Objective: We tested the hypothesis that the development of abnormal glucose metabolism after gestational diabetes can be predicted readily by means of available clinical variables.
Study design: Three hundred fifty nonpregnant former gestational diabetic women delivered during the previous 10 years underwent glucose tolerance tests. Variables including body mass index before the index pregnancy, pregnancy glucose tolerance test values, gestational diabetes treatment, complications, gestational age at diagnosis of gestational diabetes, and time elapsed since pregnancy were analyzed with logistic regression.
Results: Variables that distinguished subjects who later developed diabetes or impaired glucose tolerance included prepregnancy body mass index (28.5 +/- 7 versus 25 +/- 5 kg/m2, p < 0.001) and fasting glucose on the pregnant oral glucose tolerance test (109 +/- 20 vs 92 +/- 15 mg/dl, p < 0.001). Logistic results with these two variables plus time since the index pregnancy predict subsequent glucose tolerance test abnormality by the following equation: estimated risk = 1/[1 + e-(-10.37 + 0.04 (fasting plasma glucose) + 0.08 (body mass index) + 0.03 (months since delivery))].
Conclusion: The risk for subsequent glucose abnormality among individuals with previous gestational diabetes is quantifiable based on prepregnant body mass index and fasting plasma glucose during pregnancy.