Type 2 Diabetes Mellitus Remission Models Following Laparoscopic Gastric Bypass: a 4-Model Analysis in a Latino Population.

BACKGROUND

Bariatric surgery is indicated for major weight loss and for the control of associated comorbidities, particularly type 2 diabetes. Remission prediction scores have been proposed for this end, such as: DiaRem, Ad-DiaRem, ABCD, and DiaBetter. Nevertheless, they have not been evaluated all together in a specific population.

METHODS

Retrospective study with Mexican patients submitted to gastric bypass with at least 12 months follow-up. All patients had BMI > 30 kg/m and type 2 diabetes. The primary objective was to evaluate the remission prediction performance of scores. A baseline analysis (anthropometric, biochemical, and metabolic) and remission rates were obtained. Remission scores and cut-off values were assigned based on original descriptions. A ROC analysis was performed for sensibility and specificity.

RESULTS

A total of 95 patients were included. Mean age 44 years, 85.6% female with mean BMI of 44.1 kg/m, and mean HbA1C of 7.2%. At 12 months, complete remission was obtained in 76.8%. ROC curves were plotted showing that DiaRem had 75.3% sensitivity and 68.2% specificity (AUC 0.723 p = 0.001), Ad-DiaRem had 84.9% and 50% (AUC 0.702 p = 0.002), ABCD had 57.5% and 77.3% (AUC 0.0.690 p = 0.002), and DiaBetter had 72.6% and 77.3% (AUC 0.748 p < 0.001).

CONCLUSION

In Mexican patients with obesity and type 2 Diabetes, submitted to gastric bypass, remission prediction could be assessed with any current model showing satisfactory sensibility and specificity. Among such models, DiaBetter obtained the best statistical performance in our population. Type 2 diabetes remission rate at 1 year is similar to any other race or ethnicity.