Volume 93, Issue 1 p. 73-79
Main Research Article
Free Access

Does regular exercise in pregnancy influence duration of labor? A secondary analysis of a randomized controlled trial

Kjell Å. Salvesen

Corresponding Author

Kjell Å. Salvesen

Clinical Sciences, Obstetrics and Gynecology, Lund University, Lund, Sweden

National Center for Fetal Medicine, Department of Obstetrics and Gynecology, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway

Correspondence

Kjell Å. Salvesen, Clinical Sciences, Obstetrics and Gynecology, Lund University, Klinikgatan12, SE-221 85 Lund, Sweden. E-mail: [email protected]

Search for more papers by this author
Signe N. Stafne

Signe N. Stafne

Clinical Services, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway

Faculty of Medicine, Department of Public Health and General Practice, Norwegian University of Science and Technology, Trondheim, Norway

Search for more papers by this author
Torbjørn M. Eggebø

Torbjørn M. Eggebø

Department of Obstetrics and Gynecology, Stavanger University Hospital, Stavanger, Norway

Search for more papers by this author
Siv Mørkved

Siv Mørkved

Clinical Services, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway

Faculty of Medicine, Department of Public Health and General Practice, Norwegian University of Science and Technology, Trondheim, Norway

Search for more papers by this author
First published: 16 September 2013
Citations: 33
The authors have stated explicitly that there are no conflicts of interest in connection with this article.

Abstract

Objectives

To study effects of regular physical exercise in pregnancy on duration of the active phase of labor and the proportions of women with prolonged active second stage.

Design

A two-armed, two-center randomized controlled trial.

Setting

St. Olavs Hospital, Trondheim University Hospital and Stavanger University Hospital.

Population

A total of 855 women were randomized to intervention or control groups.

Methods

The intervention was a 12-week exercise program, including aerobic and strengthening exercises, conducted between the 20th and 36th week of gestation. One weekly group session was led by physiotherapists and home exercises were encouraged twice a week. Controls received standard antenatal care.

Main outcome measures

The duration of active phase of labor and number of prolonged second stage deliveries (active pushing > 60 min). We also studied labor outcomes. Supplementary analyses were done in a subgroup of nulliparous women with a singleton cephalic fetus and spontaneous start of term delivery.

Results

Duration of labor was similar in the two groups, and there were no differences in labor outcomes. In a subgroup analysis the duration of active second stage labor was shorter in the control group (p = 0.01).

Conclusions

Regular physical exercise during pregnancy did not influence duration of the active phase of labor or the proportion of women with prolonged active second stage.

Abbreviations

  • BMI
  • body mass index
  • Key Message

    Regular exercise during pregnancy does not influence duration of the active phase of labor.

    Introduction

    Exercise is defined as regular, leisure-time physical activities to improve physical condition, ability or health. Previous studies indicate that women with uncomplicated pregnancies are fit for exercise with few restrictions, without any risk of harm to the mother or fetus 1. A myth prevails that athletes, gymnasts and horsewomen may experience difficult labors due to strong pelvic floor muscles, and these questions are frequently discussed on chatting websites for pregnant women 2. However, training of the pelvic floor muscles may produce strong and well controlled muscles that will facilitate labor, and one randomized controlled trial has demonstrated that intensive pelvic floor muscle training in pregnancy prevented a prolonged second stage of labor in about one in eight women 3. Improved aerobic capacity and general muscular strength may also be beneficial and facilitate labor, but this has been sparsely studied in randomized controlled trials 4.

    The present study was designed to evaluate effects of regular exercise in pregnancy. The primary outcome was gestational diabetes, and results have been published elsewhere 5. This report deals with two of five pre-specified secondary outcomes of the trial. The aim of this report was to study the influence of exercise on duration of the active phase of labor and the proportion of women with prolonged active second stage.

    Material and methods

    This was a two-armed, two-center randomized controlled trial of a 12-week exercise program vs. standard antenatal care. Pregnant women booking for second trimester ultrasound at St. Olavs Hospital, Trondheim University Hospital and Stavanger University Hospital in 2007–2009 were invited to participate. During the inclusion period from April 2007 to June 2009 in Trondheim and October 2007 to January 2009 in Stavanger approximately 12 000 pregnant women had ultrasound at the two hospitals, and 875 women consented to participate in the trial. Inclusion criteria were 1 age ≥18 years and 2 a singleton live fetus. Exclusion criteria were high-risk pregnancies and/or diseases that could interfere with participation. For practical reasons we also excluded women who lived too far from the hospitals to attend weekly training groups. The study population was predominantly white European and representative of pregnant Norwegian women with respect to body mass index (BMI) and physical activity 5, 6.

    Randomization in blocks of 30 was performed at the Unit for Applied Clinical Research, Norwegian University of Technology and Science, by a web-based computerized procedure. The staff involved with training or outcome assessments had no influence on the randomization procedure. Because of the nature of the study it was not blinded. However, labor outcomes were analyzed blind for group allocation.

    The allocation of the study population is shown in Figure 1. One group of women (n = 429) was randomly allocated to a standardized exercise program. This consisted of 30–35 min low impact aerobics, 20–25 min strength exercises, including pelvic floor muscle training, and 5–10 min light stretching and body awareness. Women trained in groups with a physiotherapist once a week over a period of 12 weeks between the 20th and the 36th week of pregnancy. In addition, they were encouraged to follow a written 45-min home exercise program at least twice a week (30 min aerobic activity and 15 min strength and balance exercises). Performing the exercise program was recorded in the women's personal training diary and through reports from the physiotherapists. In addition, physical activity was recorded in a questionnaire. Women in the control group (n = 426) received standard antenatal care and the customary information given by their midwife or general practitioner.

    Details are in the caption following the image
    Flow-chart of the study participants.

    The primary outcome of this trial was gestational diabetes. Based on a prevalence estimate of gestational diabetes of 9% and a reduction to 4% in the intervention group, a study population of 381 women in each group was needed, and 855 women were included 5. Five other hypotheses regarding secondary outcomes were pre-specified in the study protocol; urinary incontinence, anal incontinence, lumbopelvic pain, duration of labor and number of prolonged second stage deliveries. The results regarding urinary and anal incontinence and lumbopelvic pain have been published elsewhere 7, 8. For the secondary outcomes presented in this report we did a power calculation based on the fixed sample size determined by the primary outcome of the trial. We assumed that 45% of all women in the trial (n = 855) would be nulliparous with term cephalic deliveries (n = 385). In a previous trial of nulliparous women we found that 21% in the intervention group and 34% in the control group had prolonged second stage labor 3. Assuming the same difference, a sample size calculation with α = 0.05 and a power of 80% showed that 185 women in each group would be sufficient. Thus, we planned a subgroup analysis of nulliparous women with a singleton fetus in cephalic position and spontaneous start of labor after the 37th week of gestation.

    Data from labor and delivery were recorded in partograms by birth attendants. Birth attendants were blinded to the randomization. The start of the active phase of labor was defined when the cervix was dilated 4 cm and the woman had regular contractions. The start of the second stage was defined when the cervix was fully dilated, and the active phase of the second stage was defined as the time of active pushing. Prolonged active second stage was defined when the woman had pushed actively more than 60 min. Labor outcomes were recorded by one research assistant after the delivery from hospital records. She was not involved in training of the women and was blinded to group allocation while recording and plotting the data.

    Gestational length was calculated based on a mid-trimester ultrasound scan. Preterm delivery was defined as delivery before pregnancy week 37 + 0. Small-for-gestational age was defined as birthweight below 2 SD (−22%) for gestational age according to Marsal et al. 9.

    The statistical analyses were done according to the intention-to-treat principle. We analyzed categorical variables using the chi-squared test and normally distributed continuous variables using the t-test. We used Kaplan–Meier curves to analyze duration of labor and compared groups using the log rank Mantel–Haenschel test. Cesarean sections and cases with duration of labor >12 h were censored investigating duration of active phase of labor, and operative deliveries and active second stage >60 min were censored investigating duration of second stage. We also performed a Cox regression analysis investigating duration of active second stage with possible confounding variables, such as maternal age, BMI, birthweight, augmentation of labor and epidural analgesia. Data were analyzed using the statistical software package PASW statistics version 18.0 (Chicago, IL, USA).

    The study was in accordance with ethical standards of research and the Helsinki declaration. The women received written information and signed informed consent forms. Participants were not compensated financially. The study was approved by the Regional Committee for Medical and Health Research Ethics (REK 4.2007.81) and registered with Clinical trial gov (NCT 00476567) www.clinicaltrials.gov.

    Results

    In all, 875 women consented to participate in the trial. Twenty women were excluded or withdrew before the first examination: five miscarried, two had twin pregnancies and 13 did not meet the inclusion criteria. A total of 855 women were randomly allocated to an intervention group or a control group (Figure 1). Two intervention group women had missing data on delivery outcomes. Data from 427 intervention group women and 426 control group women were included in a complete case analysis.

    The groups had similar baseline characteristics (Table 1). There were 247 (58%) nulliparous women in the intervention group and 239 (56%) in the control group, and 92% of all parous women had previous vaginal deliveries.

    Table 1. Baseline characteristics of the study population.
      Intervention group (n = 429) Control group (n = 426)
    n % n %
    Age, yearsa 30.5 ± 4.4 30.4 ± 4.3
    Nulliparous 247 58 239 56
    Booking BMI, kg/m2a 24.7 ± 3.0 25.0 ± 3.4
    Booking weight, kga 70.4 ± 9.8 70.8 ± 10.3
    Exercise regularly 228 53 216 51
    Exercise regularly ≥3 times per week at high intensity 60 14 50 12
    • a Mean ± SD.
    • BMI, body mass index.

    We found no significant differences in the prevalence of gestational diabetes or levels of insulin resistance between the intervention group and control group (the primary outcome of the trial; results have been reported elsewhere) 5. There was no difference between groups in weight gain, weight or BMI at a follow-up at 36 weeks' gestation (data not shown). Adherence to protocol (exercising three days per week or more at moderate to high intensity) in the intervention group was 55% at follow-up at 36 weeks' gestation. Only 10% of women in the control group exercised at this level (p < 0.001) at 36 weeks' gestation.

    The mean duration of labor in the intervention group was 289 min vs. 281 min in the control group. In a survival analysis the duration of labor was similar (p = 0.13, log rank test) (Figure 2). The mean duration of active second stage was 32 min in the intervention group vs. 29 min in the control group (p = 0.12, log rank test). The frequencies of prolonged second stage in women with a vaginal birth was 12% (43/368) in the intervention group and 11% (38/358) in the control group (p = 0.65). Data on labor and delivery are presented for nulliparous women in Table 2 and parous women in Table 3. As can be seen, nulliparous but not parous women in the intervention group had a significantly longer active second stage of labor as compared with the control group.

    Table 2. Labor and delivery for nulliparous women in intervention and control groups.
    Intervention group (n = 245) Control group (n = 239) p-value
    n/N % n/N %
    Duration of labor, mina 373 ± 266 377 ± 373 0.90
    Duration of active second stage labor, mina 44 ± 27 38 ± 24 0.03
    Prolonged active second stage 41/208 20 34/201 17 0.47
    Spontaneous start of labor 176/245 72 167/239 70 0.64
    Breech presentation 7/234 3 8/228 4 0.75
    Epidural analgesia 87/239 36 88/233 38 0.76
    Oxytocin augmentation 122/240 51 115/232 50 0.78
    Cesarean section: slow progress/total number 6/31 20 2/31 7 0.13
    Operative vaginal delivery: slow progress/total numberb 23/46 50 16/37 43 0.54
    Episiotomy 64/207 31 48/201 24 0.11
    Perineal laceration grade 3 + 4 12/206 6 12/207 6 0.99
    Postpartum hemorrhage, mLa 404 ± 277 406 ± 244 0.94
    • a Mean ± SD.
    • b Unknown indication for seven deliveries in the intervention group and five in the control group.
    • n/N, number/total number in group with available data.
    Table 3. Labor and delivery for parous women in intervention and control groups.
    Intervention group (n = 182) Control group (n = 187) p-value
    n/N % n/N %
    Duration of labor, mina 182 ± 158 161 ± 170 0.25
    Duration of active second stage labor, mina 16 ± 14 16 ± 14 0.72
    Prolonged active second stage 2/160 1 4/157 3 0.40
    Spontaneous start of labor 147/182 81 147/187 79 0.61
    Breech presentation 3/178 2 3/181 2 0.98
    Epidural analgesia 34/177 19 23/183 13 0.08
    Oxytocin augmentation 32/179 18 27/184 15 0.41
    Cesarean section: slow progress/total number 2/14 14 2/18 11 0.79
    Operative vaginal delivery: slow progress/total numberb 0/6 0 6/8 75 0.005
    Episiotomy 17/165 10 10/164 6 0.17
    Perineal laceration grade 3 + 4 2/164 1 4/164 2 0.41
    Postpartum hemorrhage, mLa 303 ± 137 336 ± 187 0.06
    • a Mean ± SD.
    • b Unknown indication for three deliveries in the intervention group.
    • n/N, number/total number in group with available data.
    Details are in the caption following the image
    Kaplan–Meier plot of the active phase of labor differentiated between intervention group (----) and control group (- - -). Cases with caesarean sections and duration of labor >12 h were censored.

    There were no differences between groups in mean gestational age at birth (280 vs. 281 days, p = 0.22) or mean birthweight (3515 vs. 3523 g, p = 0.82), and no differences in the numbers of preterm births (20 vs. 19, p = 0.88) or small for gestational age babies (8 vs. 7, p = 0.80).

    One subgroup analysis of a secondary outcome was pre-specified in the study protocol, and a power calculation had been done. In this subgroup analysis of nulliparous women (n = 387) with a fetus in cephalic position and spontaneous start of labor at term, women in the control group had a significantly shorter duration of the active second stage labor (p = 0.01) (Figure 3). In Cox regression analyses the unadjusted hazard ratio analyzing the duration of second stage was 1.37 (95% CI 1.05–1.80; p = 0.02), and the difference remained statistically significant when possible confounders were entered in the model [hazard ratio 1.34 (95% CI 1.02–1.77; p = 0.04]. However, the frequencies of women with prolonged second stage was 22% (37/171) in the intervention group vs. 17% (27/164) in the control group (p = 0.23), and there were no differences in the duration of active phase (total length) of labor (p = 0.36). Other labor outcomes are presented in Table 4 for nulliparous women, and in Table 5 for parous women.

    Table 4. Maternal characteristics and labor in intervention and control groups restricted to nulliparous women with a singleton cephalic fetus and spontaneous start of labor after 37 weeks.
    Maternal characteristics and labor outcomes Intervention group (n = 201) Control group (n = 186) p-value
    n/N % n/N %
    Maternal age, yearsa 29.1 ± 3.9 28.9 ± 4.2 0.64
    BMI at 18  weeks, kg/m2a 24.6 ± 3.2 24.8 ± 3.5 0.64
    Duration of labor, mina 390 ± 277 377 ± 338 0.69
    Duration of active second stage labor, mina 47 ± 28 38 ± 23 <0.01
    Prolonged second stage 37/171 22 27/164 17 0.23
    Epidural analgesia 65/197 33 68/182 37 0.37
    Oxytocin augmentation 98/198 50 89/181 49 0.95
    Operative vaginal delivery 40/201 20 30/186 16 0.34
    Cesarean section 24/201 12 17/186 9 0.37
    Episiotomy 54/172 31 40/163 25 0.16
    Perineal laceration grade 3 + 4 10/170 6 9/168 5 0.83
    Postpartum hemorrhage, mLa 410 ± 291 403 ± 237 0.80
    • a Mean ± sd.
    • BMI, body mass index.
    • n/N, number/total number in group with available data.
    Table 5. Maternal characteristics and labor in intervention and control groups restricted to parous women with a singleton cephalic fetus and spontaneous start of labor after 37 weeks.
    Maternal characteristics and labor outcomes Intervention group (n = 158) Control group (n = 156) p-value
    n/N % n/N %
    Maternal age, yearsa 32 ± 4 32 ± 4 0.91
    BMI at 18 weeks, kg/m2a 25 ± 3 25 ± 3 0.11
    Duration of labor, mina 182 ± 153 154 ± 138 0.11
    Duration of active second stage labor, mina 16 ± 15 16 ± 13 0.89
    Prolonged second stage 2/138 1 3/132 2 0.62
    Epidural analgesia 31/154 20 21/153 14 0.14
    Oxytocin augmentation 25/156 16 23/154 15 0.79
    Operative vaginal delivery 8/158 5 5/156 3 0.41
    Cesarean section 13/158 8 16/156 10 0.54
    Episiotomy 14/144 10 9/138 7 0.33
    Perineal laceration grade 3 + 4 2/141 1 3/138 2 0.63
    Post partum hemorrhage, mLa 298 ± 137 339 ± 194 0.03
    • a Mean ±SD.
    • BMI, body mass index; n/N, number/total number in group with available data.
    Details are in the caption following the image
    Kaplan–Meier plot of the duration of active second stage labor among nulliparous women with a fetus in cephalic position and spontaneous start of labor at term differentiated in intervention group (-----) and control group women (- - -). Cases with operative deliveries and duration of active second stage >60 min were censored.

    Since only 55% of the intervention group women followed the recommended exercise protocol, we performed a post-hoc per protocol analysis (not pre-specified in the study protocol) comparing 217 women in the intervention group exercising ≥ three times per week with 426 women in the control group. The duration of the active phase of labor was similar between groups (p = 0.37), and the frequency of women with prolonged second stage was also similar (p = 0.39). However, the duration of active second stage tended to be shorter in the control group (p = 0.05).

    Discussion

    No differences were found in the duration of labor between the intervention and control groups, the proportion of women with prolonged active second stage or labor outcomes. In a subgroup analysis of nulliparous women, the active second stage labor was shorter in the control group.

    A strength of this study was that it was an adequately sized, randomized controlled trial with blinded analyses of outcomes. One limitation is that this report is a secondary analysis of the trial, and the only significant finding was found in a subgroup. However, this subgroup analysis was pre-specified in the study protocol, and the study was adequately powered for this analysis. Still, the results from the subgroup analysis should be viewed with caution.

    Another limitation is that only 875 (7%) of 12 000 women receiving a study invitation letter with their scan appointment, participated in the trial. The exact number of eligible women is unknown because we only included singleton low-risk pregnancies in the trial, and women living too far from the hospital to attend weekly training groups were not eligible. A low inclusion rate may influence the external validity (generalizability) but the internal validity (comparison between groups) of a randomized controlled trial is not affected by a low inclusion rate. The present trial included women who had a BMI within the normal range (24.8 ± 3.2), and 32% of the women exercised regularly prior to pregnancy at moderate to high intensity three times per week or more. Thus, the generalizability of the results should be interpreted with caution in other pregnant populations with higher BMI, less physically active women and in ethnically diverse populations.

    The present trial is in line with previous randomized trials reporting on labor, delivery and neonatal outcomes. A Cochrane review from 2010 included 14 trials involving 1014 women 4. Eleven trials reported on pregnancy outcomes, and no statistically significant associations were found. This review concluded that available data were insufficient to infer important risks or benefits for the mother or infant 4. The present trial is larger than all previous trials grouped together in the Cochrane review. Thus, it is reassuring that no significant associations were found. In addition, two recent trials from 2012 show similar results 10, 11. Price et al. 10 included 91 women in a randomized controlled trial and found significantly fewer Cesarean sections in the intervention group, but no other significant differences. Barakat et al. 11 included 290 women in a randomized controlled trial and found reduced proportions of instrumental deliveries and Cesarean sections in the intervention group, but other pregnancy and newborn outcomes were unaffected. In our trial, there were no differences in numbers of operative deliveries or cesarean sections (overall or for prolonged labor).

    The finding of a shorter active second stage labor among nulliparous women in the control group contrasts with findings from a previous trial demonstrating that intensive pelvic floor muscle training during pregnancy prevented a prolonged second stage labor in about one in eight nulliparous women 3. In that study the neonates were slightly smaller, there was less need to perform episiotomy, and there were significantly fewer breech presentations in the intervention group 3. This could not be verified in the present trial, where we found similar birthweights, breech presentations and rates of episiotomy. One explanation of the contradictory results may be differences in the training program of the two trials. The training program of one trial focused specifically on pelvic floor muscle training to prevent and treat urinary incontinence. In the other trial the focus was aerobic physical training and general strength exercises. Pelvic floor muscle training was part of the training program but was not highlighted to the same extent. Another explanation may be that there is no true effect of physical exercise on labor and delivery and the contradictory results may be due to chance. One trial found a shorter second stage and the other a longer second stage of labor after exercise in pregnancy. These findings were of borderline statistical significance (p = 0.01–0.06 in various analyses in both trials).

    Fifty-five percent of the intervention group women followed the recommended exercise protocol of training at moderate intensity of 45–60 min for three days per week. This was somewhat disappointing as this is the official recommendation for all pregnant women 1. An explorative analysis of protocol-adherent women compared with control women demonstrated no differences in labor outcomes or duration of labor except for a tendency towards a prolonged active second stage in the intervention group. This supports our interpretation that there were no major effects of physical exercise on labor and delivery.

    In conclusion, we have found that regular physical exercise during pregnancy did not influence the duration of the active phase of labor or the proportion of women with prolonged active second stage.

    Funding

    The trial was supported by the Norwegian Fund for Postgraduate Training in Physiotherapy, and the Liaison Committee for Central Norway Regional Health Authority (RHA) and the Norwegian University of Science and Technology. The sponsors of the study had no role in the study design, data collection, data analysis and interpretation, writing of the manuscript or decision to submit for publication.

    Acknowledgments

    The authors thank all physiotherapists and medical secretaries at the two hospitals for their efforts during this study.