Abstract
Glycemic responses to bariatric surgery are highly heterogeneous among patients and defining response types remains challenging. Recently developed data-driven clustering methods have uncovered subtle pathophysiologically informative patterns among patients without diabetes. This study aimed to explain responses among patients with and without diabetes to bariatric surgery with clusters of glucose concentration during oral glucose tolerance tests (OGTTs). We assessed 30 parameters at baseline and at four subsequent follow-up visits over one year on 154 participants in the Bialystok Bariatric Surgery Study. We applied latent trajectory classification to OGTTs and multinomial regression and generalized linear mixed models to explain differential responses among clusters. OGTT trajectories created four clusters representing increasing dysglycemias that were discordant from standard diabetes diagnosis criteria. The baseline OGTT cluster increased the predictive power of regression models by over 31% and aided in correctly predicting more than 83% of diabetes remissions. Principal component analysis showed that the glucose homeostasis response primarily occurred as improved insulin sensitivity concomitant with improved the OGTT cluster. In sum, OGTT clustering explained multiple, correlated responses to metabolic surgery. The OGTT is an intuitive and easy-to-implement index of improvement that stratifies patients into response types, a vital first step in personalizing diabetic care in obese subjects.
Highlights
Bariatric surgery is the single most effective treatment for type 2 diabetes (T2D), with a complete two year remission in over 78% of patients and marked improvement in over 86% of patients, as well as improved prediabetic status [1]
A new class of deep analytic methods, latent trajectory classification models, have recently been implemented to detect subtle differences among patient groups that reflect metabolic and physiological mechanisms [7]. We apply these methods to a group of patients both with and without diabetes in order to gain insight into differences in metabolic and anthropometric responses to bariatric surgery based upon these latent clusters
Of the baseline Cluster 4 subjects, 24 (92%) showed improved classification, and only two remained unchanged. Of these 24, only two returned to the “healthiest” Cluster 1 glucose response curve. These results indicate that a complete resolution of glucose response measured in these oral glucose tolerance tests (OGTTs) curve classifications depends upon the severity of the initial dysregulation—the majority of the higher baseline clusters improved only partially to Clusters 2 or 3 at 12 mo post-surgery
Summary
Bariatric surgery is the single most effective treatment for type 2 diabetes (T2D), with a complete two year remission in over 78% of patients and marked improvement in over 86% of patients, as well as improved prediabetic status [1]. A new class of deep analytic methods, latent trajectory classification models, have recently been implemented to detect subtle differences among patient groups that reflect metabolic and physiological mechanisms [7]. We apply these methods to a group of patients both with and without diabetes in order to gain insight into differences in metabolic and anthropometric responses to bariatric surgery based upon these latent clusters. This is the first time that these methods have been used for a mixed cohort of patients with and without diabetes undergoing the same bariatric surgical treatment
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
