Maternal hemoglobin concentration and birth weight1–3

Philip J Steer

ABSTRACT Pregnancy requires additional maternal absorp- major factor influencing hemoglobin concentration in pregnancy
tion of iron. Maternal iron status cannot be assessed simply from is expansion of plasma volume. How this occurs is not fully
hemoglobin concentration because pregnancy produces increases established but part of the sequence might be as follows. Heat
in plasma volume and the hemoglobin concentration decreases production by the fetoplacental unit causes a rise in body temper-
accordingly. This decrease is greatest in women with large ature. Heat loss is increased by peripheral vasodilatation, which
babies or multiple gestations. However, mean corpuscular vol- causes a drop in blood pressure. This in turn stimulates the
ume does not change substantially during pregnancy and a release of aldosterone from the adrenal gland, causing the reten-
hemoglobin concentration < 95 g/L in association with a mean tion of salts and water (2). The drop in osmolality that occurs
corpuscular volume < 84 fL probably indicates iron deficiency. reduces blood viscosity and enhances blood flow in the low-

Downloaded from [Link] by guest on April 13, 2011

Severe anemia (hemoglobin < 80 g/L) is associated with the birth pressure system of the intervillous space. Enhanced blood flow
of small babies (from both preterm labor and growth restriction), improves fetal growth. In women who are not given supplemen-
but so is failure of the plasma volume to expand. Hemoglobin tal iron, the hemoglobin concentration of the maternal blood falls
concentrations > 120 g/L at the end of the second trimester are from an average of <133 g/L in nonpregnant women to an aver-
associated with a ≤ 3-fold increased risk of preeclampsia and age of <110 g/L at 36 wk of gestation (3). The fall is steepest up
intrauterine growth restriction. The minimum incidence of low to 20 wk of gestation; the hemoglobin concentration remains
birth weight (< 2.5 kg) and of preterm labor (< 37 completed fairly constant up to 30 wk and then rises slightly thereafter (4, 5).
weeks) occurs in association with a hemoglobin concentration of These changes in hemoglobin concentration are due mainly to
95–105 g/L. This is widely regarded as indicating anemia in the changes in plasma volume; the red cell mass and total hemoglobin
pregnant woman but, if associated with a mean corpuscular actually increase during pregnancy.
volume > 84 fL, should be considered optimal. Am J Clin Failure of the plasma volume to expand adequately can lead to
Nutr 2000;71(suppl):1285S–7S. restricted fetal growth, resulting in the infant being small for ges-
tational age at birth (such infants are more vulnerable to the stress
KEY WORDS Maternal hemoglobin, birth weight, iron of labor). An alternative response is for the infant to initiate labor
deficiency, plasma volume, preterm labor, intrauterine growth prematurely. We studied a large database (data on 153 602 preg-
restriction, mean corpuscular volume, women nancies, collected in the North West Thames region of London
between 1988 and 1991, inclusive) that recorded the lowest hemo-
Women are more likely than men to have low iron stores globin concentration measured in each pregnant women during
because of blood loss at the time of menstruation. During preg- her pregnancy (usually at 28 wk) to discover the values associated
nancy, the fetal demand for iron increases maternal daily iron with the largest infants at birth. The highest mean birth weight
requirements from <1 to 2.5 mg/d in early pregnancy and occurred in association with a hemoglobin concentration of 85–
6.5 mg/d in the third trimester. The average daily diet in the 95 g/L (6). The minimum incidence of low birth weight (< 2.5 kg)
developed world contains <10–14 mg nonheme iron (1) but not and of preterm labor (< 37 completed weeks) occurred in associa-
all of this can be absorbed. Evidence from stable-isotope studies tion with a hemoglobin concentration of 95–105 g/L. Such values are
suggests that the percentage of nonheme iron absorbed from commonly considered to represent anemia [eg, by the World
food during normal pregnancy increases from 7% at 12 wk of Health Organization Expert Committee on Nutrition in 1965 (7)].
gestation to 36% at 24 wk and 66% at 36 wk. These dramatic The importance of adequate plasma volume expansion in allow-
changes enable the healthy pregnant woman to cope with the ing adequate fetal growth is attested to by several studies that
extra demands of pregnancy without becoming anemic (1), but showed an increased incidence of low birth weight in association
only if there is adequate iron in her diet. If the woman’s diet is
deficient in iron, as is the case in many developing countries, 1
From the Department of Maternal and Fetal Medicine, Imperial College
fetal requirements can be met only by additional contributions of
School of Medicine, London.
iron from maternal stores. This demand by the developing fetus 2
Presented at the symposium Maternal Nutrition: New Developments
may cause the mother to develop iron deficiency anemia if she and Implications, held in Paris, June 11–12, 1998.
had inadequate iron stores at the beginning of pregnancy. 3
Reprints not avaiable. Address correspondence to PJ Steer, Academic
Whether a pregnant woman is anemic cannot be assessed sim- Department of Obstetrics, Chelsea and Westminster Hospital, 369 Fulham
ply by measuring the blood hemoglobin concentration because a Road, London SW10 9NH, United Kingdom. E-mail: [Link]@[Link].

Am J Clin Nutr 2000;71(suppl):1285S–7S. Printed in USA. © 2000 American Society for Clinical Nutrition 1285S
1286S STEER

with either a high maternal hemoglobin concentration (8–15) or lowest hemoglobin concentrations were associated with the
high hematocrit (16–18). The mechanism by which this effect is highest birth weights, which might otherwise be thought to be
mediated is unknown but may be related to blood viscosity. Mater- advantageous. A subsequent study failed to show any evidence of
nal blood is supplied to the intervillous space of the placenta by a relation between placental ratio and hemoglobin concentration
spiral arteries, which are adapted to provide an almost continuous or mean cell volume at the first antenatal visit, and placental
low pressure flow. High pressure flow would probably cause an ratio was not associated with change in mean cell volume during
excessively high incidence of placental abruption (a condition in pregnancy or with third-trimester serum ferritin concentration
which the placenta separates prematurely from the uterine wall, (35). Barker’s group has since reported that a high maternal car-
often with fatal consequences for the fetus) and shock (due to bohydrate intake in early pregnancy is associated with a lower
hemorrhage and clotting disturbances in the mother). At low flow placental weight and a lower birth weight (36), so the interrela-
velocities, blood is very viscous and this promotes stasis and tionships are complex; as they commented in their 1991 article
thrombosis, which in any case are more likely as clotting factors (34), “specific advice to pregnant women must await further
are increased in pregnancy. A drop in viscosity would thus pro- studies.” This is particularly true given that recent studies failed
mote efficient blood flow within the placenta and vice versa. to show any relation between birth weight and blood pressure in
Another mechanism through which hemoglobin concentration young adulthood (37–42). In 1998, data were reported suggest-
might affect growth is the development of preeclampsia (a syn- ing that low birth weight is a feature of an inherited predisposi-
drome of hypertension, proteinuria, and multiorgan dysfunction), tion to hypertension, perhaps because it is associated with higher
which complicates <2% of all pregnancies. Failure of the plasma maternal blood pressure during pregnancy (43). Parental blood
volume to expand (and of the hemoglobin concentration to drop) is pressure may be an important confounding factor in the relation
associated with a ≤ 3-fold increase in the incidence of preeclamp- between low birth weight and subsequent hypertension.
sia in pregnancy (19). Thus, the poor placental flow associated with In summary, plasma volume expansion in normal pregnancy
not only high hemoglobin concentrations but also maternal vascu- causes a drop in maternal hemoglobin to concentrations com-

Downloaded from [Link] by guest on April 13, 2011

lar dysfunction might be implicated in fetal growth restriction. monly regarded as indicating anemia; in fact, concentrations of
In developed countries, it is likely that disorders of plasma vol- 95–115 g/L with a normal mean corpuscular volume (84–99 fL)
ume expansion and associated high hemoglobin concentrations are should be regarded as optimal for fetal growth and well-being
more important than is anemia in the genesis of low birth weight and are associated with the lowest risk of preterm labor. Routine
and preterm labor. Nonetheless, substantial iron deficiency anemia hematinic administration to women with values in these ranges
(usually considered to be < 80 g/L) is also associated with an is probably unnecessary.
increased incidence of low birth weight (20–27). The mechanism
by which anemia could produce this effect is unknown; it may be
that nutrient deficiencies other than iron are the important factors. REFERENCES
This hypothesis is strengthened by a review of 20 randomized, con- 1. Barrett JF, Whittaker PG, Williams JG, Lind T. Absorption of non-haem
trolled trials that showed that “routine supplementation (with iron iron from food during normal pregnancy. BMJ 1994;309:79–82.
and folic acid) has no detectable effect on any substantive measures 2. Chapman AB, Zamudio S, Woodmansee W, et al. Systemic and
of either maternal or fetal outcome, [in particular] . . . on preterm renal hemodynamic changes in the luteal phase of the menstrual
cycle mimic early pregnancy. Am J Physiol 1997;273:F777–82.
delivery, low birthweight, stillbirths or neonatal morbidity” (28).
3. Taylor DJ, Lind T. Red cell mass during and after normal pregnancy.
One study that looked at outcome in a subsequent pregnancy also Br J Obstet Gynaecol 1979;86:364–70.
found no benefit of supplementation (29). 4. Paintin DB, Thomson AM, Hytten FE. Iron and the haemoglobin level
Are there ways of assessing iron deficiency in pregnancy other in pregnancy. J Obstet Gynaecol Br Commonw 1966;73:181–90.
than measuring hemoglobin concentration and mean corpuscular 5. Meng Lu Z, Goldenberg RL, Cliver S, Cutter G, Blankson M. The
volume? Ferritin is not a useful measure because concentrations relationship between maternal hematocrit and pregnancy outcome.
decrease dramatically in pregnancy irrespective of whether iron Obstet Gynecol 1991;77:190–4.
supplementation is given (30). Bone marrow aspiration seems to 6. Steer PJ, Alam MA, Wadsworth J, Welch A. Relation between
be an excessive response to moderately low hemoglobin concen- maternal haemoglobin concentration and birthweight in different
ethnic groups. BMJ 1995;310:489–91.
trations. A pragmatic trial of oral iron therapy is probably the best
7. World Health Organization Expert Committee. Nutrition in preg-
approach unless the anemia is severe (< 60 g/L). nancy and lactation; report of a WHO Expert Committee. World
Should oral iron be given routinely to pregnant women? Over Health Organ Tech Rep Ser 1965;302:5–19.
the past 10 y, Barker et al (31–33) proposed that intrauterine 8. Gaspar MJ, Ortega RM, Moreiras O. Relationship between iron sta-
malnutrition leads to adverse sequelae in adult life, notably caus- tus in pregnant women and their newborn babies. Investigation in a
ing hypertension. These researchers proposed that not only low Spanish population. Acta Obstet Gynecol Scand 1993;72:534–7.
birth weight but also high placental weight (presumably a com- 9. Huisman A, Aarnoudse JG. Increased 2nd trimester hemoglobin
pensatory phenomenon) is a sign of intrauterine nutrition (32). concentration in pregnancies later complicated by hypertension and
Subsequently, they showed that lower hemoglobin concentra- growth retardation. Early evidence of a reduced plasma volume.
tions in pregnancy are associated with larger placentas (34) and Acta Obstet Gynecol Scand 1986;65:605–8.
10. Koller O, Sagen N, Ulstein M, Vaula D. Fetal growth retardation
suggested that this indicated that <50% of well-nourished
associated with inadequate haemodilution in otherwise uncompli-
women in the Oxford region of the United Kingdom were signi-
cated pregnancy. Acta Obstet Gynecol Scand 1979;58:9–13.
ficantly iron deficient during pregnancy. They further suggested 11. Koller O, Sandvei R, Sagen N. High hemoglobin levels during preg-
that this discordance between placental and fetal growth could nancy and fetal risk. Int J Gynaecol Obstet 1980;18:53–6.
result in long-term consequences for the infant, such as an 12. Murphy JF, O’Riordan J, Newcombe RG, Coles EC, Pearson JF.
increase in the incidence of hypertension in later life. Barker et Relation of haemoglobin levels in first and second trimesters to out-
al made these conclusions despite having also shown that the come of pregnancy. Lancet 1986;1:992–5.
 MATERNAL HEMOGLOBIN AND BIRTH WEIGHT 1287S

13. Ortner A, Zech H, Humpeler E, Mairbaeurl H. May high oxygen 28. Mahomed K. Routine iron supplementation in pregnancy. In: Enkin
affinity of maternal hemoglobin cause fetal growth retardation? MW, Keirse MJNC, Renfrew MJ, Neilson JP, eds. Pregnancy and child-
Arch Gynecol 1983;234:79–85. birth module, Cochrane database of systematic reviews. 2nd ed. Oxford,
14. Rasmussen S, Oian P. First- and second-trimester hemoglobin levels. United Kingdom: Cochrane Updates on Disk, 1993 (update software).
Relation to birth weight and gestational age. Acta Obstet Gynecol 29. Hemminki E, Merilainen J. Long-term follow-up of mothers and
Scand 1993;72:246–51. their infants in a randomized trial on iron prophylaxis during preg-
15. Sagen N, Nilsen ST, Kim HC, Bergsjo P, Koller O. Maternal hemo- nancy. Am J Obstet Gynecol 1995;173:205–9.
globin concentration is closely related to birth weight in normal 30. Fenton V, Cavill I, Fisher J. Iron stores in pregnancy. Br J Haematol
pregnancies. Acta Obstet Gynecol Scand 1984;63:245–8. 1977;37:145–9.
16. Dunlop W, Furness C, Hill LM. Maternal haemoglobin concentration, 31. Barker DJ, Osmond C, Golding J, Kuh D, Wadsworth ME. Growth
haematocrit and renal handling of urate in pregnancies ending in the in utero, blood pressure in childhood and adult life, and mortality
births of small-for-dates infants. Br J Obstet Gynaecol 1978;85: from cardiovascular disease. BMJ 1989;298:564–7.
938–40. 32. Barker DJ, Bull AR, Osmond C, Simmonds SJ. Fetal and placental
17. Forest JC, Masse J, Moutquin JM. Maternal hematocrit and albumin size and risk of hypertension in adult life. BMJ 1990;301:259–62.
as predictors of intrauterine growth retardation and preterm deliv- 33. Barker DJ. Fetal origins of coronary heart disease. BMJ 1995;
ery. Clin Biochem 1996;29:563–6. 311:171–4.
18. Lu ZM, Goldenberg RL, Cliver SP, Cutter G, Blankson M. The rela- 34. Godfrey KM, Redman CW, Barker DJ, Osmond C. The effect of
tionship between maternal hematocrit and pregnancy outcome. maternal anaemia and iron deficiency on the ratio of fetal weight to
Obstet Gynecol 1991;77:190–4. placental weight. Br J Obstet Gynaecol 1991;98:886–91.
19. Murphy JF, Newcombe RG, O’Riordan J, Coles EC, Pearson JF. 35. Perry IJ, Beevers DG, Whincup PH, Bareford D. Predictors of ratio
Relation of haemoglobin levels in first and second trimesters to out- of placental weight to fetal weight in multiethnic community. BMJ
come of pregnancy. Lancet 1986;1:992–5. 1995;310:436–9.
20. Bhargava M, Iyer PU, Kumar R, Ramji S, Kapani V, Bhargava SK. 36. Godfrey K, Robinson S, Barker DJ, Osmond C, Cox V. Maternal
Relationship of maternal serum ferritin with foetal serum ferritin, nutrition in early and late pregnancy in relation to placental and

Downloaded from [Link] by guest on April 13, 2011

birth weight and gestation. J Trop Pediatr 1991;37:149–52. fetal growth. BMJ 1996;312:410–4.
21. Hirve SS, Ganatra BR. Determinants of low birth weight: a com- 37. Seidman DS, Laor A, Gale R, Stevenson DK, Mashiach S, Danon YL.
munity based prospective cohort study. Indian Pediatr 1994;31: Birth weight, current body weight, and blood pressure in late ado-
1221–5. lescence. BMJ 1991;302:1235–7.
22. Onadeko MO, Avokey F, Lawoyin TO. Observations of stillbirths, 38. Lucas A, Morley R. Does early nutrition in infants born before term
birthweight and maternal haemoglobin in teenage pregnancy in programme later blood pressure? BMJ 1994;309:304–8.
Ibadan, Nigeria. Afr J Med Med Sci 1996;25:81–6. 39. Matthes JW, Lewis PA, Davies DP, Bethel JA. Relation between
23. Scholl TO, Hediger ML, Fischer RL, Shearer JW. Anemia vs iron birth weight at term and systolic blood pressure in adolescence.
deficiency: increased risk of preterm delivery in a prospective study. BMJ 1994;308:1074–7.
Am J Clin Nutr 1992;55:985–8. 40. Morley R, Lister G, Leeson-Payne C, Lucas A. Size at birth and
24. Scholl TO, Hediger ML. Anemia and iron-deficiency anemia: com- later blood pressure. Arch Dis Child 1994;70:536–7.
pilation of data on pregnancy outcome. Am J Clin Nutr 1994; 41. Vestbo E, Damsgaard EM, Froland A, Mogensen CE. Birth weight
59(suppl):492S–500S. and cardiovascular risk factors in an epidemiological study. Dia-
25. Singla PN, Tyagi M, Kumar A, Dash D, Shankar R. Fetal growth in betologia 1996;39:1598–602.
maternal anaemia. J Trop Pediatr 1997;43:89–92. 42. Laor A, Stevenson DK, Shemer J, Gale R, Seidman DS. Size at
26. Spinillo A, Capuzzo E, Piazzi G, Nicola S, Colonna L, Iasci A. birth, maternal nutritional status in pregnancy, and blood pressure at
Maternal high-risk factors and severity of growth deficit in small for age 17: population based analysis. BMJ 1997;315:449–53.
gestational age infants. Early Hum Dev 1994;38:35–43. 43. Walker BR, McConnachie A, Noon JP, Webb DJ, Watt GCM. Con-
27. Swain S, Singh S, Bhatia BD, Pandey S, Krishna M. Maternal tribution of parental blood pressures to association between low
hemoglobin and serum albumin and fetal growth. Indian Pediatr birth weight and adult high blood pressure: cross sectional study.
1994;31:777–82. BMJ 1998;316:834–7.