The guidelines from the American College of Obstetricians and Gynecologists (ACOG) recommend that pregnant women can continue to exercise in the absence of either obstetric or medical complications (3). The current recommendation is that pregnant women should engage in 30 min or more of moderate exercise several or all days of the week. Participation in a wide range of recreational activities appears to be safe; participation in sports with high contact or with an increased risk of falling should be avoided (3). Thus, physical activity is recommended for pregnant women to maintain the health benefits of exercise such as cardiovascular fitness; however, safety is an issue.
Fitness is measured by maximal exercise testing with or without simultaneous measures of oxygen consumption. Maximal exercise testing in pregnant women may not be safe because it requires increasing heart rate to high levels. Nevertheless, a few studies have measured changes in maximal oxygen consumption (V̇O2max) during pregnancy (10,18,19). An alternative to maximal testing is to predict V̇O2max from submaximal testing. This approach has been used in several studies examining the changes in fitness during pregnancy (8,16,21). Because pregnancy-induced changes in body composition and physiology may result in deconditioning, measuring fitness before pregnancy and postpartum will assess the net effect of pregnancy on maternal fitness. In addition to aerobic fitness, strength is an important component to physical health. No study to our knowledge has examined changes in strength in women before or after pregnancy. Although it is recognized that fitness and strength are associated with weight, it is not known whether pregnancy-related changes in these parameters through a reproductive cycle differ by maternal body mass index (BMI).
The present study was designed to measure pregnancy-related changes in physical activity, fitness, and strength in women with low, normal, and high prepregnancy BMI.
MATERIALS AND METHODS
Design and protocol.
A longitudinal study examining changes in physical activity, fitness, and strength before and after pregnancy was undertaken. Women were evaluated before pregnancy and at 6 and 27 wk postpartum. Measurements of body composition, physical activity, fitness, and strength were completed during visits to the USDA/ARS Children’s Nutrition Research Center (CNRC) and Texas Children’s Hospital.
Healthy adult women (N = 124) were recruited from the local Houston area. Enrollment criteria included nonsmoking, ages 18–40 yr, parity not greater than 4, moderately active, and no chronic medications or alcohol/drug abuse. At enrollment, the women were nonanemic, normoglycemic, and euthyroidic. The women were classified into groups based on low, normal, and high BMI, calculated as weight divided by height squared (kg·m−2), defined as low BMI (<19.8 kg·m−2), normal BMI (19.8–26.0 kg·m−2), or high BMI (>28.6 kg·m−2). During the course of the study, 76 women became pregnant and 63 women delivered term, singleton infants with birth weights greater than 2.5 kg. Twelve women were dropped from the study due to obstetric complications: three sets of twins, one set of triplets, five preterm deliveries, two miscarriages, one preeclampsia, and one woman moved away from Houston. V̇O2max (1884 vs 2043 mL·min−1; P = 0.02) and V̇O2max per kilogram of body weight (29 vs 34 mL·kg−1·min−1; P = 0.001) were lower in the women who did not become pregnant. No significant differences were seen in total physical activity or strength.
Ethnicity was established by self-report. All women provided written informed consent to participate in this study, which was approved by the institutional review board for human subject research for Baylor College of Medicine and affiliated hospitals.
Body weight was measured to the nearest 0.1 kg using a digital balance (Healthometer, Bridgeview, IL), and height was measured to the nearest centimeter using a stadiometer (Holtain Ltd., Crymmych, UK). Body composition was then assessed by a multicomponent model involving total body water (TBW), body density by hydrodensitometry, and bone mineral content (BMC) by dual-energy x-ray absorptiometry (DXA). TBW was determined by dilution of an orally administered dose of deuterium oxide (40 or 100 mg 2H2O·kg−1) (Cambridge Isotope Laboratories, Andover, MA). At 0 wk gestation and at 27 wk postpartum, TBW was estimated by extrapolation to zero-time intercept from samples collected daily for 13 d, as part of the doubly labeled water (DLW) method. At 6 wk postpartum, TBW was estimated by the plateau method from samples collected at 4–6 h postdose. Saliva samples were stored frozen at −20°C in O-ring sealed vials until analyzed for hydrogen isotope ratio measurements by gas-isotope-ratio mass spectrometry (22). Deuterium dilution space (NH) was converted to TBW by dividing by 1.04. Body density (Db) was measured with an underwater weighing system utilizing “force cube” transducers (Precision Biomedical Systems, Inc., State College, PA) (2). Body volume was corrected for residual lung volume measured by the simplified nitrogen washout method (20). DXA (QDR2000, Hologic, Inc., Madison, WI; software version 5.56) was used to measure total BMC. A four-component body composition model using body weight, TBW from 2H dilution, body volume from densitometry, and BMC from DXA was used to compute fat mass (FM) and fat-free mass (FFM) (9).
Physical activity was assessed by a version of the Taylor Questionnaire for the Assessment of Leisure Time Physical Activities (1), modified to include both leisure and occupational activities. Given a list of activities, the women recalled the duration (number of minutes per occasion) and frequency (number of times per month) of each specified activity during the last month. The activities were categorized into walking, dancing, bicycling, conditioning exercises, water activities, winter activities, sports, lawn/garden activities, home activities, fishing/hunting, and occupational activities according to Ainsworth (1). The time in hours per month was multiplied by the intensity codes or METs as defined by Ainsworth et al. for each specific activity and summed to provide an estimate of total monthly METs.
Fitness was measured by a maximal V̇O2 test on a cycle ergometer (Corival 400, Lode B.V., Gronigen, The Netherlands). The exercise protocol involved a constant power output (50 W) for 4 min. The average of minutes 3–4 constituted the steady state. Power was then increased every minute thereafter by 25 W. When the subject neared exhaustion, power was increased by 15 W for 1 min. If the subject was able to continue further, power was increased by 10 W for an additional minute. V̇O2, heart rate, ventilation (V̇E), and respiratory exchange ratio (RER) were measured. A metabolic cart (Model 2900, SensorMedics Inc., Yorba Linda, CA) was used to collect the respiratory gases for indirect calorimetry measures with values given at 20-s intervals. The metabolic cart was calibrated before each test session. Heart rate was measured by a POLAR heart rate monitor. Criteria for maximal exercise included RER >1.1 and fatigue, that is, unable to continue cycling at a specified cadence with an increase in workload.
Before any strength assessment, the subject was allowed to become familiar with the equipment and exercise techniques. Participants were given instructions on each of the machines and watched for correct completion of the exercise with no resistance. Strength was assessed on the leg press, leg extension, bench press, and latissimus pull down by the one-repetition maximum (1-RM) test, defined as the maximum amount of weight that could be lifted successfully one time. Starting with a low to moderate weight, the subjects attempted lifts with gradually increased weights (∼10% at first, decreasing to 5 and 2.5% as difficulty became evident). Successive attempts were made with a 90-s rest period between attempts until failure occurred. Approximately three to five trials were needed to reach the 1-RM.
Data are presented as means ± SD. Microsoft Access for Windows 95 (Version 7.0) was used for database management. Data description and statistical analyses were performed using Minitab for Windows (Version 12.2, State College, PA). Repeated measures analysis of variance (BMDP5V) was used to test for effects of BMI groups and time; the model included a grouping factor (low, normal, or high BMI), a time factor (0 wk of gestation or 6 and 27 wk postpartum), and interactions between BMI group and time. Post hoc pairwise comparisons between BMI groups or time intervals were performed using the Tukey method. Significance was set at P < 0.05.
The sample size (N = 63) included 49 Caucasian, 6 African American, 2 Hispanic, and 2 Asian women. Age, weight, height, BMI, and body composition data are presented for the women at baseline (Table 1). Longitudinal changes in body composition are presented elsewhere (4).
Physical activity measures by self-report are presented in Table 2 at prepregnancy (0 wk), and 6 and 27 wk postpartum. No significant BMI group by time interactions were observed. The time engaged in conditioning exercise was significantly higher in the low-BMI group compared with the high-BMI group. Total monthly METs did not differ significantly between the prepartum and postpartum time intervals. However, significant time effects were observed for specific activities, namely, walking, conditioning exercises, water activities, sports, occupational activities, and home activities (Table 2), with lower values observed postpartum. Other activities on the questionnaire (data not shown) included dancing, bicycling, winter activities, gardening, fishing, and sleeping. Only time engaged in dancing and bicycling were significantly lower at 6 wk postpartum compared with baseline (P < 0.03, data not shown).
Submaximal (steady-state) and maximal heart rate (HR), ventilation (V̇E), V̇O2, and RER are shown in Table 3. During submaximal exercise, significant group by time interactions were observed for V̇E and V̇O2. Submaximal V̇E was lower in the low-BMI group than the high-BMI group at 0 wk (P = 0.02) and at 27 wk (P = 0.03) postpartum. Also, submaximal V̇E was significantly lower in the normal-BMI than the high-BMI group at 0 and 27 wk postpartum (both P < 0.001). Submaximal V̇O2 (adjusted for weight) was significantly different between the normal- and high-BMI groups at 27 wk postpartum (P = 0.015). Submaximal V̇O2 (adjusted for FFM) was significantly different between the low- and normal-BMI groups compared with the high-BMI group at 0, 6, and 27 wk postpartum (P < 0.02). RER was lower at baseline than at 6 or 27 wk postpartum (both P < 0.0001). No group differences were observed for HR or RER during submaximal exercise.
Maximal workload, heart rate, respiration rate, V̇E, V̇O2, and RER changed over time. Compared with prepregnancy values, maximal cycle workloads were on average 27 W lower at 6 wk postpartum (P < 0.001) and 18 W lower at 27 wk postpartum (P < 0.05). Maximal HR was ∼7 bpm lower 6 wk postpartum compared with baseline (P < 0.01). Maximal ventilation was lower 6 wk postpartum compared with baseline (P < 0.0001), and was higher 27 wk postpartum compared with 6 wk postpartum (P < 0.05). Maximal respiration rate was significantly different between 0 and 6 wk postpartum (P < 0.02). Maximal V̇O2, adjusted for weight, dropped by ∼385 mL·min−1 from 0 to 6 wk postpartum (P < 0.0001) and by ∼234 mL·min−1 from 0 to 27 wk postpartum (P < 0.01). Maximal RER was also different between 0 and 6 wk postpartum (P < 0.01) and 27 wk postpartum (P < 0.05). The only significant group effects were observed for V̇O2, such that the high-BMI group had a lower V̇O2 (adjusted for weight or FFM) than the normal-BMI group at 0 and 27 wk postpartum (P < 0.05). The decline in fitness (V̇O2max) at 6 wk postpartum was positively correlated (r = 0.31, P = 0.03) with the decrease in conditioning exercises and negatively correlated with the time spent walking (r = −0.45, P = 0.001). No significant correlations between changes in fitness and activities were observed at 27 wk postpartum.
The women were classified into fitness categories (low, fair, average, good, and high), as shown in Table 4. The number of women classified into the low and fair fitness categories increased from 0 to 6 wk postpartum, whereas the number of women in the average, good, and high decreased from 0 to 6 wk postpartum.
Performance in leg press, leg extension, bench press, and latissimus pull down tests is described in Table 5. Significant interactions for BMI group by time were observed for leg press and leg extension, with no significant interactions observed for the upper-body exercises. In general, strength values for the leg and arm exercises decreased from baseline to postpartum. The leg press dropped by 24% from 0 to 6 wk postpartum (P < 0.02). The 6-wk postpartum values then increased by 44% by 27 wk postpartum values (P < 0.05). Leg extension dropped 4% from 0 to 6 wk postpartum and then increased 12% by 27 wk postpartum (P < 0.05). For the latissimus pull down, strength decreased 8% by 6 wk postpartum (P < 0.01). The latissimus pull down strength values then rose again 12% by 27 wk postpartum, which were significantly higher than the 6 wk postpartum values (P < 0.04). Bench press values also decreased 5% by 6 wk postpartum, and only recovered 1% by 27 wk postpartum, but these were not significantly different from 0 wk or from the 6 wk postpartum values. Significant group effects were observed for the leg press at 6 wk postpartum, such that the high-BMI group had a lower value than the low-BMI group (P < 0.05) and the normal-BMI group (P < 0.04). For the leg extension, the low-BMI group had a lower value than the high-BMI group (P < 0.03) at 27 wk postpartum.
In this study, we evaluated the changes in physical activity, fitness, and strength in women before and after pregnancy. Our major findings include decreases in both maximal oxygen consumption and leg strength from prepregnancy to 6 wk postpartum, with some of these observed decreases recovering by 27 wk postpartum. These changes occur regardless of the mother’s initial BMI. In these same women, postpartum activity energy expenditure and physical activity levels were significantly lower than pregravid values in the low-BMI group, but not in the normal- and high-BMI group (5). In the present study, the total self-reported physical activity did not change from baseline to postpartum; however, there were changes in specific activities. For instance, conditioning and occupational activities decreased significantly postpartum, whereas an increase in walking and home activities was observed.
Assessment of submaximal steady-state cycle exercise at a constant workload across participants demonstrated that submaximal V̇O2 did not change postpartum compared with pregravid values. Pivarnik et al. (15) found no significant changes in absolute submaximal cycle V̇O2 in both sedentary women and physically active women from 25 and 36 wk of pregnancy to 12 wk postpartum. In contrast, several studies have reported increases throughout pregnancy in submaximal absolute V̇O2 during cycle exercise and in V̇O2 (mL·kg−1·min−1) during treadmill exercise (16) and absolute V̇O2 during cycle exercise (17).
We observed that measured V̇O2max adjusted for weight or FFM decreased substantially from 0 to 6 wk postpartum, and was still lower than baseline values at 27 wk postpartum. Others have also evaluated the maximal responses to exercise during pregnancy and postpartum (7,8,13,18). Predicted V̇O2max assessed in primigravid women during each trimester and 4 wk postpartum significantly declined from the first to second to third trimester, and then rose postpartum to the second trimester levels (8). By contrast, in recreational athletes, Clapp and Capeless (7) reported an increase in absolute V̇O2max from prepregnancy to 36–44 wk postpartum. Absolute V̇O2max measured with cycle ergometer exercise did not change from 26 wk gestation to 2 or 7 months postpartum (18). Two other studies using cycle ergometry (10,12) also reported no changes in absolute V̇O2max with pregnancy. Other modes of exercise testing, namely treadmill and swimming, have been used to test pregnancy changes in V̇O2max. Absolute V̇O2max measured during treadmill walking was unchanged (12). Yet, swimming absolute V̇O2max decreased as pregnancy duration progressed (13). As stated in the review by McMurray et al. (14), nonweight-bearing exercise (e.g., cycling) appears to have a slightly greater metabolic demand for a given amount of work during pregnancy, whereas this is not the case with treadmill walking.
These different findings among studies for both maximal and submaximal exercise may be due to the way in which the data are analyzed, that is, adjusting for body weight or fat-free mass, the initial fitness level of the women, and the types of activities these women engaged in. In our study, the self-report data provides the opportunity to examine specific activities. The women changed their activities postpartum (increasing home and walking activities, decreasing conditioning activities). The changes in intensity of these activities would have an effect on the fitness levels of our participants postpartum.
Our analyses also allowed us to examine differences among women with varying BMI. We found a significant difference in submaximal V̇O2 (adjusted for weight or FFM) between the normal- and high-BMI groups at 0 and 27 wk postpartum. The high-BMI group had a lower V̇O2 (adjusted for weight or FFM) than the normal-BMI group. Clearly, one unique aspect of our study is actual prepregnancy measurements of submaximal and maximal responses to cycle ergometer exercise. The current ACOG guidelines recommend that women engage in moderate activity most days of the week during pregnancy (3). In our study, prepregnancy fitness data indicated that these women had a wide range of fitness values (low to high). By classifying the women into fitness categories (low, fair, average, good, and high), only 10 women remained in their same fitness category from 0 to 6 and 27 wk postpartum. These women were in all fitness categories, so it was not necessarily the fittest women who maintained their high level of fitness. We do not know at what time point during pregnancy these changes in fitness occur. Our study is limited in that we did not include a measure of maximal exercise during pregnancy due to safety concerns for the mother and baby. Maximal testing has been done in one study in which no adverse effects of maximal exercise testing on fetal heart rate were found (6).
Another unique component of our study was the inclusion of strength measures. We observed significant decreases in leg and arm strength from prepregnancy to 6 wk postpartum. Also, these losses in strength are not fully regained by 27 wk postpartum. We do not know how long it would take to fully recover strength in postpartum women. The loss in strength is both in the upper and lower body; however, the largest losses are in the lower body. This seems somewhat contrary to what would be expected, because maintenance of leg strength would be expected due to gestational weight gain. One limitation of our study is the lack of reliability measures of strength in this specific population.
The reasons for these postpartum changes in activity, fitness, and strength are probably a reflection of the changes in the mother’s responsibilities when becoming a parent. Physical activity levels were lower postpartum than prepregnancy values in the low-BMI group only; no differences were seen in normal- and high-BMI groups (4). The reduced occupation activities and increased home activities from the self-report data show a shift toward lower energy expenditure activities. Hinton and Olson (11) reported that exercise self-efficacy and BMI were positive predictors of change in physical activity. The authors then suggested that exercise self-efficacy be increased in interventions aimed toward helping women maintain or increase their activity during the perinatal period (11).
In conclusion, relative to prepregnancy performance, physical fitness and strength declined in the early postpartum period, but improved by 27 wk postpartum.
1. Ainsworth, B. E., W. L. Haskell, A. S. Leon, et al. Compendium of physical activities: classification of energy costs of human physical activities. Med. Sci. Sports Exerc
. 25:71–80, 1993.
2. Akers, R., and E. R. Buskirk. An underwater weighing system utilizing “force cube” transducers. J. Appl. Physiol
. 26:649–652, 1969.
3. American College of Obstetricians and Gynecologists. Exercise during pregnancy and the postpartum period. ACOG Committee Opinion No. 267. Washington, DC: ACOG, 2002, pp. 1–3.
4. Butte, N. F., K. J. Ellis, W. W. Wong, J. M. Hopkinson, and E. O. Smith. Composition of gestational weight gain impacts maternal fat retention and infant birth weight. Am. J. Obstet. Gynecol
. 189:1423–1432, 2003.
5. Butte, N. F., W. W. Wong, M. S. Treuth, K. J. Ellis, and E. O. Smith. Energy requirements during pregnancy based on total energy expenditure and energy deposition. Am. J. Clin. Nutr
. 79:1078–1087, 2004.
6. Carpenter, M. W., S. P. Sady, B. Hoegsberg, et al. Fetal heart rate response to maternal exertion. JAMA
7. Clapp, J. F., and E. Capeless. The V̇O2max
of recreational athletes before and after pregnancy. Med. Sci. Sports Exerc
. 23:1128–1133, 1991.
8. Dibblee, L., and T. E. Graham. A longitudinal study of changes in aerobic fitness, body composition, and energy intake in primigravid patients. Am. J. Obstet. Gynecol
. 147:908–914, 1983.
9. Fuller, N. J., S. A. Jebb, M. A. Laskey, W. A. Coward, and M. Elia. Four-component model for the assessment of body composition in humans: comparison with alternative methods, and evaluation of the density and hydration of fat-free mass. Clin. Sci
. 82:687–693, 1992.
10. Heenan, A. P., L. A. Wolfe, and G. A. Davies. Maximal exercise testing in late gestation: maternal responses. Obstet. Gynecol
. 97:127–134, 2001.
11. Hinton, P. S., and C. M. Olson. Predictors of pregnancy-associated change in physical activity in a rural white population. Matern. Child. Health J
. 5:7–14, 2001.
12. Lotgering, F. K., M. B. Van Doorn, P. C. Struijk, J. Pool, and H. C. S. Wallenberg. Maximal aerobic exercise in pregnant women: heart rate, O2
production, and ventilation. J. Appl. Physiol
. 70:1016–1023, 1991.
13. McMurray, R. G., A. C. Hackney, V. Katz, M. Gall, and W. J. Watson. Pregnancy induced changes in maximal oxygen uptake during swimming. J. Appl. Physiol
. 71:1454–1459, 1991.
14. McMurray, R. G., M. F. Mottola, L. A. Wolfe, R. Artal, L. Millar, and J. M. Pivarnik. Recent advances in understanding maternal and fetal responses to exercise. Med. Sci. Sports Exerc
. 25:1305–1321, 1993.
15. Pivarnik, J. M., N. A. Ayres, M. B. Mauer, D. B. Cotton, B. Kirshon, and G. A. Dildy. Effects of maternal aerobic fitness on cardiorespiratory responses to exercise. Med. Sci. Sports Exerc
. 25:993–998, 1993.
16. Pivarnik, J. M., W. Lee, and J. F. Miller. Physiological and perceptual responses to cycle and treadmill exercise during pregnancy. Med. Sci. Sports Exerc
. 23:470–475, 1991.
17. Sady, S. P., M. W. Carpenter, P. D. Thompson, M. A. Sady, B. Haydon, and D. R. Coustan. Cardiovascular response to cycle exercise during and after pregnancy. J. Appl. Physiol
. 66:336–341, 1989.
18. Sady, M. A., B. B. Haydon, S. P. Sady, M. W. Carpenter, P. D. Thompson, and D. R. Coustan. Cardiovascular response to maximal cycle exercise during pregnancy and at two and seven months postpartum. Am. J. Obstet. Gynecol
. 162:1181–1185, 1990.
19. South-Paul, J. E., K. R. Rajagopal, and M. F. Tenholder. The effect of participation in a regular exercise program upon aerobic capacity during pregnancy. Obstet. Gynecol
. 71:175–179, 1988.
20. Wilmore, J. H. A simplified method for determination of residual lung volumes. J. Appl. Physiol
. 27:96–100, 1969.
21. Wong, S. C., and D. C. McKenzie. Cardiorespiratory fitness during pregnancy and its effect on outcome. Int. J. Sports Med
. 8:79–83, 1987.
22. Wong, W. W., L. S. Lee, and P. D. Klein. Deuterium and oxygen-18 measurements on microliter samples of urine, plasma, saliva, and human milk. Am. J. Clin. Nutr
. 45:905–913, 1987.