A postnatal fasting plasma glucose is useful in determining which women with gestational diabetes should undergo a postnatal oral glucose tolerance test

AIMS: It is recommended that women with gestational diabetes (GDM) should have a 6-week postnatal oral glucose tolerance test (OGTT). As this test may be unpleasant, time-consuming and has resource implications, we evaluated whether the 6-week postnatal fasting glucose could be used to determine which women should undergo an OGTT. METHODS: All women with GDM, diagnosed according to the World Health Organization criteria, who were delivered at the Princess Anne Hospital, Southampton between May 2000 and May 2002, were recommended to have an OGTT. The results of the fasting plasma glucose concentration were assessed in relation to the 2-h glucose value. RESULTS: One-hundred and fifty-two women with GDM were delivered. Thirty (19.7%) women refused an OGTT or failed to attend. In the 122 OGTTs, three (2.4%; 95% confidence interval 0.8, 7) women had diabetes, three had impaired glucose tolerance and four had impaired fasting glycaemia. No woman with a normal test had fasting glucose of > or =6.0 mmol/l. Fasting glucose was correlated with the 2-h glucose (r=0.62, P<0.0001). Only 10 (8.1%) of the OGTTs would have been performed if only women with fasting glucose of > or =6.0 mmol/l underwent the test. The sensitivity and specificity of this approach for the diagnosis of postnatal diabetes is 100% and 94%, respectively. Linear regression methods indicate that it would miss fewer than three in 10 000 cases. CONCLUSIONS: In our population, a 6-week postnatal fasting plasma glucose is useful in determining which women with gestational diabetes should undergo an OGTT. Consequently we now perform OGTT only in women whose postnatal fasting plasma glucose is > or =6.0 mmol/l.