Clinical predictors for a high risk for the development of diabetes mellitus in the early puerperium in women with recent gestational diabetes mellitus.

OBJECTIVE: The purpose of this study was to identify which maternal, antepartum, or neonatal clinical parameters were predictive for a high risk of diabetes mellitus in the puerperium in women with recent gestational diabetes mellitus and to calculate the associated diabetes mellitus rates and odds ratios. STUDY
DESIGN: One thousand six hundred thirty-six women underwent an oral glucose tolerance test within 1 to 4 months of delivery. Demographic, historic, and antenatal glycemic parameters and neonatal outcome parameters were tested by univariate and multivariate logistic regression for risk of postpartum diabetes mellitus. Continuous variables were divided into quartiles that compared the upper to lower quartile adjusted odds ratio and prevalence of diabetes mellitus.
RESULTS: Postpartum diabetes mellitus was diagnosed in 230 women (14.1%) according to the American Diabetes Association criteria (1997). No maternal demographic or neonatal parameters were significantly associated with diabetes mellitus. The final model of independent predictors in decreasing significance included the highest fasting plasma glucose level during pregnancy, any fasting plasma glucose level of > or = 105 mg/dL (class A(2)), the area under the curve of pregnancy oral glucose tolerance test, gestational age at diagnosis, previous gestational diabetes mellitus history, and 50-g glucose challenge test results. The fasting plasma glucose level was the best discriminator, with a 21-fold (95% CI, 4.6-96.3) increased odds ratio comparing the 4th quartile (fasting plasma glucose level, >121 mg/dL; diabetes mellitus rate, 36.7%) to 1st quartile (fasting plasma glucose level, < 95 mg/dL; diabetes mellitus rate, 0.5%). The presence of previous gestational diabetes mellitus or current class A(2) gestational diabetes mellitus approximately doubled the odds ratio for diabetes mellitus. The odds ratio increased 3- to 4-fold when the area under the curve was > or = 33.36 min small middle dot g/dL (4th quartile) or the glucose challenge test was > or = 155 mg/dL (2nd-4th quartiles) and decreased > 50% if gestational diabetes mellitus was diagnosed at > 27 weeks (3rd-4th quartile).
CONCLUSION: During pregnancy, the highest fasting glucose level, followed by the severity of glucose intolerance, and earlier gestational diabetes mellitus diagnosis were the best predictors for postpartum diabetes mellitus. Diabetic education should begin during pregnancy, especially for women who are identified to be at a high risk when they are highly motivated and under medical care.