HYPOTHESIS

Hypothesis

Short interpregnancy intervals and unfavourable pregnancy

outcome: role of folate depletion

Luc J M Smits, Gerard G M Essed

There is no generally accepted explanation for the excess risk of adverse pregnancy outcome after short
interpregnancy intervals. In this paper, we present a hypothesis that is both biologically plausible, empirically testable,
and able to explain many observations. Maternal serum and erythrocyte concentrations of folate decrease from the
fifth month of pregnancy onwards and remain low for a fairly long time after delivery. Women who become pregnant
before folate restoration is complete have a raised risk of folate insufficiency at the time of conception and during
pregnancy. As a consequence, their offspring have higher risks of neural tube defects, intrauterine growth retardation,
and preterm birth. We make several predictions based on our hypothesis and suggest ways of testing them empirically.
The proposed mechanism implies, among other things, that postpartum supplementation with folic acid might prevent
excess risk of unfavourable pregnancy outcome in women with short interpregnancy intervals.

Published studies of pregnancy outcome in relation to
pregnancy spacing have often shown raised risks of adverse
outcome after short interpregnancy intervals.1–14 There has
been disagreement on the causality of these associations.
Sceptics have argued that short interpregnancy intervals
merely designate women already at higher reproductive
risk, either because of underlying disorders or
socioeconomic status (SES), or lifestyle factors.15,16
Research into both crude and adjusted outcome measures
generally shows some reduction in excess risks of adverse
pregnancy outcome after controlling for preceding
pregnancy outcome, lifestyle factors (such as smoking) or
SES.2–4,6–11 However, investigators who controlled for all
three types of confounders still reported raised risks of
congenital malformation, intrauterine growth retardation,
and preterm delivery.2–6,9,11,13 A number of causal
mechanisms have been proposed to explain the effects of
short interpregnancy intervals (typically defined as
intervals shorter than 6 months). These include
postpartum hormonal imbalance continuing into the new
pregnancy,8 preovulatory ageing of the oocyte due to an
extended follicular phase of the first few ovulatory cycles,17
maternal stress induced by the new pregnancy,9 and
maternal depletion of nutrients.2,8,13 None of these
hypotheses have been tested empirically, except for
maternal depletion of nutrients, the role of which has been
estimated by adjusting for prepregnancy and postpartum
maternal bodyweight, maternal age, and parity in the
statistical analysis, with inconsistent results.2,3 The
maternal depletion hypothesis is explained by changes in
maternal weight, protein and energy balance, and essential
fatty acids.8,18,19 However, not only macronutrients but also
micronutrients such as trace minerals and vitamins might
be depleted in the course of pregnancy and in the first few
months postpartum, leading to deficiencies if a new
pregnancy occurs. Our hypothesis is that depletion of
maternal folate is a major contributor to the excess risk of
adverse pregnancy outcome after short interpregnancy
Lancet 2001; 358: 2074–77
Departments of Epidemiology (L J M Smits PhD) and Obstetrics and
Gynaecology (G G M Essed PhD), Universiteit Maastricht,
PO Box 616, 6200 MD Maastricht, Netherlands
Correspondence to: Dr Luc J M Smits
(e-mail: Luc.Smits@epid.unimaas.nl)
intervals. This hypothesis is biologically plausible, able to
explain many hitherto observed phenomena, serves as a
basis for new predictions, and is readily testable.
Course of folate concentrations during and
after pregnancy
Human beings are fully dependent on dietary sources or
dietary supplements for their folate supply. Folate is
required for cell division because of its role in DNA
synthesis.20 During pregnancy, folate demand is
increased.21 Without adequate folate supplementation,
concentrations of folate in maternal serum, plasma, and
red blood cells decrease from the fifth month of pregnancy
onwards.22–24 Concentrations continue to decrease for
several weeks after pregnancy,22,23,25-27 and by the second to
third month postpartum a third of all mothers can have
subnormal concentrations of folate in serum and red blood
cells.23,24 By 6 months postpartum, 20% of mothers were
reported as still folate deficient.27
Folic acid supplementation during pregnancy and
lactation ensures adequate maternal folate concentrations.
Lactating mothers supplemented with 0·8–1 mg folic acid
per day had folate concentrations in red blood cells (6–12
weeks postpartum) higher than or equal to those of
unsupplemented, non-pregnant women.25 The folate
concentration of breast milk is quite insensitive to folic
acid supplementation and even increases as lactation
progresses, at the expense of maternal tissue stores.25
Folate deficiency-related adverse pregnancy
outcome
A number of studies have addressed the relation between
pregnancy outcome and either maternal blood folate
concentrations, folate intake, or hyperhomocysteinaemia
(the effect of inadequate folate intake or abnormal folate
metabolism).
Periconceptional supplementation with folate has been
shown to reduce the risk of neural tube defects by almost
three-quarters.28 Studies in animals suggest that
homocysteine is the teratogen which causes excess neural
tube defects in folate deficiency.29 This study also
indicated that homocysteine could cause congenital
defects of the heart, which was confirmed by the results
of a randomised trial of periconceptional folate
supplementation (50% reduction of mainly conotruncal
and septal defects).30

2074 THE LANCET • Vol 358 • December 15, 2001

For personal use. Only reproduce with permission from The Lancet Publishing Group.
HYPOTHESIS

The effects of folate and folate supplementation on the
risk of spontaneous abortion are less clear. Women who
habitually miscarry generally have higher plasma
concentrations of homocysteine, caused by the increased
frequency of a genetic disorder of folate metabolism; these
concentrations can be normalised by administration of
folic acid.31 On the other hand, no significant decrease or
increase in the risk of spontaneous abortion was reported
in normal women receiving periconceptional folate
supplements.30 The precise role of folate and folic acid
supplementation in sporadic spontaneous abortion (apart
from their role in neural tube defects) remains unclear.20,32
Mothers of growth retarded fetuses generally have lower
folate intake, lower blood folate concentrations, and
higher homocysteinaemia rates than mothers of normal
weight fetuses.20 The same is true for mothers who deliver
prematurely.20 Supplementation of folic acid during the
entire pregnancy in healthy women has yielded
inconsistent results with respect to birth outcome,33 but
occurring after a (spontaneous or elective) abortion, short
interpregnancy intervals caused no increased risks, an
expected result in view of the relatively short-term
burdening of maternal folate reserves in case of early
pregnancy termination.
A nearly 70% higher crude risk of dying from congenital
heart disease was reported for children conceived within 6
months after a livebirth compared with children conceived
within 18–23 months after a livebirth,1 consistent with the
(inverse) relation of folate with congenital heart defects.30
Only a small excess risk of spontaneous abortion (⫹10%)
was reported for pregnancies starting within 6 months
A
First Interpregnancy Second
pregnancy interval pregnancy

Maternal red
highly favourable effects on birthweight were observed

cell folate
among South-African women with otherwise folate-
deficient diets.34

Our hypothesis

deficiency
We hypothesise that the excess risk of adverse pregnancy

Folate
outcome after short (<6 months), as opposed to longer,
interpregnancy intervals is largely attributable to
insufficient repletion of maternal folate resources.
The figure (parts A and B) gives a schematic Time (months)
representation of the expected course of folate B
concentrations in women with short and longer First Interpregnancy Second
interpregnancy intervals. As of the fifth month of pregnancy interval pregnancy
pregnancy, maternal folate concentrations decrease. They
continue to do so during the first months postpartum,
both with and without lactation. If the second pregnancy
Maternal red

starts after a sufficiently long restoration period, the

cell folate

probability of maternal folate deficiency is equal to that in

the first pregnancy (figure). A new pregnancy starting
before complete folate restoration has taken place,
however, will have a higher risk of maternal folate

deficiency
deficiency (figure).

Folate
Maternal iron stores, like those of folate, might remain
depleted for several months after pregnancy,24 and iron-
deficiency anaemia during early pregnancy is related to
Time (months)
preterm delivery (and associated low birthweight).35,36
Therefore, part of the excess risk of preterm birth after C
short interpregnancy intervals could be attributable to First Interpregnancy Second
insufficiently repleted maternal iron stores. However, pregnancy interval pregnancy
there seems to be no relation between maternal iron
deficiency and other reported effects of short
interpregnancy intervals, such as neural tube defects and
Maternal red

intrauterine growth restriction.36 Concentrations of other

cell folate

micronutrients such as zinc and vitamins A, B6, and B12

also fall during pregnancy, but either return to normal
within a few weeks postpartum or do not have a relation
with pregnancy outcome, or both.24,26,27,37 Therefore, folate
deficiency

deficiency seems the most important nutritional factor in

the excess risk of poor pregnancy outcome after short
Folate

interpregnancy intervals.

Earlier observations in line with our hypothesis Time (months)

Many earlier observations in studies of the effects of short
interpregnancy intervals on pregnancy outcome can be
explained by maternal folate depletion. A two-fold
increase in the risk of neural tube defects was observed for
conceptions within 6 months after a livebirth compared
with conceptions 1 to 2 years after a livebirth.13 The
investigators controlled for multivitamin use and other
potential confounders. Within the group of pregnancies
Expected course of maternal erythrocyte folate concentrations
across consecutive pregnancies in three different populations
A: Women with interpregnancy intervals sufficiently long for complete
folate restoration. B: Women with short interpregnancy intervals. C:
Women with short interpregnancy intervals and adequate postpartum
folate supplementation. Thick horizontal line: mean maternal folate
concentration in red blood cells in the population. Shaded area: range of
folate concentrations in the population. Thin undulating line: cut-off point
for folate deficiency (usually 340 nmol/L).

THE LANCET • Vol 358 • December 15, 2001 2075

For personal use. Only reproduce with permission from The Lancet Publishing Group.

after a livebirth,14 corresponding with the uncertain and
debated role of folate in sporadic spontaneous abortion.28,32
The effect of short interpregnancy intervals on the risk
of smallness for gestational age (SGA) has been addressed
in a number of studies.2–6,10,11 Reported adjusted odds ratio
point estimates for pregnancies within 6 months after the
preceding delivery (with variable reference intervals)
indicate a 20–50% raised risk of SGA.2,4,6,11 The risk of
SGA was noted to decrease with increasing interval
length.10
The risk of preterm delivery (<37 weeks) has been
reported to be increased by 10–60% (based on adjusted
odds ratios) in fetuses conceived after intervals of
6 months and less compared with variable reference
categories.6,11,12 Adjusted odds ratios for short pregnancy
intervals were noted to increase with increasing length of
the preceding pregnancy,8 consistent with a more excessive
depletion of maternal folate resources in longer
pregnancies. In another study, Afro-Americans were seen
to have excess risks of preterm birth and low birthweight
from interpregnancy intervals up to 9 months, whereas for
Caucasian Americans (whose daily folate intake is
generally higher38) the effect was limited to intervals up to
3 months.5
Predicted effects of our hypothesis
On the assumption that the hypothesis is true, several
other effects can be expected. In breastfeeding women,
the extra risk connected with short interpregnancy
intervals is expected to be greater because of a higher
probability of folate depletion during their first months
postpartum. In addition, the excess risk is expected to last
longer because of a longer duration of folate insufficiency.
In statistical analyses of the effects of the length of the
interpregnancy interval on pregnancy outcome, effects of
short intervals are expected to diminish or disappear after
adequate adjustment for maternal folate status.
Adequate folate supplementation during the
interpregnancy interval and during the new pregnancy is
expected to diminish or eliminate the adverse effects of
short interpregnancy intervals on pregnancy outcome.
Testing the hypothesis
The most rigorous test of the hypothesis would be to do a
randomised, placebo-controlled trial of folate supp-
lementation in a large group of women who are at risk of
becoming pregnant in the near future and who have
delivered a baby in the recent past (so that short
interpregnancy intervals might occur). These women
would have to be followed until the relevant outcome
could be measured. If the hypothesis is not true, short
interpregnancy intervals would still be a factor explaining a
large part of the excess risk of adverse pregnancy outcome
in the group of supplemented women, as would be
expected to be the case in the group of unsupplemented
women.
A less rigorous, though potentially informative, test
would be to measure, in one population, both
interpregnancy interval, pregnancy outcome, and maternal
folate status during the critical stages of embryonic and
fetal development, and analyse whether the effects of short
intervals diminish after adequate control for maternal
folate status in the analysis. If they do not decrease, one
reason could be that the hypothesis is not true or only
partly explains excess risks.
Conclusion
If the proposed mechanism correctly explains the excess
unfavourable pregnancy outcome seen after short
2076
For personal use. Only reproduce with permission from The Lancet Publishing Group.
HYPOTHESIS
interpregnancy intervals, several types of preventive
intervention are possible. First, women could be advised
to take postpartum supplementation of folic acid to ensure
adequate folate concentrations during the first months
after delivery (figure). With supplementation, postpartum
folate concentrations could be normalised within 1⋅5 to
3 months, even in lactating mothers.25 Overall, 6–12% of
all non-first pregnancies are conceived within a half year
after another pregnancy (US data).6 Most of these
pregnancies are unplanned, and only a small proportion of
mothers involved are expected to have taken
periconceptional folate supplements. These women and
their offspring would benefit from shifting the weight from
periconceptional folic acid intake (as currently advised) to
postpartum supplementation.
Second, those who have already conceived shortly after
another pregnancy might be advised to start taking folate
supplements as quickly as possible. Although this may not
reduce the risk of neural tube defects and other congenital
anomalies much, unfavourable outcomes in late
pregnancy such as preterm labour and fetal growth
retardation could be prevented in this way.
Another corollary would be that women who wish to
become pregnant soon after another pregnancy, would not
have to be advised to wait a minimum period until
spacing-associated risks are normalised, but instead
should focus on achieving adequate concentrations of
folate before conception. We, however, do not intend to
ignore the fact that the fetus is only one of the parties
potentially affected by narrow pregnancy spacing.
Unplanned early conceptions can cause mental stress in
the parents; furthermore, because of the burden of the
new pregnancy, the preceding child might not get the care
and attention it would otherwise have received.
We thank R A de Bie for comments on an early version of this paper and
P A van den Brandt for helpful discussions.
References
1 Fedrick J, Adelstein P. Influence of pregnancy spacing on outcome of
pregnancy. BMJ 1973; 4: 753–56.
2 Ferraz EM, Gray RH, Fleming PL, Maia YM. Interpregnancy interval
and low birth weight: findings from a case-control study.
Am J Epidemiol 1988; 128: 1111–16.
3 Klebanoff MA. Short interpregnancy interval and the risk of low birth
weight. Am J Public Health 1988; 78: 667–70.
4 Kallan JE. Effects of interpregnancy intervals on preterm birth,
intrauterine growth retardation, and fetal loss. Soc Biol 1992; 39:
231–45.
5 Rawlings JS, Rawlings VB, Read JA. Prevalence of low birth weight
and preterm delivery in relation to the interval between pregnancies
among white and black women. N Engl J Med 1995; 332: 69–74.
6 Kallan JE. Reexamination of interpregnancy intervals and subsequent
birth outcomes: evidence from US linked birth/infant death records.
Soc Biol 1997; 44: 205–12.
7 Smits LJ, Willemsen WNP, Zielhuis GA, Jongbloet PH. Conditions at
conception and risk of menstrual disorders. Epidemiology 1997; 8:
524–29.
8 Basso O, Olsen J, Knudsen LB, Christensen K. Low birth weight and
preterm birth after short interpregnancy intervals. Am J Obstet Gynecol
1998; 178: 259–63.
9 Khoshnood B, Lee KS, Wall S, Hsieh HL, Mittendorf R. Short
interpregnancy intervals and the risk of adverse birth outcomes among
five racial/ethnic groups in the United States. Am J Epidemiol 1998;
148: 798–805.
10 Shults RA, Arndt V, Olshan AF, Martin CF, Royce RA. Effects of
short interpregnancy intervals on small-for-gestational age and preterm
births. Epidemiology 1999; 10: 250–54.
11 Zhu BP, Rolfs RT, Nangle BE, Horan JM. Effect of the interval
between pregnancies on perinatal outcomes. N Engl J Med 1999; 340:
589–94.
12 Fuentes-Afflick E, Hessol NA. Interpregnancy interval and the risk of
premature infants. Obstet Gynecol 2000; 383: 383–90.
THE LANCET • Vol 358 • December 15, 2001
HYPOTHESIS

13 Todoroff K, Shaw GM. Prior spontaneous abortion, prior elective
termination, interpregnancy interval, and risk of neural tube defects.
Am J Epidemiol 2000; 151: 505–11.
14 Wohlfahrt J, Andersen AMN, Melbye M. Interval between pregnancies
and risk of spontaneous abortion. Epidemiology 2000; 11: 92–93.
15 Erickson JD, Bjerkedal T. Interval between pregnancies. Lancet 1979;
1: 52.
16 Klebanoff MA. The interval between pregnancies and the outcome of
subsequent births. N Engl J Med 1999; 340: 643–44.
17 Jongbloet PH. The effect of preovulatory overripeness of human eggs
on development. In: Blandau RJ, ed. Aging gametes. Basel,
Switzerland: S Karger, 1975, 300–29.
18 Winkvist A, Rasmussen KM, Habicht JP. A new definition of the
maternal depletion syndrome. Am J Public Health 1992; 82: 691–94.
19 Merchant K, Martorell R, Haas J. Maternal and fetal responses to the
stresses of lactation concurrent with pregnancy and of short
recuperative intervals. Am J Clin Nutr 1990; 52: 280–88.
20 Scholl TO, Johnson WG. Folic acid: influence on the outcome of
pregnancy. Am J Clin Nutr 2000; 71 (suppl): 1295–303.
21 McPartlin J, Halligan A, Scott JM, Darling M, Weir DG. Accelerated
folate breakdown in pregnancy. Lancet 1993; 341: 148–49.
22 Bates CJ, Fuller NJ, Prentice AM. Folate status during pregnancy and
lactation in a West African rural community. Hum Nutr Clin Nutr
1986; 40C: 3–13.
23 Qvist I, Abdulla M, Jägerstad M, Svensson S. Iron, zinc and folate
status during pregnancy and two months after delivery. Acta Obstet
Gynecol Scand 1986; 65: 15–22.
24 Açkurt F, Wetherilt H, Löker M, Hacibekiro M. Biochemical
assessment of nutritional status in pre- and post-natal Turkish women
and outcome of pregnancy. Eur J Clin Nutr 1995; 49: 613–22.
25 Smith AM, Picciano MF, Deering RH. Folate supplementation
during lactation: maternal folate status, human milk folate content,
and their relationship to infant folate status. J Pediatr Gastroenterol Nutr
1983; 2: 622–28.
26 Dostálová L. Vitamin status during puerperium and lactation. Ann
Nutr Metab 1984; 28: 385–408.
27 Bruinse HW, Van den Berg H. Changes of some vitamin levels during
and after normal pregnancy. Eur J Obstet Gynecol Reprod Biol 1995; 61:
31–37.
28 Lumley J, Watson L, Watson M, Bower C. Periconceptional
supplementation with folate and/or multivitamins for preventing neural
tube defects (Cochrane Review). In: The Cochrane Library, Issue 1,
2001. Oxford: Update Software.
29 Rosenquist TH, Ratashak SA, Selhub J. Homocysteine induces
congenital defects of the heart and neural tube: effect of folic acid.
Proc Natl Acad Sci USA 1996; 93: 15227–32.
30 Czeizel AE. Reduction of urinary tract and cardiovascular defects by
periconceptional multivitamin supplementation. Am J Med Genet
1996; 62: 179–83.
31 Wouters MGAJ, Boers GHJ, Blom HJ, et al. Hyperhomocysteinemia:
a risk factor in women with unexplained recurrent early pregnancy loss.
Fertil Steril 1993; 60: 820–25.
32 Hook EB, Czeizel AE. Can terathanasia explain the protective effect of
folic-acid supplementation on birth defects? Lancet 1997; 350: 513–15.
33 Mahomed K. Folate supplementation in pregnancy (Cochrane
Review). In: The Cochrane Library, Issue 1, 2001. Oxford: Update
Software.
34 Baumslag N, Edelstein T, Metz J. Reduction of incidence of
prematurity by folic acid supplementation in pregnancy. BMJ 1970;
1: 16–17.
35 Scholl TO, Hediger ML, Fischer RL, Shearer JW. Anemia vs iron
deficiency: increased risk of preterm delivery in a prospected study.
Am J Clin Nutr 1992; 55: 985–88.
36 Allen LH. Anemia and iron deficiency: effects on pregnancy outcome.
Am J Clin Nutr 2000; 71 (suppl): 1280–84.
37 Tamura T, Goldenberg RL, Johnston KE, DuBard M. Maternal
plasma zinc concentrations and pregnancy outcome. Am J Clin Nutr
2000; 71: 109–13.
38 Ford ES, Bowman BA. Serum and red cell blood folate
concentrations, race, and education: findings from the third National
Health and Nutrition Examination Survey. Am J Clin Nutr 1999; 69:
476–81.

Uses of error: Confessions of a neurologist

Many years ago, a very anxious young man with an unintelligible accent came to my
outpatient clinic having had a sudden, short-lived headache 6 weeks earlier. He had
attended our emergency department at the time, received a diagnosis of migraine,
and gone home. I could not obtain a clear history. I reasoned that the casualty
officer had not suspected a sub-arachnoid haemorrhage, and the period of the
highest risk of re-bleeding had passed. The only way to identify an aneurysm would
have been to admit him and do an angiogram. I did not think that it was appropriate
to discuss the risks with such an anxious patient. I reassured, and discharged him.
He died later from a ruptured cerebral aneurysm.
Neurologists have few opportunities to make mistakes with procedures. I shudder
when I think of the cisternal punctures I did without radiographic control in the
1970s. Once, a patient on whom I had just performed a lumbar puncture with great
difficulty, began to pass blood in his urine. Despite measuring the needle, and
consulting Gray's Anatomy, I spent a sleepless night. I was reassured the next
morning when the patient passed a renal stone.
After a sabbatical year away from clinical practice, I nearly made a ghastly error
with a prescription. I was telephoned about a patient with a myasthenic crisis. I
ordered neostigmine but forgot that the parenteral dose is 15 times less than the oral
dose. I arrived at the bedside to stop the nurse injecting the contents of an enormous
syringe, containing the entire hospital supply of neostigmine.
When history taking is difficult, error is more likely. Being a good physician
requires constant practice, and drugs come in containers of appropriate sizes.

Richard Hughes
Guy's, King's and St Thomas' School of Medicine, London SE1 9RT, UK

THE LANCET • Vol 358 • December 15, 2001 2077

For personal use. Only reproduce with permission from The Lancet Publishing Group.