The purpose of this study was to evaluate the effect of structured physical exercise programs during pregnancy on the course of labor and delivery.
We conducted a systematic review and metaanalysis using the following data sources: Medline and The Cochrane Library. In our study, we used randomized controlled trials (RCT) that evaluated the effects of exercise programs during pregnancy on labor and delivery. The results are summarized as relative risks.
In the 16 RCTs that were included there were 3359 women. Women in exercise groups had a significantly lower risk of cesarean delivery (relative risk, 0.85; 95% confidence interval [CI], 0.73–0.99). Birthweight was not significantly reduced in exercise groups. The risk of instrumental delivery was similar among groups (relative risk, 1.00; 95% CI, 0.82–1.22). Data on Apgar score, episiotomy, epidural anesthesia, perineal tear, length of labor, and induction of labor were insufficient to draw conclusions. With the use of data from 11 studies (1668 women), our analysis showed that women in the exercise groups gained significantly less weight than women in control groups (mean difference, –1.13 kg; 95% CI, –1.49 to –0.78).
Structured physical exercise during pregnancy reduces the risk of cesarean delivery. This is an important finding to convince women to be active during their pregnancy and should lead the physician to recommend physical exercise to pregnant women, when this is not contraindicated.
The number of cesarean delivery deliveries increased steadily during the last decades. Although cesarean delivery has become safer, it remains an invasive procedure with potential morbidity and death for both mother and her child. Short-term maternal risks include infection, thromboembolism and hemorrhage, sometimes severe enough to be life threatening. The increase of cesarean deliveries also raises concerns about longer term morbidity such as uterine rupture and placenta previa and/or accreta during subsequent pregnancies and adhesions with chronic pain as a consequence of the surgical procedure. Obstetricians therefore should restrict the use of cesarean delivery for sound medical reasons only, and interventions effective in decreasing the risk would be welcome.
Observational studies have suggested that regular exercise during pregnancy reduces the risk of cesarean deliveries. Several randomized controlled trials have attempted to measure the effect of structured exercise programs on various outcomes, including gestational diabetes mellitus and preeclampsia. Some trials were conducted to evaluate directly the relation between physical activity and cesarean delivery risk; only 1 trial found a significant effect. Most of the studies were of relatively small sample size and were not powered to show a difference in obstetric outcomes. A metaanalysis of these studies may increase the likelihood to identify a benefit from exercise programs.
We performed a systematic review of randomized controlled trials to determine whether structured physical training programs during pregnancy can improve the course of labor and delivery.
Materials and Methods
This review was conducted in accordance with the “preferred reporting items for systematic reviews and metaanalyses” statement.
We searched electronic databases, Medline (PubMed), and the Cochrane Library, using the words: «pregnancy AND exercise AND (randomized OR randomized)». We scanned the reference lists of identified relevant articles. We imposed neither language nor publication date restrictions. The last search was run in March 2013.
The eligibility criteria were (1) randomized controlled trials (ie, mentioned as randomized in title, abstract, or full text), (2) included women of any age, parity, and body mass index (BMI) with a singleton pregnancy and none of the absolute obstetrics contraindications to exercise according to the American College of Obstetricians and Gynecologists (ACOG), (3) comparison of exercise program vs no exercise program (the exercise program should include resistance or aerobic exercise; minimum 1 session per week, supervised, to ensure it was carried out with a minimum level of intensity and regularly, thus providing some homogeneity in the intervention group; studies with an intervention limited to pelvic floor exercises, stretching, or relaxation were excluded), (4) trials that reported the mode of delivery, that included the percentage of cesarean and instrumental delivery and/or any of the following outcomes: Apgar score, duration of labor, episiotomy, epidural anesthesia, induced labor, and delivery lacerations.
Contact was attempted with authors of articles that included an adequate intervention but that did not report the prespecified outcomes. Among 17 authors who we contacted, 13 authors responded, and 1 author provided additional unpublished data. Barakat et al assured us that there was no overlap of participants in their reports.
Two of the authors (I.D. and M.B.) performed the first screening, studies appraisal, and data extraction; the third author (B.K.) verified the data collection. We extracted data on (1) type of participants: inclusion and exclusion criteria, age, BMI, parity, prepregnancy level of physical activity, type of recruitment, percentage of dropout, and pregnancy weight gain; (2) type of intervention in the exercise group: number of hours per training and per week, when the intervention started and finished, type of exercise, how the women were supervised, and compliance; (3) control group: if asked not to exercise and how any exercise habits were assessed; and (4) outcomes mentioned earlier and birthweight (which could be considered as a potential effect of the intervention or as confounding factor).
The ACOG recommends 30 minutes of exercise on most days of the week, which means a minimum of 2 hours per week. An intervention that follows these guidelines during 2 trimesters would total approximately 50 hours of exercise. For each study, we calculated the number of hours of exercise planned and used the percentage of compliance (available for 11 studies) to estimate the amount of exercise carried out.
Risk of bias
To ascertain the validity of eligible randomized trials, we determined adequacy of concealment of allocation, sequence generation, blinding of the obstetrician and data collector, and percentage of loss to follow up. We used the Jadad score, which is a scale that ranges from 0–5 points to assess the quality of randomization. Because blinding of the women was impossible, a maximum of 3 points could be obtained.
Our primary analysis was a comparison of the risk of cesarean delivery in women who participated in a structured physical activity program compared with women in the control groups. We also assessed the risk ratio for the other outcomes. Statistical analysis was performed with Review Manager software (version 5.2; US Cochrane Collaboration, Baltimore, MD); we entered the data that were collected in each report. We have not made attempts to obtain raw data from the authors of the original studies to perform individual patients’ data metaanalysis. Risk ratios were calculated with 95% CI with the Mantel-Haenszel method. We used a fixed effects model. Heterogeneity was evaluated with I 2 , and we would have used a random effect model in case of I 2 to indicate heterogeneity (I 2 >50%) Sensitivity analysis was done excluding 2 studies in which there was a marked imbalance in parity and 5 studies that had a dropout proportion of >25%. Publication bias was assessed by the examination of a funnel plot.
We screened the abstracts of the 376 identified articles and excluded 203 clearly irrelevant references and 92 observational studies ( Figure 1 ).
The full text of the 81 potentially relevant articles were obtained and assessed with the use of the eligibility criteria. Twenty-six articles were excluded because they reported none of the outcomes, even after attempted contact with the authors; 33 articles reported interventions that did not correspond to the inclusion criteria. A total of 19 reports of 16 studies were identified ( Table 1 ). The study by Barakat et al was reported in 3 different articles, and the study by Baciuk et al and Cavalcante et al were reported in 2 articles. We used the information of the 19 reports to assess the quality of the studies and used the data of the 16 studies for the metaanalysis. The studies conducted by Brankston et al and Jovanovic-Peterson et al, which were included in the Cochrane Review on exercise in pregnant women with diabetes mellitus, were not included because they did not report the cesarean percentage; the study by Bung et al was not included because the comparison was exercise vs insulin. No additional study was found by screening of reference lists.
|Study||Each group, n||Drop-outs, n||Lost to follow up, n||Primiparas, %||Pregestational body mass index, kg/m 2 a||Pregestational inactivity||Adequate allocation concealment||Adequate sequence generation|
|Barakat et al, 2012||EG: 160||22 (14%)||0||60.9||24.0 ± 4.3||Not participating in another physical activity program or exercising >4 times/wk||Yes||Yes|
|CG: 160||8 (5%)||0||54.6||23.6 ± 4.0||One-third of the women had an active profession|
|Barakat et al, 2012||EG: 50||10 (20%)||0||65.0||22.7 ± 2.8||One-half of the women were active or very active before pregnancy||Unclear||Unclear|
|CG: 50||7 (14%)||0||48.8||23.0 ± 2.9||One-third of the women had an active profession|
|Barakat et al, 2011||EG: 40||6 (15%)||0||76.5||23.9 ± 3.0||No information||Unclear||Yes|
|CG: 40||7 (17%)||0||36.4||24.8 ± 4.0||One-half of the women had an active profession|
|Price et al, 2012||EG: 43||12 (28%)||0||58.1||26.6 ± 3.1||No aerobic exercise >1 time/wk for at least 6 mo||Yes||Unclear|
|CG: 48||17 (35%)||0||58.1||28.7 ± 5.4||No information about their profession|
|Marquez-Sterling et al, 2000||EG: 10||1 (10%)||0||100||22.8 ± 4.0||No exercise on regular basis for at least 1 year before conception||Yes||Unclear|
|CG: 10||2 (20%)||2 (20%)||100||24.5 ± 4.5||No information about their profession|
|Stafne et al, 2012||EG: 429||33 (8%)||21 (5%)||58.0||24.7 ± 3.0||One-third of the women exercised 3 times/wk||Yes||Yes|
|CG: 426||49 (11%)||50 (12%)||56.0||25.0 ± 3.4||No information about their profession|
|Barakat et al, 2009||EG: 80||8 (10%)||0||72.2||24.3 ± 0.5||Not exercising >20 min on >3 d/wk||Yes||Unclear|
|CG: 80||5 (6%)||5 (6%)||57.1||23.4 ± 0.5||24% of the women had an active profession|
|Kihlstrand et al, 1999||EG: 129||5 (3%)||1 (1%)||35.4||—||No information||Yes||Unclear|
|CG: 129||11 (8%)||0||26.6||—|
|Baciuk et al, 2008||EG: 34||12 (35%)||1 (3%)||47.1||24.1 ± 4.5||Not exercising regularly||Unclear||Yes|
|CG: 37||10 (27%)||0||62.2||23.4 ± 3.8|
|Ramirez-Velez et al, 2011||EG: 33||0||9 (27%)||100||—||No information||Yes||Unclear|
|CG: 31||0||5 (16%)||100||—|
|Barakat et al, 2013||EG: 255||45 (18%)||0||—||24.1 ± 4.1||Not exercising >20 min on >3 d/wk||Yes||Unclear|
|CG: 255||37 (14%)||0||—||23.7 ± 3.8||One-third of the women had an active profession|
|Hui et al, 2012||EG: 112||0||10 (9%)||—||24.9 ± 5.4||No information||Yes||Yes|
|CG: 112||0||24 (21%)||—||25.7 ± 5.1|
|Oostdam et al, 2012||EG: 62||9 (15%)||13 (21%)||38.3||33.0 ± 3.7||No information||Yes||Yes|
|CG: 59||0||14 (24%)||28.0||33.9 ± 5.6|
|Avery et al, 1997||EG:16||1 (7%)||0||—||28.4 ± 7.6||No regular exercise regimen for >30 min twice a week||Yes||Yes|
|CG: 17||3 (18%)||0||—||25.5 ± 5.5|
|Nascimento et al, 2011||EG: 40||1 (2%)||0||30.0||34.8 ± 6.6||No regular exercise||Unclear||Yes|
|CG: 42||1 (2%)||0||23.8||36.4 ± 6.9|
|Lee, 1996||EG: 182||0||8 (4%)||100||—||No information||Unclear||Unclear|
|CG: 188||0||11 (5%)||100||—|
The 16 identified studies were randomized controlled trials that were published in English and included a total of 3359 participants.
Three studies included only primiparous women. The studies including both primiparous and multiparous women were balanced, with the exception of 2. In the 2011 study of Barakat et al, 26 primiparous women were assigned randomly to the exercise group, and only 12 women were in the control group. In the study by Avery et al, the mean parity was 1.5 in the exercise group and 0.4 in the control group. In most studies, mean BMI was normal (BMI, <25 kg/m 2 ). In 4 studies, the women were overweight (BMI, >25 kg/m 2 ) or obese (BMI >30 kg/m 2 ). Despite those differences among studies, there were no significant differences between groups for a given study. The studies included mostly women without specific conditions, except the study by Avery et al, in which all women had gestational diabetes mellitus. The mean age of participants in the studies ranged from 25-32 years, except for the study by Ramirez-Velez et al (mean, 19.5 years).
Most authors chose to mix aerobic and resistance exercise ( Table 2 ). One study (Barakat et al ) planned training that was composed exclusively of resistance exercise; 4 studies focused on aerobic exercise. The training took place in swimming pools, in sports halls, or outdoors. The study by Barakat et al in 2012 mixed aquatic and land-based activities. Exercise sessions were supervised by a physiotherapist or a fitness specialist. The intervention started during the first trimester in some studies and at the end of the second trimester in others. This led to substantial differences in the number of weeks of intervention and in the number of hours of exercise between studies. Seven of the selected studies planned an intervention of ≥50 hours of exercise.
|Study||Period of exercise||Hours of exercise||Hours × compliance||Compliance, %||Intervention||Control group|
|Barakat et al, 2012||From wk 6-9 to wk 38-39||57||49||87||Light aerobic and resistance exercises (7-8 min warm-up walking and static stretching, 25 min of aerobics and muscle strengthening exercises, 7-8 min cool-down walking); <70% HRmax||40-45 min 3 times/wk; all supervised all land sessions||No exercise program (checked by phone)|
|Barakat et al, 2012||From wk 6-9 to wk 38-39||57||48||85||Light aerobic and resistance exercises (7-8 min warm-up walking and static stretching, 25 min of aerobics and resistance <3 kg barbells, Theraband, sets of 10 repetitions, 7-8 min cool-down walking, pelvic floor exercise); <70% HRmax; |
aquatic sessions: aerobic and light resistance
|35-45 min 3 times/wk all supervised; 2 land sessions; 1 aquatic session||—|
|Barakat et al, 2011||From wk 6-9 to wk 38-39||57||51||90||Same as Barakat 2012||35-45 min 3 times/wk; all supervised; all land sessions||—|
|Price et al, 2012||From wk 12-14 to wk 36-40||60-112||56-104||93||Aerobic and resistance exercise (1 session step aerobics, 1 session walking, 1 session of alternating of treadmill and weight machines); individual brisk walking 30-60 min; 12-14 on Borg scale of perceived exertion||45-60 min 4 times/wk; 3 supervised; all land sessions||No exercise, except for household or work|
|Marquez-Sterling et al, 2000||Unclear (during 15 wks)||45||—||—||Aerobic exercises (treadmill, bicycle, rowing, cycling, walking); <155 beat/min (<80% HRmax if 30 years old)||60 min 3 times/wk; all land sessions; All supervised||—|
|Stafne et al, 2012||From wk 20 to wk 36||30||16||55||Supervised: 30-min aerobic without jumping or running, 20 min strength exercises using body weight as resistance, 3 sets of 10 repetitions for each exercise; 10-min stretching and relaxation; 13-14 on Borg scale; individually: 30 min of aerobic and 15 min of resistance exercises||60 min 1 time/wk, supervised; 45 min 2 times/wk, individually; all land sessions||Not discouraged to exercise|
|Barakat et al, 2009||From wk 12-13 to wk 38-39||47||42||90||Light resistance exercises (8-min warm-up, <60% HRmax; 20-min exercise < 80% HRmax; 8-min cool down, <60% HRmax), toning, stretching, <3 kg barbells, Theraband; sets of 10 repetitions)||35 min 3 times/wk; all supervised; all land sessions||Maintain their level of activity (<60 min/wk)|
|Kihlstrand et al, 1999||From wk 20 to wk 40||10||5||50||Aquatic light aerobic and resistance exercises (one-half hour; no specifics given)||30 min 1 time/wk; all aquatic sessions; all supervised||Not offered to exercise|
|Baciuk et al, 2008||From wk 20 to wk 40||50||21||—||Aquatic aerobic exercise (no specifics given); <70% HRmax||50 min 3 times/wk; all aquatic; All supervised||No regular physical activity|
|Ramirez-Velez et al, 2011||Unclear; from wk 20 to wk 40||48||32||75||Aerobic and resistance exercises (30-min aerobic exercises, 25-min resistance exercise, 5-min stretching); intensity “moderate-to-vigorous”||60 min 3 times/wk; all land; all supervised||No exercise|
|Barakat et al, 2013||From wk 10-12 to wk 38-39||70||67||95||Light aerobic and resistance exercise (10-12 min warm-up walking and static stretching, 25 min of resistance training with <3 kg barbells, theraband; sets of 10 repetitions; 10-12 min cool-down walking, pelvic floor exercise, and 1 session per wk of aerobic dance); <70% HRmax||50 min 3 times/wk; all land sessions; All supervised||Not discouraged to exercise|
|Hui et al, 2012||From wk 20-26 to wk 36||Unclear (25?)||—||Aerobic and resistance exercises (no specifics given; an instruction video was provided for the home-based exercises); intensity “mild to moderate”||30-45 min 3-5 times/wk; 1 supervised; all land sessions||Standard prenatal care|
|Oostdam et al, 2012||From wk 12 to wk 40||40||<20||<40||Aerobic and resistance exercises (5-10 min warm-up: 50 watt cycling; 40-min core session: alternation of 1-2 aerobic exercises with 4-6 strength exercises, 5-10 min cool-down) 12 on Borg scale||60 min 2 times/wk; all land sessions; All supervised||Standard prenatal care|
|Avery et al, 1997||Unclear||Unclear (15?)||—||Aerobic exercises at 70% HRmax; Supervised sessions (cycle ergometer 5-min warm-up, 20-min core session, 5-min cool down); individual session: walking or cycling 1-2 sessions of 30 min||30 min 2-4 times/wk; all land sessions; 2 supervised||Not to change their activity (<60 min/wk)|
|Nascimento et al, 2011||Unclear; from wk 14-24||36||22||62||Resistance training: 10-min stretching, 22-min resistance; <140 beat/min (<70% HRmax); individual session 5 times/wk: walking or resistance exercises||30 min 6 times/wk; all land sessions; 1 supervised session||Routine prenatal care|
|Lee, 1996||From wk 20 to wk 40||60||20||Unclear; 33||Aerobic exercises (no specifics given)||60 min 3 times/wk; all land sessions; all supervised||No intervention|