The effects of a physical activity training program on insulin sensitivity and plasma glucose levels in women at high risk for gestational diabetes

Background: Pregnancy is a period in the lives of women that may result in decreased physical activity. This change often occurs despite the fact that there is no medical reason for women to reduce their physical activity levels. However, maintaining adequate levels of physical activity during pregnancy may be important for mother and child. Studies of physical activity and the occurrence of maternal pregnancy related disorders suggest that moderate physical activity during pregnancy may be associated with reductions in the risk of gestational diabetes mellitus (GDM). However, at the moment, physical activity is not routinely adviced or prescribed for women at risk for gestational diabetes in the Netherlands. The overall aim of the study: To assess whether a physical activity program can improve insulin sensitivity and fasting plasma glucose levels of women at high risk for gestational diabetes, assuming that this will normalise their risk of gestational diabetes. Research questions are: 1) what is the effect of a physical activity program on fasting plasma glucose and insulin sensitivity in pregnant women at high risk for gestational diabetes? 2) what is the effect of a physical activity program on birth weight? 3) what is the compliance with the physical activity program, and what factors contribute to the succes or failure of the program? Design: In this randomised trial, 160 women at risk for gestational diabetes will be randomly allocated to two groups: 1) an intervention group will receive a physical activity program in addition to usual care, and 2) a control group will receive usual care. Outcomes: Measurements will be conducted before randomisation, at 25 and 35 weeks of pregnancy and 12 weeks postpartum. Primary maternal outcome measures are fasting plasma glucose and relative increase in insulin resistance. Primary neonatal outcome is birth weight. Secondary outcome measures are: serum triglycerides, maternal weight gain during pregnancy, physical activity level, fetal growth. Analysis: Longitudinal, linear regression analyses will be performed. In these analyses, the correlation between multiple measurements within one individual is taken into account. Data will be analysed according to the intention-to-treat principle. Cost-effectiveness will be assessed by a bootstrap technique and cost-utility will be determined by modelling a linear decision tree.