What Mechanisms Account for the Observed Reductions in GDM and Preeclampsia Risk?
Results from clinical studies suggest that hypertension, chronic systemic inflammation, dyslipidemia, and oxidative stress are common pathophysiologies of GDM and preeclampsia, respectively. Taken together with evidence from observational studies, controlled clinical trials, and animal studies, this suggests that physical activity may impact the occurrence of GDM and preeclampsia through a number of overlapping and independent biological pathways (Fig. 3). For instance, evidence from studies primarily conducted in men and nonpregnant women indicate that physiological effects of physical activity include improved insulin sensitivity, reduced blood pressure, decreased concentrations of proinflammatory cytokines in peripheral circulation, reduced oxidative stress, and improved plasma lipid and lipoprotein concentrations. Next, we provide a brief overview of some of literature that describes the relationship between physical activity and concentrations of biological markers indicative of GDM and/or preeclampsia risk.
During pregnancy, the increased fetal demand for glucose, the major substrate for growth and development, produces a shift in maternal metabolism toward greater fat production and storage, resulting in decreased glucose utilization. Moderate- and vigorous-intensity physical activity is associated with improved insulin sensitivity and reductions in fat mass (4,6). Aerobic dance and walking, the most popular forms of exercise among pregnant and nonpregnant women (12), have been shown to result in a reduction in plasma insulin even during pregnancy (4). Importantly, carefully designed and conducted metabolic studies suggest that pregnancy may be characterized by a set of compensatory mechanisms that attenuate acute adverse influences exercise may have in glucose handling. Clapp and Capeless reported that the typical exercise-induced hyperglycemia observed in nonpregnant subjects appears to be reversed in healthy pregnant women who regularly engage in recreational physical activities. Although the authors did not measure maternal hepatic lipase synthesis, they attributed the variation in exercise-induced response to a pregnancy-associated decrease in hepatic glucose synthesis and an increase in fractional glucose use by maternal muscle during exercise (4). Overall, these data provide evidence to support the biological plausibility of using recreational physical activity as a disease prevention modality in pregnant women.
Preeclampsia, though characterized clinically by maternal high blood pressure and proteinuria, is also characterized by subclinical metabolic disorders such as hypertriglyceridemia, excessive lipid peroxidation or oxidative stress, insulin resistance, systemic chronic inflammation, and elevated plasma homocysteine (2,14). Physical activity profoundly affects many of these physiological functions.
Results from observational studies and from randomized trials have consistently demonstrated an inverse relation between recreational physical activity and blood pressure in nonpregnant and pregnant women. For instance, results from randomized controlled trials indicate that aerobic exercise performed by nonpregnant women aged 18 yrs or older results in statistically significant reductions in resting systolic and diastolic blood pressures. Exercise therapy is also associated with reductions in diastolic blood pressures in pregnant women with a history of mild hypertension, gestational hypertension, or family history of hypertensive disorder.
Recreational physical activity is also associated with improvements in lipid concentrations in men and nonpregnant women, specifically, reduced plasma triglycerides and increased high-density lipoproteins. Every aspect of lipid metabolism is dramatically altered during pregnancy. For instance, maternal serum or plasma cholesterol and triglyceride concentrations increase 1.5- and threefold, respectively, above nonpregnant levels by the midthird trimester (2,15). Pregnancy-associated hyperlipidemia is further exaggerated in women with preeclampsia and gestational diabetes (2,15). In 2004, Butler et al. reported that mean triglyceride concentrations were lower (−23.6 mg·dL−1) among women in the highest tertile (>12 h·wk−1) of time performing physical activity as compared with inactive women. Reductions in mean total cholesterol were also observed among women in the highest levels of time performing physical activity, energy expenditure, and maximal intensity of maternal exercise. Linear relationships were observed across levels of physical activity measures for triglyceride and total cholesterol (Fig. 4). There was no association between physical activity and high density lipoprotein (HDL) cholesterol (2). These data suggest that habitual physical activity performed during pregnancy may mitigate the pregnancy-associated dyslipidemia commonly noted in hypertensive and diabetic pregnancies.
Moderate-intensity physical activity results in decreased concentrations of proinflammatory cytokines and C-reactive protein in peripheral circulation. Clapp and Kiess reported recently that regular weight-bearing exercise during pregnancy influences alterations in plasma tumor necrosis factor-α during pregnancy (5). Women randomized to the physical activity group, as compared with inactive women, experienced greater attenuation of the proinflammatory cytokine concentrations during pregnancy (5). Because late pregnancy is characterized by an increase in insulin resistance and elevated circulating tumor necrosis factor-α levels, regular weight-bearing exercise during pregnancy may moderate insulin resistance.
Strenuous physical activity has also been noted to increase the risk of oxidative stress by some but not all investigators (14). Evidence suggests that physical activity also results in increased antioxidant enzyme activity; this increase compensates for any possible increase in oxidative stress that may be attributable to vigorous physical activity. Given the central role of oxidative stress in the pathophysiology of preeclampsia (11), more studies are needed to clarify the relation between maternal and fetal enzymatic and nonenzymatic antioxidant response to physical activity during pregnancy.
Lastly, recreational physical activity has been linked to improvements in emotional well-being and reductions in stress and anxiety (14). Investigators have shown that pregnant women experiencing anxiety and/or depression are at a threefold increased risk for preeclampsia (14). It is reasonable to speculate that the risk of preeclampsia, particularly those cases attributable to maternal psychosocial stress, may be impacted by the psychophysiological benefits of regular physical activity.
Are There Adverse Effects of Exercise in Pregnant Women?
As noted, historically, pregnant women were persuaded to take a “pregnant pause” in terms of participation in recreational physical activity. We offer some insights into why scholars and clinicians may have adopted the stance of discouraging physical activity during pregnancy during the early decades of the 20th century.
Results from animal studies published before the 1970s raised concern that physical activity during pregnancy could cause decreased uterine blood flow, fetal bradycardia, and poor outcomes such as preterm delivery and reduced fetal growth. Research findings, published since the 1970s, however, provide reason to believe that most of the previous findings in animals cannot be directly applied to humans. Exercise stimuli used in animals were often not exercise per se, but rather thermal, nutritional, or biophysical stressors. Moreover, the physiological impact of stimuli used in these studies did not simulate physiologically relevant levels of physical stress likely to be experienced by healthy, well-nourished pregnant women engaged in recreational exercise. As recently reviewed by several investigators (1,6), pregnancy is known to be characterized by several important compensatory mechanisms that offset adverse effects of exercise-induced changes in uteroplacental blood flow, fetal bradycardia, maternofetal thermoregulatory control, and musculoskeletal function, among others. We provide a brief summary of evidence from clinical and epidemiological studies that assess hypothesized adverse effects of physical activity during pregnancy.
Decrease in uteroplacental blood flow secondary to maternal participation in physical activity has been a topic of many reports in the obstetric literature. These studies were in part motivated by concerns regarding exercise-induced diversions of maternal blood flow away from the uteroplacental compartment to exercising skeletal muscles and skin. Researchers have hypothesized that such diversions may induce placental hypoxemia and compromise fetal growth by depriving the fetus of oxygen and nutrients. These concerns were reinforced somewhat by data that indicated that moderate- to high-intensity physical activity can result in a 50–80% reduction in uteroplacental blood flow. Results from studies involving healthy pregnant women who engage in recreational physical activity, however, have not supported these early concerns. Overall, maternal participation in recreational physical activity has not been associated with reduced in infant birth weight.
Concerns about exercise-induced alterations in uteroplacental blood flow also led some investigators to predict an increased incidence of transient fetal bradycardia secondary to reduced availability of oxygen to the fetus during maternal exercise. Fetal heart rate has been reported to increase, decrease, and remain unchanged during and after maternal exercise. Importantly, fetal heart rates have been shown to increase during maternal participation in activities characterized by high levels of exertion. Data from studies that evaluated fetal outcomes at birth indicate that increased maternal physiological fitness caused by sustained exercise during pregnancy is associated with similar or better outcomes than those observed among inactive and presumably less physically fit mothers.
Concerns about exercise-induced uterine irritability resulting in premature uterine contractions and preterm delivery have also been expressed. Investigators once reasoned that maternal physical activity results in a metabolic milieu consistent with an increase in the synthesis and release of prostaglandins and norepinephrine, potent uterine stimulants capable of inducing premature contractions. A review of relevant studies has failed to support early concerns. Results from more rigorously designed and executed epidemiological studies are consistent with a reduction in risk of preterm delivery among women engaged in recreational physical activity as compared with inactive women.
Exercise-induced increases in body temperature and evidence of the teratogenic effects of hyperthermia have led some investigators to speculate that maternal exercise during pregnancy may induce adverse fetal outcomes via fetal hyperthermia. Available data, however, suggest that maternal heat dissipation is enhanced during pregnancy. Clapp et al. noted that the increase in rectal temperature during exercise performed during the third trimester was statistically significantly lower than when the same exercise was performed during the period before conception (6).
Several authors have hypothesized that the combination of pregnancy-induced joint laxity and recreational physical activity could increase the risk of musculoskeletal injuries such as sprains and strains (1). In early pregnancy, relaxin secretion increases to soften the cartilage in the pelvis to accommodate delivery; during this process, other joints become lax as well. Changes in the center of gravity, weight gain, other alterations in balance, and coordination during pregnancy may also potentially augment the chance of injury during exercise. Researchers should evaluate the determinants, frequency, types, and adverse fetomaternal outcomes associated with exercise-related injuries experienced during pregnancy. Results from such studies will contribute to developing maternal physical activity regimens and adopting guidelines for the treatment and prevention of injuries.
Occupational physical activity during pregnancy deserves special mention here because available evidence suggests that the effects may be unlike those of recreational physical activity (3). Working during pregnancy per se is not a risk factor for adverse outcomes, although excessive stair climbing, standing, and bending have been associated with low birth weight, preterm delivery, and spontaneous abortion. It is possible that standing for long periods of time may disrupt uteroplacental flow, alter hormonal balance, and increase intraabdominal pressure. These physiological changes may contribute to adverse maternal and fetal outcomes as premature labor and restricted fetal growth.
What Are the Current Gaps in Knowledge?
Despite increased attention to physical activity in pregnancy over the past two decades, much remains unknown about the long- and short-term fetomaternal effects of exercise during pregnancy. By necessity, much attention has been paid to testing hypotheses motivated by theoretical concerns of adverse fetal outcomes associated with maternal exercise. Consequently, evidence amassed from studies that have followed healthy pregnant women does not support these early concerns. We believe the time is right to motivate a paradigm shift towards identifying maternal and fetal benefits of exercise in pregnancy.
Future investigations should be designed to study in greater detail the types, frequency, duration, and dose-response patterns of recreational physical activity in relation to pregnancy-related disorders. Studies should also assess the joint and potentially divergent effects of occupational physical activity, as well as activity associated with daily living (e.g., transportation and child care) on pregnancy outcomes. Carefully designed controlled metabolic studies are required to evaluate the effect of physical activity on endocrinological, hemodynamic, respiratory, immunological, and uteroplacental physiology during pregnancy.
Questions remain unanswered concerning contraindications for recreational physical activity during pregnancy. For instance, although physical activity is often prescribed as a nonpharmacological therapy for mild hypertension in nonpregnant individuals, in pregnancy women presenting with mild hypertension may be directed to go on bed-rest. Should women with high-normal or slightly elevated blood pressures in early pregnancy be encouraged to begin a supervised moderate-intensity physical activity regimen? Prospective studies, conducted in diverse populations, are needed to address this question.
Lastly, as noted, there are virtually no data specific to the epidemiology of sports- or exercise-related injury among active pregnant women. We believe that now, more than ever, there is a clear need to design and conduct surveys of pregnant women as a first step toward better understanding the distribution and determinants of musculoskeletal injury likely to be experienced during pregnancy. Knowledge gained from such studies will provide insights toward preventing sports-related injuries among expectant mothers.
The current literature does not support concerns that once promoted many to think about pregnancy as a period of confinement. It is encouraging that the ACOG and other similar organizations now publish guidelines that more assertively promote physical activity during pregnancy. Additional research is needed to increase the precision with which exercise programs may be designed for all groups of pregnant women (e.g., those with pregestational diabetes, essential hypertension, or those who are obese). We recommend that investigators now shift their efforts toward documenting the epidemiology of exercise-related injuries likely to be experienced by active reproductive-aged and pregnant women. We also support a research agenda that aggressively and rigorously tests hypotheses concerning health benefits of exercise during pregnancy.
In the mean time, we know that for most women the risks of exercise during pregnancy are far outweighed by confirmed health benefits. We also suspect that women who maintain their exercise habits during pregnancy may set a pattern of behavior that will confer health benefits well beyond their reproductive years. Hence, with proper medical evaluation and approval, women need not take a “pregnant pause” from their active lifestyle.
A more complete list of references than that provided here can be obtained from Dr. Michelle A. Williams, Center for Perinatal Studies, Swedish Medical Center, 747 Broadway (Suite 4 North), Seattle, Washington, 98122. E-mail: Mwilliam@u.washington.edu.
1. Artal, R., and M. O’Toole. Guidelines of the American College of Obstetricians and Gynecologists for exercise
and the postpartum period. Br. J. Sports Med
. 37:6–12, 2003.
2. Butler, C.L., M.A. Williams, T.K. Sorensen, I.O. Frederick, and W.M. Leisenring. Relationship between maternal recreational physical activity
and plasma lipids in early pregnancy
. Am. J. Epidemiol
. 160:350–359, 2004.
3. Clapp, J.F. III. Pregnancy
outcome: physical activities inside versus outside the workplace. Semin. Perinatol
. 20:70–76, 1996.
4. Clapp, J.F. III, and E.L. Capeless. The changing glycemic response to exercise
. Am. J. Obstet. Gynecol
. 165:1678–1683, 1991.
5. Clapp, J.F. III, and W Kiess. Effects of pregnancy
on concentrations of the metabolic markers tumor necrosis factor α and leptin. Am. J. Obstet. Gynecol
. 182:300–306, 2000.
6. Clapp, J.F. III, R. Rokey, J.L. Treadway, M.W. Carpenter, R.J. Artal, and C. Warrnes. Exercise
. Med. Sci. Sports Exerc
. 24 Supp 6:S294-S300, 1992.
7. Dempsey, J.C., T.K. Sorensen, I-M. Lee, M.L. Thompson, R.S. Miller, I.O. Frederick, and M.A. Williams. A case-control study of maternal recreational physical activity
and risk of gestational diabetes mellitus
. Diabetes Res. Clin. Practice
8. Dempsey, J.C., T.K. Sorensen, M.A. Williams, I-M. Lee, R.S. Miller, E.E. Dashow, and D.A. Luthy. Prospective study of gestational diabetes mellitus
in relation to maternal recreational physical activity
before and during pregnancy
. Am. J. Epidemiol
. 159:663–670, 2004.
9. Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General
. National Cancer Center for Chronic Disease Prevention and Health Promotion, Atlanta, Georgia, United States, 1996.
10. Marcoux, S., J. Brisson, and J. Fabia. The effect of leisure time physical activity
on the risk of preeclampsia
and gestational hypertension. J. Epidemiol. Community Health
. 43:147–152, 1989.
11. National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy
: Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy
. Am. J. Obstet. Gynecol
. 183:S1-S22, 2000.
12. Ning, Y., M.A. Williams, J.C. Dempsey, T.K. Sorensen, I.O. Frederick, and D.A. Luthy. Correlates of recreational physical activity
in early pregnancy
. J. Matern. Fetal Neonatal. Med
. 13:385–393, 2003.
13. Saftlas, A.F., N. Logsden-Sackett, W. Wang, R. Woolson, and M.B. Bracken. Work, leisure-time physical activity
, and risk of preeclampsia
and gestational hypertension. Am. J. Epidemiol
. 160:758–765, 2004.
14. Sorensen, T.K., M.A. Williams, I-M. Lee, E.E. Dashow, M.L. Thompson, and DA Luthy. Recreational physical activity
and risk of preeclampsia
15. Williams, M.A., and R. Mittendorf. Maternal Morbidity. In: Women & Health
, edited by M.B. Goldman and M. Hatch. New York: Academic Press Inc., 2000, p. 172–181.
Keywords:©2005 The American College of Sports Medicine
exercise; physical activity; pregnancy; pregnancy-related disorders; gestational diabetes mellitus; preeclampsia