Highly Accurate Method Developed Predicts Postpartum Diabetes

The SWIFT study screened women with oral glucose tolerance tests at two months after delivery and then annually thereafter to evaluate the impact of breastfeeding and other characteristics on the development of type 2 diabetes after a pregnancy complicated by gestational diabetes.

The team conducted metabolomics with baseline fasting plasma and identified 21 metabolites that significantly differed by incident type 2 diabetes (T2D) status. Machine learning optimization resulted in a decision tree modeling that predicted T2D incidence with a discriminative power of 83.0% in the training set and 76.9% in an independent testing set, being far superior to fasting plasma glucose alone. The new method may also be able to predict individuals who may develop type 2 diabetes in the general population which would be a major advance at a time when more than 300 million people suffer from the preventable form of this disease. A next-generation blood test that's more simple and accurate than the current options could help to identify individuals who would benefit most from more timely and effective interventions to prevent type 2 diabetes.

Michael B. Wheeler, PhD, a professor in the Department of Physiology and a senior author of the study said, “After delivering a baby, many women may find it very difficult to schedule two hours for another glucose test. What if we could create a much more effective test that could be given to women while they're still in the hospital? Once diabetes has developed, it's very difficult to reverse.” The study was published in the June 2016 issue of the journal Diabetes.

Related Links:
University of Toronto