Complete Micronutrient Nestle

Nutrition
Pediatric
Nestlé Nutrition Institute Workshop Series | Vol. 70
Meeting Micronutrient
Requirements for
Health and Development
Zulfiqar A. Bhutta
Richard F. Hurrell
Irwin H. Rosenberg
Meeting Micronutrient Requirements for Health and Development
Nestlé Nutrition Institute
Workshop Series
Vol. 70
Meeting Micronutrient
Requirements for
Health and Development
Editors
Zulfiqar A. Bhutta Karachi, Pakistan

Richard F. Hurrell Zurich, Switzerland
Irwin H. Rosenberg Boston, MA, USA
Nestec Ltd., 55 Avenue Nestlé, CH–1800 Vevey (Switzerland)
S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com
Library of Congress Cataloging-in-Publication Data
Meeting micronutrient requirements for health and development / editors,

Zulfiqar A. Bhutta, Richard F. Hurrell, Irwin H. Rosenberg.
p. ; cm. -- (Nestlé Nutrition Institute workshop series, ISSN
1664-2147 ; v.70)
Includes bibliographical references and index.
ISBN 978-3-318-02111-0 (hard cover : alk. paper) -- ISBN 978-3-318-02112-7
(electronic version)
I. Bhutta, Zulfiqar Ahmed, 1955- II. Hurrell, Richard. III. Rosenberg,
Irwin H. IV. Series: Nestlé Nutrition Institute workshop series ; v.70.
1664-2147
[DNLM: 1. Micronutrients--Congresses. 2. Nutritional
Requirements--Congresses. 3. Child. 4. Deficiency Diseases--Congresses.
W1 NE228D v.70 2012 / WS 130]
572'.515--dc23
2012025488
The material contained in this volume was submitted as previously unpublished material, except in the instances in
which credit has been given to the source from which some of the illustrative material was derived.
Great care has been taken to maintain the accuracy of the information contained in the volume. However, neither
Nestec Ltd. nor S. Karger AG can be held responsible for errors or for any consequences arising from the use of the
information contained herein.
© 2012 Nestec Ltd., Vevey (Switzerland) and S. Karger AG, Basel (Switzerland). All rights reserved. This book is
protected by copyright. No part of it may be reproduced, stored in a retrieval system, or transmitted, in any form or
by any means, electronic, mechanical, photocopying, or recording, or otherwise, without the written permission of
the publisher.
Printed on acid-free and non-aging paper

ISBN 978–3–318–02111–0
e-ISBN 978–3–318–02112–7
ISSN 1664–2147
e-ISSN 1664–2155
Basel · Freiburg · Paris · London · New York · New Delhi · Bangkok ·

Beijing · Tokyo · Kuala Lumpur · Singapore · Sydney
Contents
IX Preface
XI Foreword
XIV Contributors
Introduction to Micronutrient Requirements

1 Global Micronutrient Deficiencies in Childhood and Impact on Growth
and Survival: Challenges and Opportunities
Imdad, A.; Bhutta, Z.A. (Pakistan)
11 Micronutrients in the Treatment of Stunting and Moderate
Malnutrition
Penny, M.E. (Peru)
22 Discussion on Micronutrient Requirements
Solomons, N.W. (Guatemala)
Zinc in Maternal and Child Health

27 Zinc Requirements: Assessment and Population Needs
Hambidge, K.M.; Miller, L.V.; Krebs, N.F. (USA)
37 Role of Zinc in Child Health and Survival
Black, R.E.; Fischer Walker, C. (USA)
43 Discussion on Zinc in Maternal and Child Health
Wasantwisut, E. (Thailand)
Multiple Micronutrient Deficiencies in Pregnancy

49 Global Burden and Significance of Multiple Micronutrient Deficiencies
in Pregnancy
Darnton-Hill, I. (Australia/USA)
V
61 Intervention Strategies to Address Multiple Micronutrient Deficiencies
in Pregnancy and Early Childhood
Imdad, A.; Bhutta, Z.A. (Pakistan)
74 Summary on Micronutrient Requirements and Deficiencies in
Maternal and Child Nutrition
Bhutta, Z. (Pakistan)
Vitamin A in Childhood: Evidence and Controversy

79 Vitamin A Supplementation, Infectious Disease and Child Mortality:
A Summary of the Evidence
Thorne-Lyman, A.; Fawzi, W.W. (USA)
91 Issues and Controversies with Vitamin A in Childhood
Murguía Peniche, M.T. (Mexico)
103 Discussion on Vitamin A Supplementation in Childhood
Bhatia, J. (USA)
Iron
107 Influence of Inflammatory Disorders and Infection on Iron Absorption
and Efficacy of Iron-Fortified Foods
Hurrell, R.F. (Switzerland)
117 Safety of Iron Fortification and Supplementation in Malaria-Endemic
Areas
Brittenham, G.M. (USA)
128 Discussion on Iron
Lynch, S. (USA)
134 Summary on Vitamin A and Iron
Hurrell, R.F. (Switzerland)
Iodine
137 Are Weaning Infants at Risk of Iodine Deficiency Even in Countries
with Established Iodized Salt Programs?
Zimmermann, M.B. (Switzerland)
147 Current Challenges in Meeting Global Iodine Requirements
Eastman, C.J. (Australia); Jooste, P. (South Africa)
Folate, Vitamin B12 and Brain

161 Folate and Vitamin B12: Function and Importance in Cognitive
Development
Troen, A.M. (Israel)
VI Contents
172 Discussion on Folate and Vitamin B12 Importance in Cognitive
Development
Rosenberg, I.H. (USA)
175 Pros and Cons of Increasing Folic Acid and Vitamin B12 Intake by
Fortification
Allen, L.H. (USA)
184 Discussion on Vitamin B12 and Folic Acid Fortification
187 Summary on Iodine Folate and Vitamin B12
191 Subject Index
For more information on related publications, please consult the NNI website:
www.nestlenutrition–institute.org
Contents VII
Preface
In the past two decades, micronutrients have been recognized as having a

more prominent position in the global effort to control and overcome hun-
ger and malnutrition. The three micronutrients whose deficiencies have the
greatest prevalence and have attracted the greatest attention from interna-
tional agencies and as well as institutions and not-for-profit agencies are vita-
min A, iodine and iron. Recently, zinc has received much attention for its
role in the treatment of diarrhea and its preventive potential for childhood
diarrhea and pneumonia. Even with this increased attention and enhanced
research, a number of important questions remain. What is the importance
of these micronutrients and others, including folate and vitamin B12, as lim-
iting nutrients in growth and development and their potential for prevent-
ing stunting and global malnutrition? What are the forms and doses of these
micronutrients that may be required for the prevention of stunting and mal-
nutrition or the treatment of moderate and severe malnutrition? What are
the effective interventions including individual or multiple micronutrient
interventions, and what are the effective methods of addition of these micro-
nutrients to the food supply and in the form of fortification?In which form
are these micronutrients most effective and most appropriately or most effec-
tively absorbed and bioavailable? Are there any potential safety concerns? Do
interactions restrict potential coadministration? And perhaps most impor-
tantly, what are the appropriate target populations for interventions in treat-
ment or prevention?
To address the current knowledge in these and other related questions,
Nestlé Nutrition Institute invited three nutrition scientists to organize a sym-
posium with other leading scientists knowledgeable and active in the field of
micronutrient nutrition. These scientists were invited to give their presentations
at the meeting in Cebu, Philippines, in April 2011, along with an invited group
of knowledgeable leaders in the fields of pediatric nutrition and micronutrient
nutrition from all parts of the world. The papers that were presented at this
meeting and a summary of the discussions form the substance of this publica-
tion and report. We hope that this publication is another milestone in the path
IX
towards a better understanding of the importance of micronutrients in nutrition
and development, and that it willcontribute to the growing urgency for the con-
trol of childhood malnutrition.
Zulfiqar A. Bhutta
Richard F. Hurrell
Irwin H. Rosenberg
X Bhutta · Hurrell · Rosenberg

Foreword
Meeting macro- and micronutrient requirements during pregnancy and early

childhood (‘the first 1,000 days’) is crucial for short- and long-term health and
cognitive function. Recent surveys on the timing of growth retardation and
stunting indicate that the onset is maternal under nutrition before and during
pregnancy. Inappropriate weaning food and frequent infections hamper growth
and development during the ‘critical window’ of the first 24 months after birth.
Stunting (178 million children world wide) and severe acute malnutrition (wast-
ing –19 million) are associated with micronutrient deficiencies. Epidemiology
indicates that the most common micronutrient deficiencies – vitamin A, iron,
zinc, and iodine – could be related to an estimated 1 million child deaths per
year and 9% of global childhood DALYs [1]. Meta-analyses confirm that supple-
mentation or fortification of food with the ‘big four’ is efficacious to reduce risk
of infectious disease, improves growth and cognitive outcome. More recently,
folate and vitamin B12 deficiencies during pregnancy have been shown to be
associated with poor neurodevelopmental outcome and childhood obesity.
Because of the high prevalence of micronutrient deficiencies in populations
from developing countries, the challenge is to develop the right intervention
strategies. Governmental agencies and international non-governmental organi-
zations need strong support from scientific institutions, and industry should
cooperate with all relevant stakeholders.
The 70th Nestlé Nutrition Institute Workshop, which took place in March
2011 in Cebu, Philippines, was the third to address micronutrient needs. Two
previous NNI workshops addressed the needs during the first months of life
(NNIW 52) and the weaning period (NNIW 54). It became clear during the last
years and was addressed at this workshop that maternal and fetal deficiencies
can induce inadequate metabolic programming in the offspring with increased
risk for non-communicable diseases later in life. In order to answer questions
and lead scientific discussions, we asked world-renowned experts in the area of
health science and nutrition to clarify the pathogenesis of micronutrient defi-
ciencies in pregnancy and childhood, preventive methods and strategies, and
opportunities for treatment.
XI
To discuss the most recent findings and outcome strategies, leading experts in
the fields of epidemiology and nutritional intervention met with those in genet-
ics, epigenetics, and metabolic outcome. We would like to warmly acknowledge
the excellent program conceived by the chairpersons – Prof. Zulfiqar A. Bhutta,
Pakistan, Prof. Richard F. Hurrell, Switzerland, and Prof. Irwin H. Rosenberg,
USA. We are also indebted to all the renowned speakers giving the presenta-
tions, and discussants leading the debates of this important topic. We thank all
experts who came from across the globe to review and discuss the importance
of the micronutrients in child life and the opportunity to meet them.
Finally, we wish to thank Dr. Marco Turini and Dr. Grace L. Uy with their
teams from the Nestlé Nutrition Institutes in South East Asia – Philippines for
excellent logistic support and hospitality which allowed the participants to enjoy
both the scientific program and the wonderful cultural spirit of Cebu.
Reference
1 Imdad A, Bhutta ZA: Global micronutrient
deficiencies in childhood and impact on
growth andsurvival: challenges and opportu-
nities. Nestlé NutrInst Workshop Ser. Nestec,
Vevey/Basel, Karger, vol 70, 2012, pp 1–10.
Prof. Ferdinand Haschke, MD, PhD Natalia Wagemans, MD, PhD

Chairman Global Medical Advisor
Nestlé Nutrition Institute Nestlé Nutrition Institute
Vevey, Switzerland Vevey, Switzerland
XII Haschke · Wagemans

70th Nestlé Nutrition Institute Workshop
Cebu, Phillipines, March 27–30, 2011
Contributors
Chairpersons & Speakers
Prof. Lindsay H. Allen Prof. Ian Darnton-Hill

Department of Nutrition The Friedman School of Nutrition
USDA ARS Western Human Nutrition Science and Policy
Research Center Tufts University
430 West Health Sciences Dr. c/o 71 Broadway, Apt 2E
University of California New York, NY 10006
Davis, CA 95616 USA
USA E-Mail: ian.darnton-hill@sydney.edu.au
E-Mail: lhallen@ucdavis.edu
Prof. Wafaei W. Fawzi
Prof. Zulfiqar A. Bhutta Harvard School of Public Health
Aga Khan University Hospital Department of Global Health and
Stadium Road Population
PO Box 3500 655 Huntington Avenue
Karachi 74800 Building I, Room 1102A
Pakistan Boston, MA 02115
E-Mail: zulfiqar.bhutta@aku.edu USA
E-Mail: mina@hsph.harvard.edu
Prof. Robert E. Black
Johns Hopkins University Prof. K. Michael Hambidge
Bloomberg School of Public Health Pediatrics
Department of International Health University of Colorado Anschutz Medical
615 N. Wolfe Street, E-8527 Center
Baltimore, MD 21205 12700 East 19th Avenue, Box C225,
USA Room 5022
E-Mail: rblack@jhsph.edu Aurora, CO 80045
USA
Prof. Gary M. Brittenham E-Mail: michael.hambidge@ucdenver.edu
Columbia University Medical Center
CHONY, Room CHN 10-08 Prof. Richard F. Hurrell
3959 Broadway Swiss Federal Institute of Technology
New York, NY 10032 ETH
USA Institute of Food, Nutrition and Health
E-Mail: gmb31@columbia.edu ETH Zentrum, LFV D20
Schmelzbergstrasse 7
8092 Zurich
Switzerland
E-Mail: richard.hurrell@ilw.agrl.ethz.ch
XIV
Prof. Pieter Jooste Prof. Michael B. Zimmermann
Nutritional Intervention Research Unit Swiss Federal Institute of Technology ETH
Medical Research Council Laboratory for Human Nutrition
PO Box 19070,Tygerberg 7505 Institute of Food, Nutrition and Health
Cape Town Schmelzbergstrasse 7, LFVE19
South Africa 8092 Zurich
E-Mail: Pieter.Jooste@mrc.ac.za Switzerland
E-Mail: michael.zimmermann@ilw.agrl.
Prof. MarÍa Teresa Murguia Peniche ethz.ch
Infancy Area
National Center for Child and Moderators
Adolescent Health(CeNSIA)
Francisco delP. Miranda 177-1er piso Dr. Genesis C. Rivera
Col Merced Gomez, Delegacion Alvaro Philippine Children’s Medical Center
Obregón Agham Road
CP 01600 Mexico City Quezon City
Mexico Philippines
E-Mail:teresamurguiap@gmail.com E-Mail: docgen.pps@gmail.com
Prof. Mary Edith Penny Dr. Socorro de Leon-Mendoza

Institute of Nutritional Investigation Asian Hospital and Medical Center
Av. La Molina 1885 Civic Drive, FilinvestAlabang
La Molina Muntinlupa City
Lima 12 Philippines
Peru E-Mail: drsookee@yahoo.com
E-Mail: mpenny@iin.sld.pe
Dr. Carmencita d. Padilla
Prof. Irwin H. Rosenberg University of the Philippines
Nutrition and Neurocognition PGH Medical Center
Laboratory Pedro Gil St.
Jean Mayer USDA HNRCA Manila
at Tufts University Philippines
711 Washington Street E-Mail: cdpadilla@upm.edu.ph
Boston, MA 02111-1524
USA Discussants
E-Mail: irwin.rosenberg@tufts.edu
Prof. Jatinder Bhatia
Prof. Aron M. Troen Division of Neonatology
The Hebrew University of Jerusalem Medical College of Georgia
The Robert H. Smith Faculty of 1120 15th Street, BIW 6033
Agriculture, Food and Environment Augusta, GA 30912-3740
Institute of Biochemistry USA
Food Science and Nutrition E-Mail: jatindeb@mcg.edu
Nutrition and Brain Health Laboratory
PO Box 12 Prof. Wafaei W. Fawzi
Rehovot 76100 Harvard School of Public Health
Israel Department of Global Health and
E-Mail: Troen@agri.huji.ac.il Population
655 Huntington Avenue
Building I, Room 1102A
Boston, MA 02115
USA
E-Mail: mina@hsph.harvard.edu
Contributors XV
Prof. Sean Lynch Mario De Curtis/Italy
151 Breezy Point Drive Vassilios Fanos/Italy
Yorktown, VA 23692 Betty Samburu/Kenya
USA Mohamed El Sayeed/Kuwait
E-Mail: snlnch6@gmail.com Kongsy Bangsitthideth/Laos
Joseph Hadad/Lebanon
Prof. Noel Solomons Chon Joey, Malaysia
Center for Studies of Sensory Mei Ching Wong, Malaysia
Impairment Mohamed Maslina/Malaysia
Aging, and Metabolism (CeSSIAM) Mohd Nor Azam/Malaysia
Avenida 17, 16-89 (interior) Soh Lay Sian/Malaysia
Zona 11 (AnilloPeriférico) Agostinho Joaquim Anaunama/
Guatemala, 01011, C.A. Mozambique
Guatemala Kay ThiHtun/Myanmar
Email: cessiam@guate.net.gt Melinda Atienza/Philippines
Efren Balanag/Philippines
Prof. Emorn Wasantwisut Grace Battad/Philippines
Department of Nutrition Consolacion Botin/Philippines
Mahidol University Cheryl Bullo/Philippines
Phuttamonthon 4 Rd. Monina Cristina Cabral/Philippines
Salaya, Phuttamonthon Pamela Caedo/Philippines
Nakhon Pathorn 73170 Mario Capanzana/Philippines
Thailand Josie Grace Castillo/Philippines
E-Mail: numdk@mahidol.ac.th Babylyn Cayabyab/Philippines
Vivina Chiu/Philippines
Participants Sylvia Estrada/Philippines
Rodolfo Florentino/Philippines
Peter Fryer/Australia Felizardo Gatcheco/Philippines
Shao Zhou/ Australia Arturo Ludan/Philippines
H.S. Kamrul Alam/Bangladesh Mike Manalaysay/Philippines
Jahangir Alam/Bangladesh Jacinto Blas Mantaring/Philippines
Safiul Hoque/Bangladesh Socorro Mendoza/Philippines
Reaz Mobarak/Bangladesh Allan Nacion/Philippines
Fabíola Souza/Brazil Romeo Nuguid/Philippines
Kirtchhoof Keo/Cambodia Carmencita Padilla/Philippines
Inocent Gouado/Cameroun Genesis Rivera/Philippines
Celestin Nsibu/Congo Republic Jose Salazar/Philippines
Anjan Bhattacharya/India Christine Tanbonliong/Philippines
Bikash Bhattacharya/India Randy Urtula/Philippines
Sanjeev Ganguly/India Grace L. Uy/Philippines
Pankaj Garg/India Natalya Migacheva/Russia
Radha Krishna Kandula/India Natalia Wagemans/Russia
Sunil Kumar/India Guelaye Sall/Senegal
Anand N. Pandit/India Joyce Fong/Singapore
Premlata N. Sheth/India Fung Chi, Angelin Lin/Singapore
Pramita Dwipoerwantoro/Indonesia Wendy Sinnathamby/Singapore
Aryono Hendarto/Indonesia Ratna Sridjaja/Singapore
Hanifah Oswari/Indonesia Marco Turini/Singapore
Ines Yumahana/Indonesia Mohammed Ali Dhansay/South Africa
Morteza Abdollahi/Iran Y.R. Savithri De Silva/Sri Lanka
Majid Hajifaraji/Iran Theophilus Ajitha Fernando/Sri Lanka
Naser Kalantari/Iran Denis Barclay/Switzerland
Nasrin Omidvar/Iran Stefan Bodenstab/Switzerland
XVI Contributors
Urs Bruegger/Switzerland Nalinee Chongviriyaphan/Thailand
Ferdinand Haschke/Switzerland Ladda Mo-Suwan/Thailand
Emma Jacquier/Switzerland Krittawan Sirisomboonwong/Thailand
Petra Klassen/Switzerland Pattanee Winichagoon/Thailand
Monika Potter/Switzerland Hanan Anwar/United Arab Emirates
Jörg Spieldenner/Switzerland Tran Ngoc Diep/Vietnam
Pascal Volery/Switzerland Nguyen Thi Linh Giang/Vietnam
Simon Wieser/Switzerland Nguyen Thanh Ha/Vietnam
Ahmad Al Sahhar/Syria Ngo Minh Xuan/Vietnam
Contributors XVII
Bhutta ZA, Hurrell RF, Rosenberg IH (eds): Meeting Micronutrient Requirements for Health and
Development. Nestlé Nutr Inst Workshop Ser, vol 70, pp 1–10,
Nestec Ltd., Vevey/S. Karger AG., Basel, © 2012
Global Micronutrient Deficiencies in

Childhood and Impact on Growth and
Survival: Challenges and Opportunities
Aamer Imdad ⭈ Zulfiqar A. Bhutta
Division of Women and Child Health, Aga Khan University, Karachi, Pakistan
Abstract
Despite numerous advances and improvements in child health, malnutrition still remains
as one of the main public health challenges of the 21st century, particularly in developing
countries. It undermines the survival, growth and development of children, and is associ-
ated with almost 35% of all deaths in children under the age of 5 years worldwide. An
estimated 178 million children are stunted globally, and an additional 19 million children
have severe acute malnutrition (wasting). These conditions are very often associated with
concomitant micronutrient deficiencies, and among these, vitamin A, iron, zinc and iodine
deficiencies are the most prevalent in childhood. Vitamin A and zinc deficiency is associ-
ated with an estimated 1 million child deaths and 9% of global childhood disability-
adjusted life years. Recent data on the timing of growth retardation and stunting in infants
suggest that the onset is commensurate with inappropriate complementary feeding and
potentially compounded by maternal undernutrition and intrauterine growth retardation,
and that the first 24 months represent a critical window of opportunity for intervention.
Given the wide prevalence of multiple micronutrient deficiencies in malnourished children
in developing countries, the challenge is to implement intervention strategies that com-
bine appropriate infant and young child feeding with micronutrient interventions at scale.
Emerging data from community intervention trials now provide evidence that this is both
tangible and can lead to alleviation of childhood undernutrition. Some of these recent
findings will be discussed. Copyright © 2012 S. Karger AG, Basel
Introduction
Despite numerous advances and improvements in child health, malnutrition

still remains as one of the main public health challenges of the 21st century,
particularly in developing countries [1, 2]. It undermines the survival, growth
and development of children, and is associated with almost 35% of all deaths
in children under the age of 5 years worldwide [3]. An estimated 178 million
children are stunted globally and an additional 19 million children have severe
acute malnutrition (wasting) [3]. Many of these conditions are associated with
concomitant micronutrient deficiencies, and among these, vitamin A, iron, zinc
and iodine deficiencies are the most prevalent in childhood. Vitamin A and
zinc deficiency is associated with an estimated 1 million child deaths and 9% of
global childhood disability-adjusted life years [3]. Recent data on the timing of
growth retardation and stunting in infants suggest that the onset is commensu-
rate with inappropriate complementary feeding and potentially compounded by
maternal undernutrition and intrauterine growth retardation, and that the first
24 months represent a critical window of opportunity for intervention [4].
The relationship between micronutrient deficiencies and increased risk of
infections and mortality is well established [5–7]. Several micronutrients play
an important role in the immune system, and are critical in determining the
outcome of host microbe interactions [8, 9]. Infections on the other hand are a
risk factor for malnutrition as during an episode of infection, there is a general
decrease in nutrient intake, increase in losses (e.g. GI losses, fluid losses, etc.)
and altered metabolic pathways [9]. This condition is more prevalent among
poor children whose micronutrient status is already marginal, and they also have
a high frequency of infectious disease. This leads to a vicious cycle where mal-
nutrition is a health outcome as well as a risk factor for disease and exacerbation
of malnutrition [10]. The resulting complex and mutually adverse interactions
with infections constitute the major determinants of childhood morbidity and
mortality among the underprivileged preschool children in several developing
countries [11].
Knowledge of prevalence of micronutrient deficiencies and their role in
childhood morbidity and mortality is of great importance in planning compre-
hensive strategies to promote child health and survival in developing countries
[5]. The purpose of this chapter was to summarize the current knowledge on
micronutrient deficiencies and their role in reducing morbidity and mortality
during childhood. A literature search was conducted on PubMed, Cochrane
library and WHO and UNICEF databases. We discuss the epidemiology, effec-
tive intervention and challenges and opportunities for vitamin A, zinc, iodine,
iron and other micronutrients.
Vitamin A
Vitamin A has been termed an anti-infectious agent [12, 13], and it plays an
important role in the visual system [14]. Vitamin A deficiency (VAD) impairs
numerous body functions, and can lead to many adverse health consequences
including xerophthalmia (dry eyes), infectious morbidity, mortality, suboptimal
2 Imdad · Bhutta
physical growth and anemia [12, 15]. VAD is a major nutritional public health
problem in the developing world. According to the latest report of the WHO,
globally about 190 million preschool-aged children and 19.1 million pregnant
women are vitamin A deficient (i.e. serum retinol <0.70 μmol/l) [16]. This cor-
responds to 33.3% of preschool-aged children and 15.3% of pregnant women
in populations at risk of VAD. According to current estimates, 122 countries
are classified as having a moderate to severe public health problem based on
biochemical VAD in preschool-aged children, and 88 countries based on bio-
chemical VAD in pregnant women [16]. Xerophthalmia, resulting from VAD,
is the primary preventable cause of blindness, and of the world’s children with
xerophthalmia, nearly half reside in South or South-East Asia, of whom more
than 85% live in India [17]. About 5.2 million preschool-aged children and 9.8
million pregnant women suffer from night blindness, which represents 0.9 and
7.8% of the population at risk of VAD, respectively. The estimates show that
Africa and South-East Asia contain the highest proportions of preschool-aged
children and pregnant females with biochemical VAD and night blindness [16].
Evidence from community trials has shown that vitamin A supplementation
(VAS) reduces all-cause mortality and diarrhea- and measles-specific mortality
[18–21]. It also reduces incidence of measles and diarrhea infection [21]. VAS
has also been shown to reduce prevalence of xerophthalmia (RR 0.31, 95% CI
0.22–0.45) and night blindness (RR 0.32, 95% CI 0.21–0.50) [21]. VAS as an
adjunct in the treatment of measles has been shown to reduce mortality (RR
0.18; 95% CI 0.03–0.61), pneumonia-specific mortality (RR 0.33; 95% CI 0.08–
0.92) and incidence of croup (RR 0.53; 95% CI 0.29–0.89) [22].
World Health Organization recommends two annual high-dose supplements
of vitamin A for every child at risk of VAD [23]. Administering vitamin A is a
simple act that can be performed by a trained health worker, community worker
or volunteer. It is one of the most cost-effective large-scale child survival inter-
ventions [24]. VAS is inexpensive on a per-child basis, with the cost of single
capsule approximately USD 0.02 [25]. The total cost of two annual doses var-
ies according to the region, and it has been estimated that supplementing each
child with two doses per year, it costs USD 1.20 per child per year for South Asia
and sub-Saharan Africa, with relatively higher costs for Central Asia and Latin
America [26].
Since 1998, large VAS programs are in place in about 193 UNICEF-targeted
countries to deliver the required dose of vitamin A [27]. Figure 1 shows the
progress of increase in VAS from 1999–2007. It can be noted that in 1999, just
16% of children in these countries received full VAS, while in 2005 this num-
ber increased to 72% [28]. In 2007, the rate of coverage dropped to 62% partly
because India increased its target group from 3 to 5 years of age [28]. An impor-
tant point to note is that least developed countries showed the most impres-
sive progress in achieving two-dose coverage [29] (fig. 2). While these figures
are encouraging, significant gaps in VAS coverage remain and continue to
Micronutrient Deficiencies in Childhood and Impact on Growth and Survival 3

80
75 72
70 61
59 72 62
60
50
Childern (%)
50 52
40
36
30
20
16 Children receiving at least one dose
10
Children receiving two doses
0
1999 2001 2003 2005 2007
Fig. 1. Progress with VAS for children 6–59 months. From UNICEF Global Database 2008.
Africa 73
Asia 70
Least developed
85
countries
Developing
71
countries
0 20 40 60 80 100
Percent
Fig. 2. VAS coverage levels: two doses. From UNICEF [29].
undermine children’s health [30] (fig. 3). An effort should be made to improve
the current strategies of vitamin A delivery to achieve at least 80% coverage on
persistent basis. Special efforts should be made to cover the hard-to-reach chil-
dren through complementary strategies, such as special outreach programs, to
reach the final 20% who have not been reached through regular programs. To
maintain the sustainability, resources should be mobilized in national budgets,
and VAS programs should be made part of integrated delivery strategies, with
continuous monitoring and tracking of progress.
Zinc
Zinc is the second most abundant transition metal in humans after iron. It is an
essential part of about 100 specific enzymes and also serves as structural ions
in transcription factors [31]. The association of zinc deficiency in children with
4 Imdad · Bhutta
≥80 %
50–79 %
<50 %
Vitamin A priority country;

data not reported
Not a vitamin A
priority country
Fig. 3. VAS coverage levels by country: two doses (2005). From UNICEF [30].
growth retardation and hypogonadism was first described in 1963 in a study

from Iran [32], and it is now well established based on animal and human stud-
ies demonstrating that zinc plays a critical role in cellular growth, cellular dif-
ferentiation and metabolism [33], and in turn promotes immunity, resistance to
infection, and the growth and development of the nervous system [31]. Although
zinc deficiency is increasingly being recognized as a widespread problem, there
is very limited nationally representative data on the magnitude and severity of
this deficiency [34]. Some of this is due in part to the lack of reliable biomarkers
of zinc status. Recently, Wuehler et al. [35], using data from national food bal-
ance sheets compiled by the Food and Agricultural Organization, estimated that
20% of the world’s population is at risk of low zinc intakes. The global preva-
lence of low intakes by region indicates that 26% of population in South Asia
and 28% in Sub-Saharan Africa are at risk of deficiency [35].
Evidence from community trials has shown that both therapeutic and pre-
ventive zinc supplementations are effective in reducing morbidity and mortality
in children <5 years of age [36–40]. Therapeutic zinc supplementation as an
adjunct in the treatment of diarrhea has been shown to reduce the duration
of acute diarrhea by 0.5 days and that of persistent diarrhea by 0.68 days [37].
Preventive zinc supplementation reduces incidence of diarrhea by 20% (RR =
0.80; 95% CI, 0.71–0.90) and that of pneumonia by 15% (RR = 0.85; 95% CI,
0.75–0.97) [36]. Preventive zinc supplementation has been shown to reduce the
rate of stunting as well [36, 41, 42].

Countries that have adopted zinc policy as
part of their national child health policy
Fig. 4. Countries that have adopted national policy on zinc supplementation. From Zinc
Task Force using data from the UNICEF, USAID and WHO.
In 2004, the WHO and UNICEF formulated a new recommendation to

administer zinc for 10–14 days as an adjunct treatment for diarrhea, along with
low-osmolarity oral rehydration solutions and continuation of feeding [43].
Since then, the WHO and UNICEF in collaboration with USAID and Johns
Hopkins University has worked to ensure the availability of zinc products [44],
and about 91 million tablets had been provided in the year 2008 [45]. Despite
all these efforts, zinc supplementation is not part of national programs around
the globe. Figure 4 shows that only 46 countries have adopted zinc policy as
part of their national child health policy [45]. It is therefore required to scale up
the zinc supplementation, and it should be incorporated into national diarrhea
management policy. Adequate funds with possible public and private partner-
ships should be sorted out, and general awareness about the effectiveness of zinc
supplementations should be raised through media campaigns.
Iodine
Iodine is the key element required for thyroid hormone synthesis, and is also
important for brain development during fetal and early years of life [46]. Iodine
deficiency is the primary cause of preventable mental retardation and brain dam-
age and also increases the chance of infant mortality, miscarriage and stillbirth
6 Imdad · Bhutta
Category of public health significance
(based on median urinary iodine)
Moderate iodine deficiency (20–49 μg/l)
Mild iodine deficiency (50–99 μg/l)
Optimal (100–199 μg/l)
Risk of iodine-induced hyperthyroidism (200–299 μg/l)
Risk of adverse health consequences (>300 μg/l)
No data
Fig. 5. Prevalence of iodine deficiency in school-aged children [47].
[46]. Children born to iodine-deficient mothers may appear normal at birth, but
might have suffered brain damage and loss in IQ points, affecting their ability to
develop to their full potential. These seemingly normal children will later have dif-
ficulty learning in school and staying in school. It has been shown that in commu-
nities where iodine intake is sufficient, average IQ is on average 13 points higher
than in iodine-deficient communities [46]. According to an estimate, about 2 bil-
lion people have insufficient iodine intake around the globe and about 31.5% of
school-age children (266 million) have insufficient iodine intake (fig. 5) [47].
Salt iodization is one of the exemplary success stories of food fortification
offering great benefits for the intellectual health of nations that have embraced it
[48]. The number of countries where iodine deficiency disorders were a public
health concern has been reduced by more than half from 1993 to 2007 [47].
Thirty-four developing countries have achieved the universal salt iodization
goal, and an additional 38 countries are considered ‘on track’ for elimination
of iodine deficiency disorders [48]. These are countries that have either shown
increases in coverage of at least 20% over the previous decade or that have
reached between 80 and 89% coverage with no indication of possible decline
[48]. Despite this progress, many countries are lagging far behind. Twenty-four
countries have experienced no growth in coverage rates or have even experi-
enced a decline since the mid-1990s [48]. In 12 countries, less than 20% of the
population are consuming adequately iodized salt [48].

In order to achieve universal salt iodization, country level legislation should
be done, and adequate funds should be ensured to enforce it. At the same time,
media campaigns should be launched for general awareness of the importance
of salt iodization. Population monitoring systems should be strengthened so
that program adjustments can be made as habits and diets change over time.
Other Micronutrients
Iron is an essential mineral for human development and function. It is required

for hemoglobin and is critical for motor and cognitive development in child-
hood, and for physical activity in all humans [49]. Iron is also critical to the
health of a pregnant mother and her unborn child. A woman needs more iron
during pregnancy because the fetus and placenta both need additional iron.
Iron supplementation during pregnancy lowers the risk of maternal mortality
due to hemorrhage, the cause of more than 130,000 maternal deaths each year
[27]. Supplementation also helps to lower the risks of premature birth and low
birthweight. Studies have shown that infants with anemia caused by iron defi-
ciency have lower mental scores and lower motor scores than infants without
anemia [50]. Ensuring sufficient iron levels in the first months and years of life
is, therefore, critical.
Conclusions
Micronutrients like vitamin A, zinc, iodine and iron are important for growth
and survival of children. Given the wide prevalence of multiple micronutrient
deficiencies in malnourished children in developing countries, the challenge is
to implement intervention strategies that combine appropriate infant and young
child feeding with micronutrient interventions at scale. Emerging data from
community intervention trials now provide evidence that this is both tangible
and can lead to alleviation of childhood undernutrition.
References
1 Khan Y, Bhutta ZA: Nutritional deficiencies 3 Black RE, Allen LH, Bhutta ZA, et al:
in the developing world: current status and Maternal and child undernutrition: global and
opportunities for intervention. Pediatr Clin regional exposures and health consequences.
North Am 2010;57:1409–1441. Lancet 2008;371:243–260.
2 Harrison GG, Tirado MC, Galal OM: 4 Victora CG, de Onis M, Hallal PC, et al:
Backsliding against malnutrition. Asia Pac J Worldwide timing of growth faltering:
Public Health 2010;22(suppl 3):246S–253S. revisiting implications for interventions.
Pediatrics 2010;125:e473–e480.
8 Imdad · Bhutta
5 Bhutta ZA: Micronutrient needs of malnour- 19 Fawzi WW, Chalmers TC, Herrera MG,
ished children. Curr Opin Clin Nutr Metab Mosteller F: Vitamin A supplementation
Care 2008;11:309–314. and child mortality. A meta-analysis. JAMA
6 Shann F, Duke T: Paediatrics. Micro- 1993;269:898–903.
nutrients, measles, malaria, and mortality. 20 Glasziou PP, Mackerras DE: Vitamin A
Lancet 1995;346(suppl):s16. supplementation in infectious diseases: a
7 Caulfield LE, de Onis M, Blossner M, Black meta-analysis. BMJ 1993;306:366–370.
RE: Undernutrition as an underlying cause of 21 Imdad A, Herzer K, Mayo-Wilson E, et al:
child deaths associated with diarrhea, pneu- Vitamin A supplementation for preventing
monia, malaria, and measles. Am J Clin Nutr morbidity and mortality in children from 6
2004;80:193–198. months to 5 years of age. Cochrane Database
8 Katona P, Katona-Apte J: The interaction Syst Rev 2010;CD008524.
between nutrition and infection. Clin Infect 22 Yang HM, Mao M, Wan CM: Vitamin A
Dis 2008;46:1582–1588. for treating measles in children. Cochrane
9 Solomons NW: Malnutrition and infection: Database Syst Rev 2005;CD001479.
an update. Br J Nutr 2007;98(Suppl 1):S5– 23 World Health Organization: Vitamin A
S10. supplements: a guide to their use in preven-
10 Khan YP, Yakoob MY, Bhutta ZA: Maternal tion and treatment of vitamin A deficiency
mineral and vitamin supplementation in and xerophthalmia. Geneva, World Health
pregnancy. Expert Rev Obstet Gynecol Organization, 1997.
2010;5:241–256. 24 Fiedler JL, Dado DR, Maglalang H, et al:
11 UNICEF: Tracking Progress on Child and Cost analysis as a vitamin A program design
Maternal Nutrition. New York, UNICEF, and evaluation tool: a case study of the
2009. http://www.unicef.org/publications/ Philippines. Soc Sci Med 2000;51:223–242.
index_51656.html. 25 The micronutrient initiative, vitamin A, the
12 Bates CJ: Vitamin A. Lancet 1995;345:31–35. scope of the problem. http://www.micronu-
13 Green H, Mellanby E: Vitamin A as an anti- trient.org/English/view.asp?x=577.
infective agent. BMJ 1928;2:691–696. 26 Horton S: Best Practice Paper – Micro-
14 Sommer A, Tarwotjo I, Djunaedi E, et al: nutrient Supplements for Child Survival
Impact of vitamin A supplementation on (Vitamin A and Zinc). Copenhagen,
childhood mortality. A randomised con- Copenhagen Consensus Center, 2008.
trolled community trial. Lancet 1986;1: 27 UNICEF: The State of the World’s Children
1169–1173. 2009. New York, UNICEF, 2009.
15 Rice AL, West KP Jr, Black RE: Vitamin A 28 UNICEF: Vitamin A Supplementation: A
Deficiency. Global and Regional Burden of Decade of Progress. New York, UNICEF, 2007.
Disease Attributable to Selected Major Risk 29 UNICEF: Eastern and Southern Africa.
Factors. Geneva, World Health Organization, Multi-country Evaluation of Child Health
2004, vol 1. Days in Eastern and Southern Africa. New
16 World Health Organization: Global York, UNICEF, 2008, pp v–vi.
Prevalence of Vitamin A Deficiency in 30 UNICEF: Progress for Children. http://www.
Populations at Risk 1995–2005. WHO Global unicef.org/progressforchildren/2007n6/
Database on Vitamin A Deficiency. Geneva, index_41510.htm.
World Health Organization, 2009. 31 Prasad A: Biochemistry of Zinc. New York,
17 West KP Jr: Extent of vitamin A deficiency Plenum Press, 1993.
among preschool children and women of 32 Moynahan EJ: Letter: acrodermatitis entero-
reproductive age. J Nutr 2002;132(9 suppl): pathica: a lethal inherited human zinc-
2857S–2866S. deficiency disorder. Lancet 1974;ii:399–400.
18 Beaton G, Martorell R, Aronson K, et al: 33 Prasad AS: Discovery of human zinc defi-
Effectiveness of vitamin A supplementation ciency and studies in an experimental human
in the control of young child morbidity and model. Am J Clin Nutr 1991;53:403–412.
mortality in developing countries. ACC/ 34 Ramakrishnan U: Prevalence of micronu-
SCN State of the Art Series, Nutrition Policy trient malnutrition worldwide. Nutr Rev
Discussion. 1993;13. 2002;60(5 Pt 2):S46–S52.

35 Wuehler SE, Peerson JM, Brown KH: Use of 43 WHO/UNICEF: Clinical management of
national food balance data to estimate the acute diarrhoea. 2004. http://www.who.
adequacy of zinc in national food supplies: int/child_adolescent_health/documents/
methodology and regional estimates. Public who_fch_cah_04_7/en/index.html.
Health Nutr 2005;8:812–819. 44 Fischer Walker CL, Fontaine O, Young MW,
36 Brown KH, Peerson JM, Baker SK, Hess Black RE: Zinc and low osmolarity oral
SY: Preventive zinc supplementation rehydration salts for diarrhoea: a renewed
among infants, preschoolers, and older call to action. Bull World Health Organ 2009;
prepubertal children. Food Nutr Bull 2009; 87:780–786.
30(1 suppl):S12–S40. 45 UNICEF: Zinc for management of diar-
37 Haider BA, Bhutta ZA: The effect of rhoea: consultation with pharmaceutical
therapeutic zinc supplementation among manufacturers. 2008. www.unicef.org/
young children with selected infections: supply/files/4a__Zinc_Jan_Komrska_and_
a review of the evidence. Food Nutr Bull Francisco_Blanco(1).pdf.
2009;30(suppl 1):S41–S59. 46 Zimmermann MB, Jooste PL, Pandav
38 Walker CL, Black RE: Zinc for the treatment CS: Iodine-deficiency disorders. Lancet
of diarrhoea: effect on diarrhoea morbidity, 2008;372:1251–1262.
mortality and incidence of future episodes. 47 de Benoist B, McLean E, Andersson M,
Int J Epidemiol 2010;39(suppl 1):i63–i69. Rogers L: Iodine deficiency in 2007:
39 Patel A, Mamtani M, Dibley MJ, et al: global progress since 2003. Food Nutr Bull
Therapeutic value of zinc supplementation 2008;29:195–202.
in acute and persistent diarrhea: a systematic 48 UNICEF: Sustainable Elimination of Iodine
review. PLoS One 2010;5:e10386. Deficiency: progress since the 1990 World
40 Lassi ZS, Haider BA, Bhutta ZA: Zinc Summit for Children, New York, UNICEF,
supplementation for the prevention of 2008.
pneumonia in children aged 2 months to 59 49 Stoltzfus RJ, Mullany L, Black RE: Iron
months. Cochrane Database Syst Rev 2010; deficiency anaemia; in Ezzati M, Lopez AD,
CD005978. Rodgers A, Murray CLJ (eds): Comparative
41 Imdad A, Bhutta ZA: Effect of preventive Quantification of Health Risks: Global and
zinc supplementation on linear growth in Regional Burden of Disease Attributable to
children under 5 years of age in developing Selected Major Risk Factors. Geneva, World
countries: a meta-analysis of studies for input Health Organization, 2004, pp 163–209.
to the Lives Saved Tool. BMC Public Health 50 Walker SP, Wachs TD, Gardner JM, et al:
2011;11(suppl 3):S22. Child development: risk factors for adverse
42 Ramakrishnan U, Nguyen P, Martorell R: outcomes in developing countries. Lancet
Effects of micronutrients on growth of chil- 2007;369:145–157.
dren under 5 y of age: meta-analyses of single
and multiple nutrient interventions. Am J
Clin Nutr 2009;89:191–203.
10 Imdad · Bhutta
Micronutrients in the Treatment of

Stunting and Moderate Malnutrition
Mary Edith Penny
Instituto de Investigación Nutricional, Lima, Perú
Abstract
Linear growth retardation or stunting may occur with or without low weight-for-age, but
in both cases stunted or moderately malnourished children are deficient in micronutrients.
Pregnancy and the first 2 years are critical periods. Dietary deficiency of zinc, iron, calcium,
and vitamin A are especially common and often occur together. Zinc is essential for ade-
quate growth, and supplements have been shown to increase intrauterine femur length
and to prevent stunting. However, in general, supplements which provide a mixture of
micronutrients have been more successful in preventing stunting and are simpler to take
and distribute. Multiple micronutrients together with energy and macronutrients are also
needed for the management of moderate malnutrition. Multiple micronutrients may be
delivered as medicinal-like supplements, but may also be combined with food, for instance
in milk drinks, in fortified dried cereal mixes used to supplement complementary foods or
in lipid nutrition supplements. The latter also provide essential fats necessary for growth.
Micronutrient powders for home fortification are effective in preventing anemia, but pres-
ent combinations do not prevent stunting. Improving the diets of infant and young chil-
dren is also possible, and increased intake of animal source foods can improve growth.
Copyright © 2012 S. Karger AG, Basel
Introduction
Infant and child nutritional status is expressed as a weight-for-age, height-for-

age, and weight-for-height z scores in relation to the median (50th centile) of
a reference population. Linear growth retardation leads to stunting, which is
defined as length or height-for-age z score less than –2. Stunting is a chronic
process that represents a personal history of deprivation and has lifetime con-
sequences [1]. Low weight-for-age can be a consequence of acute malnutrition
associated with loss of fat and lean body mass resulting in thinness (low weight-
0.8
0.6
0.4
Z score 0.2
0
–0.2
–0.4
–0.6
–0.8
South Asia Latin and Central America
Mean height-for-age Z score

Mean weight-for-height Z score
Mean weight-for-age Z score
Fig. 1. Patterns of growth of 0- to 6-month-old children in South Asia and Latin and
Central America from selected national demographic and health surveys.
for-height) or result from chronic low growth rate when it is often associated
with stunting and intrauterine growth retardation. Moderate malnutrition
(MM) is defined variously as weight-for-age z scores between –2 and –3 SD
[2] or weight-for-length-/height-for-age (wasting) between –2 and –3 SD [2].
Children with MM are at risk of long-term adverse consequences [3].
The new WHO growth reference standards have provided a benchmark against
which population and individual child growth and nutritional status can be com-
pared [4], and the strikingly similar growth pattern of children from different
ethnic and geographical conditions adds to existing evidence that in the first 5
years it is possible to use a single reference to compare growth across the world.
While normal growth seems similar across populations, the pattern of growth
of children in different low resource situations is not. Figure 1 is an example of
the considerable difference in early growth between babies born in South Asia
compared with babies in South and Central America. In the latter area, the most
common growth deficit is stunting in the absence of low weight, whereas in
South Asia low birthweight is common, and more than 40% of young children
are underweight. Treatment and prevention guidelines need to take this into
consideration, and I will consider the role of individual micronutrients in the
prevention and treatment of stunting and MM separately.
Stunting
Linear growth retardation often begins in utero, but is most marked during
the vulnerable period of complementary feeding, the transition from a diet
of breast milk to the family food [5]. While dietary intakes of breast milk plus
12 Penny
complementary foods usually meet protein requirements, energy intakes may
be low, and diets are almost always low in critical micronutrients, especially cal-
cium, iron and zinc, both in absolute terms and in terms of nutrient density.
Vitamin A is also deficient in some areas, and dietary deficiency of niacin, ribo-
flavin, thiamine, B6, B12, vitamin C, vitamin D, magnesium, phosphorus and
potassium have been reported [6–9]. These micronutrients, except vitamin C,
are typically associated with animal source foods (ASFs). ASFs are especially
good sources of minerals such as iron, zinc and calcium because of relatively
high concentrations and bioavailability. Inhibitors, such as phytates and fiber,
present in cereals and legumes, reduce the dietary utility of minerals from plant
sources. Diets that are high in cereal content and low in ASFs have been associ-
ated with stunting [10, 11].
Zinc
Zinc is essential for many physiological processes, and severe zinc deficiency
leads to dwarfism [12]. Zinc deficiency is common in many populations [13].
Two out of three recent meta-analyses [14–16] of clinical trials of zinc supple-
mentation in children report a large and significant effect of daily oral zinc
supplements on linear growth especially in stunted children and in developing
countries. Imdad and Bhutta [16] reported the greatest impact when zinc was
given alone, and estimated that in children under 5 years there was a net gain of
0.37 ± 0.25 cm for a dose of 10 mg zinc daily for 24 weeks.
Antenatal zinc supplementation has been shown to increase fetal long bone
growth [17], of particular relevance because nutritional stunting is due to short
femur length [18]. A study in Nepal showed a small but significantly greater height
in school children whose mothers received antenatal zinc supplements [19], but a
large study of antenatal zinc supplements reported no effect on postnatal growth,
although the offspring of supplemented mothers had reduced prevalence of diar-
rhea and other infections [20]. Zinc supplements have been repeatedly shown to
prevent respiratory complications and diarrhea in infancy, themselves associated
with stunting. The benefit of zinc may be limited by the continuing infection or by
the presence of other nutrient deficiencies. A study combining zinc and antipara-
site treatment reported a positive effect of zinc on linear growth that was reduced
by the presence of Giardia lamblia and Ascaris lumbricoides [21].
Other Single Micronutrients
No other micronutrient has been so well studied in relation to linear growth as

zinc, but large scale trials assessing the importance of other single micronutri-
ents such as vitamin A and iron have often measured growth.
Micronutrients for Stunting and Moderate Malnutrition 13

Clinical vitamin A deficiency is associated with poor growth, but although
a few studies report increased height gain when children with severe deficiency
of vitamin A were given supplements [22], most studies have not reported any
significant effect on linear growth or weight gain [15].
Iron deficiency is common, and the hematological and cognitive effects are
sufficient reason for supplementation. The impact of iron supplementation on
linear growth has been inconsistent. A systematic review by Ramakrishnan et
al. [23] reported a nonsignificant tendency towards a positive effect in children
who were stunted at baseline, and a review by Sachdev et al. [24] reported sig-
nificant effects only in some subpopulations: children older than 5 years, in
malaria hyperendemic areas and those who received the supplement for longer
than 6 months. An update of recent evidence [25] concluded that growth was
neither positively nor negatively affected by iron supplementation.
Calcium is essential for bone formation and often deficient in diets low in
ASFs, especially when dairy products are absent. Trials of calcium supplemen-
tation as a single nutrient have tended to focus on school-aged children rather
than infancy when linear growth faltering is maximal. A review of randomized
trials concluded that calcium supplements had no impact on growth [26].
Many other micronutrients have a role in normal growth and potentially may
contribute to stunting. Severe iodine deficiency for instance causes cretinism, a
form of dwarfism, but since important deficits in cognitive development occur
even in moderate deficiency, this has been the primary outcome of intervention
trials, and linear growth has not been measured.
Other vitamins commonly deficient in cereal-based diets include vitamin
B12, folic acid, thiamine, niacin and vitamin D, and because of probable dietary
deficiencies these are often included in multi-micronutrient supplements, but
their role in preventing stunting has not been established.
Multiple Micronutrients
Given that micronutrient deficiencies rarely occur in isolation, attention has

turned to multimicronutrient supplements as an alternative approach. This
brings challenges because of interactions, product stability and acceptability,
but this approach is very attractive to programs since the logistics and cost of
attempting multiple separate daily supplements is prohibitive.
Multiple micronutrients (MMNs) have been assessed in clinical trials in dif-
ferent parts of the world. A review in 2003 [27] of community-based micronu-
trient supplementation trials concluded that supplements that contained zinc,
vitamin A and iron could have positive effects in deficient populations. A meta-
analysis in 2004 [23] also reported overall positive effects of MMN supplements
on growth in contrast to single nutrient supplements. A recent review by Allen
et al. [28] was able to include a further 13 studies. The analysis included studies
14 Penny
that had compared MMNs against single nutrients, mainly iron. This analysis
concluded that overall length and height were improved with an effect size of
about 0.13 (95% CI: 0.055–0.21). Some studies reported positive effects only in
subgroup analyses; for instance, in Mexico a multinutrient beverage was only
effective in children under 12 months [29], and in Vietnam [30] the subgroup of
children who were stunted at the start of the study showed the largest effect size.
MMNs were delivered in a variety of ways, but the analysis was unable to detect
whether any particular delivery mode was better than another.
In conclusion, MMN supplements given to infants and young children seem
to have a fairly small but positive effect on length or height, and two recent
analyses suggest that this is greater than the effect of single micronutrients.
Given that birthweight is associated with postnatal infant growth, micro-
nutrients given antenatally would be expected to reduce infant malnutrition at
least during the first few months. A Cochrane review [31] reported that prenatal
MMNs were no more effective than iron and folic acid alone in reducing low
birthweight and small for gestational age. However, a recent cluster-randomized
trial of the INIMMAP MMN prenatal supplement in Niger significantly reduced
low birthweight and increased average birthweight but only by 67 g [32]. An
effectiveness trial in Vietnam [33] showed an impact on birthweight and also on
the height of 2-year-old children including 10% less stunting in the communi-
ties where the prenatal supplements had been used. A more recent combined
analysis of original data from 12 randomized trials concluded that compared
with iron-folic acid, MMNs resulted in a small increase in birthweight and a
reduction of low birthweight by about 10% [34]. Information about the infant’s
postnatal growth was not included.
Thus, is seems that MMNs are an effective intervention to prevent or treat
stunting in infants and young children, but more needs to be studied about
means of delivery and the potential impact of prenatal supplementation, par-
ticularly in preventing the growth retardation in the first 6 months that has been
revealed by the WHO growth reference [35].
Moderate Malnutrition
Moderate malnutrition may arise either because a previously ‘normal’ child has
lost weight or failed to gain weight and crosses ‘growth centiles’ or because a
low birthweight baby continues to track along a low ‘centile’ corresponding to
less than 2 SD below the norm. These two scenarios are very different in clinical
terms, but cannot be distinguished by a single measurement, and for this reason
in evaluating individuals longitudinal growth data whenever available should
always be examined. Thus, moderate malnutrition, in contrast to stunting does
not necessarily represent a chronic dietary inadequacy; however, recovery from
moderate malnutrition will always require an increase in body mass implying

increased energy and protein needs. On the other hand, the treatment of stunting
with additional calories, especially in the context of Latin America where stunt-
ing is often associated with above average weight-for-height, may potentially
lead to increased overweight or obesity, and should be carefully controlled.
A recent World Health organization call for action and consultation on the
management of moderate malnutrition in children under 5 years of age has
addressed the issues related to managing moderate malnutrition and provides
an excellent review of all the issues involved in rising to the challenge of treat-
ing these children [2]. The nutrient needs of children have been considered
and a number of nutrients including protein, potassium, phosphorus as well
as energy, essential fatty acid and MMNs are considered important. In terms of
micronutrients, there is emphasis on the need to ensure sufficient intake of all
type 2 micronutrients, those associated with growth. Although single micronu-
trients such as iron [24, 25], vitamin A [22] and zinc have been associated with
weight gain, since children with moderate malnutrition are especially likely to
have multiple nutrient deficiencies, the role of single micronutrients may be less
significant, and MMNs have usually been considered; in fact, the expert consul-
tation went as far to say that ‘approaches putting emphasis on single nutrients
are misguided and should be abandoned’ [2].
There is considerable debate over the amounts of micronutrients that should
be included in multi-micronutrient mixes whether in food or as supplements,
and the expert consultation suggested that the requirements be set between the
daily recommended intakes for eutrophic children and the contents of the F100
diet designed for management of severe malnutrition with the latter providing
a safe upper limit. The same consultation provided detailed background and
thoughtful recommendations not only on the content of foods but also relevant
foods and ingredients including anti-nutrients, and the reader is referred to this
report for further information [2]. In addition, attention is given to the impor-
tant complementary activity of dietary counseling [2].
Delivery of Multiple Micronutrients
The evidence so far suggests that both for the treatment and prevention of stunt-
ing and for moderate malnutrition MMNs are indicated. So what options are
available for delivery of these micronutrients, and what are the issues involved?
Many different formulations have been tried in clinical trials of micronutri-
ents, but there are some emerging leaders in the field. For more than 30 years,
food assistance programs have distributed food aid for the prevention and treat-
ment of malnutrition and for emergencies as fortified food blends [2]. These
are intended to replace or add to local traditional complementary foods. These
mixtures usually contain blends of cereals such as corn, soy beans, sugar and
oil and are distributed by international agencies. In many countries there are
16 Penny
also local products such as Incaparina of Guatemala, and Mi Papilla in Ecuador.
These products can be successful in reducing anemia, but have had less impact
on stunting or moderate malnutrition. Ecuador children attending participating
health facilities who were offered counseling and ‘mi papilla’ had highly signifi-
cant reductions in anemia compared with those attending control facilities who
only received counseling, but differences in height were not significant when
confounding variables were controlled [36]. Studies in India, Mexico and Perú
have shown less encouraging results with no effect on linear growth faltering,
and a review of the PL480 title II foods program included in the moderate mal-
nutrition consultancy [2] illustrated some of the problems of low acceptance
and intrafamily sharing of the product that reduced the nutritional benefit to
the target child.
Two other categories of fortified foods: complementary food supplements
and micronutrient powders are extensively reviewed by the Ten Year Strategy
to Reduce Vitamin and Mineral Deficiencies, Maternal, Infant and Young Child
Nutrition Working Group: formulation subgroup of the moderate malnutrition
consultancy [2]. Complementary food supplements are food based and con-
tain macronutrients and micronutrients designed to complement, rather than
replace, other foods prepared in the family. Micronutrient powders are mix-
tures of vitamins and minerals that are added to traditional foods. They do not
include macronutrients in significant dietary amounts.
Complementary Food Supplements
These supplements include the highly nutrient-dense spreads described by

Briend [37] also known as ready to use therapeutic food, and now being suc-
cessfully used for the treatment of severe and moderate malnutrition [2].
These lipid nutrient supplements (LNS) are based on high-fat products such
as peanut butter or soy beans with added micronutrients with or without
milk. They have the benefit of being highly palatable with good acceptabil-
ity. They provide an excellent medium for micronutrients, and essential fatty
acids can be added, and because of the low water content, bacterial growth
is inhibited. A study in Malawi confirmed better weight gain in moderately
malnourished children compared with a cereal product [38], but there was
no difference in linear growth. These products are now being developed for
use as a food complement in smaller doses and with micronutrients adjusted
for treatment of stunting and moderate malnutrition. In addition, a trial of
multi-micronutrients delivered as micronutrient powders, LNS and crushable
tablets in Ghana showed that while all three supplements reduced anemia and
resulted in improved motor function, only the LNS improved linear growth
[39]. Cost is a concern with LNS, but it is hoped that local production will
address this problem.

Micronutrient Powders
Micronutrient powders presented in small single-use sachets to add to a serv-

ing of complementary food have proved successful in preventing and treating
anemia [40, 41]. The potential for adding other micronutrients such as zinc has
been explored, but as yet no formula has been shown to prevent stunting or
promote linear growth.
Improving the Infant and Young Child’s Diet
Another alternative to improve stunting and moderate malnutrition is to

improve the child’s normal diet by increasing the micronutrient content and
availability. Two approaches have been tried: reducing inhibitors to absorption
to enhance bioavailability and increasing consumption of ASFs.
Complementary foods can be processed, for instance by fermentation to
reduce phytate content. In Tanzania, processed complementary food reduced
phytate but did not improve growth compared with usual household prepara-
tions [42]. Another attempt to use maize that had been selected for low phytate
content to improve growth was not successful in infants in rural Guatemala [43].
The reduction of inhibitors to mineral absorption may be seen as a complemen-
tary strategy to improve complementary foods, but has not yet been shown to
enhance growth.
Enhanced nutrition education through the health services in a population
where low-cost ASFs are available, resulted in higher intakes of ASFs associ-
ated with prevention of stunting [44]. Intake of ASFs was associated with better
height gain in Kenyan school children, and weight gain and increased lean body
mass was associated with increased meat intake, but only milk intake resulted
in height gain [45].
Milk deserves special mention because of its potential role as a vehicle for
fortification. Fortified milk given to Indian children resulted in increased lin-
ear growth [46] compared with unfortified milk. Milk is of particular interest
because it stimulates IGF-I secretion which stimulates growth. Further research
is needed to explore milk, fortified or not in programmatic conditions.
Discussion and Future Directions
The evidence suggests that MMNs can reverse to some extent the impact of
linear growth retardation in infants and young children whose dietary intake
is insufficient. MMN combinations may be more effective than single nutri-
ents and have logistic advantages. New strategies are encouraging, and many
programs and countries are now considering how best to implement MMN
18 Penny
supplementation at scale. The combination of micronutrients makes it pos-
sible to tailor interventions to the particular needs of different populations
addressing the multiple effects of inadequate dietary intakes. We need to take
a holistic approach that also includes the health hazards of both deficiency
and excess, especially where increasing rates of childhood obesity coexist with
stunting and anemia.
References
1 Victora CF, Adair L, Fall C, et al: Maternal 9 Kimmons JE, Dewey KG, Haque E, et al:
and Child Undernutrition Study Group. Low nutrient intakes among infants in rural
Maternal and child malnutrition: conse- Bangladesh are attributable to low intake and
quences for adult health and human capital. micronutrient density of complementary
Lancet 2008;371:340–357. foods. J Nutr 2005;135:444–451.
2 World Health Organization: Joint WHO/ 10 Marquis GS, Habicht JP, Lanata CF, et
UNICEF/WFP/UNHCR consultation on the al: Breastmilk or animal product foods
dietary management of moderate malnutri- improve linear growth of Peruvian toddlers
tion. Food Nutr Bull 2009;30(suppl):S1–S474. consuming marginal diets. Am J Clin Nutr
3 Black RE, Allen LH, Bhutta ZA, et al: Global 1997;66:1102–1109.
and regional exposures and health conse- 11 Allen LH, Backstrand JR, Stanek EJ, et
quences. Lancet 2008;371:243–260. al: The interactive effect of dietary qual-
4 WHO Multicentre Growth Reference Study ity on the growth and attained size of
Group: WHO growth standards based on young Mexican children. Am J Clin Nutr
length/height, weight and age. Acta Paediatr 1992;56:353–364.
Suppl 2006;450:76–85. 12 Prasad AS: Clinical manifestations of zinc
5 Dewey K, Huffman SL: Maternal, Infant and deficiency. Ann Rev Nutr 1985;5:341–363.
young child nutrition: combining efforts 13 Brown KH, Wuehler SE, Peerson JM: The
to maximize impacts on child growth and importance of zinc in human nutrition and
micronutrient status. Food Nutr Bull 2009; estimation of the global prevalence of zinc
30:S187–S189. deficiency. Food Nutr Bull 2001;22:113–125.
6 Campos R, Hernandez L, Soto-Mendez MJ, 14 Brown KH, Peerson JM, Baker SK, et al:
et al: Contribution of complementary food Preventive zinc supplementation among
nutrients to estimated total food nutrient infants, preschoolers, and older prepubertal
intake in Guatemalan infants in the second children. Food Nutr Bull 2009;30(suppl
semester of life: Asia Pac J Clin Nutr 2010; 1):S12–S40.
19:481–490. 15 Ramakrishnan U, Nguyen P, Martorell R:
7 Lander R, Enkhjragal TS, Batjargal J, et Effects of micronutrients on growth of chil-
al: Poor dietary quality of complementary dren under 5y of age: meta-analysis of single
food is associated with multiple micronu- and multiple nutrient interventions. Am J
trient deficiencies during early childhood Clin Nutr 2009;89:191–203.
in Mongolia. Public Health Nutr 2009; 16 Imdad A, Bhutta ZA: Effect of preventive
13:1303–1313. zinc supplementation on linear growth in
8 Allen LH: How common is vita- children under 5 years of age in developing
min B12 deficiency? Am J Clin Nutr countries: a meta-analysis of studies for input
2009;89(suppl):693S–696S. to the Lives Saved tool. BMC Public Health
2011;11(suppl 3):S22.

17 Merialdi M, Caulfield LE, Zavaleta N, et al: 28 Allen LH, Peerson JM, Olney DK: Provision
Randomized controlled trial of prenatal zinc of multiple rather than two or fewer micro-
supplementation and fetal bone growth. Am nutrients more effectively improves growth
J Clin Nutr 2004;79:826–830. and other outcomes in micro-nutrient
18 Frisancho AR: Relative leg length as a biolog- deficient children and adults. J Nutr 2009;
ical marker to trace the developmental his- 139:1022–1030.
tory of individuals and populations: growth 29 Rivera JA, Gonzalez-Cossio T, Flores M, et
delay and increased body fat. Am J Hum Biol al: Multiple micronutrient supplementation
2007;19:703–710. increases the growth of Mexican infants. Am
19 Stewart CP, Christian P, LeClerk SC, et al: J Clin Nutr 2001;74:657–663.
Antenatal supplementation with folic acid 30 Thu BD, Schultink W, Dillon D, et al: Effect
+ iron + zinc improves linear growth and of daily and weekly micronutrient supple-
reduces peripheral adiposity in school-aged mentation on micronutrient deficiencies and
children in rural Nepal. Am J Clin Nutr growth in young Vietnamese children. Am J
2009;90:132–140. Clin Nutr 1999;69:80–86.
20 Osendarp SJM, van Raaij JMA, Darmstadt 31 Haider BA, Bhutta Z: Multiple micronutrient
GL, et al: Zinc supplementation during supplementation for women during preg-
pregnancy and effects on growth and mor- nancy (review). Cochrane Data Base Review
bidity in low birthweight infants: a random- 2006;CD004905.
ized placebo controlled trial. Lancet 2001; 32 Zagré NM, Desplats G, Adou P, et al:
357:1080–1085. Prenatal multiple micronutrient supplemen-
21 Rosado JL, Caamaño MC, Montoya YA, et tation has greater impact on birthweight than
al: Interaction of zinc or Vitamin A supple- supplementation with iron and folic acid: a
mentation and specific parasite infections on cluster-randomized, double-blind, controlled
Mexican infants’ growth: a randomized clini- programmatic study in rural Niger. Food
cal trial. Eur J Clin Nutr 2009;63:1176–1184. Nutr Bull 2007;28:317–330.
22 West KP, LeClerq SC, Shrestha SR, et al: 33 Huy ND, Hop LT, Shrimpton R, et al: An
Effects of vitamin A on growth of vitamin effectiveness trial of multimicronutrient sup-
A-deficient children: field studies in Nepal. J plementation during pregnancy in Vietnam:
Nutr 1997;127:157–165. impact on birthweight and on stunting in
23 Ramakrishnan U, Aburto N, McCabe G, et children around 2 years of age. Food Nutr
al: Multimicronutrient interventions but Bull 2009;30:S506–S516.
not vitamin A or iron interventions alone 34 Fall CHD, Fisher DJ, Osmond C, et al:
improve child growth: results of three meta- Maternal micronutrient supplementation
analyses. J Nutr 2004;134:2592–2602. study group (MMSSG). Multiple micronutri-
24 Sachdev H, Gera T, Nestel P: Effect of iron ent supplementation during pregnancy in
supplementation on physical growth: system- low-income countries: a meta-analysis of
atic review of randomized controlled trials. effects on birth size and length of gestation.
Public Health Nutr 2006;9:904–920. Food Nutr Bull 2009;30:S533–S546.
25 Stoltzfus RJ, Heidcamp R, Kenkel D, et al: 35 Victora CG, de Onis M, Hallal PC, et al:
Iron supplementation in young children: Worldwide timing of growth faltering:
learning from the new evidence. Food Nutr revisiting implications for interventions.
Bull 2007;28:S572–S584. Pediatrics 2010;125:e473–e480.
26 Winzenberg T, Shaw K, Fryer J, et al: 36 Lutter CK, Rodriguez A, Fuenmayor G, et al:
Calcium supplements in healthy children do Growth and micronutrient status in children
not affect weight gain, height, or body com- receiving a fortified complementary food. J
position. Obesity 2007;15:1789–1798. Nutr 2008;138:379–388.
27 Rivera JA, Hotz C, Gonzalez-Cossio T, et 37 Briend A: Highly nutrient dense spreads a
al: The effect of micronutrient deficiencies new approach to delivering multiple micro-
on child growth: a review of results from nutrients to high risk groups. Br J Nutr 2001;
community-based supplementation trials. J 85(suppl 2):S175–S179.
Nutr 2003;133:S4010–S4020.
20 Penny
38 Thakwalakwa C, Ashorn P, Phuka J, et al: 43 Mazariegos M, Hambidge KM, Westcott JE,
A lipid based nutrient supplement but not et al: Neither a zinc supplement nor phytate-
corn-soy blend modestly increased weight reduced maize nor their combination enhance
gain among 6–18 month old moderately growth of 6- to 12-month-old Guatemalan
underweight children in rural Malawi. J Nutr infants. J Nutr 2010;140:1041–1048.
2010;140:2008–2013. 44 Penny ME, Creed-Kanashiro HM, Robert
39 Adu-Afarwuah S, Lartey A, Brown KH, et RC, et al: Effectiveness of an educational
al: Randomized comparison of 3 types on intervention delivered through the health
micronutrient supplement for home forti- services to improve nutrition in young chil-
fication of complementary foods in Ghana: dren: a cluster-randomised controlled trial.
effect on growth and motor development. Lancet 2005;365:1863–1872.
AJCN 2007;86:412–420. 45 Neumann CG, Murphy SP, Gewa C, et al:
40 Zlotkin SH, Schauer C, Cristofides A, et al: Meat supplementation improves growth,
Micronutrient sprinkles to control childhood cognitive, and behavioural outcomes in
anemia. PLos Med 2005;2:e10. Kenyan children. J Nutr 2007;137:1119–
41 Menon P, Ruel MT, Loechl CLL, et al: 1123.
Micronutrient sprinkles reduce anemia 46 Sazawal S, Dhingra U, Dhingra P, et al:
among 9–24-mo-old children when deliv- Micronutrient fortified milk improves iron
ered through an integrated health and status, anemia and growth among children
nutrition program in rural Haiti. J Nutr 1–4 years: a double masked, randomized
2007;137:1023–1030. controlled trial. Plos One 2010;5:e12167.
42 Mamiro PS, Kolsteren PW, van Camp JH,
et al: Processed complementary food does
not improve growth or hemoglobin status of
rural Tanzanian infants from 6–12 months of
age in Kilosa district, Tanzania. J Nutr 2004;
134:1084–1090.

Discussion on Micronutrient
Requirements
Setting the Terms of Discussion for the Workshop

The 70th Workshop of the Nestlé Nutrition Institute was entitled ‘Micronutrient
Requirements for Health and Development’. Two distinguished scientist with
experience in developing countries with important micronutrient deficiencies:
Dr. Zulfiqar Bhutta and Dr. Mary Penny took part in this workshop. The inter-
action of nutrition, growth and child survival are prominent public health issues
in both the South Asian subcontinent and the Andean region, providing ample
opportunity for the two speakers in the inaugural session to gain experience and
synthesize their reflections.
One of the important terms of reference in the Workshop title is that of ‘micro-
nutrient requirements’. For the introductory talks, and throughout the Workshop,
however, randomized trials with single or multiple micronutrients formed the
evidence for a discussion of a requirements theme. Most often, levels of nutrients
well above the ‘requirement’ (daily recommended intake) amounts are delivered
in such experimental settings. The variance in health and developmental out-
comes that can derive from improving the micronutrient offering from dietary
interventions with high-quality foods and beverages was only rarely alluded to.
Such a perspective would have been more consonant with the explicit terms of
reference in the title of the Workshop. To her credit, Dr. Penny was one of the
speakers who used the prism of dietary variation, although the title and message
of her presentation dealt with treatment – of stunting and of severe malnutrition.
Another term of reference is health. Impaired health is associated with altered
nutrient requirements, usually a selective increase in the needs for macro- and
micronutrients. Dr. Penny’s discussion of prescription of nutrients to assist what is
called in the title as ‘moderate malnutrition’ (underweight), derives from her years
of experience in defining the safe and effective rescue, recovery and rehabilitation
of children with clinical syndromes of undernutrition. An opportunity was missed
in this session, and the subsequent program, to examine the term ‘health’ in terms
of nutrient requirements. What are established as reference intakes by bodies
such as the UN Systems Food and Agricultural Organization and World Health
Organization or by the Food and Nutrition Board of the Institute of Medicine in
the USA are explicitly applicable to ‘healthy populations’. Children in low-income
settings are more often than not infested by protozoan and helminthic parasites
and beset by recurrent communicable diseases affecting the respiratory and gas-
trointestinal tracts. These alter intermediary metabolism and can interfere with
nutrient intake and provoke excessive nutrient wastage. Are the environmental
and epidemiological correlates of living in disadvantaged circumstance a motive
for adjusting the recommendations to compensate for the ambient stress?
The nutrients most associated with growth and survival, as mentioned in
this session, were vitamin A, iron, zinc and iodine. Severe deficiency of all four
nutrients is, indeed, associated with decreased survival, and even marginal
vitamin A deficiency increases mortality from common childhood infections.
Where both Dr. Bhutta and Dr. Penny produced a potential disconnect across
the title topic was in the arena of what degree of recovery of linear growth we
should expect from nutrient exposures across the full panorama of interven-
tion possibilities. Almost any nutrient at severe enough restriction will cause
livestock, poultry or laboratory rodents to falter in their growth; this is the basis
for the classical demonstration of specific food factors as essential nutrients.
Profound effects on linear growth in free-living populations can only be related
to zinc and iodine, with vitamin A and iron producing limited growth faltering.
In places like Afghanistan, Angola, and the Horn of Africa, and in my region
of Guatemala, the accumulated deficit in height at 5 years of age reaches 10 cm
or more. However, when it comes to preservation in linear growth, effect sizes
of only a few cm are the maximal outcomes even in the most successful single
and multiple micronutrient interventions on record. It may not be the dosage or
combination of micronutrients that limit this effect, but rather the intrinsic bio-
logical nature of short stature. Evidence can be mobilized from livestock man-
agement and experimental science that inflammation and the stress of living in
an unsanitary environment are important determinants of growth faltering, and
will not be overcome by any amount of additional nutrient exposures.
Salient Points of the Session’s Presentations

The titular theme of Dr. Bhutta’s presentation involved ‘global micronutrient
deficiencies’; using maps, he illustrated the estimated prevalence of vitamin A,
zinc and iodine deficiency. Not surprisingly, the concentration of micronutri-
ent deficiency problems aligned with the poorest nations of Latin America,
Africa and Asia. He accompanied this geography lesson with a narrative on the
impacts on health and development of the deficiency of these micronutrients in
early life, pointing out that vitamin A and zinc deficiency is associated with an
estimated 1 million child deaths and 9% of global childhood disability-adjusted
life years. In this respect, he harkened to the 50-year-old paradigm of the inter-
action of nutrition and infection introduced by Nevin Scrimshaw, noting that
child health is an outcome as well as a risk factor.
Discussion on Micronutrient Requirements 23

He reinforced the relevance of the period from conception to 24 months of
life, the so-called 1,000 days’ window of opportunity, involving maternal health
in pregnancy and the adequacy of complementary feeding. He also mentioned
that other nutrients beyond the primary three, namely iron, was also associated
with limitations of human development, especially with an onset in infancy. The
opportunity to mention vitamin D deficiency as an impending public health
problem was missed both in this session, and throughout the Workshop.
On repeated opportunities in the presentation, the pediatrician from Pakistan
cited the challenge as being ‘to implement intervention strategies that combine
appropriate infant and young child feeding with micronutrient interventions at
scale’. There is a sense of detachment or dissociation of the appropriate feeding
and the interventions. One wonders if the concentration on what are appropri-
ate estimates for required intakes of nutrients from young-child diets in low-
income societies would not obviate the need for large-scale interventions with
micronutrient supplements.
The treatment of moderate undernutrition involves correction of inappro-
priate weight deficit. Although energy and protein are the primary nutritional
considerations when repleting body tissue mass, the British pediatrician in
Peru, Dr. Penny, pointed out how micronutrients could not be ignored in this
therapy. The broad consideration of micronutrients, beyond those of primary
public health interest, is exemplified in the comprehensive therapeutic approach
to moderate malnutrition. In fact, this led to a discussion of multiple micronu-
trients administered in combination. Surprisingly, again in light of the title of
the Workshop, if we take the whole program and its 14 invited presentations
into perspective, only Dr. Penny’s piece in the Introduction, the two presenta-
tions in the pregnancy session (Dr. Darnton-Hill; Dr. Bhutta), and the fortified
food topic of Dr. Allen put the emphasis on multiple micronutrients in combi-
nation. The other 11 are monotonic for single nutrients.
The treatment of stunting or short stature is also addressed by Dr. Penny,
and she invokes both single-nutrient and multi-nutrient formats. To her credit,
she provides a balanced review of the literature, and shows the mixed – and
not overwhelming – evidence for efficacy of reversing linear growth faltering
with single or combined nutrients in stunted populations. The most interesting
insight into the biology of stunting is that the length of the head, neck and torso
of stunted children is equivalent to that of a normally grown peer. The growth
faltering is largely the result from the failure of elongation of the long bones of
the legs. This could point to some disruption of the hormonal cascade that sup-
ports bone growth at the epiphyses.
Summary of the Question and Answer Participation

In the question and answer discussion period, Dr. Bhutta was challenged on sev-
eral points. Since both issues of bodyweight (underweight) and diverse micro-
nutrient deficiencies seem to predict and predispose to under-five mortality,
24 Solomons
how does one parse and separate out the specific impact of the different defi-
ciency conditions, and how does one translate that into effective programmatic
actions. The discussion moved to the inconsistencies related to the impact of
zinc interventions in achieving improvements in survival, disease resistance
and other functional benefits. One factor is the (as yet imperfect) approaches to
assessing the existence of zinc deficiency due to lack of a specific biomarker of
deficiency. Regarding the query of why vitamin D deficiency has yet to be incor-
porated into the global micronutrient deficiency agenda, it was admitted that
the so-called ‘sunshine vitamin’ can be deficient in populations of the tropics.
Insofar as this realization is of recent origins, we are only at the incipient level of
inquiry into vitamin D deficiency, beyond overt rickets, in the global statistics of
mortality, morbidity and intrauterine growth failure.
Dr. Penny extended her discussion of animal protein food in meeting require-
ments from dietary sources in response to questions as to whether milk was also
in this category. Two characteristics of milk that could produce positive trophic
effects in children were identified. The first is that bovine milk has been found
to contain insulin-like growth factor-I, which may be absorbed intact and func-
tion as a hormone in the milk consumer. Another thought was that of surrepti-
tious and serendipitous correction of an unrecognized iodine deficit by milk.
Due to the iodine-containing antiseptic solutions used to disinfect the cows’
udders prior to industrial milking, residual iodine remains in the dairy items
derived. By repleting iodine status, milk consumption might positively impact
those short children with prior iodine deficiency. Rounding out the discus-
sion on nutrients from actual foods was a focus on the limitations to obtaining
adequate micronutrient nutrition during the critical 6–12 months of life cov-
ering transition from exclusive breastfeeding to mixed feeding. In this period,
the critical quest is getting complementary foods with a high enough nutrient
density to provide the residual needs of vitamins and minerals with a low allot-
ment for additional calories. This is where animal protein foods are indispens-
able because of their generally more nutrient dense composition.
The remainder of the discussion centered on deriving useful insight on the
Workshop topic from published trials. The question of how the various study
designs included together in a review of intervention studies are casting light
on micronutrients and growth was raised. Obvious issues such as the nutrient
dosages, the mixtures of nutrients, the duration of the trial, the age and status of
the children, and the degree of preexisting stunting in the trial populations are
all features that could influence the findings of benefits. The issues of nutrient–
nutrient interactions in multiple nutrient interventions were raised within this
context. That interactions might be responsible for the variability of results from
trial to trial was raised. Most of the interest centered on the competitive interac-
tion of iron and zinc. The possibility of certain interactions having the effect of
accentuating the actions of the nutrients was also raised as a theoretical possibil-
ity when multinutrient combinations are at play.
Discussion on Micronutrient Requirements 25

The status of parasitosis with protozoan and helminthic pathogens in the
various intervention trials was advanced as another issue of potential relevance
to the interpretation of the findings. Parasite loads vary from setting to setting
depending upon residual infections in the population and ecological factors sup-
porting parasite transmission and propagation. One commentator opinion was
that multicellular pathogens are much more common as agents of severe gas-
troenteritis episodes and malabsorption than has been previously recognized.
Finally, an interesting suggestion on transgenerational effects, i.e. that attention
to micronutrients in the mother will have beneficial effects for her daughters
and her granddaughters, was raised in the discussion.
Conclusion of the Discussant

The issue of micronutrient requirements from the connotation of diet has
become divorced from the thrust of the field research on nutrient supplemen-
tation conducted over past quarter of a century. These years have allowed for
the proliferation and collation of well-conducted and statistically powerful field
trials, now coalesced into systematic reviews and meta-analyses. By their very
design and nature, they allow for pharmacological actions of high-dose nutri-
ents to be misinterpreted as nutritional functions of the nutrients and produce
imbalances that can distort the normal metabolism of nutrients taken in from
the diet itself. For all of their citation by the lead-off speakers and by the majority
of presentations to follow, their capacity to inform us regarding ‘micronutrient
requirements’ should be viewed with healthy skepticism. Aside from the ques-
tion of the long-term sustainability of intervention programs as public policy,
their limitations in eradicating the survival, growth, health and developmen-
tal issues may derive from the simple reality that other, nutrient-independent
environmental assaults are influencing or even determining the outcomes. An
alternative view might be to let the best quality diet do what diet can do, and
address the collateral causes of poor health and development with a comprehen-
sive array of complementary actions directed at the residual ecological origins.
Noel W. Solomons
26 Solomons
Zinc Requirements: Assessment and

Population Needs
K. Michael Hambidge ⭈ Leland V. Miller ⭈ Nancy F. Krebs
University of Colorado Denver, Aurora, CO, USA
Abstract
Reliable estimates of zinc requirements have assumed greater priority as the global pub-
lic health importance of preventing zinc deficiency has gained increasing recognition. On
a global public health basis, our first most evident goal is reliable estimates of average
population requirements. Despite expectations of rapid advances towards simpler and
more sophisticated strategies, estimations of zinc requirements continue to depend on a
factorial approach. Since the Dietary Reference Intakes (DRIs) were published, there have
been important advances in techniques for the factorial approach but also confusion
resulting from the subsequent publication of conflicting ‘international’ estimates. The rea-
sons for these differences have now been fully elucidated, removing an obstruction to
continuing progress and refinements of our knowledge base. A key advance has been the
development and validation of a model that can be simply applied to determine the
inhibitory effects of phytate on zinc absorption. Better understanding of maternal and
young child zinc requirements continues to present a challenge of special importance.
Introduction
Zinc deficiency now has an established place among the small group of micro-
nutrient deficiencies having acknowledged public health importance of global
dimensions. This recognition has underscored the need for reliable, broadly
accepted estimates of dietary needs for all population groups consuming a
broad variety of diets. Though there have been expectations for some years of
new successful approaches for the assessment of zinc requirements, perhaps
especially, zinc status-related changes in one or more of the zinc absorption
transporter proteins or corresponding mRNA, these have yet to be validated.
A factorial approach, therefore, remains not only the strategy of choice but the
Ingested Zn
Slowly
exchanging
pools Intestine Ingested
phytate
New Rapidly Absorption

tissue exchanging
pools
Secretion
Reabsorption
Milk Zn
Integumental Semen/ Urine Endogenous Non-absorbed

Zn menses Zn Fecal Zn
Zn
Fig. 1. Schematic of total body zinc homeostasis.
only available method. Fortunately, though data remain limited, new concepts
and effective modeling over the past decade have resulted in quite fundamental
advances in the effectiveness of the factorial approach. The specifics and impli-
cations of these advances as well as the resolution of conflicting estimates are the
focus of this review. Though a principal focus of this workshop is on develop-
ment, this paper will be directed to adult requirements. Reliable estimates for all
women of child-bearing age are important for the in utero development of their
offspring. Estimates of zinc requirements for young children depend primarily
on extrapolation from adult estimates. Therefore, reliable estimates of adult zinc
requirements provide an essential cornerstone for understanding requirements
during all stages of development. These are also an essential starting point for
our understanding of the effects of inflammation and chronic/recurrent disease
on zinc requirements.
The factorial estimation of average zinc requirements of populations
depends on reliable data that can best be considered under three major head-
ings. These are: (1) physiological requirements (PR) defined as the quantity of
zinc absorbed daily (TAZ) that is just adequate to match inevitable excretion
of endogenous zinc by all routes together with any zinc retention required for
new tissue (fig. 1); (2) the estimated average daily zinc requirement (EAR) of
bioavailable zinc necessary to achieve absorption of the PR, and (3) the bioavail-
ability of the ingested zinc which is determined at least primarily by the quantity
of dietary phytate ingested. This paper will review each of these aspects includ-
ing the resolution of recent disparities in estimates of zinc requirements. The
accurate determination of zinc absorption requires measurement over a period
28 Hambidge · Miller · Krebs

of a minimum of one day rather than single meal studies, and all data covered in
this review are expressed in terms of mg Zn/day.
Physiological Requirements
Typically, except during early lactation, the intestine is the major route of
excretion. The quantity of zinc excreted via this route is regulated in response
to current zinc absorption and to an undefined but probably limited extent
by ‘zinc status’. Other routes of excretion of endogenous zinc are via the kid-
neys (also regulated but with relatively minor implications for requirements),
reproductive system, integument, and, during lactation, the mammary gland.
The demands of new tissue add to the PR during childhood and pregnancy,
though these may be offset during pregnancy, as may the extra excretion of
zinc via the mammary gland, by up-regulation of absorption [1, 2] and, pos-
sibly, by down-regulation of endogenous zinc excreted via the intestine [1].
A limitation in concept of human zinc homeostasis in one case and errors in
determining the relationship between endogenous fecal zinc (EFZ) and TAZ
in another resulted in mistaken estimates of PR in two of three relatively recent
publications considered in this review, each of which has had and continues
to have an impact globally. In the first of these, in 1996 [3], it was concluded,
during this initial (and notable) endeavor to provide a strong evidence base
for zinc requirements, that EFZ and, to a small extent, urine varied with adap-
tation to recent intake or/and ‘status’. This led to the concept of ‘basal’ (fully
adapted) and ‘normative’ (non-adapted) PR. However, in both cases these lev-
els were based on low zinc intakes and were regarded as ‘static’ numbers which
did not increase with increasing absorption of zinc. This resulted in extremely
low and erroneous estimates of PR [4] in what was otherwise an outstanding
contribution to the evolution of emphasis on an evidence base for assessing
zinc requirements.
A key advance was contributed by the Institute of Medicine (IOM) in 2001
[5] with recognition of the occurrence and significance of a strong positive cor-
relation for EFZ versus the quantity of TAZ which, importantly, was apparent at
levels of TAZ well below the PR [6]. The implication is that regulation of EFZ
is imperfect, as zinc excretion increases with increasing TAZ when the latter is
still below that needed to meet physiologic requirements. This concept and sup-
porting data resulted in estimates of PR for adult males [5] that were more than
threefold those estimated by the WHO [3]. The validity of this approach has not
been questioned, though relatively minor numerical adjustments are likely as
the current limited data base is enlarged, especially with studies of zinc homeo-
stasis while ingesting habitual diets.
Publication of the IOM estimates [5] was followed 3 years later by ‘inter-
national’ estimates [7]. It should be noted that these ‘international’ estimates
Zinc Requirements: Assessment and Population Needs 29

Women
IZiNCG IOM
1.9 3.2
1.5 2.0 2.5 3.0 3.5

Zn (mg/day)
Men
IZiNCG IOM
2.7 3.9
2.5 3.0 3.5 4.0

Zn (mg/day)
Fig. 2. Discrepancies between adult PR for zinc estimated by the IOM in 2001 [5] and
IZiNCG in 2004 [7].
were based on studies of human zinc homeostasis in the USA or UK with

one study of ours in China [8] being the only exception. Two errors in these
estimates, both related to the determination of the relationship between EFZ
and TAZ, were primarily responsible for major underestimates of PR [9].
Because these estimates have been widely disseminated and utilized globally,
they have been the cause of much confusion and have inhibited progress in
our quantitative understanding of zinc homeostasis and requirements for sev-
eral years. Correction of these errors together with using the same reference
data for adult size and correction of an IOM overestimate of menstrual zinc
losses results in close reconciliation [9]. The very small residual differences
(fig. 2–4) are attributable to the use of different datasets, notably with IOM
utilizing male data only to define the relationship between EFZ and TAZ [5]
and International Zinc Nutrition Consultative Group (IZiNCG) [7] using
data from both genders. The closeness of the reconciliation is remarkable in
view of the limited database, the interpretation of which remains complex for
females [10] and the range of laboratory techniques and strategies that have
been used.

Women
IZiNCG IOM
1.9 3.2
2.7 3.0
Data
Menses
correction
1.5 2.0 2.5 3.0 3.5

Zn (mg/day)
Men
IZiNCG IOM
2.7 3.9
3.4
Data
correction
2.5 3.0 3.5 4.0

Zn (mg/day)
Fig. 3. Residual discrepancies in estimates of PR for zinc after correction of identified

errors in estimates.
Estimated Average Zinc Requirements for Populations
The EAR is determined from the intercept between the PR and the model
describing the relationship between TAZ and ingestion of bioavailable zinc.
The conceptual advance that has occurred subsequent to the publication of the
DRIs by the IOM is that this relationship is best fit by a saturation response
model [4]. This is not surprising as the absorption of zinc by the enterocyte has
long been known to be a saturable process in mammals [11]. Recognition of the
appropriateness of this model has not only provided the best fit for data, but has
proved a major visual advance in our understanding of human zinc homeostasis
including advancing our understanding of the relationship between the quan-
tity of bioavailable zinc ingested and the efficiency of zinc absorption. As with
the regulation of EFZ, it is apparent that the regulation of zinc absorption is
not a completely efficient phenomenon [4] (fig. 5). If it were, the model would
coincide with the line of equality between dietary zinc and TAZ until the PR is
met and then remain horizontal. The gap between the line of equality and the
saturation response model is consistent with the abundant evidence that human

Women
IZiNCG IOM
1.9 3.2
2.7 2.9 3.0
Body weight
Data
Menses
correction
1.5 2.0 2.5 3.0 3.5

Zn (mg/day)
Men
IZiNCG IOM
2.7 3.9
3.4 3.7
Data
correction Body
weight
2.5 3.0 3.5 4.0

Zn (mg/day)
Fig. 4. Effect of selection of different adult weight reference standards in estimating

excretion of endogenous zinc via the integument. Minor residual differences between
estimates attributable to use of different datasets.
zinc deficiency, from the very mild to severe, does indeed occur and is no sur-
prise. As up- and down-regulation of proteins involved with zinc absorption
via the enterocyte occurs within a short time span, it is concluded that the less
than perfect efficiency of absorption at levels of ingestion of bioavailable zinc
less than EAR occurs despite maximal up-regulation of transporters and other
proteins involved with zinc absorption. Despite this limitation on adaptation,
fractional absorption is quite favorable at levels of ingestion less than EAR, aver-
aging approximately 40%. It is noted that the estimated average requirement
approximates half maximal absorption of zinc, below which absorption is rela-
tively favorable. Also of note is the low and increasingly poor fractional absorp-
tion of zinc when ingestion of bioavailable zinc exceeds the estimated average
requirement.
While search continues for the ‘holy grail’ of a sensitive, reliable biomarker
of zinc status, we should not underrate, especially at the population level, how
much we can learn about status from reliable determinations of dietary zinc
intakes and, from there, the percentage of the population with zinc intakes

6
A
Absorbed Zn (mg/day) 5 B
4 C
0
0 2 4 6 8 10 12 14 16 18
Ingested Zn (mg/day)
Fig. 5. Estimates of daily requirements for ingested zinc for adult males. A: line of equal-
ity between ingested zinc and absorbed zinc; B: saturation response curve fitted to adult
male data utilized by IOM [5]; C: predicted effect of 1 mmol of daily dietary phytate.
Vertical lines from intercept of estimated PR and saturation response curves indicate esti-
mated average requirements for ingested zinc. Adapted from Hambidge et al. [17].
below the EAR [12]. The only substantial additional information required is the
bioavailability of the ingested zinc, a topic to be addressed in the next section.
For the individual, biomarkers are certainly needed as is adequate information
to provide a reliable measure of the variance around the EAR.
Bioavailability
Phytate is the one dietary factor recognized to have a major effect on the absorp-
tion of ingested zinc apart from the quantity of ingested zinc itself. The inhibi-
tory effect of phytate on zinc bioavailability is notable and of most profound
importance in populations largely dependent on plant products, especially
unrefined cereal grains and beans. In developing the DRIs for zinc, the IOM
concluded that, at that time, there were insufficient data to reliably estimate the
quantitative effect of phytate on daily absorption of zinc. Subsequent trivari-
ate modeling of TAZ as a function of both daily zinc and daily phytate intake
appears to have quite successfully addressed the estimation of the quantitative
effect of different quantities of dietary phytate on zinc absorption [13] over the
spectrum of intake of dietary zinc. Though minor modifications of this model
remain in progress as the effects of other dietary components are examined,
the quality of both the parameter estimates and fit of the model to the data are
reassuring. Moreover, it has been validated independently [14]. Based on this

model, the effect of realistic quantities of dietary phytate on estimated require-
ments of ingested zinc necessary to meet physiological zinc requirements in
adults is illustrated in figure 5. This model has been used, for example, to pre-
dict the level of enhanced zinc absorption achieved by zinc biofortification of
wheat [15]. It is apparent that existing and future strategies to reduce the phytate
content of meals prior to consumption, without equivalent reductions in zinc,
can have a profound effect on zinc requirements and the prevalence of zinc defi-
ciency in populations dependent on high-phytate diets.
Infants and Children
Data on zinc homeostasis in infants and young children, the most vulnerable to
the effects of zinc deficiency, are limited. It is known that the same saturation
response modeling as for adults serves to fit existing experimental data well [16].
In line with lower physiological zinc requirements, the models have lower levels of
absorbed zinc and lower estimates of maximal zinc absorption [16, 17]. It has been
noted that if these models, even for premature infants, are adjusted for differences
in length of small intestine, they fit closely with the adult model [16]. However,
direct measurements of the inhibitory effect of phytate in young children remain
pending at this time, and measurements of endogenous zinc losses are limited.
Conclusions
At least 15 years ago, there was a perception that stable isotopic investigations
were outdated, almost before their application had begun and long before we
knew how to use them optimally in estimating zinc requirements. There was a
palpable feeling that molecular techniques, in particular, would sweep all before
them. In fact, we need these ‘out-of-date’ techniques more than ever and depend
on them to estimate zinc requirements with reasonable precision. However, the
necessary data are still remarkably limited, and the variance remains uncertain.
Moreover, existing data continue to be evaluated, and it is apparent that addi-
tional modifications of current understanding of human zinc homeostasis and
requirements will follow [Miller et al., unpubl.]. This underlines the importance
of building onto an existing base that is as reliable as current data permit.
The greatest challenge is presented by those for whom we need it most. At
this time, we still depend to a very large extent on extrapolation from adults
to estimate zinc requirements in young children. Maternal estimates are com-
plicated by changes in zinc homeostasis that include regulation of absorption
and probably of intestinal excretion of endogenous zinc and, during the entire
course of pregnancy, by the interrelationships between the maternal, placental
and fetal regulation of zinc homeostasis.

References
1 Sian L, Krebs NF, Westcott JE, et al: Zinc 9 Hambidge KM, Miller LV, Krebs NF:
homeostasis during lactation in a population Physiological requirements for zinc. Int J Vit
with a low zinc intake. Am J Clin Nutr 2002; Nutr Res, 2011.
75:99–103. 10 Hambidge KM, Miller LV, Krebs NF:
2 Donangelo CM, Zapata CL, Woodhouse LR, Relationship between endogenous fecal zinc
et al: Zinc absorption and kinetics during and zinc absorbed revisited. FASEB J 2009;
pregnancy and lactation in Brazilian women. 23:216–218.
Am J Clin Nutr 2005;82:118–124. 11 Steele L, Cousins RJ: Kinetics of zinc absorp-
3 World Health Organization: Trace Elements in tion by luminally and vascularly perfused rat
Human Nutrition and Health. Geneva, 1996. intestine. Am J Physiol 1985;248:G46–G53.
4 Hambidge KM, Miller LV, Westcott JE, et al: 12 Institute of Medicine: Dietary Reference
Zinc bioavailability and homeostasis. Am J Intakes: Applications in Dietary Assessment.
Clin Nutr 2010;91:1478S–1483S. Washington, National Academy Press, 2000.
5 Food and Nutrition Board, Institute of 13 Miller LV, Krebs NF, Hambidge KM: A math-
Medicine: Dietary Reference Intakes for ematical model of zinc absorption in humans
Vitamin A, Vitamin K, Boron, Chromium, as a function of dietary zinc and phytate. J
Copper, Iodine, Iron, Manganese, Molyb- Nutr 2007;137:135–141.
denum, Nickel, Silicon, Vanadium and Zinc. 14 Hunt JR, Beiseigel JM, Johnson LK:
Washington, National Academy Press, 2001. Adaptation in human zinc absorption as
6 Hambidge KM, Krebs NF: Interrelationships influenced by dietary zinc and bioavailability.
of key variables of human zinc homeostasis: Am J Clin Nutr 2008;87:1336–1345.
relevance to dietary zinc requirements. Ann 15 Rosado JL, Hambidge KM, Miller LV, et al:
Rev Nutr 2001;21:429–452. The quantity of zinc absorbed from wheat in
7 International Zinc Nutrition Consultative adult women is enhanced by biofortification.
Group (ed): Assessment of the Risk of J Nutr 2009;139:1920–1925.
Zinc Status in Populations and Options for 16 Hambidge KM, Krebs NF, Westcott JE, Miller
the Control of Zinc Deficiency. Boston, LV: Changes in zinc absorption during devel-
International Nutrition Foundation for opment. J Pediatr 2006;149:S64–S68.
United Nations University Press, 2004. 17 Hambidge KM, Miller LV, Tran CD, Krebs
8 Sian L, Mingyan X, Miller LV, et al: Zinc NF: Measurements of zinc absorption:
absorption and intestinal losses of endog- application and interpretation in research
enous zinc in young Chinese women with designed to improve human zinc nutriture.
marginal zinc intakes. Am J Clin Nutr 1996; Int J Vitam Nutr Res 2005;75:385–393.
63:348–353.

Role of Zinc in Child Health and Survival

Robert E. Black ⭈ Christa Fischer Walker
Department of International Health, Johns Hopkins University, Bloomberg School of Public Health,
Baltimore, MD, USA
Abstract
Zinc deficiency has been estimated to result in more than 450,000 child deaths annually
by increasing the risk of diarrhea and pneumonia mortality. Trials of daily supplemental
zinc have shown preventive benefits in childhood diarrhea with a 20% reduction in inci-
dence. Use of zinc in treatment of diarrhea has also been successful in shortening the
duration of the episode by 10% and reducing the number of prolonged episodes. The
World Health Organization recommends that zinc supplements be used for 10–14 days
for every episode of childhood diarrhea along with oral hydration and feeding. Large-
scale effectiveness trials of these recommendations in Bangladesh and India have found a
reduction in hospitalizations due to diarrhea and pneumonia and in child mortality. Trials
have also demonstrated a reduction in the incidence childhood pneumonia with zinc
supplements and some, but not all, studies have found a therapeutic benefit of zinc as
adjunctive treatment along with antibiotics as well. Preventive zinc also improves the
growth of children in developing countries and reduces total deaths in 1- to 4-year-old
children by 18%. Zinc supplementation is an intervention with proven effectiveness and
broad application to address pneumonia and diarrhea, the two most important child-
hood infectious diseases globally.
A large proportion of the world’s population is at risk of zinc deficiency which

has serious consequences for health and survival, especially in children under
5 years of age [1]. The initial descriptions of severe human zinc deficiency in
adolescent boys included stunting of growth, immune compromise, high rates
of infectious diseases and substantially reduced survival [2]. More recently, it
has been demonstrated, largely through controlled trials of zinc supplementa-
tion, that even less severe zinc deficiency in preschool children is associated
with growth faltering, increased risk for serious infectious diseases and excess
mortality.
Childhood Growth and Development
Stunting of linear growth in early childhood, a good overall measure of child

health, occurs in a third of children worldwide [1]. Zinc deficiency is one con-
tributor to stunting. Most meta-analyses of trials demonstrate that zinc supple-
mentation enhances growth [3], and subgroup analyses suggest that the effect is
greatest in growth-retarded children in less developed countries [4]. A number
of trials of zinc supplementation also find benefits on child activity and devel-
opmental measures, although the evidence is more limited and less clear than
for the effects on growth [4].
Childhood Diarrhea
Diarrheal diseases remain one of the top two causes of child death globally in
spite of reductions over several decades associated with improved nutritional
status and increased implementation of preventive and therapeutic interven-
tions [5]. Children with zinc deficiency are more susceptible to infections caus-
ing diarrhea, and the loss of zinc during diarrhea exacerbates dietary inadequacy
of zinc, contributing to a vicious cycle of infection and nutritional deficiency.
Numerous trials comparing oral zinc supplements, usually given daily, with
placebo have been done and summarized in meta-analyses. The most recently
published of these included 33 comparisons and more than 16,000 child partici-
pants [3]. This analysis found an overall effect of zinc supplementation of a 20%
lower incidence of diarrhea. One of the initial trials of zinc supplementation
reported 27% lower rate of diarrhea in children 12 months of age or more but no
effect in infants [6]. The meta-analysis likewise found a 27% (95% confidence
interval, CI: 13–39%) lower diarrheal incidence in children with an initial age
greater than 12 months and no effect in infants [3]. The effect of zinc was also
found to be significantly greater in children who were stunted at enrollment.
Diarrhea treatment with oral rehydration solution has become a mainstay
for controlling diarrhea mortality, but this has no effect on diarrheal duration.
In contrast, zinc supplementation has been found to reduce the duration and
severity of the episode. Since the first definitive trial of zinc for treatment of
diarrhea [7], there have been numerous trials in low- and middle-income coun-
tries. There have been several meta-analyses of these trials, demonstrating ben-
efits with regard to episode duration, proportion of episodes persisting beyond
7 days and stool frequency. The most recently published systematic review
found that zinc in treatment of diarrhea reduced the mean duration by 19.7%
38 Black · Fischer Walker

[8]. There is heterogeneity in the trial results, and the benefits have been found
in some but not all studies in infants 1–5 months of age.
Several early trials of treatment of diarrhea provided zinc for 2 weeks during
and following the episode. These trials were summarized in a pooled analysis
to show that there was a 34% (95% CI: 17–48%) reduction in the prevalence of
diarrhea and a 26% non-statistically significant reduction in the incidence of
pneumonia in the 2–3 months following the diarrheal episode [9].
In addition, there have been two large trials in which health service areas
were randomly assigned to provide for children with diarrhea ORS or ORS plus
daily zinc supplements for 14 days during and after diarrhea. In Bangladesh,
children in the zinc intervention service clusters had a 24% (95% CI: 10–35)
shorter duration and 15% (95% CI: 4–24) lower incidence of diarrhea than chil-
dren in the comparison group [10]. Admission to hospital for diarrhea was 24%
(95% CI: 2–41%) lower in the intervention group and admission for pneumonia
was 19% lower, but this was not statistically significant. The rate of non-injury
child deaths in intervention clusters was 51% (95% CI: 6–75%) lower. In India,
children in the zinc intervention clusters had lower 24-hour prevalences of diar-
rhea and acute respiratory infections than children in the comparison group
[11]. All-cause, diarrhea and pneumonia hospitalizations were reduced in the
intervention areas by 59, 31 and 71%, respectively (all statistically significantly
different from comparison areas).
Childhood Pneumonia
Pneumonia is the leading cause of child mortality, and zinc deficiency is a risk
factor contributing to the incidence and severity of the disease [5]. As with
diarrhea, the strongest evidence for the role of zinc deficiency comes from zinc
supplementation trials. Because of the lower incidence of pneumonia than of
diarrhea, the data are more limited. Furthermore, the use of a range of illness
definitions for acute respiratory infections (ALRI)/pneumonia makes summa-
rizing these results more complicated than for diarrhea. The ten available trials
of zinc supplementation were analyzed with regard to the outcome definitions
used [12]. Zinc supplementation provided daily or weekly reduced the inci-
dence of ALRI defined using specific clinical criteria by 35% (95% CI: 18–48%).
There was no effect found using less-specific outcome case definitions based on
caregiver report. A separate meta-analysis found an overall reduction of ALRI
of 15% (95% CI: 3–25%), but a larger effect in children who had lower initial
height-for-age (stunted) and in studies using more specific clinical criteria for
the outcome [3].
Zinc has also been used in trials as an adjunctive treatment along with anti-
biotics for pneumonia. An initial trial in Bangladesh of zinc for treatment of
severe pneumonia found a reduction in the episode duration of one day (hazard
Role of Zinc in Child Health and Survival 39

ratio 0.75, 95% CI: 1–43) [13]. However, subsequent trials have not confirmed
this result [14, 15], and the conclusion concerning a therapeutic effect in pneu-
monia will need to await several ongoing trials.
Childhood Malaria
Malaria is an important cause of child mortality, especially in sub-Saharan

Africa [5], and zinc deficiency may be a risk factor for malaria infection or ill-
ness. Several randomized controlled trials have investigated the role of zinc
supplementation for prevention of clinical malaria. Trials in The Gambia and
Papua New Guinea found 32 and 38%, respectively, reductions in clinic visits for
malaria, but a trial in Burkina Faso found no effect on the incidence of malaria
diagnosed by home visits [3]. It is possible that the effect is on the severity of
illness rather than the incidence, but the data are insufficient to determine if
there is an important preventive effect. A multi-site trial of zinc for adjunctive
treatment of malaria did not find any benefit [16].
Child Mortality
Two very large trials of daily zinc supplementation were done in Zanzibar and
Nepal. These trials were designed to examine the effects on child mortality [17,
18]. In Zanzibar, children 1–36 months old were randomized to receive 10 mg
of zinc (5 mg for infants) daily and followed for a total of 56,507 child-years
[17]. Overall, there was a non-significant reduction of 7% in all-cause mortal-
ity in the zinc-supplemented group. The effect seemed to differ by age with a
reduction of 18% (95% CI: 0–32%) in children 12–48 months and no effect in
infants. There were non-significant trends for lower mortality due to malaria,
diarrhea and other infections in the zinc group. In Nepal, children 1–35 months
old were randomized using the same doses as Zanzibar and followed for total of
60,636 child-years [18]. Overall, there was a non-significant reduction of 8% in
all-cause mortality. There was a non-significant reduction of 18% in mortality in
children 12 months or older, but no effect in infants. There were non-significant
trends for reductions in mortality due to diarrhea and ALRI in the zinc group.
In a meta-analysis of the effects of zinc supplementation on mortality, the
study results were stratified by age and whether the zinc was given with iron and
folic acid [3]. There was no effect of zinc supplementation in infants, but chil-
dren 12 months old or more had an 18% reduction in all-cause mortality (95%
CI: 4–30). When iron and folic acid were given with zinc, there was no benefit
in either age group on mortality. Another meta-analysis of the effects of zinc
supplementation on cause-specific mortality suggested effects on diarrhea and
pneumonia of 15–18%, but these effects were non-significant [19].

Conclusions
Zinc deficiency is a prevalent condition that has important effects on child-

hood infectious diseases and mortality. It has been estimated that this defi-
ciency results in more than 450,000 child deaths annually and nearly 4% of the
disability-adjusted life years lost in children worldwide [1]. The preventive use
of zinc supplements or fortified foods would be expected to reduce stunting and
mortality, especially that due to diarrhea and pneumonia [20]. The benefits of
zinc as adjunctive treatment for diarrhea are well demonstrated, and the World
Health Organization and UNICEF since 2004 have recommended that zinc be
used for treatment of all childhood diarrhea [21]. The usefulness of zinc for pre-
venting malaria and treating pneumonia is unclear until the results of additional
trials are available.
References
1 Black RE, Allen LH, Bhutta ZA, et al, the 8 Patel A, Mamtani M, Dibley MJ, et al:
Maternal and Child Undernutrition Study Therapeutic value of zinc supplementation
Group: Maternal and child undernutrition: in acute and persistent diarrhea: a systematic
global and regional exposures and health review. PLoS One 2010;5:1–10.
consequences. Lancet 2008;371:243–260. 9 Zinc Investigators’ Collaborative Group,
2 Prazad AS, Miale A, Farid Z, et al: Bhutta ZA, Black RE, Brown KH, et al:
Biochemical studies on dwarfism, hypogo- Prevention of diarrhea and pneumonia by
nadism and anemia. Arch Internal Med zinc supplementation in children in develop-
1962;111:407–428. ing countries: pooled analysis of randomized
3 Brown KH, Peerson JM, Baker SK, Hess SY: controlled trials. J Pediatr 1999;135:689–697.
Preventive zinc supplementation among 10 Baqui AH, Black RE, Arifeen SE, et al:
infants, preschoolers, and older prepubertal Effect of zinc supplementation started dur-
children. Food Nutr Bull 2009;30(suppl 1): ing diarrhea on morbidity and mortality in
S12–S40. Bangladeshi children: community random-
4 Fischer Walker CL, Black RE: Functional ized trial. BMJ 2002;325:1–7.
indicators for assessing zinc deficiency. Food 11 Bhandari N, Mazumder S, Taneja S, et
Nutr Bull 2007;28(suppl):S454–S479. al: Effectiveness of zinc supplementation
5 Black RE, Cousens S, Johnson HL, et al: plus oral rehydration salts compared with
Global, regional and national causes of child oral rehydration salts alone as a treatment
mortality. Lancet 2010;375:1969–1987. for acute diarrhea in a primary care set-
6 Sazawal S, Black RE, Bhan MK, et al: Efficacy ting: a cluster randomized trial. Pediatrics
of zinc supplementation in reducing the inci- 2008;121:1279–1285.
dence and prevalence of acute diarrhea – a 12 Roth DE, Richard SA, Black RE: Zinc
community-based, double-blind, controlled supplementation for the prevention of acute
trial. Am J Clin Nutr 1997;66:413–418. lower respiratory infection in children in
7 Sazawal S, Black RE, Bhan MK, et al: Zinc developing countries: meta-analysis and
supplementation in young children with meta-regression of randomized trials. Int J
acute diarrhea in India. N Eng J Med Epidemiol 2010;39:795–808.
1995;333:839–844.
Role of Zinc in Child Health and Survival 41

13 Brooks WA, Yunus M, Santosham M, et al: 18 Tielsch JM, Khatry SK, Stoltzfus RJ, et al:
Zinc for severe pneumonia in very young Effect of daily zinc supplementation on child
children: double-blind placebo-controlled mortality in southern Nepal: a community-
trial. Lancet 2004;363:1683–1688. based, cluster randomised placebo-controlled
14 Bose A, Coles CL, Gunavathi JH, et al: trial. Lancet 2007;370:1230–1239.
Efficacy of zinc in the treatment of severe 19 Yakoob MY, Theodoratou E, Jabee A, et al:
pneumonia in hospitalized children <2 years Preventive zinc supplementation in developing
old. Am J Clin Nutr 2006;83:1089–1096. countries: impact on mortality and morbid-
15 Valentiner-Branth P, Srestha PS, Chandyo ity due to diarrhea, pneumonia and malaria.
RK, et al: A randomized controlled trial BMC Public Health 2011;11(suppl 3):S23.
of the effect of zinc as adjuvant therapy in 20 Bhutta ZA, Ahmed T, Black RE, et al, the
children 2–35 months of age with severe or Maternal and Child Undernutrition Study
nonsevere pneumonia in Bhaktapur, Nepal. Group: Maternal and child undernutrition:
Am J Clin Nutr 2010;91:1667–1674. what works? Interventions to affect maternal
16 The Zinc against Plasmodium Study and child undernutrition and survival glob-
Group: Effect of zinc on the treatment of ally. Lancet 2008;371:417–440.
Plasmodium faciparum malaria in children: 21 World Health Organization, UNICEF:
a randomized controlled trial. Am J Clin WHO-UNICEF Joint statement on the clini-
Nutr 2002;76:805–812. cal management of acute diarrhea; in World
17 Sazawal S, Black RE, Ramson M, et al: Effect Health Assembly, Geneva, 2004.
of zinc supplementation on mortality in chil-
dren aged 1–48 months: a community-based
randomized placebo-controlled trial. Lancet
2007;369:927–934.

Discussion on Zinc in Maternal and

Child Health
Zinc and Infection

There are ongoing studies in infants and young children of developing coun-
tries providing data to begin estimating prevalence of inadequate zinc intakes.
Relatively good models now exist to examine the efficiency of absorption in var-
ious situations and the effect of phytate as the major factor affecting absorption.
The issue was raised about the value of conducting more bioavailability studies,
while the major contributing factor is the prevalence of diarrhea in the popula-
tion (Dr. Allen). In response, the speaker (Dr. Hambidge) agreed with the point
made, and indicated that the cornerstone of normal condition is required before
expanded to cover infectious diseases. Eventually, both models are needed. Data
have begun to emerge in relation to possible inflammation role in the gut, and
further studies are required to elucidate whether part or whole of zinc deficiency
problem is due to infection.
The issue of how acute infections including diarrhea and respiratory tract
infections affect zinc requirements in children followed (Dr. Mobarak). Dr.
Hambidge clarified that earlier works have shown that excessive zinc is lost in
diarrheal fluid, so more zinc is needed to compensate for this loss. In addition,
one may need additional zinc to correct for deficiency which is likely to be sig-
nificant in this situation. The inquiry was then made of why given only zinc in
diarrhea when other micronutrients could have been lost as well. The response
was that while oral rehydration solution (ORS) is the core treatment, the ben-
efit of zinc in diarrhea was clearly shown in earlier works by Bhutta et al. [1].
However, the same strength of evidence on other micronutrients and ORS for
diarrhea treatment is lacking at the present time.
A question of whether zinc requirement in HIV population differs from
normal population (Dr. Samburu) was addressed by Dr. Fawzi. So far, there are
a few trials that investigated whether giving zinc supplements to HIV-infected
individuals have a benefit. Earlier observational studies reported a sort of mixed
evidence, and some of the observational studies suggested that zinc was benefi-
cial as low serum zinc was associated with harmful outcomes in terms of dis-
ease progression. Another observational study demonstrated that actually zinc
intake was also harmful, so possibly reflecting a U-shaped relationship. These
studies were all observational with certain limitations. As far as clinical trials
are concerned, there are a few providing a sort of mixed evidence. In South
Africa, zinc was beneficial in reducing diarrhea, while one study in Tanzania
among pregnant women suggested no effect on CD4, CD8 or clinical out-
comes. A third study more recently in the US among HIV drug users suggested
that zinc supplements were actually beneficial in slowing disease progression
particularly among individuals with advanced HIV. These were individuals
with zinc deficiency, and the outcome was zinc supplements reduced the risk
of failing on their antiretroviral therapy. So, the evidence is mixed and requires
further studies (refer to the article by W. Fawzi in this book).
Zinc and Infant Growth

The issue concerns infants who are born growth retarded and whether zinc
requirements differ between infants who are small for gestational age (SGA) and
normal infants, and if there are any data available in terms of tracking of zinc
from those pregnancies (Dr. Bhutta). Dr. Hambidge refers to the work of Dr.
Nancy Krebs in this area with exchangeable zinc pools that shows lesser quanti-
ties of zinc in the zinc pools of SGA infants compared to those of controls. The
follow-on comment (Dr. Solomons) pointed out that in animals with restricted
growth, the nutrient requirements are reduced. Similarly, if the infant is pri-
marily not growing and zinc in growing infants has to do with laying down of
tissues, then zinc requirement will be reduced. On the other hand, when infant
enters growth spurts, then requirements will go up because the amount of tissue
being laid down is rapidly expanding.
Because zinc is one of those nutrients that dramatically falls in breast milk
over a period of time, this phenomenon leads to the issue of zinc supplementa-
tion for both the growth-retarded and/or preterm babies, besides providing iron
to prevent iron deficiency anemia (Dr. Mendoza). Dr. Hambidge indicated that
it’s quite extraordinary in the term infant situation how rapidly the physiologic
decline in milk zinc takes place, and levels are very low by 6 months in the term
infant; similar situation could occur in the preterm infant with a lot of other
issues and challenges happening at the same time.
The next discussion took the direction of iron and zinc absorption. The con-
cern is that in preterm nutrition, the suggestion has been made to start giving
iron quite early. Therefore, if iron affects zinc absorption, then zinc requirement
should be increased in these preterm infants (Dr. Garg). Dr. Hambidge agreed
that zinc should be given too.
A comment was also made that studies in infants and adults showed that when
iron is given together with zinc in foods, there is no effect on zinc absorption.
44 Wasantwisut
Bioavailability
A question was raised concerning bioavailability of different zinc compounds,
citing animal studies done in Beijing that reported zinc gluconate as the most
bioavailable, followed by zinc sulfate and zinc lactate, respectively (Dr. Ludan).
In response, Dr. Hambidge indicated that all water-soluble compounds are bio-
available. With respect to iron, calcium and phytate, the old data suggested that
calcium phytate reduced zinc bioavailability. However, current modeling where
iron and calcium data are available showed that absorption of zinc is increased,
not decreased. Based on the assumption that the zinc and the iron were taking
up some of the binding sites on the phytate, it increased the R2 value from 0.81
to 0.87. Therefore, this issue needs to be revisited.
Stable Isotope Technique

The issue on the use of the stable isotope technique to measure zinc status and
the usefulness of this technique to determine the exchangeable zinc pools was
raised (Dr. Hurrell). The response is that there is no answer on this matter yet.
However, at the time being, with the absorption figures as shown in the pre-
sentation as well as good dietary data that include zinc and phytate values and
random sub-samples of serum zinc from diverse studies, these should result in
useful estimates of population zinc status.
IOM and IZiNCG Estimates

Reconciliation of IOM [2] and IZiNCG [3] estimates is recommended. IZiNCG
is planning a review on zinc requirement estimates to take into consideration
new data from studies published after 2001. The model as presented on the
inhibitory effects of phytate on zinc absorption should be taken into consider-
ation (Dr. Wasantwisut).
Overall, the paper is well presented and raised valid points for consideration
with regard to zinc requirements, including the role of infection, preterm infant
growth, interaction of iron and zinc, bioavailability of zinc in relation to iron,
calcium and phytate, as well as the usefulness of stable isotope technique and
modeling for estimation of population zinc status.
Discussion on ‘The Role of Zinc in Child Health and Survival’ by R.E. Black
and C.F. Walker
Zinc and Diarrhea

A question was raised regarding the reason why zinc given with ORS for the
treatment of diarrhea is more effective in developing countries compared to
developed countries (as demonstrated in ESPGAN study in Europe). Infants
who are exclusively breastfed in the first 6 months are unlikely to be zinc defi-
cient (Dr. Haschke). The response (Dr. Black) is that little information exists
Discussion on Zinc in Maternal and Child Health 45

on zinc in treatment of diarrhea in high-income countries; there is a potential
deficiency in the infant age group. In the first 6 months, this may not be the
issue, but the 6–11 months is certainly more equivocal, and zinc deficiency is
likely to occur. A comment followed that Bangladeshi children have subclinical
zinc deficiency due to low zinc in soil and foods being produced. Zinc therapy
for diarrhea is recommended; the question is whether 20 mg of zinc in young
children is too high and whether it could have an adverse effect (Dr. Alam). Dr.
Black clarified that often a lower dose (10 mg) is used in very young children. In
fact, where 20 mg has been used for a short time period, there are virtually no
side effects. In the meta-analysis of all of the trials, there is a few percent addi-
tional vomiting in the zinc versus placebo groups, but there is no interference
with oral rehydration therapy or clinical outcome. Similarly, an inquiry was
made on possible side effects in certain studies which gave as much as 45 mg
dose to the under-five children (Dr. Kalantari). In response, the 45-mg dose was
used in a couple of therapeutic trials, not in a daily preventive dose, and resulted
in more side effects. Such high dose apparently did not lead to any more benefit
than using a lower dose in the therapy of diarrhea. The WHO recommendation
is 20 mg except for younger children where the dose is lower [4].
With respect to the mechanism of diarrhea for the 20-mg zinc dose (Dr.
Hurrell), this may be due to potential immune mediation, probable physi-
ologic or pharmacologic reasons and anti-secretory effects observed in animal
models. Since many of the causes of diarrhea have very different pathogenesis
with respect to mucosa and secretory factors and so forth, an inquiry was made
whether zinc effect is more likely to be observed with certain kinds of diarrhea
such as bacterial versus viral (Dr. Rosenberg). The response is that most trials
did not conduct ideological investigation. There are two trials reporting benefits
of zinc in Shigella and cholera. It’s quite important in syndromes like cholera or
ideologic specific diarrhea like Shigella to show zinc does work because those
are the ones that would stand out. From a public health standpoint, the overall
effect such as that of ORS matters eventually, and implementation should not be
held back because of lack of ideological reasons.
For the benefits in diarrhea, this appears to be specific to zinc per se (Dr.
Manalaysay), while the probable usefulness of other micronutrients (Dr.
Mobarak) would require supportive evidence. Additional comment was made
(Dr. Bhattacharya) that in clinical practice, pediatric patients with bacterial
infection and bloody diarrhea received i.v. fluid, antibiotics and zinc supple-
ments upon discharge. The follow-up results were excellent with the trend of
improved appetite and faster weight gain.
Another inquiry was whether the existing knowledge on formulations of zinc
for the treatment of diarrhea is sufficient to indicate that the type of preparation
does not matter. In addition, there has been some suggestion that maybe the
effect seen in diarrhea is pharmacologically dependent upon the dosage, and if
the same amount is divided into 3 doses over a 24-hour period, one doesn’t see
46 Wasantwisut
the same effect (Dr. Bhutta). The response is that there are no data to respond to
these issues, and further studies are required.
The consistent lack of zinc effect in the <12 months age group has been raised
(Dr. Wasantwisut), and the reason could be the increased demand due to rapid
growth spurt as well as the immune system which cannot be met through the zinc
dose given. In addition, a good biomarker of zinc status is needed besides serum
zinc, which increases during the period of supplementation. In a developing
country, recommendation on zinc prevention or treatment (Dr. Hadad) should
be based on dietary data or population level of serum zinc measurement.
Other Benefits of Zinc

An issue was raised with respect to routine provision of zinc to children beyond
treatment of diarrhea (Dr. Fawzi). On the one hand there is clear benefit on
prevention of diarrhea, on the other hand when zinc is given with iron there is
some negative interaction. In the context of Sprinkles, the question is whether
there is a certain ratio that should be avoided or promoted. In response, there is
good evidence that zinc supplements have a benefit on infectious disease out-
comes, and that the benefit on growth may be small. Some equivocal evidence
exists of interference or reduction of that effect if zinc is given with iron. It’s not
yet known concerning the right balance of zinc versus iron in a supplement or
a Sprinkle micronutrient powder. The suggestion to go for equal molar ratios
lacks supportive evidence on functional outcomes or increases in serum zinc.
The supplementation trials that gave zinc only, almost always demonstrated an
increase in serum zinc. In this context, a comment was made (Dr. Bhutta) that
so far, the three large Sprinkle studies have not been able to show any impact on
zinc, using either a 7.5- or a 10-mg dosage.
Besides the treatment of diarrhea, an inquiry was made whether zinc may
benefit pneumonia or urinary tract infection and other inflammatory disor-
ders (Dr. Manalaysay). The response is that solid evidence in human trials is
required prior to issuing any recommendation. The potential role of zinc in
treating asthmatic symptoms with respect to pneumonia has been mentioned
(Dr. Bhattacharya), and deserves further attention.
Supplements versus Foods

The people who have got sustained zinc supplementation as prevention for
infectious morbidities would in fact over time meet and exceed the require-
ments. On the other hand, a question was raised whether the preventive aspect
of the supplemental exposure would withhold if people were fortified or their
diet diversified to meet their nutrient requirements within the range of recom-
mendations (Dr. Solomons). In response, in zinc-replete populations, there
would be no therapeutic benefit. With supplements, one could possibly achieve
normal zinc status during the period of supplementation and a couple of months
afterwards. In the case of fortification, the same scale of achievement has not
Discussion on Zinc in Maternal and Child Health 47

been shown. The lack of evidence on improved zinc status from fortified foods
or Sprinkles versus supplements could be due to the fact that the absorption
of zinc in supplements is much higher than from foods, so a higher fortified
dose may be required to reach similar outcomes (Dr. Hurrell). Another factor
could be that there are no sensitive methodologies or biomarkers to measure the
impact of zinc in foods. More studies are on the way to unravel the issue sur-
rounding the formulation and whether or not adding Sprinkles or adding zinc
to Sprinkles per se or zinc to Sprinkles without iron is a means of assessing their
bioavailability (Dr. Bhutta).
An inquiry was made regarding multiple interventions, in this case zinc in
Sprinkles and zinc supplements for diarrhea, of whether there is a risk of toxic-
ity (Dr. Penny). The response was that there has to be evidence on the impact of
zinc in Sprinkles on improving zinc status before judging potential excess level
of intake. In addition, the safety margin with zinc is incredibly high. Additional
comment was made that the efficacy of supplements is different from Sprinkles
or micronutrient powder in food due to phytic acid. A study has just been com-
pleted where micronutrient powder that contains endogenous phytase actually
reduces the phytic acid content of the complementary food resulting in a sig-
nificant increase in serum zinc (Dr. Zimmermann). In addition, a randomized
control trial using different dosages of zinc against the background of Sprinkles
is underway and should provide the results by the end of the year (Dr. Bhutta).
Overall, the paper was well presented and generated a lot of discussion,
primarily on the benefits of zinc in diarrhea and other infectious diseases and
the mechanism involved, the ideological spectrum of pathogen-specific envi-
ronment, the differences in efficacy trials of zinc supplements compared to
food-based strategies, and the concern of excess level of intake with multiple
interventions.
Emorn Wasantwisut
References
1 Bhutta ZA, Black RE, Brown KH, et al: 3 International Zinc Nutrition Consultative
Prevention of diarrhea and pneumonia by Group (ed): Assessment of the Risk of
zinc supplementation in children in devel- Zinc Status in Populations and Options for
oping countries: pooled analysis of random- the Control of Zinc Deficiency. Boston,
ized controlled trials. Zinc Investigators’ International Nutrition Foundation for
Collaborative Group. J Pediatr 1999; United Nations University Press, 2004.
135:689–697. 4 World Health Organization, UNICEF:
2 Food and Nutrition Board, Institute of WHO-UNICEF Joint statement on the
Medicine: Dietary Reference Intakes for Clinical Management of Acute Diarrhea.
Vitamin A, Vitamin K, Boron, Chromium, Geneva, World Health Assembly, 2004.
Copper, Iodine, Iron, Manganese,
Molybdenum, Nickel, Silicon, Vanadium
and Zinc. Washington, National Academy
Press, 2001.
48 Wasantwisut
Global Burden and Significance of

Multiple Micronutrient Deficiencies
in Pregnancy
Ian Darnton-Hill
Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney,
Sydney, NSW, Australia, and Friedman School of Nutrition Science and Policy, Tufts University,
New York, NY, USA
Abstract
Maternal mortality, low birthweight infants and childhood stunting continue to be major
global public health problems, part of a recurring cycle of disadvantage. Maternal undernu-
trition in particular is one of the most neglected aspects of nutrition in public health. One
possible low-cost public health intervention that might help address these problems is the
antenatal provision of multiple micronutrient supplements. If the evidence base could be
established, cost-effectiveness found to be acceptable and safety ensured, supplementa-
tion could ameliorate the impact of poor nutrition and diets, high disease burdens and the
sociocultural factors contributing to these problems. There have been good studies in over
a dozen countries addressing some of these issues but with conflicting results. Consequently,
at least three meta-analyses have been undertaken to establish significant findings that
could help guide policies and programs. They concluded that multimicronutrient supple-
mentation improves birthweight and likely reduces the number of infants born low birth-
weight. Supplementation with iron-folic acid or multimicronutrients also appears to have
positive longer-term impacts on the health and development of the offspring. There remain
concerns about possible increased infant mortality in some populations. Given the results
of the meta-analyses, cautious scaling-up of country effectiveness trials appears justified
with careful monitoring and evaluation. Copyright © 2012 S. Karger AG, Basel
Introduction
High levels of maternal mortality, infants of low birthweight (LBW) and child-
hood stunting remain major public health problems, especially in resource-
poor settings. They are also problems that have remained largely intractable,
although there has been some recent progress. Maternal undernutrition in
particular is one of the most neglected aspects of nutrition in public health
globally [1]. Consequently, low-cost public health interventions that might
help ameliorate the impact of poor nutrition and diets, high disease burdens
and the sociocultural factors contributing to the high levels of these prob-
lems are needed. One of these may be the antenatal provision of multiple
micronutrients. This paper looks at the evidence for micronutrient deficien-
cies and the likely impact of these, especially in women of low- and middle-
income countries, the global burden of such deficiencies, their significance,
and recent evidence that supplements of multimicronutrients may help to
address poor health and nutrition of women and improve subsequent repro-
ductive outcomes, including even longer-term benefits to the children of
women who received the supplementation. The paper by Bhutta [2] discusses
intervention strategies to address such multiple micronutrient deficiencies in
pregnancy.
Micronutrient Deficiencies in Women of Reproductive Age
Concurrent deficiencies of micronutrients are well documented among young

pregnant women (and young children), especially in low- and moderate-
income countries [3–5]. The commonest deficiency is iron, but often there is
at least one other deficiency [4]. These deficiencies in maternal micronutrient
status are a result of poor-quality diets, high fertility rates, repeated pregnan-
cies, and short inter-pregnancy intervals, increased physiological needs, as well
as inadequate health systems with poor capacity, poverty and inequities, and
sociocultural factors such as early marriage and adolescent pregnancies and
some traditional dietary practices [6]. A systematic review identifying all stud-
ies that had been published between 1988 and 2008 reporting on micronutrient
intakes in women living in resource-poor environments showed that, except
for vitamin A (29%), vitamin C (34%), and niacin (34%), the reported mean/
median intakes in over 50% of the studies were below the estimated average
requirements, demonstrating that inadequate intakes of multiple micronutri-
ents are common amongst women living in resource-poor settings [7], includ-
ing urban settings [8].
Data on vitamin and mineral metabolism and requirements during preg-
nancy are surprisingly imprecise, largely because of the complexity of maternal
metabolism during pregnancy. Requirements for many, but not all, micronutri-
ents increase during pregnancy. Studies of micronutrient status in adolescents,
including when pregnant, have found poor micronutrient intakes and status [9],
including in the UK [10], and increased risk of small for gestational age (SGA)
and LBW infants at birth [6].
50 Darnton-Hill
Global Burden
Information on the burden of diseases can help countries assess their compara-
tive importance in causing premature death, loss of health and disability, and
so assist countries in deciding health policies and programs [11]. Although the
numbers of pregnant women affected by micronutrient deficiencies, especially
iron deficiency and anemia, are considerable, the actual global burden of multi-
ple micronutrient deficiencies during pregnancy has not been estimated [3, 11].
It would be difficult to estimate as the burden rests not only on women’s mortal-
ity, morbidity and reproductive health outcomes [1], but also inter-generational
effects that affect both the mother and her neonate’s immediate burden and that
of her children, especially in terms of intellectual and physical development and
the later incidence of noncommunicable diseases [5, 12–16].
The micronutrients, for which the global burden is reported upon, are iodine
deficiency, vitamin A deficiency, and iron deficiency anemia, and an estimate for
zinc has also been made [3, 11]. However, while the numbers of those affected
may be great, e.g. up to two billion for iron deficiency or those living in areas
of poor iodine availability, the burden (as measured by disability-adjusted life
years – DALYs) on women of child-bearing age (15–59 years) is limited. Except
for anemia, most of the impact is on the infants and children, e.g. the num-
ber of child deaths attributed to zinc, vitamin A and iron deficiencies was esti-
mated in 2002 to be 19% of all child deaths or just over two billion children [3].
In total, micronutrient deficiencies were estimated in the 2002 World Health
Report to cause about 6% of global DALYs [3]. Since then, progress has contin-
ued to be made in the prevention and control of iodine deficiency and vitamin
A deficiency, where the burden of disease for females caused by iodine and vita-
min A deficiency is only 0.2% and 0.0% of the global total of DALYs, respec-
tively [3, 11]. By contrast, little progress has been made with iron deficiency,
and estimates attribute it causing a fifth of early neonatal mortality and a tenth
of maternal mortality, along with the additional morbidity of reduced cogni-
tive development in children and work performance in adolescents and adults.
About 800,000 deaths and 2.4% of global DALYs were attributed to iron defi-
ciency in 2002 [3] with the global figures, based on 2004 estimates, in table 1,
with updated figures due this year (2011) [11].
Although the global burden for reproductive-aged women from deficiencies
of a range of vitamins and minerals is not estimated as such, WHO does give
estimates of the three most prevalent micronutrient deficiencies by gender and
by WHO region, and which show marked differences. The impact on women of
poor micronutrient status is double that of males, and is far greater in low-income
countries, so that the burden (as measured by DALYs) of iron deficiency anemia
in low-income countries e.g. is 12.5 times that of high-income countries [11]. In
the eastern Mediterranean, South Asia (SEAR) and Sub-Saharan Africa (AFR),
3 out of every 10 deaths are due to ‘communicable, reproductive or nutritional
Multiple Micronutrients in Pregnancy 51

Table 1. Deaths and burden of disease (as measured by DALYs) as percent of total deaths and
total DALYs, by nutritional cause and by female sex (based on [3] and tables A1, A2, and A5a in
Annexes of WHO [11])
Nutritional cause Deaths as % of total (n) Burden by DALYs as % of total

1 2
both sexes females both sexes females
Protein-energy malnutrition 0.4 (251) 0.4 (16) 1.1 1.2

Iodine deficiency 0.0 (5) 0.0 (0) 0.2 0.2
Vitamin A deficiency 0.0 (17) 0.0 (0) 0.0 0.0
Iron deficiency anemia 0.3 (153) 0.4 (55) 1.1 1.3
Figures in parentheses indicate total numbers in thousands.

1
Includes children.
2
For numbers only – age range for females is 15–59 years.
conditions’ [11]. Most of the deaths from iron deficiency anemia in women (15–
59 years) occur in low-income countries (about 70,000 poorer women annually
to virtually none in high-income countries) contributing an estimated 5,792,000
DALYs (or about 1.6%) [11]. By region, SEAR bears 4.1% of the global burden for
iron deficiency anemia, 3.8% in the eastern Mediterranean and 3.0% in AFR. The
burden of maternal conditions in AFR and SEAR is responsible for 8% of the total
global burden in women aged 15–59 years [11]. Almost all of this loss of healthy
years of life is avoidable. In equivalent years of healthy life lost for women from
anemia, the number in low- and middle-income countries for women is 2.4%
or 7.4 million women [11]. There is no information on the probably fairly small
contribution to the global burden of disease from other micronutrients of public
health importance such as folate, vitamin B12 and other B vitamins and selenium,
although the total number of women deficient in two or more of these is con-
siderable, e.g. 46% of rural Ethiopian women are said to have folate deficiency.
Another way of looking at this is the global burden of prematurity and LBW (to
which maternal nutrition and health contribute importantly), which is 2.9% of
total DALYs, and which again rises to 3.9% in low-income countries [11].
Significance
The significance of this likely relatively large burden (in numbers if not percentages
of total) due to multiple micronutrient deficiencies is that it is heavily weighted
against women and especially those in low-income countries. In resource-poor
settings where diagnosis and treatment may be difficult to access, solutions are
likely to need a large public health component approach [6]. Because LBW is
associated with higher infant mortality rates, stunting and impaired intellectual
52 Darnton-Hill
development, there is a need to shift, at a population level, the whole birthweight
curve towards adequate weight in low and middle-income countries [6]. The
risks in doing this are likely to have been overemphasized, but it still means that
health service improvements and sufficient monitoring and evaluation need to be
strengthened, along with maternal antenatal nutritional interventions [17].
Multimicronutrient Supplementation
Over the last decade, there have been studies in at least 12 countries testing the
impact of antenatal micronutrient supplementation [6], and at least four meta-
analyses [12, 14, 15, 18]. However, the many confounding factors, different epi-
demiological environments and diverse methodological approaches mean that it
remains difficult to have a consistent body of evidence in order to come to con-
clusions and make concrete recommendations, although there are some clear
directions emerging. While most of the studies and meta-analyses have appro-
priately guarded conclusions, it has been noted that in public health, action can
rarely wait for perfect knowledge [6]. The prevalence of HIV in communities
and mothers has provoked a lot of the multiple micronutrient research, using
composition of supplements ranging from single micronutrients, iron-folic acid
supplements (IFAs) and multiple vitamins and minerals of varying composition
and multiples of RNIs/RDAs. Women’s stature probably leads to different study
results – from the slight, undernourished women of rural Nepal, to the some-
what better nourished women of China, or the underweight but taller women
of parts of Africa and so on [6]. Further confusing issues are the measured out-
comes e.g. miscarriage rates, fetal deaths, neonatal deaths, size at birth, micro-
nutrient status of the newborn, development of the young child and so on.
Maternal Impact
Clinically speaking, iron deficiency and anemia in pregnancy are associated
with higher maternal mortality and morbidity, premature delivery, LBW, and
iron-deficiency and anemia in infants [19]. Supplementation with iron and folic
acid during the antenatal period has long been WHO policy and the policy of
most countries. Surprisingly few of the antenatal supplementation studies have
looked at the micronutrient status of mothers themselves, but where it has been
examined for, there has generally been an improvement in the levels of other
micronutrients as well, although not always. Long-term intermittent multiple
micronutrient supplementation enhanced hemoglobin and micronutrient status
more than IFA in Bangladeshi adolescent girls with nutritional anemia [9]. In the
SUMMIT study, there was also a 33% decrease in the number of anemic moth-
ers [20]. One meta-analysis that was done of ten randomized controlled effi-
cacy and effectiveness trials covering 12 countries showed, among other things,
supplementation can improve outcomes beyond anemia, including deficiencies

of other micronutrients and birthweight and that the multiple micronutrients
worked as well, if not better, than IFA in terms of reducing anemia [6]. Multiple
micronutrient supplementation increased hemoglobin synthesis to the same
extent as supplementation with iron or iron-folic acid supplements, although
they often contained lower amounts of iron (30 mg as opposed to 60 mg) [4].
Nevertheless, a high prevalence of various micronutrient deficiencies persisted
in those receiving the multiple micronutrient supplements [4].
Supplementation with multivitamins has also been shown to reduce morbid-
ity and mortality in non-pregnant HIV-infected women [21]. Another finding
in many of the studies was that adolescent girls and women in real-life situations
can have very adequate compliance/adherence rates when there are adequate
supplies and some support [4, 6, 9, 17].
Neonatal Impact
In an early study in low-income urban US women, supplement use was associ-
ated with an approximately twofold reduction in risk of preterm delivery and
in overall risk of LBW [22]. However, a Cochrane review in 2005 found that
taking vitamin supplements prior to pregnancy or in early pregnancy did not
prevent women from experiencing miscarriage or stillbirth, but they were less
likely to develop pre-eclampsia, but also more likely to have a multiple preg-
nancy [23]. In a study of 4,752 US women (of whom 95% self-reported vitamin
supplement use in early pregnancy), it was found that any use of vitamins dur-
ing pregnancy was associated with decreased odds of miscarriage in compari-
son with no exposure [24]. On the other hand, Alwan et al. [25] in a prospective
birth cohort study of UK women in Leeds found that regular multimicronutri-
ent supplement use during pregnancy in a developed country setting, while not
associated with size at birth was associated with reduced preterm birth if taken
daily in the third trimester. In the context of HIV, a placebo-controlled trial in
Tanzania showed that HIV-infected pregnant women who took multivitamins
of multiples of RDAs gave birth to fewer low-weight and preterm infants [21].
Interestingly, a recent paper demonstrated that multivitamins at multiple and
single doses of the RDA had similar effects on the risk of LBW [21]. An Indian
study found that early neonatal morbidity was decreased by 58% [26].
Using three sets of DHS data, an Australian group found that IFA during
pregnancy in Indonesia significantly reduced the risk of early neonatal death
[27]. In a double-blind cluster-randomized trial in Lombok, where Indonesian
women were given multimicronutrient supplements by midwives to be taken
daily, their infants had an 18% reduction in early infant mortality compared
with those of women given IFAs, and infants of those mothers who were under-
nourished or anemic at enrollment had a reduction in early infant mortality of
25 and 38%, respectively [20].
However, some important questions remain. Findings from two sepa-
rate, and different, studies in Nepal showed an increase, not statistically
54 Darnton-Hill
significant, in the risk of early neonatal mortality associated with the use of
multiple micronutrient supplements compared with IFA and vitamin A, but
which was significant when the studies were combined [28, 29]. The findings
of an increased neonatal mortality have not been supported by at least three
of the systematic reviews, as well as by data from other studies in different
populations [2, 6, 12]. The meta-analysis of Ronsmans et al. [15] concluded
that there was no overall difference in deaths, although when the SUMMIT
results were excluded from the analysis because of significant heterogeneity (p
< 0.10), there was a statistically higher odds ratio of early neonatal mortality
[6]. Nevertheless, overall the meta-analysis reported by the United Nations
System Standing Committee on Nutrition (SCN) found no statistically signifi-
cant differences between infants of mothers supplemented with IFC and those
supplemented with multimicronutrients, and noted that increased birthweight
has consistently been found to be associated with reduced risk of dying in
infancy in other situations [6]. A recent review by Bhutta’s group concluded
that there was no significant increase in the risk of neonatal mortality where
skilled birth care is available and the majority of births take place in facilities
[2], a not common situation in many populations with the greatest prevalence
of multiple micronutrient deficiencies.
Effects on Birthweight
The meta-analysis reported on by the SCN and others observed a mean increase
in birthweight of 22 g (4.9–75.5 across studies), and a reduction in the preva-
lence of LBW of about 10%, with a larger impact on birthweight in infants of
heavier women [6, 14]. There was a positive shift in the entire birthweight dis-
tribution, with decreases in the numbers of LBW babies and SGA babies with
no differences in birth length or head circumference [6]. It was noted that the
small significant increase in mean birthweight among infants of supplemented
mothers is of similar magnitude to that produced by food supplementation dur-
ing pregnancy [6]. Haider and Bhutta in their meta-analysis for the Cochrane
review, found a relative risk of 0.84 reduction in LBW compared with iron and
folic acid [12]. There was a 14% reduction in LBW with a 33% decrease in ane-
mic mothers in the large Lombok study [20].
Another meta-analysis and systematic review of 13 published trials that
included HIV-positive women also found that prenatal multimicronutrient
supplementation in resource-poor settings was associated with a significantly
reduced risk of LBW by 17–19% and with an improved mean birthweight of
54 g when compared with IFAs [18]. They found no significant effect of the
supplementation on the risk of preterm birth or SGA infants [18]. They con-
cluded that: ‘With the possibility of reducing the incidence of low birth weight
by 17%. . .providing pregnant women with multimicronutrient supplementa-
tion offers the highest possible return for the investment. . .and could avert 1.5
million births of LBW infants globally each year’ [18].

Increases in birthweight with antenatal multiple micronutrient supplementa-
tion have now also been found in studies reported since two of the meta-analyses
were published but included in that of Shah and Ohlsson [18]. One was with
well-nourished French women in a hospital setting where there was an increase
in the average birthweight by 251 g as compared with a placebo [Hininger et al.,
cited in 6]. The other was an Indian study, also hospital-based, of thin women
(BMI <18.5), and where the infants of the micronutrient-supplemented moth-
ers group were heavier by a mean 98 g and measured 0.8 cm longer and 0.2
cm larger in mid-arm circumference compared with the placebo group [26].
Incidence of LBW declined from 43.1 to 16.2% with multimicronutrient supple-
mentation (a 70% decrease).
Effects on Infant and Child Development of Maternal Micronutrient

Supplementation
Besides the health and well-being of the mother herself, and successful imme-
diate reproductive outcomes, the subsequent health and development of the
infant and older child can be clearly affected by the intrauterine environment.
Early iron deficiency is known to alter the neuroanatomy, biochemistry, and
metabolism in the infant leading to changes in the neurophysiologic processes
that support cognitive and motor development [5]. In a long-term study of
Finns, there was a direct positive association between Hb levels of the preg-
nant women and the educational achievement of their children later in life
(at 14 and 16 years of age) with also increased odds of having a higher level
of education at the age of 31 years [30]. In a rural area of Nepal where iron
deficiency is prevalent, aspects of intellectual functioning including working
memory, inhibitory control and fine motor functioning among offspring were
positively associated with prenatal IFA supplementation [5]. Positive func-
tional and developmental milestones of the Nepalese children whose mothers
were supplemented in pregnancy with multiple micronutrients showed small
improvements in weight and a decrease in peripheral adiposity after 2 years
[5]. Although in several of the studies in Nepal, the addition of other micro-
nutrients or zinc attenuated or negated the impact of the IFA, this is the
opposite to a study in China where multimicronutrient prenatal supplemen-
tation found modest improved changes in cognitive outcomes compared with
IFA [31]. Similarly, in Bangladesh there were small but significant improve-
ments in motor scores and activity ratings compared with mothers receiving
iron/folic acid [32]. This was also found for psychomotor but not cognitive
development indices of children born to multimicronutrient-supplemented
HIV-positive Tanzanian mothers (compared to placebo) [33]. Reductions in
subsequent stunting, e.g. in Vietnam, have also been reported [6, 17]. The
public health significance of such findings is not yet known, but may be set-
ting these children on a different developmental trajectory that mitigates
the risk of chronic disease in adult life [6], a problem that is now greater
56 Darnton-Hill
in low- and moderate-income countries, and in rural populations as well as
urban ones [8, 16].
Significance for Interventions
Public health antenatal and obstetric interventions on a larger scale, as the

above evidence seems to suggest, would also require monitoring for effective-
ness, sustainability and impact. WHO already has recommendations regard-
ing supplementation to improve iron stores in adolescent girls and women
before they enter pregnancy in areas where anemia rates are greater than 40%,
and more recently a position statement suggested intermittent IFA where the
prevalence of anemia in women of reproductive age is greater than 20% [19].
It is also policy to provide HIV-infected pregnant women multiple micronu-
trients (of a single RDA), and multimicronutrients for pregnant and lactat-
ing women in emergency situations have been encouraged in a WHO/WFP/
UNICEF statement. Nevertheless, cost-effectiveness has not been established,
and neither has there been clear proof that there is an advantage over IFA,
although it does seem likely that a lower dose of iron is effective in a multimi-
cronutrient supplement.
There are some other positive aspects. Most of the research appears to be
heading in largely the same direction, at least in terms of much improved ante-
natal maternal care, including supplementation where relatively high adherence
rates have been shown to be achievable in field settings where it is emphasized
[6, 9, 17]. There is increased policy and donor attention to maternal and early
child nutrition and health, including through a wish to achieve the relevant
MDGs by 2015. From a public health perspective, the maternal and child
nutrition Lancet series concluded that effective micronutrient interventions
for pregnant women should include supplementation with iron-folic acid, and
noted that multiple micronutrients reduced the risk of LBW at term by 16%
[34]. There is general agreement that supplementation programs should be
part of a larger improvement in antenatal care programs including increased
contact, improved nutrition, deworming and access to clinic care and delivery
and so on. A very recent review by Haider et al. [35] concluded that there is
good evidence ‘of a significant benefit of MMN supplementation during preg-
nancy on reducing SGA births as compared to iron-folate, with no significant
increase in the risk of neonatal mortality in populations where skilled birth
care is available and majority of births take place in facilities’. Nevertheless,
some outstanding issues remain: the possible increased risk of neonatal mor-
tality in some populations; the appropriate levels of the multiple vitamins and
minerals in supplements [6], and issues of cost-effectiveness [13] as interven-
tions are needed that are both efficacious and effective [1]. Nevertheless, taking
all the studies together, and with experience in countries such as Indonesia,

Mexico, Vietnam and elsewhere, it is likely that effective distribution and pro-
motion systems can be developed for different target groups and settings [36]
‘in the context of available services in health systems and [with] birth outcomes
monitored’ [35].
Conclusions
There are substantial inequalities in maternal and newborn health, and in access
to health care. In an ideal world, young women would be born with adequate
birthweight, develop appropriately without stunting, and go into their first
pregnancy no earlier than 18 years, and be adequately nourished. Their health,
sanitation and social support should also be optimal, not least because it seems
likely the significantly better survival and other results found in Lombok may
well have been due to the attention paid to antenatal services and maternal care
in general. In the absence of this happening, all evidence-based, simple and cost-
effective measures must be adequately tried in low-resource settings. It has been
frequently noted that antenatal supplement with multiple vitamins and miner-
als is the norm in many affluent countries and often for the affluent in poorer
countries. More programmatic information will be coming from the limited
number of countries using multiple micronutrient interventions to pregnant
women, but these need to be closely monitored. WHO is currently using its new
NUGAG process to develop guidance and then policy for recommendations.
Building on the commentaries written around the different meta-analyses,
some conclusions might be drawn.
• Deficiencies in micronutrients affect many women of reproductive age, and
are associated with adverse maternal and perinatal outcomes and have even
likely longer-term impacts into adulthood
• Micronutrients likely to be important for maternal, infant and child outcomes
include iron, vitamin B12, folate, vitamin D and selenium and probably zinc
and maybe others (along with appropriate dietary energy intakes)
• In addition to multimicronutrient supplementation, optimal interventions
to improve maternal nutrition need to address household food insecurity,
reduce the burden of maternal infections such as HIV and malaria, and
actively address gender and social disadvantage
• In the meantime, if proven effective and safe in representative health
care systems, supplementation with multimicronutrients should replace
supplementation with iron and folic acid in vulnerable populations as a
way of breaking the intergenerational reality of LBW infants growing up
disadvantaged and stunted to repeat the cycle.
58 Darnton-Hill
References
1 Bhutta ZA, Haider BA: Prenatal micronutri- 11 World Health Organization: Health Statistics
ent supplementation: are we there yet? CMAJ and Health Information Systems. Geneva,
2009;180:1188–1189. World Health Organization, 2004.
2 Bhutta ZA: Intervention strategies to address 12 Haider BA, Bhutta ZA: Multiple-
multiple micronutrient deficiencies in preg- micronutrient supplementation for women
nancy; in Bhutta Z, Hurrell RF, Rosenberg during pregnancy. Cochrane Database Syst
IH: Meeting Micronutrient Requirements for Rev 2006;CD004905.
Health and Development. Nestlé Nutr Inst 13 Darnton-Hill I, Webb P, Harvey PWJ, et
Workshop Ser. Nestec, Vevey/Karger, Basel, al: Micronutrient deficiencies and gender:
2011, vol 70, pp 61–73. social and economic costs. Am J Clin Nutr
3 Black R: Micronutrient deficiency – an 2005;819(suppl):1198S–1205S.
underlying cause of morbidity and mortality. 14 Fall CHD, Fisher DJ, Osmond C,
Bull World Health Org 2003;81:79. Margetts BM, the Maternal Micronutrient
4 Allen LH, Peerson JM, the Maternal Supplementation Study Group: Multiple
Micronutrient Supplementation Study micronutrient supplementation dur-
Group: Impact of multiple micronutrient ing pregnancy in low-income countries:
versus iron-folic acid supplements on mater- a meta-analysis of effects on birth size
nal anemia and micronutrient status in preg- and length of gestation. Food Nutr Bull
nancy. Food Nutr Bull 2009;30:S527–S532. 2009;30:S533–S546.
5 Christian P, Murray-Kolb LE, Khatry S, et 15 Ronsmans C, Fisher DJ, Osmond C, et al, the
al: Prenatal micronutrient supplementation Maternal Micronutrient Supplementation
and intellectual and motor function in early Study Group: Multiple micronutrient supple-
school-aged children in Nepal. J Am Med mentation during pregnancy in low-income
Assoc 2010;304:2716–2723. countries: a meta-analysis of effects on still-
6 Shrimpton R, Huffman SL, Zehner ER, et births and on early and late neonatal mortal-
al: Multiple micronutrient supplementation ity. Food Nutr Bull 2009;30:S547–S555.
during pregnancy in developing country 16 Andersen LG, Ängquits L, Eriksson JG, et al:
settings: policy and program implications of Birth weight, childhood Body Mass Index
the results of a meta-analysis. Food Nutr Bull and risk of coronary heart disease in adults:
2009;30:S556–S573. combined historical cohort studies. PLoS
7 Torheim LE, Ferguson EL, Penrose K, One 2010;5:e14126.
Arimond M: Women in resource-poor 17 The United Nations System Standing
settings are at risk of inadequate intakes Committee on Nutrition: 6th Report on
of multiple micronutrients. J Nutr the World Nutrition Situation. Progress
2010;140:2051S–2058S. in Nutrition. Geneva, World Health
8 Matthews Z, Channon A, Neal S, Osrin D, Organization.
et al: Examining the ‘urban advantage’ in 18 Shah PS, Ohlsson A, on behalf of the
maternal health care in developing countries. Knowledge Synthesis Group on deter-
PLoS Med 2010;7:e1000327. minants of low birth weight and preterm
9 Ahmed F, Khan MR, Akhtaruzzaman M, births: Effects of prenatal multimicronutri-
et al: Efficacy of long-term twice-weekly ent supplementation on pregnancy out-
supplementation with multiple micronutri- comes: a meta-analysis. Can Med Assoc J
ents compared to iron-folic acid supplements 2009;180:E99–E108.
on hemoglobin and micronutrient status of 19 World Health Organization: Weekly Iron-
rural adolescent girls in Bangladesh. J Nutr Folic Acid Supplementation in Women of
2010;140:1879–1886. Reproductive Age: Its Role in Promoting
10 Baker PN, Wheeler SJ, Sanders TA, et al: A Optimal Maternal and Child Health.
prospective study of micronutrient status Position Statement. Geneva, World Health
in adolescent pregnancy. Am J Clin Nutr Organization, 2010.
2009;89:1114–1124.

20 Supplementation with Multiple 29 Osrin D, Vaidya A, Shrestra Y, et al: Effects
Micronutrients Intervention Trial Study of antenatal multiple micronutrient supple-
Group, Shankar AH, Jahari AB, Sebayang SK, mentation on birthweight and gestational
et al: Effect of maternal multiple micronutri- duration in Nepal: double-blind, randomised
ent supplementation on fetal loss and infant controlled trial. Lancet 2005;365:955–962.
death in Indonesia: a double-blind cluster- 30 Fararouei Md, Robertson C, Whittaker J,
randomised trial. Lancet 2008;371:215–227. et al: Maternal Hb during pregnancy and
21 Kawai K, Kupka R, Mugusi F, et al: A ran- offspring’s educational achievement: a pro-
domized trial to determine the optimal dos- spective cohort study over 30 years. Br J Nutr
age of multivitamin supplements to reduce 2010;104:1363–1368.
adverse pregnancy outcomes among HIV- 31 Li Q, Yan H, Zeng L, et al: Effects of maternal
infected women in Tanzania. Am J Clin Nutr development of infants in rural Western
2010;91:391–397. China: follow-up evaluation of a double-
22 Scholl TO, Hediger ML, Bendich A, et al: Use blind, randomized, controlled trial. Pediatr
of multivitamin/mineral prenatal supple- 2009;123:e685–e692.
ments: influence on the outcome of preg- 32 Tofail F, Persson LA, El Arifeen S, et al:
nancy. Am J Epidemiol 1997;146:134–141. Effects of prenatal food and micronutrient
23 Rumbold A, Middleton P, Crowther CA: supplementation on infant development:
Vitamin supplementation for preventing a randomized trial from the Maternal
miscarriage. Cochrane Database Syst Rev and Infant Nutrition Interventions,
2005;CD004073. Matlab (MINIMat) study. Am J Clin Nutr
24 Hasan R, Olshan AF, Herring AH, et al: Self- 2008;87:704–711.
reported vitamin supplementation in early 33 McGrath N, Bellinger D, Robins J, et al:
pregnancy and risk of miscarriage. Am J Effect of maternal multivitamin supple-
Epidemiol 2009;169:1312–1318. mentation on the mental and psychomotor
25 Alwan NA, Greenwood DC, Simpson NA, et development of children born to HIV-1-
al: The relationship between dietary supple- infected mothers in Tanzania. Pediatrics
ment use in late pregnancy and birth out- 2006;117:e216–e225.
comes: a cohort study in British women. Br J 34 Bhutta ZA, Ahmed T, Black RE, et al:
Obstet Gynaecol 2010;117:821–829. Maternal and child undernutrition study
26 Gupta P, Ray M, Dua T, et al: group. What works? Interventions for mater-
Multimicronutrient supplementation for nal and child undernutrition and survival.
undernourished pregnant women and the Lancet 2008;371:417–440.
birth size of their offspring: a double-blind, 35 Haider BA, Yakoob MY, Bhutta ZA: Effect
randomized, placebo-controlled trial. Arch of multiple micronutrient supplementation
Pediatr Adolesc Med 2007;161:58–64. during pregnancy on maternal and birth
27 Titaley CR, Dibley MJ, Roberts CL, et al: Iron outcomes. BMC Public Health 2011;11(suppl
and folic acid supplements and reduced early 3):S19–S26.
neonatal deaths in Indonesia. Bull World 36 Casey GJ, Jolley D, Phuc TQ, et al: Long-
Health Org 2010;88:500–508. term weekly iron-folic acid and de-worming
28 Christian P, Khatry SR, Katz J, et al: Effects of is associated with stabilized haemoglo-
alternative maternal micronutrient supple- bin and increasing iron stores in non-
ments on low birth weight in rural Nepal: pregnant women in Vietnam. PLoS One
double blind randomised community trial. 2010;5:e15691.
BMJ 2003;326:571–576.
60 Darnton-Hill
Intervention Strategies to Address

Multiple Micronutrient Deficiencies in
Pregnancy and Early Childhood
Aamer Imdad ⭈ Zulfiqar A. Bhutta
Division of Women’s and Children’s Health, Aga Khan University, Karachi, Pakistan
Abstract
Deficiencies of multiple micronutrients are prevalent among women of reproductive age
and young children, and represent a risk factor for increased morbidity and mortality in
these women and children. The role of multiple micronutrient supplementation during
pregnancy and early childhood has been evaluated in randomized trials. Multiple micro-
nutrient supplementation during pregnancy has a positive effect on birthweight and
reduces prevalence of low birthweight and small for gestational age babies. It had com-
parable effects on prevalence of anemia regarding iron-folate supplementation. Multiple
micronutrient supplementations in children have been shown to improve linear growth,
weight, hemoglobin, serum zinc, serum retinol levels and motor development. Some of
the most commonly used strategies to deliver multiple micronutrients include powders
(e.g. Sprinkles®), crushable tablets (e.g. Foodlets), etc. Multiple micronutrient supplemen-
tation during pregnancy and early childhood seems to be an effective way of prevention
of micronutrient deficiencies and has a significant protective effect against adverse out-
comes related to their deficiencies. Their use on a larger scale should be considered to
improve the survival and decrease morbidity and mortality in children and women.
Micronutrient deficiencies are widespread, and are a major global health prob-
lem worldwide [1]. The World Health Organization (WHO) estimates that
more than 2 billion people are deficient in key vitamins and minerals, partic-
ularly vitamin A, iodine, iron and zinc [2]. Most of these people live in low-
income countries and are typically deficient in more than one micronutrient.
The groups most vulnerable to micronutrient deficiencies are pregnant women,
lactating women and young children [3, 4]. Pregnancy represents a state of
Table 1. Reference intake of micronutrients for children [52]
Vitamin RNI 0–3 months RNI 7–9 months RNI 1–3 years Tolerable upper intakes
and 4–6 months or 10–12 months
A, mg 350 350 400 800 mg/day (1–3 years)

D, μg 8.5 7 7 25 μg/day (0–24 months)
E, mg 0.4 mg/g PUFA 0.4 mg/g PUFA 0.4 mg/g PUFA 10 mg/100 kcal of formula
K, μg 10 10 10 not given
B1 (thiamine), mg 0.2 0.2/0.3 0.5 not given
B2 (riboflavin), mg 0.4 0.4 0.6 not given
Niacin (equivalents), mg 3 4/5 8 2 mg/day (1–3 years)
B6 (pyridoxine), mg 0.2 0.3/0.4 0.7 5 mg/day (1–3 years)
B12, μg 0.3 0.4 0.5 not given
Biotin, μg not given not given not given 7.5/100 kcal
Pantothenate, mg 1.7 1.7 1.7 1.2/100 kcal
Folic acid, μg 50 50 70 200 μg/day (1–3 years)
C, mg 25 25 30 30 mg/100 kcal
PUFA = Polyunsaturated fatty acids; RNI = reference nutrient intake. Vitamin A expressed as retinol equivalent (RE): 1
μg RE = 3.33 IU. Vitamin D (calciferol): 1 μg = 40 IU. Vitamin E, α-tocopherol equivalent: 1 mg = 1 IU. Vitamin C expressed
as ascorbic acid.
increased metabolic requirements, and intake of key micronutrients by preg-

nant women especially in developing countries is usually inadequate. This inad-
equate intake and increased requirement during pregnancy further exacerbates
the preexisting maternal deficiency [5]. The resulting micronutrient deficiencies
may lead to potentially adverse effects on the mother such as anemia, hyperten-
sion, complications of labor and even death [6] and adverse perinatal outcomes
such as low birthweight, small for gestational age and preterm birth [3]. The
time from conception until the age of 2 years is the most critical period for any
child. During pregnancy, maternal malnutrition (macro-/micronutrient) can
adversely affect the birth outcomes. After birth, if exclusive breastfeeding is not
practiced during the first 6 months of life or if the solid foods introduced after
that period are nutrient poor, young children are likely to suffer vitamin and
mineral deficiencies [4]. Table 1 shows daily requirement of some of the micro-
nutrients according to different age groups. Figure 1 shows the consequence of
mineral and vitamin deficiencies during the life cycle [7].
Globally, about 1.62 billion people are anemic, with the highest prevalence in
preschool age children (47%) and the second highest in pregnant women (42%).
For pregnant women, over 80% of the countries have a moderate or severe pub-
lic health problem [8]. According to the latest report of the WHO, globally about
190 million preschool age children and 19.1 million pregnant women are vita-
min A deficient (i.e. serum retinol <0.70 μmol/l) [9]. Approximately 100 million
62 Imdad · Bhutta
Baby
• Low birthweight
Elderly • Higher mortality rate
• Increased morbidity • Impaired mental
(osteoporosis, mental development
impairment, etc) • Increased risk of chronic
• Increased mortality disease
Child
• Stunted
• Reduced mental capacity
Inadequate vitamin • Frequent infections
and mineral status • Inadequate growth catch-up
Pregnant • Reduced productivity
Adult women • Higher mortality rate
• Reduced productivity
• Increased mortality
• Poor socioeconomic
• Increased perinatal
status
complications
• Malnourished
• Reduced productivity Adolescent
• Stunted
• Reduced mental capacity
• Fatigue
• Increased vulnerability to
infection
Fig. 1. Consequences of mineral and vitamin deficiencies during the life cycle [7].
women of reproductive age suffer from iodine deficiency [10]. An estimated

82% of pregnant women worldwide have inadequate intakes of zinc to meet the
normative needs of pregnancy [11]. Suboptimal vitamin B6 and B12 status has
been observed in many developing countries [12].
Deficiency of micronutrients is a risk factor for increased incidence and
severity of infectious illness and of dying from diarrhea, measles, malaria and
pneumonia [13]. Data from intervention studies have shown that unsupple-
mented vitamin A-deficient children have increased risk of diarrhea mortality
(RR 1.47, 95% CI: 1.25–1.75) and measles mortality (RR 1.35, 95% CI: 0.96–
1.89) [4]. Similarly, risk associated with zinc deficiency for children below 5
years is RR 1.27 (95% CI: 0.96–1.63) for diarrhea, 1.18 (95% CI: 0.90–1.54) for
pneumonia and 1.11 (95% CI: 0.94–1.30) for malaria [4]. Anemia during preg-
nancy is a risk factor for maternal mortality [14]; however, anemia in child-
hood has not been shown to be associated with an increased risk of mortality
[13]. However, it is well established that anemia during childhood can adversely
affect the developmental outcomes [15]. Evidence suggested that iron deficiency
anemia accounted for a loss of 1–2 intelligence quotient points [16, 17]. Iodine
deficiency has adverse effects on both pregnancy outcome and child develop-
ment, and it has been shown that chronic iodine deficiency can lead to a 13.5
Intervention Strategies to Address Multiple Micronutrient Deficiencies 63

Table 2. Global deaths and disease burden measured in disability-adjusted years (DALYs) in
children <5 years due to micronutrient deficiency in 2004 [4]
Measure Deaths Death in children Disease burden DALYs in children

<5 years, % 1,000 DALYs <5 years, %
Vitamin A deficiency 667,771 6.5 22,668 5.3

Zinc deficiency 453,207 4.4 16,342 3.8
Iron deficiency 20,854 0.2 2,156 0.5
Iodine deficiency 3,619 0.03 2,614 0.6
point reduction in intelligence of the affected population [4]. Table 2 presents

the number of deaths and disability-adjusted life years lost to micronutrient
deficiencies.
Given the significant impact of deficiencies of key micronutrients during
pregnancy and early childhood [4], there has been an increased interest in sup-
plementation with multiple micronutrients, and supplementation with multiple
micronutrients during pregnancy may be a feasible public health strategy [18].
The purpose of this chapter was to summarize the current evidence on strat-
egies of multiple micronutrient supplementations during pregnancy and early
childhood. A literature search was conducted on PubMed, Cochrane library
and WHO/UNICEF databases.
Maternal Micronutrient Supplementation during Pregnancy
Many workers had attempted augmentation of iron-folate supplementation

in pregnancy with additional micronutrients, but the first systematic efforts
to undertake this were almost a decade ago [5]. In 1999, the UNICEF/WHO/
UN University proposed a prenatal supplement UNIMAPP containing fifteen
micronutrients, including iron and folic acid which could provide one recom-
mended daily allowance of each and potentially replace standard iron-folate
supplements for pregnant women in low- and middle-income countries [19].
A Cochrane review on the subject [20] indicated that iron-folate and multiple
micronutrient supplementation had a comparable effect on maternal anemia
and a significant effect on incidence of low birthweight babies and small for
gestational age babies (fig. 2, 3). Another systematic review was undertaken by
a team commissioned by UNICEF/WHO/SCN which analyzed data from 12
trials using the UNIMAPP formulation and evaluated effects on maternal and
pregnancy outcomes [21]. The pooled results had shown that multiple micro-
nutrient supplementation was associated with an increase in mean birthweight
(mean difference 22.4 g, 95% CI: 8.3–36.4 g), a reduction in the prevalence of low
birthweight (odds ratio, OR = 0.89, 95% CI: 0.81–0.97) and small for gestational
64 Imdad · Bhutta
Study or subgroup log[] SE Weight IV, random, 95% CI IV, random, 95% CI
3.3.1 All trials
Bhutta 2009 0.1838 0.19606 3.0% 1.20 [0.82, 1.76]
Christian 2003 –0.1625 0.0674 20.6% 0.85 [0.74, 0.97]
Fawzi 2007 –0.1863 0.07255 18.3% 0.83 [0.72, 0.96]
Friis 2004 –0.3011 0.2537 1.8% 0.74 [0.45, 1.22]
Gupta 2007 –0.9675 0.3026 1.3% 0.38 [0.21, 0.69]
Kaestel 2005 –0.1278 0.1954 3.1% 0.88 [0.60, 1.29]
Osrin 2005 –0.2876 0.1138 8.5% 0.75 [0.60, 0.94]
Ramakrishnan 2003 –0.04082 0.2571 1.8% 0.96 [0.58, 1.59]
Roberfroid 2008 –0.09431 0.17166 3.9% 0.91 [0.65, 1.27]
SUMMIT 2008 –0.1508 0.0819 15.0% 0.86 [0.73, 1.01]
Sunawang 2009 0.1988 0.2027 2.8% 1.22 [0.82, 1.81]
Tofail 2008 –0.1508 0.09778 11.1% 0.86 [0.71, 1.04]
Zagre 2007 –0.1392 0.1487 5.2% 0.87 [0.65, 1.16]
Zeng 2008 –0.1054 0.1797 3.6% 0.90 [0.63, 1.28]
Subtotal (95% CI) 100.0% 0.86 [0.79, 0.92]
Heterogeneity: Tau2 = 0.00; Chi2 = 15.55, df = 13 (P = 0.27); I2 = 16%

Test for overall effect: Z = 4.14 (P < 0.0001)
Fig. 2. Multiple micronutrient supplements during pregnancy: effect on incidence of

low- birthweight babies [20].
Risk ratio Risk ratio

Study or subgroup log [risk ratio] SE Weight IV, random, 95% CI IV, random, 95% CI
3.2.1 All trials
Bhutta 2009 –0.3147 0.3223 0.7% 0.73 [0.39, 1.37]
Christian 2003 –0.0726 0.0471 33.7% 0.93 [0.85, 1.02]
Fawzi 2007 –0.2357 0.0617 19.6% 0.79 [0.70, 0.89]
Friis 2004 –0.2107 0.2347 1.4% 0.81 [0.51, 1.28]
Gupta 2007 –0.51082 0.2068 1.7% 0.60 [0.40, 0.90]
Kaestel 2005 –0.2744 0.255 1.1% 0.76 [0.46, 1.25]
Osrin 2005 –0.2357 0.1928 2.0% 0.79 [0.54, 1.15]
Ramakrishnan 2003 –0.1508 0.22806 1.4% 0.86 [0.55, 1.34]
Roberfroid 2008 –0.1863 0.1247 4.8% 0.83 [0.65, 1.06]
SUMMIT 2008 –0.0305 0.0787 12.1% 0.97 [0.83, 1.13]
Sunawang 2009 –0.1392 0.16468 2.8% 0.87 [0.63, 1.20]
Tofail 2008 –0.1053 0.1355 4.1% 0.90 [0.69, 1.17]
Zagre 2007 –0.1984 0.1344 4.1% 0.82 [0.63, 1.07]
Zeng 2008 –0.1165 0.0843 10.5% 0.89 [0.75, 1.05]
Subtotal (95% CI) 100.0% 0.87 [0.83, 0.92]

Test for overall effect: Z = 4.97 (P < 0.00001)
Fig. 3. Multiple micronutrient supplements during pregnancy: effect on incidence of

small for gestational age babies [20].

age birth (OR = 0.90, 95% CI: 0.82–0.99), and an increase in the prevalence of
large for gestational age birth (OR = 1.13, 95% CI: 1.00–1.28). There were no
significant effects of multiple micronutrient supplementation on birth length,
head circumference or the duration of gestation [22].
Recently, there has been interest in prenatal food supplementation with both
macro- and multiple micronutrients. A randomized trial from Burkina Faso has
shown that combined supplementation with balanced protein energy macronu-
trient and multiple micronutrient has more pronounced effect on birth length
compared to multiple micronutrients alone [23]. An effort should be made to
reproduce results of this study in other parts of the developing world as defi-
ciencies of macro- and micronutrients coexist.
Prenatal Multiple Micronutrient Supplementation in HIV-Affected

Populations
Observational studies have demonstrated an association with low biochemical and

dietary levels of micronutrients and increased risk of mother to child transmission
(MTCT) in HIV-infected mothers [24–26]. However, a review of randomized tri-
als by Mills et al. [27] showed conflicting evidence without strong evidence of ben-
efit or harm on micronutrients. The review included three trials on vitamin A, of
whom two suggested no difference in MTCT, while the third and largest trial sug-
gested an increased risk of MTCT (RR 1.35, 95% CI: 1.11–1.66, p = 0.009). Two of
the vitamin A trials addressed the impact of supplementation on preterm delivery;
one suggested a benefit (RR 0.65, 95% CI: 0.44–0.94) and the other no difference.
Two of the included trials looked at multivitamin use, and one reported data on
MTCT with a non-significant relative risk of 1.04 (95% CI: 0.82–1.32). There was
no significant effect on preterm delivery or child mortality at 1 year [27].
Multiple Micronutrient Supplementation among Children at Risk of

Deficiencies
As deficiencies of important micronutrients like iron, zinc vitamin A, etc. are

prevalent in children in developing countries, efforts have been made to sup-
plement infants and children with multiple micronutrients [28]. A review by
Ramakrishnan et al. [29] based on 20 randomized trials has shown that multiple
micronutrient interventions improved linear growth (effect size 0.09; 95% CI:
0.008–0.17). Another review by Allen et al. [28] has shown that in children,
multiple micronutrient interventions resulted in small but significantly greater
improvements in length or height (effect size 0.13; 95% CI: 0.055–0.21) and
weight (effect size 0.14; 95% CI: 0.029–0.25), hemoglobin (effect size 0.39; 95%
CI: 0.25–0.53), serum zinc (effect size 0.23; 95% CI: 0.18–0.43), serum retinol
66 Imdad · Bhutta
(effect size 0.33; 95% CI: 0.050–0.61) and motor development. In addition to
these benefits, multiple micronutrients have a beneficial effect on mental devel-
opment of children. A review by Eilander et al. [30] has shown that multiple
micronutrient supplementation during childhood has a significant effect on
academic performance (effect size 0.30 SD, 95% CI: 0.01–0.58). There was how-
ever no significant effect on fluid intelligence (effect size 0.14 SD, 95% CI: –0.02
to 0.29) or crystallized intelligence (effect size –0.03 SD, 95% CI: –0.21 to 0.15).
Delivery of Multiple Micronutrients to Pregnant Women and Children
Several strategies have been applied to deliver multiple micronutrients to preg-

nant women and children [31, 32]. Table 3 gives composition of some of the
products delivering multiple micronutrients [33]. Sprinkles® are the most exten-
sively studies product among these, and consist of sachets containing micronu-
trients in a powdered form, to be sprinkled onto a portion of food just before
it is consumed [34, 35]. Crushable or chewable tablets (Foodlets) are multiple
micronutrient tablets that can be easily dissolved in a small amount of liquid,
or crushed and added to foods [36]. The other micronutrients packages in table
3 have been used at different location; however, the efficacy of some of them
needs to be tested in randomized trials.
In pregnant women, different approaches have been used for increasing
maternal intake of multiple micronutrients [37]. One of them is to improve
dietary quality, which in many situations might require increasing consumption
of animal source foods, fruits and vegetables.
In some situations, well-designed nutrition education programs can improve
dietary quality and pregnancy outcome [38]. Other approach is to provide mul-
tiple micronutrient supplements (described above) to women during pregnancy.
In the Cochrane review on the topic that included 17 trials, 7 trials enrolled
participants in the first trimester of pregnancy, one when participants were <28
weeks of gestation, 3 trials enrolled in the second trimester, 2 trials enrolled in
both second and third trimester, whereas 3 trials enrolled pregnant women who
were less than 37 weeks of gestation [20]. In all the included studies, micronutri-
ents were given in the form of supplements, except one study that used fortifica-
tion for the delivery of micronutrients.
In children, the most commonly used approach is the home fortification of
complementary foods. Dewey et al. [31] has reviewed home fortification strate-
gies of complementary foods. Home fortification was highly effective at reduc-
ing iron deficiency and decreased the prevalence of anemia by half. There was
no effect on growth with micronutrients only; however, there was a significant
effect when micronutrients were combined with a small amount of energy
(including fat and protein). These results were based on the pooled data from
two efficacy trials in Africa [39, 40]. Home fortification also had a beneficial

Table 3. Composition of different multiple micronutrient supplements for use in children
Plumpy’Nut B51 Sprinkles
micronutrients amount micronutrients amount micronutrients amount
Vitamin A 910 μg Vitamin B1 0.52 mg Vitamin A 300 μg

Vitamin D 16 μg Vitamin B2 0.52 mg Vitamin D 5 μg
Vitamin E 20 mg Niacin 6.5 mg Vitamin E 6 IU
Vitamin C 53 mg Vitamin B6 0.87 mg Vitamin B1 0.5 mg
Vitamin B1 0.6 mg Vitamin B12 1.3 μg Vitamin B2 0.5 mg
Vitamin B2 1.8 mg Vitamin C 40 mg Niacin 6 mg
Vitamin B6 0.6 mg Vitamin A 1567 IU Vitamin B6 0.5 mg
Vitamin B12 1.8 μg Vitamin D 172 IU Vitamin B12 0.9 μg
Vitamin K 21 μg Vitamin E 3.5 mg Vitamin C 30 mg
Biotin 65 μg Folic acid 130 μg Folic acid 160 μg
Folic acid 210 μg Pantothenic acid 2.2 mg Iron 12.5 mg
Pantothenic acid 3.1 mg Biotin 62.5 μg Zinc 5 mg
Niacin 5.3 mg Calcium 600 mg Copper 0.3 mg
Calcium 320 mg Magnesium 60 mg Iodine 90 μg
Phosphorus 394 mg Iron 10 mg
Potassium 1,111 mg Phosphorus 600 mg
Magnesium 92 mg Zinc 10 mg
Zinc 14 mg Iodine 100 μg
Copper 1.78 mg
Iron 11.53 mg
Iodine 110 μg
Sodium < 290 mg
Selenium 30 μg
1
Efficacy not proven in trials.
impact on morbidity in high-risk populations in some studies; however, there

was no overall significant impact. Acceptability of home fortification by caregiv-
ers and young children was high, and side effects were rare [31].
Side Effects
There is considerable debate on the potential adverse effects of providing mul-

tiple micronutrient supplements during pregnancy with a concern about an
increase in neonatal mortality in less developed health systems that have sub-
optimal maternal care [41, 42]. There are a few studies that reported mortality
beyond the neonatal period with maternal multiple micronutrient supplementa-
tion. Christian et al. [43] reported infant deaths (0–3 months) in the multiple
68 Imdad · Bhutta
Table 3. Continued
Foodlet Unimix1 Anuka1
micronutrients amount micronutrients amount micronutrients amount
Vitamin A 375 μg Vitamin B1 2.8 mg Vitamin A 300 IU

Vitamin D 5 μg Vitamin B2 8.2 mg Vitamin C 10 mg
Vitamin E 6 mg Niacin 50 mg Iron 12 mg
Vitamin B1 0.5 mg Vitamin B6 1.65 mg
Vitamin B2 0.5 mg Pantothenic acid 28 mg
Niacin 6 mg folic acid 2 mg
Vitamin B6 0.5 mg Vitamin B12 13 μg
Vitamin B12 0.9 μg Vitamin C 600 mg
Vitamin C 35 mg Vitamin A 23,000 IU
Folic acid 150 μg Vitamin D 2,000 IU
Iron 10 mg Vitamin E 75 IU
Zinc 10 mg Calcium 2,600 mg
Copper 0.6 mg Zinc 120 mg
Iodine 59 μg Iron 80 mg
micronutrient recipients versus controls (receiving vitamin A only) with a non-

significant higher risk in the micronutrient group (RR = 1.07; 95% CI: 0.75–1.58).
In contrast, Shankar et al. [44] in the considerably larger SUMMIT trial, reported
a statistically significant 18% reduction in early infant mortality (0–3 months)
in the multiple micronutrient group compared to iron-folate (RR = 0.82; 95%
CI: 0.70–0.95; p = 0.010), with comparable results for postneonatal mortality –
from 29 to 90 days after birth (RR = 0.70; 95% CI: 0.55–0.89). A pooled analysis
in a review for Live Saved Tool [45] showed a non-significant overall effect on
neonatal mortality (RR = 1.17; 95% CI: 0.95–1.44). However, a subgroup analy-
sis based on provision of skilled attendance at birth showed that there was an
increased risk of neonatal mortality in places where most of the deliveries were
conducted at home (fig. 4). There was no increased risk in settings where the
majority of births took place at facilities. These findings suggest that the use of

Risk ratio Risk ratio
Study or subgroup log [risk ratio] SE Weight IV, random, 95% CI Year IV, random, 95% CI
4.14.1 Facility-based births greater than or equal to 60%
Kaestel 0.1655 0.2964 9.3% 1.18 [0.66, 2.11] 2005
Osrin 0.4318 0.3809 6.3% 1.54 [0.73, 3.25] 2005
Sunawang –0.2877 0.3336 7.8% 0.75 [0.39, 1.44] 2008
Zeng 0.1398 0.3401 7.5% 1.15 [0.59, 2.24] 2008
SUMMIT –0.1053 0.08626 28.7% 0.90 [0.76, 1.07] 2008
Subtotal (95% CI) 59.5% 0.94 [0.81, 1.09]
Test for overall effect: Z = 0.82 (P = 0.41)
4.14.2 Facility-based births less than 60%
Christian 0.3988 0.2324 13.0% 1.49 [0.94, 2.35] 2003
Tofail 0.2927 0.2827 10.0% 1.34 [0.77, 2.33] 2008
Robertfroid 0.7466 0.5077 3.8% 2.11 [0.78, 5.71] 2008
Bhutta 0.3646 0.2231 13.7% 1.44 [0.93, 2.23] 2009
Subtotal (95% CI) 40.5% 1.47 [1.13, 1.92]
Test for overall effect: Z = 2.87 (P = 0.004)
Total (95% CI) 100.0% 1.17 [0.95, 1.44]
Test for overall effect: Z = 1.52 (P = 0.13) 0.01 0.1 1 10 100
Favours experimental Favours control
Fig. 4. Multiple micronutrient supplements during pregnancy: effect on neonatal mor-

tality [45].
multiple micronutrient supplements in populations to address maternal anemia

and reduce the incidence of small for gestational age must be accompanied by
the provision of skilled care at delivery and facility births to offset any potential
increase in the risk of obstructed labor and birth asphyxia [46, 47].
As different strategies are being evaluated to supplement children with iron
to prevent anemia, there are concerns about increased risk of severity of infec-
tious disease in the presence of malaria and/or undernutrition [48]. Given the
reported increase in malaria-associated morbidity following the use of large-
scale iron supplementation in Zanzibar, the entire rationale for blanket iron
supplements in children has been questioned [49]. Nutrient interactions, espe-
cially in complex formulations, are another area for study as they may hinder
absorption of some nutrients [50].
Conclusions
Multiple micronutrient supplementation during pregnancy and early child-

hood in populations at risk is an effective way of prevention of micronutrient
deficiencies. Although preparations and modes of administration differ and
there are uncertainties around interactions, such strategies have protective
effects against adverse outcomes associated with micronutrient deficiencies.
70 Imdad · Bhutta
Further studies are needed on cost-effectiveness and delivery strategies in
health system settings.
References
1 Tulchinsky TH: Micronutrient deficiency 13 Bhutta ZA: Micronutrient needs of malnour-
conditions: global health issues. Public ished children. Curr Opin Clin Nutr Metab
Health Rev 2010;32:243–255. Care 2008;11:309–314.
2 World Health Organization. World Health 14 Allen LH: Anemia and iron deficiency:
Report. Geneva, World Health Organization, effects on pregnancy outcome. Am J Clin
2000. Nutr 2000;71(suppl 5):1280S–1284S.
3 Black RE: Micronutrients in pregnancy. Br J 15 Olney DK, Pollitt E, Kariger PK, Khalfan SS,
Nutr 2001;85(suppl 2):S193–S197. et al: Combined iron and folic acid supple-
4 Black RE, Allen LH, Bhutta ZA, et al: mentation with or without zinc reduces
Maternal and child undernutrition: global time to walking unassisted among Zanzibari
and regional exposures and health conse- infants 5- to 11-month old. J Nutr 2006;
quences. Lancet 200819;371:243–260. 136:2427–2434.
5 Huffmann SL, Baker J, Schumann J, Zehner 16 Lozoff B, Jimenez E, Smith JB: Double bur-
ER: The Case for Promoting Multiple den of iron deficiency in infancy and low
Vitamin/Mineral Supplements for Women of socioeconomic status: a longitudinal analysis
Reproductive Age in Developing Countries. of cognitive test scores to age 19 years.
Washington, The LINKAGES Project, 1998. Arch Pediatr Adolesc Med 2006;169:1108–
6 Ramakrishnan U, Manjrekar R, Rivera J, et 1113.
al: Micronutrients and pregnancy outcome. 17 Sachdev H, Gera T, Nestel P: Effect of iron
Nutr Res 1999;19:103–159. supplementation on mental and motor devel-
7 United Nations Administrative Committee on opment in children: systematic review of
Coordination Sub-Committee on Nutrition randomised controlled trials. Public Health
(ACC/SCN): Fourth Report on the World Nutr 2005;8:117–132.
Nutrition Situation. Geneva, International 18 Bhutta ZA, Ahmed T, Black RE, et al: What
Food Policy Research Institute, 2000. works? Interventions for maternal and child
8 World Health Organization: Worldwide undernutrition and survival. Lancet 2008;
Prevalence of Anaemia 1993–2005. WHO 71:417–440.
Global Database on Anaemia Information. 19 UNICEF, WHO, and UNU: Composition
Geneva, World Health Organization, 2008. of a Multi-Micronutrient Supplement to Be
9 World Health Organization: Global Used in Pilot Programmes among Pregnant
Prevalence of Vitamin A Deficiency in Women in Developing Countries. New York,
Populations at Risk 1995–2005. WHO Global UNICEF, 1999.
Database on Vitamin A Deficiency. Geneva, 20 Haider BA, Bhutta ZA: Multiple-micro-
World Health Organization, 2009. nutrient supplementation for women during
10 Leslie J: Women’s nutrition: the key to pregnancy. Cochrane Database Syst Rev
improving family health in developing coun- 2006;CD004905.
tries? J Health Policy Plan 1991;6:1. 21 Dalmiya N, Darnton-Hill I, Schultink W,
11 Caulfield LE, Zavaleta N, Shankar AH, Shrimpton R (eds): Multiple Micronutrient
Merialdi M: Potential contribution of mater- Supplementation during Pregnancy in
nal zinc supplementation during pregnancy Developing Country Settings. Nutrition
to maternal and child survival. Am J Clin Policy Paper No 22. Geneva, UN Standing
Nutr 1998;68(suppl 2):499S–508S. Committee on Nutrition, 2009.
12 McLean E, de Benoist B, Allen LH: Review
of the magnitude of folate and vitamin B12
deficiencies worldwide. Food Nutr Bull 2008;
29(suppl 2):S38–S51.

22 Fall CH, Fisher DJ, Osmond C, Margetts 32 Sazawal S, Dhingra U, Dhingra P, et al:
BM: Multiple micronutrient supplemen- Micronutrient fortified milk improves iron
tation during pregnancy in low-income status, anemia and growth among children
countries: a meta-analysis of effects on birth 1–4 years: a double masked, randomized,
size and length of gestation. Food Nutr Bull controlled trial. PLoS One 2010;5:e12167.
2009;30(suppl 4):S533–S546. 33 http://www.idpas.org/index.html.
23 Huybregts L, Roberfroid D, Lanou H, et 34 de Pee S, Kraemer K, van den Briel T, et al:
al: Prenatal food supplementation fortified Quality criteria for micronutrient powder
with multiple micronutrients increases birth products: report of a meeting organized by
length: a randomized controlled trial in rural the World Food Programme and Sprinkles
Burkina Faso. Am J Clin Nutr 2009;90:1593– Global Health Initiative. Food Nutr Bull
1600. 2008;29:232–241.
24 Dreyfuss ML, Fawzi WW: Micronutrients 35 Young SL, Blanco I, Hernandez-Cordero S:
and vertical transmission of HIV-1. Am J Organoleptic properties, ease of use, and
Clin Nutr 2002;75:959–970. perceived health effects are determinants
25 Semba RD, Miotti PG, Chiphangwi JD, of acceptability of micronutrient supple-
et al: Maternal vitamin A deficiency and ments among poor Mexican women. J Nutr
infant mortality in Malawi. J Trop Pediatr 2010;140:605–611.
1998;44:232–234. 36 Smuts CM, Lombard CJ, Benade AJ, et
26 Semba RD, Miotti PG, Chiphangwi JD, et al: Efficacy of a foodlet-based multiple
al: Infant mortality and maternal vitamin A micronutrient supplement for preventing
deficiency during human immunodeficiency growth faltering, anemia, and micronutri-
virus infection. Clin Infect Dis 1995;21:966– ent deficiency of infants: the four country
972. IRIS trial pooled data analysis. J Nutr
27 Mills EJ, Wu P, Seely D, Guyatt GH: Vitamin 2005;135:631S–638S.
supplementation for prevention of mother- 37 Allen LH: Multiple micronutrients in preg-
to-child transmission of HIV and pre-term nancy and lactation: an overview. Am J Clin
delivery: a systematic review of randomized Nutr 2005;81:1206S–1212S.
trial including more than 2800 women. AIDS 38 Long VA, Martin T, Janson-Sand C: The
Res Ther 2005;2:4. great beginnings program: impact of a nutri-
28 Allen LH, Peerson JM, Olney DK: Provision tion curriculum on nutrition knowledge,
of multiple rather than two or fewer diet quality, and birth outcomes in preg-
micronutrients more effectively improves nant and parenting teens. J Am Diet Assoc
growth and other outcomes in micronutri- 2002;102(suppl 3):S86–S89.
ent deficient children and adults. J Nutr 39 Adu Afarwuah S, Lartey A, Brown KH, et al:
2009;139:1022–1030. Randomized comparison of 3 types of micro-
29 Ramakrishnan U, Nguyen P, Martorell R: nutrient supplements for home fortification
Effects of micronutrients on growth of chil- of complementary foods in Ghana: effects on
dren under 5 y of age: meta-analyses of single growth and motor development. Am J Clin
and multiple nutrient interventions. Am J Nutr 2007;86:412–420.
Clin Nutr 2009;89:191–203. 40 Kuusipalo H, Maleta K, Briend A, et al:
30 Eilander A, Gera T, Sachdev HS, et al: Growth and change in blood haemoglobin
Multiple micronutrient supplementation concentration among underweight Malawian
for improving cognitive performance in infants receiving fortified spreads for
children: systematic review of random- 12 weeks: a preliminary trial. J Pediatr
ized controlled trials. Am J Clin Nutr Gastroenterol Nutr 2006;43:525–532.
2010;91:115–130. 41 Bhutta ZA, Haider BA: Prenatal micronutri-
31 Dewey KG, Yang Z, Boy E: Systematic review ent supplementation: are we there yet? CMAJ
and meta-analysis of home fortification of 2009;180:1188–1189.
complementary foods. Matern Child Nutr 42 Bhutta ZA, Haider BA: Maternal micronu-
2009;5:283–321. trient deficiencies in developing countries.
Lancet 2008;371:186–187.
72 Imdad · Bhutta
43 Christian P, West KP, Khatry SK, et al: Effects 47 Lee AC, Darmstadt GL, Khatry SK, et al:
of maternal micronutrient supplementation Maternal-fetal disproportion and birth
on fetal loss and infant mortality: a cluster- asphyxia in rural Sarlahi, Nepal. Arch Pediatr
randomized trial in Nepal. Am J Clin Nutr Adolesc Med 2009;163:616–623.
2003;78:1194–1202. 48 Oppenheimer SJ: Iron and its relation to
44 Shankar AH, Jahari AB, Sebayang SK, et al: immunity and infectious disease. J Nutr
Effect of maternal multiple micronutrient 2001;131(suppl 2):616S–633S, discussion
supplementation on fetal loss and infant 633S–635S.
death in Indonesia: a double-blind cluster- 49 Sazawal S, Black RE, Ramsan M, et al: Effects
randomised trial. Lancet 2008;371:215–227. of routine prophylactic supplementation with
45 Haider BA, Yakoob MY, Imdad A, Bhutta iron and folic acid on admission to hospital
ZA: Effect of multiple micronutrient supple- and mortality in preschool children in a high
mentation during pregnancy on maternal malaria transmission setting: community-
and birth outcomes. BMC Public Health based, randomised, placebo-controlled trial.
2011;11(suppl 3):S19. Lancet 2006;367:133–143.
46 Christian P, Khatry SK, LeClerq SC, Dali SM: 50 Ramakrishnan U, Neufeld LM, Flores R, et
Effects of prenatal micronutrient supplemen- al: Multiple micronutrient supplementation
tation on complications of labor and delivery during early childhood increases child size at
and puerperal morbidity in rural Nepal. Int J 2 y of age only among high compliers. Am J
Gynaecol Obstet 2009;106:3–7. Clin Nutr 2009;89:1125–1131.

Summary on Micronutrient
Requirements and Deficiencies in
Maternal and Child Nutrition
Following on the fabulous summary from Wafaei Fawzi, the important thing
to take home is that some of these interventions or potential strategies are
really quite interconnected. I don’t know how many of you might have seen a
recent analysis of the factors that determined how countries have been able to
reduce their child mortality in the last 3–4 decades. This analysis by Prof. Chris
Murray’s team at the Institute for Health Metrics and Evaluation evaluated a
range of factors influencing child mortality trends over time and determined
that the strongest determinant amongst all affecting child mortality change was
maternal education. It actually contributed to 54% of the total reduction that
was seen across a huge range of data sets in countries. So I can’t underscore
more the importance of maternal education and supporting something as fun-
damental as this not just for improving health and nutrition but from a develop-
ment perspective.
This has been an exceptionally informative day. This was just the opening
innings of the entire micronutrient workshop, and I am sure there will be much
more. Today, we focused principally on setting the scene and addressing some
of the key issues around micronutrients; what is it that we do and don’t know?
We began by looking at the landscape in terms of what exists in relation to bur-
den data of deficiencies, key micronutrients and then moved on to very specific
nutrients and then combinations thereof.
I am not going to repeat the points that have already been made, but one
of the key messages from the first session and the latter part of the afternoon
was the importance of getting good representative data. It is probably a fair
summation that on the interface of maternal newborn and child nutrition as
well as adolescent nutritional status, our information and data bases are very
skimpy, and therefore a concerted effort needs to be made across the public
health community to try and strengthen this database. It is also a responsibil-
ity of the nutrition industry working in areas of global nutrition to try and see
how they can help plug some of those data gaps, because these are clearly not
only of public health importance; they also have importance in terms of their
own R&D for products that may be of value to public health and nutrition
outcomes.
A large proportion of the discussion this morning focused on the relation-
ship of zinc to growth and outcomes, and I am going to just very briefly outline
the things that I have picked up in areas that you may consider in your delib-
erations of the next couple of days, and these are not necessarily in the order
of the sessions or the presentations. The first important issue was the need for
much better information around zinc requirements, and we were really pleased
to hear from Michael Hambidge that there has been progress in terms of rec-
onciling the desperate recommendations between two large technical groups
looking at this issue, one that I sit on as well, and the other being the Institute
of Medicine and some of the WHO people. And it’s great to see this coming
together. We also heard how uncertain we currently are in trying to match the
science of stable isotope methodology with some of the fundamental data gaps
around requirements and around the whole issue of status. Several amongst
the audience pointed out to the difficulty that we have in assessing zinc at the
population level given the poor association of plasma or serum zinc with status
measurements. This being such an important issue, I would make the case that
we should focus our efforts on trying to come up with measures of assessing
zinc status and requirements thorough easily applicable protocols that can be
implemented in developing country settings, and I know that there are some
that are currently being planned. We had quite a lot of discussion in various
presentations on the heterogeneity of the data around zinc and growth, par-
ticularly around preventive zinc supplementation, and these data are probably
amenable to further analysis using some of the approaches that we had in the
multiple micronutrient field. You have also seen how several groups working
independently have been able to bring the science together, and approaching
the issue completely independent of each other have come up with very compa-
rable conclusions. It is also gratifying to see this because it tells you exactly what
needs to be done around zinc too. So, rather than just do pure meta-analysis and
come up with perfunctory forest plots, we need robust meta-regressions across
studies to try and understand some of the contextual differences. I believe we
need to control analysis across studies for factors such as intrauterine growth
retardation and the proportion of those children who start off with deficits that
may already be too late to change. Today’s presentations and discussions have
raised a number of ideas as to how we can try and shed more light on the zinc
and growth literature both in terms of understanding it and also importantly
determine future studies. One issue that came up in the discussion around zinc
was that we do not necessarily have a good vehicle for administration, that there
are serious concerns that Sprinkles or micronutrient powders may not be an
appropriate strategy for administering zinc because we have so limited data on
Summary on Micronutrient Requirements and Deficiencies in Maternal and Child Nutrition 75

bioavailability of zinc therefrom. So, I certainly think a lot of attention needs to
be focused on getting those information gaps resolved.
Robert Black presented findings from the zinc mortality studies on potential
interactions of zinc and iron. This is a hugely important issue that needs to be
resolved perhaps employing stable isotope methodology and studies that also
look at some of the potential mechanisms that may exist for such interaction.
There were some uncertainties which were also highlighted around the infor-
mation which should be priorities areas for research. One of those is the impact
of zinc on both the treatment of diarrhea and other biological outcomes in the
prevention mode in relatively zinc-replete populations. This is an issue which is
increasingly highlighted because of the negative studies in Europe recently and
some of the previous studies from Australia which have failed to show an effect.
I do think we have to address this, and perhaps best through well-designed pro-
spective trials to answer the question if there are populations which will not
benefit from zinc supplementation. Secondly, given the burden of morbidity
and mortality therein, the whole issue of younger infants under 6 months is
important. I think the data do point out that zinc therapy for diarrhea is not
effective in this population, but our understanding of the reasons why is inex-
act. This information gap does cast a shadow on the entire zinc and diarrhea
treatment literature, so we need studies that also intercalate some mechanistic
information.
There were lots of questions raised on the lack of information on diarrhea
etiology in relation to response to zinc: Should we be giving zinc for treatment
of diarrhea in a background of preventive zinc dosing? Does fortification work
as a strategy, so if you have got fortified foods coming in, do you really need
additional zinc supplementation? I don’t have the answers to those questions,
but they are all very important and need to be addressed in a robust manner.
In particular, the role of Sprinkles as a vehicle for supplying zinc is an issue, not
just because my group works on them, but because there is now a global rollout
of micronutrient powders and Sprinkles. If there is a question mark in terms of
availability of zinc from Sprinkles, we need to resolve that very quickly. I was
pleased to learn that there is a study going on in Kenya using methodologies to
address whether Sprinkles with zinc and Sprinkles with zinc and iron are com-
parably effective. These are high-priority studies, and they probably need to be
done with diverse dietary background so that we know how much of the issue is
diet and how much is actually the bioavailability from Sprinkles.
The other issue that did come up this morning and is not really partitioned
in the context of zinc or others is this whole issue of the confusion that we cur-
rently have of what to do for acute versus chronic malnutrition. When we did
the Lancet series, a lot of emphasis was put on using stunting as a measure of
undernutrition because we felt from robust analysis that it was a much better
marker of undernutrition than the currently available MDG1 target of under-
weight. We also recognized at the time when we did the analysis that if you
76 Bhutta
looked at the overlap of severe acute malnutrition and stunting, the overlap was
relatively minimal, and therefore conceptually as a public health indicator stunt-
ing and severe acute malnutrition would allow us to target those populations
at risk appropriately, one being a much more chronic process, the other being
an acute process that you targeted with therapeutic feeding. I think the confu-
sion that has arisen has been largely around now putting up the moderate acute
malnutrition category which overlaps to a fairly large extent with stunting, and
therefore countries and practitioners are getting very confused whether moder-
ate acute malnutrition also represents acute malnutrition or there are subgroups
within it with chronic malnutrition. I would make a plea to the nutritionists
who are also on a public health nutrition platform to try and harmonize this
because it’s causing enormous confusion at scale. I just got a message from a col-
league in Pakistan who is involved in this controversy right now, and it’s causing
huge policy confusion whether we should be addressing moderate acute malnu-
trition in populations with therapeutic feeding at scale. This is not a trivial issue,
and I would underscore that as something that came up today, and we should
definitely keep it on the table.
Lastly, I think the whole multiple micronutrient session despite the obvious
overlap between Ian Darnton-Hill’s and my presentation was an extremely illu-
minating one for me personally because we have got a lot of feedback from the
group. I would like to underscore a few things. We don’t need more small-scale
studies in this area. I think what we need is larger effectiveness kind of evalua-
tion of this intervention layered on top of fairly robust in-depth scientific evalu-
ation of what multiple micronutrients mean, what kind of deficiencies we are
tackling and what kind of outcomes we should be preparing for. If the malaria
community can raise 30–40 million USD to address malaria issues and response
to treatment with some of the best genomics platforms that you can find, I can’t
understand why we, as a nutrition community, cannot address this huge issue
that is associated with 18–20 million low birthweight births every year. There
was also an understandable note of caution of moving carefully before we make
this a programmatic priority because of potential questions that we haven’t
answered, and I entirely agree with this. I think if large-scale effectiveness evalu-
ations around some of the multiple micronutrient questions can be designed,
the safety elements can be easily answered.
Zulfiqar Bhutta
Summary on Micronutrient Requirements and Deficiencies in Maternal and Child Nutrition 77

Vitamin A Supplementation, Infectious

Disease and Child Mortality: A Summary
of the Evidence
Andrew Thorne-Lymana ⭈ Wafaie W. Fawzia–c
Departments of aNutrition, bEpidemiology and cGlobal Health and Population, Harvard School of
Public Health, Boston, MA, USA
Abstract
This manuscript reviews the evidence related to the effects of vitamin A (VA) supplemen-
tation of women and children on child health and mortality. VA supplementation of
children aged 6–59 months has been well studied, and meta-analyses have consistently
demonstrated effects on all-cause mortality, yet its mechanisms and the reasons for het-
erogeneous effects on mortality across trials continue to be debated. Recent meta-analy-
sis of cause-specific mortality suggests beneficial effects on diarrheal mortality, with null
but potentially beneficial effects also present for mortality from measles, lower respira-
tory infection, and meningitis. Some evidence suggests that pneumonia severity may
increase with VA supplementation in this age group, particularly among well-nourished
children. Maternal supplementation with VA during pregnancy has not shown benefits on
neonatal mortality in large trials. A recent meta-analysis suggested that high-dose sup-
plementation of lactating women immediately following delivery did not affect child sur-
vival. There is still uncertainty around the benefits of neonatal VA supplementation that
should be resolved once the findings of ongoing trials are reported.
Introduction
Next year will mark the 100th anniversary of the isolation of ‘fat soluble A’ by
McCollum and Davis [1]. Although interest in vitamin A (VA) waned over much
of the subsequent century, a renewal of interest in the 1980s led to numerous
randomized trials testing the efficacy of VA supplementation on adverse out-
comes among preschool age children, infants and neonates, and pregnant and
lactating women. In this chapter, we provide an overview of the epidemiological
evidence on routine preventive VA supplementation of neonates, infants, and
children and child health outcomes. It is important to note that VA supplemen-
tation has long been demonstrated as efficacious in the treatment of measles and
both treatment and prevention of xerophthalmia, and in the interest of brevity
we have not extensively reviewed that evidence here [2].
Epidemiology and Causes of Vitamin A Deficiency during Childhood
VA deficiency is a problem of greatest public health concern among children less

than 5 years of age and pregnant and lactating women. It has been estimated that
190 million preschool children and 19.1 million pregnant women have low serum
retinol concentrations (<0.70 μmol/l) in countries at risk [3]. Night blindness,
the first stage of clinical VA deficiency, affects 5.2 million preschool age children
and 9.75 million pregnant women in countries at risk. Among the WHO regions,
South East Asia and Africa have the highest prevalence of preschool subclinical
VA deficiency (at 49 and 44%, respectively), although the absolute number of
children at risk is much higher in South East Asia (91 vs. 56 million children) [3].
The mean requirement for VA during early childhood ranges from 180 μg
retinol equivalents (RE)/day for neonates and infants to 200 μg RE/day up to
age 6 years, with recommended safe intake set at 375 and 450 μg RE/day, respec-
tively [4]. In children, clinical symptoms of VA deficiency include night blind-
ness as the first physical manifestation, proceeding to conjunctival and corneal
involvement which then leads to corneal ulceration and scarring resulting in
blindness if uncorrected [2]. It is well accepted that the immune-compromis-
ing effects of VA deficiency manifest long before the physical appearance of
deficiency [2].
VA deficiency during infancy and early childhood is caused by multiple
factors including maternal VA deficiency (and low concentrations of retinol in
breast milk), suboptimal breastfeeding, complementary foods low in VA, and
severe or repeated episodes of infectious disease [5]. Sources of performed VA
include liver, eggs, milk, and fortified foods, all of which are consumed infre-
quently in developing countries. Generally, foods containing provitamin A car-
otenoids such as green leafy vegetables and yellow-orange fruits such as mango
and papaya are thought to provide the majority of dietary VA in such countries.
Following the downward adjustment of conversion factors for provitamin A
carotenoids for estimating intake of VA from plant foods, it has been generally
assumed that such foods contribute relatively little towards meeting VA require-
ments [6–8]. While plant sources of food are unlikely by themselves to fully
meet requirements when consumed in the quantities typical of many develop-
ing country diets, there is some evidence that their consumption may still offer
protection against xerophthalmia, mortality, and growth faltering, particularly
among children not supplemented with VA [9–12].
80 Thorne-Lyman · Fawzi
Supplementation during Pregnancy
Early work by Mellanby and Green reported in 1929 suggested the potential for VA
to reduce puerperal septicemia, although little follow-up research on the potential
for VA to reduce mortality or morbidity among pregnant women occurred over
the next 60 years [13]. More recently, a large randomized controlled trial in Nepal
reported a 44% reduction in maternal mortality associated with daily supplemen-
tation with VA or β-carotene throughout pregnancy [14]. Based on these promis-
ing findings, large confirmatory trials were launched in Bangladesh and Ghana,
neither of which reported significant decreases in pregnancy-related mortality up
to 6 weeks [15, 16]. Pooled analysis of these three trials suggests no overall effect
of supplementation on maternal mortality, though heterogeneous findings have
spurred debate over whether the differences are attributable to chance or context
[17]. One consistent observation across all of these large studies was the lack of
an effect on neonatal mortality, effectively putting to rest the suggestion that VA
supplementation during pregnancy could be an effective child survival initiative
although other potential benefits are being explored [17].
Postpartum Vitamin A Supplementation
Postpartum VA supplementation was originally devised as a strategy to improve

VA status of the mother/infant dyad, and was once implemented as a program
in many countries. A recent meta-analysis examined the effect of maternal post-
partum VA supplementation on infant mortality and morbidity [18]. The analy-
sis reported no significant effect on mortality (RR = 1.05, 95% CI: 0.92–1.20)
or risk of diarrhea or acute respiratory infection [18]. Based on a review of the
evidence, WHO recently issued guidelines stating that VA supplementation of
postpartum women is not recommended for the prevention of maternal and
infant morbidity or mortality (table 1) [19].
Maternal Vitamin A Supplementation and HIV Transmission
Observational studies in the 1990s suggested that low maternal VA status

was associated with greater risk of HIV transmission from mother to child as
well as higher rates of infant mortality [20–22]. Accordingly, four random-
ized placebo-controlled trials exploring various combinations of VA and/or
β-carotene supplementation to women and infants were launched in Tanzania,
Malawi, South Africa, and Zimbabwe [23–26]. The design and results of these
trials are presented in table 2. In a pooled analysis of these trials, antenatal or
postpartum maternal supplementation with VA or VA/BC was found to have no
effect on HIV transmission (RR = 1.04, 95% CI: 0.87–1.24) [27].
Vitamin A Supplementation, Infectious Disease and Child Mortality 81

Table 1. WHO Recommendations for preventive VA supplementation
Population subgroup Recommendations
Pregnant women Not recommended for the prevention of maternal and infant
[53] morbidity and mortality, but in areas where VA deficiency is a severe
public health problem, VA supplementation is recommended for
prevention of night blindness. Quality of evidence: high for maternal
mortality, moderate for other outcomes.
Postpartum women High-dose VA supplementation is not recommended for the

[19] prevention of maternal and infant morbidity and mortality. Quality
of evidence: high for all-cause infant mortality and very low for
cause-specific infant mortality and morbidity.
Neonatal Not recommended as a public health intervention to reduce infant

supplementation [39] morbidity and mortality. Quality of evidence: moderate for
mortality-related outcomes.
Early infants Not recommended as a public health intervention for the reduction
(1–5 months) [54] of morbidity and mortality. Quality of evidence: moderate for infant
mortality, low for other outcomes.
Older infants and Infants 6–11 months of age: 100,000 IU once. Children 12-59
young children 6–59 months of age: 200,000 IU every 4–6 months. Quality of evidence:
months of age high for all-cause mortality, moderate to very low for other
(including HIV+) [49] outcomes.
It is important to note heterogeneity in the content of supplements and

timing of supplementation across these trials. Findings from two of the trials
suggest that VA supplementation may increase the risk of HIV transmission.
The Tanzania trial, which tested the effects of supplementation with VA/BC
during pregnancy and postpartum reported a 38% increase in HIV transmis-
sion or death compared to the control group (RR = 1.38, 95% CI: 1.09–1.76, p =
0.009) [23]. In the Zimbabwe trial, risk of either HIV transmission or mortality
was higher among infants whose mothers received a one-time 400,000 IU dose
or who directly received a 50,000 IU dose as neonates [26]. Paradoxically, the
group that had received both maternal and child supplementation showed no
difference in risk compared with the control group. Yet, among those infants
who were PCR negative at 6 weeks, all three groups including either maternal
postnatal and/or neonatal VA supplementation had approximately double the
mortality risk compared with the placebo group (table 2).
There is some evidence from in vitro and other epidemiological studies sup-
porting the hypothesis that VA could increase transmission [28]. An in vitro
study suggested that VA could increase the expression of CCR5 receptors lead-
ing to increased replication of HIV-1 and increased susceptibility of monocytes
and macrophages to HIV [29]. Evidence from a prospective cohort study in the
Table 2. Randomized controlled trials exploring the effect of VA or VA/BC supplementation on HIV transmission
from mother to child
First author and Setting Design and participants Findings

year
Coutsoudis, South Africa Randomized controlled trial of 728 No difference in risk of HIV infection by 3
1999 [25] HIV-infected women recruited at months of age was observed between the
17–39 weeks’ gestation. Women VA group (20.3% HIV+, 95% CI: 15.7–24.9)
received daily oral dose of (1) VA and the placebo group (22.3% HIV+, 95%
supplement containing 5,000 IU CI: 17.5–27.1), and no significant
retinyl palmitate and 30 mg differences in fetal or infant mortality.
β-carotene or (2) placebo through
delivery. Women in the VA group also
received 200,000 IU VA.
Fawzi, 2002 [23] Tanzania Randomized 2 × 2 factorial trial of The total risk of infection was
800 HIV-1 infected pregnant women significantly higher in the VA group
enrolled from 12–27 weeks gestation. compared with the non-VA group (RR =
Women received: (1) VA + βC (30 mg 1.38, 95% CI: 1.09–1.76, p = 0.009). No
β-carotene plus 5,000 IU preformed significant effect on mortality by 24
VA), (2) multivitamins excluding VA + months (RR = 1.00, 95% CI: 0.80–1.24),
βC, (3) MV including VA + βC or (4) two or on composite HIV infection/mortality
tablets of placebo from enrollment end point (RR = 1.13, 95% CI: 0.94–1.36).
onwards. Women in VA groups also
received 200,000 IU VA at delivery.
Kumwenda, 2002 Malawi Randomized placebo-controlled trial No significant differences between

[24] among 697 HIV-infected pregnant groups in the proportion of HIV-infected
women enrolled at 18–28 weeks’ infants at 6 weeks (VA: 26.6% vs. placebo:
gestation. Women received daily oral 27.8%), 12 months (VA: 27.3% vs.
dose of either (1) VA (10,000 IU; n = placebo: 32.0%), and 24 months (VA:
340) or (2) placebo (n = 357). Women 27.7% vs. placebo: 32.8%), all p values
in both groups received 200,000 IU >0.20.
VA at delivery.
Humphrey, 2006 Zimbabwe Randomized factorial trial among Neither postnatal maternal nor neonatal
[26] 14,110 mother-infant pairs including VA supplementation significantly
(1) maternal supplementation with affected postnatal HIV transmission or
400,000 IU VA administered overall mortality at 24 months, although
postpartum and 50,000 IU VA both group 2 and 3 had increased risk of
administered to the neonate, (2) the infection or death outcome at 12
maternal VA as above with placebo months (40.3 and 40.0%, respectively)
administered to neonate, (3) compared with placebo (35.4%). Among
maternal placebo and 50,000 IU VA PCR-negative infants at baseline, all three
administered to the neonate, (4) VA groups appeared to have heightened
maternal placebo and neonatal risk of mortality vs. placebo, though one
placebo. group narrowly missed statistical
significance (group 1: HR = 2.05, 95% CI:
1.14–3.67, p = 0.02; group 2: HR = 1.82,
95% CI: 0.99–3.31, p = 0.05; group 3: HR =
1.89, 95% CI: 1.05–3.40, p = 0.03)

United States suggested a U-shaped relationship between dietary intake of VA
and HIV disease progression and mortality [30, 31]. Further analysis of breast
milk samples from the Tanzanian trial revealed that VA/β-carotene supple-
mentation was associated with increased risk of subclinical mastitis as well as
increased HIV viral load in breast milk, suggesting a potential mechanism for
increased transmission [32, 33].
Neonatal Vitamin A Supplementation
Two recent meta-analyses have consolidated the findings of studies of VA

supplementation within the first 30 days of life, but reached different conclu-
sions [34, 35]. The first reported no significant effect on mortality risk during
a 12-month follow-up period (6 studies, RR = 0.92, 95% CI: 0.75–1.12) [36].
The second study presented findings disaggregated by time to follow-up, and
found a significant reduction of mortality risk among all infants at 6 months
(5 studies, RR = 0.86; 95% CI: 0.77–0.97) but not at 12 months (4 studies, RR
= 1.02; 95% CI: 0.87–1.20) [35]. The latter paper argued that routine program-
matic VA supplementation reached many children above 6 months in most of
the included study countries, potentially leading to an attenuation of findings
when using a follow-up period exceeding the first 6 months of life. A WHO
technical consultation on neonatal VA supplementation also noted disparate
findings in effects of supplementation by region; studies from Asian contexts
appeared more suggestive of benefit than studies in Africa, although such differ-
ences were not statistically significant in meta-regression [37]. A separate meta-
regression has also suggested that a protective effect appears to be more evident
among neonates supplemented in the first 48 h of life living in areas of prevalent
VA deficiency [38].
At this time, the WHO does not recommend neonatal VA supplementation
[39]. In response to recommendations from a WHO technical consultation on
neonatal supplementation research priorities, trials are currently underway in
Tanzania, Ghana, India, and Pakistan to better understand the potential contex-
tual differences noted above and improve estimates of the efficacy of neonatal
supplementation [37]. More work is also needed to understand potential mech-
anisms through which neonatal supplementation might affect mortality risk; it
has been hypothesized that it may do so through effects on the immune system
or organ maturation [37].
Vitamin A Supplementation among Infants 1–6 Months of Age
In contrast to older (or younger) age groups, there is little evidence that VA
supplementation of infants 1–6 months of age is protective against mortality.
Several studies in Asia, one in Africa, and a multicenter study have been under-
taken, and none showed statistically significant benefit; a recent pooled estimate
was also null (5 studies, RR = 1.05, 95% CI: 0.88–1.26) [40].
Vitamin A Supplementation for 6- to 59-Month-Olds
All-Cause Mortality
The most recent and comprehensive meta-analysis of published studies reported
that VA supplementation of children 6–59 months of age led to a 24% reduction
in all-cause mortality (RR = 0.76, 95% CI: 0.69–0.83), an estimate that remained
largely the same as previous analyses [41, 42]. Heterogeneity in the findings of
included studies was also apparent (I2 = 48%, p = 0.02). Many explanations have
been proposed for heterogeneity across trials including coexisting micronutrient
deficiencies or differences in dietary fat intake (which may influence absorption
or utilization of VA), differences in dosing frequency or amount, and differences
in the incidence/prevalence of infectious disease. In the most recent review,
significant beneficial effects were apparent in both Africa and Asia, among both
boys and girls, and among multiple age strata (6–12 and 12–60 months) [41].
An interesting nuance is that more frequent dosing than the currently recom-
mended WHO frequency of 4–6 months was associated with a 54% reduction
in all-cause mortality (compared with a 19% reduction among those supple-
mented every 4–6 months). This suggests the possibility that at least in some
settings, a more frequent dosing schedule than is currently implemented could
increase the efficacy of the intervention (provided that it could be implemented
with similar coverage rates). Paradoxically, the analysis also found that those
who received less frequent dosing appeared to have greater mortality benefit
than those who received supplementation every 4–6 months, an observation
that cannot be easily explained [41].
Another important recent development in the evidence base related to supple-
mentation of this age group was the presentation of preliminary findings of the
DEVTA trial, the largest randomized controlled trial of VA supplementation to
date (conducted in India and involving over 1 million participants), a trial that
remains unpublished at the time of this publication (2012). Findings available
online suggested that supplementation every 6 months with 200,000 IU had only
a small (RR = 0.96, 99% CI: 0.88–1.05) statistically non-significant benefit on all-
cause mortality [8]. In sensitivity analysis of all trials conducted to date including
these findings, the overall effect estimate of the meta-analysis remained statistically
significant, but the benefit was cut in half (from 24 to 12%) [41]. Some have used
these findings to support arguments questioning the efficacy of VA [43]. Others
have argued that the continued presence of VA deficiency in the VA arm of the
trial (2.2% had Bitot’s spots and 11% had plasma retinol <0.35 μmol/l) is evidence
that the dose may not have been adequate to achieve maximal efficacy [8, 44].

Cause-Specific Mortality and Morbidity
VA has a number of important roles in immunity including maintenance of
epithelial integrity, regulating differentiation and function of monocytes, and
influencing differential Th1/Th2 responses [45]. Early research in the 1930s
generated interest around the ‘anti-infective’ properties of VA, and an early
trial by Ellison demonstrated that VA dramatically lowered mortality rates of
children with measles [1, 46, 47]. Since that time, a number of hospital-based
trials have firmly established the efficacy of VA in the treatment of measles
[45, 48].
In the most recent meta-analysis of epidemiological studies, VA supplemen-
tation significantly reduced the risk of mortality due to diarrhea by 28% (RR
= 0.72, 95% CI: 0.57–0.91) as well as incidence of diarrhea by 15% (13 stud-
ies, RR = 0.85 (95% CI: 0.82–0.87), but no effect on prevalence of diarrhea was
observed (2 studies, RR = 1.08, 95% CI: 1.05–1.12). Based on the relationship
between supplementation and mortality, it has been posited that VA may help to
protect against severe diarrhea, but might not affect more mild forms, although
recent meta-analyses have not disaggregated by severity of the outcome [2]. The
mechanism through which VA affects incidence or severity of diarrhea is uncer-
tain, although the vitamin plays an important role in maintenance of epithe-
lial integrity and in mucus secretion, both of which may help to protect against
diarrhea [48].
It is well documented that measles is a precipitating factor of a great deal of
child xerophthalmia and that supplementation with VA as treatment during a
measles episode strongly reduces the risk of mortality [2]. Recent meta-anal-
ysis of trials of routine preventive supplementation also suggests reductions in
measles-related mortality, although this finding was not statistically significant
(5 trials, RR = 0.80, 95% CI: 0.51–1.24).
In contrast to diarrhea and measles, evidence for an effect of VA supple-
mentation on lower respiratory tract infections including pneumonia has been
greatly inconsistent. The most recent review of the effects of supplementation
on cause-specific morality attributable to lower respiratory tract infections
indicated a non-significant trend toward benefit (7 studies, RR = 0.78, 95%
CI: 0.54–1.14), while finding a non-significant increase in incidence of lower
respiratory tract infections (7 studies, RR = 1.14, 95% CI: 0.95–1.37). Another
recent meta-analysis of the effects of VA on mortality specifically attributable to
pneumonia also reported null effects, though the risk estimate was significantly
attenuated compared with the above analysis, and some of the included studies
differed (7 studies, RR = 0.94, 95% CI: 0.67–1.30) [40].
There is evidence from both hospital-based studies and community-based
trials that VA supplements may increase the risk of acute lower respiratory tract
infections among well-nourished children [48]. It has been speculated that
high-dose VA might cause dysregulation of the immune system or a proinflam-
matory immune response, although this is uncertain [45, 48].
Programmatic Considerations
WHO recently issued guidance recommending high-dose VA supplementation

for infants and children 6–59 months of age in settings where VA deficiency is a
public health problem (table 1) [49]. Economic analyses have repeatedly ranked
VA supplementation of children 12–59 months of age as one of the most cost-
effective health interventions available, it was recently ranked at the top of the
list along with zinc supplementation of interventions chosen by the Copenhagen
consensus. Biannual distribution of capsules to children 6–59 months containing
200,000 IU VA is a relatively inexpensive intervention to implement and requires
little additional human resource mobilization. Initially coupled in many countries
with National Immunization Days for polio (for at least one round per year), in
recent years efforts have been made to distribute VA through child health days
along as part of a package often including insecticide-treated bednets, anti-helm-
inth medication, and other core child health interventions. Critics of VA capsule
supplementation programs have expressed concern that capsule supplementation,
originally considered to be a short-term solution to controlling the deficiency has
been implemented in many countries for decades, and that it might be holding
back funding of other approaches to controlling VA deficiency [43].
There are also programmatic uncertainties and issues requiring further
research. Two-round coverage rates are often quite high, nearing 80% in many
countries, but children who are missed by one or both rounds are often socio-
economically worse off than those reached, and could therefore be at greatest risk
of deficiency [50]. It is difficult to quantify how many lives are saved each year
from such programs, as few effectiveness evaluations have been undertaken. It is
possible that more frequent supplementation could be more efficacious, particu-
larly since it is known that serum retinol declines 4 months after supplementation
[2], yet more frequent supplementation also requires more resources and risks
donor and community fatigue. There is some evidence suggesting that the effect
of VA supplementation on risk of all-cause mortality may differ by vaccination
status of children [51, 52]. These findings are of potential concern to programs
given that vaccination is a convenient potential health contact for the adminis-
tration of VA. Lastly, the rapid expansion of other strategies for addressing VA
deficiency including home fortification with micronutrient powders, fortifica-
tion, and homestead food production, suggests the need for more work to develop
strategies to balance or target different approaches including consideration of the
potential adverse effects of VA fortification in areas of high HIV prevalence.
Conclusions
The basis for VA supplementation of 6- to 59-month-old infants and young

children is well established through many trials. More work is needed to

understand the mechanisms through which VA reduces all-cause and cause-
specific mortality and to more firmly establish the evidence related to supple-
mentation of other groups, particularly neonates.
References
1 Semba RD: Vitamin A as ‘anti-infective’ ther- 12 Jayarajan P, Reddy V, Mohanram M: Effect
apy, 1920–1940. J Nutr 1999;129:783–791. of dietary fat on absorption of beta carotene
2 Sommer A, West KP Jr: Vitamin A from green leafy vegetables in children.
Deficiency: Health, Survival, and Vision. Indian J Med Res 1980;71:53–56.
New York, Oxford University Press, 1996. 13 Mellanby E, Green H: Vitamin A as an anti-
3 World Health Organization W: Global infective agent. Its use in the treatment of
Prevalence of Vitamin A Deficiency in puerperal septicaemia. BMJ 1929;1:984–986.
Populations at Risk, 1995–2005. WHO 14 West KP Jr, Katz J, Khatry SK, et al: Double
Global Database on Vitamin A Deficiency. blind, cluster randomised trial of low dose
Geneva, World Health Organization, 2009. supplementation with vitamin A or beta
4 World Health Organization and Food and carotene on mortality related to pregnancy in
Agriculture Organization: Vitamin and Nepal. The NNIPS-2 study group. BMJ 1999;
Mineral Requirements in Human Nutrition, 318:570–575.
ed 2. Geneva, World Health Organization, 15 Kirkwood BR, Hurt L, Amenga-Etego S, et
2004. al: Effect of vitamin A supplementation in
5 Miller M, Humphrey J, Johnson E, et al: Why women of reproductive age on maternal sur-
do children become vitamin A deficient? J vival in Ghana (OBAAPAVITA): A cluster-
Nutr 2002;132:2867S–2880S. randomised, placebo-controlled trial. Lancet
6 de Pee S, West CE, Muhilal, et al: Lack 2010;375:1640–1649.
of improvement in vitamin A status with 16 West K, Christian P, Labrique A, et al: Effects
increased consumption of dark-green leafy of Vitamin A or beta carotene supplementa-
vegetables. Lancet 1995;346:75–81. tion on pregnancy-related mortality and
7 de Pee S, Bloem MW: The bioavailability of infant mortality in rural Bangladesh: a clus-
(pro) vitamin A carotenoids and maximizing ter randomized trial. J Am Med Assoc 2011;
the contribution of homestead food produc- 305:1986–1995.
tion to combating vitamin A deficiency. Int J 17 Thorne-Lyman A, Fawzi W: Vitamin A and
Vitam Nutr Res 2007;77:182–192. carotenoids during pregnancy and maternal
8 Awasthi S, Peto R, Read S, Bundy D: Six- and infant health outcomes: a systematic
Monthly Vitamin A from 1 to 6 Years of review. Paediatr Perinat Epidemiol, in press.
Age. DEVTA: Cluster-Randomised Trial 18 Gogia S, Sachdev HS: Maternal postpartum
in 1 Million Children in North India: Vitamin A supplementation for the preven-
Micronutrient Forum. Istambul, ILSI, 2007. tion of mortality and morbidity in infancy: a
9 Campbell AA, Akhter N, Sun K, et al: systematic review of randomized controlled
Relationship of homestead food production trials. Int J Epidemiol 2010;39:1217–1226.
with night blindness among children below 19 World Health Organization: Guideline:
5 years of age in Bangladesh. Public Health Vitamin A Supplementation in Postpartum
Nutr 2011;14:1627–1631. Women. Geneva, World Health
10 Fawzi WW, Herrera MG, Willett WC, et Organization, 2011.
al: Dietary vitamin A intake and the risk of 20 Greenberg BL, Semba RD, Vink PE, et
mortality among children. Am J Clin Nutr al: Vitamin A deficiency and maternal-
1994;59:401–408. infant transmissions of HIV in two met-
11 Fawzi WW, Herrera MG, Willett WC, et al: ropolitan areas in the United States. AIDS
Dietary vitamin A intake in relation to child 1997;11:325–332.
growth. Epidemiology 1997;8:402–407.
21 Semba RD, Miotti PG, Chiphangwi JD, et al: 31 Tang AM, Graham NM, Kirby AJ, et al:
Maternal vitamin A deficiency and mother- Dietary micronutrient intake and risk of pro-
to-child transmission of HIV-1. Lancet gression to acquired immunodeficiency syn-
1994;343:1593–1597. drome (aids) in human immunodeficiency
22 Semba RD, Miotti PG, Chiphangwi JD, et virus type 1 (HIV-1)-infected homosexual
al: Infant mortality and maternal Vitamin A men. Am J Epidemiol 1993;138:937–951.
deficiency during human immunodeficiency 32 Villamor E, Koulinska IN, Aboud S, et al:
virus infection. Clin Infect Dis 1995;21:966– Effect of vitamin supplements on HIV shed-
972. ding in breast milk. Am J Clin Nutr 2010;
23 Fawzi WW, Msamanga GI, Hunter D, et al: 92:881–886.
Randomized trial of vitamin supplements in 33 Arsenault JE, Aboud S, Manji KP, et al:
relation to transmission of HIV-1 through Vitamin supplementation increases risk of
breastfeeding and early child mortality. AIDS subclinical mastitis in HIV-infected women.
2002;16:1935–1944. J Nutr 2010;140:1788–1792.
24 Kumwenda N, Miotti PG, Taha TE, et al: 34 Gogia S, Sachdev HS: Neonatal vitamin A
Antenatal vitamin A supplementation supplementation for prevention of mortal-
increases birth weight and decreases anemia ity and morbidity in infancy: systematic
among infants born to human immunodefi- review of randomised controlled trials. BMJ
ciency virus-infected women in Malawi. Clin 2009;338:b919.
Infect Dis 2002;35:618–624. 35 Haider BA, Bhutta ZA: Neonatal vitamin
25 Coutsoudis A, Pillay K, Spooner E, et al: A supplementation for the prevention of
Randomized trial testing the effect of vitamin mortality and morbidity in term neonates
A supplementation on pregnancy outcomes in developing countries. Cochrane Database
and early mother-to-child HIV-1 transmis- Syst Rev 2011;CD006980.
sion in Durban, South Africa. South African 36 Gogia S, Sachdev HS: Vitamin A supple-
Vitamin A study group. AIDS 1999;13:1517– mentation for the prevention of morbidity
1524. and mortality in infants six months of age
26 Humphrey JH, Iliff PJ, Marinda ET, et al: or less. Cochrane Database Syst Rev 2011;
Effects of a single large dose of vitamin A, CD007480.
given during the postpartum period to HIV- 37 World Health Organization: WHO Technical
positive women and their infants, on child Consultation on Neonatal Vitamin A
HIV infection, HIV-free survival, and mor- Supplementation Research Priorities.
tality. J Infect Dis 2006;193:860–871. Geneva, World Health Organization, 2009.
27 Wiysonge CS, Shey M, Kongnyuy EJ, et al: 38 Rotondi MA, Khobzi N: Vitamin A supple-
Vitamin A supplementation for reducing mentation and neonatal mortality in the
the risk of mother-to-child transmission of developing world: a meta-regression of
HIV infection. Cochrane Database Syst Rev cluster-randomized trials. Bull World Health
2011;CD003648. Organ 2010;88:697–702.
28 Fawzi WW: The benefits and concerns 39 World Health Organization: Guideline:
related to vitamin A supplementation. J Neonatal Vitamin A Supplementation.
Infect Dis 2006;193:756–759. Geneva, World Health Organization, 2011.
29 MacDonald KS, Malonza I, Chen DK, et al: 40 Imdad A, Yakoob MY, Sudfeld C, et al:
Vitamin A and risk of HIV-1 seroconver- Impact of vitamin A supplementation on
sion among Kenyan men with genital ulcers. infant and childhood mortality. BMC Public
AIDS 2001;15:635–639. Health 2011;11(suppl 3):S20.
30 Tang AM, Graham NM, Saah AJ: Effects of 41 Mayo-Wilson E, Imdad A, Herzer K, et
micronutrient intake on survival in human al: Vitamin A supplements for preventing
immunodeficiency virus type 1 infection. mortality, illness, and blindness in children
Am J Epidemiol 1996;143:1244–1256. aged under 5: systematic review and meta-
analysis. BMJ 2011;343:d5094.

42 Fawzi WW, Chalmers TC, Herrera MG, 48 Villamor E, Fawzi WW: Vitamin A supple-
Mosteller F: Vitamin A supplementation mentation: implications for morbidity
and child mortality. A meta-analysis. JAMA and mortality in children. J Infect Dis
1993;269:898–903. 2000;182(suppl 1):S122–S133.
43 Latham M: The great vitamin A fiasco. 49 World Health Organization: Guideline:
World Nutr 2010;1:12–45. Vitamin A Supplementation in Infants and
44 Thurnham D: Vitamin A supplementation: Children 6–59 Months of Age. Geneva,
beneficial effects on mortality and morbid- World Health Organization, 2011.
ity in children aged six months to five years. 50 UNICEF: The State of the World’s Children
Sight Life Mag, in press. 2011. New York, UNICEF, 2011.
45 Villamor E, Fawzi WW: Effects of Vitamin 51 Benn CS, Aaby P, Nielsen J, et al: Does vita-
A supplementation on immune responses min A supplementation interact with routine
and correlation with clinical outcomes. Clin vaccinations? An analysis of the Ghana vita-
Microbiol Rev 2005;18:446–464. min A supplementation trial. Am J Clin Nutr
46 Ellison JB: Intensive vitamin therapy in 2009;90:629–639.
measles. BMJ 1932;2:708–711. 52 Benn CS, Rodrigues A, Yazdanbakhsh M, et
47 Semba RD: On Joseph Bramhall Ellison’s al: The effect of high-dose vitamin A supple-
discovery that vitamin A reduces measles mentation administered with BCG vaccine
mortality. Nutrition 2003;19:390–394. at birth may be modified by subsequent DTP
vaccination. Vaccine 2009;27:2891–2898.
Issues and Controversies with Vitamin A

in Childhood
María Teresa Murguía Peniche
Infancy Area, National Center for Child and Adolescent Health, Mexico City, Mexico
Abstract
Vitamin A deficiency is common in the developing world. Vitamin A supplementation
(VAS) has been used to prevent or treat vitamin A deficiency and to decrease mortality and
morbidity in children. However, there are still controversial issues in relation to the role of
universal VAS in different populations. Thus, studies that look at mortality outcomes reveal
that VAS decreases mortality in children >6 months of age; however, there is still contro-
versy on the extent to which reduction in morbidity from diarrhea and respiratory infec-
tion, other than measles, decreases mortality. Studies in infants 1–5 months old show no
protective effect of VAS on mortality; whether this is secondary to environmental influ-
ences (breastfeeding), or interactions with DTP vaccine, needs to be further investigated.
Studies with VAS in newborns have resulted in contrasting results in countries in Africa and
Asia; trials are underway to better understand this. VAS does not have a universal protec-
tive effect on lower respiratory tract infection in children; some studies reveal an increase
in respiratory morbidity associated with VAS, especially in well-nourished children; in con-
trast, VAS may confer some protection to malnourished children. The interaction of VAS
with different vaccines is under current debate; some discussions are presented.
Introduction
Vitamin A deficiency (VAD) is common in the developing world. About 190

million children under 5 years and 19.1 million pregnant women are vitamin A
deficient (i.e. serum retinol <0.70 μmol/l), representing about 33% of children
under 5 years in populations at risk of VAD. Xerophthalmia, a condition associ-
ated with VAD, is the world’s leading preventable cause of blindness [1].
Vitamin A has been described as an anti-infectious vitamin because of its
role in regulating human immune function and epithelial integrity; studies in
animals and humans have reported an association between VAD and increased
susceptibility to infections [2].
In addition to its preventive and therapeutic effect against xerophthalmia,
vitamin A supplementation (VAS) has been used to prevent or treat VAD and
to decrease mortality and morbidity in children. However, there are still con-
troversial issues, especially in relation to VAS’s role in deaths or morbidity not
related to measles in the pediatric population.
The WHO recommends VAS: 100,000 IU for infants 6–12 months of age
and 200,000 IU for children over 12 months of age, every 4–6 months [3].
The international community recommends VAS for all children between the
ages of 6 months and 5 years in all countries where over 70 in 1,000 children
die before the age of 5 years, ‘as this is the internationally accepted proxy
to indicate that VAD is a public health problem’ [4]. However, worldwide, a
large proportion of children receiving it are not vitamin A deficient, stunted
or wasted. The effect of such a public health policy on this population needs
to be carefully assessed.
Mortality Outcomes: Vitamin A Supplementation in >6-Month-Old

Children
In the 1930s, Ellison [5] first documented the protective effect of vitamin
A on measles mortality. The first evidence of the role of prophylactic VAS
in preventing death was demonstrated the 1980s in Indonesia; it concluded
that children who received massive-dose vitamin A supplements had a 34%
lower mortality from all causes than those not receiving the supplement [6].
These results were dramatic; however, no causes of death were reported. Also,
the control children had more clinical signs of VAD and poorer growth than
the children receiving VAS. During the following years, several other trials
were performed. A meta-analysis of these initial studies concluded that VAS
to patients hospitalized due to measles was highly protective against mor-
tality. In addition, when given periodically to children at the community
level, it decreased overall mortality (0.70; 0.56–0.879) [7]. It is important to
emphasize that these studies were performed mostly in Asia, in countries
with a high incidence of VAD and also in population with not universal mea-
sles immunization. Could it be that the significant reduction in mortality
rates in children receiving vitamin A supplements in some studies was due
to a reduction in measles deaths? This question has never been specifically
addressed. Latham [8] suggested that the statistical difference in deaths might
disappear if measles mortality were excluded from some studies of VAS.
‘‘The ‘causes’ of death in such studies were established by ‘verbal autopsies’.
It appeared entirely feasible, based on many years’ experience in the field in
Africa, that many deaths recorded as due to respiratory infections, diarrhea
92 Murguía Peniche
or fever, might in fact be measles deaths. Measles can cause all of these symp-
toms’’. Whether VAS might still have a positive effect in countries with high
measles immunization coverage remains to be established, and is an area of
controversy.
In a recent meta-analysis by Imdad et al. [9], including 43 trials, involving
215,633 >6-month-old children, there was an observed reduction in the risk of
all-cause mortality (24%) and a 28% overall reduction in diarrheal mortality. Of
interest, there was no effect on the mortality associated with lower respiratory
tract infection (LRTI). Again, the authors stated that ‘the causes of morbidity
and mortality were characterized by uncertainty’. So, the controversy of measles’
cofounding action remains on the scene. Moreover, one of the largest studies
exploring whether massive-dose vitamin A administration is associated with
a reduction in childhood mortality was taken up in India between 1999 and
2004. In that study, children were given 6-monthly massive doses of vitamin A,
6-monthly deworming, or both, or neither. Approximately 1 million children
were followed, and mortality rates in children 1–6 years of age were recorded.
There was no significant difference in death rates between children who received
the massive dose of vitamin A and those who did not [10]. Unfortunately, this
trial has not been published to date, and details that might explain the difference
between this study and previous meta-analyses are not available so far.
Mortality Outcomes: Children 1–5 Months of Age
Studies where children were supplemented with vitamin A between 1

and 5 months of age did not show any positive effect on overall mortality
[11–14] (see tables 1 and 2). Plausible explanations for these findings have
focused on differences in environmental influences on this age group in rela-
tion to older infants (e.g. the protection of breastfeeding against malnutrition
and infection) or on differences among the populations studied (e.g. different
prevalence rates of infectious diseases among study sites may affect under-
lying mortality rates). Whether this could be explained by the interactions
of vaccines given at this age, such as inactivated diphtheria-tetanus-pertussis
(DTP) vaccine, with vitamin A, needs to be further investigated (see below).
Vitamin A in Newborns
In developing countries, infants are born with low stores of vitamin A and
depend on external sources. Sometimes breastfeeding is not enough and new-
borns and infants need VAS; however, results obtained with this strategy have
been contradictory. Studies from Bangladesh [15], India [16] and Indonesia
[17] have shown reductions in all-cause mortality (15, 22 and 63%, respectively)
Controversies with Vitamin A Supplementation 93

Table 1. Main controversies related to periodical VAS in children (mortality outcomes)
Age at intervention Results Controversial issues
>6 months [9, 10] Decreased all-cause To what extent is this related to measles?
and diarrhea-specific
mortality No effect
1–5 months [11–14] No effect on mortality Are these results explained by breastfeeding
practices?
To what extent may vaccination at this age
interfere with a positive response?
Newborns African trials: no effect Are these results secondary to regional or

African trials [19–21] on mortality biological factors?
Asian trials [15–17] Asian trials (most): Could a difference in the prevalence of VAD
decreased mortality in pregnant women explain differences?
Table 2. Effect of VAS on mortality in infants 1–5 months of age
First author Recipients Vitamin A IU RR (95% CI)
Daularie [14] 1,058 50,000 0.99 (0.41–2.41)

West [11] 11,127 100,000
1,020 (1–3 months) 1.26 (0.88–1.80)
4,021 (4–5) months 0.84 (0.49–1.45)
Rahman [12] 199 25,000 × 3 0.97 (0.29–3.25)
WHO [13] 9,424 25,000 × 3 0.96 (0.73–1.27)
in infants who received VAS relative to controls. Also, VAS has shown to reduce
diarrhea case-fatality rates and the incidence of fever [18]. One study performed
in Nepal showed no overall effect on early infant mortality, and there was a
tendency for the relative risk of mortality among vitamin A recipients to rise
with improved nutritional status [19]. Trials in Africa, in countries like Guinea-
Bissau [20] and Zimbabwe [21] suggest a lack of benefit from VAS. A meta-
analysis by Gogia and Sachdev [22] proposes that there is insufficient evidence
to support neonatal supplementation with vitamin A, and there is an increased
risk of acute respiratory infection or respiratory difficulty (1.11; 1.02–1.21).
Although the study is appropriate, its application of general inclusion criteria
may limit its ability to ascertain the role of the prevalence of VAD in infant
mortality. Many speculations have so far been formulated in order to under-
stand these controversial results. A recent meta-regression analysis performed
by Rotondia and Khobzia [23] suggests that VAS to neonates within the first
94 Murguía Peniche
2 days of life may confer benefit, especially in regions where the prevalence of
vitamin A deficiency is 22% or more among pregnant women. This is an impor-
tant finding given the current debate as to whether giving neonates vitamin A
supplements helps reduce infant mortality in populations where endemic VAD
and high infant mortality exist. However, not all neonatal VAS trials support a
strong association between vitamin A status and baseline infant mortality, and
the VAS effect. Thus, in the Indonesian trial mentioned above, mothers had a
good vitamin A status, and infants a very good effect of VAS [17]. Biological
and regional differences among populations need to be further studied to better
understand contrasting results.
And finally, there has also been a debate in relation to a possible difference to
neonatal VAS by sex. However, results from a recent meta-analysis performed
by Kirkwood at al. [24] indicated that there is no differential effect of the inter-
vention in boys and girls. Overall, there was no benefit from VAS.
From these results, it is clear more research is needed in this area before
sound recommendations can be presented. It is possible that we will have better
idea of the impact of this intervention by the year 2013 when information from
ongoing trials in Ghana, India and Tanzania become available.
Morbidity Outcomes: Diarrhea, Lower Respiratory Tract Infection and

Measles
In relation to morbidity, another controversy arises: to what extent is the

decrease in mortality due to reduction in morbidity from diarrhea and respi-
ratory infection (other than measles). One study in India, with measles-
vaccinated and not severely malnourished children, concluded that VAS did
not reduce respiratory and diarrheal morbidity as compared to the placebo
controls [25].
The meta-analysis by Imdad et al. [9] revealed a reduced incidence of
diarrhea (0.85; 0.82–0.87) and measles (0.50; 0.37–0.67) for VAS compared
with control. But again, it included studies where it was difficult to control
for measles comorbidity. Of interest, there was no effect on the incidence of
respiratory disease or the incidence of hospitalizations due to diarrhea or
respiratory disease. All of these studies were performed in developing coun-
tries, cointerventions were not analyzed, and there was statistical heteroge-
neity. Again, the controversy remains because of this and the uncertainty of
morbid conditions.
Even more controversial, VAS has been shown to be detrimental in pediat-
ric population. In a blinded, randomized controlled trial (RCT) performed in
Mexico [26], where measles vaccination coverage is very high and mortality in
children <5 years has been estimated at 17.4/1,000 live births (Information from
DGIS, SS Mexico, 2009), VAS to infants 6–15 months old was associated with a

Table 3. Controversies surrounding VAS in children (diarrheal and respiratory morbidity)
Morbidity Results Controversial issues
Incidence of Decrease/increase/no change Are these differences secondary to

diarrhea [9, 25, 26] pathogen-specific responses to
vitamin A?
To what extent is this related to
measles?
Incidence of No change/increase/decrease Better results with lower nutritional

respiratory status?
infections [9, 26, 29] Pathogen-specific responses?
Respiratory severity Therapeutic: no effect/increase Are these differences pathogen-

[29] Prophylactic: no effect/increase specific?
Are these results related to different
doses? Lower doses better results?
27% increase in diarrheal disease and a 23% increase in cough with fever (RR:
1.23; 1.02–1.47). The authors postulated that the increase in diarrhea in this
study might reflect the effect of VAS to downregulate the Th1 response, which
protects against rotavirus. Similarly, they speculated that negative respiratory
outcomes might be explained by the adverse effect of VAS in respiratory syncy-
tial virus infections [27, 28].
A meta-analysis by Chen et al. [29] concluded that VAS does not have a uni-
versal protective effect as prophylaxis for LRTI in children; indeed, three stud-
ies showed an increase in respiratory morbidity associated with VAS (table 3).
There was also post-hoc evidence to show that VAS in children with poor nutri-
tional status may decrease the incidence of LRTI while the opposite was true for
well-nourished children.
Another systematic review by Mathew [30], which included 11 prophylac-
tic and 9 therapeutic trials concluded that VAS has no benefit for childhood
community-acquired pneumonia.
Wu et al. [31] published a meta-analysis involving six trials with 1,740 chil-
dren, some of them <6 months of age. They concluded that disease severity after
supplementary high-dose vitamin A was significantly worse compared with
placebo (one study). However, low-dose vitamin A (10,000 IU twice daily for
the first 6 days, followed by 1,500 IU/day administered for the next 20 days)
significantly reduced the recurrence rate of bronchopneumonia (OR 0.12; 0.03–
0.46). Moderate vitamin A doses (100,000 IU of vitamin A to children aged one
year and older, and 50,000 IU to infants under the age of one year) significantly
reduced the time to remission of signs in children with normal serum retinol
(>200 μg/l).
96 Murguía Peniche
Further RCTs, possibly with measured vitamin A levels and varying vitamin
A doses, may provide sufficient evidence to clarify the role of vitamin A in non-
measles pneumonia and other conditions.
Measles and Vitamin A Supplementation

A recent review which included eight trials (2,574 participants) revealed
that there was no significant reduction in the risk of mortality in the vita-
min A group when all the studies were pooled (0.70; 0.42, 1.15), and the
pooled estimate from four studies suggested the risk of pneumonia-specific
mortality (0.57; 0.24–1.37) was not significantly reduced in supplemented
children. However, there was evidence of benefit with two mega-doses of
VAS on consecutive days. This approach reduced the mortality in children
aged <2 years (0.21; 0.07–0.66) and the risk of pneumonia-specific mortal-
ity (0.33; 0.08–0.92); it also reduced the incidence of croup by 41% [32, 33].
It is important to recognize that benefit was demonstrated in hospitalized
children aged <2 years, given two doses of VA and in areas where the case
fatality rate was greater than 10%. Controversy still remains as to whether
these findings could be generalized to populations with high rates of measles
immunization (low exposure), with low case fatality rates and in nonhospi-
talized children.
Apart from its role in decreasing mortality associated with measles pneumo-
nia in children <2 years, the best target population for VAS in respiratory condi-
tions, the best dosage, and the best timing are not clear yet.
Effect of Vitamin A Supplementation on Vaccine’s Response: Good, Bad,

Indifferent?
The interaction of VAS with different vaccines may vary depending on the
immunogen, the host, vitamin A dosage and many other factors. Benn et al. [34]
have suggested that VAS could interact with the vaccines by amplifying their
nonspecific effects. According to this thesis, VAS would have beneficial effects
when administered with live vaccines such as BCG at birth or measles after 6
months of age, but no effect or even negative effects when administered with
inactivated DTP vaccine between the age of 1 and 5 months. This hypothesis
remains to be tested.
Some relevant findings related to VAS and immunizations are summarized
in the following paragraphs.
Bacille Calmette-Guerin Vaccine

The effect of VA administered at birth along with this vaccine has been studied
in a large (n = 1,894 and 1,426 in final analyses) randomized, placebo-controlled
trial in Guinea-Bissau. The percentage of purified protein derivative of

Mycobacterium tuberculosis responders was significantly lower in 2-month-old
boys who received VAS than in placebo recipients [35].
Measles Vaccine
There is some evidence that the effect of VAS is highly dependent on age at vac-
cination. One RCT study in Indonesia in which infants received the Schwarz
measles vaccine (MV) and either the supplement (30 mg retinol equivalent) or a
placebo at 6 months of age reported that supplemented infants had lower sero-
conversion rates at age 12 months, but only in infants with high levels of specific
antibodies before vaccination (titers >8). It was suggested that the combined
immune effects of VA and circulating maternal antibodies could neutralize the
vaccine antigens [36, 37].
In children >6 months, there have been controversial results regarding the
effect of VAS on MV rate of seroconversion or serologic titers. Some RCTs
have shown higher seroconversion rates, some higher serum titers and other
no changes associated with VAS [38–42]. Contradictory results from these
studies may be explained by various factors, including preintervention VA sta-
tus, age at vaccination and supplementation and hence presence/absence of
maternal antibodies, vaccine strain, and antibody serological assays. There is
some evidence to suggest that the most important factor responsible for the
effect of VAS on MV response is the baseline-specific antibody titers. From a
meta-analysis based on 3 studies in which this information was available [36,
42, 43], it was found that VAS was associated with reduced seroconversion
rates in children having high baseline antibody levels [OR 0.57; 0.35–0.94]. In
contrast, VAS did not affect seroconversion rates in children with low baseline
antibody titers [44].
Diphtheria-Tetanus-Pertussis Vaccine
Most studies have not shown an improved tetanus-specific antibody response
associated with VAS, as compared to controls [45, 46]. However, in an Indonesian
study, the oral administration of 200,000 IU of vitamin A to tetanus-naive chil-
dren 3–6 years of age resulted in significantly higher titers of anti-tetanus toxoid
after immunization compared to those in the placebo group. In this study, vita-
min A was given 2 weeks before immunization [47].
These controversial results may be related to differences in the VAS timing,
basal vitamin A status or other factors that need to be further studied.
A group of researchers in Denmark have raised questions regarding the
safety of VA administration with DTP vaccines. Thus, reanalyzing data from
a study performed in Ghana, VAS had a negative effect in measles-vaccinated
girls who were missing one or more doses of DTP at enrollment, a group who
often received DTP during follow-up (mortality RR: 2.60; 1.41–4.80) [48].
The controversy regarding a potential detrimental effect of VA administered
with DTP vaccines has led to considerable discussion. The possible negative
98 Murguía Peniche
interaction is biologically plausible; however, the evidence produced so far is
not abundant.
Polio Vaccine
VA has no effect on seroconversion to any of the 3 serotypes of poliomyeli-
tis vaccine, although one study has shown a positive effect on seroprotection
against poliovirus type 1 [49].
Antagonism with Vitamin D
Animal studies suggest that vitamin A is an antagonist of vitamin D action.

Massive doses of vitamin A have been shown to intensify the severity of bone
demineralization and to inhibit the ability of vitamin D to prevent such demin-
eralization [50]. The effect of supplementing with mega-doses of vitamin A
vitamin D-deficient children born to malnourished or not sun-exposed women
with a low calcium intake needs to be seriously addressed.
Conclusions
There is evidence that VAS in children >6 months of age decreases childhood mor-
tality; however, there is controversy on the relation of the cause-specific protective
effect. Could the difference in deaths possibly disappear if measles mortality were
excluded from some studies of VAS? Might those in fact be measles deaths? If this
is the case, could VAS still have a positive effect in countries with high measles
immunization coverage? VAS in newborns and infants <6 months seems not to
have a protective effect on mortality; however, there is controversy if newborn VAS
could be useful in regions where pregnant women have a high prevalence of VAD.
There is evidence to suggest VAS might be beneficial for diarrheal morbidity;
however, this is controversial when considering measles comorbidity. Even though
there are theoretical reasons that VAS might be effective for acute LRTI, most
studies have not supported this concept or even have revealed the opposite. The
reasons for these findings are not clear. The modulatory effects of vitamin A in
the immune system, its potential interaction with the specific immune responses
to different viral/bacterial pathogens, its role in several pathways of disease and
the dose effect on disease development/protection remain to be elucidated to
understand these complex results. VAS reduces measles mortality if mega-doses
are given to children <2 years of age in regions with high case fatality rates. VAS
improves immune response to MVs if circulating maternal antibodies are low or
absent. There is controversy regarding the interaction between DTP and VAS; a
single research group has reported possible adverse interaction; however, there is
no conclusive evidence.

References
1 World Health Organization: Global 13 Randomised trial to assess benefits and
Prevalence of Vitamin A Deficiency in safety of vitamin A supplementation linked
Populations at Risk 1995–2005. WHO Global to immunisation in early infancy. WHO/
Database on Vitamin A Deficiency 2009. CHD Immunisation-Linked Vitamin A
2 Semba RD: Vitamin A and immunity to Supplementation Study Group. Lancet
viral, bacterial and protozoan infections. 1998;352:1257–1263.
Proc Nutr Soc 1999;58:719–727. 14 Daularie NM, Starbuck ES, Hosuton RM,
3 WHO Guideline: Vitamin A supplementa- et al: Childhood mortality after high dose
tion in infants and children 6–59 months vitamin A in a high risk population. BMJ
of age. Geneva, world Health Organization, 1992;304:207–210.
2011. 15 Klemm RD, Labrique AB, Christian P, et
4 UNICEF: Vitamin A supplementation: al: Newborn vitamin A supplementation
Progress for child survival. Working paper reduced infant mortality in rural Bangladesh.
prepared by UNICEF nutrition section. New Pediatrics 2008;122:e242–e250.
York, UNICEF, 2005. 16 Rahmathullah L, Tielsch JM, Thulasiraj RD,
5 Ellison JB: Intensive vitamin therapy in et al: Impact of supplementing newborn
measles. BMJ 1932;ii:708–711. infants with vitamin A on early infant mor-
6 Sommer A, Djunaedi E, Loeden AA, et al: tality: community based randomised trial in
Impact of vitamin A supplementation on southern India. BMJ 2003;327:254.
childhood mortality. Lancet 1986;1:1169– 17 Humphrey JH, Agoestina T, Wu L, et al:
1173. Impact of neonatal vitamin A supplementa-
7 Fawzi WW, Chalmers TC, Herrera MG, et tion on infant morbidity and mortality. J
al: Vitamin A supplementation and child Pediatr 1996;128:489–496.
mortality: a meta-analysis. J Am Med Assoc 18 Tielsch JM, Rahmathullah L, Thulasiraj RD,
1993;269:898–903. et al: Newborn vitamin A dosing reduces the
8 Latham MC: Vitamin A and childhood mor- case fatality but not incidence of common
tality. Lancet 1993;342:549. childhood morbidities in South India. J Nutr
9 Imdad A, Herzer K, Mayo-Wilson E, et al: 2007;137:2470–2474.
Vitamin A supplementation for preventing 19 West KP Jr, Katz J, Shrestha SR, et al:
morbidity and mortality in children from 6 Mortality of infants <6 mo of age supple-
months to 5 years of age. Cochrane Database mented with vitamin A: a randomized,
Syst Rev 2010;CD008524. double-masked trial in Nepal. Am J Clin
10 Awasthi S, Peto R, Read S, et al: Six-monthly Nutr 1995;62:143–148.
vitamin A from 1 to 6 years of age. DEVTA: 20 Benn CS, Diness BR, Roth A, et al: Effect of
cluster-randomised trial in 1 million children 50,000 IU vitamin A given with BCG vaccine
in North India. www.ctsu.ox.ac.uk/projects/ on mortality in infants in Guinea-Bissau:
devta/istanbul-vit-a-lecture.ppt. randomised placebo controlled trial. BMJ
11 West KP Jr, Katz J, Shrestha SR, et al: 2008;336:1416–1420.
Mortality of infants <6 mo of age supple- 21 Malaba LC, Iliff PJ, Nathoo KJ, et al:
mented with vitamin A: a randomized, ZVITAMBO Study Group Effect of postpar-
double-masked trial in Nepal. Am J Clin tum maternal or neonatal vitamin A supple-
Nutr 1995;62:143–148. mentation on infant mortality among infants
12 Rahman MM, Mahalanabis D, Wahed MA, born to HIV-negative mothers in Zimbabwe.
et al: Administration of 25,000 IU vitamin Am J Clin Nutr 2005;81:454–460.
A doses at routine immunisation in young 22 Gogia S, Sachdev HS: Neonatal vitamin A
infants. Eur J Clin Nutr 1995;49:439–445. supplementation for prevention of mortal-
ity and morbidity in infancy: systematic
review of randomised controlled trials. BMJ
2009;338:b919.
100 Murguía Peniche

23 Rotondia MA, Khobzia N: Vitamin A 35 Diness BR, Fisker AB, Roth A, et al: Effect of
supplementation and neonatal mortality in high-dose vitamin A supplementation on the
the developing world: a meta-regression of immune response to Bacille Calmette-Guerin
cluster-randomized trials. Bull World Health vaccine. Am J Clin Nutr 2007;86:1152–1159.
Organ 2010;88:697–702. 36 Semba RD, Munasir Z, Beeler J, et al:
24 Kirkwood B, Humphrey J, Moulton L, et al: Reduced seroconversion to measles in infants
Neonatal vitamin A supplementation and given vitamin A with measles vaccination.
infant survival. Lancet 2010;6737:61895– Lancet 1995;345:1330–1332.
61898. 37 Benn CS, Whittle H, Aaby P, et al: Vitamin
25 Ramakrishnan U, Latham MC, Abel R, et al: A and measles vaccination. Lancet
Vitamin A supplementation and morbidity 1995;346:503–504.
among preschool children in South India. 38 Benn CS, Aaby P, Bale C, et al: Randomised
Am J Clin Nutr 1995;61:1295–1303. trial of effect of vitamin A supplementa-
26 Long KZ, Montoya Y, Hertzmark E, et al: A tion on antibody response to measles vac-
double-blind, randomized, clinical trial of cine in Guinea-Bissau, west Africa. Lancet
the effect of vitamin A and zinc supplemen- 1997;350:101–105.
tation on diarrheal disease and respiratory 39 Benn CS, Balde A, George E, et al: Effect
tract infections in children in Mexico City, of vitamin A supplementation on measles-
Mexico Am J Clin Nutr 2006;83:693–700. specific antibody levels in Guinea-Bissau.
27 Jiang B, Snipes-Magaldi L, Dennehy P, et Lancet 2002;359:1313–1314.
al: Cytokines as mediators for or effectors 40 Bhaskaram P, Rao KV: Enhancement in
against rotavirus disease in children. Clin seroconversion to measles vaccine with
Diagn Lab Immunol 2003;10:995–1001. simultaneous administration of vitamin A in
28 Bresee JS, Fischer M, Dowell SF, et al: 9-months-old Indian infants. Indian J Pediatr
Vitamin A therapy for children with respira- 1997;64:503–509.
tory syncytial virus infection: a multicenter 41 Cherian T, Varkki S, Raghupathy P, et al.
trial in the United States. Pediatr Infect Dis J Effect of Vitamin A supplementation on the
1996;15:777–782. immune response to measles vaccination.
29 Chen H, Zhuo Q, Yuan W, et al: Vitamin Vaccine 2003;21:2418–2420.
A for preventing acute lower respiratory 42 Semba RD, Akib A, Beeler J, et al: Effect of
tract infections in children up to seven vitamin A supplementation on measles vac-
years of age. Cochrane Database Syst Rev cination in nine-month-old infants. Public
2008;CD006090. Health 1997;111:245–247.
30 Mathew J: Vitamin A supplementation for 43 Bhaskaram P, Rao KV: Enhancement in
prophylaxis or therapy in childhood pneu- seroconversion to measles vaccine with
monia: a systematic review of randomized simultaneous administration of vitamin A
controlled trials. Indian Pediatr 2010;47:255– in 9-months-old Indian infants. Indian J
261. Pediatr. 1997;64:503–509.
31 Wu T, Ni J, Wei J: Vitamin A for non-measles 44 Benn CS, Whittle H, Aaby P, et al: Vitamin
pneumonia in children. Cochrane Database A and measles vaccination. Lancet
Syst Rev 2005;CD003700. 1995;346:503–504.
32 D’Souza RM, D’Souza R: Vitamin A for treat- 45 Brown KH, Rajan MM, Chakraborty J,
ing measles in children. Cochrane Database et al: Failure of a large dose of vitamin
Syst Rev 2002;CD001479. A to enhance the antibody response to
33 Yang HM, Mao M, Wan C: Vitamin A for tetanus toxoid in children. Am J Clin Nutr
treating measles in children. Cochrane 1980;33:212–217.
Database Syst Rev 2005;CD001479. 46 Kutukculer N, Akil T, Egemen A, et al:
34 Benn CS, Balé C, Sommerfelt H, et al: Adequate immune response to tetanus toxoid
Hypothesis: vitamin A supplementation and failure of vitamin A and E supplementa-
and childhood mortality: amplification of tion to enhance antibody response in healthy
the non-specific effects of vaccines? Int J children. Vaccine 2000;18:2979–2984.
Epidemiol 2003;32:822–828.

47 Semba RD, Muhilal AL, Scott G, et al: 49 Bahl R, Bhandari N, Kant S, et al: Effect
Depressed immune response to tetanus in of vitamin A administered at Expanded
children with vitamin A deficiency. J Nutr Program on Immunization contacts on anti-
1992;122:101–107. body response to oral polio vaccine. Eur J
48 Benn CS, Aaby P, Nielsen J, et al: Does vita- Clin Nutr 2002;56:321–325.
min A supplementation interact with routine 50 Rohde CM, Manatt CE, Clagett-Darne M, et
vaccinations? An analysis of the Ghana al: Vitamin A antagonizes the action of vita-
Vitamin A Supplementation Trial 1 – Am J min D in Rats. J Nutr 1999;129:2246–2250.
Clin Nutr 2009;90:629–639.
102 Murguía Peniche

Discussion on Vitamin A
Supplementation in Childhood
Vitamin A deficiency (VAD) is a very common public health problem in more

than half of the world, especially in Africa and Southeast Asia, with young
children and pregnant women in low-income countries being hardest hit. The
WHO estimates that 250 million preschool children have VAD and that a sub-
stantial proportion of pregnant women in VAD geographic areas is vitamin A
deficient. Globally, between 1995 and 2005, ~5.2 million preschool children
and ~9.7 million pregnant women were affected by night blindness. If retinol
concentrations <70 μmol/l are reviewed, 190 million children have low levels
of retinol compared to 19 million pregnant women. For children, VAD signifi-
cantly increases the risk of severe illness, and even death from common illnesses
such as diarrhea and measles. For pregnant women, VAD occurs during the last
trimester and the mother presents with night blindness. The relationship or
association of VAD and vertical transmission of HIV remains to be clarified.
Dietary deficiency can begin in the neonatal period due to lack of breastfeed-
ing and continue into adulthood due to diets deficient in vitamin A-containing
foods such as liver, cheese, milk, eggs, fruits and vegetables and fortified foods.
Randomized controlled studies of vitamin A supplementation in women before,
during or after pregnancy have yielded conflicting results. West et al. [1] in a
double-blind, placebo controlled study in Nepal, demonstrated that weekly 7,000
retinol equivalents as vitamin A or provitamin A β-carotene were associated with a
44% reduction in maternal mortality. In the SUMMIT trial conducted in Indonesia,
early infant mortality was reduced in infants born to mothers receiving micronu-
trient supplementation including 800 μg of retinol (35.5/1,000 vs. 43/1,000, RR
0.82, 95% CI: 0.70–0.95), but there was no effect on maternal mortality [2, 3]. More
recently, Kirkwood et al. [3], found no significant effect on pregnancy-related
mortality or on all-cause mortality with vitamin A supplementation. There were
no effects on the rate of still birth, perinatal, neonatal or infant death.
The effect of VAD on all cause and diarrheal mortality remains controversial.
In one study of over 200,000 children, vitamin A supplementation decreased
mortality, whereas in the DEVTA study from India, no differences in mortality
were observed [4]. Further, a question raised was whether vitamin A supplemen-
tation would still have a beneficial effect on mortality in countries with a high
measles immunization rate. In 1995, Ramakrishnan and colleagues found that
respiratory and diarrheal morbidity were similar in children less than 3 years of
age who were vaccinated against measles but were not severely malnourished,
suggesting that overall nutrition is a very important confounding factor. There
appears to be an inverse relationship between vitamin A supplementation and
all-cause mortality with declining mortality with increased number of doses per
child per year. On the other hand, vitamin a supplementation in doses ranging
from 1,000 to 9,000 IU/kg per day for 28 days increased markers of oxidative
stress in animal models.
In children, protection from VAD comes from breastfeeding in the neona-
tal period, and later providing a periodic supply of high-dose vitamin A has
demonstrated a reduction in mortality. Since breast milk is a good source of
vitamin A, and its level can be influenced by supplementation of the lactating
mother, promoting breastfeeding is a good strategy to prevent VAD. For defi-
cient children, a periodic supply of high doses of vitamin A has been demon-
strated to reduce mortality by 23% and up to 50% in those with measles. The
exact number of doses and the dose itself is still not clear. Overall, in an analysis
by Kirkwood et al. [3], there was a favorable trend towards reduced mortality in
boys, but not in girls. However, of the 6 studies used in the analyses for boys, the
confidence interval crossed 1 in 5 of the 6, whereas they crossed 1 in all 6 studies
for girls. There were no significant differences in outcomes that included infant
mortality, newborn mortality, respiratory or diarrheal morbidity [5]. Vitamin
A supplementation during infancy does not appear to adversely interact with
common immunizations with no differences in mortality in the first year in
supplementation vs. placebo and DPT/polio vaccination [6].
Control of VAD can be done by different approaches: improving the avail-
ability and intake of vitamin A through dietary intake of vitamin A-rich foods;
increasing the dietary intake through fortification of foods as has been done in
Central and South America; periodic delivery of 200,000 IU of vitamin A in
preschool children with half the dose given between 6 and 11 months of age
appears very successful [7, 8]. The periodic approach to supplementation has
been shown to reduce the risks of xerophthalmia by ~90% and mortality by up
to 23% in young children [9]. Many high-risk countries have also adopted the
WHO policy of supplementing mothers within 6 weeks after delivery to increase
vitamin A content of breast milk.
There appear, however, to be geographic differences with trials in Africa of
vitamin A supplementation in infants that did not alter mortality, and those in
Asia that did. In infants between 1 and 5 months of age, there was no effect on
mortality. In infants >6 months of age, vitamin A supplementation decreased
all-cause and diarrhea-specific mortality in some studies, but not others.
104 Bhatia
Thus, for the continuum of vitamin A sufficiency, one would include breast-
feeding for 6 months, appropriate timely vitamin A supplementation between
6 months and 6 years of age, and food fortification after that (Guatemala, for
example, where sugar is fortified). For specific vulnerable populations, growing
fruits and vegetables complements dietary diversity and food fortification.
These three approaches combined with other public health initiatives to
promote breastfeeding, oral rehydration therapy, vaccination and birth spacing
can all lead to significant reductions in VAD and related morbidities.
Jatinder Bhatia
References
1 West KP Jr, Katz J, Khatry SK, et al: Double 5 Gogia S, Sachdev HS: Neonatal vitamin A
blind, cluster randomised trial of low dose supplementation for prevention of mortal-
supplementation with vitamin A or beta ity and morbidity in infancy: systematic
carotene on mortality related to pregnancy review of randomised controlled trials. BMJ
in Nepal. The NNIPS-2 Study Group. BMJ 2009;338:b919.
1999;318:570–575. 6 WHO/CHD Immunisation-Linked
2 Supplementation with Multiple Vitamin A Supplementation Study Group:
Micronutrients Intervention Trial Randomised trial to assess benefits and
(SUMMIT) Study Group, Shankar AH, Jahari safety of vitamin A supplementation linked
AB, Sebayang SK, et al: Effect of maternal to immunisation in early infancy. Lancet
multiple micronutrient supplementation 1998;352:1257–1263.
on fetal loss and infant death in Indonesia: 7 Arroyave G, Aguilar JR, Flores M, Guzmán
a double-blind cluster-randomised trial. MA: Evaluation of Sugar Fortification with
Lancet 2008;371:215–227. Vitamin A at the National Level. Scientific
3 Kirkwood BR, Hurt L, Amenga-Etego S, publication 384. Washington, Pan American
Tawiah C, et al: Effect of vitamin A supple- Health Organization, 1979.
mentation in women of reproductive age on 8 World Health Organization, UNICEF,
maternal survival in Ghana (ObaapaVitA): a IVACG Task Force: Vitamin A Supplements:
cluster-randomised, placebo-controlled trial. A Guide to Their Use in the Treatment of
Lancet 2010;375:1640–1649. Vitamin A Deficiency and Xerophthalmia, ed
4 Imdad A, Herzer K, Mayo-Wilson E, et al: 2. Geneva, World Health Organization, 1997.
Vitamin A supplementation for preventing 9 Bhutta ZA, Ahmed T, Black RE, et al: What
morbidity and mortality in children from 6 works? Interventions for maternal and
months to 5 years of age. Cochrane Database child undernutrition and survival. Lancet
Syst Rev 2010;CD008524. 2008;371:417–440.
Discussion on Vitamin A Supplementation in Childhood 105

Iron
Influence of Inflammatory Disorders

and Infection on Iron Absorption and
Efficacy of Iron-Fortified Foods
Richard F. Hurrell
ETH Zurich, Zurich, Switzerland
Abstract
The provision of iron-fortified foods is a common strategy to prevent iron deficiency;
however, ensuring adequate iron absorption is a challenge. Iron bioavailability depends
on the choice of iron compound, the presence enhancers and inhibitors of absorption in
the food matrix, and the physiological state of the consumer, including iron status, other
nutritional deficiencies and inflammatory disorders. Inflammation associated with infec-
tions and inflammatory disorders would be expected to decrease iron absorption and
reduce the efficacy of iron-fortified foods. The decreased absorption is due to an increase
in circulating hepcidin in response to inflammatory cytokines. Hepcidin degrades ferro-
portin and blocks the passage of iron from the intestinal cell to the plasma. This is the
innate immune response to infections and aims to restrict pathogen growth by restricting
iron supply. Stable isotope studies have reported women and children with chronic
malaria parasitemia or febrile malaria to have increased inflammatory cytokines, increased
hepcidin and much decreased iron absorption. No studies have specifically investigated
the efficacy of iron-fortified foods in the absence and presence of infections. In contrast,
inflammation and increased hepcidin associated with adiposity in overweight have been
linked to both lower iron absorption and the decreased efficacy of iron-fortified foods.
Introduction
Approximately 90% of daily iron needs can be met by the reutilization of red cell
iron after the breakdown of circulating red cells at the end of their natural life
span. Although there is no physiological mechanism for active iron excretion
from the body, there are obligatory iron losses from skin, intestine and urinary
tract and additional losses during menstruation in women of child-bearing
age. To maintain iron balance, obligatory and menstrual iron losses, plus the
iron required for growth in infants, children, and adolescents must come from
the diet. This represents about 1–2 mg of absorbed iron per day. Many young
women, infants and children fail to meet their iron needs from diet, and these
population groups make up the majority of the 2 billion people worldwide esti-
mated to be iron deficient [1]. The situation is worse in the developing world
due to low iron bioavailability from plant-based diets, and iron deficiency has
more serious negative health and economic consequences including poor preg-
nancy outcome, poor cognitive development in children, and decreased physi-
cal performance and productivity [2].
Four strategies are currently being used to target iron deficiency. These
are dietary diversification, supplementation, food fortification and biofortifi-
cation, although fortification and supplementation are often the only practical
options. Improving iron bioavailability by including more animal tissue foods,
fruits and vegetables in the diet is usually limited by cost and supply, espe-
cially in the developing world, and biofortification of plant staples by breeding
or genetic engineering is not yet ready for introduction into the food supply.
Iron supplementation can be both therapeutic and preventive and involves
the provision of relatively large amounts of medicinal iron as tablets or syr-
ups to population known or expected to be iron deficient. Supplements are
recommended to be taken in the absence of foods as food per se decreases
iron absorption [3]. Iron-fortified foods contain lower doses of iron than sup-
plements, and are regarded as the best long-term approach for the preven-
tion of iron deficiency. Iron-fortified staple foods and condiments are usually
planned and mandated by government, and will reach all population groups
including women and older children. Iron-fortified manufactured weaning
foods are targeted specifically at infants and young children and, for the lower
socioeconomic groups, micronutrient powders and micronutrient-fortified
fat-containing pastes have been introduced in recent years. These products,
usually in the form of food aid, are designed to be added at time of consump-
tion to the cereal gruels commonly eaten by these infants and young children.
The food industry additionally markets a range of iron-fortified foods (milk
products, breakfast cereals, chocolate drinks) targeted at infants, children,
adolescents and young women.
For the many infants, children and women at risk of iron deficiency, iron-
fortified foods can fill the gap between iron intake from the regular diet and
iron needs. However, ensuring adequate absorption of the fortification iron is
far from easy and iron bioavailability from fortified foods depends on the choice
of iron compound [1], the food matrix and the physiological state of the con-
sumer. Iron-fortified foods often contain inhibitors of iron absorption such as
phytic acid and polyphenols in cereal and legume foods [4, 5] and calcium in
milk products [6]. In order to overcome the negative effects of these inhibitors,
108 Hurrell
the food manufacture usually adds ascorbic acid, although NaFeEDTA and
phytase can also be added [7].
However, even if the food matrix is optimized, iron absorption ultimately
depends on the physiological state of the consumer. There is an inverse corre-
lation between iron status and iron absorption [8] with a 50% decrease in iron
stores doubling of iron absorption, although the magnitude of the inhibitory or
enhancing effects are independent of iron status. Genetic disorders, nutritional
deficiencies, infection and inflammation can also influence iron absorption [9].
Homozygotes for thalassemia and hemochromatosis may have increased iron
absorption and must be under medical surveillance, although dietary iron absorp-
tion by single gene carriers of these traits is little affected. Vitamin A deficiency
can influence several stages of iron metabolism, and riboflavin deficiencies may
decrease iron incorporation into hemoglobin. Chronic inflammation [10] and
obesity [11] increase hepcidin expression and would be expected to decrease iron
absorption [12]. This review evaluates the potential influence of inflammatory
disorders and infection on iron absorption and efficacy of iron-fortified foods.
Iron Metabolism during Infection and Inflammation
Iron needed for new red cell synthesis and for replenishing iron enzymes and
myoglobin is transported to the bone marrow or body tissues via transferrin in
the plasma. Most of the iron supply enters the plasma via macrophages after the
destruction of senescent red cells. The remainder enters as absorbed iron via the
mucosal cells and, in times of need, from the iron stored as ferritin mainly in hepa-
tocytes. The passage into the plasma is mediated via the transport protein ferro-
portin situated on the cell membrane [13], and its entry is strictly controlled by the
regulatory hormone, hepcidin. This protein is secreted by the liver when iron status
is adequate and inhibits the transport of iron into the plasma from both the mac-
rophages and the intestinal cells [14]. Hepcidin binds ferroportin at the cell mem-
brane, causing internalization and degradation [15], and when iron status is low,
hepcidin release from the liver is decreased and iron absorption is maximized.
The innate immune response to microbial infection is to increase hepcidin via
an inflammatory response and to restrict microbial growth by restricting the entry
of iron into the plasma [16]. The anemia of infection results from an interruption
in the recycling of red cell iron. Macrophage iron is not released resulting in insuf-
ficient iron for erythropoiesis. The outcome of many infectious diseases depends
on preventing the invading pathogen from obtaining its iron supply, and provision
of high iron doses to an infected patient can worsen the infection. In addition to
preventing iron release from the macrophage, the inflammatory response would
also be expected to prevent iron release from the mucosal cell and thus restrict iron
absorption. Recent evidence indicates that the inflammatory response to malarial
parasitemia decreased iron absorption from an iron-fortified sorghum gruel [12],
Inflammation and Iron Absorption 109

Table 1. Iron absorption and iron status following malaria treatment of
anemic 8- to 36-month-old Gambian children and subsequent iron supple-
mentation [20]
Percent iron Hb, g/l Serum ferritin, μg/l

absorption
Children with malaria

Day 1 after treatment 8.7 84 74
Day 15 after treatment 15.5 100 22
Day 30 after treatment – 111 20
Children free of malaria,

no treatment
Day 1 26.6 91 6
Day 15 24.0 97 13
Day 30 – 103 16
Children were supplemented with 2 mg Fe/kg bodyweight per day from day 1
to day 30.
and that the inflammation associated with obesity and overweight decreases iron
absorption and reduces the efficacy of iron-fortified foods [17].
Influence of Infection on Iron Absorption and on the Efficacy of

Iron-Fortified Foods
Recent studies provide good evidence that the acute-phase response to malarial
infection [18] increases cytokines such as IL-6 which upregulate hepcidin synthe-
sis [19] and decrease iron absorption. Doherty et al. [20] investigated iron absorp-
tion from an iron-fortified orange juice in young Gambian children 1 day and 15
days after the treatment of malaria. The acute-phase response to the infection, as
indicated by a raised serum ferritin in anemic children, was still present on day 1
after treatment but was much decreased at day 15 (table 1). In this study, thirty-
seven 8- to 36-month-old, anemic (Hb <110 g/l) children with uncomplicated
malaria followed a 3-day treatment with chloroquine and fansidar. On day 1 after
the treatment, they consumed an orange juice fortified with iron and ascorbic
acid and labeled with 57 iron stable isotope as ferrous sulfate. Iron absorption
was quantified by measuring the incorporation of the stable isotope into eryth-
rocytes at 14 days. On day 15 after treatment, a second iron absorption measure-
ment was made on the same children who again consumed an iron- and ascorbic
acid-fortified orange juice, but this time labeled with 58 ferrous sulfate. Fourteen
days later, erythrocyte incorporation of the isotope was similarly quantified. The
control group consisted of anemic children in the same age range but with no
110 Hurrell
malaria and no malarial treatment. From day 1 to 30, all children received iron
supplementation with liquid iron glycine sulfate at 2 mg Fe/kg bodyweight, and
the hemoglobin and serum ferritin response of the postmalarial and nonmalarial
anemic children to iron supplementation was compared.
Selected results from this study are summarized in table 1. The serum fer-
ritin values have been recalculated from the published graphics. Iron absorption
in the children who had been treated for malaria increased from 8.7% on day 1
after treatment to 15.5% on day 15, but on both days iron absorption was still sig-
nificantly lower than in the noninfected control children, whose absorption was
ca. 25% (p < 0.001). Serum ferritin in the children treated for malaria decreased
from 74 μg/l on day 1 to 22 μg/l on day 15 after treatment, even though the
children received iron supplements over this period. Thus, the malarial treat-
ment appeared to have decreased the acute-phase response, although not com-
pletely. Providing iron supplements to the control children over the same time
period increased their serum ferritin from 6 to 16 μg/l. Although the children
treated for malaria absorbed less iron than the controls over the supplementa-
tion period, they showed a much greater increase in hemoglobin concentration
between days 1–15 (p < 0.001) and between days 16–30 (p < 0.001). This can
be explained by an increased store of macrophage iron from the destruction of
senescent red cells being prevented by hepcidin from entering the serum during
the malarial infection, but being released following malarial treatment.
In contrast to the Gambian study which investigated iron absorption in chil-
dren with uncomplicated febrile malaria, a more recent study in Benin inves-
tigated iron absorption in women with malarial parasitemia but no fever or
other symptoms. While acute febrile malaria affects individuals only a few days
a year, in areas of perennial transmission, asymptomatic parasitemia affects
much of the population for most of the year giving rise to protracted low-level
inflammation [21]. In the Benin study [12], women with relatively high loads of
Plasmodium falciparum (>500 parasites/μl blood) but no fever or other infec-
tions, consumed a sorghum porridge labeled with 57 ferrous sulfate before and
after a 10-day treatment with MalaroneTM. Iron absorption was measured as
incorporation of the stable isotopes into hemoglobin 14 days after consumption
of the test meals. Hepcidin and other inflammation indices were measured at
baseline and on day 25 after clearance of the infection and immediately prior to
the second test meal. The results are shown in table 2. Treatment of the malar-
ial parasitemia increased mean iron absorption from 10.2 to 17.6% (p < 0.01),
which could be explained by a decrease in inflammation as demonstrated by a
decrease in IL-6, -8 and -10 and an almost 50% fall in plasma hepcidin.
From these studies, it would be expected that inflammatory disorders and
infections that lead to an inflammatory response would lead to a hepcidin-
modulated decrease in iron absorption. Any long-term or chronic condition
would then be expected to blunt the absorption of iron-fortified foods and the
efficacy of iron fortification programs. To date, the only infection investigated

Table 2. Influence of malarial parasitemia on fractional iron absorption and indicators of
inflammation in young Beninese women consuming an iron-fortified sorghum porridge [12]
With malarial After treatment: p value

parasitemia no parasitemia
Parasitemia, parasites/μl 880 (123–2,760) 0 (0) <0.01

Percent iron absorption 10.2 (4.42; 23.5) 17.6 (9.17; 33.8) <0.01
Serum ferritin, μg/l 71 (29–99) 37 (21–57) <0.001
Hepcidin, nmol/l 2.7 (1.0–4.6) 1.4 (0.7–2.4) <0.005
IL-6, pg/ml 1.32 (0.96–1.92) 1.27 (0.7–1.32) <0.05
IL-8, pg/ml 7.31 (4.58–10.0) 4.18 (2.60–5.67) <0.001
IL-10, pg/ml 7.38 (4.44–13.9) 2.9 (1.91–2.94) <0.001
All values are given as median with 25–75 percentiles in parentheses, except for iron absorp-
tion, which is a geometric mean (±SD).
in relation to iron absorption is malaria, although other common infections

such as tuberculosis [22] and Schistosoma haematobium [23] also lead to an
inflammatory response and increase hepcidin concentrations. There is no direct
evidence however that the efficacy of iron fortification programs is blunted in
malaria-endemic areas or regions of widespread infections. Lack of efficacy of
electrolytic iron-fortified cereal products in Ivory Coast [24] and Kenya [25] has
been blamed on widespread infections, but lack of gastric acid in malnourished
populations could be an alternative explanation. Electrolytic iron-fortified wheat
flour products were efficacious in China and Thailand [26]. While no studies
have directly compared efficacy in infected and noninfected subjects from the
same populations, there are efficacy studies with similarly fortified salt or cereals
which have been made in populations with little or no infection or with wide-
spread infections. The salt studies with ferric pyrophosphate in Morocco (no
infections) [27] and Ivory Coast [28] reported similar improvements in serum
ferritin and transferrin receptor as did the NaFeEDTA-fortified cereal studies
in India (no infections) [Muthayya et al., unpubl.] and Kenya [25]. This failure
to demonstrate reduced efficacy in the presence of infection is probably due to
differences in baseline iron status and other physiological differences in the sub-
jects, as well as differences in food vehicle, diet, and fortification levels.
Obesity and Overweight in Relation to Iron Absorption and the Efficacy of

Iron-Fortified Foods
American national surveys have consistently shown that overweight toddlers,

children, adolescents and adults are more likely to be iron deficient than their
112 Hurrell
1.8
y = –0.029x + 1.5032
R2 = 0.051
Log fractional Fe absorption (%) 1.5
1.2
0.9
0.6
0.3
0.0
15 18 21 24 27
BMI
Fig. 1. Relationship between log fractional iron absorption and body mass index in
healthy premenopausal Thai women (n = 92) who consumed meals of rice and vegetables
labeled with ≈4 mg of [57Fe/58Fe]-ferrous sulfate [17].
normal weight counterparts, and recent evidence suggests that adiposity-related

inflammation may play a central role through its regulation of hepcidin [11].
In a recent Swiss study, overweight children were reported to consume similar
amounts of bioavailable iron as normal weight children but have a lower iron
status, higher serum hepcidin and higher subclinical inflammation as measured
by IL-6 and C-reactive protein [29]. The adipokine leptin was also upregulated,
and both IL-6 and leptin can increase hepcidin expression [11].
We have earlier reported that adiposity in women and children in transition
countries predicts decreased iron absorption, iron deficiency and a decreased
response to iron fortification [17]. The stable isotope iron absorption studies
were made in 92 premenopausal Thai women of which some 20% were iron defi-
cient and 22% overweight. They all consumed a reference meal of vegetable soup
and white rice labeled with 57 or 58 ferrous sulfate. Iron absorption was quanti-
fied from erythrocyte incorporation of the isotope after 14 days. Independent
of iron status, a higher BMI was associated with a lower iron absorption (p <
0.03; fig. 1). After adjustment for the differences in iron status (serum ferritin),
fractional iron absorption was negatively correlated with C-reactive protein (p <
0.001) and BMI (p < 0.05).
In order to evaluate the influence of adiposity on the efficacy of iron fortifica-
tion programs, we analyzed baseline (n = 1,688) and intervention (n = 727) data
from previously published efficacy studies in schoolchildren in Morocco and
India to look for associations between BMI z scores and iron status measures.
In this cohort of children, 42% were iron deficient but only 6.3% overweight;

4
y = –0.0509x – 0.7041
R2 = 0.0246
3
1
BMI Z score
0
–10 –5 0 5 10 15 20
–1
–2
–3
–4
–5
Change in body Fe (mg/kg bodyweight) during intervention
Fig. 2. BMI z score and the change in body iron stores (calculated from the ratio of SF/TfR)
in iron fortification intervention studies in Moroccan and Indian children (n = 727) [17].
however, a lower BMI z score predicted poorer iron status at baseline (p < 0.001)
and less improvement in iron status on consumption of the iron-fortified food
(fig. 2). Iron status was measured as the serum ferritin-to-transferrin receptor
ratio. This influence of adiposity on iron intervention has been confirmed in a
recent Italian study in 15 obese children which reported that weight loss was
associated with a significant decrease in circulating hepcidin and an increased
response to iron supplements [30]. Overweight is increasing in some low-
income countries at 2–4 times the rate in the developing world, and the high
levels of overweight in women and school children in transition countries such
as Thailand and India [11] could reduce the impact of iron interventions.
Conclusions
Inflammatory disorders and inflammation associated with infections would be

expected to decrease iron absorption and reduce the efficacy of iron-fortified
foods. The likely explanation is that the passage of iron from the intestinal cell to
the plasma is prevented by the degradation of the transport protein ferroportin
by higher hepcidin levels produced as a response to the inflammatory cytokines.
While iron absorption has been reported to be much reduced in response to
114 Hurrell
chronic malarial parasitemia and acute malarial infection, there is no evidence
to confirm a reduced efficacy of iron-fortified foods in areas of widespread
malaria and other infections. In contrast, increased inflammation and circulat-
ing hepcidin levels as a result of adiposity have been linked to both decreased
iron absorption and a reduced efficacy of iron-fortified foods. While this might
be due to the additional influence of the adipokine leptin in further increas-
ing hepcidin levels in overweight individuals, it seems likely that inflammation
associated with infections will decrease the impact of iron-fortified foods, but
well-planned studies have yet to be made.
References
1 World Health Organization/Food and 10 Nemeth E, Valore EV, Territo M, et al:
Agricultural Organization: Guidelines on Hepcidin, a putative mediator of anemia of
Food Fortification with Micronutrients. inflammation, is a type II acute-phase pro-
Geneva, World Health Organization, 2006. tein. Blood 2003;101:2461–2463.
2 Zimmermann MB, Hurrell RF: Nutritional 11 Cepeda-Lopez AC, Aeberli I, Zimmermann
iron deficiency. Lancet 2007;370:511–520. MB: Does obesity increase risk for iron
3 Cook JD: Diagnosis and management of deficiency? A review of the literature and the
iron-deficiency anaemia. Best Pract Res Clin potential mechanisms. Int J Vitam Nutr Res
Haematol 2005;18:319–332. 2010;80:263–270.
4 Petry N, Egli I, Zeder C, et al: Polyphenols 12 Cercamondi CI, Egli IM, Ahouandjinou E,
and phytic acid contribute to the low iron et al: Afebrile Plasmodium falciparum para-
bioavailability from common beans in young sitemia decreases absorption of fortification
women. J Nutr 2010;140:1977–1982. iron but does not affect systemic iron utiliza-
5 Hallberg L, Brune M, Rossander L: Iron tion a double stable-isotope study in young
absorption in man: ascorbic acid and dose- Beninese women. Am J Clin Nutr 2010;92:
dependent inhibition by phytate. Am J Clin 1385–1392.
Nutr 1989;49:140–144. 13 Donovan A, Lima CA, Pinkus JL, et al: The
6 Hallberg L, Brune M, Erlandsson M, et al: iron exporter ferroportin/Slc40a1 is essential
Calcium: effect of different amounts on non- for iron homeostasis. Cell Metab 2005;1:191–
heme- and heme-iron absorption in humans. 200.
Am J Clin Nutr 1991;53:112–119. 14 Ganz T: Hepcidin – a regulator of intesti-
7 Troesch B, Egli I, Zeder C, et al: Optimiz- nal iron absorption and iron recycling by
ation of a phytase-containing micronutrient macrophages. Best Pract Res Clin Haematol
powder with low amounts of highly bio- 2005;18:171–182.
available iron for in-home fortification of 15 Nemeth E, Tuttle MS, Powelson J, et al:
complementary foods. Am J Clin Nutr 2009; Hepcidin regulates cellular iron efflux by
89:539–544. binding to ferroportin and inducing its inter-
8 Cook JD, Dassenko SA, Lynch SR: nalization. Science 2004;306:2090–2093.
Assessment of the role of nonheme-iron 16 Wander K, Shell-Duncan B, McDade TW:
availability in iron balance. Am J Clin Nutr Evaluation of iron deficiency as a nutritional
1991;54:717–722. adaptation to infectious disease: an evolu-
9 Hurrell R, Egli I: Iron bioavailability and tionary medicine perspective. Am J Hum
dietary reference values. Am J Clin Nutr Biol 2009;21:172–179.
2010;91:1461s–1467s.

17 Zimmermann MB, Zeder C, Muthayya S, et 24 Rohner F, Zimmermann MB, Amon RJ, et
al: Adiposity in women and children from al: In a randomized controlled trial of iron
transition countries predicts decreased iron fortification, anthelmintic treatment, and
absorption, iron deficiency and a reduced intermittent preventive treatment of malaria
response to iron fortification. Int J Obes for anemia control in Ivorian children, only
(Lond) 2008;32:1098–1104. anthelmintic treatment shows modest ben-
18 O’Donnell A, Fowkes FJI, Allen SJ, et al: The efit. J Nutr 2010;140:635–641.
acute phase response in children with mild 25 Andang’o PEA, Osendarp SJM, Ayah R, et al:
and severe malaria in Papua New Guinea. Efficacy of iron-fortified whole maize flour
Trans R Soc Trop Med Hyg 2009;103:679–686. on iron status of schoolchildren in Kenya:
19 de Mast Q, van Dongen-Lases EC, Swinkels a randomised controlled trial. Lancet 2007;
DW, et al: Mild increases in serum hepcidin 369:1799–1806.
and interleukin-6 concentrations impair iron 26 Hurrell R, Ranum P, de Pee S, et al: Revised
incorporation in haemoglobin during an recommendations for iron fortification of
experimental human malaria infection. Br J wheat flour and an evaluation of the
Haematol 2009;145:657–664. expected impact of current national wheat
20 Doherty CP, Cox SE, Fulford AJ, et al: Iron flour fortification programs. Food Nutr Bull
incorporation and post-malaria anaemia. 2010;31:S7–S21.
PLoS One 2008;3:e2133. 27 Zimmermann MB, Wegmueller R, Zeder C,
21 Imrie H, Fowkes FJ, Michon P, et al: Low et al: Dual fortification of salt with iodine
prevalence of an acute phase response in and micronized ferric pyrophosphate: a ran-
asymptomatic children from a malaria- domized, double-blind, controlled trial. Am J
endemic area of Papua New Guinea. Am J Clin Nutr 2004;80:952–959.
Trop Med Hyg 2007;76:280–284. 28 Wegmueller R, Fatoumata C, Zimmermann
22 Vander Beken S, Al Dulayymi JR, Naessens MB, et al: Salt dual fortified with iodine and
T, et al: Molecular structure of the Myco- micronized ground ferric pyrophosphate
bacterium tuberculosis virulence factor, affects iron status but not hemoglobin in chil-
mycolic acid, determines the elicited dren in Côte d’Ivoire. J Nutr 2006;136:1–7.
inflammatory pattern. Eur J Immunol 29 Aeberli I, Beljean N, Lehmann R, I’Allemand
2011;41:450–460. D, Spinas GA, Zimmermann MB: The
23 Ayoya MA, Spiekermann-Brouwer GM, increase of fatty acid-binding protein aP2 in
Stoltzfus RJ, et al: Alpha 1-acid glycoprotein, overweight and obese children: interactions
hepcidin, C-reactive protein, and serum fer- with dietary fat and impact on measures of
ritin are correlated in anemic schoolchildren subclinical inflammation. Int J Obes (Lond)
with Schistosoma haematobium. Am J Clin 2008;32:1513–1520.
Nutr 2010;91:1784–1790. 30 Amato A, Santoro N, Calabro P, et al: Effect
of body mass index reduction on serum hep-
cidin levels and iron status in obese children.
Int J Obes (Lond) 2010;34:1772–1774.
116 Hurrell
Iron
Safety of Iron Fortification and

Supplementation in Malaria-Endemic
Areas
Gary M. Brittenham
Division of Pediatric Hematology, Department of Pediatrics, Columbia University, New York, NY, USA
Abstract
This review considers the safety of iron supplementation and fortification for the preven-
tion and correction of iron deficiency in malaria-endemic areas, with a focus on potential
means whereby provision of additional iron might heighten the risks of malaria and other
infections. Iron deficiency itself may increase the risk of morbidity and mortality from
malaria and other infections. The available evidence indicates that iron interventions are
safe in settings without endemic malaria, and, with adequate health care, in regions with
high transmission of malaria and other infections. Without regular surveillance and treat-
ment of malaria and other infections, iron supplementation of individuals who are iron
deficient seems safe, but individuals who are iron replete may have an increased risk of
adverse outcomes. The mechanisms responsible for harmful effects with iron supplemen-
tation have not been established. These are likely to include the effects of (a) increased
amounts of absorbed iron, with the production of plasma non-transferrin-bound iron, (b)
increased amounts of iron in the gastrointestinal tract, with effects on gastrointestinal
structural integrity and on gut microflora, and (c) the complex immune effects of iron
interventions. Iron fortification appears to be generally safe, although more data from
malaria-endemic areas are needed.
Introduction
This review examines the safety of iron supplementation and fortification for
the prevention and correction of iron deficiency in malaria-endemic areas.
The focus is on potential means whereby provision of additional iron might
heighten the risks of malaria and other infections. In settings without endemic
malaria, studies have generally found no indication of iron-induced increases
in infectious illness [1–5]. After summarizing the available evidence from pro-
spective trials of iron interventions in malarial areas, proposed mechanisms of
adverse interactions of iron interventions with malaria and other infections will
be considered.
Iron supplements are iron preparations given orally to population groups for
prevention and treatment of iron deficiency [6], typically without food in doses
of 1–2 mg/kg bodyweight. Iron fortification usually has the primary goal of pre-
venting iron deficiency and includes both (a) the addition of iron to staple foods
or condiments during processing and (b) the use of iron-containing powders,
crushable tablets or spreads that are intended to be added to complementary
foods prepared at home [6]. Iron-fortified processed foods and condiments pro-
vide lower levels of iron than supplements and are generally for adults, older
children and adolescents. Iron preparations for home fortification supply inter-
mediate levels of iron and are mainly for infants and young children. Iron for-
tification has generally been considered to be safe, but has not been specifically
evaluated in malaria-endemic areas.
Iron Interventions, Malaria and Other Infections
Nearly half the world’s population now lives in regions endemic for malaria in
Africa, Asia and South America [7, 8]. In these same regions, iron deficiency, the
most common form of malnutrition, further increases the risks of disability and
death [9]. Lack of iron causes anemia, impairs cognitive and behavioral devel-
opment, decreases work capacity, and when severe, increases mortality during
pregnancy, infancy and childhood [10]. Iron deficiency and malaria usually are
found together with other nutritional deficiencies and infectious pathogens, as
well as with inherited red blood cell abnormalities, such as the thalassemias and
sickle cell disorders. In these settings, iron nutrition can be improved by modify-
ing the diet, by fortifying foods, treating helminth infections and by other public
health interventions. Along with these measures, supplemental iron is needed
for prevention and correction of iron deficiency, especially among infants, chil-
dren and pregnant women [11].
Concerns about the safety of programs to provide additional iron have been
long standing [1, 12]. Earlier studies had generally suggested that the benefits
of iron supplementation exceeded the risks [2, 13]. In 2006, publication of the
results of a large, randomized, controlled trial in Pemba, Zanzibar, Tanzania
(hereafter, the Pemba trial) [14], was followed by a sustained reexamination
of the risks of iron supplementation and fortification. Because of the impor-
tance of these results, the design and major findings of the Pemba trial will be
outlined, together with subsequent appraisals of the safety of iron interven-
tions in malaria-endemic regions by a World Health Organization (WHO)
118 Brittenham
Consultation, by a systematic review from the Cochrane Collaboration, and by
a Technical Working Group (TWG) convened by the Eunice Kennedy Shriver
National Institute of Child Health and Human Development of the US National
Institutes of Health and the Bill and Melinda Gates Foundation.
Pemba Main Study

The Pemba trial was a randomized, placebo-controlled trial of children aged
1–35 months living in Pemba, Zanzibar, Tanzania [14], in a setting where sub-
optimum care seeking and treatment in the routine public health system was
associated with high mortality from infectious diseases. Children were assigned
to daily oral supplementation with iron (12.5 mg) and folic acid (50 μg), iron,
folic acid, and zinc, or placebo; children aged 1–11 months received half the
dose. Most children were given supplements between meals. Based on available
weights, the mean (SD) iron dose in mg/kg per day for infants 0–5 months old
was 0.92 (0.23), for infants 6–11 months old was 0.80 (0.14), and for children
aged 12 months or older was 1.15 (0.23). On the recommendation of the data
and safety monitoring board, the iron and folic acid-containing groups of the
trial were stopped early. At that time, 24,076 children contributed a follow-up of
25,524 child-years. Those who received iron and folic acid with or without zinc
were 12% (95% CI: 2–23%, p = 0.02) more likely to die or need treatment in hos-
pital for an adverse event and 11% (1–23%, p = 0.03) more likely to be admit-
ted to hospital; there were also 15% (–7 to 41%) more deaths in these groups,
although this difference was not statistically significant (p = 0.19). Overall, the
analyses of cause-specific admissions to hospital and deaths in the main study
indicated that malaria and other infectious diseases were significantly increased
in children given iron and folic acid.
Pemba Substudy
The Pemba substudy included 2,413 children in whom baseline iron status and
anthropometry were assessed. Although not an explicit aim, the children in
the substudy were also provided with active surveillance for malaria and free
treatment for infection in an attempt to balance the benefits and burdens of
study participation [4, 15]. In the children enrolled in the substudy, iron status
at baseline seemed to modify the risk of adverse events. An elevated erythrocyte
zinc protoporphyrin (≥80 μmol/mol heme) was used as the criterion for iron
deficiency. As shown in figure 1, iron deficiency greatly increased the risk of
severe illness and death. The rate of adverse events (hospitalizations and deaths)
in iron-deficient children treated with placebo (9.76 events/100 child-years) was
increased by 202% compared to that in iron-replete children (4.83 events/100
child-years; p = 0.04). Iron and folate supplementation benefited iron-deficient
children, reducing their rate of adverse events by 38% (to 6.00 events/100 child-
years; p = 0.02). By contrast, iron and folate supplementation seemed to harm
iron-replete children, increasing their risk of severe illness and death by 63% (to
Safety of Iron Fortification and Supplementation 119

Deaths and hospitalizations
10
100 child-years
8
6
4
2 ZnPP ≥ 80
0
ZnPP < 80
Placebo
Iron + folate
Fig. 1. Effects of supplementation with iron and folic acid on adverse events by iron
status as assessed by erythrocyte zinc protoporphyrin (ZnPP) in the Pemba substudy
among 2,413 children [14]. Iron and folate supplementation benefited iron-deficient
(ZnPP ≥ 80 μmol/mol heme) children but harmed iron-replete (ZnPP < 80) children.
7.86 events/100 child-years), although this result was not statistically significant
(p = 0.24). Unlike the increase in adverse events observed in the main study, the
overall effect of supplementation with iron and folic acid in the substudy was a
non-significant reduction in adverse events. The different results between the
main study and the substudy were interpreted as the result of more intensive
diagnosis and management of children with malaria and other infections in the
substudy. Although the increase in adverse events in iron-replete children given
iron and folate in the substudy was not statistically significant, the investigators
used this result to suggest that the increase in hospitalizations and deaths in the
main study was the consequence of giving iron and folic acid supplementation
to iron-replete children [14].
WHO Consultation on Prevention and Control of Iron Deficiency in Infants and

Young Children in Malaria-Endemic Areas
After publication of the results of the Pemba trial, the WHO and the United
Nations Children’s Fund (UNICEF) issued a joint statement [16], and convened
a WHO Consultation to examine the results of the Pemba trial. The WHO
Consultation then made a series of recommendations:
‘Universal iron supplementation for children under the age of 2 years is not
recommended in malaria-endemic areas. However, iron therapy may have an important
positive impact on child survival if it is directed to iron-deficient children in the setting
of appropriate treatment of malaria and the complicating bacterial infections. Prior
screening to identify iron-deficient children should be a necessary component of any
such intervention’ [6].
‘The safety of iron preparations administered through home fortification of
complementary foods for infants and young children, i.e. powders, crushable tablets,
and fat-based spreads, is uncertain in malaria-endemic areas. Although there is reason
120 Brittenham
to believe that those preparations may be safer than iron supplements, they cannot be
recommended until this has been demonstrated’ [6].
The WHO Consultation also noted that the safety of the use of processed foods
fortified with iron in malarial areas had not been documented. Overall, because
of the difficulties of diagnosing iron deficiency in resource-limited settings, the
practical consequence of following these recommendations would be to halt pro-
grams of iron supplementation and fortification in malaria-endemic areas.
Cochrane Review: Oral Iron Supplementation for Preventing or Treating Anemia

among Children in Malaria-Endemic Areas
A subsequent systematic review, published in 2009, examined additional data
and information that were not included in the WHO Consultation [5]. The
meta-analysis found that (a) the risk of clinical malaria was not increased by
iron supplementation (RR: 1.00, 95% CI: 0.88–1.13; 22,724 children, 14 tri-
als) and (b) the risk was similar among children who were non-anemic (and
thus less likely to be iron deficient) at baseline (RR: 0.96, 95% CI: 0.85–1.09).
Nonetheless, in trials that did not provide malaria surveillance and treatment,
an increased risk of malaria with iron was observed (RR: 1.16, 95% CI: 1.03–
1.31). Growth and other infections were generally not affected by iron supple-
mentation. The conclusions of the review were that:
‘Iron does not increase the risk of clinical malaria or death, when regular malaria
surveillance and treatment services are provided. There is no need to screen for anæmia
prior to iron supplementation’ [5].
‘Based on our review, routine iron supplementation should not be withheld for
children living in malaria-endemic areas. Iron supplementation in malaria-endemic
areas need not be based on baseline iron assessment; rather, emphasis should be on
appropriate malaria prevention and surveillance’ [5].
National Institute of Child Health and Human Development /Gates Foundation

Technical Working Group Review
For regions with a high prevalence of malaria and other infections, the TWG
evaluated potential mechanisms that might be responsible for adverse effects of
iron, assessed the usefulness of biomarkers for assessing iron status, and reviewed
the available evidence for the safety and effectiveness of iron supplementation
and fortification [17], taking account of the results of the WHO Consultation
and Cochrane Review. The TWG Iron and Malaria Technical Report [17]
identified an overall consistency between the findings of the Pemba trial and
the conclusions of previous reviews, the WHO Consultation, and the Cochrane
review. All agreed that when iron supplementation is given in settings with inad-
equate health care services, the risk of malaria and other infections is increased.
Iron supplementation with adequate medical surveillance and prompt treat-
ment of malaria and other infections appeared to be safe. Altogether, the TWG
noted the practical difficulties of determining the adequacy of health care and

emphasized the need for further research to provide an improved understand-
ing of the mechanisms underlying the adverse effects of iron interventions.
Potential Mechanisms for Adverse Effects of Iron in Settings with a High

Prevalence of Malaria and Other Infections and Inadequate Heath Care
The adverse effects of iron observed in trials of iron supplementation almost

certainly involve a complex interaction between iron status, the specific iron
intervention, and malaria and other infections. Broadly, underlying potential
mechanisms for adverse effects with iron supplementation or fortification may
be considered as those resulting from (a) the increased amounts of iron absorbed,
(b) the increased amounts of iron in the gastrointestinal tract, and (c) the immune
effects of iron interventions. While a single mechanism might underlie the harm-
ful effects of iron administration reported in the Pemba and other trials, a variety
of mechanisms acting in concert is more likely to be responsible.
Mechanisms for Adverse Effects of Iron Interventions Involving Increased Iron

Absorption
The WHO Consultation [6] and the TWG report [17] identified plasma non-
transferrin-bound iron as a probable cause of the adverse events in the Pemba
trial:
‘When bolus doses of iron are administered parenterally or orally, especially without
food, they may increase plasma iron concentrations and transferrin saturation and
exceed the binding capacity of transferrin, leading to the appearance of non-transferrin-
bound iron (NTBI). NTBI is potentially toxic because it may promote free radical
formation and be more readily available to pathogens’ [6].
While direct toxic effects mediated by harmful free radical reactions [18, 19]
could be involved, analyses of cause-specific admissions to hospital and deaths
in the Pemba main study identified malaria and other infectious diseases as the
most common causes of hospitalization and death in children given iron and
folic acid [14].
For nonmalarial infections, the most likely explanation for an increased risk
would seem to be plasma non-transferrin-bound iron being available to pathogens
reaching the blood stream and enhancing their growth [20]. Concomitant bacte-
remia is common in patients with malaria infection. Bacterial virulence is greatly
enhanced by freely available iron and the appearance of plasma non-transferrin-
bound iron could transform bacteremia into a serious or fatal infection.
With respect to the potential adverse effects of plasma non-transferrin-bound
iron on malarial infection, direct donation of iron to Plasmodium falciparum
seems unlikely. The available evidence indicates that the growth of P. falciparum
in infected red blood cells is independent of host iron status [19]. Extracellular
iron does not seem to be required for intra-erythrocytic parasite growth. Iron
122 Brittenham
used by the parasite seems to be transported directly across the parasite plasma
membrane from the infected red blood cell pool of cytosolic iron rather than
from the digestive vacuole or from heme [21].
Plasma non-transferrin-bound iron has been reported to increase the expres-
sion of vascular endothelial adhesion molecules involved in sequestration of
P. falciparum [22], providing a potential explanation for the more severe clinical
course in children in the Pemba trial [14] given iron and folic acid. Increased
sequestration of infected erythrocytes would increase the risk of more severe
forms of malaria, especially of cerebral malaria, by worsening microvascular
obstruction. Another potential mechanism would involve the hepatic phase of
malarial infection [21]. Plasma non-transferrin-bound iron is rapidly removed
from the portal circulation by hepatocytes. Increased iron availability within
hepatocytes might enhance merozoite production, resulting in higher para-
sitemia during the erythrocytic phase of the infection.
To date, measurements of plasma non-transferrin-bound iron have not
been reported in patients with malaria. In noninfected subjects, oral adminis-
tration of doses of iron of about 1 mg/kg bodyweight, similar to those used in
the Pemba trial, resulted in the appearance of plasma non-transferrin-bound
iron in iron-deficient, anemic women [23] and in volunteers with normal
iron stores [24]. Our own studies [25] in women with low iron stores (serum
ferritin <25 μg/l), summarized graphically in figure 2, have confirmed the
appearance of plasma non-transferrin-bound iron after oral administration of
ferrous sulfate, 60 mg iron, or about 1 mg iron/kg, similar to the dose of iron
used in the Pemba trial. Giving the same dose of iron with a meal (rice with
a vegetable sauce) greatly reduced the amounts of plasma non-transferrin-
bound iron formed. Giving a dose of iron of 6 mg/kg, an amount comparable
to that used in iron fortification, with the same meal failed to produce a sig-
nificant increase in plasma non-transferrin-bound iron. Similar studies are
needed in subjects in malaria-endemic areas, but these preliminary results
suggest that oral iron supplements can produce plasma non-transferrin-
bound iron, that administration of iron supplements with food can decrease
the amounts formed, and that lower doses, like those used in iron fortifi-
cation, given with meals can further reduce or eliminate the production of
plasma non-transferrin-bound iron.
Mechanisms for Adverse Effects of Iron Interventions Involving Increased

Amounts of Iron in the Gastrointestinal Tract
Because only a fraction of supplementation and fortification iron is absorbed,
most of the dose given passes into the lower small intestine and colon. Iron
and iron status affect the structural and immunological integrity of the gastro-
intestinal tract as well as the gastrointestinal microflora, potentially promoting
invasion by pathogenic enteric bacteria. Because P. falciparum sequesters within
the gastrointestinal microvasculature, the inflammatory effects of malaria could

A – 60 mg Fe suppl.
3.0 B – 60 mg Fe suppl. + rice/veg. meal
C – 6 mg fortification Fe in rice/veg. meal
2.5
2.0
NTBI (μmol/l)
1.5
1.0
0.5
0.0
0 1 2 3 4 5 6 7 8
–0.5
Time after meal consumption (h)
Fig. 2. Plasma non-transferrin-bound iron (NTBI) depends on the quantity of iron con-
sumed and the presence of food [25]. In brief, 16 women with low iron stores (serum fer-
ritin <25 μg/l) were administered 60 mg iron as a ferrous sulphate solution, with and
without a meal of rice and vegetable sauce, or the same meal without a supplement but
containing 6 mg iron as ferrous sulphate added as fortification iron to the meal. On com-
bining results from the three test administrations, there was a strong correlation between
increase in serum iron and the appearance of NTBI (R2 = 0.58).
interact with oxidant effects of iron on the gut wall, facilitating systemic bacte-
rial invasion, bacteremia, and septicemia. Increased intestinal mucosal perme-
ability has been reported in iron-supplemented Zambian children [26].
The normal gastrointestinal tract is populated by beneficial barrier bacte-
ria that help protect against infection by preventing pathogenic colonization.
Increases in gastrointestinal iron could alter this balance by enhancing the
colonization and virulence of pathogenic enterobacteria [27]. Some studies have
reported an increase in diarrhea in groups receiving iron supplements [2]. In a
recent trial in African children, administration of iron-fortified biscuits resulted
in a decrease in protective lactobacilli and an increase in fecal enterobacteria,
together with an increase in mean fecal calprotectin, an indicator of gut inflam-
mation [28]. In this population, iron fortification seemed to result in a more
pathogenic gastrointestinal microbiota in association with evidence of increased
gastrointestinal inflammation.
124 Brittenham
Mechanisms for Adverse Effects of Iron Interventions Involving Immune Effects of
Iron Interventions
Both iron and malaria modify host immune responses in complex and still
poorly characterized fashions. An extensive cross-regulatory network inter-
connects iron metabolism and immune function. Iron alters cytokine secre-
tion and transcription factors involved in the immune response, while both
cellular and systemic iron homeostasis are modulated by acute-phase proteins
and mediators derived from immune cells [29]. The withholding of iron from
pathogens is a central component of host defense. Supplemental and forti-
fication iron, by compromising iron withholding and interfering with the
regulation and coordination of innate and adaptive immune defenses both
systemically and within the gastrointestinal tract, might impair defenses
against a variety of pathogens. In addition, malaria-induced dysregulation of
innate and adaptive immune responses also interferes with protection against
a variety of microorganisms [30].
Conclusion
Iron deficiency itself may increase the risk of morbidity and mortality from malaria
and other infections. The available evidence indicates that efforts to prevent and
control iron deficiency by iron supplementation and fortification are safe in set-
tings without endemic malaria, and, with adequate health care, in regions with
high transmission of malaria and other infections. Without regular surveillance
and treatment of malaria and other infections, iron supplementation of individuals
who are iron deficient seems safe, but individuals who are iron replete may have
an increased risk of adverse outcomes. The mechanisms responsible for adverse
effects with iron supplementation have not been established. These are likely to
include the results of (a) increased amounts of iron absorbed, with the produc-
tion of plasma non-transferrin-bound iron, (b) increased amounts of iron in the
gastrointestinal tract, with effects on gastrointestinal structural integrity and on
gut microflora, and (c) the complex immune effects of iron interventions. Iron for-
tification appears to be generally safe, although more data from malaria-endemic
areas are needed. Further research is needed to provide an improved understand-
ing of the mechanisms underlying the adverse effects of iron interventions.
Acknowledgements
Supported, in part, by grants from the US National Institutes of Health, including the
Eunice Kennedy Shriver National Institute of Child Health and Human Development,
the NIH Office of Dietary Supplements, the National Institute of Neurological Disorders
and Stroke, the Fogarty International Center and from the Bill and Melinda Gates
Foundation Gift Fund (U01 HD061233; R21 U01 HD064827; NS055348).

References
1 Oppenheimer SJ: Iron and its relation to 13 INACG: Consensus Statement: Safety of
immunity and infectious disease. J Nutr Iron Supplementation Programs in Malaria-
2001;131:616S–633S, discussion 633S–635S. Endemic Regions. Washington, ILSI, 1999.
2 Gera T, Sachdev HP: Effect of iron supple- 14 Sazawal S, Black RE, Ramsan M, et al: Effects
mentation on incidence of infectious ill- of routine prophylactic supplementation with
ness in children: systematic review. BMJ iron and folic acid on admission to hospital
2002;325:1142. and mortality in preschool children in a high
3 Tielsch JM, Khatry SK, Stoltzfus RJ, et malaria transmission setting: community-
al: Effect of routine prophylactic supple- based, randomised, placebo-controlled trial.
mentation with iron and folic acid on pre- Lancet 2006;367:133–143.
school child mortality in southern Nepal: 15 Stoltzfus R: Commentary: Cochrane Review
community-based, cluster-randomised, on Oral iron supplementation for prevent-
placebo-controlled trial. Lancet 2006;367: ing or treating anaemia among children in
144–152. malaria-endemic areas. Int J Epidemiol 2010;
4 Stoltzfus RJ, Heidkamp R, Kenkel D, Habicht 39:34–35.
JP: Iron supplementation of young children: 16 World Health Organization, UNICEF:
learning from the new evidence. Food Nutr Iron Supplementation of Young Children
Bull 2007;28:S572–S584. in Regions Where Malaria Transmission
5 Ojukwu JU, Okebe JU, Yahav D, Paul M: Is Intense and Infectious Disease Highly
Oral iron supplementation for preventing or Prevalent. Geneva, World Health
treating anaemia among children in malaria- Organization 2006.
endemic areas. Cochrane Database Syst Rev 17 Raiten D, Namasté S, Brabin B:
2009;CD006589. Considerations for the safe and effective use
6 WHO Secretariat on behalf of the partici- of iron interventions in areas of malaria bur-
pants to the Consultation: Conclusions and den. Int J Vitam Nutr Res 2011;81:57–71.
recommendations of the WHO consulta- 18 Esposito BP, Breuer W, Sirankapracha P, et
tion on prevention and control of iron al: Labile plasma iron in iron overload: redox
deficiency in infants and young children in activity and susceptibility to chelation. Blood
malaria-endemic areas. Food Nutr Bull 2007; 2003;102:2670–2677.
28:S621–S627. 19 Hershko C: Mechanism of iron toxicity. Food
7 Hay SI, Guerra CA, Tatem AJ, et al: The Nutr Bull 2007;28:S500–S509.
global distribution and population at risk 20 Ratledge C: Iron metabolism and infection.
of malaria: past, present, and future. Lancet Food Nutr Bull 2007;28:S515–S523.
Infect Dis 2004;4:327–336. 21 Scholl PF, Tripathi AK, Sullivan DJ:
8 Snow RW, Guerra CA, Noor AM, et al: The Bioavailable iron and heme metabolism in
global distribution of clinical episodes of Plasmodium falciparum. Curr Top Microbiol
Plasmodium falciparum malaria. Nature Immunol 2005;295:293–324.
2005;434:214–217. 22 van Tits LJ, Jacobs EM, Swinkels DW, et
9 Stoltzfus RJ: Iron deficiency: global preva- al: Non-transferrin-bound iron is associ-
lence and consequences. Food Nutr Bull ated with plasma level of soluble intercel-
2003;24:S99–S103. lular adhesion molecule-1 but not with in
10 Zimmermann MB, Hurrell RF: Nutritional vivo low-density lipoprotein oxidation.
iron deficiency. Lancet 2007;370:511–520. Atherosclerosis 2007;194:272–278.
11 Stoltzfus R: Defining iron-deficiency anemia 23 Hutchinson C, Al-Ashgar W, Liu DY, et
in public health terms: a time for reflection. J al: Oral ferrous sulphate leads to a marked
Nutr 2001;131:565S–567S. increase in pro-oxidant nontransferrin-bound
12 Murray MJ, Murray AB, Murray MB, Murray iron. Eur J Clin Invest 2004;34:782–784.
CJ: The adverse effect of iron repletion on 24 Dresow B, Petersen D, Fischer R, Nielsen
the course of certain infections. BMJ 1978;2: P: Non-transferrin-bound iron in plasma
1113–1115. following administration of oral iron drugs.
Biometals 2008;21:273–276.
126 Brittenham
25 Egli I, Brittenham GM, Forman JT, et al: 28 Zimmermann MB, Chassard C, Rohner F, et
The influence of supplementation and al: The effects of iron fortification on the gut
fortification iron doses on the formation of microbiota in African children: a random-
non-transferrin-bound iron (NTBI), unpub- ized controlled trial in Cote d’Ivoire. Am J
lished, 2011. Clin Nutr 2010;92:1406–1415.
26 Nchito M, Friis H, Michaelsen KF, et al: Iron 29 Nairz M, Schroll A, Sonnweber T, Weiss G:
supplementation increases small intestine The struggle for iron – a metal at the host-
permeability in primary schoolchildren in pathogen interface. Cell Microbiol 2010;
Lusaka, Zambia. Trans R Soc Trop Med Hyg 12:1691–1702.
2006;100:791–794. 30 Hawkes M, Li X, Crockett M, et al: Malaria
27 Raffatellu M, Baumler AJ: Salmonella’s iron exacerbates experimental mycobacterial
armor for battling the host and its micro- infection in vitro and in vivo. Microbes
biota. Gut Microbes 2010;1:70–72. Infect 2010;12:864–874.

Discussion on Iron
Discussion of Session V was focused on two important aspects of iron nutrition

that have a significant impact on current efforts to reduce the worldwide prev-
alence of iron deficiency. The first presentation by Dr. Hurrell dealt with the
effects of inflammation on iron absorption, and the second by Dr. Brittenham
with safety issues and the putative risks of supplementation in regions where
malaria is endemic.
The currently available strategies for alleviating nutritional iron deficiency
are dietary diversification, supplementation, food fortification and biofortifica-
tion. Fortification of staple foods and/or condiments is currently regarded as the
best long-term approach to reaching all segments of the population. However,
there are several obstacles to ensuring adequate absorption. The choice of
iron compound, the food matrix and interactions with other nutrients are all
important, but physiological control is the major determinant of absorption [1].
Hepcidin is the systemic regulator of iron absorption. Plasma levels rise in con-
cert with increasing iron stores. Hepcidin binds to ferroportin, the only known
cellular iron exporter on the surface membrane of intestinal mucosal cells and
macrophages, causing it to be internalized and degraded, thereby inhibiting iron
absorption and its release from stores.
Hepcidin is also a type II acute-phase reactant. Iron absorption is therefore
decreased in infectious and inflammatory disorders adding another level of com-
plexity to the correction of iron deficiency in regions where malaria and other
infectious disorders are prevalent. Dr. Hurrell presented experimental evidence
from studies employing stable isotopes demonstrating increased concentrations
of inflammatory cytokines, C-reactive protein and hepcidin, and significantly
decreased iron absorption in children with febrile malaria and women with afe-
brile malarial parasitemia. Dr. Sall asked whether minor infections such as skin
infections and rhinitis that are common in Senegal have the same effect. In reply,
Dr. Hurrell indicated that there is as yet no information on iron absorption in
the presence of other infections, but that his group’s studies suggest that there
is some degree of inhibition of absorption in regions with widespread infection.
In response to a question from Dr. Rosenberg, Dr. Hurrell said that there is,
however, no direct evidence of reduced efficacy of fortification in regions where
malaria is endemic. Salt fortified with ferric pyrophosphate was equally effica-
cious in Morocco (no malaria) and Ivory Coast (prevalent malaria). However,
the studies compared different groups of children. More research is needed to
clarify this issue. Efficacy should be compared in children drawn from the same
population who are either healthy or infected. Dr. Hurrell also mentioned that
based on their observations, the incorporation of iron into hemoglobin is not
affected by infection.
The optimal approach to nutritional iron deficiency in the presence of infec-
tion was discussed briefly. Clinicians have traditionally withheld iron supple-
ments until the infectious process has been controlled because the iron is
unlikely to be absorbed and could be harmful. It may increase the virulence of
pathogens. Iron requirements are reduced during episodes of infection because
erythropoiesis is suppressed. Iron is redistributed from the functional com-
partment to stores. This iron is readily mobilized once normal erythropoiesis
is restored. The effect of recurrent periods of infection on the iron status of
iron-sufficient individuals may consequently be relatively limited. The impact is
likely to be greater in those who are iron deficient because their ability to restore
optimal iron balance is limited by infection. Individualized supplementation in
the intervals between episodes of infection is unlikely to be feasible. Moreover,
the intake of staple foods or condiments fortified with iron cannot easily be
adjusted for recurrent infections. The best approach may be the use of highly
bioavailable iron compounds that allow rapid upregulation of iron absorption
during the intervals between episodes of infection.
Dr. Hurrell’s presentation describing the putative relationship between obe-
sity and iron absorption generated a lot of discussion. Dr. Bhatia pointed out
that the correlation between body mass index and iron absorption in the Thai
women reported by Zimmermann et al. [2] is not very strong. This was, how-
ever, a retrospective analysis using data from an earlier study. The women were
only mildly overweight with an average body mass index of 27.
The postulated etiology of reduced iron absorption in obese individuals is
increased hepcidin production caused by proinflammatory cytokines secreted
by visceral fat. While accepting the evidence of poor iron status in obese chil-
dren, Dr. Haschke questioned the pathogenetic relationship between obesity
and impaired iron absorption at this age. He stated that the distribution of fat
in children is different from that in adults. Children have less central obesity
and visceral fat which is the site of cytokine production. However, the evidence
presented by Dr. Hurrell strongly supports an inflammatory process as the link
between overweight and reduced iron absorption. Iron intake did not seem to
be an important factor. It was the same in overweight and normal-weight chil-
dren in their Swiss study. Dr. Zimmermann pointed out that children as young
as 6 years of age have increased IL-6 and C-reactive protein associated with
obesity. This relationship has been observed in healthy children in Switzerland,
Discussion on Iron 129

Italy and the United States. He went on to mention a soon to be published
prospective study in South Africa which includes obese, normal-weight and
stunted children. The observed odds ratio of remaining iron deficient after a
period of supplementation among obese children is twice that of the other two
groups.
Childhood obesity and its potential long-term consequences are the sub-
ject of intensive research at the present time. To some extent, the disagreement
between the discussants may have been their failure to specify the age groups
of the children that they were commenting on. There is a considerable body of
evidence supporting an association between obesity and low-grade inflamma-
tion in children over the age of 3–6 years, the age group that Dr. Zimmermann
was referring to. The situation may be more complex in younger children [3].
Finally, Dr. Zhou asked about the practical implications of the reduced iron
absorption associated with obesity. In Dr. Hurrell’s opinion, it is too early to
make specific recommendations. More research is needed to define the magni-
tude of the effect as well as the long-term consequences.
One aspect of the relationship between inflammatory and infectious disor-
ders and iron absorption was not discussed. Some conditions affect the duode-
nal mucosa directly. Chronic Helicobacter infections may lead to achlorhydria
and reduce iron absorption in older men and women [4]. Gluten enteropathy
(celiac disease) may cause significant impairment of iron absorption. It is preva-
lent in several western societies. Finally, the possible role of tropical enteropathy
and recurrent diarrhea deserves further study.
In reply to a question from Dr. Bhatia, Dr. Hurrell stated that it is difficult to
document the impact of mass fortification in western countries. The practice
was introduced without any prior efficacy or effectiveness studies. The evidence
for an impact of fortification of infant foods on the prevalence of early child-
hood anemia is more convincing. Mass fortification is nevertheless generally
considered to be a contributor to the very low prevalence of iron deficiency in
the United States. At least 20% of the iron consumed is derived from fortifica-
tion. On the other hand, the impact of current policies addressing mass fortifica-
tion in the developing world may be considerably smaller. Dr. Hurrell described
briefly an analysis carried out a few years ago to evaluate the potential impact
of wheat flour fortification in 78 countries that have established national pro-
grams to fortify wheat flour with iron as well as other micronutrients [5]. The
conclusion drawn from the analysis was that an impact on iron nutrition could
be expected in only 6 countries. The primary reason for the anticipated lack
of effectiveness is the widespread use of iron compounds that are not bioavail-
able, particularly hydrogen-reduced iron powders. They are favored by millers
because of their low cost and lack of chemical reactivity. The workshop partici-
pants recommended that efforts be made to convince millers to use electrolytic
iron or other more absorbable iron compounds instead of hydrogen-reduced
iron powders.
130 Lynch
There were two questions related to the risk of fortification. Dr. Ganguly
asked about the effect on potentially pathogenic bacteria in the gastrointesti-
nal tract. Dr. Hurrell replied that in their study in the Ivory Coast there was
a fivefold increase in pathogenic enterobacteria and a decrease in lactobacilli
among the children receiving biscuits fortified with electrolytic iron [6]. They
also reported evidence of gut inflammation. However, more research is needed
to establish the practical implications of this finding. Dr. Bhatia expressed his
concern about iron overload in individuals who are iron sufficient and consum-
ing a fortified diet. Dr. Hurrell replied that the risk for the general population is
low. Individuals who are single trait carriers of the common form of hemochro-
matosis (HFE hemochromatosis) or the thalassemia syndromes do not accumu-
late excess iron. The problem is limited to the minority of individuals who are
hemochromatosis homozygotes and those who have clinically evident thalas-
semia. The risk is further reduced in HFE hemochromatosis because of the low
penetrance of the disorder.
Dr. Brittenham introduced his presentation by describing the dilemma faced
by nutritionists because of the observations reported from the iron, folic acid and
zinc supplementation trial carried out in Pemba, Tanzania [7]. The results indi-
cate that the risk for severe morbidity and mortality from malaria is increased
in young children who are iron deficient. On the other hand, universal iron and
folic acid supplementation raised the risk for severe morbidity and mortality
in the population of children under the age of 3 years as a whole. The results of
this trial were discussed in some depth. It was pointed out that earlier smaller
studies raised concerns about the risk of iron supplementation in regions where
malaria is endemic. However this very large, well-controlled trial is the first to
provide a more definitive answer. It would be valuable to know whether we are
dealing with a relatively minor although important factor that affects the whole
population or a more powerful contributor in an as yet unidentified subgroup.
Dr. Black stated that the data from the Pemba study do not provide the informa-
tion that would be needed to address this issue. A redesigned trial with the sta-
tistical power to answer the question does not seem feasible at the present time.
Dr. Brittenham said that alternative approaches that are focused on developing
a better understanding of the underlying pathophysiological mechanisms are
more likely to be fruitful. Such studies should include pregnant women.
Dr. Rosenberg suggested that folic acid supplementation could have played
a role since it was always given with the iron, and the antimalarial drug being
used at the time is an antifolate agent. Dr. Black discounted this possibility
because responses in community and hospital-treated children did not differ.
Furthermore, earlier studies tend to incriminate iron.
The putative risk of iron supplementation in the setting of other infectious
diseases was discussed briefly. Dr. Brittenham stated that there is evidence
implicating iron in HIV disease and tuberculosis, and that the observations
made by Zimmermann et al. [6] suggest that it may well play a role in diarrheal

diseases. They reported that iron fortification favors the growth of potentially
pathogenic gut bacteria associated with inflammation of the gastrointestinal
tract in African children.
The postulated mechanisms for the adverse effects of iron supplements
were discussed at some length. A pivotal role for non-transferrin-bound iron
(NTBI) is favored at the present time. It could promote microbial virulence.
Dr. Brittenham described recent unpublished observations that he and his col-
leagues have made that support this contention. When blood that has been
stored for some time is transfused, the red cells are rapidly sequestered and pro-
cessed in the spleen resulting in the release of NTBI. They transfused stored
blood into healthy human volunteers. Blood samples from these volunteers con-
tained NTBI. They allowed bacteria that cause infections in human beings to
grow more readily, suggesting that NTBI may well increase the risk of severe
morbidity and mortality from bacteremia.
The mechanisms that lead to increased morbidity in malaria may be differ-
ent. Dr. Brittenham listed several possibilities in his presentation. One involves
increased expression of endothelial adhesion molecules, which might lead to the
sequestration of Plasmodium falciparum-infected erythrocytes in the microvas-
culature causing obstruction and severe clinical manifestations such as cerebral
malaria. Dr. Brittenham pointed out that if this postulate proves to be correct,
supplemental iron may not be a risk factor in vivax malaria since vascular adhe-
sion does not occur. It might also provide an explanation for potentially disas-
trous consequences of P. falciparum malaria in pregnancy.
The concentration of NTBI is proportional to the serum iron level. The
serum iron rises rapidly after a dose of supplemental iron. The height of the
rise is determined by dose, the presence of absence of food and the iron status
of the individual. The highest levels occur when supplemental iron is consumed
in the fasting state. When taken with food, the increment is smaller. There is no
measurable increase with fortification iron. It seems paradoxical that the high-
est peak serum iron levels are encountered in iron-deficient individuals who
seem to benefit from iron supplements. One would anticipate the generation of
more NTBI and increased risk in them. However, Dr. Solomons said that they
have unpublished observations demonstrating that the areas under the curves
for NTBI concentrations after a dose of supplemental iron are similar in iron-
sufficient and iron-deficient volunteers. In addition, NTBI clearance may be
accelerated in iron deficiency because of the increased demand for iron.
Several questions related to the best pragmatic approach to iron deficiency
in individuals suffering from malaria or other infections. Dr. Brittenham rec-
ommended that the infection be treated before giving iron since there is usu-
ally no immediate urgency to correct iron deficiency. Dr. Solomons pointed
out that although the correction of iron deficiency is not regarded as urgent in
the clinical sense, periods of iron deficiency in early childhood may have long-
term consequences for physical, cognitive and emotional development. Iron
132 Lynch
deficiency should therefore be corrected as soon as possible. Dr. Brittenham
concluded by saying that there are no simple, universally applicable answers to
these questions. The guiding principle should be avoidance of the production
of NTBI by using lower doses of iron over longer time periods. He also empha-
sized the importance of prevention in infancy and childhood. Enhancement of
the mother’s iron status during pregnancy and the correction of iron deficiency
in women before pregnancy are essential.
Sean Lynch
References
1 Bothwell TH, Charlton RW, Cook J, Finch C: 5 Hurrell R, Ranum P, de Pee S, et al: Revised
Iron Metabolism in Man. Oxford, Blackwell recommendations for iron fortification
Scientific Publications, 1979. of wheat flour and an evaluation of the
2 Zimmermann MB, Zeder C, Muthayya S, et expected impact of current national wheat
al: Adiposity in women and children from flour fortification programs. Food Nutr Bull
transition countries predicts decreased iron 2010;31:S7–S21.
absorption, iron deficiency and a reduced 6 Zimmermann MB, Chassard C, Rohner F, et
response to iron fortification. Int J Obes al: The effects of iron fortification on the gut
(Lond) 2008;32:1098–1104. microbiota in African children: a random-
3 Tam CS, Clement K, Baur LA, Tordjman J: ized controlled trial in Cote d’Ivoire. Am J
Obesity and low-grade inflammation: a paedi- Clin Nutr 2010;92:1406–1415.
atric perspective. Obes Rev 2010;11:118–126. 7 Sazawal S, Black RE, Ramsan M, et al: Effects
4 Hershko C, Skikne B: Pathogenesis and man- of routine prophylactic supplementation with
agement of iron deficiency anemia: emerging iron and folic acid on admission to hospital
role of celiac disease, Helicobacter pylori, and and mortality in preschool children in a high
autoimmune gastritis. Semin Hematol 2009; malaria transmission setting: community-
46:339–350. based, randomised, placebo-controlled trial.
Lancet 2006;367:133–143.

Summary on Vitamin A and Iron
I think it was really an excellent discussion both in the vitamin A section and in
the iron section. I have made three points here with the vitamin A, and I really
didn’t know so much about the vitamin A when I came to this meeting. It seems
to me that vitamin A supplementation historically has been targeted to reduce
morbidity and mortality. As Noel Solomons said, the paradigm of death reduc-
tion is not to reduce vitamin A deficiency, so really maybe we should now revise
these recommendations based on more evidence of low vitamin A intakes or
status. There seems to be good evidence at least from Asia that this universal
vitamin A supplementation decreases mortality in children less than 6 months
of age even when this new India study was included in the meta-analysis. There
is less good information from Africa and from Mexico, where there is less
vitamin A deficiency, and larger doses of vitamin A to children with no vitamin
A deficiency seem to increase infections or respiratory infections and HIV and
maybe also antagonize vitamin D metabolism. This was my first point.
The second point, large doses of vitamin A may specifically improve only
measles outcome, so the effectiveness in measles-vaccinated children is less
clear, and the effectiveness in other disease such as diarrhea may be a little bit
clearer. Concerning respiratory infections in the children less than 6 months
of age, there seems to be no effect of vitamin A supplementation, so maybe as
Lindsay Allen was saying we should target lactation. Inconsistent results with
newborns, interactions with vaccinations and perhaps universal supplementa-
tion with vitamin A should be rethought, and more evidence should be put in
the area of food fortification. This would improve the quality of the diet.
With the iron absorption, ensuring adequate absorption from iron-fortified
foods is not easy, it’s a challenge in addition to the iron compound inhibitors, etc.
We need now to reflect on inflammation, hepcidin and decreased absorption.
It’s still not clear whether efficacy is affected or whether erythropoiesis over-
rules inflammation. The decrease in absorption has only been demonstrated
with malaria and overweight, and we need to look at other more common infec-
tions. It’s still not clear what iron requirements are in a population with wide-
spread infections or inflammatory disorders including overweight. We need to
think about that as well as fortification levels.
The last points will be on the iron supplementation. It’s accepted that with inad-
equate health care and not good malaria surveillance, iron supplementation to
children in malaria-endemic areas can increase morbidity and mortality. The
negative impacts on other infections, as Gary Brittenham said, are expected but
not yet demonstrated. The most plausible mechanism is still the non-transferrin-
bound iron (NTBI); at the beginning we didn’t really believe it existed, but it
does exist, and Noel Solomons has also reported some studies with NTBI; it is
possibly coupled with the stimulation of the growth of the pathogenic organ-
isms by the iron which goes through to the gut; the most likely explanation is
the sequestration of the infected erythrocytes in the brain or the villi, and the
villi could lead to the breach of the intestinal barrier and bacteremia. I think the
strategy is by understanding the mechanism of this negative effect with supple-
mentation, we should be able to confirm the safety of food fortification and
hopefully also the Sprinkles.
Richard F. Hurrell
Summary on Vitamin A and Iron 135

Iodine
Are Weaning Infants at Risk of Iodine

Deficiency Even in Countries with
Established Iodized Salt Programs?
Michael B. Zimmermann
Human Nutrition Laboratory, Institute of Food, Nutrition and Health, Swiss Federal Institute of
Technology (ETH) Zürich, and the International Council for the Control of Iodine Deficiency
Disorders (ICCIDD), Zurich, Switzerland
Abstract
Because iodine deficiency (ID) during infancy can irreversibly impair neurodevelopment
and increase mortality, it is critical that dietary iodine is adequate in this vulnerable group.
Lactating mothers consuming iodized salt can transfer adequate iodine to the infant via
breast milk, but during the weaning period, infants are at risk for ID for several reasons: (1)
requirements per kg bodyweight for iodine and thyroid hormone during infancy are
higher than at any other time in the life cycle; (2) experts recommend no extra salt (iodized
or not) be given to infants during the first year; (3) cow’s milk (a major source of dietary
iodine in many countries) is also not recommended for infants during the first year; and
(4) iron deficiency, a common disorder during infancy, can impair iodine metabolism and
reduce thyroid hormone production. For many weaning infants in industrialized coun-
tries, iodine fortified into commercial infant foods becomes important. This has recently
been demonstrated in Switzerland, where a long-standing iodized salt program provides
adequate iodine to pregnant women and school-age children, but new national data sug-
gest weaning infants not receiving iodine-containing commercial baby foods have inad-
equate iodine intakes. Thus, even in countries with effective iodized salt programs, infants
may be at risk of ID during weaning and may need additional dietary and/or supplemen-
tal sources of iodine. Copyright © 2012 S. Karger AG, Basel
Iodine deficiency (ID) during early life may cause irreversible damage to the
developing brain [1]. In regions of moderate-to-severe ID, many infants and
pregnant women have low circulating levels of thyroid hormone [2]. Thyroid
hormone is required for normal neuronal migration and myelination of the
brain during fetal and early postnatal life. Hypothyroidism during these
critical periods can cause mental retardation and neurological abnormalities [3].
Prevention of ID during infancy not only improves neurodevelopment, it may
also reduce infant mortality. In a randomized, placebo-controlled trial of oral
iodized oil (100 mg iodine) in Indonesian infants (n = 617) treated at ≈6 weeks
of age, there was a 72% decrease in risk of infant death [4]. Similarly, in a large
cross-sectional study in Indonesia, use of adequately iodized salt was associated
with a lower infant mortality rate [5]. The potential adverse effects of mild-to-
moderate ID during infancy remain unclear, as no well-controlled studies have
tested the effects of iodine repletion at this age. However, infant requirements
per kg bodyweight for iodine and thyroid hormone are much higher than later
in life. Even in areas of iodine sufficiency, a newborn’s thyroidal iodine reserve is
small, only ≈300 μg [6], and thyroxine turnover is high, with estimated produc-
tion rates of 5–6 μg/kg bodyweight per day in infancy [7]. Thus, the infant needs
a regular supply of dietary iodine to maintain euthyroidism.
National programs to control ID should therefore emphasize this vulnerable
period. In most countries, the most effective strategy against ID is salt iodiza-
tion because salt is one of few foodstuffs regularly consumed by most of the
population through the year [8]. Since 1990, a major global effort has increased
the number of households using iodized salt from <20% to >70%, dramatically
reducing ID [9]. However, the International Council for the Control of Iodine
Deficiency Disorders (ICCIDD) estimates 1.88 billion individuals globally still
have an insufficient iodine intake, including nearly 1/3 of all school-age chil-
dren, and ID remains a public health problem in 32 countries [10]. ID is not only
a problem in developing regions; it also affects many industrialized countries.
About 50% of continental Europe remains mildly iodine deficient, and iodine
intakes in other industrialized countries, including the US, the United Kingdom
and Australia, have fallen in recent years. ID has reappeared in Australia and
the United Kingdom, mainly due to declining iodine residues in milk products
because of decreased iodophor use by the dairy industry. In the US, the median
UI is 160 μg/l (95% CI: 146–172), still adequate but half the median value of
321 μg/l found in the 1970s [11], and pregnant women who consume few dairy
products may be at risk of ID [12]. In most industrialized countries, because
70–90% of salt consumption is from purchased processed foods, if only house-
hold salt is iodized, it will not supply adequate iodine. In order to successfully
control ID, it is critical that the food industry use iodized salt whenever possible
[13]. If both household and industry salt are iodized, salt iodization can deliver
adequate iodine to most age groups, including children and adults. Whether it
can also deliver adequate iodine during infancy is, however, uncertain [14].
A central problem to monitoring iodine status in infancy is the iodine require-
ment for this age group has not been well defined. The US Institute of Medicine
[15] estimated an adequate intake (AI) for infants because there was not suf-
ficient scientific evidence to calculate an Estimated Average Requirement. The
AI, 110 μg and 130 μg/day for ages 0–6 and 6–12 months, respectively, is derived
138 Zimmermann
from the mean iodine intake of healthy infants fed human milk, based on a
median breast milk iodine content (BMIC) of 146 μg/l in US women in the early
1980s [15]. But because iodine intakes in the US population were excessive at
that time, the BMIC used to set the AI was at the upper end of the range of 78–
167 μg/l reported for iodine-sufficient countries [16]. Although high maternal
iodine intakes can result in high BMIC, iodine intakes by the infant greater than
requirements will simply be excreted in the urine. Thus, iodine requirements
during lactation should be based on infant balance studies, and a single small
balance study in Belgian infants found iodine retention was 7.3 μg/kg per day
[17]. If the reference bodyweight at 6 months of age is 7 kg [15], an infant at this
age would be in positive balance at a daily intake of ≈50 μg. The WHO has set
a Recommended Nutrient Intake (RNI) of 90 μg/day for iodine during infancy
[8]. But clearly more data are needed, preferably from a large, carefully con-
trolled balance study, to better define the iodine requirement during infancy.
Lactating mothers consuming iodized salt transfer iodine to the infant via
breast milk. During lactation, the mammary gland actively concentrates iodine
via the Na/I symporter [18] and secretes it into breast milk. Because the gland
is able to concentrate iodine, iodine supply to the newborn via breast milk
may be at least partially maintained even during maternal ID [19]. However,
a recent New Zealand study in iodine-deficient mothers reported a significant
fall in BMIC during the first 6 months of lactation [20]. In contrast, in iodine-
sufficient lactating women in the US with a median urinary iodine concentra-
tion (UIC) of 114 μg/l, the median BMIC was 155 μg/l [21]. A review of BMIC
in iodine-sufficient countries found a wide range of mean or median concentra-
tions, from 50 μg/l in Finland to 270 μg/l in the US, but sample sizes were small
and not representative [16].
Although breast milk can supply adequate iodine to infants, as they are
weaned from breast milk, usually in the second half of the first year, their dietary
iodine intakes may fall. With the exception of some sea foods, the native iodine
content of most foods is low [22, 23]. Residues from iodophors used during
dairying and transport of milk can increase the iodine content of dairy products
[13]. But iodized salt, either used in the household or added to processed foods,
is the primary source of iodine in the diets of many countries. For example, the
two main sources of dietary iodine in the US and Switzerland are bread con-
taining iodized salt and dairy products [23, 24]. However, iodized salt may not
contribute significantly to infant iodine intakes because pediatricians and nutri-
tionists recommend no extra salt (iodized or not) be given to infants during
the first year. After breastfeeding for 6 months, mothers are encouraged to feed
their infants home-prepared complementary foods (CF) without added salt [25,
26]. Moreover, cow’s milk (a major adventitious source of dietary iodine in older
children) is also not recommended for infants during the first year [26].
Thus, for many weaning infants in industrialized countries, iodine added
to commercial infant foods becomes an important iodine source. In the New
Weaning and Risk of Iodine Deficiency in Industrialized Countries 139

Zealand Total Diet study, which simulated typical diets, iodine-containing
infant formula/foods provided 60% of iodine intakes for infants older than 6
months [27]. In the US Total Diet Study, 90% of iodine intake in infants older
than 6 months was provided by infant formula/foods and dairy products [28].
In national programs to control ID, regular monitoring of iodine status in tar-
get populations is important to detect both low and excessive intakes of iodine.
For monitoring, WHO recommends using the median UIC from a representative
sample of spot urine collections to classify a population’s iodine status [8]. Because
>90% of dietary iodine eventually appears in the urine, UIC is an excellent indi-
cator of recent iodine intake. In monitoring programs, it has been traditionally
assumed that if the general population is iodine sufficient, infants will be also
iodine sufficient. But this assumption has not been rigorously tested as there have
been no national studies assessing infant iodine status in countries with estab-
lished iodized salt programs where the general population has adequate iodine
intakes. One reason for the lack of data in infants is the difficulty of obtaining spot
urine samples to measure UIC. However, a simple noninvasive method for col-
lection of spot urine samples from infants has recently been developed and vali-
dated [29]. An absorbent pad (e.g. a feminine hygiene pad that is free of iodine)
is inserted inside the diaper and the infant is breastfed. Several milliliters of urine
absorbed by the pad can be aspirated into a syringe for measurement. WHO rec-
ommendations state a median (m)UIC ≥100 μg/l in a representative population of
infants indicates they have adequate iodine nutrition [8].
Switzerland has a model iodized salt program that was initiated in 1922; in
national surveys in 1999 and 2004, >90% of households were using iodized salt,
and school children were iodine sufficient [30, 31]. The objectives of a recent
Swiss study [32] were to first measure UIC in a national sample of pregnant
women and school children to confirm that the Swiss population remains
iodine sufficient, and then to collect UIC data from a nationally representative
sample of infants. The mUIC (95% CI) in school children (n = 916) and preg-
nant women (n = 648) was 120 (120–128) and 162 (144–177) μg/l, respectively,
indicating iodine sufficiency in both groups. Spot urine samples were collected
from infants at 3–4 days after delivery, at 6 months ± 6 weeks or at 12 months
± 6 weeks. Inclusion criteria for the mother-infant pairs were: (1) full-term,
healthy pregnancy; (2) parental residence in Switzerland for ≥12 months before
delivery and since delivery; (3) no history of thyroid disorders in the mother; (4)
no ingestion of iodine-containing drugs or contrast media during gestation; (5)
delivery without use of iodine-containing disinfectants, and (6) no health prob-
lems in the infant. Breast milk iodine concentrations (BMIC) were measured
in the mothers of the infants at 6 and 12 months. The iodine concentration of
commercial infant foods was directly analyzed, including infant formulas, fol-
low-on formulas and commonly-consumed baby cereal products. The relative
contributions of BMIC, infant formula milk (IFM) and CF to iodine intakes in
the infants were estimated.
140 Zimmermann
Table 1. UICs in infancy in Switzerland (2005–2009)
Age of the infants
3–4 days 6 months 12 months
Total 368 279 228

Male/female 171/197 142/136 106/122
Age 6.2±0.4 12.3±0.4
UIC, μg/l
Median (95% CI) 91 (82–99) 91 (79–103) 103 (92–116)
Range 4–922 11–759 6–951
<20% 2 2 3
20–49% 20 22 17
50–99% 36 32 28
<100% 58 55 48
>300% 5 6 4
Data from Andersson et al. [32].
Twenty-four participating clinics provided samples from exclusively breast-

fed infants on days 3 or 4 after birth (n = 368: day 3, n = 248, day 4, n = 120).
Overall, mUIC was 91 μg/l; at day 3, mUIC was 87 μg/l and at day 4, it was 100
μg/l (table 1). Among the mothers, 65% (n = 241) were taking supplements,
but only 0.8% (n = 3) were consuming iodine-containing supplements (dur-
ing pregnancy or currently), and 12% (n = 42) were using non-iodized salt.
For the older infants, eighteen clinics provided 507 infant/mother pairs.
The mUICs (95% CI) in the 6- and 12-month-old infants were 91 (79–103)
and 103 (92–116) μg/l, respectively, and were not significantly different (table
1). The overall infant mUIC was 98 μg/l (95% CI: 89–105; table 1). Girls had
higher UICs than boys (mUIC, 103 vs. 88 μg/l; p < 0.05). Among the mothers,
mUIC was 75 μg/l (95% CI: 69–81) and mBMIC was 49 μg/kg. Fifty-seven
percent of the 6-month-old infants and 18% of the 12-month-old infants
were being breastfed fully or partly at the time of sampling. Breastfed infants
with or without IFM had a lower mUIC than infants not currently breastfed
(82 μg/l, n = 196, vs. 105 μg/l, n = 311; p < 0.001). About 60% of all infants
were receiving IFM, and their mUIC was higher than in those not receiving
IFM (109 μg/l, n = 304, vs. 73 μg/l, n = 203; p < 0.001). Infants (breastfed and/
or CF) receiving IFM had higher mUIC than breastfed weaning infants who
did not receive IFM (109 μg/l, n = 304 vs. 70 μg/l, n = 131; p < 0.01; fig. 1).
Weaned infants not receiving breast milk or IFM did not differ in UIC (89
μg/l, n = 72) from the other two groups. Eighty four percent of mothers were
using iodized salt at home, 8% of mothers were not using iodized salt and
8% were unsure; there were no significant differences in UIC of the mothers

*
120
100
Urinary iodine (μg/l)
80
60
40
20
0
Group 1 Group 2 Group 3
Fig. 1. Differences in median UIC (error bars show the 95% CI) by mode of feeding in
6- and 12-month-old Swiss infants. Group 1 (n = 131): infants receiving breast milk,
partly CF, but no IFM; group 2 (n = 304): infants receiving infant formula, partly com-
bined with breast milk and/or CF; group 3 (n = 72): infants receiving no breast milk or
infant formula, but CF. * p < 0.01, significant differences between groups 1 and 2 (Mann-
Whitney U test with Bonferroni correction). The dashed horizontal line indicates the
WHO cutoff value for the median urinary iodine indicating adequate iodine intakes in
infancy [8].
or infants among these 3 groups. Among the 6-month-old infants, nearly 4

out of 5 were already receiving CF, and at 6 and 12 months, 7 and 95% of
infants were receiving some foods from the family table. The mUIC of infants
receiving some iodized salt in CF (103 μg/l, n = 287) was higher, but not
significantly different from the mUIC of infants not receiving iodized salt (89
μg/l, n = 189). In a multivariate regression of predictors of UIC in the 6- and
12-month-old infants, BMIC (β = 0.320, p < 0.0001) and current consump-
tion of IFM (yes/no; β = 0.201, p = 0.010) were significant, while gender, age
and maternal UIC were not.
Thirty-two percent of the mothers were taking nutritional supplements (n =
158), but only 3% of women were consuming iodine-containing supplements.
The agreement between the labeled and analyzed iodine content of the IFMs
and infant cereals was high: the mean (± SD) difference (%) between labeled and
measured values was 13.5 ± 9.1% for the formulas and 4.5 ± 2.6% for the cereals.
None of the formula milks or cereals exceeded the recommended maxima of 50
and 35 μg/100 kcal [18].
This study is important in that it is the first national representative study
assessing infant iodine status in a country with an established long-standing
142 Zimmermann
iodized salt program where school age children and pregnant women have ade-
quate iodine intakes [32]. The mUIC in the infants in this study at 3–4 days
and at 6 months was just below the 100 μg/l cutoff that indicates iodine suffi-
ciency in infancy [8]. Infants who were not receiving iodine-fortified IFM dur-
ing the weaning period were clearly deficient, with an mUIC of ≈70 μg/l. Several
reports of mUIC in infants (<2-year-olds) in countries with more than adequate
or excess iodine intakes have found higher values than in this study [33–36].
But other studies in European infants have generally reported mUIC similar to
the Swiss value [33, 37–40]. In the Swiss infants, only 58% of the 6-month-olds
and 18% of the 12-month-olds were being breastfed. Infants who were breastfed
and given home-prepared CF (that contain little or no added salt) were at high-
est risk of low iodine intakes. In Germany, using a dietary model, it was esti-
mated that the iodine intake of an 8-month-old breastfed infant who receives
home-prepared CF would be only ca. 45 μg/day compared to ≥125 μg/day in a
formula-fed infant who receives commercial CF [41].
An additional factor that may aggravate low iodine intakes in weaning infants
is the high prevalence of iron deficiency at this age. Iron deficiency reduces
heme-dependent thyroperoxidase activity in the thyroid and impairs produc-
tion of thyroid hormone. In iodine-deficient children, iron deficiency anemia
blunts the efficacy of iodized oil and iodized salt [42]. In pregnant women, poor
iron status predicts higher thyroid-stimulating hormone and lower thyroxine
concentrations in an area of borderline ID [43]. Thus, iron-deficient infants
may be at higher risk of poor thyroid function during low iodine intakes. But
there are no published studies of the influence of iron status on thyroid function
in infants with ID.
The Swiss study and others raise the concern that specific public health mea-
sures, in addition to salt iodization, should be considered to improve iodine
intakes at this age. These could include increasing public awareness of the
importance of providing at least some iodine-containing infant foods/formula
during weaning. Infant food manufacturers should be strongly encouraged to
fortify their products with iodine [14]. In Europe, the required level of iodiza-
tion for IFMs is 10–50 μg/100 kcal (2.5 μg/100 kJ), but for cereal-based and
other CFs, there are no requirements for minimum iodization while the allowed
upper level is 35 μg/100 kcal [44]. In the US, iodine fortification of infant for-
mula is mandatory at a minimum level of 5 μg/100 kcal (maximum level is 75
μg/100 kcal) [45]. In Germany, it is estimated only ≈50% of CFs are fortified
with iodine [41].
Another strategy to increase iodine intakes in weaning infants would be
iodine supplementation. Supplementation of lactating women can increase
BMIC and could provide additional iodine, particularly during weaning when
infants are being partially breastfed. In Danish mothers (n = 147), median
BMIC on the 5th day postpartum was significantly higher (57 μg/l) in those
receiving supplementation with 150 μg/day of oral iodine, compared to those

not supplemented (34 μg/l) [46]. In Germany, 60 mothers who received 200 μg/
day of oral iodine had significant higher mean iodine concentrations in breast
milk (76 μg/l) than those who did not (55 μg/l) [47]. Thus, iodine supplemen-
tation of breastfeeding women can significantly improve iodine supply to the
newborn. In the Swiss study, supplements containing iodine were consumed by
<5% of lactating women. Alternatively, iodine supplements could be given to
infants directly, as is often done for iron or vitamin D. Currently, American and
European pediatric societies do not recommend iodine supplements for infants
on well-balanced diets [48]. Similarly, in countries such as Switzerland with an
effective iodized salt program, WHO does not recommend iodine supplemen-
tation for infants or lactating women [49]. These recommendations may need
to be reconsidered if the Swiss findings are confirmed in other industrialized
countries.
References
1 Zimmermann MB, Jooste PL, Pandav CS: 10 Andersson M, Karumbunathan V,
Iodine-deficiency disorders. Lancet 2008; Zimmermann MB: Global iodine status in
372:1251–62. 2011 and trends over the past decade. J Nutr
2 Morreale de Escobar G, Obregon MJ, 2012;142:744–750.
Escobar del Rey F: Role of thyroid hormone 11 Caldwell KL, Miller GA, Wang RY, et al:
during early brain development. Eur J Iodine status of the U.S. population, National
Endocrinol 2004;151(suppl 3):U25–U37. Health and Nutrition Examination Survey
3 McCarrison R: Observations on endemic 2003–2004. Thyroid 2008;18:1207–1214.
cretinism in the Chitral and Gilgit Valleys. 12 Perrine CG, Herrick K, Serdula MK, Sullivan
Lancet 1908;ii:1275–1280. KM: Some subgroups of reproductive age
4 Cobra C, Muhilal, Rusmil K, et al: Infant women in the United States may be at risk
survival is improved by oral iodine supple- for iodine deficiency. J Nutr 2010;140:1489–
mentation. J Nutr 1997;127:574–578. 1494.
5 Semba RD, de Pee S, Hess SY, et al: Child 13 Zimmermann MB: Iodine deficiency in
malnutrition and mortality among fami- industrialised countries. Proc Nutr Soc
lies not utilizing adequately iodized salt in 2010;69:133–143.
Indonesia. Am J Clin Nutr 2008;87:438–444. 14 Dunn JT: Iodine should be routinely added
6 Delange F: Screening for congenital hypothy- to complementary foods. J Nutr 2003;
roidism used as an indicator of IDD control. 133:3008S–3010S.
Thyroid 1998;8:1185–1192. 15 Institute of Medicine, Academy of Sciences:
7 Brown RS: Chapter 15b – Disorders of the Dietary Reference Intakes for Vitamin A,
Thyroid Gland in Infancy, Childhood and Vitamin K, Arsenic, Boron, Chromium,
Adolescence. http://www.thyroidmanager. Copper, Iodine, Iron, Manganese,
org; accessed January 20, 2011. Molybdenum, Nickel, Silicon, Vanadium
8 WHO/UNICEF/ICCIDD: Assessment of and Zinc. Washington, National Academy
Iodine Deficiency Disorders and Monitoring Press, 2001.
Their Elimination, ed 3. Geneva, World 16 Zimmermann MB: The impact of iodised
Health Organization, 2007. salt or iodine supplements on iodine status
9 Delange F, Bürgi H, Chen ZP, Dunn JT: during pregnancy, lactation and infancy.
World status of monitoring iodine deficiency Public Health Nutr 2007;10:1584–1595.
disorders control programs. Thyroid 2002;
12:915–924.
144 Zimmermann
17 Delange F, Bourdoux P, Vo Thi LD, et al: 29 Dorey CM, Zimmermann MB: Reference
Negative iodine balance in preterm infants. values for spot urinary iodine concentra-
Ann Endocrinol 1984;45:77. tions in iodine-sufficient newborns using
18 Eskandari S, Loo DD, Dai G, et al: a new pad collection method. Thyroid
Thyroid Na+/I- symporter. Mechanism, 2008;18:347–352.
stoichiometry, and specificity. J Biol Chem 30 Hess SY, Zimmermann MB, Torresani T, et
1997;272:27230–27238. al: Monitoring the adequacy of salt iodiza-
19 Vermiglio F, Lo Presti VP, et al: Enhanced tion in Switzerland: a national study of
iodine concentration capacity by the mam- school children and pregnant women. Eur J
mary gland in iodine deficient lactating Clin Nutr 2001;55:162–166.
women of an endemic goiter region in Sicily. 31 Zimmermann MB, Aeberli I, Torresani T,
J Endocrinol Invest 1992;15:137–142. Burgi H: Increasing the iodine concentration
20 Mulrine HM, Skeaff SA, Ferguson EL, et al: in the Swiss iodized salt program markedly
Breast-milk iodine concentration declines improved iodine status in pregnant women
over the first 6 mo postpartum in iodine- and children: a 5-y prospective national
deficient women. Am J Clin Nutr 2010; study. Am J Clin Nutr 2005;82:388–392.
92:849–856. 32 Andersson M, Aeberli I, Wüst N, et al: The
21 Pearce EN, Leung AM, Blount BC, et al: Swiss iodized salt program provides adequate
Breast milk iodine and perchlorate concen- iodine for school children and pregnant
trations in lactating Boston-area women. J women, but weaning infants not receiving
Clin Endocrinol Metab 2007;92:1673–1677. iodine-containing complementary foods as
22 Pennington JAT, Schoen SA, Salmon GD, well as their mothers are iodine deficient. J
et al: Composition of core foods in the U.S. Clin Endocrinol Metab 2010;95:5217–5224.
food supply, 1982–1991. J Food Comp Anal 33 Delange F, Heidemann P, Bourdoux P, et al:
1995;8:171–217. Regional variations of iodine nutrition and
23 Haldimann M, Alt A, Blanc A, Blondeau K: thyroid function during the neonatal period
Iodine content of food groups. J Food Comp in Europe. Biol Neonate 1986;49:322–330.
Anal 2005;18:461–471. 34 Pearce EN, Leung AM, Blount BC, et al:
24 Pearce EN, Pino S, He X, et al: Sources Breast milk iodine and perchlorate concen-
of dietary iodine: bread, cows’ milk, and trations in lactating Boston-area women. J
infant formula in the Boston area. J Clin Clin Endocrinol Metab 2007;92:1673–1677.
Endocrinol Metab 2004;89:3421–3424. 35 Rossi L, Branca F: Salt iodisation and public
25 Heird WC, Ziegler P, Reidy K, Briefel R: health campaigns to eradicate iodine defi-
Current electrolyte intakes of infants and ciency disorders in Armenia. Public Health
toddlers. J Am Diet Assoc 2006;106(suppl Nutr 2003;6:463–469.
1):S43–S51. 36 Wang Y, Zhang Z, Ge P, Wang S: Iodine
26 Agostoni C, Decsi T, Fewtrell M, et al: status and thyroid function of pregnant,
Complementary feeding: a commentary by lactating women and infants (0–1 yr) resid-
the ESPGHAN Committee on Nutrition. J ing in areas with an effective Universal Salt
Pediatr Gastroenterol Nutr 2008;46:99–110. Iodization program (Abstract). Asia Pac J
27 Thomson BM, Vannoort RW, Haslemore Clin Nutr 2009;18:34–40.
RM: Dietary exposure and trends of expo- 37 Hoang Truong T, Gerber H, Haenel AF,
sure to nutrient elements iodine, iron, Burgi H: Iodine supply at various periods in
selenium and sodium from the 2003–4 New life and ultrasonographic thyroid volume in
Zealand Total Diet Survey. Br J Nutr 2008; school children in a region of Switzerland (in
99:614–625. German). Schweiz Med Wochenschr 1997;
28 Murray CW, Egan SK, Kim H, et al: US Food 127:715–721.
and Drug Administration’s Total Diet Study: 38 Heidemann PH, Stubbe P, Thal H, et al:
dietary intake of perchlorate and iodine. J Influence of iodine prophylaxis on thyroid
Expo Sci Environ Epidemiol 2008;18:571–580. function and iodine excretion in newborn
infants and their mothers: comparison
between Sweden and Germany. Acta
Endocrinol 1986;3:47–48.

39 Dlouhy P, Rambouskova J, Wiererova O, et 45 US Food and Drug Administration: The
al: Iodine saturation of Roma neonates in Federal Food, Drug, and Cosmetic Act, 21
Prague is not at an optimum level. Ann Nutr USC §350a, Sec 412(i), 1980. Requirements
Metab 2006;50:242–246. for Infant Formula. Code of Federal
40 Thamm M, Ellert U, Thierfelder W, et Regulations.
al: Iodine intake in Germany. Results of 46 Nohr SB, Laurberg P, Bortum KG, et al:
iodine monitoring in the German Health Iodine status in neonates in Denmark:
Interview and Examination Survey for regional variations and dependency on
Children and Adolescents (KiGGS) maternal iodine supplementation. Acta
(in German). Bundesgesundheitsblatt Paediatr 1994;83:578–582.
Gesundheitsforschung Gesundheitsschutz 47 Seibold-Weiger K, Wollmann H, Rendl J, et
2007;50:744–749. al: Iodine concentration in the breast milk of
41 Alexy U, Drossard C, Kersting M, Remer mothers of premature infants. Z Geburtshilfe
T: Iodine intake in the youngest: impact of Neonatol 1999;203:81–85.
commercial complementary food. Eur J Clin 48 Ernährungskommission der Schweizerischen
Nutr 2009;63:1368–1370. Gesellschaft für Pädiatrie. Empfehlungen für
42 Zimmermann MB: The influence of iron die Säuglingsernährung 2008 (in German).
status on iodine utilization and thyroid func- Paediatrica 2008;19:19–21.
tion. Ann Rev Nutr 2006;26:367–389. 49 WHO Secretariat on behalf of the partici-
43 Zimmermann MB, Bürgi H, Hurrell RF: Iron pants to the Consultation: Prevention and
deficiency predicts poor maternal thyroid control of iodine deficiency in pregnant and
status during pregnancy. J Clin Endocrinol lactating women and in children less than
Metab 2007;92:3436–3440. 2-years-old: conclusions and recommenda-
44 Commission of the European Communities: tions of the technical consultation. Public
Commission Directive 2006/125/EC of 5 Health Nutr 2007;10:1606–1611.
December 2006 on processed cereal-based
foods and baby foods for infants and young
children. Official J Eur Union L 2006;
339:16–35.
146 Zimmermann
Iodine
Current Challenges in Meeting Global

Iodine Requirements
Creswell J. Eastmana ⭈ Pieter Joosteb
a
International Council for Control of Iodine Deficiency Disorders, Sydney Medical School, University
of Sydney, Sydney, NSW, Australia; bInternational Council for Control of Iodine Deficiency Disorders,
Nutritional Intervention Research Unit, Medical Research Council, Cape Town, South Africa
Abstract
Iodine deficiency is a global problem of immense magnitude afflicting 2 billion of the
world’s population. The adverse effects of iodine deficiency in humans, collectively termed
iodine deficiency disorders, result from decreased thyroid hormone production and action,
and vary in severity from thyroid enlargement (goiter) to severe, irreversible brain damage,
termed endemic cretinism. Thyroid hormone is essential throughout life, but it is critical for
normal brain development in the fetus and throughout childhood. During pregnancy,
maternal thyroid hormone production must increase by 25–50% to meet maternal-fetal
requirements. The principal sources of iodine in the diet include milk and dairy products,
seafoods and foods with added iodized salt. Vegetables, fruits and cereals are generally
poor sources of iodine because most of our soils and water supplies are deficient in iodine.
The accepted solution to the problem is Universal Salt Iodization where all salt for human
and animal consumption is iodized at a level of 20–40 μg/g. In principle, mandatory fortifi-
cation represents the most effective public health strategy where safety and efficacy can
be assured and there is a demonstrated need for the nutrient in the population. Voluntary
fortification of salt and other foods has many limitations and few benefits. Iodine supple-
mentation is a useful, but expensive, inefficient and unsustainable strategy for preventing
iodine deficiency. The current worldwide push to decrease salt intake to prevent cardiovas-
cular disease presents an entirely new challenge in addressing iodine deficiency in both
developing and developed countries. Copyright © 2012 S. Karger AG, Basel
Introduction
Iodine is a simple element that is widely distributed in nature occurring in

soils and food, but found most abundantly in the ocean in marine creatures
and seaweeds. Its importance as a micronutrient rests on its essential role as
Brain damage
hypothyroidism
goitre
Hypothyroidism
Risk
UIE 100 μg/l goitre
Hypothyroidism Hyperthyroidism
goitre
Exposure
Fig. 1. Relationship between exposure to a certain iodine intake level over a long period
of time and the risk of developing a thyroid disorder. Reproduced with permission from
Laurberg et al. [4].
a building block for the synthesis of the thyroid hormones tri-iodothyronine

(T3) and tetra-iodothyronine or thyroxine (T4). Iodine is an integral part of the
thyroid hormone molecule, making up approximately two thirds of the mass of
T3 and T4. The thyroid secretes predominantly T4, with only a small amount of
T3 coming directly from the gland. T4 is a prohormone, converted in periph-
eral tissues by deiodinase enzymes to the more metabolically active T3. Thyroid
hormone exerts multiple physiologic actions in the developing fetus, growing
child and the mature adult. Besides being the principal regulator of the meta-
bolic rate in humans, it is the most potent hormonal stimulus for growth and
maturation of both the brain and skeleton [1].
The distribution of iodine in soil and water is quite variable, generally being
adequate or abundant in coastal regions and becoming deficient the further one
travels inland. Severe environmental iodine deficiency is invariably present in
the mountainous zones of the world. By contrast, there are areas where iodine
is present in excessive quantities in the environment, usually related to subter-
ranean water sources. While iodine deficiency is an important and serious cause
of disease, iodine excess may also cause illnesses in predisposed individuals
[2]. The major consequences of deficient or disordered iodine metabolism are
a result of too little (hypothyroidism) or too much (hyperthyroidism) thyroid
hormone [3, 4] (fig. 1).
Iodine Metabolism and Thyroid Hormone Production
Iodine is readily absorbed, in the form of iodide, into the blood from the gastro-
intestinal tract, and is actively transported by an iodine transporter into thyroid
148 Eastman · Jooste

Table 1. Spectrum of IDD throughout the human life cycle
Life stage Clinical expression and adverse outcomes
Fetus Miscarriage or stillbirth

Increased perinatal morbidity and mortality
Neurological and/or myxoedematous cretinism
Decreased IQ, deafness and neuropsychomotor deficits
Neonate Neonatal goiter

Neonatal hypothyroidism
Decreased IQ and neuropsychomotor deficits
Child Endemic goiter

Juvenile hypothyroidism
Stunted physical growth and impaired mental development
Adult Endemic goiter

Hypothyroidism and its complications
Risk of iodine-induced hyperthyroidism
Data from Hetzel [6].
cells where it is concentrated by the thyroid gland. Iodide rapidly inhibits its
own transport when the thyroid gland is exposed acutely to an abrupt increase
in the plasma iodide concentration. Once incorporated into the thyroid gland,
the iodide is oxidized and bound to tyrosine residues in the thyroglobulin mol-
ecule to form T3 and T4. These processes are enzymatically regulated within the
gland and are under the overall control of thyrotropin secreted by the pituitary
gland. The normal adult thyroid needs to trap at least 60 μg of iodide daily to
maintain adequate thyroid hormone production. The newborn infant’s thyroidal
iodine reserve is very small and turnover very rapid. The body conserves iodine
by recycling iodine enzymatically stripped off the T3 and T4 molecules during
metabolic breakdown. Approximately 90% of absorbed iodine is excreted each
day in the urine with insignificant quantities appearing in the feces. Given this
relationship between the amount absorbed and the amount excreted, measure-
ment of urinary iodine excretion (UIE) accurately reflects daily iodine intake,
and has become the most widely used index of nutritional iodine status [5].
Iodine Deficiency Disorders
The adverse effects of iodine deficiency in humans are collectively termed iodine
deficiency disorders or IDD (table 1) [6]. Until recently, iodine deficiency in
a population was usually equated with endemic goiter, without recognition of
the widely prevalent, often subtle, and devastating neurological or psychomotor
Challenges in Meeting Iodine Requirements 149

Fig. 2. Large multinodular goiter in a
mature-age woman from an area of mod-
erate to severe iodine deficiency in north-
ern Thailand.
damage in these populations (fig. 2 and 3). It has long been known that very
severe iodine deficiency affecting people in remote underdeveloped regions of
the world is associated with severe mental and physical impairment (termed
endemic cretinism), but it was not appreciated until recently that less severe
forms of brain damage were so highly prevalent [6, 7]. The WHO says: ‘Iodine
deficiency is the world’s most prevalent – yet easily preventable – cause of brain
damage’ [8]. Iodine deficiency affects approximately 2 billion people worldwide,
but the prevalence has decreased by half over the past 2 decades [9]. In popula-
tions where iodine deficiency is severe, IQ levels are, on average, 13–15 points
below comparable iodine-replete populations, and irreversible neurodevelop-
mental abnormalities are highly prevalent [5, 7]. However, there remains con-
siderable debate about the consequences in the offspring born to mothers with
mild to moderate iodine deficiency.
Thyroid hormone is essential throughout life, but it is critical at certain life
stages. Maternal T4 is the only source of hormone for the fetus early in gestation
before the development of the fetal thyroid (fig. 4) [10]. The adverse effects of
IDD are dependent on the timing of the iodine deficiency and the severity of
the insult from lack of thyroid hormone. During pregnancy, thyroid hormone
synthesis must increase by between 25 and 50% to meet maternal fetal require-
ments [11]. The damage to the developing brain of the fetus, neonate and infant
from lack of thyroid hormone due to iodine deficiency is largely irreversible
[12]. Coexisting deficiencies of selenium, iron and vitamin A may contribute
to the severity of the expression of iodine deficiency [5]. Therefore, our efforts
to prevent brain damage mean we must target our prevention programs at the
pregnant woman and developing child.

Fig. 3. A man afflicted by neurological
cretinism living in a rural area of severe
iodine deficiency in China.
Major period of insult to the CNS: in congenital hypothyroidism

in severe iodine deficiency
T4
From the child
From the mother

Major formative events in human CNS development
Cochlea Myelination
Cerebral cortex
Striatum
Subarachnoid
pathways
Cerebellum
Corpus callosum
Face Dentate of hippocampus
Eye
0 1 2 3 4 5 6 7 8
Gestational age (months) Birth
Fig. 4. Diagrammatic representation of the relationship between gestational age, neuro-

logical development in the fetus and transfer of thyroxine from mother to fetus. Adapted
from Morreale de Escobar et al. [10].

Table 2. Food sources of iodine
Food Level of iodine, μg
Fruit 1 serve <2

Vegetables 1 serve <5
Cheese 40 g 13.6
Seaweed (nori) 1 sheet 16
Eggs 55 g 20
Milk 200 ml 30–60
Iodized salt 1 g 20–40
Sea fish 100 g 35
Adapted from FSANZ [21].
Sources of Iodine in the Diet
The principal sources of iodine in the diet vary from one country and region
to another. In less developed regions of the world lacking iodine in the soil or
water, the most valuable source is likely to be iodized salt. By contrast, in devel-
oped countries where most varieties of food are readily available, milk and dairy
products, seafood, and the discretionary use of iodized cooking and table salt are
the major sources of iodine in the diet (table 2). Vegetables, fruit and cereals are
generally poor sources of iodine because most of our soils and water supplies are
deficient in iodine. The quantity of iodine in salt-water fish and shellfish is very
variable and small compared with its abundance in certain seaweeds. This is the
reason for the very high iodine intake in the traditional diet of the Japanese and
Korean peoples. Other common adventitious sources of iodine include medica-
tions such as amiodarone, cough mixtures, throat lozenges, iodine-containing
antiseptics, multi-vitamin preparations, kelp tablets, and foods and beverages
containing the iodine-containing food coloring agent erythrosine. Many of the
diagnostic contrast agents employed in medical imaging, for example CT scans,
contain huge amounts of iodine.
While a proportion of the iodine in dairy milk comes naturally from pas-
tures, most is the result of contamination by iodine residues in the milk from
iodine-containing sanitizers called iodophors. The dairy industries in Australia
and New Zealand – as many other countries in the world – have employed
iodophors as sanitizing agents for the past 4–5 decades [13]. These chemicals
are disinfectants consisting of iodine coupled to an anionic detergent. Iodine
is released into the solution to exert its bactericidal activity. When used in the
dairy industry, the amount of residual iodine in milk is highly variable and, in
our experience, may result in anything from negligible quantities to 500 μg/l.
Iodine-contaminated milk has been the major source of this micronutrient in

Table 3. Recommended daily intake of iodine
Population group Recommended daily intake of Excessive iodine intake, μg/day

iodine, μg/day
WHO IOM (USA) NHMRC WHO NHMRC

(Australia) upper level of intake
Infants 90 110–130 90 >200 200 (0–3 years)

Children 120 90–120 90–120 >500 300 (4–8 years)
600–900 (9–18 years)
Adult men and women 150 150 150 >500 1,100
Pregnant women 250 220 220 >500 1,100
Breastfeeding women 250 290 270 >500 1,100
Data sourced from Eastman and Zimmermann [3]. IOM = Institute of Medicine; NHMRC =
National Health and Medical Research Council of Australia.
the Australian diet for the past 40 years [13, 14]. This has been termed silent pro-
phylaxis resulting in ‘an accidental public health triumph’. Unfortunately, over
the past decade there has been a substantial decline in the sales of iodophors in
Australia with chlorine-based disinfectants replacing the iodine-based ones. We
have been monitoring iodine concentrations in retail milk samples in Sydney
and found that these levels have been declining and currently are between 90
and 200 μg/l with an average of approximately 130 μg/l [14]. These levels are
less than half of what we measured a decade or so ago, and have resulted in the
reemergence of iodine deficiency in Australia and New Zealand [15]. It is quite
an extraordinary situation that the Australian population has, for decades, relied
upon this adventitious source of iodine to maintain adequate iodine nutrition,
and we are now facing a significant public health problem because of declining
iodine contamination of dairy milk.
Recommended Intakes of Iodine at Different Stages of Life
Several countries and organizations throughout the world have adopted recom-
mended dietary intakes, allowances or reference values for iodine nutrition (table
3) [3]. Most of these recommendations are very similar; however, recently there has
been a tendency to increase recommended intakes. The recommendations are very
arbitrary, and most are not based on solid scientific evidence. Unfortunately, we do
not have rigorous experimental-based evidence to make these recommendations
and must rely on indirect evidence. Indeed, most of the early recommendations
on iodine intake were aimed at providing sufficient iodine in the diet simply to
prevent development of palpable or visible goiter, and were made in the absence of

our understanding of subtle but significant deficits in neurological and intellectual
function occurring in response to minor decreases in maternal thyroid function.
Recommended daily intake for iodine is the average daily intake to meet the iodine
requirement of 97.5% of healthy individuals at any specific life stage.
Excessive Iodine Intake and Recommended Upper Limit of Normal Intake
The normal adult thyroid gland adjusts to both abrupt and sustained increases
in iodine intake by in-built mechanisms that regulate absorption of iodine at the
level of the thyroid cell. However, fetal and neonatal thyroid glands cannot do
this very well, and exposure to large quantities of iodine may cause goiter and
hypothyroidism in the developing baby. Similarly, individuals suffering from
autoimmune thyroid disease are unable to compensate and may develop hypo-
thyroidism or hyperthyroidism when exposed to iodine excess [4]. An arbitrary
iodine intake of around 1,000–1,100 μg/day has been set as the upper limit of
normal in Australia and most other countries in the world. While most of the
public health emphasis has been on efficacy to ensure an adequate daily intake,
insufficient emphasis has been placed on the concept of safety and how these
two objectives can be satisfactorily achieved.
Current Scientific Issues in Iodine Deficiency Disorders Research
Zimmermann has recently emphasized the need for more research to resolve
issues relating to (a) iodine nutrition in pregnancy and infancy, (b) potential
adverse effects of iodine deficiency on cognitive development, especially the
relative contributions of maternal versus childhood iodine deficiency to neu-
rodevelopment, (c) safe upper limit of iodine intake, (d) laboratory methods for
monitoring salt iodine content, (e) metabolic interactions of iodine and other
micronutrient deficiencies.
Fortification of Food with Iodine
In principle, mandatory fortification represents the most effective public health

strategy where safety and efficacy can be assured, and there is a demonstrated
need for the nutrient in the population. Voluntary fortification has many limita-
tions and few benefits. [16]. It is generally inequitable, poses major difficulties
in monitoring, and it usually fails because it is not sustainable. By contrast, man-
datory fortification is a more equitable strategy, reaching most of the popula-
tion, and is readily monitored allowing adjustment in the fortification level, and
hence is more likely to be sustainable (table 4) [16].

Table 4. Assessment of potential public health strategies to correct iodine deficiency in a population [16]
Mandatory Voluntary Fortification of other Supplements Dietary education

USI fortification foods, e.g. water, bread
Effectiveness yes possible incomplete yes no

Equity yes no no no no
Efficiency yes no yes/no no no
Certainty yes no no no no
Feasibility yes yes yes yes yes
Sustainability yes probably not yes probably not probably not
Universal Salt Iodization: Salt Fortification

The recommended strategy for control and prevention of iodine deficiency
is mandatory universal salt iodization (USI) that requires the iodization of
all salt for human and animal consumption at a level to provide sufficient
iodine in the diet for most of the population without exposing them to an
excessive iodine intake. The recommended level of iodization is 20–40 mg
iodine/kg salt [17]. Salt is usually considered the best food to fortify with
iodine because it is widely consumed by most people in the world regard-
less of income; it is consumed consistently in small amounts by individuals;
it is inexpensive to iodize, and there is no chemical reaction altering taste or
color of the salt.
The WHO estimates that approximately two thirds of 5 billion people liv-
ing in areas of iodine deficiency now have access to iodized salt. Over the past
decade, the number of countries employing USI has doubled from around 50
to over 100, but there are over a dozen developing countries where there has
been little or no progress [9, 17]. In the developed world, USI has not gen-
erally been adopted. Australia is a good example. Iodized salt is available in
Australia, but the market size for iodized salt is only 25% of the total edible salt
market. Food standard codes regulate iodine concentrations at between 25 and
45 μg/g. Why are the sales of iodized table salt in Australia so poor compared
with non-iodized salt? We do not have a definitive answer to this question, but
it appears to be due to several factors; the consumer is unaware of the benefits
of iodized salt, health authorities do not promote its use and the food indus-
try does not market it. To our knowledge, iodized salt is not used in the food
manufacturing industry or at the retail or fast food outlets. With increasing
consumer awareness of the adverse effects of high salt intake on blood pressure
and cardiovascular disease, we might reasonably expect that iodine intake from
iodized salt will decrease even further with time. It is likely that the informed
consumer will first reduce the discretionary salt intake and hence exacerbate
iodine deficiency.

Barriers to Adoption of Universal Salt Iodization
Major efforts have been made throughout the world to achieve the goal of USI,
particularly in developing countries with high prevalence of iodine deficiency.
There are many outstanding examples of countries who have achieved a pen-
etration of over 90% of households using iodized salt. Two of the best exam-
ples in the Asia-Pacific region are the People’s Republic of China and Vietnam.
Essential elements in successful programs require legislation underpinning USI
and national monitoring programs to assess the effectiveness of these programs.
Many countries claim to have implemented USI, but in effect the iodization of
salt is not universal. Despite the continuing belief by international agencies that
USI can be implemented universally and can be sustained, this is simply not
possible in many countries for several reasons.
In developed countries, barriers to USI include:
• Most of the salt intake in the diet (>75%) comes from processed foods not
from the discretionary use of household salt. While most developed countries
have household iodized salt available for purchase in food stores, non-iodized
salt is also readily available.
• Large multinational food-producing companies resist the pressure to use
iodized salt in food processing and manufacturing. The reasons for this
resistance include costs for production and labeling, costs for meeting
regulatory requirements and trade barriers erected by countries who have
legislated against the importation of foods containing iodized salt.
• Increasingly successful campaigns to reduce population salt intake [18].
In developing countries, barriers to USI include:
• A large proportion of the salt intake in the diet comes through salty
condiments not through the use of household salt in many countries,
particularly Asian countries; iodized salt is not employed in the manufacture
of these products.
• The inability of countries to implement and/or police iodized salt legislation;
the lack of political will is the usual cause for lack of implementation.
• The ready availability of less expensive non-iodized salt in the marketplace.
• Lack of political will and commitment to sustaining USI.
• Concerns about the safety of USI and the fear of potential adverse effects.
The well-orchestrated salt-reduction campaign originating in the UK in
2005 is now active in 80 countries and expanding rapidly [18]. This campaign
is founded on the premise that the current estimated high salt intake, in excess
of 10 g/day in developed countries, is a major cause of hypertension and cardio-
vascular disorders [19]. The primary goal is to decease salt intakes to the WHO
target of <5 g/day. As this movement becomes more widespread, International
Agencies (UNICEF, WHO, ICCIDD) will have to revise the recommended lev-
els of iodine in edible salt based upon population monitoring. In countries like
the People’s Republic of China there has been a public backlash against USI with
claims by several medical professional groups that the Chinese population is

ingesting excessive quantities of iodine, as a consequence of USI, and this is
causing a rising incidence of autoimmune thyroid disease and thyroid cancer
[20].
Fortification of Other Foods

The addition of iodine to bread, drinking water, rice, fruits, eggs and fish
sauce, and complementary foods has been tried in several countries with vary-
ing degrees of success. The Netherlands has continued the practice of using
iodized salt in bread for several decades without any problems. Australia and
New Zealand mandated the use of iodized salt in bread in 2009 in an attempt
to address the reemergence of iodine deficiency in both of these countries, but
recognized this initiative would not satisfy iodine requirements in pregnant and
breastfeeding women, the newborn and young infants [21]. Therefore, these
strategies may be useful and effective in limited geographical situations, but will
need to be complemented by iodine supplementation directed to the most vul-
nerable groups [22]. It is a matter of ‘think globally but act locally’.
Iodine Supplementation
Iodine supplementation to specific individuals or subgroups of an iodine-

deficient population has been widely practiced, but remains a controversial
strategy. Until recently, this strategy did not receive much support from the
international agencies, such as WHO/UNICEF/ICCIDD who argued that sup-
porting iodine supplementation programs would detract from USI. Dietary
education and supplementation are generally inefficient public health strategies
because of the level of investment necessary to implement such interventions,
and they are confounded by equity and access issues [16]. Those with higher
socioeconomic and educational levels are more likely to benefit from education
and promotion of supplements frequently making the strategies inequitable.
Education of the population and health care providers is a prerequisite for any
supplementation program. However, there is a place for iodine supplementation
during pregnancy and breastfeeding as a complementary strategy, particularly
where voluntary or mandatory salt iodization has not been implemented or is
ineffective [22].
Supplements can be given on a daily basis by oral administration of potas-
sium iodide pills (150–250 μg/day) or annually by way of iodized oil capsules
(400 mg dose) which are slow-release depot preparations that provide enough
iodine for up to 12 months’ duration [17]. While there are numerous studies
demonstrating the effectiveness of iodized oil in preventing IDD, there are no
data from randomized control trials and very little in the way of information
confirming the safety of this material. The current recommendations agreed by
WHO, ICCIDD and UNICEF are for the administration of iodine supplements

to pregnant and breastfeeding women and infants up to 2 years of age living
in areas where USI has not been implemented or is ineffective [17]. Several
influential scientific and medical bodies – including the American Thyroid
Association, the American Endocrine Society and the National and Medical
Research Council in Australia – now recommend daily iodine supplementation
for pregnant and breastfeeding women. Recommendations for infants and chil-
dren and the need to ensure optimal iodine intake in complementary foods are
covered elsewhere in this symposium.
Monitoring and Sustainability
The success of all iodine fortification or supplementation programs is depen-

dent on population monitoring. In the past, monitoring has targeted school-
age children, measuring goiter prevalence and UIE. It is clear that monitoring
should be directed to the most vulnerable groups, namely pregnant and lactat-
ing women, neonates and infants.
References
1 Zimmermann MB, Jooste PL, Pandav CS: 8 Qian M, Wang D, Watkins WE, et al: The
The iodine deficiency disorders. Lancet 2008; effects of iodine on intelligence in children: a
372:1251–1262. meta-analysis of studies conducted in China.
2 Li M, Liu DR, Qu CY, et al: Endemic goitre Asia Pacific J Clin Nutr 2005;14:32–42.
in central China caused by excessive iodine 9 De Benoist B, McLean E, Andersson M:
intake. Lancet 1987;ii:257–259. Iodine deficiency: The extent of the prob-
3 Eastman CJ, Zimmermann MB: Iodine lem; in Preedy VC, Burrow GN, Watson RR
Deficiency Disorders. Chapter 20. http:// (eds): Comprehensive Handbook on Iodine:
www.thyroidmanager.org. Nutritional, Endocrine and Pathological
4 Laurberg P, Bülow Pedersen I, Carlé A, et Aspects. Amsterdam, Elsevier, 2009, chapter
al: The U-shaped curve of iodine intake and 48, pp 461–467.
thyroid disorders; in Preedy VC, Burrow GN, 10 Morreale de Escobar G, Obregon M, Escobar
Watson RR (eds): Comprehensive Handbook del Rey F: Role of thyroid hormone during
on Iodine: Nutritional, Endocrine and early brain development. Eur J Endocrinol
Pathological Aspects. Amsterdam, Elsevier, 2004;151(suppl 3):25–37.
2009, chapter 47, pp 449–455. 11 Glinoer D: The regulation of thyroid func-
5 Zimmermann MB: Iodine deficiency. Endocr tion in pregnancy: pathways of endocrine
Rev 2009;30:376–408. adaptation from physiology to pathology.
6 Hetzel BS: Iodine deficiency disorders (IDD) Endocr Rev 1997;18:404–433.
and their eradication. Lancet 1983;ii:1126– 12 Halpern JP, Boyages SC, Maberly GF, et al:
1129. The neurology of endemic cretinism. Brain
7 Vermiglio F, Sidoti M, Finocchario MD, et al: 1991;114:825–841.
Defective neuromotor and cognitive ability 13 Eastman CJ: Where has all our iodine gone?
in iodine deficient children of an endemic Med J Aust 1999;171:455–456.
goitre region in Sicily. J Clin Endocrinol
Metab 1990;70:379–384.

14 Li M, Waite KV, Ma G, Eastman CJ: Declin- 18 Feng J, Jenner KH, Farrand MacGregor GA:
ing iodine content of milk and the re- Editorial – World Salt Awareness Week. J
emergence of iodine deficiency in Australia. Clin Hypertens 2011;13:141–145.
Med J Aust 2006;184:307. 19 Bibbins-Domingo K, Chertow GM, Coxson
15 Li M, Eastman CJ, Waite KV, et al: Are PG: Projected effect of dietary salt reductions
Australian children iodine deficient? Results on future cardiovascular disease. N Engl J
of the Australian National Iodine Nutrition Med 2010;362:590–599.
Study. Med J Aust 2006;184:165–169. 20 Teng W, Shan Z, Teng X, et al: Effect of
16 Stanley F, Eastman C, Mann J, Binns C: The iodine intake on thyroid diseases in China. N
effectiveness of mandatory fortification as Engl J Med 2006;354:2783–2793.
a public health strategy to increase nutrient 21 Food Standards Australia New Zealand 2008.
intakes, with reference to iodine and folate. Proposal P1003 – Consideration of manda-
Expert Public Health Advice; prepared for tory iodine fortification for Australia and
the Australian Health Ministers Advisory New Zealand. Dietary intake assessment
Council, 2005. report. http://www.foodstandards.gov.au/_
17 Andersson M, de Benoist B, Delange F, srcfiles/P1003%20SD10%20-%20Dietary%20
Zupan J: Prevention and control of iodine Intake%20Assessment.pdf.
deficiency in pregnant and lactating women 22 Gallego G, Goodall S, Eastman CJ: Iodine
and in children less than 2 years old: conclu- deficiency in Australia: is iodine supplemen-
sions and recommendations of the techni- tation for pregnant and lactating women
cal consultation. Public Health Nutr 2007; warranted? Med J Aust 2010;192:461–463.
10:1606–1611.

Folate and Vitamin B12: Function and

Importance in Cognitive Development
Aron M. Troen
Nutrition and Brain Health Laboratory, Institute of Biochemistry, Food Science and Nutrition,
The Robert H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University of
Jerusalem, Rehovot, Israel
Abstract
The importance of the B vitamins folate and vitamin B12 for healthy neurological develop-
ment and function is unquestioned. Folate and vitamin B12 are required for biological
methylation and DNA synthesis. Vitamin B12 also participates in the mitochondrial catabo-
lism of odd-chain fatty acids and some amino acids. Inborn errors of their metabolism and
severe nutritional deficiencies cause serious neurological and hematological pathology.
Poor folate and vitamin B12 status short of clinical deficiency is associated with increased
risk of cognitive impairment, depression, Alzheimer’s disease and stroke among older
adults and increased risk of neural tube defects among children born to mothers with low
folate status. Folate supplementation and food fortification are known to reduce incident
neural tube defects, and B vitamin supplementation may have cognitive benefit in older
adults. Less is known about folate and vitamin B12 requirements for optimal brain devel-
opment and long-term cognitive health in newborns, children and adolescents. While
increasing suboptimal nutritional status has observed benefits, the long-term effects of
high folate intake are uncertain. Several observations of unfavorable health indicators in
children and adults exposed to high folic acid intake make it imperative to achieve a more
precise definition of folate and B12 requirements for brain development and function.
Introduction – Biochemical Function of Folate and Vitamin B12
The importance of the B vitamins folate and vitamin B12 for healthy neurological
development and function is unquestioned. As essential cofactors for one-car-
bon metabolism, folate and vitamin B12 are required for biological methylation
and DNA synthesis. Methylation activity is essential for many processes that
are critical for central nervous function, such as the epigenetic regulation of
gene expression by DNA and histone methylation, the methylation of myelin
basic protein and membrane phosphatidylcholine, the synthesis of hormones
and neurotransmitters such melatonin and epinephrine, and the inactivation of
dopamine and catecholamines. Folate (in the form of 5-methyltetrahydrofolate)
and vitamin B12 are required in all tissues to maintain adequate cellular methi-
onine, which in turn is converted to S-adenosylmethionine the universal methyl
donor for these reactions. Methylations produce the sulfur amino acid homo-
cysteine, which serves as precursor for the cyclical regeneration of methionine.
Choline is another important dietary methyl donor, providing an alternative
source of labile methyl groups for regeneration of methionine by an alternative
pathway which is only expressed in the liver and kidney.
Disrupting these pathways can have wide-ranging direct and indirect con-
sequences for neural tissue. Folate and B12 deficiency inhibit methylation and
cause homocysteine to accumulate in circulation. Homocysteine may be cyto-
toxic, and mildly elevated plasma homocysteine is associated with increased
risk of neurodegenerative and cardiovascular disease. By increasing choline
utilization or by depleting membrane phosphatidylcholine, folate deficiency
might also indirectly limit the synthesis of the key neurotransmitter ace-
tylcholine. Folate and B12 are also required for activity that is independent
of each other. Folate is needed to synthesize thimidylate and purine nucle-
otides. Thus, folate deficiency can compromise DNA replication and integ-
rity. Vitamin B12 serves as a coenzyme for methylmalonyl-CoA mutase which
catalyzes the final step in the mitochondrial degradation of odd chain fatty
acids and branched amino acids. Thus, in addition to impinging upon folate
and methylation, B12 deficiency leads to abnormal membrane fatty acid com-
position and elevation of circulating methylmalonic acid. These biochemical
effects of deficiency can be theoretically linked to the observed neuropatho-
logical, cognitive and developmental abnormalities that are associated with
deficiency of these vitamins [1].
Neurological Consequences of Acquired and Congenital Defects in One-

Carbon Metabolism
The importance of the B vitamins folate and vitamin B12 for healthy brain devel-
opment and function is evident in the serious neurological consequences of
acquired nutritional deficiencies and inborn errors of their metabolism. Severe
acquired deficiencies of vitamin B12 and folate manifest with either hematologi-
cal or neurological abnormalities or a combination of both [2–6]. A large pro-
portion of cases develop neurological symptoms without exhibiting any major
hematological abnormality, and this tendency is thought to be more preva-
lent among the elderly [7]. The primary neurological feature is a progressive
162 Troen
neuropathy in which sensation and motor control are gradually lost, typically
beginning with the lower limbs [6]. In addition, these conditions are often
accompanied by behavioral changes ranging from mild irritability to severe
depression, hallucinations, confusion and memory loss [8]. If the condition is
left untreated, it may eventually lead to sub-acute combined degeneration of
the spinal cord. The dominant histopathological finding in this condition is a
patchy loss of myelin in the white matter of the spinal cord where myelin-laden
macrophages surround the empty spaces, although brain and peripheral myelin
are also affected [7]. Prognosis depends on intervention. Although vitamin sup-
plementation is an effective treatment in many cases, cumulative neurodegener-
ative damage may become irreversible if it is allowed to progress too far. It is not
clear whether the behavioral changes are due to vitamin-related neurochemical
imbalances or to structural changes in the brain caused by demyelination [8].
Genetic impairment of one-carbon metabolism can produce similar
pathology. For example, mutations that severely limit the synthesis of 5-meth-
yltetrahydrofolate by methylene tetrahydrofolate reductase (MTFHR) typi-
cally result in abnormal neurological development, progressive demyelinating
neuropathy and cognitive impairment [9–11]. Although congenital defects
in MTFHR activity usually become evident soon after birth, they have been
known occasionally to remain silent until the onset of neurological symp-
toms and psychotic or schizophrenic behavior in adolescence or adulthood
[12–15]. Gene-gene and gene-nutrient interactions appear to influence the
severity and expression of phenotype in these cases. This is implied by cases
where some family members carrying the same mutation are symptomatic
while others are not [13, 14]. In many cases, MTFHR deficiency is partially
responsive to supplementation with folate or betaine [12, 13]. Similar find-
ings are observed in mice with a targeted mutation in the MTHFR gene. In
homozygotes for the mutation, CNS development is compromised, but this
can be partially mitigated by dietary supplementation with betaine [16].
Cerebral folate deficiency is another example of folate-dependent impairment
of CNS development and function. In this rare pediatric autoimmune condi-
tion, auto-antibodies to the choroid plexus high-affinity folate transporter are
associated with low cerebrospinal folate despite normal serum folate, and with
progressive impairment of neurological development, myelination and cog-
nition [17]. In some cases, the symptoms can be partially mitigated, at least
temporarily, by folinic acid (5-formyl tetrahydrofolate), which can enter the
CNS through an alternative reduced folate transporter. Vitamin B12 deficiency
in newborns and infants who have smaller liver and tissue stores [18] is often
due to maternal deficiency. Symptoms of clinical deficiency can be treated
with intramuscular injections and high-dose oral supplementation, leading to
hematologic and neurologic recovery; however in the long-term, resolving the
deficiency and acute pathology does not guarantee full restoration of cogni-
tive development [19].
Folate and Vitamin B12: Function and Importance in Cognitive Development 163
Subclinical Deficiencies in CNS Development and Aging
The neuropathology common to both congenital and severe acquired vitamin

B12 and folate deficiencies demonstrates the importance of methylation for
brain development and function, yet these deficiencies are relatively rare. More
often, individuals who fall in the lower range of population folate and vitamin
B12 intake and plasma concentrations, but who are not clinically deficient, have
significantly higher risk of poor neurological and cognitive outcomes. These
include increased risk of neural tube defects (NTDs) among children born to
mothers with low folate status, and increased risk of cognitive impairment,
depression, Alzheimer’s disease and stroke among older adults.
With respect to aging, the association between low B vitamin status and
increased risk of neurodegenerative disease has been extensively reviewed else-
where [20, 21]. Data from over 100 observational cross-sectional and prospec-
tive studies, encompassing over 50,000 subjects in total, provide compelling
evidence for the association, with approximately 90% of the studies reporting
significant associations and the remainder not. Taken together, they provide
the basis for the tenable hypothesis that ‘low-normal intake or blood concen-
trations of B vitamins (folate, B12 and B6) and/or moderately elevated plasma
total homocysteine increase the risk of brain atrophy and developing cognitive
impairment in the elderly’ [1, 21, 22]. As demyelination is not a dominant fea-
ture of these diseases, a variety of alternative mechanisms has been proposed to
account for these associations, relating both vitamin deficiencies and elevated
homocysteine to different aspects of cognitive dysfunction and neuropathology
[23]. The efficacy of B vitamin supplementation for neurocognitive protection
has yet to be determined, with several trials reporting null findings [24–26] and
others suggesting benefit [27, 28].
With respect to early life, the prevention of NTDs by folic acid provides
strong evidence of folate’s critical importance to CNS development, although
the precise mechanism for the phenomenon is still unknown. Early observa-
tions that poor folate status is associated with increased risk of spina bifida and
anencephaly led to highly successful clinical trials in which use of periconcep-
tual folic acid supplements reduced the risk of incident NTDs by as much as
50% or more [29]. Further evidence for folate’s importance for neural develop-
ment comes from genetic association studies. A meta-analysis of more than 17
different studies including several thousand subjects calculated a pooled odds
ratio giving an approximately twofold increase in risk of an NTD for mothers
and children who are TT homozygotes compared with CC homozygotes for the
C677->T single nucleotide polymorphism in the MTHFR gene, where the TT
genotype encodes a slightly less active enzyme. Risk for NTDs among the TT
homozygotes decreased with increased folate status [30]. Folate supplementa-
tion before conception is critical for preventing NTDs given that neural tube
closure occurs early in embryonic development when the mother may not be
164 Troen
aware of the pregnancy. Despite this, most women of childbearing age do not
take supplements, particularly those at risk for nutritional deficiency. For this
reason, mandatory food fortification of all cereal grains with folic acid has been
enacted in over 50 countries. These public health programs have been highly
successful with respect to reduction of incident NTDs. They have also signifi-
cantly increased folic acid intake across populations and lowered the prevalence
of hyperhomocysteinemia. In the Framingham study in the USA, the prevalence
of folate deficiency decreased from 4.9 to 1.9% after fortification [31], but at the
same time, the proportion of individuals exceeding the recommended upper
intake limit of 1 mg folic acid/day grew from 1.3 to 11.3% [32]. Interestingly,
following fortification, vitamin B12 and choline status have been found to deter-
mine NTD risk [33, 34].
Folate and B12 Supplementation and Cognitive Development
All of these examples relate to pathological consequences of vitamin deficiency.

But what effect does supplementation beyond replacement have on brain devel-
opment and attainment of full intellectual potential? Here, the data are surpris-
ingly scarce. Answering this question is important in light of the possibility that
exposure to high levels of folic acid may have unintended effects, for better or
for worse. The upper limit of folate intake for adults was conservatively set at 1
mg per day out of concern for the possibility that in individuals with frank B12
deficiency, high folate might either exacerbate the symptoms or ‘mask’ the asso-
ciated anemia, delaying early detection and treatment and thus allowing it to
progress [35]. This concern relates to a relatively small though not insignificant
proportion of the population. There is deeper controversy over whether forti-
fication has increased cancer risk and decreased risk of cardiovascular disease
and stroke [36–38]. Finally, despite scant data, concerns have also been raised
over theoretical mechanisms through which high folate intake might exert
subtle but substantial life-long effects on metabolism and neurological develop-
ment through altered epigenetic programming [39, 40].
A small number of recent observations show unexpected evidence for meta-
bolic interaction of high folate status with low-marginal B12 status. Cross-sectional
data from the NHANES and SALSA cohorts show that the metabolic effects of
vitamin B12 deficiency are more pronounced in adults with high folic acid intake.
MMA and Hcy levels are elevated in individuals with low B12 and high folate sta-
tus compared to those with normal folate and normal or low B12 [41, 42]. These
metabolic findings are mirrored by poorer neuropsychological test scores on
the symbol digit test and by lower hemoglobin values [43]. Similar associations
between worse metabolic and functional impairments with B12 deficiency were
found when unmetabolized folic acid in plasma was used as a putative proxy for
high folic acid intake [44]. Irrespective of B12 status, high folic acid intake has also
been observed to associate with more rapid cognitive decline in older adults [45]
and with decreased immune function in older postmenopausal women [46].
The same pattern of a potentially undesirable interaction between high folate
and low B12 has been seen in children. A large study of 2,812 Columbian school
children aged 5–12 years found that hemoglobin concentrations decreased with
increasing erythrocyte folate [47]. There was a significant interaction between
folate and B12 status such that differences in hemoglobin concentrations
between the highest and lowest quartile of folate were greatest in children with
low B12 status (<148 pmol/l) followed by marginal B12 status (between 188–221
pmol/l) and least in B12-sufficient children. Notably, the prevalence of anemia
in this cohort was only 3.7%, and data were adjusted for relevant biological and
socioeconomic factors. The same interaction pattern has been observed in an
entirely different setting, the Maternal Nutrition Study in Pune, India, where
higher maternal folate during pregnancy predicted greater insulin resistance in
offspring 6 years later, particularly in mothers with low B12 [48].
Cognitive development was not reported in these studies; however, a handful
of other studies yield inconsistent associations between maternal folate and B12
status with cognitive development in children. In a cohort of 536 children from
Mysore, India, higher maternal folate was associated with better performance
on a battery of neuropsychological tests after appropriate adjustment for con-
founding, irrespective of vitamin B12 status, and despite the fact that 42.5% of
mothers had low vitamin B12 during pregnancy. In this cohort, maternal B12 and
homocysteine concentrations did not predict children’s cognitive performance
[49]. In contrast to these findings, maternal B12 status predicted cognitive per-
formance in 9-year-old offspring in a small subset of the Pune cohort. Children
born to mothers in the lowest B12 decile (<77 pmol/l) performed significantly
worse than those whose mothers were in the highest decile of B12 (>224 pmol/l)
on two out of 7 neuropsychological tests used for the examination (digit span
backward and color trail test A). Plasma folate, homocysteine and methylma-
lonic acid were not associated with cognition [50]. A cross-sectional study of
598 Indian children aged 6–10 years revealed an unexpected inverse association
between plasma B12 status and factor scores for cognitive domains of short-term
memory, retrieval ability and an overall composite score for mental processing.
In this cohort, increasing hemoglobin concentration was positively associated
with better cognitive performance, but iron, iodine, fatty acids and folate showed
no association [51]. A comparison of cognitive performance among adolescents
(age 10–16) who were either vegan-macrobiotic or omniverous until the age of
6, found poorer performance on cognitive tests among the previously macrobi-
otic group, which also had a higher prevalence of low but not deficient vitamin
B12 status. Finally, an early study found severe mental retardation among chil-
dren born to mothers with folate deficiency [3], but a more recent study found
no relation of neuropsychological development in 5-year-old children to mater-
nal folate status during the second half of pregnancy [52].
166 Troen
A Role of Folate and Vitamin B12 in Epigenetic Programming of
Neurocognitive Development?
The prevention of NTDs by relatively modest doses of folic acid demonstrates

the profound effect that folate-dependent metabolism has during CNS develop-
ment. However, there is no reason to assume that the mechanism for this effect
is specific only to neural tube closure. A leading hypothesis for the mechanism
involves epigenetic mechanisms – the stable and heritable regulation of gene
without altering gene sequence that is regulated by DNA and histone methy-
lation. Epigenetic patterns that are established early in life program life-long
effects on development, function and aging. Aberrant DNA methylation is criti-
cally involved in the neurodevelopmental disorders Praeder Willi and Angelman
syndromes, and is also hypothesized to be involved in Rett syndrome, schizo-
phrenia and autism [53, 54]. Moreover, recent animal studies have shown that
experience-dependent, reversible changes in DNA and histone methylation are
critically involved in the regulation of synaptic plasticity, learning and memory
in nondividing neurons of the adult brain. Thus, it is reasonable to hypothesize
that even mild alterations of methylation during brain development as a func-
tion of intake of folic acid, vitamin B12, choline or other nutrients would have
long-lasting impact on brain development and function [55, 56].
Evidence that methyl donor availability can determine stable epigenetic
changes in brain development and cognitive function comes largely from ani-
mal studies. In rodents, providing dams with choline at critical windows dur-
ing embryonic development changes DNA methylation and gene expression
in the brain of the offspring, induces structural changes involving neuronal
stem cell proliferation and angiogenesis in hippocampus, and enhances mem-
ory. Moreover maternal choline supplementation prevents normal age-related
cognitive decline in elderly rats [57]. However, the beneficial effects of supple-
mentation may have upper limits. Folic acid has been shown to significantly
improve nerve regeneration following spinal cord injury in adult rats. The effect
of folate followed a nonlinear dose-dependent relationship with regeneration,
where the enhancement of regeneration peaked at 80 μg folic acid per kg body
weight and declined significantly thereafter. This inverse U-shaped relationship
was paralleled by increasing expression of the DNA methyltransferase enzyme 3
(DNMT3) up to 80 μg/kg and decreased expression at higher doses [58].
Although data from human studies are limited, the available data show that
even modest supplementation within the normal range of dietary intake can
have significant epigenetic effects. A recent study documented the effect of using
periconceptual folic acid (400 μg/day) on epigenetic regulation of the insulin-
like growth factor 2 gene (IGF2) in offspring. Lymphocyte DNA methylation in
the promoter region of the IGF2 gene was significantly (4.5%) higher in chil-
dren born to mothers who took folic acid supplements (n = 86), compared to
those not exposed to folic acid (n = 34). Although birthweight was similar in the
two groups, a 1.7% higher methylation was associated with one standard devia-
tion decrease in birthweight (584 g) independent of periconceptual exposure
to folic acid or gestational age at delivery [59]. Another human study found a
nonlinear U-shaped association of plasma folate with DNA telomere length in
a non-fortified population of older Italian adults. Telomere length (a marker
for biological aging) declined with increasing folate until the median folate
level, and then increased again as folate increased above the median [60]. The
authors speculated that epigenetic mechanisms were involved. If such changes
to peripheral DNA are also reflected in the brain, this would suggest that brain
development, cognitive function and aging can be modulated by folate and B12
availability across the current range of vitamin intake from diet, fortification
and supplement use. Given the clear importance of folate and vitamin B12-
dependent metabolism in brain development and function, it would not be sur-
prising if these processes also influenced the development and realization of full
cognitive and intellectual potential.
Conclusions
Taken together, the demonstrable importance of folate supplementation for NTD

prevention, the paucity of data on the impact of folic acid and B12 supplementa-
tion on neurocognitive development in human populations, the observation of
both favorable and potentially unfavorable outcomes in association with folate
intake alone or in association with low B12 status, and the uncertainty regard-
ing the mechanisms that underlie such observations all suggest it is important
to improve folate and vitamin B12 status in deficient populations, possibly in
conjunction with choline, and to avoid excess. In order to improve on current
guidelines and recommendations by revisiting the upper limits, and to delineate
their risks and benefits, we require a much more precise understanding than is
currently available of the impact of these nutrients on brain development and
cognitive function, and the mechanisms through which they act on the brain
throughout the lifespan.
References
1 Selhub J, Bagley LC, Miller J, et al: B vita- 3 Gross JS, Weintraub NT, Neufeld RR, et al:
mins, homocysteine, and neurocognitive Pernicious anemia in the demented patient
function in the elderly. Am J Clin Nutr without anemia or macrocytosis. A case
2000;71:614S–620S. for early recognition. J Am Geriatr Soc
2 Botez MI: Folate deficiency and neuro- 1986;34:612–614.
logical disorders in adults. Med Hypotheses 4 Parry TE: Folate responsive neuropathy.
1976;2:135–140. Presse Med 1994;23:131–137.
168 Troen
5 Rosenblatt DS, Whitehead VM: Cobalamin 16 Chen Z, Schwahn BC, Wu Q, et al: Postnatal
and folate deficiency: acquired and heredi- cerebellar defects in mice deficient in meth-
tary disorders in children. Semin Hematol ylenetetrahydrofolate reductase. Int J Dev
1999;36:19–34. Neurosci 2005;23:465–474.
6 Weir DG, Scott JM: The biochemical basis 17 Ramaekers VT, Rothenberg SP, Sequeira JM,
of the neuropathy in cobalamin deficiency. et al: Autoantibodies to folate receptors in
Ballieres Clin Haematol 1995;8:479–497. the cerebral folate deficiency syndrome. N
7 Lindenbaum J, Healton EB, Savage DG, et al: Engl J Med 2005;352:1985–1991.
Neuropsychiatric disorders caused by coba- 18 Dror DK, Allen LH: Effect of vitamin B12
lamin deficiency in the absence of anemia or deficiency on neurodevelopment in infants:
macrocytosis. N Engl J Med 1988;318:1720– current knowledge and possible mechanisms.
1728. Nutr Rev 2008;66:250–255.
8 Allen RH, Stabler SP, Lindenbaum J: 19 Graham SM, Arvela OM, Wise GA: Long-
Relevance of vitamins, homocysteine and term neurologic consequences of nutritional
other metabolites in neuropsychiatric vitamin B12 deficiency in infants. J Pediatr
disorders. Eur J Pediatr 1998;157(suppl 1992;121:710–714.
2):S122–S126. 20 Troen A, Rosenberg I: Homocysteine and
9 Clayton PT, Smith I, Harding B, et al: cognitive function. Semin Vasc Med 2005;
Subacute combined degeneration of the 5:209–214.
cord, dementia and parkinsonism due to an 21 Smith AD: The worldwide challenge of
inborn error of folate metabolism. J Neurol the dementias: a role for B vitamins and
Neurosurg Psychiatry 1986;49:920–927. homocysteine? Food Nutr Bull 2008;
10 Hyland K, Smith I, Bottiglieri T, et 29(suppl):S143–S172.
al: Demyelination and decreased 22 Rosenberg IH, Miller JW: Nutritional
S-adenosylmethionine in 5,10-methyl- factors in physical and cognitive func-
enetetrahydrofolate reductase deficiency. tions of elderly people. Am J Clin Nutr
Neurology 1988;38:459–462. 1992;55(suppl):1237S–1243S.
11 Surtees R, James L, Austin S: Association of 23 Troen AM: The central nervous system in
demyelination with deficiency of cerebro- animal models of hyperhomocysteinemia.
spinal fluid S-adenosylmethionine in inborn Prog Neuropsychopharmacol Biol Psychiatry
errors of methyl-transfer pathway. Lancet 2005;29:1140–1151.
1991;338:1550–1554. 24 Aisen PS, Schneider LS, Sano M, et al:
12 Freeman JM, Finkelstein JD, Mudd SH: High-dose B vitamin supplementation
Folate-responsive homocystinuria and and cognitive decline in Alzheimer dis-
schizophrenia. A defect in methylation ease: a randomized controlled trial. JAMA
due to deficient 5,10-methylenetetrahy- 2008;300:1774–1783.
drofolate reductase activity. N Engl J Med 25 Brady CB, Gaziano JM, Cxypoliski RA, et
1975;292:491–496. al: Homocysteine lowering and cognition
13 Haworth JC, Dilling LA, Surtees RA, et al: in CKD: the Veterans Affairs homocysteine
Symptomatic and asymptomatic methyl- study. Am J Kidney Dis 2009;54:440–449.
enetetrahydrofolate reductase deficiency 26 McMahon JA, Green TJ, Skeaff CM, et al: A
in two adult brothers. Am J Med Genet controlled trial of homocysteine lowering
1993;45:572–576. and cognitive performance. N Engl J Med
14 Marquet J, Chadefaux B, Bonnefont JP, et al: 2006;354:2764–2772.
Methylenetetrahydrofolate reductase defi- 27 Durga J, van Boxtel MP, Schouten EG, et al:
ciency: prenatal diagnosis and family studies. Effect of 3-year folic acid supplementation
Prenatal Diagn 1994;14:29–33. on cognitive function in older adults in the
15 Pasquier F, Lebert F, Petit H, et al: FACIT trial: a randomised, double blind,
Methylenetetrahydrofolate reductase defi- controlled trial. Lancet 2007;369:208–216.
ciency revealed by a neuropathy in a psy-
chotic adult. J Neurol Neurosurg Psychiatry
1994;57:765–766.
28 Smith AD, Smith SM, de Jager CA, et al: 39 Leeming RJ, Lucock M: Autism: is there
Homocysteine-lowering by B vitamins slows a folate connection? J Inherit Metab Dis
the rate of accelerated brain atrophy in mild 2009;32:400–402.
cognitive impairment: a randomized con- 40 Main PA, Angley MT, Thomas P, et al:
trolled trial. PLoS One 2010;5:e12244. Folate and methionine metabolism in
29 Heseker HB, Mason JB, Selhub J, et al: Not all autism: a systematic review. Am J Clin Nutr
cases of neural-tube defect can be prevented 2010;91:1598–1620.
by increasing the intake of folic acid. Br J 41 Miller JW, Garrod MG, Allen LH, et al:
Nutr 2009;102:173–180. Metabolic evidence of vitamin B-12 defi-
30 Botto LD, Yang Q: 5,10-Methylenetetra- ciency, including high homocysteine and
hydrofolate reductase gene variants and methylmalonic acid and low holotranscoba-
congenital anomalies: a HuGE review. Am J lamin, is more pronounced in older adults
Epidemiol 2000;151:862–877. with elevated plasma folate. Am J Clin Nutr
31 Choumenkovitch SF, Jacques PF, Nadeau 2009;90:1586–1592.
MR, et al: Folic acid fortification increases 42 Selhub J, Morris MS, Jacques PF, et al: Folate-
red blood cell folate concentrations in the vitamin B-12 interaction in relation to cogni-
Framingham study. J Nutr 2001;131:3277– tive impairment, anemia, and biochemical
3280. indicators of vitamin B-12 deficiency. Am J
32 Choumenkovitch SF, Selhub J, Wilson PW, Clin Nutr 2009;89:702S–706S.
et al: Folic acid intake from fortification in 43 Morris MS, Jacques PF, Rosenberg IH, et
United States exceeds predictions. J Nutr al: Folate and vitamin B-12 status in rela-
2002;132:2792–2798. tion to anemia, macrocytosis, and cognitive
33 Ray JG, Wyatt PR, Thompson MD, et al: impairment in older Americans in the age
Vitamin B12 and the risk of neural tube of folic acid fortification. Am J Clin Nutr
defects in a folic-acid-fortified population. 2007;85:193–200.
Epidemiology 2007;18:362–366. 44 Morris MS, Jacques PF, Rosenberg IH, et al:
34 Shaw GM, Finnell RH, Blom HJ, et al: Circulating unmetabolized folic acid and
Choline and risk of neural tube defects in 5-methyltetrahydrofolate in relation to ane-
a folate-fortified population. Epidemiology mia, macrocytosis, and cognitive test perfor-
2009;20:714–719. mance in American seniors. Am J Clin Nutr
35 DRI Dietary Reference Intakes for Thiamin, 2010;91:1733–1744.
Riboflavin, Niacin, Vitamin B6, Folate, 45 Morris MC, Evans DA, Bienias JL, et al:
Vitamin B12, Pantothenic Acid, Biotin, and Dietary folate and vitamin B12 intake and
Choline. Washington, National Academy cognitive decline among community-dwell-
Press, 2000. ing older persons. Arch Neurol 2005;62:641–
36 Mason JB, Dickstein A, Jacques PF, et al: 645.
A temporal association between folic acid 46 Troen AM, Mitchell B, Sorensen B, et al:
fortification and an increase in colorectal Unmetabolized folic acid in plasma is associ-
cancer rates may be illuminating important ated with reduced natural killer cell cytotox-
biological principles: a hypothesis. Cancer icity among postmenopausal women. J Nutr
Epidemiol Biomarkers Prev 2007;16:1325– 2006;136:189–194.
1329. 47 Arsenault JE, Mora-Plazas M, Forero Y, et al:
37 Oakley GP Jr: Inertia on folic acid fortifica- Hemoglobin concentration is inversely asso-
tion: public health malpractice. Teratology ciated with erythrocyte folate concentrations
2002;66:44–54. in Colombian school-age children, especially
38 Yang Q, Botto LD, Erickson JD, et al: among children with low vitamin B12 status.
Improvement in stroke mortality in Canada Eur J Clin Nutr 2009;63:842–849.
and the United States, 1990 to 2002.
Circulation 2006;113:1335–1343.
170 Troen
48 Yajnik CS, Deshpande SS, Jackson AA, et 54 Fagiolini M, Jensen CL, Champagne FA:
al: Vitamin B12 and folate concentrations Epigenetic influences on brain develop-
during pregnancy and insulin resistance in ment and plasticity. Curr Opin Neurobiol
the offspring: the Pune Maternal Nutrition 2009;19:207–212.
Study. Diabetologia 2008;51:29–38. 55 Crews D: Epigenetic modifications of brain
49 Veena SR, Krishnaveni GV, Srinivasan K, et and behavior: theory and practice. Horm
al: Higher maternal plasma folate but not Behav 2011;59:393–398.
vitamin B-12 concentrations during preg- 56 Day JJ, Sweatt JD: Cognitive neuroepige-
nancy are associated with better cognitive netics: a role for epigenetic mechanisms in
function scores in 9- to 10-year-old children learning and memory. Neurobiol Learn Mem
in South India. J Nutr 2010;140:1014–1022. 2011;96:2–12.
50 Bhate V, Deshpande S, Bhat D, et al: Vitamin 57 Zeisel SH: The fetal origins of memory:
B12 status of pregnant Indian women and the role of dietary choline in optimal brain
cognitive function in their 9-year-old chil- development. J Pediatr 2006;149:S131–S136.
dren. Food Nutr Bull 2008;29:249–254. 58 Iskandar BJ, Rizk E, Meier B, et al: Folate reg-
51 Eilander A, Muthayya S, van der Knaap ulation of axonal regeneration in the rodent
H, et al: Undernutrition, fatty acid and central nervous system through DNA methy-
micronutrient status in relation to cogni- lation. J Clin Invest 2010;120:1603–1616.
tive performance in Indian school chil- 59 Steegers-Theunissen RP, Obermann-Borst
dren: a cross-sectional study. Br J Nutr SA, Kremer D, et al: Periconceptional mater-
2010;103:1056–1064. nal folic acid use of 400 micro g per day is
52 Tamura T, Goldenberg RL, Chapman VR, et related to increased methylation of the IGF2
al: Folate status of mothers during pregnancy gene in the very young child. PLoS One
and mental and psychomotor develop- 2009;4:e7845.
ment of their children at five years of age. 60 Paul L, Cattaneo M, D’Angelo A, et al:
Pediatrics 2005;116:703–708. Telomere length in peripheral blood mono-
53 Bale TL, Baram TZ, Brown AS, et al: Early nuclear cells is associated with folate status in
life programming and neurodevelopmental men. J Nutr 2009;139:1273–1278.
disorders. Biol Psychiatry 2010;68:314–319.
Discussion on Folate and Vitamin B12

Importance in Cognitive Development
Dr. Troen has provided a contemporary insight into the ways in which folic acid
and vitamin B12, their interactions and deficiencies might impact upon cogni-
tive and brain development in the child and even at the other extremity of the
life cycle, older age. We need to remember that when we speak about folic acid,
we need to add our specificity, especially when we refer to the possibility of
adverse effects. Folic acid is a synthetic form of folate. The synthesis was accom-
plished in the late 1940s. The synthetic form of the folic monoglutamate is not
the same as folate in foods which tend to be reduced and tend also to be in the
conjugated form of polyglutamates. As we think about the effects of folic acid
administered as a supplement before or during pregnancy, we need to speak in
terms of whether there are special effects of synthetic folic acid, which has to
traverse a somewhat different metabolic pathway than so-called ‘natural’ folate.
That distinction is probably going to turn out to be important as we explore
further the matter of epigenetics and the way in which folic acid or other methyl
donors may be influencing gene and gene expression. Also the pathways leading
to methylation of the DNA genome are going to be a very interesting approach
to epigenetic regulation.
Regarding brain development, as with physical growth and development,
deficiency of almost any essential vitamin and mineral (as we have heard with
iodine and iron and other B vitamins, thiamine, niacin and vitamin B6) can
affect brain development and cause disease. The complex relationship between
nutritional status of micronutrients and brain development is before us. The
special consideration that was emphasized by Dr. Troen is the interaction of the
last two of the discovered vitamins. I remind you that the last two vitamins that
were discovered in the 1940s and 1950s were folic acid (which became vitamin
B9) and vitamin B12. (It turned out that vitamin B10 and B11 were both forms of
folic acid.) But the history is entrained in a very important way so that before
we had the identification of vitamin B12, while we already had synthetic folic
acid, folic acid was used in the treatment of pernicious anemia even before it
was known that pernicious anemia was vitamin B12 deficiency conditioned
by an absorption defect of intrinsic factor. The result was that many patients
were reported to have a worsening of their neurologic problems when they
were treated with high doses of folic acid before we learned that vitamin B12
was the therapeutic need. These observations had a large impact on the later
definition of the safe upper level of folic acid, which was defined in the DRI
(Dietary Reference Intake) report for the US Institute of Medicine in terms of
what would happen if you gave large doses of synthetic folic acid in the pres-
ence of limiting vitamin B12 status. That continues to be a challenge for us now,
and it reemphasizes the importance of updating our information about what
should be the safe levels of folic acid supplementation using synthetic folic acid.
I am not talking about any adverse effect of folate as food folate. Perhaps in the
next version of the DRIs we are going to have to look at a much more complex
interaction using some of the kinds of methodologic evaluations that Dr. Troen
presented (including epigenetic gene expression) to arrive at an understanding
of not only how folate and B12 influence brain development and cognitive devel-
opment but perhaps how their interaction might be important in the generation
and maintenance of health.
When Dr. Troen was asked whether he thinks that the closure of the devel-
oping neural tube in the presence of perhaps adequate amounts of folic acid
is a good model for us to be thinking about the effect of folic acid on brain
development, he replied that we still don’t understand the mechanism despite
the powerful effect of folic acid supplementation. There are clear trophic effects
of this metabolic pathway on different aspects of brain development at differ-
ent critical windows. Perhaps the problem we have with fortification is that we
undertook the public health initiative to prevent neural tube defects without
full knowledge of observations in populations that are exposed with respect to a
whole range of other important health outcomes and risks.
Dr. Haschke added two questions related to epigenetic phenomena which
Dr. Troen described, the first one is the Indian study on maternal B12 and cogni-
tive function at 9 years of age. We must be very careful because this is a highly
selective population in terms of nutrition. The low vitamin B12 cohort is prob-
ably a vegetarian cohort, and it might be that the children have also very low B12
intake as children from parents who are vegetarians. And the second phenom-
enon described is the periconceptional folic acid supplementation. The study by
Steeger seems to show lower birthweight if the mothers are supplemented. This
was not the primary outcome of the study, and the sample size is much too small
to really show differences in birthweight. You need at least 80–100 per group
to rule out the type 2 error. If folic acid when given perinatally results in lower
birthweight, this potentially important finding has to be confirmed in adequate
studies.
Dr. Troen added that we don’t know that the Steeger’s study demonstrates
that there is an effect of periconceptional folic acid on birthweight. We are look-
ing at one gene. His point was that by virtue of our standard practice we can
Discussion on Folate and Vitamin B12 Importance in Cognitive Development 173

show a measurable effect on a measurable biological change that has a known
consequence in terms of expression of the gene product and downstream effect.
But that’s one gene out of our entire genome, the entire genome is exposed, the
whole panel of imprinted genes is exposed, we don’t know what this does over-
all. So, he was not unduly concerned, and he doesn’t think that we have reduced
birthweight across the folic acid-fortified population.
Irwin H. Rosenberg
174 Rosenberg
Pros and Cons of Increasing Folic Acid

and Vitamin B12 Intake by Fortification
Lindsay H. Allen
USDA, ARS Western Human Nutrition Research Center, University of California, Davis, CA, USA
Abstract
There is no doubt that folic acid fortification can be effective for reducing the incidence of
neural tube defects. The degree of efficacy depends on both the level of folate depletion
and other, yet to be fully characterized, genetic and/or environmental factors. This article
summarizes briefly data on neural tube defect reduction and other benefits of folic acid
fortification as these have been reviewed in more detail elsewhere. More attention is
drawn to questions that have been raised about the possible adverse effects of folic acid
fortification including the incidence of colorectal cancer and immune function. The main
question addressed here is whether folic acid fortification can exacerbate the adverse
effects of vitamin B12 deficiency. Most analyses of this question have been conducted in
wealthier countries based on data from elderly populations – which have the highest
prevalence of vitamin B12 deficiency. However, of potentially greater concern is the
increasingly common practice of folic acid fortification in developing countries, where
folate status is probably often adequate even prior to fortification, and vitamin B12 deple-
tion or deficiency is common. To add to this information, data from a group of Chilean
elderly with a range of vitamin B12 status and exposed to high levels of folic acid fortifica-
tion will be presented. Copyright © 2012 S. Karger AG, Basel
Introduction
Flour fortification with folic acid has the primary goal of preventing neural tube
defects (NTDs) by improving the periconceptional folate status of women at risk
of delivering an infant with these abnormalities. At least 52 countries are cur-
rently mandating flour fortification with folic acid [1], although none of them
are in Europe where there are stronger concerns about the safety of folic acid
fortification and/or its likely effectiveness in regions where industry is already
voluntarily fortifying foods, such as cereals, with the vitamin. After mandatory
or voluntary fortification was instigated in different locations, reports began to
appear of associations between the timing of fortification, and/or folate status,
and a range of positive and negative health outcomes. Mandatory fortification
with vitamin B12 is much less common, although the prevalence of deficiency is
probably much higher. There are numerous documented and potential adverse
effects of this nutrient deficiency that could justify fortification policies. Dual
fortification with both vitamins may make sense based on their metabolic inter-
actions and concern about the potential (as yet unproven) for folic acid fortifi-
cation to exacerbate signs and symptoms of vitamin B12 deficiency.
Folic Acid Fortification
Advantages
Benefits of folic acid fortification have been documented in a number of coun-
tries. For example, in the United States where fortification of flour became man-
datory in 1998, the incidence of NTDs has been reduced by 19–40% [2–4]. In
addition, the prevalence of folate deficiency has fallen from ≈20–25% to ≈1%
[5–7]. Hyperhomocysteinemia has been reduced from 17 to 9% [5], which has
been associated with reduction in mortality from stroke in the US and Canada
[8], although it is not clear that this has had any effect on risk of CVD [9]. High
plasma homocysteine in pregnant women is associated with an increased risk of
other adverse pregnancy outcomes [10], lower birthweight of the infant [11] and
in rats, impaired glucose regulation in the offspring [12], but these associations
have not been tested in randomized controlled trials. In Canada, the prevalence
of open NTDs fell by 48%, from 1.13 to 0.58 per 1,000 pregnancies within a few
years of the start of fortification [13]. There was also a significant, 6%/year fall
in the birth prevalence of severe congenital heart defects [14].
Potential for Adverse Effects

Concern has been raised about potential adverse effects of folic acid fortifica-
tion, as reviewed elsewhere [15, 16]. Folic acid is a substrate for the synthesis of
thymidine which is incorporated into DNA, and in folate deficiency there is less
thymidine available and more uracil is incorporated into DNA, which can cause
breaks in the DNA base sequence. In animal models, this increases the risk of
cancer initiation. However, after cancer has been initiated, there may actually
be a reduction in proliferation when folate is restricted, which is the basis of
treatment with anti-folate drugs such as methotrexate. Thus, fortification with
folic acid can theoretically prevent cancer initiation in depleted individuals but
stimulate proliferation of preexisting tumors. This hypothesis is supported by
an increase in colorectal cancer rates in the United States since folic acid addi-
tion to foods commenced [17] and a similar situation in Chile [18]. Although
176 Allen
improvements in ability to detect and diagnose colon cancer were ruled out by
the investigators as the explanation for the increase in colon cancer, it is clear
that more research is needed before causality can be proven. Moreover, increased
mortality from colon cancer has not been demonstrated.
Questions have also been raised about whether high folic acid intakes
reduce natural killer (NK) cell cytotoxicity, which is an important component
of immune function. Postmenopausal women in the US whose diet was low
in folate (<233 μg/day) had better NK cell cytotoxicity if they took folic acid
supplements, but those who had a folate-rich diet and in addition took >400 μg/
day as a supplement had poorer NK cell cytotoxicity [19]. This association was
supported by the cytotoxicity being inversely related to levels of unmetabolized
folic acid in plasma; the latter increase with folic acid intake.
Potential for High Folate Status to Exacerbate Vitamin B12 Deficiency

Evidence has been accumulating to suggest that high folic acid intakes might
exacerbate the adverse effects of vitamin B12 deficiency. Concerns were first
raised in the 1940s and 1950s about the possibility that folic acid supplementa-
tion might exacerbate both the hematological and the neurological symptoms
of vitamin B12 deficiency. These concerns arose during clinical examination of
patients with megaloblastic anemia who were treated with folic acid because they
were mistakenly diagnosed with folate deficiency, when in fact they were B12
deficient. The folic acid ameliorated the anemia for a few years but the neurologi-
cal symptoms of vitamin B12 deficiency were reportedly exacerbated or appeared
suddenly when folic acid supplementation was started [20, 21]. Higher doses of
folic acid, given for a longer time, were most likely to produce these effects, and
of 38 cases of vitamin B12 deficiency treated with <1 mg folic acid, only 6 had
neurological deterioration, and those cases had been treated longer [21]. Thus, 1
mg/day is generally accepted as being the upper limit for folic acid intake.
Morris et al. [22] analyzed data from the 1999–2002 National Health and
Nutrition Examination Survey (NHANES) to investigate whether individuals
with higher serum folate concentrations had poorer vitamin B12 status based
on biochemical markers. Values were available for 1,459 persons after exclud-
ing those with evidence of renal disease, alcoholism, history of stroke, or liver,
thyroid or coronary artery disease. Their analyses revealed that having both
vitamin B12 deficiency (serum B12 <148 pmol/l or serum methylmalonic acid,
MMA >210 μmol/l) and high serum folate (>59 nmol/l, the 80th percentile
of the population) increased the risk of both anemia and cognitive impair-
ment (assessed with the Digit Symbol Coding subtest of the Wechsler Adult
Intelligence Scale III, which assesses processing speed and memory). The risk
of anemia and cognitive impairment was around three times higher in the low-
B12, high-folate group compared to those with serum folate ≤59 nmol/l. On the
other hand, high serum folate protected against cognitive impairment in those
with adequate B12 status.
Pros and Cons of Folic Acid and Vitamin B12 Fortification 177
The same team later performed similar analyses using data from NHANES
conducted in 1991–1994 and from 1999 to 2002, this time adding values for
total homocysteine (tHcy) and MMA [23]. In that analysis, they found that
persons with vitamin B12 deficiency – defined as low serum vitamin B12 (<148
pmol/l) – in combination with high serum folate (>32.6 nmol/l) had signifi-
cantly higher plasma tHcy and serum MMA than those with B12 deficiency and
lower folate status. In the low serum B12 group, tHcy and MMA increased when
serum folate started to rise above 20 nmol/l. Subsequent analyses explored
additional relationships among data from the 1999–2002 NHANES [24]. Low
B12 status was defined as serum B12 <148 pmol/l or serum MMA >210 nmol/l.
Compared to subjects with plasma folate ≤59 nmol/l and normal B12 status,
the odds ratios of having anemia and impaired cognitive function respectively
were 2.1 and 1.7 with low B12 and normal folate, and 4.9 and 5.0 for low B12
and high folate (>59 nmol/l). With increasing serum folate, in participants with
serum B12 <148 pmol/l there was again a surprising increase in homocysteine
and MMA, whereas there was the opposite trend for those with serum B12 ≥148
pmol/l. These results suggest that the coenzyme functions of vitamin B12 are
more impaired at high serum folate concentrations in people with poor vitamin
B12 status.
In California, we found a similar pattern in elderly Latinos (participants
in the Sacramento Area Latino Study on Aging) [25]. In this analysis, vitamin
B12 deficiency was again defined as <148 pmol/l, while high serum folate was
defined as >45.3 μmol/l, a cutoff selected because it was the upper limit of accu-
racy for the assay. The group with B12 deficiency and high serum folate had a
significantly higher plasma tHcy and MMA, and lower holoTC and holoTC:B12
ratio, than groups with B12 deficiency and high folate, or adequate B12 status and
normal or high folate. However, neither cognitive function (3MSE and delayed
recall) nor depressive symptoms were different across the four groups. In con-
trast, low serum folate was associated with dementia and impaired cognitive
function in the same population [26].
In contrast to these reports, a study from Oxford, UK, found that older indi-
viduals with a holoTC concentration in the lowest tertile had a 1.8-fold higher
risk of cognitive impairment, while for the lower tertile of plasma folate, it was
1.55-fold higher [27]. However, there was no evidence that high folate status
(serum folate >30 or >60 nmol/l) increased the risk of anemia or cognitive
impairment.
Possible Explanations for the Observed Adverse Associations with Folate Status
For adverse effects of high folic acid intakes to be plausible, there needs to be
a plausible explanation for the observed associations. Several such explana-
tions have been offered. Certainly folate and vitamin B12 metabolism are closely
linked, and both 5-methyl-tretrahydrofolate and B12 are cofactors for the synthe-
sis of methionine. Thus, an abnormal folate cycle or vitamin B12 deficiency can
178 Allen
have similar effects on both hematopoiesis and the nervous system. In vitamin
B12 deficiency plasma folate increases because there is a block in the utilization of
methyl folate; this can lead to megaloblastic anemia and neurological problems.
In addition, the apparent associations between vitamin B12 deficiency and high
serum folate may to some extent be explained by this phenomenon rather than
high folate causing B12 depletion. Some investigators have hypothesized that the
increase in circulating folic acid, which rises progressively with intakes of folic
acid in supplements or fortification [28], may have adverse effects on immune
function or the rate of oxidation of forms of methionine synthase [23]. An alter-
native interpretation is that individuals with high serum folate are those taking
multivitamin supplements containing folic acid, in addition to fortified prod-
ucts [29]. Such supplements usually contain vitamin B12, so individuals with low
serum B12 but high folate may have problems absorbing vitamin B12, i.e. they
may have pernicious anemia, and ensuing long-term negative effects on hema-
tology or neurological function. Clearly, all of these mechanisms remain to be
confirmed, logically with the inclusion of both animal models and clinical trials.
Estimating the Appropriate Level of Folic Acid Addition to Flour

Because of the low occurrence of NTDs (0.8/1,000 in the US to 14/1,000
in China, for example), it has been pointed out that relatively few NTDs are
prevented relative to the number of people exposed to folic acid fortification
[15]. For example, it has been estimated that increasing folic acid intake in the
UK may prevent up to 162 NTDs/year in a population of 60 million people.
Increasing intake by 0.2 mg/day would prevent 49 NTDs per year in Australia
and 11 in New Zealand [30] out of a population of ≈27 million people. In the
UK, were flour to be fortified with 300 μg/100 g, 77–162 NTDs/year would be
prevented, but 370,000–780,000 people exposed to excess folic acid per NTD
prevented [15]. Thus, it is important to ensure that the level of exposure of the
general public to folic acid is not excessive as a result of fortification.
The underlying metabolic and genetic defects that cause NTDs are not com-
pletely understood, but it is apparent that low maternal folate status interacts
with genetic and environmental risk factors to increase risk [31, 32]. Thus, folic
acid fortification is more effective at reducing NTDs in population groups with
poor folate status. The folate status of populations varies widely, and tends to be
lower in industrialized countries than in poorer regions of the world where there
is a relatively high intake of green leafy vegetables and legumes, which are good
sources of the vitamin [33]. Countries such as Chile, the United States, Canada,
and China had a low intake and generally poor folate status before fortification,
and thus would be expected to benefit more from fortification. Moreover, it is
clear that the reduction in NTDs after fortification was greatest in areas with the
highest rates of NTD and poorer folate status (e.g. Newfoundland compared to
Quebec and Manitoba in Canada, Northern China compared to the South, New
England compared to the rest of the US) [34]. The question then arises about
whether it is reasonable to expect similar rates of NTD reduction in developing
countries, where folate status might be adequate before fortification. For exam-
ple, it has been estimated that serum folate needs to be 16 nmol/l for maximum
prevention of NTDs, and that this can be achieved with a folic acid intake of 279
μg/day, assuming no other dietary intake [35]. It is rational to obtain data on
folate status and intake in populations with a low intake of animal source foods
(ASF; and/or vitamin B12 deficiency) prior to assuming that folic acid fortifica-
tion will have sufficient benefit to outweigh any potential harm.
Vitamin B12 Fortification
There are reasons why vitamin B12 fortification should be considered, with or
without folic acid fortification, especially for the elderly and in populations with
a low intake of ASF.
Prevalence of Deficiency
It is now apparent that the prevalence of vitamin B12 deficiency (serum B12
<148 pmol/l or <200 pg/ml) and depletion (serum B12 ≥148 to <221 pmol/l) is
high in people of all ages in populations with a low intake of ASF, since these
are the only natural source of the vitamin. Based on 2005 data from the World
Health Organization, the prevalence of B12 deficiency in adults was highest in
India (50%) and was about 5–30% in most other countries [36]. The prevalence
of vitamin B12 depletion is likely to be at least as high as that of deficiency; in
studies in Latin America, for example, prevalence of serum B12 <148 pmol/l is
about 15–20%, while over 20% are depleted. This prevalence also applies to a
nationally representative sample in Mexico [37]. In all developing country stud-
ies where prevalence was reported, serum vitamin B12 concentrations were cor-
related with intake of the vitamin, or intake of ASF.
Vitamin B12 deficiency has several well-documented adverse effects and is
associated with many others. The severe deficiency that occurs in pernicious
anemia, strict vegetarianism, and occasionally in elderly with malabsorption of
the vitamin from food due to gastric atrophy can result in megaloblastic anemia.
However, populations chronically deficient or depleted in the vitamin due to
low ASF intake are much less likely to suffer from anemia as a consequence [38],
nor does supplementation with B12 affect any hematological symptoms. Vitamin
B12 status was a predictor of NTDs in Ireland where flour is not fortified with
folic acid [39], and Canada after folic acid fortification of flour was introduced
[40]. Neurological disorders including loss of memory, cognitive impairment
[41] and rate of brain volume loss with aging [42, 43] and the severity of white
matter lesions [44], and depression [Jones et al., unpubl. data] are all associ-
ated with poor vitamin B12 status. The results of supplementation trials are not
yet definitive for vitamin B12 alone, but homocysteine lowering as a result of
180 Allen
supplementation with vitamin B12 (0.5 mg/day), vitamin B6 and folic acid sig-
nificantly reduced the rate of brain volume loss with aging [44]. A potentially
major but poorly documented problem is the low concentration of the vitamin
in breast milk produced by mothers who are deficient or depleted in the vita-
min [Deegan et al., unpubl. data], which is a concern since infants breastfed by
deficient mothers have an increased risk of developmental delays, which may be
permanent [45], including delayed motor development [46]. Poor vitamin B12
status has also been implicated in bone loss [47] possibly as a result of hyper-
homocysteinemia, although randomized controlled supplementation trials are
needed for confirmation.
Vitamin B12 Fortification

Based on the high prevalence of vitamin B12 deficiency and depletion, and the
confirmed and potential adverse effects of deficiency, it has been recommended
that flour be fortified with 2 μg vitamin B12/100 g flour in the form of cyanoco-
balamin in developing countries [48]. This level of addition is restricted by the
relatively high cost of the vitamin; in fortification, it is recommended that the
total cost of fortificants should not exceed 2% of the total cost of the product.
Unfortunately, there are no large-scale fortification trials of the efficacy of this
dose. We have conducted a small-scale pilot study of the bioavailability of the
labeled vitamin from fortified bread rolls given to healthy adult volunteers, and
found absorption from a 0.8 μg dose to be approximately 50%. The efficiency
of absorption of the vitamin decreases substantially as intake increases, to about
30% from 5 μg and 5% with 20 μg; above about 25 μg only 1% will be absorbed,
by passive transport [49]. There are no known adverse effects of high intakes of
vitamin B12 in supplements, or when taken intramuscularly.
Thus, it may be prudent policy to include both folic acid, at a reasonable
level, and vitamin B12 in flour fortification programs for developing countries.
Many elderly are also likely to benefit in countries where ASF intake is higher,
since vitamin B12 deficiency is more common in this group presumably due to
gastric atrophy and poor absorption of the vitamin from food. Clearly, trials of
dual fortification should be conducted.
References
1 Flour Fortification Initiative. http://www.sph. 3 Williams LJ, Mai CT, Edmonds LD, et al:
emory.edu/wheatflour/index.php. 2010. Prevalence of spina bifida and anencephaly
2 Honein MA, Paulozzi LJ, Mathews TJ, et al: during the transition to mandatory folic acid
Impact of folic acid fortification of the US fortification in the United States. Teratology
food supply on the occurrence of neural tube 2002;66:33–39.
defects. JAMA 2001;285:2981–2986.
4 Centers for Disease Control and Prevention. 15 Smith AD, Kim YI, Refsum H: Is folic acid
Trends in spina bifida, United States, 1991– good for everyone? Am J Clin Nutr 2008;
2005. http://www.cdc.gov/Features/dsSpin- 87:517–533.
aBifida. 2010. 16 Smith AD: Folic acid fortification: the good,
5 Jacques PF, Selhub J, Bostom AG, et al: The the bad, and the puzzle of vitamin B-12. Am
effect of folic acid fortification on plasma J Clin Nutr 2007;85:3–5.
folate and total homocysteine concentra- 17 Mason JB, Dickstein A, Jacques PF, et al: A
tions. N Engl J Med 1999;340:1449–1454. temporal association between folic acid forti-
6 Dietrich M, Brown CJ, Block G: The effect of fication and an increase in colorectal cancer
folate fortification of cereal-grain products rates may be illuminating important biologi-
on blood folate status, dietary folate intake, cal principles: a hypothesis. Cancer Epidemiol
and dietary folate sources among adult non- Biomarkers Prev 2007;16:1325–1329.
supplement users in the United States. J Am 18 Hirsch S, Sanchez H, Albala C, et al: Colon
Coll Nutr 2005;24:266–274. cancer in Chile before and after the start
7 Pfeiffer CM, Johnson CL, Jain RB, et al: of the flour fortification program with
Trends in blood folate and vitamin B-12 con- folic acid. Eur J Gastroenterol Hepatol
centrations in the United States, 1988–2004. 2009;21:436–439.
Am J Clin Nutr 2007;86:718–727. 19 Troen AM, Mitchell B, Sorensen B, et al:
8 Yang Q, Botto LD, Erickson JD, et al: Unmetabolized folic acid in plasma is associ-
Improvement in stroke mortality in Canada ated with reduced natural killer cell cytotox-
and the United States, 1990 to 2002. icity among postmenopausal women. J Nutr
Circulation 2006;113:1335–1343. 2006;136:189–194.
9 Anderson JL, Jensen KR, Carlquist JF, et al: 20 Reynolds E: Vitamin B12, folic acid, and the
Effect of folic acid fortification of food on nervous system. Lancet Neurol 2006;5:949–
homocysteine-related mortality. Am J Med 960.
2004;116:158–164. 21 Savage DG, Lindenbaum J: Neurological
10 Vollset SE, Refsum H, Irgens LM, et al: complications of acquired cobalamin
Plasma total homocysteine, pregnancy com- deficiency: clinical aspects. Baillieres Clin
plications, and adverse pregnancy outcomes: Haematol 1995;8:657–678.
the Hordaland Homocysteine Study. Am J 22 Morris MS, Jacques PF, Rosenberg IH,
Clin Nutr 2000;71:962–968. Selhub J: Folate and vitamin B-12 status in
11 Murphy MM, Fernandez-Ballart JD: relation to anemia, macrocytosis, and cogni-
Homocysteine in pregnancy. Adv Clin Chem tive impairment in older Americans in the
2011;53:105–137. age of folic acid fortification. Am J Clin Nutr
12 Maloney CA, Hay SM, Young LE, et al: A 2007;85:193–200.
methyl-deficient diet fed to rat dams during 23 Selhub J, Morris MS, Jacques PF: In vitamin
the peri-conception period programs glucose B12 deficiency, higher serum folate is associ-
homeostasis in adult male but not female ated with increased total homocysteine and
offspring. J Nutr 2011;141:95–100. methylmalonic acid concentrations. Proc
13 Ray JG, Meier C, Vermeulen MJ, et al: Natl Acad Sci USA 2007;104:19995–20000.
Association of neural tube defects and folic 24 Selhub J, Morris MS, Jacques PF, Rosenberg
acid food fortification in Canada. Lancet IH: Folate-vitamin B-12 interaction in rela-
2002;360:2047–2048. tion to cognitive impairment, anemia, and
14 Ionescu-Ittu R, Marelli AJ, Mackie AS, Pilote biochemical indicators of vitamin B-12 defi-
L: Prevalence of severe congenital heart ciency. Am J Clin Nutr 2009;89:702S–706S.
disease after folic acid fortification of grain
products: time trend analysis in Quebec,
Canada. BMJ 2009;338:b1673.
182 Allen
25 Miller JW, Garrod MG, Allen LH, et al: 37 Allen LH: Folate and vitamin B12 status in
Metabolic evidence of vitamin B-12 defi- the Americas. Nutr Rev 2004;62:S29–S33,
ciency, including high homocysteine and discussion S4.
methylmalonic acid and low holotranscoba- 38 Metz J: A high prevalence of biochemical evi-
lamin, is more pronounced in older adults dence of vitamin B12 or folate deficiency does
with elevated plasma folate. Am J Clin Nutr not translate into a comparable prevalence of
2009;90:1586–1592. anemia. Food Nutr Bull 2008;29:S74–S85.
26 Ramos MI, Allen LH, Mungas DM, et al: 39 Molloy AM, Kirke PN, et al: Maternal vitamin
Low folate status is associated with impaired B12 status and risk of neural tube defects in
cognitive function and dementia in the a population with high neural tube defect
Sacramento Area Latino Study on Aging. Am prevalence and no folic acid fortification.
J Clin Nutr 2005;82:1346–1352. Pediatrics 2009;123:917–923.
27 Clarke R, Sherliker P, Hin H, et al: Folate and 40 Ray JG, Wyatt PR, Thompson MD, et al:
vitamin B12 status in relation to cognitive Vitamin B12 and the risk of neural tube
impairment and anaemia in the setting of defects in a folic-acid-fortified population.
voluntary fortification in the UK. Br J Nutr Epidemiology 2007;18:362–366.
2008;100:1054–1059. 41 Smith AD, Refsum H: Vitamin B-12 and
28 Sweeney MR, Staines A, Daly L, et al: cognition in the elderly. Am J Clin Nutr
Persistent circulating unmetabolised folic 2009;89:707S–711S.
acid in a setting of liberal voluntary folic acid 42 Vogiatzoglou A, Refsum H, Johnston C,
fortification. Implications for further manda- et al: Vitamin B12 status and rate of brain
tory fortification? BMC Public Health 2009; volume loss in community-dwelling elderly.
9:295. Neurology 2008;71:826–832.
29 Yeung LY, Yang Q, Berry RJ: Contributions of 43 Smith AD, Smith SM, de Jager CA, et al:
total daily intake of folic acid to serum folate Homocysteine-lowering by B vitamins slows
concentrations. JAMA 2008;300:2486–2487. the rate of accelerated brain atrophy in mild
30 Bower C, de Klerk N, Hickling S, et al: cognitive impairment: a randomized con-
Assessment of the potential effect of incre- trolled trial. PLoS One 2010;5:e12244.
mental increases in folic acid intake on 44 de Lau LM, Smith AD, Refsum H, et al:
neural tube defects in Australia and New Plasma vitamin B12 status and cerebral white-
Zealand. Aust N Z J Public Health 2006; matter lesions. J Neurol Neurosurg Psychiatry
30:369–374. 2009;80:149–157.
31 Beaudin AE, Stover PJ: Insights into metabolic 45 Dror DK, Allen LH: Effect of vitamin B12
mechanisms underlying folate-responsive neu- deficiency on neurodevelopment in infants:
ral tube defects: a minireview. Birth Defects current knowledge and possible mechanisms.
Res A Clin Mol Teratol 2009;85:274–284. Nutr Rev 2008;66:250–255.
32 Beaudin AE, Abarinov EV, Noden DM, et al: 46 Jones KM, Ramirez-Zea M, Zuleta C, Allen
Shmt1 and de novo thymidylate biosynthesis LH: Prevalent vitamin B-12 deficiency in
underlie folate-responsive neural tube defects twelve-month-old Guatemalan infants is pre-
in mice. Am J Clin Nutr 2011;93:789–798. dicted by maternal B-12 deficiency and infant
33 Allen LH: Causes of vitamin B12 and folate diet. J Nutr 2007;137:1307–1313.
deficiency. Food Nutr Bull 2008;29:S20–S34, 47 Dhonukshe-Rutten RA, Pluijm SM, de Groot
discussion S5–S7. LC, et al: Homocysteine and vitamin B12
34 Heseker HB, Mason JB, Selhub J, et al: Not all status relate to bone turnover markers, broad-
cases of neural-tube defect can be prevented band ultrasound attenuation, and fractures
by increasing the intake of folic acid. Br J in healthy elderly people. J Bone Miner Res
Nutr 2009;102:173–180. 2005;20:921–929.
35 Dary O: Nutritional interpretation of folic 48 Allen LH, Rosenberg IH, Oakley GP, Omenn
acid interventions. Nutr Rev 2009;67:235–244. GS: Considering the case for vitamin
36 McLean E, de Benoist B, Allen LH: Review B12 fortification of flour. Food Nutr Bull
of the magnitude of folate and vitamin B12 2010;31:S36–S46.
deficiencies worldwide. Food Nutr Bull 49 Allen LH: How common is vitamin B-12 defi-
2008;29:S38–S51. ciency? Am J Clin Nutr 2009;89:693S–696S.
Discussion on Vitamin B12 and

Folic Acid Fortification
Forty years after the discovery of vitamin B12, there is still much work left to
be done. This connects also with Dr. Troen’s remarks on folic acid. The power-
ful effects of the randomized controlled trial that led to the decision in some
countries to fortify flour with folic acid in order to prevent neural tube defects
present additional challenges for interpretation. The message of those trials was
not just that there needs to be better folic acid status in the preconceptional
period in order to limit, even if not to altogether prevent, the incidence of neu-
ral tube defects. Another message of those trials is that we need to be much
more concerned about the entire nutritional status of women in the periconcep-
tional period. This isn’t just about folic acid, nor is it even just about those other
micronutrients which may be involved in prevention of neural tube defects. We
have heard throughout these few days that the maternal nutritional status before
as well as during pregnancy and even after pregnancy is a very important target
with really substantial public health implications much beyond the prevention
of neural tube defects. Perhaps this returns us to the observation that was made
by Dr. Jooste. I will paraphrase his story about ‘think globally and act locally’ to
say we need to think critically globally and act locally upon a much better nutri-
tional database than we currently have. The notion to me of a flour fortification
initiative that says that we need to use data from a limited number of countries
to make public health decisions about fortification or even levels of fortification
throughout the world, without knowing nearly enough about the status of either
folate or B12 or other factors, strikes me as not thinking critically globally. As Dr.
Allen states, we have a need for a lot more information about B12 status around
the world, and about folate status around the world, not just to consider the
incidence of neural tube defects.
When asked to comment further on what we know about the efficiency of
absorption of either folate or B12 from breast milk (since one mistake of the
DRIs was to suggest that the absorption of breast milk folate was less efficient
than the absorption of synthetic folic acid), Dr. Allen stated that it’s clear that
early in infancy the mechanism of absorption is different from later in infancy.
Vitamin B12 is probably absorbed bound to haptocorrin early in infancy and
then later there is enough intrinsic factor and peptic digestion for B12 to be
absorbed as free B12. B12 is bound incredibly tightly to haptocorrin in milk,
which is why the assay has been difficult. Interestingly, Dr. Lönnerdal showed
that the haptocorrin reduces the growth of Escherichia coli in the intestine, so
that’s another reason why it’s also probably really important. She didn’t know
what percent is broken down and absorbed. Because the assays were wrong
formerly, we underestimated requirements of those infants and breastfeeding
women. The values are going to be double. There is a big study going on in
Denmark where they don’t have supplementation of women in pregnancy and
lactation to get normal values for breast milk vitamin B12.
In response to Dr. Kalantari’s comment that the B12 content of breast milk is
dependent on the daily diet of the mothers, Dr. Allen noted that one interesting
thing about B12 is that the efficiency of absorption falls off quickly with increas-
ing doses, even in the usual range of intakes. For example, she has just done a
study with eggs labeled with isotopic B12, and you get half the absorption from
two eggs as you do from one egg. So, it appears not to make any difference
whether you eat one or two eggs. Even in the low range of intake, if you give a
dose in one bolus, then you have poor efficiency of absorption, and so that’s what
we are doing when we give these high-dose supplements to lactating women.
That’s why when treating B12 deficiency, physicians give high doses, and that’s
because only 1% of those high doses are absorbed compared to 50% of really
low doses. So, you really need a system with B12, where you are delivering it in
small amounts during the day to have an effect. And then the other unknown is
what comes from the mother’s liver. Clearly, mother’s status is related to infant’s
status and infant’s liver content even 2 years after they were born. Perhaps some
of the B12 in breast milk comes from stores and not from current intake. We are
not certain about that with practically any nutrient, how much comes currently
from the mother’s diet versus from stores that she already has. But she thinks the
efficiency of absorption is the issue here.
Dr. Allen added one really important comment about folate and B12 that the
folate concentration in breast milk is not much affected by maternal status. What
happens is that the mother gets more and more depleted as lactation goes on if
she is not replacing it from the diet, and the breast milk levels change hardly at
all, so you don’t get infant depletion as a result of low breast milk levels of folate,
nor does high folic acid intake affect breast milk levels very much.
Dr. Penny asked whether Dr. Allen could give some idea about how much
animal source food is needed and for how long in terms of amount for women
and for children.
Dr. Allen replied that in Kenya we fed sort of the RDA as animal source
foods as milk or meat to schoolers of whom about 80% were deficient or
depleted. After a year, we had a substantial improvement, after 2 years the severe
Discussion on Vitamin B12 and Folic Acid Fortification 185

deficiency went away, but we still had a lot of depletion and even deficient lev-
els. In Guatemala, what I didn’t show were the results of this intervention study
where we gave meat to these kids or the RDA as supplements, and that had
absolutely no effect on their serum levels of B12 over one meal a day feeding 70
g of beef or the equivalent of about a microgram of B12. I don’t think it’s malab-
sorption because those that were drinking cow’s milk were doing just fine, and
the cow’s milk had much higher levels of B12 in it than we were giving. So, it’s
hard to replete a B12-deficient person once he/she is depleted, and we need to
understand much more about that.
Irwin H. Rosenberg
186 Rosenberg
Summary on Iodine Folate and

Vitamin B12
Perhaps folic acid and vitamin B12 have been orphan micronutrients for too
long. Even though they are the last of the vitamins to have been identified, I
think we have had over 60 years to gather more information about the status of
these vitamins and their importance, not only in nutritional health but in child
development. This challenge was put forward both by Lindsay Allen and Aron
Troen. Aron Troen has shown us a number of examples of the significant extent
to which our understanding of the biologic importance of these nutrients have
advanced. The advances that we heard from Aron Troen on our understanding of
the biology of folic acid and folic acid and vitamin B12 interactions give us a very
strong platform for application of knowledge about vitamin status to address
some of the important current public health considerations. Lindsay Allen indi-
cated there are more than 52 countries that are fortifying flour with folic acid.
This is a remarkable demonstration of thinking globally but not thinking criti-
cally. As there is increasing information about those processes in human biol-
ogy which are influenced by folic acid through methylation and epigenetics, we
must be even more cognizant of local information that helps us make decisions
about fortification and about interactions. Arguments put forward by Lindsay
Allen suggesting that we have inadequate database about vitamin B12 in a num-
ber of countries should raise the question about whether this is a micronutrient
that deserves to be increased in the public health priority as a way of promoting
health and approaching prevention of deficiency, and whether in fact fortifica-
tion with vitamin B12 does have as strong a scientific basis as the fortification
with folic acid.
The matter of the interaction of the two vitamins has become more fascinat-
ing and current. These two nutrients were intertwined in their early history,
and now we realize that we have these national experiments of fortification with
folic acid (driven by experiences in North America with control of neural tube
defects) in countries which did not even have information about the folic acid
status, vitamin B12 status, or even neural tube birth incidence. We have now
the challenge of deciding to obtain information that will allow us to judge the
relative importance and relative risk of fortifying with folic acid in order to pre-
vent neural tube defects in countries with minimal reporting data. While neural
tube defect control can be demonstrated to be an important goal of folic acid
fortification, the question remains, what is the exposure with regard to safety
and interaction of the vitamins? We are talking about synthetic folic acid, not
about food folate. We are now able to measure the circulating levels of unme-
tabolized folic acid and analyze some of these observations in terms of whether
we are looking at the effects of folic acid per se or total folate intake in relation
to vitamin B12. The approaches that were presented in this session allow us to
begin to look at these interactions in a way which is much more analytical in use
of growing information about the fact that we actually can measure the status
of these nutrients. This should help us put at a higher priority level the consid-
eration of both decisions about folic acid and vitamin B12, and that can be done
with an increasing foundation of a solid and growing biological database.
Let me return again to the notion put forward by Pieter Jooste of ‘thinking
globally but acting locally’. This is very much a challenge with many if not all the
micronutrients that we have considered throughout these few days. Certainly it’s
true of the micronutrients that we considered today with regard to iodine, folate
and vitamin B12, and if we are to make our decisions about how to target our
public health interventions, we are going to have to upgrade the level of informa-
tion that we have in individual countries. We should not take as a global impera-
tive or an initiative which is global, arguments for doing universal fortification.
Sometimes that would be expedient when there may not be enough information
from other sources. Lindsay Allen will remember and as will Richard Hurrell, the
WHO consultation on Fortification [1] whose recommendations were clear that
the decisions about fortification for any of these nutrients should be based to
the extent possible on local information and public health priorities. Sometimes
that local information is actually going to be regional. The points that Michael
Zimmerman made in his discussion about iodine nutrition focus on ways in
which we might improve the information on the infant or the neonatal child
by using some new techniques of measurement such as steroid-stimulating
hormones. There are ways forward with regard to being able to not only think
locally on the basis of local information but to be able to actually define the tar-
get populations and whether the intervention that we have in mind is well suited
to those target populations. Where a universal approach is not adequate to reach
all the target populations, then we are clearly challenged to find ways to improve
on the impact of whatever interventions we can do.
Both of the sessions this morning should challenge us to be much more cog-
nizant of the nutritional status, nutritional intake, not just of women in preg-
nancy, not just women who are breastfeeding and not even perhaps just in the
preconceptional period and women of childbearing age, but also that we heard a
great deal of support for the importance of targeting even the adolescent women
who in many countries will soon become mothers. Their nutritional status will
188 Rosenberg
have a great deal of influence not only on birth outcomes but even on outcomes
that can be projected to even later times in the life cycle. Thus, by focusing on
a number of these nutrients such as we did in this session and throughout this
meeting, we can mount some very strong arguments for targeting our interven-
tion to the populations which are the important targets. I would hope that we
can use some of this kind of information and this discussion, to which all have
contributed so well, to have this publication lead to a list of important targets
and recommendations that could have an influence on how we go forward with
health and development programs.
Irwin H. Rosenberg
Reference
1 Allen L, de Benoist B, Dary O, Hurrell
R: Guidelines on Food Fortification with
Micronutrients. Geneva, World Health
Organization and Food and Agricultural
Organization of the United Nations, 2006.
Summary on Iodine Folate and Vitamin B12 189

Subject Index
Absorption Diphtheria-tetanus-pertussis (DPT),

breast milk vitamins vitamin A supplementation effects on
folic acid 184, 185 response 98, 99
vitamin B12 184, 185
iron Epigenetics, programming of
increase and adverse effects 122, neurocognitive development 167, 168,
123, 132, 133, 135 173
malaria studies 111, 112
obesity effects 112–114, 129, Flour fortification, see Folic acid;
130 Vitamin B12
overview 108, 109 Folic acid
zinc bioavailability 33, 34, 45 absorption from breast milk 184,
185
Bacille Calmette-Guerin (BCG), deficiency
vitamin A supplementation effects neurological consequences 162, 163
on response 97, 98 subclinical deficiency in central
Birthweight, see Stunting nervous system development
Breast milk and aging 164, 165
folic acid absorption 184, 185 epigenetic programming of
iodine composition 139, 140 neurocognitive development 167,
vitamin B12 absorption 184, 168, 173
185 flour fortification
advantages 176
Calcium, stunting studies 14 adverse effects
Cobalamin, see Vitamin B12 overview 176, 177
Complementary food supplements, vitamin B12 deficiency 177–179
stunting studies 17 estimation of levels 179, 180
functions 161, 162
Diarrhea supplementation
vitamin A supplementation cognitive development studies 165,
studies 95, 104 166
zinc supplementation studies 38, 39, malaria studies 131
43–46, 76 safety 173
191
Growth, see Stunting functions 8
infection effects on efficacy of fortified
Hepcidin, iron status marker 111–113, foods 110–112
115, 134 metabolism
Human immunodeficiency virus (HIV) absorption 108, 109
micronutrient deficiencies in infection and inflammation
pregnancy effects 109–112, 128, 129
overview 54, 57 obesity effects on absorption 112–
supplementation 66 114, 129, 130
vitamin A supplementation and overview 107, 108
maternal transmission supplementation
effects 81–84 malaria risks and benefits
zinc deficiency 43, 44 absorption increase and adverse
effects 122, 123, 132, 133,
Impaired health 22 135
Inflammation, iron absorption Cochrane review 121
effects 109–112, 128, 129 gastrointestinal tract levels and
Iodine adverse effects 123, 124
breast milk 139, 140 immune response and adverse
deficiency effects 125
disorders throughout life overview 118
cycle 149–151 Pemba substudy 119, 120
infant monitoring 138, 139 Pemba trial 119, 131
mortality and disease burden 64 technical working group
scope of problem 6, 7, 137, 138 review 121, 122
dietary sources 152, 153 WHO Consultation
excess intake 154 recommendations 120, 121
food fortification 157 overview 8
functions 147, 148 stunting studies 14
infant formula 139, 140 zinc interactions 47, 75, 76
intake recommendations 153, 154
iron deficiency effects on status 143 Lipid nutrient supplements (LNS),
metabolism 148, 149 stunting studies 17
salt iodization Lower respiratory tract infection, vitamin
barriers 156, 157 A supplementation studies 96
overview 7, 8
Switzerland 140–144 Malaria
universal salt iodization 155 iron absorption effects 111, 112
supplementation iron supplementation risks and
delivery 157, 158 benefits
recommendations 158 absorption increase and adverse
thyroid hormone production 148, 149 effects 122, 123
urine monitoring 140, 141 Cochrane review 121
Iron gastrointestinal tract levels and
deficiency adverse effects 123, 124
iodine status effects 143 immune response and adverse
mortality and disease burden 64 effects 125
192 Subject Index

overview 118 Pregnancy
Pemba substudy 119, 120 micronutrient deficiencies
Pemba trial 119, 131 global burden 51, 52
technical working group overview 50
review 121, 122 significance 52, 53
WHO Consultation multiple micronutrient
recommendations 120, 121 supplementation
zinc supplementation studies 39, 40 birthweight effects 55, 56, 64–66
Measles, vitamin A supplementation delivery 67, 68
pneumonia studies 97, 134 developmental effects 56
vaccine response studies 98 human immunodeficiency virus
Methylentetrahydrofolate reductase studies 66
(MTHFR), gene defects 163, 164 maternal impact 53, 54
Micronutrient powders, stunting neonatal impact 54, 55
studies 18 side effects 68–70
Milk, fortification for stunting 18, 25 significance for interventions 57
Mortality vitamin A supplementation 81
vitamin A supplementation impact
all-cause mortality impact 85 Reference nutrient intake (RNI),
cause-specific mortality and micronutrients in children 62
morbidity 86
children older than six months 92, Salt iodization, see Iodine
93 Stunting
neonates 93–95 milk fortification 18, 25
one-to-five months of age 93 moderate malnutrition 15, 16
zinc supplementation impact in multiple micronutrients
children 40 birthweight effects 55, 56, 64–66
Multiple micronutrients, see Pregnancy; complementary food
Stunting supplements 17
delivery 16, 17, 67, 68
Neural tube defect (NTD), folic acid micronutrient powders 18
deficiency 164, 165 side effects 68–70
flour fortification advantages 176, supplement studies 14, 15, 66,
179, 180 67
Non-transferrin-bound iron (NTBI), normal growth patterns 11, 12
adverse effects 122, 132, 133, 135 overview of dietary factors 12, 13
prospects for study 18, 19
Obesity, iron absorption effects 112–114, zinc supplementation 13, 37, 38, 44
129, 130
Thyroid hormone, production 148, 149
Phytate, zinc bioavailability effects 33, 34 Tuberculosis, see Bacille Calmette-Guerin
Pneumonia
measles pneumonia studies of Vitamin A
vitamin A supplementation 97, 134 deficiency
zinc supplementation studies 39 epidemiology in children 80
Polio, vitamin A supplementation effects mortality and disease burden 64
on vaccine response 99 scope of problem 2, 3, 91, 92, 103
Subject Index 193

supplementation epigenetic programming of
children older than six months 92, neurocognitive development 167,
93 168
diarrhea studies 95, 104 flour fortification 181
infants up to six months 84, 85, 93 functions 161, 162
lower respiratory tract infection supplementation and cognitive
studies 96 development studies 165, 166
maternal human Vitamin D
immunodeficiency virus deficiency 24, 25
transmission effects 81–84 vitamin A supplementation and
measles pneumonia studies 97 antagonism 99
neonates 84, 93–95, 104
pregnancy 81 Zinc
programs 3, 4, 87 bioavailability 33, 34, 45
recommendations 3, 92, 105 deficiency
six-to-59-months of age mortality and disease burden 64
all-cause mortality impact 85 scope of problem 4
cause-specific mortality and homeostasis 28
morbidity 86 iron supplementation interactions 47,
stunting studies 14 75, 76
vaccine response studies requirements
bacille Calmette-Guerin 97, 98 estimated average
diphtheria-tetanus-pertussis 98, requirements 31–33
99 infants and children 34
measles 98 physiological requirements 29, 30
polio 99 reconciliation of estimates 45
vitamin D antagonism 99 stable isotope technique for status
Vitamin B12 evaluation 45
absorption from breast milk 184, 185 supplementation
deficiency diarrhea 38, 39, 43–46, 76
folic acid induction 177–179 food fortification versus
neurological consequences 162, supplements 47, 48
163 malaria 39, 40
prevalence 180, 181 mortality impact 40
subclinical deficiency in central outcomes 5
nervous system development pneumonia 39
and aging 164, 165 programs 6
dietary sources 180, 185, 186 stunting studies 13, 37, 38, 44
194 Subject Index

Nutrition
Pediatric
Nestlé Nutrition Institute Workshop Series
69 Sports Nutrition: More Than Just Calories – Triggers for Adaptation

Editors: R.J. Maughan, Loughborough; L.M. Burke, Bruce
XIV + 154 p., 14 fig., 2 tab., hard cover, 2011
ISBN 978–3–8055–9697–8
70 Meeting Micronutrient Requirements for Health and Development

Editors: Z.A. Bhutta, Karachi; R.F. Hurrell, Zurich; I.H. Rosenberg, Boston, MA
XVIII + 194 p., 24 fig., 16 tab., hard cover, 2012
ISBN 978–3–318–02111–0
Meeting macro- and micronutrient requirements during pregnancy and early childhood is
crucial for short- and long-term health and cognitive function. Meta-analyses confirm that sup-
plementation or fortification of food with the ‘big four’ (vitamin A, iron, zinc, and iodine) is
efficacious to reduce the risk of infectious disease and improves growth and cognitive outcome.
More recently, folate and vitamin B12 deficiencies during pregnancy have been shown to be
associated with poor neurodevelopmental outcome and childhood obesity.
The papers collected in the book at hand address the fact that maternal and fetal deficiencies
can induce inadequate metabolic programming in the offspring, with increased risk for non-
communicable diseases later in life. World-renowned experts in the fields of epidemiology and
nutritional intervention met with those in genetics, epigenetics, and metabolic outcome to
clarify the pathogenesis of micronutrient deficiencies in pregnancy and childhood, preventive
methods and strategies, and opportunities for treatment.
CV4482

Complete Micronutrient Nestle

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Complete Micronutrient Nestle

Transféré par

Droits d'auteur :

Formats disponibles

Nutrition

Nestlé Nutrition Institute Workshop Series | Vol. 70

Zulfiqar A. Bhutta Karachi, Pakistan

Library of Congress Cataloging-in-Publication Data

Meeting micronutrient requirements for health and development / editors,

Printed on acid-free and non-aging paper

Basel · Freiburg · Paris · London · New York · New Delhi · Bangkok ·

Introduction to Micronutrient Requirements

Zinc in Maternal and Child Health

Multiple Micronutrient Deficiencies in Pregnancy

Vitamin A in Childhood: Evidence and Controversy

Folate, Vitamin B12 and Brain

191 Subject Index

In the past two decades, micronutrients have been recognized as having a

X Bhutta · Hurrell · Rosenberg

Meeting macro- and micronutrient requirements during pregnancy and early

Prof. Ferdinand Haschke, MD, PhD Natalia Wagemans, MD, PhD

XII Haschke · Wagemans

Chairpersons & Speakers

Prof. Lindsay H. Allen Prof. Ian Darnton-Hill

Prof. Mary Edith Penny Dr. Socorro de Leon-Mendoza

Global Micronutrient Deficiencies in

Despite numerous advances and improvements in child health, malnutrition

Micronutrient Deficiencies in Childhood and Impact on Growth and Survival 3

Fig. 2. VAS coverage levels: two doses. From UNICEF [29].

Vitamin A priority country;

growth retardation and hypogonadism was first described in 1963 in a study

Micronutrient Deficiencies in Childhood and Impact on Growth and Survival 5

In 2004, the WHO and UNICEF formulated a new recommendation to

Fig. 5. Prevalence of iodine deficiency in school-aged children [47].

Micronutrient Deficiencies in Childhood and Impact on Growth and Survival 7

Iron is an essential mineral for human development and function. It is required

Micronutrient Deficiencies in Childhood and Impact on Growth and Survival 9

Micronutrients in the Treatment of

Infant and child nutritional status is expressed as a weight-for-age, height-for-

Mean height-for-age Z score

Other Single Micronutrients

No other micronutrient has been so well studied in relation to linear growth as

Micronutrients for Stunting and Moderate Malnutrition 13

Given that micronutrient deficiencies rarely occur in isolation, attention has

Micronutrients for Stunting and Moderate Malnutrition 15

Delivery of Multiple Micronutrients

Complementary Food Supplements

These supplements include the highly nutrient-dense spreads described by

Micronutrients for Stunting and Moderate Malnutrition 17

Micronutrient powders presented in small single-use sachets to add to a serv-

Improving the Infant and Young Child’s Diet

Another alternative to improve stunting and moderate malnutrition is to

Discussion and Future Directions

Micronutrients for Stunting and Moderate Malnutrition 19

Micronutrients for Stunting and Moderate Malnutrition 21

Setting the Terms of Discussion for the Workshop

Salient Points of the Session’s Presentations

Discussion on Micronutrient Requirements 23

Summary of the Question and Answer Participation

Discussion on Micronutrient Requirements 25

Conclusion of the Discussant

Zinc Requirements: Assessment and

New Rapidly Absorption

Integumental Semen/ Urine Endogenous Non-absorbed

Fig. 1. Schematic of total body zinc homeostasis.

28 Hambidge · Miller · Krebs

Zinc Requirements: Assessment and Population Needs 29

1.5 2.0 2.5 3.0 3.5