The Preventive Effect of Physical Activity on Gestational Diabetes Mellitus: A Korean Longitudinal Prospective Cohort Study
Article information
Abstract
Background
To assess longitudinally physical activity patterns and intensity from pre-pregnancy to postpartum and evaluate the association between timing and type of physical activity and the development of gestational diabetes mellitus (GDM).
Methods
The Korean Pregnancy Outcome Study (KPOS) is a prospective cohort study conducted from 2013 to 2017. Our study included 3,457 participants with singleton pregnancies in KPOS, classified according to the pregnancy outcome: GDM (n=231) and normal (n=3,226). Physical activity data were collected at five time points using the short form of the International Physical Activity Questionnaire: before pregnancy, at 12, 24, and 36 gestational weeks (GW), and 6–8 weeks postpartum. Pre-pregnancy physical activity was collected through recall at 12 GW.
Results
Maternal age, pre-pregnancy body mass index, educational status, smoking, mini dietary assessment index, first-degree family history of diabetes, hypertension, parity, pre-existing GDM, and previous macrosomia showed significant differences between the GDM group and the normal group (P<0.05 for all). Pre-pregnancy muscle-strengthening activity was significantly associated with a lower risk of GDM (adjusted odds ratio, 0.46; 95% confidence interval, 0.25 to 0.85).
Conclusion
These findings suggest that physical activity, such as muscle-strengthening activities before pregnancy, could be a preventive strategy to reduce GDM risk. Although the study does not provide evidence that physical activity during pregnancy and postpartum reduces GDM risk, it offers significant insights into the importance of maintaining a healthy level of physical activity from pre-pregnancy to prevent GDM.
Highlights
• GDM is a common pregnancy complication caused by metabolic abnormalities.
• GDM poses significant risks to both expectant mothers and their babies.
• Physical activity was prospectively assessed from before pregnancy to postpartum.
• Pre-pregnancy muscle-strengthening activities lowered the risk of developing GDM.
INTRODUCTION
Gestational diabetes mellitus (GDM) is the development of carbohydrate intolerance during pregnancy [1]. GDM poses substantial risks for both expectant mothers and their babies, leading to various complications. Recognized risk factors for GDM include being overweight or obese, physical inactivity, having a first-degree relative with diabetes, race or ethnicity, history of fetal macrosomia, previous GDM, hypertension, history of cardiovascular disease, and polycystic ovary syndrome [1]. However, many of these GDM risk factors are not modifiable. Only overweight, obesity, and physical inactivity are modifiable through lifestyle changes. Therefore, physical activity is one of the main targets for reducing GDM. The American College of Obstetricians and Gynecologists recommended that physical activity during pregnancy is safe and desirable, and pregnant women should be encouraged to continue or start safe physical activity. They also suggested that further research is needed to determine the optimal type, intensity, and frequency of physical activity [2]. Although recent meta-analyses suggest that physical activity during pregnancy helps prevent GDM, inconsistencies remain regarding the timing, type, and intensity of physical activity [3-6].
Moreover, despite the obvious benefits, pregnant people worldwide exhibit reduced physical activity compared to prepregnancy, with more pronounced declines observed among Asian women [7-9]. Further research is essential to understand these physical activity patterns and establish specific evidence-based guidelines for preventing GDM. Therefore, this study aims to longitudinally assess physical activity patterns and intensity from pre-pregnancy to postpartum and evaluate the association between timing and type of physical activity and the development of GDM.
METHODS
Study population
This study used a Korean Pregnancy Outcome Study (KPOS) dataset, a longitudinal prospective cohort study examining the epidemiology of pregnancy-related complications and risk factors, conducted by two secondary hospitals in Seoul, Korea [10]. The KPOS was conducted between 2013 and 2017 and included Korean pregnant participants who visited study hospitals for prenatal care at approximately 12 gestational weeks (GW). Pregnancies with triplets or more were excluded. The analysis included 3,457 participants with singleton pregnancies in KPOS, classified according to pregnancy outcomes: GDM (n=231) and normal (n=3,226) (Supplementary Fig. 1). GDM was diagnosed using a two-step method. Universal screening with a 50 g glucose challenge test (GCT) was conducted between 24 and 28 GW. If the result of the GCT was 140 mg/dL or more, a confirmatory oral glucose tolerance test was performed [10]. The KPOS study was approved by the Institutional Review Boards (IRB) of Cheil General Hospital (CGH-IRB-2013–10) and CHA Hospital (2013-14-NC13–018). This analysis using data from the KPOS study was also approved by the CHA Bundang Medical Center (CHAMC 2021-04-052).
Demographic data analysis
The study analyzed demographic data from the KPOS dataset. Socio-demographic information (maternal age, educational status, and household income) and health-related behaviors (smoking, alcohol consumption, and dietary behaviors) were collected through interview-based questionnaires under the supervision of trained interviewers. Information on past pregnancies (parity, pre-existing GDM, and previous macrosomia), as well as medical history (hypertension and polycystic ovary syndrome) and first-degree family history of diabetes were gathered by interview-based questionnaires or clinical data from medical records [10]. The smoking status was categorized into three groups: ‘never smoked’ (those who had never smoked), ‘pre-pregnancy smoking’ (those who smoked before pregnancy and quit after becoming pregnant), and ‘smoking in early pregnancy’ (those who continued smoking after becoming pregnant and were still smoking at the time of enrollment). Alcohol consumption was categorized into two groups: ‘never drank’ (those who had never consumed alcohol) and ‘drinker’ (those who drank before or during pregnancy). Dietary behaviors were assessed using the validated mini dietary assessment (MDA) index, which consists of 10 items. Each item is rated on a 5-point Likert scale (1=not at all true; 5=very true), with a maximum total score of 50 [11].
Assessment of physical activity
Self-reported physical activity data from the International Physical Activity Questionnaire Short Form (IPAQ-SF) were analyzed at five different time points: pre-pregnancy, at 12, 24, and 36 GW, and 6–8 weeks postpartum [12]. Pre-pregnancy physical activity was collected through recall at 12 GW. In longitudinal data, there were missing responses to physical activity questionnaires at five time points. In the analysis, only participants with a complete physical activity questionnaire for all items were included. Therefore, there was a discrepancy in the overall totals at five time points (Supplementary Fig. 1).
The intensity levels of physical activity were calculated using min/week and the metabolic equivalent of task (MET) values using the formula for the different activity categories: 8.0 MET for vigorous activity, 4.0 MET for moderate activity, and 3.3 MET for walking, respectively [13]. In addition to the IPAQ-SF questions, we examined strength training (including exercises such as push-ups, sit-ups, dumbbells, weights, and barbells) in the past 7 days and multiplied the results by the 2024 Adult Compendium of Physical Activities’ muscle-strengthening activity equivalent of 5.0 METs [14].
We evaluated the frequency of individuals achieving four key performance indicators for physical activity, according to criteria in Korea Health Statistics 2020 [15]: (1) walking performance: a minimum of 10 minutes in a single session and at least 30 minutes per day for a minimum of 5 days in the past week; (2) moderate to vigorous physical activity (MVPA) performance: in moderate-intensity physical activity for 2 hours and 30 minutes or vigorous-intensity physical activity for 1 hour and 15 minutes per week or a combination of moderate and vigorous-intensity physical activity (1 minute of vigorous-intensity equals 2 minutes of moderate-intensity) equivalent to the prescribed duration for each activity in a week; (3) muscle-strengthening activity performance: at least 2 days in the past week; and (4) performance for both MVPA and muscle-strengthening activity. MVPA performance is the same measure as 150 minutes of MVPA per week, which is the amount of physical activity recommended for pregnant people by the World Health Organization and others. We further analyzed the key performance indicators of physical activity by maternal age and pre-pregnancy body mass index (BMI) subgroups.
Statistical analysis
Continuous variables are presented as average±standard deviation or 95% confidence interval (CI), while categorical variables are presented as numbers and percentages. Differences in the averages between the two groups were evaluated using independent sample t-tests or the Mann-Whitney U test. Pearson chi-square test was used for the analysis of categorical variables. Adjusted odds ratios (OR) for each physical activity category were calculated using binomial logistic regression analysis, with adjustments for potential confounders. Propensity score matching (PSM) was performed using a 1:1 ratio in R with the MatchIt package (R Foundation for Statistical Computing, Vienna, Austria), applying nearest-neighbor matching to adjust for the potential confounders when comparing the GDM and normal groups. A P<0.05 or 95% CI not including 1.0 was considered statistically significant. The data were analyzed using SPSS version 28.0 software (IBM Co., Armonk, NY, USA).
RESULTS
Participant characteristics at baseline
Table 1 summarizes the baseline characteristics. Physical activity was analyzed in a total of 3,457 participants in the KPOS dataset across five time periods, from pre-pregnancy to postpartum. Among these participants, 231 were diagnosed with GDM, resulting in a GDM prevalence of 6.7% (Supplementary Fig. 1). Maternal age, pre-pregnancy BMI, educational status, smoking, MDA index, hypertension, first-degree family history of diabetes mellitus, parity, pre-existing GDM, and previous macrosomia were significantly higher in the GDM group compared to the normal group (P<0.05 for all). Household income, alcohol consumption, and polycystic ovary syndrome did not differ significantly.
Baseline characteristics at enrollment
Analysis of physical activity intensity
Physical activity decreased rapidly from pre-pregnancy through the first trimester and into the postpartum period (Fig. 1). Among the various activities, only pre-pregnancy muscle-strengthening activity was significantly lower in the GDM group compared to the normal group (P<0.05) (Fig. 1, Supplementary Table 1).
The intensity of moderate to vigorous physical activity and muscle-strengthening activity between gestational diabetes mellitus (GDM) and normal from pre-pregnancy to postpartum. (A) Moderate to vigorous physical activity. (B) Muscle-strengthening activity. MET, metabolic equivalent of task. aP<0.05 by Mann-Whitney U test.
Frequency analysis of physical activity key performance indicators
Table 2 presents the frequency analysis results using key performance indicators of physical activity. During pre-pregnancy, walking (36.3% vs. 44.5%, P=0.016), muscle-strengthening activity (5.3% vs. 11.6%, P=0.004), and both MVPA and muscle-strengthening activity (4.9% vs. 10.6%, P=0.006) were significantly lower in the GDM group compared to the normal group. No significant differences were observed between the groups during pregnancy and postpartum. Furthermore, the proportion of participants meeting the recommended amount of physical activity (i.e., MVPA performance) ranged from 59.7% to 60.7% before pregnancy, 30.7% to 50.0% during pregnancy, and 13.9% to 14.9% postpartum.
The frequency of key performance indicators of physical activity between GDM and normal from pre-pregnancy to postpartum
Association of physical activity with GDM risk
To determine whether participation in key performance indicators of physical activity was associated with GDM, we performed logistic regression analysis after adjusting for maternal age, pre-pregnancy BMI, educational status, smoking, MDA index, hypertension, first-degree family history of diabetes mellitus, and pre-existing GDM (Table 3). Pre-pregnancy muscle-strengthening activity was significantly associated with a reduced risk of GDM (adjusted OR, 0.46; 95% CI, 0.25 to 0.85). Furthermore, meeting both MVPA and muscle-strengthening activity guidelines was associated with a lower risk of GDM development (adjusted OR, 0.46; 95% CI, 0.25 to 0.87).
Odds ratios and 95% confidence intervals for the risk of GDM by meeting key performance indicators of physical activity in pre-pregnancy
To correct for potential confounding in the longitudinal observational data, PSM analysis was conducted [16]. Using a 1:1 PSM, 226 GDM cases were matched with 226 normal controls, resulting in a total of 452 participants included in the analysis. No significant differences in baseline characteristics were observed between the two groups, indicating that the PSM was performed appropriately (Supplementary Table 2). Additionally, logistic regression analysis of the 452 participants confirmed that pre-pregnancy muscle-strengthening activity significantly reduced the risk of GDM (adjusted OR, 0.44; 95% CI, 0.20 to 0.94), consistent with the primary analysis results without PSM (Supplementary Table 3).
Subgroup analysis by maternal age and pre-pregnancy BMI
Subgroup analyses were conducted by dividing participants into two groups based on maternal age (<35 and ≥35 years) and into three groups based on pre-pregnancy BMI: normal (<23 kg/m2), overweight (23 to 24.9 kg/m2), and obese (≥25 kg/m2). The ORs and 95% CI were analyzed for the association between meeting key performance indicators of physical activity and GDM in these subgroups. The interaction P values for maternal age and pre-pregnancy BMI subgroups were not statistically significant (P>0.05 for all) (Fig. 2), indicating that the protective effect of muscle-strengthening activities against GDM is not influenced by maternal age or pre-pregnancy BMI stratification.
Subgroup analysis of odds ratio (OR) and 95% confidence interval (CI) for gestational diabetes mellitus risk by key performance indicators of pre-pregnancy physical activity. (A) Maternal age. (B) Pre-pregnancy body mass index (BMI). MVPA, moderate to vigorous physical activity.
DISCUSSION
The main results of our study demonstrate that participants who engaged in muscle-strengthening activity before pregnancy experienced a protective effect against the risk of developing GDM. Subgroup analysis based on maternal age and pre-pregnancy BMI revealed that the protective effect of muscle-strengthening activity was independent of both factors. These findings align with previous studies suggesting that prepregnancy resistance training may reduce the risk of GDM in early pregnancy [5]. Furthermore, a randomized controlled trial on Chinese patients with GDM reported that resistance training is more suitable for pregnant women than aerobic exercise [17].
GDM, a form of glucose intolerance detected during pregnancy, shares similarities with type 2 diabetes mellitus, with insulin resistance being a key feature. Skeletal muscle plays a crucial role in insulin-mediated glucose uptake after meals and is essential for glycogen synthesis [18]. Increasing muscle mass through resistance training enhances glucose transporter protein type-4 (GLUT-4) sensitivity to insulin, resulting in increased glucose uptake by muscles and improved glycemic control [19]. These mechanisms may explain the association between muscle-strengthening activities and a reduced risk of GDM observed in our study.
The importance and benefits of physical activity during pregnancy and the postpartum period are well-established in previous studies and guidelines [2]. However, several studies have demonstrated a decrease in physical activity during pregnancy compared to pre-pregnancy levels [7,20-22]. Research analyzing physical activity by race and ethnicity has shown a more significant decline in Asian women compared to Western women during pregnancy [23-25]. Similarly, our results show a decrease in physical activity during pregnancy and postpartum compared to pre-pregnancy levels. This reduction in physical activity may explain why the preventive effect of physical activity on GDM was not observed during pregnancy and postpartum in our study. Although the findings did not confirm the importance of physical activity during pregnancy and postpartum, we suggest that pre-pregnancy muscle-strengthening activities could serve as a valuable strategy for GDM prevention in the childbearing population.
The strengths of this study include its prospective cohort design, with a relatively large sample size participants, and the longitudinal assessment of physical activity at multiple time points. Specifically, we analyzed both the intensity (MET-min/week) and frequency of key performance indicators of physical activity. A key strength of this study is the inclusion of potential confounding variables, such as dietary factors and disease history, allowing for a more comprehensive understanding of the relationship between GDM and physical activity. Additionally, the PSM approach used in this study provided greater statistical precision and control for confounding variables compared to traditional multivariate regression analysis, allowing for more powerful adjustment for potential confounding variables. However, this study has several limitations. First, the assessment of physical activity relied on questionnaires, which may be subject to recall bias and less accurate than objective measurement tools. Second, missing data in the physical activity questionnaires at each time point led to a reduced sample size. Third, the study population was limited to Korean women, and the sample was drawn from individuals who visited the research institutions, potentially introducing selection bias and limiting the generalizability of the findings. Finally, other variables not included in the analysis may influence the relationship between physical activity and GDM risk. Despite these limitations, our study benefits from the prospective follow-up of a large cohort of pregnant individuals in South Korea. This allowed us to analyze the association between physical activity, including muscle-strengthening activities, and GDM risk through continuous evaluation throughout pregnancy.
In conclusion, promoting muscle-strengthening activities before pregnancy may serve as a crucial public health strategy for reducing the incidence of GDM, ultimately improving maternal and fetal health outcomes. Our findings indicate that engaging in muscle-strengthening activities before pregnancy reduces the risk of developing GDM, regardless of maternal age or pre-pregnancy BMI. Educating women of childbearing age on the benefits of physical activity, particularly muscle-strengthening activities, could significantly contribute to reducing the risk of GDM.
SUPPLEMENTARY MATERIALS
Supplementary materials related to this article can be found online at https://doi.org/10.4093/dmj.2024.0280.
Physical activity intensity (MET-min/week) between GDM and normal group from pre-pregnancy to postpartum
Baseline characteristics at enrollment after propensity score matching
Odds ratios and 95% confidence intervals for the risk of GDM by met key performance indicators of physical activity in pre-pregnancy after propensity score matching
Flow chart of study participant selection process. GDM, gestational diabetes mellitus.
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: M.Y.K., H.M.R.
Acquisition, analysis, or interpretation of data: J.H.L., M.H.K., H.J.H., Y.J.H., J.H.C., D.W.K., M.Y.K., H.M.R.
Drafting the work or revising: J.H.L., M.H.K.
Final approval of the manuscript: all authors.
FUNDING
This work was supported by grants from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI21C1779 and RS2022-KH129953).
ACKNOWLEDGMENTS
We express our sincere gratitude to the Division of Maternal-Fetal Medicine in Cheil General Hospital and CHA Hospital who were involved in the KPOS research (the Korea National Institute of Health research project: 2015-E6302-00).