A Periconceptional Nutritional Origin for Noninfectious Preterm Birth

See allHide authors and affiliations

Science  25 Apr 2003:
Vol. 300, Issue 5619, pp. 606
DOI: 10.1126/science.1080803

The incidence of preterm birth is increasing and remains the major cause of perinatal morbidity and mortality. While approximately 40% of preterm birth is inflammatory in origin (1), the cause of idiopathic preterm birth is unknown, and the lack of suitable experimental paradigms has limited research. Elevated corticotropin releasing hormone levels in mid-pregnancy in mothers who subsequently enter preterm labor (PTL) suggest that idiopathic PTL has its origin well before the peripartum period and involves alterations in the feto-placental endocrine system (2). Preliminary evidence suggests that severe maternal undernutrition (3) or stress (4) early in human pregnancy may result in preterm birth. However, there is little experimental evidence to support this.

In all species studied, there is a surge in circulating fetal cortisol concentrations before birth. This cortisol surge is critical for the maturation of many organ systems in preparation for extrauterine life and, in sheep, is essential for the initiation of normal parturition (5). Here, we demonstrate in sheep that moderate maternal undernutrition around the time of conception results in a precocious fetal cortisol surge and preterm birth.

Singleton-bearing ewes were either well fed (ad lib, n = 8) or undernourished (UN) to reduce maternal weight by 15% (n = 10) from 60 days before until 30 days after conception (term = 145 days), with ad libitum intake thereafter (fig. S1). Fetal cortisol and adrenocorticotropin (ACTH) concentrations were measured frequently in late gestation, and ewes were allowed to lamb normally (6).

Fetuses from UN ewes were delivered earlier than ad lib controls (mean gestation length 139 versus 146 days, P < 0.05) (Fig. 1A). Fetal plasma cortisol concentrations also increased earlier in fetuses from UN ewes (P < 0.0001) (Fig. 1B). In half of these, there was a precocious rise in cortisol and the fetuses were delivered preterm (>2 SD below the mean for this flock) (Fig. 1A). Fetal ACTH concentrations were higher in all UN fetuses, regardless of timing of birth (P < 0.01) (Fig. 1C).

Fig. 1.

(A) Kaplan-Meier survival analysis of gestation length. Solid line, UN; dashed line, ad lib. (B) Cortisol and (C) ACTH concentrations in fetal plasma from 125 days gestation until delivery [ad lib (▢), UN delivering at term (⚫), individual UN fetuses delivering preterm (×, ♦, ▾, △, ◯)]. Values are mean ± SE. Ad lib, n = 8; UN, n = 10. ***P < 0.0001, **P < 0.01, *P < 0.05 (UN versus ad lib).

We excluded infection as a cause of delivery by measuring maternal and fetal circulating concentrations of interleukin-6 (IL-6) and tumor necrosis factor–α (TNF-α) (6). IL-6 concentrations were not elevated in any fetus [UN, 0.03 (0.01–1.04) ng/ml; ad lib, 0.03 (0.01–0.14) ng/ml; (median (range)], and maternal IL-6 and fetal and maternal TNF-α concentrations were undetectable.

Our finding that a modest reduction in food intake commencing before conception and continuing for only 30 days thereafter induces premature delivery in sheep may have major implications. Fetal nutrient requirements are extremely small at 30 days gestation, and thus maternal undernutrition is unlikely to have limited nutrient availability for fetal growth. Consistent with this, fetuses of both groups were of similar size. Rather, the periconceptional nutritional insult led to accelerated fetal hypothalamic-pituitary-adrenal axis (HPAA) maturation. Although a precocious cortisol surge occurred in only half of the UN fetuses, ACTH concentrations were increased in all fetuses, and the UN group was delivered early across the whole gestational age spectrum. We hypothesize that maturation of the fetal HPAA must reach a threshold level to initiate the “feed forward” loop, when pituitary ACTH production increases concomitantly with increasing cortisol concentrations, which eventually leads to initiation of labor, and that this threshold was reached only in half the UN group. It is of interest that only 50% of women who enter PTL actually deliver prematurely (7) and that gestation length is shortened in women with poor nutritional reserves at the beginning of pregnancy (3).

A modest nutritional restriction around the time of conception, therefore, may have far-reaching consequences for the fetus, with accelerated maturation of the fetal adrenal gland resulting in preterm birth. If these findings are applicable to human pregnancy, then a focus on events around the time of conception may hold the key to prevention of one of the major causes of preterm birth.

Supporting Online Material

Materials and Methods

Fig. S1

References and Notes

Navigate This Article