Study Design:
This study is a secondary analysis of a single-blinded, four-arm randomized controlled exercise intervention trial that focused on determining the influence of different modes of maternal exercise during pregnancy on neonatal health outcomes[12]. Pregnant women were randomized to one of four intervention groups: aerobic training, resistance training, combination (resistance and aerobic) training and non-exercising control group. The study was 24 + weeks in duration, following women from the early 2nd trimester (13–16 weeks of gestation) until delivery. All protocols were approved by the East Carolina University Institutional Review Board. The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board of East Carolina University (#12-002524); the study was registered on clinicaltrial.gov #NCT03838146, on 12/02/2019. Written informed consent was obtained from all participants as well as a clearance letter from their obstetric provider confirming their pregnancy and ability to engage in moderate-intensity exercise.
Study Population: Healthy women with a low-risk, singleton pregnancy was recruited from local obstetric clinics via fliers and email announcements. Women were eligible for the study if they met the following criteria: gestational age ≤ 16 weeks, ages 18–40 years, pre-pregnancy body mass index (BMI) of 18.50-34.99 kg/m2, not currently using alcohol, tobacco, recreational drugs, or medications for chronic disorders, no contraindications to exercise in pregnancy as outlined by the American College of Obstetricians and Gynecologists, with no pre-existing diabetes, hypertension, or other cardiovascular disease. Women diagnosed with gestational diabetes during the study were still included in the intervention.
Pre-Intervention Exercise Testing
Prior to randomization, all participants completed a validated submaximal modified Balke treadmill exercise test[13] to assess aerobic exercise capacity as well as the one-repetition maximum (1RM) test to assess strength. During the treadmill test, oxygen and carbon dioxide levels were assessed via indirect calorimetry (Parvo Medics, TrueOne 2400, Sandy, UT) to determine aerobic capacity, i.e., fitness level, VO2peak (ml O2⋅kg− 1⋅min− 1) and maternal heart rate (HR), the latter of which was continuously measured (Polar FS2C heart rate monitor). Following the test, target heart rate zones (THR) were determined in order to establish target heart rates (THRs) corresponding to 40 to 59% VO2peak for moderate-intensity exercise[14]. During the 1RM tests, participants performed resistance exercises based on the American College of Sports Medicine (ACSM) protocol[15] to determine maximum strength encompassing a full array of upper and lower body exercises on machines, free weights, and body weight. Participants randomized to a group with resistance exercise started at 60% of their 1RM test weight.
Exercise Intervention
Participants engaged in 50 minutes of moderate-intensity exercise sessions three times per week for 24 + weeks. All exercise sessions were supervised by certified trainers at one of two university-affiliated facilities and followed a standard protocol as described previously[12]. Exercise sessions consisted of a 5-minute warmup, 50 minutes of moderate-intensity exercise, and a 3-5-minute cooldown. Exercise intensity was monitored via maternal HR, using a Polar FS2C heart rate monitor, as well as the 15-point Borg scale of perceived exertion to ensure compliance with the protocol as previously published[16]. The non-exercising control group performed stretching of all muscle groups and breathing exercises with inhalation and exhalation[17, 18]. This approach is aimed at maintaining contact and engagement with control participants throughout the study and is critical to retention[19]. This also ensures equal contact between exercisers and control participants, thus attenuating varying participant engagement as a potential covariate. Exercise adherence was tracked for all participants as the number of sessions attended divided by the total sessions possible. Participants completing at least 80% of sessions were considered exercise adherent and included in the per protocol analyses. Participant activities during pregnancy were also quantified as MET∙min∙wk− 1 of exercise based on the calculation of (frequency X duration of session in minutes) then multiplied by the published MET (metabolic equivalent) level for their specific exercise. The totals for all weeks were summed and averaged for the pre-pregnancy MET∙min∙wk− 1 value.
Maternal Covariates
Maternal demographic and pregnancy-related characteristics including age, race/ethnicity, parity, pre-pregnancy weight and height, gestational diabetes mellitus status (yes or no) were abstracted from various sources including pre-screening eligibility questionnaires and electronic health records. Pre-pregnancy BMI was calculated using self-reported height and weight collected from the pre-screening eligibility questionnaire at enrollment via the following established equation
\(BMI= \left(\right({W}{t} \left[{k}{g}\right]/\left({{h}{t} [{m}}^{2}\right]\left)\right)\) [5]
BMI classifications used were previously established[20]:18.5–24.9 kg/m2 was designated as normal, 25.0-29.9 kg/m2 was overweight, and ≥ 30.0 kg/m2 was obese. Additionally, gestational weight gain (GWG) was determined by calculating the differences in maternal weight gain between pre-pregnancy weight and delivery weight acquired from electronic health records.
Pregnancy and Delivery Outcomes
Delivery mode (spontaneous vaginal delivery or unplanned/emergency or planned cesarean birth, reason for delivery mode), gestational age in weeks, and birth weight in kilograms and infant sex were acquired from electronic health records. Since our question focused on the potential physiological influence of exercise on the delivery process, 14 women with planned cesarean deliveries (6 exercise, 8 controls) were excluded from the analysis of risk for cesarean birth. Preterm birth was defined as a delivery < 37 weeks of gestation as previously established by medical professionals as well as research showing an association with poorer health outcomes 21.
Statistical Analysis: Between-group differences in maternal demographic characteristics, pregnancy and delivery outcomes were determined via Student’s t-tests and Pearson Chi-Square tests, where appropriate. Between-group differences were analyzed using intention-to-treat analysis as well as the per-protocol approach based on those who were exercise adherent. Conditional distributions evaluated for the assumptions of linear and Poisson regression were satisfied. Following intention-to-treat, ANCOVA and Poisson regression models were performed to evaluate the effects of prenatal exercise on the association between maternal BMI and pregnancy as well as delivery outcomes. The primary outcomes for this study were: occurrence of non-elective cesarean births, birth weight (kg; continuous), preterm (< 37 or ≥ 37 weeks). The main effects between maternal BMI and the outcomes of interest were assessed first, followed by the effects of prenatal exercise (MET∙min∙week− 1) to evaluate the potential attenuation of the relationship between maternal BMI and delivery outcomes consequent to prenatal exercise. Where appropriate, the following covariates were then considered: maternal race, age, gestational age, GWG, aerobic capacity, and parity. Statistical analyses were performed using SAS, version 9.4 (Cary, NC). Statistical significance was determined a priori at p < 0.05 with two-tailed analyses.