The number of women who exercise regularly has been increasing since the mid-1970s. Women may run, cycle, perform aerobics or lift weights on a daily basis. Many women participate in competitive activities such as team sports, races, or even marathons. Collegiate women's sports are very popular, and the creation of the Women's National Basketball Association (WNBA), with its first games held in May 1997, has supported and encouraged organized women's sports. The formation of the WNBA has helped draw attention to the lack of scientific data on the effect of exercise on women. No guidelines are available to ensure uniformity of recommendations to women before, during and after pregnancy. As the authors of the article in this issue point out,1 many women now view exercise as an integral, if not vital, part of their daily lives.
Physically active women who become pregnant invariably ask their physicians about the safety of exercise and its impact on pregnancy. Should they continue to exercise at their current level or make modifications to their program? Physicians need to be able to answer these questions for their patients. Moreover, physicians and pregnant women need to be assured that the answers are supported by solid research.
Although most earlier studies measured the effect of “moderate” exercise on pregnancy outcome, the definitions used for moderate exercise ranged from five minutes of exercise at a heart rate of 100 beats per minute to 50 or more minutes of exercise at a heart rate of over 150 beats per minute. Because the literature varies on this point, it is important that a standard definition be used for moderate exercise. The American College of Obstetricians and Gynecologists defines moderate exercise as the level of exercise at which cardiovascular conditioning can occur.2 The target heart rate range can be calculated by the following formula: (220 – age) × 60 to 80 percent. This target range needs to be maintained for 20 to 30 minutes for a minimum of three times a week.
Numerous case reports can be found of women who ran extensively throughout their pregnancies and who delivered healthy, normal birth weight infants. One example is Sue Olsen, ranked fourth nationally for 24-hour races. She ran a four-hour marathon when she was eight and one-half months pregnant. The following week she partly ran and partly walked a 24-hour race posting 62.9 miles.3 One week later she delivered a healthy infant. Although these reports lack comparison groups and are not generalizable, they lend support to the conclusion that healthy women with normal, uncomplicated pregnancies may exercise with few restrictions without adversely affecting their infants or themselves.
The most frequently measured pregnancy outcome is birth weight. Other measures are premature birth, miscarriage, complications during pregnancy or during labor and delivery, gestational age and Apgar score. Women with jobs requiring prolonged standing4 may be at increased risk for preterm delivery, but no evidence has associated miscarriage, pregnancy complications or premature delivery with exercise during pregnancy.5–7
The relationship between exercise and birth weight is not firmly established. One study8 looked at 800 women categorized into groups of sedentary, mild, moderate and previously conditioned moderate exercisers. The mild exercisers delivered infants with a mean birth weight approximately 100 g (3.6 oz) higher than the birth weight among infants of sedentary women, while the infants born to the moderate exercisers had an even higher mean birth weight. However, another study9 found birth weight to be 400 g (14.3 oz) lower in the infants of women who exercised throughout their pregnancy, compared with the infants born to women who stopped exercising early in pregnancy. Of note in the latter study was the heavy exercise performed throughout pregnancy by the subjects.
Results of these studies, although seemingly contradictory, suggest that moderate levels of exercise during pregnancy may be beneficial whereas heavy exercise may have a detrimental effect on birth weight. This conclusion is supported by a study10 in which a group of previously conditioned, moderate exercisers were compared with a group of nonexercisers. The study found that the mean birth weight of infants born to the exercisers decreased as the amount of weekly exercise increased. Women who exercised three times a week delivered infants with higher mean birth weights than those of the women in the control group, whereas the infants of women exercising four to seven times a week had lower mean birth weights than the infants of women in the control group.
However, studies have not shown a consistent reduction in birth weight with exercise. Sternfeld and colleagues11 looked at 388 women with levels of exercise classified as moderate, mild to moderate, mild, and sedentary. Neither mean birth weight nor gestational age was found to be related to the level of exercise. A meta-analysis12 concluded the mean birth weight of infants born to exercisers was not significantly different from the birth weight of infants born to sedentary women. An important limitation of these studies was the lack of randomization. Many did not report how complications in pregnancy were handled, and others were retrospective in design, relying on participants' recall of their amount of exercise. Additionally, many of the participants in the studies significantly reduced their level of activity in the last trimester. These factors could attenuate any possible impact on birth weight.
More recent large-scale studies have found results similar to those of the meta-analysis. Megann13 and colleagues evaluated 2,743 women grouped as sedentary and mild, moderate and heavy exercisers. Important differences were revealed between each group, including age, height, socioeconomic status, smoking practice, parity and race. Multivariate analyses showed that the apparent effects of exercise were due entirely to confounding variables and not to the level of activity. It is quite possible that groups in previous studies were not as closely matched as originally intended and, therefore, it is important to interpret with caution the results of any study that failed to account for these confounding variables.
In a large population-based case-control study,14 the relationship of exercise, employment and other daily activities with pregnancy outcome were examined. Mothers of infants with very low or normal birth weight were surveyed on these issues. The major finding of this study was that mothers of very low-birth-weight infants were much less likely to exercise during pregnancy than the mothers with other pregnancy outcomes. No significant increased risks were found between employment during pregnancy, types of work activities and conditions, or other daily activities and adverse pregnancy outcome. An even more recent study showed that healthy and well-conditioned women could participate in a moderate or high intensity exercise program during pregnancy without experiencing adverse fetal or maternal outcomes.15
Based on these studies and the meta-analysis of 18 observational and interventional studies,12 it is reasonable to conclude that mild to moderate exercise during pregnancy has no obvious, important adverse effects on the developing fetus. Although there are discrepancies among study results, these discrepancies are most likely explained by differences in study design and methods, data analysis, bias in self-selection of study groups and other confounding variables. Even so, the discrepancies may not be that clinically important. For example, among the studies in which birth weight was found to be lower among exercising women, infant weight was still within the range of normal birth weight and the risk of premature birth or miscarriage was never increased.
Despite abundant experimental data on the effects of exercise on pregnancy, our knowledge of the subject is still insufficient. For example, research demonstrates that the body makes significant adaptations to maintain fetal well-being during exercise, but it is not clear whether previously conditioned women who exercise throughout pregnancy have better adaptive mechanisms than other women. Could these previously conditioned women exercise at even higher performance levels without causing harm to the fetus? It is worth mentioning that many of the body's adaptive mechanisms described above are dependent on exercise intensity or duration, or both, and it is unclear how the body adapts to more intense or prolonged bouts of exercise.
Research on physiologic adaptations and pregnancy outcomes does not adequately address the issue of more prolonged or strenuous exercise. One possible way to study the effects of strenuous exercise on pregnancy outcome would be a randomized clinical trial. Women who intend to continue a heavy exercise regimen could be recruited, and after an explanation of the study, the women would be randomized to either continue a heavy exercise routine or start a routine that was reduced to a moderate level. The effects of moderate and heavy exercise on birth weight could be compared. The daily caloric intake and monthly maternal weight gains of the two groups should be tracked because it is unclear whether heavy exercisers gain similar amounts of weight during pregnancy as do moderate or nonexercisers. Differences in maternal weight gain affect birth weight of the newborn.
It is unlikely that we will see studies measuring the absolute limits to which pregnant women can exercise because of the potential adverse effects that could be imposed on the fetus. However, it is important to continue to clarify safe limits. Thus, future studies should examine previous physical conditioning, gestational stage, activity type and moderate versus heavy levels of exercise.
Once more information is known about the effects of physical conditioning on maternal-fetal adaptations and the effects of strenuous exercise on pregnancy outcome, guidelines can be revised to reflect these new understandings. Revised guidelines could include more specific information on the levels and types of exercise best suited for each trimester, better information on ways to avoid overheating, and nutritional guidelines before, during and after more prolonged bouts of exercise.