The efficacy of moderate-to-vigorous resistance exercise during pregnancy: a randomized controlled trial
Abstract
Objectives
To assess the effect and safety of moderate-to-vigorous resistance exercise during pregnancy.
Design
Randomized controlled study.
Setting
Two antenatal clinics in Gothenburg, Sweden.
Population
Ninety-two healthy pregnant women.
Methods
The intervention was administered during gestational weeks 14–25. The intervention group received supervised resistance exercise twice a week, performed at an activity level equivalent to within moderate-to-vigorous (n = 51). The control group received generalized exercise recommendation, a home-based training program and a telephone follow up (n = 41).
Main outcome measures
Health-related quality of life, physical strength, pain, weight, blood pressure, functional status, activity level, and perinatal data.
Results
Functional status deteriorated during the intervention in both groups and pain increased. Significant differences between the groups were obtained only for birthweight. Newborns delivered by women who underwent resistance exercise during pregnancy were significantly heavier than those born to control women; 3561 (±452) g vs. 3251 (±437) g (p = 0.02), a difference that disappeared when adjustment was made for gestational age (p = 0.059). Both groups showed normal health-related quality of life, blood pressure, and perinatal data.
Conclusions
These findings indicate that supervised, moderate-to-vigorous resistance exercise does not jeopardize the health status of healthy pregnant women or the fetus during pregnancy, but instead appears to be an appropriate form of exercise in healthy pregnancy.
Abbreviations
-
- BP
-
- blood pressure
-
- DRI
-
- disability rating index
-
- GDM
-
- gestational diabetes mellitus
-
- HRQoL
-
- health-related quality of life
-
- RCT
-
- randomized controlled trials
-
- SF-36
-
- Short Form-36 Health Survey
Key Message
Supervised regular, moderate-to-vigorous resistance exercise has no adverse effect on childbirth outcome, pain, or blood pressure. Further investigation into the use of resistance exercise during pregnancy is needed.
Introduction
Maintaining a physically active lifestyle is associated with many benefits, including lower risk of cardiovascular disease, diabetes, hypertension, some types of cancer, and depression 1. Exercise is defined as a planned, structured, and repetitive subset of physical activity that improves or maintains physical fitness, overall health or well-being as an intended intermediate or final objective 2. Pregnant, healthy women are recommended to do 30 min or more of light to moderate exercise a day on most, if not all, days of the week 3, 4. Exercise during pregnancy is associated with reduced back pain 5, improved health perception 6, and weight gain control 7. In the literature investigations have reported on the effect of aerobic exercise when adopted in pregnancy. Only a few randomized controlled trials (RCT) have, however, examined the efficacy and safety of resistance exercise during pregnancy 8-10. One study found resistance exercise performed with an elastic band effective in improving glycaemic control in women with gestation diabetes mellitus (GDM) 9. No adverse impact on the delivery 10, or the newborn has been seen 8. Hence, in their review Nascimento et al. 11 suggested adding resistance exercise to the exercise recommendations. Although the exercise recommendations do not provide specific guidance for vigorous intensity exercise 3, 4, Nascimento et al. 11 suggested that their results supported the promotion of moderate-to-vigorous prenatal exercise. Ruchat et al. 12 concluded that a walking program of vigorous intensity is safe and beneficial to both the mother and fetus, and also promotes decline in glucose concentrations among women at risk of GDM 13. In a prospective study Jukic et al. found an association between first-trimester vigorous exercise and longer gestation, and reduced risk of preterm births 14. To the best of our knowledge, no RCT has been conducted that studies prenatal resistance exercise at moderate-to-vigorous intensity.
The aim of this study was to evaluate the health effect and safety of moderate-to-vigorous intensity resistance exercise when free weights are used by healthy women during pregnancy, with regard to health-related quality of life (HRQoL), pain location, physical strength, body weight gain, blood pressure (BP), functional status, activity level, and childbirth outcomes.
Material and methods
Participants for this study were recruited from February 2006 through November 2006, and from September 2008 through April 2009, all from two antenatal clinics in Gothenburg, Sweden. Background variables did not significantly differ in between the two study periods. The participants were verbally informed about the study by midwives or they received written information available at the antenatal clinic. Those interested in participating contacted the research coordinator. The criteria of inclusion were: (i) Pregnancy <14 weeks of gestation; (ii) single pregnancy; (iii) absence of medical or obstetric diseases; (iv) ability to understand verbal and written Swedish. Criteria of exclusion followed the recommendation regarding contraindications according to the American College of Obstetricians and Gynecologists 4. Approximately 1500 pregnant women had been registered at the antenatal clinics at that time (99% of Swedish women receiving combined obstetric and midwifery care have their deliveries in the public health system, via antenatal clinics) 15. Of 116 women willing to participate, 24 were excluded for either twin pregnancy or being ≥14 weeks of gestation. Hence, 92 women were included (Figure 1). The study was approved by the Regional Ethical Review Board at the University of Gothenburg (D No. 353-05) and registered in the Research and Development Center in Sweden (registration number VGFOUGSB-4388). Written informed consent for participation was obtained from each participant. The CONSORT guidelines were followed while conducting the study.
The women were enrolled at gestational week 13, at which time the research coordinator (physiotherapist) met with each and provided them with general exercise recommendations and a home training program. Participants were randomly assigned to either the intervention group or the control group (allocation ratio 1:1). The research coordinator performed the randomization by using opaque sealed envelopes, which were randomly picked out before the meeting with each participant. All data were collected at a primary health care location by an investigator (physiotherapist) who was blinded to participant group status. The intervention group practised high-repetition, resistance training twice a week for 12 weeks (pregnancy weeks 14–25), performed using light barbells [1–10 lbs (0.45–4.5 kg)] and weight plates [2.5–10 lbs (1.1–4.5 kg)], which was carried out while listening to music in a supervised (by the research coordinator) group exercise setting 16. The exercises were self-adjusted to each woman's condition of pregnancy and performed at a self-estimated, moderate to vigorous intensity 17. Each session was 60 min long, including warm-up and wind-down. All major muscle groups were trained repeatedly (50–80 repetitions for each muscle group) during 3–5 min, including shorter breaks. The training was inspired by BODYPUMP™ 16 but adjusted to pregnant women: warm-up and relaxation periods were extended a few more minutes, squat jumps were exchanged for heel raises, the squats were less deep and the abdominal training was exchanged for pelvic-lift and static-abdominal training. All such training was performed with good form for trunk- and pelvic-floor muscle control and these were supervised by the research coordinator. Besides this training, additional exercises were recommended to the participants, including walking, bicycling, water-gymnastics, Pilates, yoga and resistance exercises. Compliance was defined as participation in at least two-thirds of all training sessions.
Control group participants received general exercise recommendations during pregnancy and were informed of appropriate exercise and exercise frequency. Additionally, they were given written guidance and instruction on a general training program containing home exercises suitable for pregnancy, including pelvic-floor muscle training. They were offered a free visit to a physiotherapist if desired. Exercises were recommended to the control participants, including walking, bicycling, water-gymnastics, Pilates, yoga and resistance exercises. In pregnancy week-18 the research coordinator contacted the controls by telephone for follow up. Before and after the intervention the participants filled in questionnaires and performed the following tests: HRQoL was measured using the Short Form-36 Health Survey (SF-36; Swedish Acute Version 1.0) 18 SF-36 comprises 36 items, 35 of which are arranged in eight physical and mental health scales, and one being a self report of health change. Scores may be obtained ranging from 0 to 100. Higher scores positively correlate with higher levels of functioning. The eight sub-scales are: (i) physical functioning; (ii) role, physical; (iii) bodily pain; (iv) general health; (v) vitality; (vi) social functioning; (vii) role, emotional; and (viii) mental health. Pain was measured using a graphic (drawings) depiction of body locations, upon which the participant would localize the area of pain. Scores were assigned to indicate the presence or absence of pain in each of 45 body areas 19. Five areas located to the lower back and pelvis were selected for analysis.
Hand-grip isometric strength was defined as the peak, voluntary, hand-grip force applied using the Grippit® electronic hand-grip strength detector (Grippit®AB, Gothenburg, Sweden), administered in accordance with standard laboratory protocol 20. Grip strength correlates with upper extremity strength 21. Weight gain and BP were obtained from medical records at the perinatal clinic, Primary Health Care, Gothenburg, Sweden. Weight gain was measured in kilograms, wearing clothes but with no shoes. Functional status was measured by the Disability Rating Index (DRI) and recorded on a visual analogue scale, i.e. a self-estimation scale from 0 to 100, where 0 indicated no restricted activity and 100 indicated inability to perform the activity. The DRI includes 12 items that measure functions, including: dressing, outdoor walks, climbing stairs, sitting for a longer time, standing bent over a sink, carrying a bag, making bed, running, light physical activity, heavy physical activity, lifting heavy objects and participating in exercises/sports. During the intervention period all of the participants kept a structured diary of daily physical activity, according to the short form of International Physical Activity Questionnaire 22. Each activity was divided into “vigorous intensity”, “moderate intensity”, “walking” and “sedentary”. The participants estimated the number of minutes per day that they had performed the different activities.
Perinatal data, including birthweight, birth length, gestational age at time of delivery and cesarean section rates were obtained from the Sahlgrenska University Hospital's perinatal records.
Statistical analysis
The sample size calculation was calculated on a difference in HRQoL measured by SF-36. To detect a clinically significant difference between groups with 80% power and at the 5% significance level, and a difference between groups of 10 units, 40 participants were needed in each group. Group-mean differences in physical strength, body mass index, BP, activity level, birthweight, birth length, and gestational age were analyzed by the Student's t-test or chi-squared test using statistical software SPSS version 22.0 (IBM Corp, Armonk, NY, USA). Subgroup scores in SF-36 and DRI were analyzed using the Mann–Whitney U-test. Birthweights adjusted for gestational week were converted to Z-scores. A p-value <0.05 (two-tailed) was defined as significant. Only participants who fully completed the trial according to the protocol were considered compliant and analyzed.
Ninety-two women met the inclusion criteria and were enrolled (Figure 1). No significant differences in background characteristics were found between the group of women who dropped out and those for whom complete data were obtained, up to and including gestational week 25. Sixty percent of the women in the intervention group estimated their own resistance exercise as “vigorous” and 40% rated their exercise as “moderate” intensity. The mean attendance rate in the intervention group was 80% (67–100%).
Results
The group means for the intervention group and control group did not significantly differ on multiple variables, including: age, education, parity, pre-pregnancy body mass index, physical activity, marital status, and tobacco status (Table 1). HRQoL did not differ between the intervention group and controls (Figure 2), neither number of women with back pain nor those with pelvic pain (Table 2). Group-means scores for peak voluntary strength, weight gain, BP, and activity level did not differ significantly (Table 2). As shown in Figure 3, no significant differences were found between the groups in physical functioning, measured by the DRI. Birthweights were significantly higher in the intervention (x̅I = 3561 ±452 g) compared with the control group (x̅c = 3251 ±437 g; p = 0.02). However, when adjusted for gestational age, this difference did not remain (p = 0.059). The mean birth length was 50.7 (±2.1) cm in the intervention group and 49.4 (±2.6) in controls (p = 0.10), and gestational age was 280.1 (±12.5) days in the intervention group and 276.7 (±7.7) among controls (p = 0.16). There were no differences in cesarean section rates. In each group, five women had a cesarean section: 14% and 15% in the intervention group and the control group, respectively.
| Baseline demographics | Intervention group (n = 38) | Control group (n = 34) | p-value |
|---|---|---|---|
| Maternal age, mean (±SD) | 30.8 (±3.6) | 30.6 (±3.4) | 0.77 |
| Education, years at school, mean (±SD) | 15.4 (±1.79) | 16.3 (±1.8) | 0.07 |
| Prepregnancy body mass index, mean (±SD) | 22.6 (±2.5) | 23.0 (±2.6) | 0.50 |
| Activity level pregnancy week 13 (min/day), mean (±SD) | |||
| Walking | 52 (±47) | 47 (±58) | 0.68 |
| Moderate | 11.7 (±14.8) | 18.3 (±20.7) | 0.12 |
| Vigorous | 1.1 (±3.2) | 1.2 (±4.3) | 0.89 |
| Primipara, n (%) | 28 (74) | 28 (82) | 0.52 |
| Marital status, n (%) | |||
| Married/cohabiting, n (%) | 38 (100) | 33 (97) | 0.47 |
| Smoking, n (%) | 1 (2.5) | – | 0.52 |
| Taking snuff, n (%) | 1 (2.5) | – | 0.52 |
| Outcome | Intervention group (n = 38) | Control group (n = 34) | p-value | CI |
|---|---|---|---|---|
| Pain, n (%) | ||||
| Gw 13 | 12 (31%) | 12 (36%) | 0.40 | |
| Gw 25 | 22 (58%) | 21 (52%) | 0.52 | |
| Peak voluntary strength, mean (±SD) | ||||
| Left hand | ||||
| Gw 13 | 311 (±47) | 309 (±58) | 0.92 | −25.8 to 23.2 |
| Gw 25 | 317 (±52) | 311 (±55) | 0.65 | −31.2 to 19.4 |
| Right hand | ||||
| Gw13 | 338 (±46) | 342 (±58) | 0.73 | −20.7 to 28.1 |
| Gw25 | 344 (±54) | 248 (±67) | 0.69 | −24.3 to 32.8 |
| Weight | ||||
| Gw13 | 66.5 (±9.3) | 65.0 (±10.3) | 0.61 | −6.1 to 3.1 |
| Gw25 | 72.7 (±9.2) | 70.5 (±11.0) | 0.80 | −6.7 to 2.6 |
| Systolic blood pressure | ||||
| Gw13 | 109 (±18.8) | 111 (±10.5) | 0.39 | −5.1 to 9.2 |
| Gw25 | 112 (±8.8) | 112 (±9.7) | 0.56 | −4.7 to 4.0 |
| Diastolic blood pressure | ||||
| Gw13 | 66.2 (±8.3) | 63.7 (±7.7) | 0.35 | −6.23 to 1.28 |
| Gw25 | 65.7 (±8.2) | 63.4 (±7.8) | 0.56 | −6.02 to 1.52 |
| Activity level | ||||
| Walking | ||||
| Gw13 | 52 (±47) | 47 (±58) | 0.68 | −29.6 to 19.4 |
| Gw25 | 118 (±109) | 79 (±76) | 0.08 | −83.6 to 5.1 |
| Moderate | ||||
| Gw13 | 11.7 (±14.8) | 18.3 (±20.7) | 0.12 | −1.73 to 14.99 |
| Gw25 | 15.8 (±12.9) | 16.8 (±15.4) | 0.76 | −5.60 to 7.61 |
| Vigorous | ||||
| Gw13 | 1.1 (±3.2) | 1.2 (±4.3) | 0.89 | −1.88 to 1.64 |
| Gw25 | 11.2 (±26.9) | 4.6 (±7.1) | 0.15 | −15.9 to 2.78 |
| Sedentary | ||||
| Gw13 | 316 (±196) | 366 (±170) | 0.25 | −35.9 to 136.2 |
| Gw25 | 372 (±171) | 384 (±171) | 0.77 | −68.8 to 92.7 |
- Mean values ± SD and confidence interval (CI) or n (%).
- Gw, gestational week.
Discussion
Benefits of resistance exercise for adults are well-known 23, but in pregnancy women are recommended to proceed with caution, particularly when exercise is performed at moderate or vigorous levels 3, 4. The present study indicates that supervised regular, moderate-to-vigorous resistance exercise performed twice a week does not adversely impact childbirth outcome, pain, or BP. However, it is important to note that physical and emotional status changes accompany pregnancy, including a decrease in functional status and an increase in pain, whether one does resistance exercise or not. The intervention did not have any significant impact on HRQoL. Previous studies were inconclusive. Barakat et al. reported improved health self-perception, a component of HRQoL, among previously sedentary pregnant women 6. However, other studies failed to show a significant increase in HRQoL among exercising, overweight/obese women 24, or pregnant women performing water gymnastic training 25.
The participants self-rate the intensity of their resistance exercise as moderate-to-vigorous; few RCTs with pregnant women have exercised to such a vigorous level of intensity. Remarkably, these results show no adverse effects on the course of pregnancy or the health status of offspring 12-14. In our study, birthweight in the intervention group was higher than that of those born to control women, but when adjusted for gestational age this was not significant. In the body of literature we found inconsistent results on exercise and birthweight. Exercise may reduce the risk of low birthweight outcome 26 leading to a hypothesis that exercise enhances growth in the feto–placental unit and reduces the risk of preterm birth 14. Additionally, though in contrast, Jukic et al. reported that women who performed exercise tended to have lighter babies 14.
A few RCTs have studied resistance training during pregnancy. Light-to-moderate resistance training seems to be effective in improving glycaemic control in women with GDM 9, and does not adversely impact the type of delivery 10 or the health status of offspring 8. The present study corroborates this, supporting the finding that higher intensity resistance training is appropriate as a prenatal exercise. Back and pelvic pain are common during pregnancy. Pennick and Liddle concluded from their review that more than two-thirds of pregnant women experience back pain and almost one-fifth experience pelvic pain 27. One might be apprehensive about performing weight-bearing exercise with barbell and weight-plates because this could negatively impact pelvic girdle pain and lumbar pain. However, the present study shows that women from both intervention and control groups reported back and pelvic pain equally as often. In a large RCT, Stafne et al. 28 concluded that prenatal exercise does not modify the risk of low back pelvic pain, although the women they studied who regularly exercised seemed to handle the disorders in a better way.
Pregnancy in itself is dynamic, of course, in many respects. This presents a methodological challenge, and it is hard to determine a baseline for many of the variables under study. The SF-36 measures physical and mental health functioning along eight dimensions, but these are conditions that change naturally as pregnancy progresses 29. This underscores the benefit of carrying out RCTs.
A limitation in the present study may be the small sample size: though our power analysis led us to choose 80 participants, the effect size in our instruments may have been underestimated, which is a possible explanation for not finding significant differences in most of the obstetric and perinatal outcomes (type 2 error). Another limitation is that both groups exhibited a relatively high level of physical activity, including that both groups were physically active at baseline, which may explain why no significant difference was found between the groups. For instance, Price et al. 30 showed that in comparison with pregnant women who remain sedentary during the gestation period, active pregnant women improve muscular strength and fitness. An alternative approach might have been to study sedentary women, or women with some kind of disability, such as back pain, obesity, GDM or reduced HRQoL, to examine group differences. Future studies might examine the efficacy of regular resistance training among pregnant women with poorer HRQoL, back pain, GDM or women sedentary at baseline.
Another limitation is the potential bias associated with the high drop-out rate, as 20 of 92 participants did not complete the intervention. During pregnancy unforeseen things do occur, including miscarriage, as occurred once in each of the groups in this study. Another woman in the intervention group dropped out due to ‘discomfort during exercise’, in her case while doing squats, a lower body resistance movement. Good trunk- and pelvic-floor muscle control is very important during moderate-to-vigorous exercise in pregnancy, especially during exercises when the hip is in flexion and a heavier load is placed on pelvic floor muscles and cervix (for example squats or cycling in a forward leaning position).
Since the study comprised women who were volunteers and motivated to participate in an exercise program during pregnancy, the result may not be generalizable to other populations. Additionally, the women were used to doing exercise, ranging from regular walks to vigorous exercise. In comparison with the background population, these women were also highly educated and had a low body mass index.
We consider that supervised, self-rated moderate-to-vigorous resistance training does not jeopardize the pregnant woman or the fetus in healthy pregnancy, although pregnant women should follow the exercise recommendations and start at an exercise level corresponding to light-to-moderate 3, 4. Within our knowledge, this is the first study to examine moderate to vigorous resistance training. In a review, Nascimento et al. 11 recommend moderate-to-high-intensity exercise for active women. Noting that no upper level of safe exercise intensity has been established 3, 4, this study warrants further investigation into the use of resistance exercise during pregnancy, to advance arguments toward new recommendations regarding appropriate exercise activity and intensity in healthy pregnancy.
Conclusion
The results of this study indicate that supervised, moderate-to-vigorous resistance training does not jeopardize the pregnant women or the fetus during pregnancy, but instead appears to be an appropriate form of exercise in healthy pregnancy.
Funding
This study was financially supported by the Wilhelm and Martina Lundgren's Research Foundation, IRIS Stipendiet, the Renée Eander fund, the local Research and Development Center of Gothenburg and South Bohuslän, and from Research and Development Primary health Care of Gothenburg (clinical trial registration number 2021).
