Commentary: Maternal nutrition, body proportions at birth, and adult chronic disease

The Dutch Famine Study, originally published as a full-length volume in 1975,1 is one of the classic texts in perinatal epidemio-logy and should be required reading for all students and practitioners in that field. It is based on the very unnatural experiment of severely limited food rations imposed by the Germans on the west of Holland during what has been termed the ‘hunger winter’ of 1944–1945. The limited rations began with a transport embargo in October 1944 and ceased abruptly with the liberation by allied forces in May 1945. Even pregnant women (who received extra rations) averaged <1000 kcal/day during the winter months. Birth records, including measurements of birthweight, birth length, and head circumference, were available from several teaching hospitals in the region. Because pregnant women were exposed at different times before, during, and after their pregnancies, the Dutch Famine Study enables a careful assessment of the effects of acute maternal undernutrition at specific time periods on the outcome of pregnancy. The major findings in the 1975 publication were that exposure to the famine in the third trimester, but not preconceptionally or during early pregnancy, resulted in major (approximately 300 g, or 10%) reductions in mean birthweight, with somewhat smaller (2–3%) relative reductions in length and head circumference. No consistent reductions were found for gestational duration. The article by Stein et al. in this issue of the International Journal of Epidemiology2 focuses on the effect of the famine on body proportions at birth (ratios and other relationships between pairs of the three measurements of weight, length, and head circumference). The authors justify this focus based on the potential relevance of newborn body proportions for the fetal origins of adult disease (FOAD). Some of the early work in the FOAD area suggested that differences in severity or timing of acute maternal undernutrition could affect body proportions. Much of the speculation about this was based on Tanner's highly schematic graphs, which suggested deceleration of growth in length after 20 weeks of gestation.3 These graphs have been repeatedly cited by investigators interested in studying body proportions at birth, but studies based on prostaglandin or hysterotomy terminations show no evidence of deceleration of growth in length after 20 weeks in gestation.4,5 Crown-heel length appears to increase linearly until at least 35 or 36 weeks. Studies have clearly shown that maternal undernutrition and overnutrition have greater relative negative and positive effects, respectively, on fetal weight than they do on length or head circumference.6 Thus body proportions at birth, which are most commonly reported using Rohrer's ponderal index (birthweight in grams × 100/birth length3 in cm3), have been shown to be highly associated with the degree of growth restriction, as reflected by reduction in birthweight for gestational age. More severely growth-restricted infants tend to be disproportionately thin (‘asymmetric’), while mildly growth-restricted infants tend to be more proportional (‘symmetric’).6 Other than these associations with severity, variations in proportions among weight, length, and head circumference have not been convincingly shown to carry any unique prognostic information with respect to subsequent growth or health, including adult chronic disease. Simple geometric considerations suggest that reductions in linear measurements should be reflected in much larger (following a cubic function) reductions in volume (and thus weight). Even taking this into account, however, reductions in birthweight are generally larger than those in length or head circumference. Using a z-transformation to standardize other anthropometric measurements for weight, length and head circumference are progressively reduced with increasingly severe reductions in weight, but the reductions are of a lesser magnitude than for weight (i.e. length and head circumference are partially ‘spared’).6 In my view, perinatal epidemiologists have given insufficient attention to overweight at birth. Several countries have reported a consistent temporal trend toward increasing birthweights, particularly at term.7 Since high birthweight is associated with an increased risk of obesity in childhood and beyond,8,9 we should be concerned that this birthweight trend may be contributing to the current obesity epidemic. The alarming increase in prevalence of obesity in both children and adults seems largely attributable to reductions in physical activity (and, perhaps, increases in energy intake). But the temporal trend toward higher prepregnancy body mass index and weight gain during pregnancy appears to be at least partly responsible for the trend toward heavier babies.10 Epidemiological research, clinical practice, and public health policy relating to the long-term effects of fetal growth on adult chronic disease should perhaps focus more on the overnutrition of mothers and infants than on fetal proportionality associated with maternal undernutrition.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.