Women diagnosed with gestational diabetes mellitus (GDM) are at increased risk of pregnancy complications1 as well as developing type 2 diabetes and metabolic syndrome in later life.2 For the fetus, elevated maternal glucose concentrations promote excessive growth and increased birth weight, increasing the risk of macrosomia, birth injury, and admission to a neonatal care unit.1 Later in life, these offspring have an increased prevalence of obesity, type 2 diabetes, and metabolic syndrome, perpetuating the burden of diabetes across generations.2
Epidemiological data indicate that being physically active before and during pregnancy is associated with a reduced risk of GDM,3 and experimental studies have shown that regular exercise during pregnancy may attenuate the typical decline in glucose tolerance.4 However, few randomized controlled trials have investigated whether regular exercise can prevent GDM,5–7 and no trials have focused on women with a history of the condition who have a high risk of recurrence (approximately 48%).8 Furthermore, compliance to prescribed exercise interventions has been reported as a limiting factor during pregnancy.7 Common barriers to exercise in this population include a lack of time associated with managing a household and caring for existing children, tiredness, physical limitations, and concerns about safety.9,10 Supervised, home-based exercise training may overcome many of these barriers.4,11 Accordingly, the aim of this study was to investigate the effect of a 14-week supervised, home-based exercise program started at 14 weeks of gestation on the recurrence of GDM. The effect of the intervention on maternal cardiovascular fitness, body anthropometrics, psychological well-being, and obstetric and neonatal outcomes was also examined.
MATERIALS AND METHODS
This single-centered randomized controlled trial was conducted in Perth, Australia, and registered at www.clinicaltrials.gov (identifier: NCT01283854). The study was approved by the Women and Newborn Health Service Ethics Committee and conducted in accordance with the Consolidated Standards of Reporting Trials Statement with the first patient enrolled on June 22, 2011. Pregnant women with a history of GDM in a previous pregnancy were recruited through antenatal clinics, obstetricians, general practitioners, and ultrasound practices. Participants were eligible for inclusion if they were less than 14 weeks of gestation, older than 18 years of age, and able to participate in a 14-week exercise program. Women with pre-existing diabetes, elevated baseline oral glucose tolerance test (OGTT), multiple pregnancy or a medical condition that restricted exercise participation, those taking oral hypoglycemic agents, or those who were already engaged in a structured exercise program were not eligible for the study. All participants provided written informed consent.
Eligible women were randomized between 12 and 14 weeks of gestation to an exercise intervention or control using a custom-designed computer program on a dedicated laptop that stratified by body mass index (calculated as weight (kg)/[height (m)]2, less than 30, 30–34.9, or greater than 35) and maternal age (younger than or 35 years or older). Randomization was performed during the first visit to the laboratory with the woman entering the stratification factors into the program to be informed of her allocation by the next screen. Participants randomized to exercise engaged in a 14-week stationary cycling program starting 7 days later. An upright cycle ergometer was delivered to each participant's home for her use during the intervention. All sessions were supervised by an exercise physiologist who would travel to the woman's home three times each week to monitor the duration and intensity of exercise.
Each session commenced with a 5-minute warm-up consisting of pedaling at an intensity that equated to 55–65% of age-predicted maximum heart rate and a rating of perceived exertion of 9–11 on the Borg scale.12 This scale ranges from 6 (no exertion) to 20 (where exercise is perceived to be “very very hard”). The subsequent conditioning period was divided into 5-minute periods of continuous moderate-intensity cycling (65–75% maximum heart rate; rating of perceived exertion 12–13) alternating with 5-minute periods of interval cycling. Two types of intervals were used; one involved an increase in pedaling rate for 15 seconds and the other involved an increase in cycling resistance for 30 seconds (target intensity 75–85% maximum heart rate; rating of perceived exertion 14–16) repeated every 2 minutes. A 5-minute cool down concluded each session (55–65% maximum heart rate; rating of perceived exertion 9–11) followed by light stretching. The duration of each session was progressively increased by 5-minute increments every 2–3 weeks, as tolerated, from 20 to 30 minutes to a maximum session duration of 60 minutes. The degree of progression was dependent on the baseline fitness level of the woman and her ongoing pregnancy symptoms.
Each participant was required to visit an independent local accredited pathology center (PathWest Laboratory Medicine, Perth, Australia) in the fasted state to complete a 75-g OGTT both preintervention and postintervention (at least 48 hours after the last exercise session). Diagnosis of GDM at the postintervention OGTT was based on the criteria adopted in Western Australia at the time of the trial (fasting venous blood glucose 5.5 mmol/L or greater [99 mg/dL] or a 2‐hour OGTT glucose 8.0 mmol/L or greater [144 mg/dL] or both). Fasting glucose and insulin concentrations, together with glucose tolerance and insulin sensitivity, were determined from the 75-g OGTT. Insulin sensitivity was determined using the homeostatic model of assessment13 and the insulin sensitivity index.14 Glycosylated hemoglobin, serum c-peptide, and cholesterol concentrations were also measured from the fasting venous blood sample. Postintervention testing was completed within a week of cessation of the intervention, but at least 48 hours after the last exercise session.
On a separate occasion, participants visited our laboratory for the assessment of secondary outcome measures preintervention and postintervention. Height, body mass, five peripheral skinfolds (biceps, triceps, subscapular, midthigh, and calf), four limb girths (upper relaxed arm, upper flexed arm, thigh, and calf), resting heart rate, and blood pressure were measured. In addition, cardiovascular fitness was assessed based on the heart rate and oxygen consumption responses to submaximal exercise on a stationary cycle ergometer. Briefly, this test involved progressive exercise until a heart rate equivalent to 75% of age-predicted maximum was attained.11 Oxygen consumption was measured for the duration of the test with fitness expressed as both power output (W) and oxygen consumption at 75% of maximum heart rate. During the same session, the Edinburgh Postnatal Depression Scale15 was administered along with the 21-item Depression Anxiety Stress Scale16 to assess general psychological distress,17 and the Social Physique Anxiety Scale18 monitored social anxiety relating to physique.
Each participant also completed a 7-day food diary preintervention and postintervention for assessment of mean daily nutritional intake (total energy, carbohydrate, fat, protein, sugar, and fiber. Physical activity (daily steps, time spent sedentary and in moderate-intensity activity) was monitored for the same 7-day period using an accelerometer. Changes in exercise habit strength (patterning of action, automaticity, stimulus-response bonds, and negative consequences for nonperformance) were assessed using the Exercise Habit Strength Questionnaire.19 After delivery, relevant obstetric and neonatal outcomes were extracted from hospital records.
Data from King Edward Memorial Hospital (Perth, Western Australia) indicated that the risk of GDM recurrence was 55% in our population, a figure consistent with the published literature.8 A sample size of 180 (90 women in each group) was required to attain 80% power to detect a 40% reduction in the incidence of GDM, from 55% to 33%, when performing a test of proportions at the 5% significance level. Continuous data were summarized using means and standard deviation, or medians and interquartile ranges, depending on data normality, and were univariately compared using a t test or Mann-Whitney test. Categorical data were summarized using frequency distributions. The incidence of GDM and other categorical outcomes was compared using a χ2 test, except for low-frequency outcomes, which were compared using a Fisher exact test. The glucose and insulin responses to the OGTT were compared using three-way (group×pre–post×time) repeated-measures analysis of variance. Maternal fitness, body anthropometrics, psychological well-being, physical activity levels, and daily nutritional intake were compared using two-way (group×pre-post) repeated-measures analysis of variance. Extreme underreporters were identified using the Goldberg method as per Black20 and excluded from the dietary analysis (8% of entries). Statistical analysis was conducted based on intention-to-treat using SPSS. All hypothesis tests were two-sided and conducted at a 5% significance level.
A total of 318 women were assessed for eligibility for the study between June 1, 2011, and July 31, 2014 (Fig. 1), and 205 provided written consent to participate and completed baseline assessments at 13±1 weeks of gestation. Of these 205 women, 33 were excluded from the trial as a result of an elevated OGTT at baseline, leaving 172 randomized to exercise (n=85) or control (n=87). The baseline characteristics of these women are shown in Table 1. The majority were of Caucasian ethnicity and lived in geographic areas in the top quintile for relative socioeconomic advantage. There were no differences in glucose regulation between groups at baseline. Two women experienced pregnancy loss at 19 weeks of gestation and one at 21 weeks of gestation (control=2; exercise=1). All remaining women completed the postintervention OGTT (28±1 weeks of gestation); however, 12 women failed to complete the postintervention assessment of fitness, body anthropometrics, and psychological well-being (secondary outcomes). The final participant delivered on January 26, 2015, completing the period of data collection. Compliance to the exercise intervention was high (median 86% of sessions completed; interquartile range 79–95%). The mean heart rate during exercise was 130±10 beats per minute (70±5% age-predicted maximum). Participants reported a mean session rating of perceived exertion of 13±1 indicating that exercise intensity was perceived as “somewhat hard.” Exercise duration increased from 28±4 minutes per session in the first week to 47±11 minutes in the last week of the intervention.
The recurrence rate of GDM was similar between groups (40% control; 40.5% exercise; relative risk 1.01, 95% confidence interval 0.70–1.46; P=.950; Table 2) and there was no difference in the severity of GDM based on the overall degree of glucose tolerance or insulin response to the OGTT between groups postintervention. Homeostatic model of assessment, insulin sensitivity index, glycosylated hemoglobin, and serum C-peptide were also similar between groups postintervention (Table 2). Advancing pregnancy was associated with an increase in resting heart rate, blood triglycerides, low-density lipoprotein and high-density lipoprotein cholesterol, and a decrease in systolic and diastolic blood pressure (P<.001); however, there were no differences between groups (Table 3). Similarly, body mass, sum of skinfolds, and girths increased as pregnancy progressed (P<.001) with no difference between groups. Maternal cardiovascular fitness, based on the power output at 75% maximum heart rate, was increased in response to exercise (P<.01), resulting in higher fitness in the exercise group compared with women in the control group postintervention (P<.01). Oxygen consumption at 75% maximum heart rate also tended to increase after the exercise intervention (P=.05; Table 3).
There were no changes in Edinburgh Postnatal Depression Scale scores over time and no differences between groups (Table 4). Similarly, the number of participants with Edinburgh Postnatal Depression Scale scores 12 or greater (indicating higher risk for depression in pregnancy) was similar between groups. In contrast, a significant difference in the 21-item Depression Anxiety Stress Scale psychologic distress scores was noted between groups postintervention (P=.04) with a reduction in response to exercise but no change in the control group. Social Physique Anxiety did not change over time and was similar between groups (Table 4). Daily nutritional intake was unaltered over time and was similar between groups preintervention and postintervention (P>.05; Table 5). With respect to daily physical activity levels, the mean number of steps taken decreased and the time spent sedentary increased as pregnancy progressed (P<.05; Table 5), but there were no differences preintervention or postintervention between groups. Time spent in moderate activity outside of the intervention was similar over time and between groups. Of note, women performing more than 20 minutes of moderate-intensity physical activity (50th percentile) at study entry had a reduced incidence of GDM independent of their group allocation compared with women performing less than 20 minutes of daily moderate activity (29% compared with 49%; P=.013). The Habit Strength Questionnaire revealed an increase in the patterning of action and exercise automaticity from preintervention to postintervention in the exercise group but no change in women in the control group (P<.05; Table 5). There were no differences in obstetric or neonatal outcomes between groups (Table 6) except that there were more male neonates born to participants in the exercise group compared with women in the control group (P<.01).
A 14-week supervised, home-based exercise program started at 14 weeks of gestation did not reduce the recurrence of GDM nor did it alter the overall degree of glucose intolerance or insulin sensitivity. However, the intervention was associated with improvements in maternal cardiovascular fitness, increases in exercise automaticity, and a reduction in general psychological distress indicated by the 21-item Depression Anxiety Stress Scale. Impaired glucose tolerance was identified before 14 weeks of gestation in 16% of women consenting to the study, suggesting that screening for GDM in women with a history of the condition may need to start earlier than is currently practiced.
The lack of effect of the intervention on the recurrence of GDM is surprising given the epidemiological data to support a reduction in GDM risk with increasing physical activity3 together with experimental studies reporting benefits of regular exercise during pregnancy for glucose tolerance.4 However, the few randomized controlled trials investigating the effect of exercise for GDM prevention have reported conflicting results, with some trials observing no effect of antenatal exercise (moderate-intensity aerobic, strength and flexibility exercises performed three times a week for 45–60 minutes) on the incidence of GDM,5,7 whereas Cordero et al6 reported reduced prevalence of GDM with a program of aerobic, strength, and flexibility exercises performed three times per week started from 10 to 14 weeks of gestation. The lack of benefit of our intervention for decreasing the recurrence of GDM was unlikely as a result of low compliance given the fully supervised home-based design. The exercise prescription was based on previous work in women with and without GDM4,11,21; however, it is possible that increasing the frequency of exercise may have altered the results, and the optimal duration and intensity of exercise for the prevention of GDM is not known. It is also unlikely that the lack of effect of exercise on GDM was the result of compensatory changes in daily nutritional intake or physical activity levels outside of the intervention given the lack of difference between groups, although it should be acknowledged that these measures were taken preintervention and postintervention only, and that self-reported food diaries are susceptible to underreporting, which is likely given the lower than expected energy intake seen here. It must also be noted that the exercise group gave birth to significantly more male neonates. This coincidental occurrence may have influenced the rate of GDM given that women carrying a male fetus may have an increased risk of GDM.22
The present intervention started at 14 weeks of gestation. Although this was initially considered a strength of the trial, it has recently been suggested that placental function and gene expression are programmed by the first trimester.23 Accordingly, future research could focus on overcoming the challenges of recruiting earlier in pregnancy and even before conception. In support of this notion, women performing more than 20 minutes of moderate-intensity physical activity at study entry had a reduced incidence of GDM independent of their group allocation. Consideration of the baseline physical activity levels of our sample also highlights that the women studied here may not be representative of the general population of pregnant women at risk of GDM. At study entry, the present sample accumulated a mean of 24 minutes of daily moderate-intensity physical activity, which is higher than that reported in the general population of pregnant women in Australia.24 Moreover, 57% of our sample lived in geographic areas of social advantage, 72% were under private obstetric care, and approximately 80% had body mass indexes less than 30 kg/m2. These indicators of socioeconomic status may influence awareness and commitment to health.9,25 In support of this notion, the GDM recurrence rate of 40% observed in our sample is lower than the expected recurrence rate based on our hospital data (55%). Indeed, volunteering for a 14-week exercise intervention itself may introduce self-selection bias, potentially suggesting a proactive approach to avoiding a second GDM diagnosis.
Despite a lack of effect of the exercise intervention on the recurrence or severity of GDM, the improvement in cardiovascular fitness is of great importance given the beneficial relationship between exercise during pregnancy and higher fitness and reduced cardiovascular risk factors later in life.26 Meanwhile, the reduction in general psychological distress indicated by the 21-item Depression Anxiety Stress Scale in the exercise group is of interest given the adverse effects of maternal distress on development in utero.27 Regardless of these benefits, there was no difference in obstetric and neonatal outcomes between groups, although the present study was not powered for these outcomes. The present intervention ceased at 28 weeks of gestation to coincide with assessment of the primary outcome measure; however, the effect of continuing an intervention of this nature until delivery remains to be determined.
In summary, supervised home-based exercise started after the first trimester of pregnancy does not reduce the recurrence of GDM or the degree of decline in glucose tolerance in women with a history of the condition at 28 weeks of gestation; however, it does benefit maternal fitness and psychological well-being. Future studies should seek to implement exercise interventions before conception, extend through to delivery, and focus on women with the lowest health, socioeconomic status, and baseline physical activity levels. Regardless of the lack of benefit observed here for the prevention of GDM, the benefits of regular exercise for the management of GDM after diagnosis are well-established.11,21
1. HAPO Study Cooperative Research Group, Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, Coustan DR, et al Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008;358:1991–2002.
2. Buchanan TA, Xiang AH, Page KA. Gestational diabetes mellitus: risks and management during and after pregnancy. Nat Rev Endocrinol 2012;8:639–49.
3. Dempsey JC, Sorensen TK, Williams MA, Lee IM, Miller RS, Dashow EE, et al Prospective study of gestational diabetes mellitus risk in relation to maternal recreational physical activity before and during pregnancy. Am J Epidemiol 2004;159:663–70.
4. Ong MJ, Guelfi KJ, Hunter T, Wallman KE, Fournier PA, Newnham JP. Supervised home-based exercise may attenuate the decline of glucose tolerance in obese pregnant women. Diabetes Metab 2009;35:418–21.
5. Barakat R, Pelaez M, Lopez C, Lucia A, Ruiz JR. Exercise during pregnancy and gestational diabetes-related adverse effects: a randomized controlled trial. Br J Sports Med 2013;47:630–6.
6. Cordero Y, Mottola MF, Vargas J, Blanco M, Barakat R. Exercise is associated with a reduction in gestational diabetes mellitus. Med Sci Sports Exerc 2015;47:1328–33.
7. Stafne SN, Salvesen KÅ, Romundstad PR, Eggebø TM, Carlsen SM, Mørkved S. Regular exercise during pregnancy to prevent gestational diabetes: a randomized controlled trial. Obstet Gynecol 2012;119:29–36.
8. Schwartz N, Nachum Z, Green MS. The prevalence of gestational diabetes mellitus recurrence—effect of ethnicity and parity: a metaanalysis. Am J Obstet Gynecol 2015;213:310–7.
9. Gaston A, Cramp A. Exercise during pregnancy: a review of patterns and determinants. J Sci Med Sport 2011;14:299–305.
10. Guelfi KJ, Wang C, Dimmock JA, Jackson B, Newnham JP, Yang H. A comparison of beliefs about exercise during pregnancy between Chinese and Australian pregnant women. BMC Pregnancy Childbirth 2015;15:345.
11. Halse RE, Wallman KE, Dimmock JA, Newnham JP, Guelfi KJ. Home-based exercise improves fitness and exercise attitude and intention in women with GDM. Med Sci Sports Exerc 2015;47:1698–704.
12. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 1982;14:377–81.
13. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–9.
14. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999;22:1462–70.
15. Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry 1987;150:782–6.
16. Henry JD, Crawford JR. The short-form version of the Depression Anxiety Stress Scales (DASS-21): construct validity and normative data in a large non-clinical sample. Br J Clin Psychol 2005;44:227–39.
17. Osman A, Wong JL, Bagge CL, Freedenthal S, Gutierrez PM, Lozano G. The Depression Anxiety Stress Scales-21 (DASS-21): further examination of dimensions, scale reliability, and correlates. J Clin Psychol 2012;68:1322–38.
18. Hart EA, Leary MR, Rejeski WJ. The measurement of social physique anxiety. J Sport Exerc Psychol 1989;11:94–104.
19. Grove JR, Zillich I, Medic N. A process-oriented measure of habit strength for moderate-to-vigorous physical activity. Health Psychol Behav Med 2014;2:379–89.
20. Black AE. Critical evaluation of energy intake using the Goldberg cut-off for energy intake: basal metabolic rate. A practical guide to its calculation, use and limitations. Int J Obes 2000;24:1119–30.
21. Halse RE, Wallman KE, Newnham JP, Guelfi KJ. Home-based exercise training improves capillary glucose profile in women with gestational diabetes. Med Sci Sports Exerc 2014;46:1702–9.
22. Retnakaran R, Kramer CK, Ye C, Kew S, Hanley AJ, Connelly PW, et al Fetal sex and maternal risk of gestational diabetes mellitus: the impact of having a boy. Diabetes Care 2015;38:844–51.
23. Catalano P, deMouzon SH. Maternal obesity and metabolic risk to the offspring: why lifestyle interventions may have not achieved the desired outcomes. Int J Obes 2015;39:642–9.
24. Wilkinson SA, Miller YD, Watson B. Prevalence of health behaviours in pregnancy at service entry in a Queensland health service district. Aust N Z J Public Health 2009;33:228–33.
25. Adler NE, Ostrove JM. Socioeconomic status and health: what we know and what we don't. Ann N Y Acad Sci 1999;896:3–15.
26. Clapp JF 3rd. Long-term outcome after exercising throughout pregnancy: fitness and cardiovascular risk. Am J Obstet Gynecol 2008;199:489.e1–6.
© 2016 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
27. Dunkel Schetter C, Tanner L. Anxiety, depression and stress in pregnancy: implications for mothers, children, research, and practice. Curr Opin Psychiatry 2012;25:141–8.