Discordance in prediction for prognosis of type 2 diabetes after metabolic surgery: comparison of the ABCD, DiaRem, and individualized metabolic surgery models

PURPOSE: Metabolic surgery has been performed as a treatment option for uncontrolled type 2 diabetes (T2D), and several scoring systems for predicting postoperative T2D remission have been proposed. This study was designed to assess consistency of 3 existing scoring systems in patients with T2D duration <1 year. METHODS: This study included 186 patients with T2D enrolled in a university hospital prospective database between 2011 and 2013. Externally validated scoring systems for predicting T2D prognosis after metabolic surgery were identified and selected through systematic literature search. We assessed concordance between ABCD, DiaRem, and individualized metabolic surgery (IMS) scores in participants using kappa statistical analysis and 1-way analysis of variance. RESULTS: Of the participants, 52 and 82 patients were expected to have favorable T2D remission after metabolic surgery with ABCD score of 10–5 and DiaRem score of 0–7, respectively, and a slight-to-fair concordance was shown between the 2 scoring systems (kappa measure, 0.07; standard error [SE], 0.05 and kappa measure, 0.25; SE, 0.19, respectively). The DiaRem score increased with T2D severity determined by IMS score (P < 0.001), while the ABCD score showed no significant association with IMS score. CONCLUSION: ABCD and DiaRem scores showed significant discordance when applied to potential metabolic surgery candidates in whom postoperative T2D remission rate was highly expected. The IMS score showed a dose-response association with DiaRem score but had no significant association with the ABCD score.