Implications of new diagnostic criteria for abnormal glucose homeostasis in women with previous gestational diabetes.

Publisher Website
Google Scholar
Abstract
OBJECTIVE: To determine the consequences of applying revised American Diabetes Association (ADA) (1997) and World Health Organization (WHO) (1998) recommendations for the classification of glucose intolerance in women with previous gestational diabetes mellitus (GDM). RESEARCH DESIGN AND METHODS: There were 192 women with previous GDM who took an oral glucose tolerance test (OGTT) 1-86 months after delivery and were classified by WHO (1985), ADA (1997, fasting glucose), and revised WHO (1998) guidelines. RESULTS: Among the 165 women without a preexisting diagnosis of diabetes, WHO-1985 and ADA-1997 provided similar estimates of diabetes prevalence (13.3% vs. 11.5%) but widely differing estimates of impaired glucose homeostasis (31.5% impaired glucose tolerance [IGT] by WHO-1985 vs. 10.9% impaired fasting glucose by ADA-1997 criteria). Overall, 56 women (34%) showed a classification discrepancy between WHO-1985 and ADA-1997 criteria, including 44 with normal fasting glucose by ADA-1997 criteria, but abnormal 2-h glucose by WHO-1985 criteria (40 IGT, 4 diabetes). The cardiovascular risk profile of these women was more favorable than that of 18 women with impaired fasting glucose. WHO-1998 recommendations reproduced ADA-1997 findings when used as a fasting screen, but behaved similarly to WHO-1985 criteria when 2-h glucose values were also analyzed. CONCLUSIONS: All criteria produced similar estimates of diabetes prevalence. However, analyses based on a single fasting glucose screen (and a threshold of 6.1 mmol/l) failed to identify 60% of women with abnormal 2-h glucose levels. Screening women with previous GDM (and by analogy, other groups at high risk of diabetes) with a single fasting glucose has low sensitivity for the detection of abnormal glucose tolerance. Recent guidelines recommending this approach require reevaluation.

This publication has 0 references indexed in Scilit: