Clinical, metabolic, and microbial responses to sleeve gastrectomy and Roux-en-Y gastric bypass: results from a randomised clinical trial

Abstract

Background: Bariatric surgery is the most effective method for the treatment of morbid obesity and type 2 diabetes (T2D). Roux-en-Y gastric bypass (RYGB) and sleeve gastrectomy (SG) are the most frequently performed bariatric surgeries in New Zealand and worldwide. RYGB and SG involve distinctly different anatomical rearrangements of the stomach and gastrointestinal (GI) tract yet produce comparable weight loss and T2D remission in 70-85% of patients. The overlapping and distinct mechanisms underpinning the latter remain unclear. Potential mechanisms by which RYGB and SG achieve their respective metabolic benefits include reduced caloric intake due to restricted stomach capacity, changes in gut hormones that impact on satiety and/or glucose regulation, and changes to the gut bacteria. While discordant gut microbial changes by type of surgery have been described, common microbial taxa that accompany T2D remission or signal likelihood of T2D remission across both types of bariatric surgery have not been clearly identified. Despite their common use, the clinical and gut microbial changes that occur after both surgeries have not been adequately compared in a single, parallel-arm, assessor-blinded randomised trial. Aims: In this thesis I aimed to compare body composition, bone mineral density, and satiety 1 year after banded-RYGB versus SG in a single, randomised cohort of adults with obesity and T2D. I aimed to systematically review the current literature on the changes to the gut bacterial communities following bariatric surgery and I aimed to identify differences in the gut microbial communities and predicted functional potential of the gut bacteria following banded-RYGB and SG. In addition, I aimed to identify potential microbial indicators of T2D remission both pre-surgery and at 1 year after surgery. Methods: A cohort of 114 individuals with obesity and T2D were randomised to either banded-RYGB or SG surgery. Two days before and 1 year after surgery, participants had body composition and bone mineral density (BMD) measured by dual-energy X-ray absorptiometry (DXA). A sub-group of 86 participants consented to further investigations at each of these two visits, including a satiety questionnaire and a 75g oral glucose tolerance test (OGTT) with 5 time point blood sampling until 120 minutes. At each visit, participants in the sub-group also provided a 5-day food dairy and a faecal sample. Food diaries were analysed using FoodWorks. Whole-metagenome shotgun sequencing was performed on paired faecal samples. Taxonomic composition and predicted functional potential of the gut bacteria were identified using HUMANn2 and annotated using MetaCyc. Results: Comparative analysis of body composition changes showed significantly more percent excess weight loss (%EWL) (p=0.006) following RYGB compared to SG. Lean body mass loss of 8.5% was observed for both RYGB and SG surgery groups. Statistically significant reductions in BMD were seen for the whole body, lumbar spine and femoral neck as well as in respective T and Z scores for all groups of participants (all p<0.001). The BMD of female participants was more negatively affected by bariatric surgery, with significant sex specific reductions in bone mineral content (BMC, p=0.001) and total BMD at the femoral neck T- and Z-scores (p=0.026, p=0.036 respectively). Postsurgery, osteopenia developed in 4 participants at the lumbar spine and 11 participants at the femoral neck. Osteoporosis developed at the lumbar spine in two osteopenic women who had RYGB. Glucose control was restored following both RYGB and SG surgery indicated by the high rates of T2D remission 1 year after RYGB (75.8%) and SG (70.6%) (p=0.78). A significant increase in glucose stimulated GLP-1, GIP, and glucagon was observed following both RYGB and SG surgery (all p<0.001), with no differences between the surgery types. A significant decrease in plasma leptin was observed 1 year after both RYGB and SG surgery (both p<0.0001). Fasting plasma ghrelin was significantly lower 1 year after surgery in participants who had SG surgery (p=0.024). Post prandial PYY response and selfreported scores of hunger did not differ between the surgery types 1 year after surgery. Overall, increased relative abundances of Firmicutes and Proteobacteria followed RYGB surgery while increased relative abundance of Bacteroidetes followed SG surgery. Lachnospiraceae (p=0.04) and Roseburia (p=0.01) species were more relatively abundant in participants who achieved T2D remission 1 year after surgery. Importantly, greater relative abundance of Eubacteriaceae (p=0.01) and Alistipes putredinis (p=0.01) were observed pre-surgery among participants who remitted from T2D 1 year after surgery. Conclusion: This thesis provides a comparison of the relative effects of the two most commonly performed bariatric surgeries in New Zealand and worldwide at one year following surgery. Percent excess weight loss was greater following RYGB, and BMD loss was significant and more severe for women. T2D remission rates were similar between RYGB and SG at 1 year post-surgery - underpinned by similar alterations in gut hormones and increased relative abundances of Eubacteriaceae and Alistipes putredinis in the presurgery faecal microbiota and Lachnospiraceae and Roseburia species in the post-surgery faecal microbiota. Future work in validating the novel pre-surgery microbial predictors of T2D remission after bariatric surgery could inform patient prioritisation and investigating the post-surgery microbial biomarkers of T2D remission could inform novel probiotic development to complement bariatric surgery.