The Effect of Calcium Supplementation on Blood Lead Levels

in Nigerian Children
Elizabeth M. Keating, BS, Philip R. Fischer, MD, John M. Pettifor, MB, BCh, PhD, Mark Pfitzner, MD, MPH,
Christian O. Isichei, BM, BCh, MS, and Tom D. Thacher, MD
Objective To determine whether calcium supplementation alters the risk of lead toxicity.
Study design Children aged 12-18 months from 3 communities in Nigeria were assigned to receive daily calcium
supplementation, as either calcium carbonate (400 mg) or ground dried fish (529  109 mg), or placebo. All children
received 2500 IU of vitamin A. Levels of blood lead, calcium, and vitamin D metabolites were measured at baseline
and after 12-18 months (n = 358).
Results The mean (SD) baseline lead level was 11.1  7.8 mg/dL (range, 1-43 mg/dL; median, 9 mg/dL); 44.7% of
subjects had a lead level >10 mg/dL. After 12-18 months, the mean lead level was 8.1  6.3 mg/dL (range, 1-48 mg/dL;
median, 6 mg/dL), with 22.6% with a level >10 mg/dL. Lead levels at baseline varied among communities (P = .01)
and were higher in children who used eye cosmetics or lived near a lead-acid battery melter (both P < .001). In a multiple regression model, the decrease in blood lead level was predicted by age, baseline lead level, and time of final
lead value at 12-18 months (R2 = 31%), but not by calcium supplementation (P = .98).
Conclusions Lead toxicity is common in Nigerian children, but calcium supplementation does not affect blood
lead levels. (J Pediatr 2011;159:845-50).

ead exposure early in life adversely affects the cognitive, neurobehavioral, and neurophysiological development of young
children.1 The major sources of lead in childrens environments are contaminated food and beverages, soil, street dust,
interior lead-based paint in older housing, and mining and industrial activity.1 In some areas, the use of lead-based eye
cosmetics is a factor.2 Regardless of the initial route of exposure, young children typically ingest environmental lead through
hand-to-mouth behaviors.3 Lead is absorbed from the gastrointestinal tract and stored in bone. During bone remodeling, lead
is released and can be toxic to the developing nervous system.3
Dietary calcium deficiency is found in many developing countries2,4-7 and occurs in the context of a diet lacking in dairy
products. In Nigerian children, the daily dietary intake of calcium is 200-300 mg, far below the dietary reference intakes
of 500 mg for children aged 1-3 years and 800 mg for children aged 4-8 years.8,9 Children with calcium-deficient diets may
preferentially absorb another divalent cation, such as lead, in place of calcium. A report from Nigeria found that up to 70%
of young children have a blood lead concentration >10 mg/dL.5
Lead is also known to substitute for calcium in a number of cellular processes, interfering with some reactions that require
calcium.4 It has been postulated that calcium supplementation can reduce the risk of lead toxicity by decreasing lead absorption
and inhibiting the adverse effects of lead on calcium-mediated functions.4 Several studies have noted decreased lead absorption
with increasing calcium supplementation,3,10-14 whereas others have found no effect.15,16 Although it is possible that childhood
lead toxicity may be aggravated by dietary calcium insufficiency, no previous studies have adequately examined whether calcium supplementation can prevent lead absorption in children. The present study was part of a longitudinal investigation to
assess the effect of dietary calcium supplementation on mineral homeostasis and rickets incidence. The primary objective of this
part of the trial was to test the hypothesis that calcium supplementation could reduce the risk of lead toxicity in Nigerian children accustomed to a low calcium intake.

Methods
This study was conducted in conjunction with a calcium supplementation trial to
examine whether rickets can be prevented in Nigerian children.8 Three geographically separate urban communities in Jos, Nigeria were selected for supplementation. Communities with similar ethnic and socioeconomic characteristics
and with well-functioning primary health facilities were selected to serve as study
centers.8 Personnel from each center were trained in data collection and study
1,25(OH)2D
25(OH)D

1,25-dihydroxyvitamin D
25-hydroxyvitamin D

From the Mayo Medical School, Rochester, MN (E.K.);

Departments of Pediatric and Adolescent Medicine (P.F.)
and Family Medicine (T.T.), Mayo Clinic, Rochester, MN;
MRC Mineral Metabolism Research Unit, Department of
Paediatrics, Faculty of Health Sciences, University of the
Witwatersrand and Chris Hani Baragwanath Hospital,
Johannesburg, South Africa (J.P.); Department of
Pediatrics, University of Utah, Salt Lake City, UT (M.P.);
and Department of Chemical Pathology, Jos University
Teaching Hospital, Jos, Nigeria (C.I.)
Supported in part by the Thrasher Research Fund and
the Primary Childrens Medical Center Foundation, Salt
Lake City, UT. The authors declare no conflicts of interest.
0022-3476/$ - see front matter. Copyright 2011 Mosby Inc.
All rights reserved. 10.1016/j.jpeds.2011.04.038

845

THE JOURNAL OF PEDIATRICS

www.jpeds.com

performance, and the quality of data collection was monitored by one of the investigators (T.T.).
Children aged 12-18 months who resided in the designated
communities were eligible for enrollment. All eligible children visiting the study center for routine growth monitoring
and immunizations were invited to participate. Information
was collected regarding age, ethnic group, religion, milk intake, source of drinking water, nature and frequency of cosmetic eye shadow use, pica, and proximity to a place where
lead-acid batteries are burned or melted. Written informed
consent was obtained from a parent, and the Ethics Committee of the Jos University Teaching Hospital and the Institutional Review Boards of the Mayo Clinic and the University
of Utah approved the study.
After thorough skin cleansing, blood was collected by
antecubital venipuncture. Two blood spots were prepared
for each subject on individually prepackaged filter paper
(LeadTech Corp, North Bergen, North Jersey) and resealed
in a plastic bag. Serum was stored at 20 C until being
transported frozen to the Mayo Clinic. Serum calcium,
phosphorus, alkaline phosphatase, and albumin concentrations were determined by standard methods. Concentrations of 25-hydroxyvitamin D [25(OH)D] and
1,25-dihydroxyvitamin D [1,25(OH)2D] were measured
by radioimmunoassay following acetonitrile precipitation
extraction (DiaSorin, Stillwater, Minnesota). Filter paper
blood spots were sent to LeadTech Corp for lead analysis
by atomic absorption spectrophotometry. Each of the 2
blood spots and a non-bloodstained portion of each filter
paper was analyzed to assess variability and exclude contamination.2,5
Intervention
Each community was randomly assigned to receive vitamin A
only, vitamin A plus calcium tablets, or vitamin A plus
ground fish (a locally available calcium source). Individual
subjects were not randomized, to simplify the study logistics
and avoid potential contamination of treatment assignments.
Vitamin A was chosen for use as a control because it does not
affect lead toxicity but does provide a health benefit to the
children enrolled in the study.17 Vitamin A was dispensed
in prefilled syringes for oral administration of 2500 U (0.1
mL) daily. Calcium carbonate was dispensed to provide
400 mg (2 chewable tablets that were crushed and mixed
with the childs regular food) of elemental calcium daily.
Ground fish was tested for palatability in 15 hospitalized children and was accepted by all of them. Local species of dried
catfish (Clarias gariepinus or Heterobranchus longifilis) were
baked and ground. A spoon was provided to administer 10
g of ground fish daily, mixed with the childs food. Twenty
monthly samples of ground fish were tested for calcium content; the mean (SD) value of elemental calcium in 10 g of
ground fish was 529  109 mg. Samples of ground fish had
no toxic concentrations of heavy metals. The lead content
of each of 5 separate batches of ground fish samples was <2
ppm. Samples were digested according to Environmental
Protection Agency Method 3050 and analyzed using induc846

Vol. 159, No. 5

tively coupled plasma spectrophotometry (American Environmental Consultants Laboratory, Salt Lake City, Utah).
Follow-Up
Study subjects were recruited between July 1998 and November 1998 and followed for 18 months, and data collection was
concluded in May 2000. Subjects returned every 4 weeks for
their assigned supplement. Intercurrent adverse events or illnesses were recorded at each visit. The volume of vitamin A,
number of calcium tablets, and weight of ground fish remaining at each visit were recorded to assess compliance. Community health workers attempted to contact the parents of
subjects who missed scheduled appointments to encourage
their return and determine the reason for missing the appointment.
Clinical evaluation and blood collection for lead measurement was repeated 12 months after enrollment. For subjects
who did not return at 12 months, evaluation and sampling
were completed by 18 months after enrollment. Baseline
blood lead results were not available until after the second
visit and blood draw. Once the results were available, parents
of children with elevated lead levels were advised to avoid using eye cosmetics on their children and to keep them away
from lead-acid battery facilities.
Statistical Methods
Data entry was performed using JMP 8.0 statistical software
(SAS Institute, Cary, North Carolina). Means of normally
distributed variables were compared using 2-sample t tests
for independent samples. Analysis of variance was used for
the comparison of nominal data with the continuous dependent variable, final lead concentration.
Multiple linear regression analyses were performed to control for differences in baseline variables among the 3 locations, with final blood lead level as the dependent variable.
Factors included in the model were age, sex, eye cosmetic
use, type of eye cosmetic, battery melting exposure, initial
blood lead level, initial serum calcium level, initial 25(OH)
D level, time of final blood lead sampling (12 or 18 months),
and calcium supplementation. The regression was performed
to determine whether calcium supplementation was a significant predictor of lead toxicity, while adjusting for other potentially important predictors of lead toxicity. Stepwise
backward elimination was used to develop a regression
model to predict final lead concentration. To address the
problem of colinearity among variables closely correlated
with one another, the single variable most closely related to
final blood lead concentration was selected for inclusion in
the model. A P value <.05 was considered significant.

Results
Baseline blood lead levels were obtained from 625 children,
and blood lead was measured at 12 or 18 months in 358 of
these children (57.2%). The 42.8% dropout rate was due to
such reasons as moving away, tiring of the inconvenience
of traveling to the clinic, and wanting to avoid blood sample
Keating et al

ORIGINAL ARTICLES

November 2011

Table I. Baseline characteristics of the children in the 3 intervention groups*

Characteristic
Age, months, mean  SD
Male sex, n (%)
Religion, n (%)
Christianity
Islam
Anthropometric characteristics, mean  SD
Height, cm
Weight, kg
Eye cosmetic use, n (%)
Battery melting exposure, n (%)
Dairy product calcium intake, mg, mean  SD
Serum calcium, mg/dL, mean  SD
25(OH)D, ng/mL, mean  SD
1,25(OH)2D, pg/mL, mean  SD
Phosphorus, mg/dL, mean  SD

Ground fish and

vitamin A (n = 103)

Calcium tablets and

vitamin A (n = 138)

Vitamin A only
(n = 117)

P value

13.8  1.5
52 (50.5)

14.8  2.1
70 (50.7)

14.5  1.8
62 (53)

<.001
.91

7 (7)
96 (93)

12 (8.7)
126 (91.3)

47 (40.2)
70 (59.8)

<.001

72.0  3.2
8.4  1.1
94 (91.3)
8 (7.8)
29  89
10.2  0.5
16.7  4.0
125  43.2
5.8  0.6

71.2  4.0
8.5  1.3
120 (87.0)
3 (2.2)
43  75
10.0  0.7
22.4  6.3
137  48.1
5.6  0.8

68.6  5.8
8.5  1.3
90 (76.9)
1 (0.9)
30  52
10.1  0.6
21.4  6.4
140  45.1
5.7  1.0

<.001
.55
.0086
.011
<.001
.048
<.001
.034
.26

*Characteristics were compared across groups using analysis of variance for continuous variables and the c2 test for categorical variables.

collection. The children who returned for blood sampling

form the basis of this report. These 358 children included
103 enrolled in the ground fish group, 138 in the calcium tablet group, and 117 controls. Baseline characteristics of the 3
groups are compiled in Table I. Significant differences in
age, height, and most biochemical measurements were
present among the groups at enrollment. Mean dairy
product calcium intake was uniformly low. The mean
blood lead level at baseline was 11.1  7.8 mg/dL (range, 143; median, 9), and 44.7% of subjects had a lead level >10
mg/dL (Figure 1). Lead toxicity (blood lead >10 mg/dL) was
found in 43 children (41.7%) in the ground fish group, in
67 children (48.6%) in the calcium tablet group, and in 50
children (42.7%) in the control group. Lead levels were
higher in children who used cosmetic eye shadow (11.8 
7.9 mg/dL vs 6.8  4.9 mg/dL; P < .001) or who lived near
lead-acid battery melters (16.5  14.2 mg/dL vs 10.9  7.4
mg/dL; P = .013). Baseline lead levels were not related to
baseline milk calcium intake (P = .26), serum phosphorus
level (P = .43), 25(OH)D level (P = .95), or 1,25(OH)2D
level (P = .84); however, baseline lead level was correlated
with serum calcium level at baseline (P = .0065; R2 = 2.3%).
Mean overall compliance for the 18-month trial, as
measured by the number of scheduled doses of vitamin

Figure 1. Blood lead levels (mg/dL) at baseline in all subjects.

At baseline, 44.7% of subjects had lead >10 mg/dL, which
qualifies as lead toxicity.

A consumed divided by the number of person-days of

follow-up, was 57%  25% in the calcium tablet group,
65%  26% in the ground fish group, and 63%  31%
in the placebo group (P = .009). In those who returned
for the 18-month follow-up visit, the total calcium consumed divided by the intended consumption was 50% 
21% in the calcium tablet group and 57%  26% in the
ground fish group (P = .02). The mean daily supplemental
calcium consumed was 226  99 mg in the calcium tablet
group and 309  137 mg in the ground fish group (P <
.001). Compared with the subjects who completed the
trial, those who dropped out came from larger families,
discontinued breast-feeding earlier, had fathers with occupations associated with lower incomes, and had greater
baseline 25(OH)D values.
Of the 358 children in the final analysis, 257 had a blood
lead measurement at 12 months and 101 had a blood lead
measurement at 18 months. Table II presents baseline,
final, and changes in blood lead levels during the
supplement trial for all 3 groups. After 12-18 months, the
mean lead level was 8.1  6.3 mg/dL (range, 1-48 mg/dL;
median, 6 mg/dL), and 22.6% had a blood lead level >10
mg/dL. Lead levels decreased significantly over time in each
community (Figure 2). Despite this decrease, however, the
decline in lead levels was not affected by calcium
supplementation (P = .98; Table III; available at www.
jpeds.com). In a multiple linear regression model, the
change in blood lead level was predicted by subject age,
baseline lead level, and time of final lead value (R2 = 31%),
but not by calcium supplementation. On multivariate
analysis, lower lead level at follow-up was associated with
lower blood lead levels at baseline, older age, and follow-up
at 12 months rather than 18 months. In a post hoc
multiple regression restricted to those with a blood lead
level >10 mg/dL at baseline, calcium supplementation did
not reduce lead levels (P = .65). When the 2 calcium
treatment groups were pooled and compared with the
placebo group, calcium supplementation still had no effect
on the change in blood lead level.

The Effect of Calcium Supplementation on Blood Lead Levels in Nigerian Children

847

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Vol. 159, No. 5

Table II. Blood lead, calcium, vitamin D, and phosphorus levels over time in 3 intervention groups
Characteristic

Ground fish and vitamin A (n = 103)

Lead, mg/dL
Baseline
Follow-up (12 or 18 months)
Change
Calcium, mg/dL
Baseline
Follow-up (12 or 18 months)
Change
25(OH)D, ng/mL
Baseline
Follow-up (18 months)
Change
1,25(OH)2D, pg/mL
Baseline
Follow-up (12 or 18 months)
Change
Phosphorus, mg/dL
Baseline
Follow-up (12 or 18 months)
Change

Calcium tablets and vitamin A (n = 138)

Vitamin A (n = 117)
9.9  6.3
7.5  4.6
2.3  6.0

P value*

10.4  6.7
6.4  4.8
4.0  7.2

12.6  9.3
9.8  7.9
2.8  8.1

10.2  0.5
9.6  0.5
0.6  0.7

10.0  0.7
9.7  0.5
0.3  0.8

10.1  0.6
9.7  0.5
0.4  0.6

.048
.56
.064

16.7  4.0
17.3  5.0
0.7  3.5

22.4  6.3
20.7  14.4
1.6  16.3

21.4  6.4
19.0  6.4
2.4  6.2

<.001
.044
.11

124.7  43.2
111.2  39.3
12.4  39.8

137.4  48.1
138.6  45.6
0.7  55.6

139.9  45.1
146.7  47.3
7.1  37.9

.034
<.001
.0081

5.8  0.6
5.0  0.8
0.8  0.9

5.6  0.8
5.1  0.7
0.5  0.9

5.7  1.0
5.1  0.7
0.6  1.2

.26
.22
.065

.012
<.001
.24

*P value from analysis of variance comparing the given characteristic across groups.
P < .05 compared with baseline by the paired t test.

Discussion
Almost half of the subjects had a blood lead level >10 mg/dL
at entry into this study, indicating that lead toxicity is common in Nigerian toddlers living in Jos. Although lead toxicity in children has been associated with a number of
different causes, in Nigeria lead exposure appears to be related mainly to eye cosmetic use and proximity to home
battery recycling.2 A large incidence of lead poisoning in
children associated with informal gold mining activity was
reported in northern Nigeria,18 but children in this study
were within the city and not in proximity to informal mining activities.
Over the course of our study, blood lead levels decreased in
all 3 communities studied. However, neither the use of eye
cosmetics nor exposure to lead-acid battery melters changed

Figure 2. Blood lead levels at baseline and follow-up according to treatment received. In the ground fish, calcium
tablet, and control groups, blood lead levels decreased by 4.0
 7.2 mg/dL, 2.8  8.1 mg/dL, and 2.3  6.0 mg/dL, respectively (P = .24). Lead levels decreased significantly over time in
each community (P < .05). At baseline, 44.7% of subjects had
lead >10 mg/dL, and after 12-18 months, only 22.6% had lead
>10 mg/dL.
848

over the course of the study. One explanation for the decline
in lead levels despite the continuing exposure to eye cosmetic
and battery melting could be decreased hand-to-mouth behaviors as the children aged.19
Although a significant decrease in lead toxicity was seen
over time, the decreasing lead level was not significantly
linked to calcium supplementation. Furthermore, calcium
supplementation had no effect on lead levels in the subgroup
of children with blood lead levels >10 mg/dL. Previous studies
have suggested that calcium has a favorable effect in reducing
lead toxicity3,10-14; however, our data and data from 2 other
studies suggest that this is not the case, at least in children.15,16 This discrepancy in findings could be related to
the fact that the preliminary reports supporting the notion
that supplemental calcium intake can reduce lead absorption
in children are based on animal studies, experiments with human adults, or cross-sectional studies of children using varying methodologies.20
Previous studies in rats have suggested that supplemental
calcium intake can reduce lead toxicity.10,21-23 One review
suggested that nutritional factors such as calcium may be important modifiers for the metabolism and toxicity of lead, and
that calcium supplementation may be just as, if not more, effective as chelation in lead toxicity treatment.1 However, these
claims were based on studies done on rats, and the interaction
between calcium and lead may be different in humans.
Other studies that have proposed an impact on lead toxicity from calcium supplementation are experiments with human adults and studies with varying methodologies. One
study that reported decreased lead absorption with increasing
dietary intake of calcium included only 8 male subjects.14 Because children have higher gastrointestinal lead absorption
per unit body weight than adults,1 it is likely that lead and calcium interactions vary with age. In addition, the small sample
size might not be representative of broader population
groups.
Keating et al

ORIGINAL ARTICLES

November 2011
Most studies on calcium and lead toxicity have been done
in rats or adult humans. Few studies have been done in children, and our study adds to this pool of information. In
a study of urban children in greater Newark, New Jersey,
the investigators speculated that calcium supplementation
in children with higher lead intakes would help reduce morbidity due to exposure to environmental lead.3 Our findings
suggest otherwise, but it is possible that the leadcalcium interactions are different in urban US children and Nigerian
children. For example, Nigerian children have a greater prevalence of lead toxicity with higher levels and lower calcium
intakes than US children; it is possible that these differences
could combine to produce the seemingly discrepant findings.
The suggested relationship between calcium supplementation and the reduced lead levels in other studies could provide a false sense of effectiveness and divert efforts from
lead elimination and behavior modification, which might
have more impact.20 Thus, it is important to keep in mind
that calciumlead interactions might not be of major importance in children, and efforts toward environmental abatement of lead is of greater public health significance.
One limitation of our study is the fact that blood lead levels
at baseline varied among the treatment groups. This limitation is likely related to a number of unmeasured variables
within the study communities that could not be adjusted
for in the regression. However, this limitation applies to all
observational cohort studies, and each child served as his
or her own control at baseline in the regression analysis. In
addition, the results are likely valid regardless, because risk
factors for higher lead levels that are not age-dependent
have been identified, such as the use of eye cosmetics and
proximity to lead-acid battery melters.2,12 Because these factors do not vary with age but did vary between treatment
groups at baseline in our study, they likely were constantly
differing among the groups throughout the study. The finding that lead levels changed similarly in each group regardless
of calcium supplementation indicates that calcium supplementation was not linked to the lead level changes.
Another possible limitation of our study is that the amount
of calcium supplementation might not have been high
enough to produce a significant effect on decreasing lead absorption. According to the Institute of Medicine, the dietary
reference intake for calcium for children aged 1-3 years is 500
mg/day.9 Our study supplemented the children with 400 mg
of calcium per day, which, if it were the childrens only source
of calcium, would be below the recommended amount. Thus,
it is possible that the leadcalcium interaction might not be
representative of that in children in areas where diets include
more calcium. In addition, the estimated supplement intake
varied from 230 to 300 mg/day over the duration of the
trial, and it is possible that these supplement doses might
have been too low to have an effect.
An additional concern is the 42.8% dropout rate. However, this rate is representative of a real-world setting. In addition, because we found that calcium did not affect lead
levels, assuming that the subjects who dropped out would
have changed this finding would require the assumption

that these subjects differed from those who completed the

study and would have benefited more from calcium. This is
possible, yet quite unlikely. In contrast, those who dropped
out were probably less likely to take the calcium supplementation, and it is unlikely that they would have altered the
study results. If anything, they would have swayed the results
more toward a null effect.
The 3 communities were located 5 km apart from one
other, and it is possible that contamination in treatment
could have occurred across communities. However, we
found no evidence to suggest that such contamination occurred. In addition, there was no way of ensuring, other
than parental report, that the supplement was actually ingested by the child and not by another family member, but
we did not receive any reports that this occurred. n
Submitted for publication Jan 12, 2011; last revision received Mar 21, 2011;
accepted Apr 20, 2011.
Reprint requests: Elizabeth M. Keating, BS, Mayo Medical School, Mayo
Clinic, 200 First Street SW, Rochester, MN 55905. E-mail: keating.elizabeth@
mayo.edu

References
1. Ahamed M, Siddiqui MKJ. Environmental lead toxicity and nutritional
factors. Clin Nutr 2007;26:400-8.
2. Wright NJ, Thacher TD, Pfitzner MA, Fischer PR, Pettifor JM. Causes of
lead toxicity in a Nigerian city. Arch Dis Child 2005;90:262-6.
3. Bruening K, Kemp FW, Simone N, Holding Y, Louria DB, Bogden JD.
Dietary calcium intakes of urban children at risk of lead poisoning. Environ Health Perspect 1999;107:431-5.
4. Kerper LE, Hinkle PM. Cellular uptake of lead is activated by depletion
of intracellular calcium stores. J Biol Chem 1997;272:8346-52.
5. Pfitzner MA, Thacher TD, Pettifor JM, Zoakah AL, Lawson JO,
Fischer PR. Prevalence of elevated blood lead levels in Nigerian children.
Ambul Child Health 2000;6:115-23.
6. Thacher TD, Fischer PR, Strand MA, Pettifor JM. Nutitional rickets
around the world: causes and future directions. Ann Trop Paediatr
2006;26:1-16.
7. Fischer PR, Thacher TD, Pettifor JM. Pediatric vitamin D and calcium nutrition in developing countries. Rev Endocr Metab Dis
2008;9:181-92.
8. Thacher TD, Fischer PR, Isichei CO, Pettifor JM. Prevention of nutritional rickets in Nigerian children with dietary calcium supplementation. J Bone Miner Res 2008;23:S500.
9. Institute of Medicine of the National Academies. Dietary reference intakes for calcium and vitamin D. Washington, DC: National Academies
Press; 2010.
10. Bogden JD, Gertner SB, Christakos S, Kemp FW, Yang Z, Katz SR, et al.
Dietary calcium modifies concentrations of lead and other metals and renal calbindin in rats. J Nutr 1992;122:1351-60.
11. Quarterman J, Morrison JN, Humphries WR. The influence of high dietary calcium and phosphate on lead uptake and release. Environ Res
1978;17:60-7.
12. Ziegler EE, Edwards BB, Jensen RL, Mahaffey KR, Fomon SJ. Absorption
and retention of lead by infants. Pediatr Res 1978;12:29-34.
13. Blake KCH, Mann M. Effects of calcium and phosphorus on gastrointestinal absorption of 203Pb in man. Environ Res 1983;30:188-94.
14. Heard MJ, Chamberlain AC. Effect of minerals and food on uptake of lead
from the gastrointestinal tract in human. Hum Toxicol 1982;1:411-5.
15. Sargent JD, Dalton MA, OConnor GT, Olmstead EM, Klein RZ.
Randomized trial of calcium glycerophosphatesupplemented infant formula to prevent lead absorption. Am J Clin Nutr 1999;
69:1224-30.

The Effect of Calcium Supplementation on Blood Lead Levels in Nigerian Children

849

THE JOURNAL OF PEDIATRICS

www.jpeds.com

16. Markowitz ME, Sinnett M, Rosen JF. A randomized trial of calcium supplementation for childhood lead poisoning. Pediatrics 2004;113:34-9.
17. Rabiu MM, Kyari F. Vitamin A deficiency in Nigeria. Niger J Med 2002;
11:6-8.
18. Moszynski P. Lead poisoning in Nigeria causes unprecedented emergency. BMJ 2010;341:c4031.
19. Xue J, Zartarian V, Moya J, Freeman N, Beamer P, Black K, et al. A metaanalysis of childrens hand-to-mouth frequency data for estimating nondietary ingestion exposure. Risk Anal 2007;27:411-20.

Vol. 159, No. 5

20. Ballew C, Bowman B. Recommending calcium to reduce lead toxicity in
children: a critical review. Nutr Rev 2001;59:71-9.
21. Miller GD, Massaro TF, Massaro EJ. Interactions between lead and essential elements: a review. Neurotoxicology 1990;11:99-120.
22. Mahaffey-Six M, Goyer RR. Experimental enhancement of lead toxicity
by low dietary calcium. J Lab Clin Med 1970;76:933-42.
23. Mahaffey KR, Haseman JD, Goyer RA. Dose-response to lead ingestion in rats on low dietary calcium. J Lab Clin Med 1973;83:
92-100.

50 Years Ago in THE JOURNAL OF PEDIATRICS

Precocious Puberty Due to Secreting Chorionepithelioma (Teratoma) of the Brain
Bruton OC, Martz DC, Gerard ES. J Pediatr 1961;59:719-25

ifty years ago in The Journal, Bruton et al wrote, It is evident that an early diagnosis could not be established in this
case, about a 7-year-old child with precocious puberty from a mixed germ cell brain tumor, diagnosed ultimately at
death. In the course of 6 months from presentation, the boy underwent testicular biopsy, multiple blood tests, two
electroencephalograms, pneumoencephalography, non-diagnostic posterior fossa craniotomy, and finally, repeat
craniotomy with an aborted attempt to explore the third ventricle. The surgeon visualized but could not resect the
tumor, so the patient was treated with 4000 cGy of irradiation before dying 12 months after the diagnosis. At autopsy,
the boy was found to have a 6- by 5- by 6-cm suprasellar mass with histology including malignant teratoma,
choriocarcinoma, and seminoma (germinoma). The authors believed that hormonal secretion from the tumor was
novel and noted chorionic gonadotropins in excess of 100 000 U/L.
Would this paper pass peer review for publication today in The Journal? Probably not. Fifty years later, there exists
an entirely new discipline of pediatric neuro-oncology, with most childrens hospitals having dedicated teams of
neuro-oncologists, neurosurgeons, neuroadiologists, and neuropathologists. Today, brain tumors are diagnosed
quickly with magnetic resonance imaging, and an elevation of serum alpha-fetoprotein, beta-human chorionic gonadotropin, or placental alkaline phosphatase would immediately raise suspicion for a germ cell tumor. We recognize that
germ cell tumors can rarely produce precocious puberty, by either endogenous hormone production or mass effect on
the pituitary. We divide germ cell tumors of the brain into pure germinomas (seminomas) or non-germinomatous
germ cell tumors (NGGCT). Nearly all children with germinomas have them cured, and most patients with NGGCT
survive for >5 years.1 Modern therapeutic protocols would call for this child to receive multi-agent chemotherapy
(even without undergoing a biopsy!) because the high gonadotropin level would have made clear the tumor was
a NGGCT with elements of choriocarcinoma. Resection might have occurred at diagnosis or after chemotherapy,
with craniospinal irradiation following to consolidate treatment. Although we can only speculate whether this child
would be healthy today, diagnosis would be easily established. As British physician Thomas Fuller wrote centuries
ago, All things are difficult before they are easy.

Paul Graham Fisher, MD

Departments of Neurology, Pediatrics, Neurosurgery, and Human Biology
Stanford University
Palo Alto, California
10.1016/j.jpeds.2011.05.050

Reference
1. Matsutani M. Clinical management of primary central nervous system germ cell tumors. Semin Oncol 2004;31:676-83.

850

Keating et al

ORIGINAL ARTICLES

November 2011

Table III. Factors predictive of final lead by multivariate analyses

Characteristic
Age at enrollment
Male sex
Eye cosmetic use
Battery melting exposure
Baseline serum calcium level
Baseline lead level
25(OH)D level
Follow-up lead measurement at 12 months
Calcium supplementation

Estimate
0.5181
0.4640
0.1160
0.2887
0.8056
0.3904
0.0952
1.8611
0.0089

The Effect of Calcium Supplementation on Blood Lead Levels in Nigerian Children

Standard error

P value

0.1628
0.3040
0.4395
0.9064
0.5206
0.0416
0.0490
0.3293
0.3298

.0016
.13
.79
.75
.12
<.001
.053
<.001
.98

850.e1