983: Rate of Hyperglycemia and Adverse Outcomes among Gestational Diabetics

Abstract

The diagnosis of gestational diabetes (GDM) typically entails blood glucose (BG) monitoring 4 times daily: fasting (F) and three postprandial (PP) values. The ideal threshold at when therapy should be initiated or modified is uncertain, with some suggesting when > 33% of values are above pre-specified thresholds and others, when > 50% are abnormal. The aim of this study was to determine if there exists a threshold in the percentage of abnormal BG values at any of the 4 scheduled testing timeframes that is associated with adverse perinatal outcomes. This was a secondary analysis of a randomized controlled trial at 5 tertiary centers evaluating the frequency of BG testing among GDM patients. BG values were defined as abnormal if: F were ≥ 95 mg/dl or if 2-hr PP values were ≥ 120 mg/dl. Composite neonatal morbidity (CNM) included any one of the following: Shoulder dystocia, large-for-gestational age (birthweight >90th% for GA), NICU admission, neonatal hypoglycemia or hyperbilirubinemia. Composite maternal morbidity (CMM) was the presence of either pre-eclampsia or indicated delivery for uncontrolled BG values. Patients with missing data for any of these outcomes were excluded. A receiver operating characteristics (ROC) analysis was used for each time point to determine if the percentage of abnormal BG was associated with CNM or CMM. Of the 293 women randomized, 282 (96%) were eligible for analysis. The rate of CNM was 35%, and of CMM, 18%. On average, women had 32 BG values for evaluation for fasting and 2hr PP. Baseline maternal characteristics were similar among women with and without CNM. Almost three-quarter of the patients did not require medical treatment. The proportion of chorioamnionitis (0.5% vs. 5%, p=0.02) and preterm labor (3.3% vs. 18%, p<0.01) were higher among women with CNM. The mean percentage of abnormal values at all each of the 4 testing timeframes was similar between both groups (Table). There was no difference in proportion of patients with ≥50% abnormal values throughout pregnancy. Area under the ROC curves demonstrated that the percentages of abnormal BG value was poor at identifying CNM or CMM (Figure). Among women with GDM, the percentage rate of abnormal BG values was poor predictor of composite maternal or neonatal morbidity; concurrently identifying an optimal threshold was not feasible.View Large Image Figure ViewerDownload Hi-res image Download (PPT)