Assessing IADPSG criteria for gestational diabetes mellitus diagnosis in Sri Lankan pregnant women

Abstract

INTRODUCTION AND OBJECTIVES: Gestational diabetes mellitus (GDM) is a condition of glucose intolerance first recognized during pregnancy. Currently, the guidelines for diagnosing GDM, as recommended by the International Association of Diabetes and Pregnancy Study Groups (IADPSG) and adopted by the World Health Organization (WHO) in 2013, include an elevation in either fasting ( 5.1 mmol/L), 1-hr (10.0 mmol/L) or 2-hour (8.5 mmol/L) venous plasma glucose level after a 75-gram glucose intake. IADPSG criteria is widely adopted in many countries. Clinicians have questioned the applicability of these diagnostic thresholds for different ethnicities. Therefore, we aimed to examine the prevalence of GDM diagnosed by three different guidelines and its association with offspring birthweight, in a hospital-based observational study among Sri Lankan pregnant women. METHODS: Medical records of 795 women with singleton pregnancy who attended two tertiary hospitals in Sri Lanka were utilized for the following information: age, height, weight and gestational age at the booking visit, 75g oral glucose tolerance test with fasting, 1- and 2-hour glucose readings after 20 weeks of gestation, and offspring birthweight. Body mass index was calculated at the booking visit. Following diagnostic thresholds were applied to define GDM: IADPSG criteria, Sri Lanka national guideline and WHO 1999. We calculated crude and age- and BMI-adjusted prevalence rates of GDM using STATA. Venn diagrams were used to represent the individual and overlapping diagnosis of GDM by each time point of glucose level. RESULTS: The crude, and age- and BMI adjusted GDM prevalence, were as follows: 31.2% and 31.2% for IADPSG criteria; 21.3% and 27.4% for Sri Lanka national guidelines; and 21.8% and 28.0% for WHO 1999. GDM diagnosed using the IADPSG criteria, or using Sri Lankan guidelines, had significant associations with birthweight in unadjusted models, yet the associations attenuated to non-significance after adjusting for age and BMI. It is notable that effect estimates for birthweight did not differ greatly among women diagnosed with GDM by IADPSG only and in those diagnosed with GDM by IADPSG and Sri Lankan guidelines. Linear regression of GDM diagnosed by WHO 1999 was not associated with birthweight in either unadjusted or adjusted models. CONCLUSIONS: The IADPSG criteria appeared to have a better diagnostic value, in terms of identifying cases of GDM and in predicting birthweight across all three guidelines. Our results suggest that adopting the IADPSG criteria for diagnosing GDM may be important in reducing hyperglycaemia-related outcomes in Sri Lankan women.