A pilot study with quasi-experimental study design was conducted among women with GDM at tertiary care Hospital of Father Muller Medical College and Hospital, Mangaluru, Karnataka, India, between June 2019 to December 2019. The sample size was calculated based on the study conducted by Wang C et al., and 30 women with GDM were recruited [15]. Ethical clearance was obtained protocol no: 2018/183 and FMMCIEC/CCM/492/2018, followed by prior permission from the hospital authorities of respective hospitals. The 30 GDM women were selected using a simple random sampling technique for the control and intervention groups. Women who were in the obstetrics Outpatient Department and were admitted to the maternal wards, fulfilled the inclusion criteria and were enrolled as study subjects.

Inclusion criteria: Study included the women in whom GDM is diagnosed when any two values are met or elevated in Glucose Tolerance Test (GTT) [16] and confirmed by the obstetrician with BMI 18.5-30 kg/m², were in their 24-26 weeks of gestation and were having smart phone.

Exclusion criteria: The women with a bad obstetric history like Previous Intrauterine Death (IUD), Intrauterine Growth Restriction (IUGR), repeated abortions, and preterm labour, Pregestational diabetes, practicing yoga, enrolled in the gym or any other exercise regimen other than walking, voluntarily opted for Lower (uterine) Segment Caesarean section (LSCS) and/or had a previous LSCS and contraindicated for aerobic exercise (According to ACOG criteria) were excluded [17].

The informed written consent was obtained from the study participants after explaining the client information sheet. A self-developed, semi-structured interview schedule was used to record the baseline profile of the women, and a pretest was conducted by sending the women’s blood investigation to the laboratory for the glucose estimation of both the groups. Venous blood was drawn to the FBS, Post Prandial Blood Sugar (PPBS), and HbA1c test.

For the women in the interventional group, exercise were taught and observed by the trained and certified investigator on the 24^th week and re-demonstrated by the subjects. The exercises were focused on the large muscles of the upper extremities. The duration of the total exercises was 45 minutes which includes the one-minute rest after every exercise. Initially, warm-up exercises were taught for five minutes and continued with aerobic exercises, which include forward pull-ups, upright row, and shoulder press. Shoulder T lifts, triceps extension, biceps curls, and lateral raise are the resistance exercises. Followed by cool-down exercises for five minutes. Borg Rating of Perceived Exertion Scale (Champaign, IL: Human Kinetics, 1998) [18] was used to maintain the intensity of exercises. Thereafter, the subjects performed it weekly thrice with the gap of two days between three schedules until completion of 35 weeks of gestation. On the first day of the recruitment, the video of the exercise was uploaded by sharing Wi-Fi connectivity to the subjects to perform these exercises. Once in a week exercises were supervised by the investigator and the remaining days in a week, a video of the exercises were uploaded to the subject’s cell phone to perform. Subjects were asked to maintain a diary of their exercise regimen and they were motivated to perform exercises through telephonic reminders. Once in two weeks subjects were called to OPD and exercises were supervised.

For both the group regular hospital treatment was given which included medication, diet, and regular walking. Post-test was done by collecting venous blood for glycaemic scores of FBS and PPBS at 28^th week, 32^nd week, and 36^th week, also blood for HbA1c collected at 36^th week was tested at the hospital laboratory. Pregnancy outcome was measured 24 hours after the delivery by using a checklist, which includes complications of GDM, pre-eclampsia, polyhydramnios, maternal distress. One mother discontinued the exercises at 28 weeks due to an accident.

Data were tabulated, analysed, and interpreted using descriptive and inferential statistics like frequency, percentage. Mann-Whitney test, ANOVA, post-hoc test, paired Eta square were also used and IBM Statistical Package for the Social Sciences (SPSS) version 23.0 was used.

[Table/Fig-1] shows the mean age of women was 28.07±2.712 years in the interventional group and 30.20±4.73 years in the control group. Out of 15 women in the interventional group, 20% of them and 33.3% in the control group had a history of GDM in the previous pregnancy. Sixty percent of women had a family history of diabetes, out of which 88.8% were first-degree relatives. Family history of hypertension was present among 66.66% of the women in the control group and all of them were first-degree relatives. On the contrary, in the interventional group, only 40% of the women had a family history of hypertension and among them, 83.33% were first-degree relatives. Concerning the glycemia values, 42.86% (n=14) of the women in the interventional group, and 51.72% (n=15) in the control group had a normal FBS. PPBS values show 93.3% of women had increased values in both groups. In an interventional group, 50% (n=14) of the women had a normal HbA1c whereas only 20% (n=3) of the women in the control group. The glycaemic scores p-value >0.05. Hence, it is inferred that there was a similarity in glycaemic level among the interventional and control group, so both the groups were comparable in terms of glycaemic values.

[Table/Fig-2] shows there is a significant difference in the Pre FBS and post FBS values in the intervention group (F=6.097, df=3.42) p-value <0.027. In control group there is no significant difference in FBS values at various time points (F=0.397, df=3.42) p-value >0.756. Paired Eta square shows 0.303 which is a moderate beneficial effect of the exercise on the glycaemic value in the interventional group.

[Table/Fig-3] shows that there is a significant difference between the 24 weeks FBS and 36 weeks FBS scores p-value=0.027 in the intervention group.

[Table/Fig-4] shows there was a significant difference in the pre PPBS and post PPBS values in the intervention group (F=9.359, df=3.42) p-value <0.008 and control group (F=5.802, df=3.42) p-value <0.002.

[Table/Fig-5] shows that there is a statistically significant difference between the 24 weeks PPBS and 36 weeks PPBS scores p-value=0.016 which is highly significant. In 28 weeks and 36 weeks, scores p-value=0.002 in the intervention group. In the control group, 28 weeks PPBS and 32 weeks show significant difference p-value=0.028.

[Table/Fig-6] shows that there is a highly significant p-value=0.01 difference between the Post PPBS (28 weeks) and post PPBS (32 weeks) among the intervention and control groups. Hence, the null hypothesis is rejected and the research hypothesis is accepted.

Data presented in the [Table/Fig-7] shows that there is no significant difference in the maternal outcome observed between intervention and control group p-value >0.05.

[Table/Fig-8] shows there is a significant difference between parity p-value=0.038, history of PCOS p-value=0.026 and BMI p-value=0.030 with HbA1c values.

In the present study, the mean age of women was 28.07±2.712 years in the interventional group and 30.20±4.73 years in the control group. These findings were congruent to the findings of the study conducted by Kokic IS et al., where they assessed the acute responses to structured aerobic and resistance exercise in women with GDM. The mean age of the women was 32.8±3.8, prepregnancy body mass index was 24.4±4.9 kg/m², and 50% of them were nulliparous [19].

In the present study Post-hoc test observes that there was a statistically significant difference between the 24 weeks PPBS and 36 weeks PPBS scores. Similar study results are shown with the conventional treatment group and aerobic exercise which, reduced fasting blood glucose (WMD=-0.35, 95% CI: -0.62 to -0.08, I2=87%), postprandial blood glucose (WMD=-0.62, 95% CI:-0.95 to -0.29, I2=84%) and HbA1c levels (WMD=-0.35, 95% CI:-0.49 to -0.20, I2=71%) in patients with GDM [20]. A different form of exercise has been included within the resistance exercise modality in Thailand for eight weeks, the studied population performed this exercise twice a week for 50 minutes. The variables analysed are fasting, postprandial glucose, and HbA1c. These three variables are lower in the intervention group, with a significant difference (p-value=0.012; p-value=0.001; p-value=0.038, respectively) [21].

Results show that there is no significant difference in the maternal outcome observed between the intervention and control group p-value >0.05. Similar results show in the study conducted by De Barros MC et al., on resistance exercise and glycaemic control in women with GDM. The 2 groups were similar in terms of the variables measured at the time of delivery (p-value >0.05). No difference in the frequency of caesarean section was observed between groups (n=21 of 32 in EG vs n=24 of 32 in CG; p-value=0.412) [22]. Exercise during pregnancy especially aerobic and resistance exercises was beneficial to the GDM women to reduce their glycaemic level at 36 weeks and to improve the maternal outcome.

Even though exercise had a good effect on the glycaemic level there was some drawback in the study. The sample size of the study was small, so generalisation was difficult and prediction of risk on maternal outcome could not be evaluated. Further studies can be done on various pattern of exercises which is moderate intensity will help to control the glucose level during pregnancy.

Maternal fitness is essential during pregnancy and delivery for a better maternal outcome. Regular moderate-intensity exercise training during pregnancy is associated with both a lower blood glucose level PPBS and maternal weight gain and also provides physical fitness. Characteristics of effective exercise programs for management of GDM appear to be exercise performed at a moderate intensity and for a minimum of three times a week. Exercise during pregnancy with gestational diabetes will improve maternal outcomes.