EndoBarrier gastrointestinal liner. Delineation of underlying mechanisms and clinical effects.

Bariatric surgery (e.g. Roux-en-Y gastric bypass (RYGB)) has proven the most effective way of achieving sustainable weight losses and remission of type 2 diabetes (T2D). Studies indicate that the effectiveness of RYGB is mediated by an altered gastrointestinal tract anatomy, which in particular favours release of the gut incretin hormone glucagon-like peptide-1 (GLP-1). The EndoBarrier gastrointestinal liner or duodenal-jejunal bypass sleeve (DJBS) is an endoscopic deployable minimally invasive and fully reversible technique designed to mimic the bypass component of the RYGB. Not only GLP-1 is released when nutrients enter the gastrointestinal tract. Cholecystokinin (CCK), secreted from duodenal I cells, elicits gallbladder emptying. Traditionally, bile acids are thought of as essential elements for fat absorption. However, growing evidence suggests that bile acids have additional effects in metabolism. Thus, bile acids appear to increase GLP-1 secretion via activation of the TGR5 receptor on the intestinal L cell. Recently FXR receptors were postulated to contribute to GLP-1 secretion too. Furthermore, metformin has been shown to increase circulating GLP-1 levels but although the exact mechanism is not fully elucidated it may involve metformin-induced inhibition of bile acid reuptake from the small intestines. Small-sized studies reported varying degrees of weight loss and, in some, improvement of glucose metabolism. Therefore, the objectives of this thesis were to collect existing information on the DJBS in order to evaluate clinical efficacy and safety (study I and II). Furthermore, since the endocrine impact of the DJBS is not fully elucidated, and DJBS is expected to mimic RYGB, we investigated postprandial metabolic changes following 26 weeks of DJBS treatment in ten obese subjects with normal glucose tolerance (NGT) and nine matched patients with T2D (study III). Finally, we studied the single and combined effects of CCK induced gallbladder emptying and single-dose metformin on human GLP-1 secretion in ten healthy subjects (study IV). We hypothesized that metformin-induced GLP-1 secretion - at least partly - would be dependent on gallbladder emptying and the presence of bile acids in the gut. DJBS appears to lead to moderate weight losses in obese subjects compared to diet or lifestyle modifications (study II). DJBS had insignificant and small effects (compared to diet) on glycaemic regulation. Adverse events consisted mainly of mild-to-moderate transient dyspepsia. Nearly 20% (n = 66) of DJBS treated subjects experienced a serious adverse event (e.g. gastrointestinal bleeding or device migration), which resulted in early device removals. No deaths or liver abscesses were reported following DJBS treatment. In our own study III we found similar, moderate weight losses as in study II. GLP-1 and PYY concentrations increased in patients with T2D (not NGT subjects) after implantation. DJBS had no or minor effects on postprandial levels of glucose, insulin, C-peptide, glucagon, GIP, CCK or gastrin. Food intake decreased in parallel with an increased sensation of satiety in obese NGT subjects, but were transient. Dyspeptic episodes occurred in nearly all participants. Five devices (21%) were explanted early due to abdominal pain, and few changes of on-going antidiabetic medication were made. Finally, study IV showed that both CCK-induced gallbladder emptying and metformin alone elicited significant GLP-1 responses that were additive upon combination. Moreover, we saw significant PYY and short-lasting glucose-dependent insulinotropic polypeptide (GIP) responses following gallbladder emptying. In conclusion, in spite of increased GLP-1 responses in patients with T2D and a modest weight loss, DJBS had no effect on postprandial glucose metabolism, and therefore no patient with T2D achieved disease remission. Based on our results, we cannot recommend DJBS to be implemented as a standard of medical care management of obese patients with T2D. Future larger trials may lead to different conclusions. In addition, the observed gut hormone responses following CCK-induced gallbladder emptying and metformin, make suggest that bile acid release into the small intestines with subsequent TGR5 and FXR involvement contributes to stimulation of GLP-1 secretion and, therefore, that metformin's mode of action encompasses both bile acid-dependent and independent stimulation of gut hormone secretion.