Screening for gestational diabetes at antenatal booking in a Malaysian university hospital: the role of risk factors and threshold value for the 50-g glucose challenge test.

BACKGROUND The best method of screening for gestational diabetes (GDM) remains unsettled. The 50-g glucose challenge test (GCT) is used in a two-stage screening process but its best threshold value can vary according to population. AIMS To evaluate the role of risk factors in conjunction with GCT and to determine an appropriate threshold for the one-hour venous plasma glucose with the GCT. METHOD In a prospective study, 1600 women at antenatal booking without a history of diabetes mellitus or GDM filled a form on risk factors before GCT. Women who had GCT >or= 7.2 mmol/L underwent the 75-g oral glucose tolerance test (OGTT). GDM was diagnosed according to WHO (1999) criteria. RESULT Thirty-five per cent had GCT >or= 7.2 mmol/L, 32.6% underwent OGTT and 34.5% of OGTT confirmed GDM. The GDM rate in our population was at least 11.4%. Examination of the receiver operator characteristic curve suggested that the best threshold value for the GCT in our population was >or= 7.6 mmol/L. Multivariable logistic regression demonstrated that only GCT >or= 7.6 mmol/L was an independent predictor for GDM (adjusted odds ratio 3.7: P < 0.001). After GCT, maternal age and anthropometry, OGTT during the third trimester, family history, obstetric history and glycosuria were not independent predictors of GDM. CONCLUSIONS Risk factors were not independent predictors of GDM in women with GCT >or= 7.2 mmol/L. GCT threshold value >or= 7.6 mmol is appropriate for the Malaysian population at high risk of GDM.