Antenatal screening using random blood glucose values.

For many years it has been established practice to test the urine of pregnant women for the presence of glucose in the belief that this is an efficient way to detect diabetes mellitus. It is now becoming recognized that one of the normal maternal physiologic adaptations during pregnancy is an increase in the renal excretion of glucose; on examination, up to 50% of healthy pregnant women will have detectable glycosuria at some stage. As the definition of diabetes mellitus is based on random blood glucose values or the concentrations achieved at defined times after an oral glucose load, it would seem logical that any antenatal screening procedure should be similarly based. A total of 2403 consecutive antenatal patients had a random venous whole blood glucose concentration determined between 28 and 32 wk gestation. Calculated 99% cutoff values were 110 mg/dl (6.1 mmol/L) within 2 h of a meal or 101 mg/dl (5.6 mmol/L) more than 2 h postprandial. Four patients were found to have previously unsuspected but unequivocal diabetes mellitus and four more had impaired glucose tolerance on the basis of the 1980 WHO criteria. Screening all antenatal patients by this method is efficient, does not inconvenience patients, and is relatively cost efficient in terms of staff and laboratory resources.