Ideal pregnancy length: an unsolved mystery

By Anne Marie Z. Jukic, Donna Baird, Clarice Weinberg, and Allen Wilcox

Pregnancy begins with conception — an event that is practically invisible. Since we can’t measure the beginning of pregnancy, it’s hard to know how far along a woman is in her pregnancy. We guess the beginning of pregnancy either from the woman’s report of her last menstrual period or from fetal size on ultrasound, both of which have errors. Everyone knows that it’s hard to predict when women will deliver, but we haven’t known how much of that uncertainty was due to errors in measurement of gestational age. If somehow we had a precise measure of pregnancy onset, wouldn’t pregnancy length be much less variable? No one had been able to look at the variability of pregnancy length using a precise measure of conception. Our study, which used daily hormone measures to detect ovulation, provides that precise measure because the egg only survives a matter of hours after ovulation if it is not fertilized. What we found, to our surprise, was that the true length of pregnancy is still quite variable, ranging over five weeks in a group of healthy women (from 35.3 to 40.6 weeks from ovulation).

While our study was small (125 women with no fertility problems who conceived naturally), the implications are intriguing. If healthy pregnancies can vary in length by five weeks, then an “ideal” gestation for one woman might be 38 weeks, while for another woman it might be 42 weeks. How can we distinguish the woman who would ideally deliver at 40 weeks, but has gone to 42, from the woman who would ideally deliver at 42 weeks? These fundamental questions currently have no answer. We do have one small clue: we found that the average length of pregnancy in a woman’s other births was correlated with the length of her study pregnancy. Thus, variability in pregnancy length was not just random; women tended to follow their own individualized clock for pregnancy length. This is not, however, a perfect measure of the “healthy pregnancy length” for a given woman since pathological conditions could be affecting the lengths of all her pregnancies. Thus, the elusive “ideal” duration of a given pregnancy remains obscure.
As research progresses, we may be better able to partition the variability in pregnancy length into its component parts. For example, predicting a woman’s “due date” might be improved by incorporating additional information (mother’s age and body weight, previous pregnancy lengths, and so on). But to do this, we need data from larger studies that develop predictive models for due dates.

Finally, early pregnancy events may provide useful information about the health of a developing fetus. While early pregnancy is difficult to observe in natural conceptions, it can be done in a research setting. For example, the timing of implantation can be defined by the first appearance of human chorionic gonadotropin (hCG) in the maternal urine. Early hCG is a signal to the corpus luteum on the mother’s ovary to continue producing progesterone in support of the developing pregnancy. This process of embryonic communication with the mother is referred to as “corpus luteum rescue”. Further research on timing of implantation and on corpus luteum rescue may help us understand the biological pathways that influence both the baby’s development and pregnancy length. Given how little is known about early pregnancy biology, further investigation may yield a host of insights into pregnancy development. A better understanding of early pregnancy biology may lead to better estimates of the right time for delivery, and perhaps, the early detection of pregnancy complications.

Dr Anne Marie Z. Jukic is a Postdoctoral Fellow in the Reproductive Epidemiology Group at the US National Institute of Environmental Health Sciences. Dr Donna D. Baird is Principal Investigator, Epidemiology, in the Women’s Health Group at the NIEHS. Dr Clarice R. Weinberg is Chief of the Biostatistics Branch of the NIEHS. Dr Allen J. Wilcox is the Principal Investigator of Environmental Toxins and Human Reproduction at the NIEHS. They are the authors of the paper ‘Length of human pregnancy and contributors to its natural variation’, which was recently published in the journal Human Reproduction.

Human Reproduction features full-length, peer-reviewed papers reporting original research, clinical case histories, as well as opinions and debates on topical issues.

Subscribe to the OUPblog via email or RSS.
Subscribe to only health and medicine articles on the OUPblog via email or RSS.
Image credit: Pregnancy women. By Chepko, via iStockphoto.