Development and validation of an interpretable machine learning-based calculator for predicting 5-year weight trajectories after bariatric surgery: a multinational retrospective cohort SOPHIA study.

BACKGROUND: Weight loss trajectories after bariatric surgery vary widely between individuals, and predicting weight loss before the operation remains challenging. We aimed to develop a model using machine learning to provide individual preoperative prediction of 5-year weight loss trajectories after surgery. METHODS: In this multinational retrospective observational study we enrolled adult participants (aged ≥18 years) from ten prospective cohorts (including ABOS [NCT01129297], BAREVAL [NCT02310178], the Swedish Obese Subjects study, and a large cohort from the Dutch Obesity Clinic [Nederlandse Obesitas Kliniek]) and two randomised trials (SleevePass [NCT00793143] and SM-BOSS [NCT00356213]) in Europe, the Americas, and Asia, with a 5 year follow-up after Roux-en-Y gastric bypass, sleeve gastrectomy, or gastric band. Patients with a previous history of bariatric surgery or large delays between scheduled and actual visits were excluded. The training cohort comprised patients from two centres in France (ABOS and BAREVAL). The primary outcome was BMI at 5 years. A model was developed using least absolute shrinkage and selection operator to select variables and the classification and regression trees algorithm to build interpretable regression trees. The performances of the model were assessed through the median absolute deviation (MAD) and root mean squared error (RMSE) of BMI. FINDINGS: 10 231 patients from 12 centres in ten countries were included in the analysis, corresponding to 30 602 patient-years. Among participants in all 12 cohorts, 7701 (75·3%) were female, 2530 (24·7%) were male. Among 434 baseline attributes available in the training cohort, seven variables were selected: height, weight, intervention type, age, diabetes status, diabetes duration, and smoking status. At 5 years, across external testing cohorts the overall mean MAD BMI was 2·8 kg/m2 (95% CI 2·6-3·0) and mean RMSE BMI was 4·7 kg/m2 (4·4-5·0), and the mean difference between predicted and observed BMI was -0·3 kg/m2 (SD 4·7). This model is incorporated in an easy to use and interpretable web-based prediction tool to help inform clinical decision before surgery. INTERPRETATION: We developed a machine learning-based model, which is internationally validated, for predicting individual 5-year weight loss trajectories after three common bariatric interventions. FUNDING: SOPHIA Innovative Medicines Initiative 2 Joint Undertaking, supported by the EU's Horizon 2020 research and innovation programme, the European Federation of Pharmaceutical Industries and Associations, Type 1 Diabetes Exchange, and the Juvenile Diabetes Research Foundation and Obesity Action Coalition; Métropole Européenne de Lille; Agence Nationale de la Recherche; Institut national de recherche en sciences et technologies du numérique through the Artificial Intelligence chair Apprenf; Université de Lille Nord Europe's I-SITE EXPAND as part of the Bandits For Health project; Laboratoire d'excellence European Genomic Institute for Diabetes; Soutien aux Travaux Interdisciplinaires, Multi-établissements et Exploratoires programme by Conseil Régional Hauts-de-France (volet partenarial phase 2, project PERSO-SURG).

Impact of Centralized Management of Bariatric Surgery Complications on 90-day Mortality.

BACKGROUND AND AIMS: The potential benefit of the centralization of Bariatric surgery (BS) remains debated. The aim of this study was to evaluate the impact on 90-day mortality of an innovative organization aiming at centralizing the care of severe postoperative complications of BS. STUDY DESIGN: The centralization of care for postoperative complication after BS was implemented by French Authorities in 2013 in the Nord-Pas-de-Calais Region, France. This unique formalized network (OSEAN), coordinated by 1 tertiary referral center, enrolled all regional institutions performing bariatric surgery. Data were extracted from the medico-administrative database providing information on all patients undergoing BS between 2009 and 2016 in OSEAN (n = 22,928) and in Rest of France (n = 288,942). The primary outcome was the evolution of 90-day mortality before and after the implementation of this policy. Rest of France was used as a control group to adjust the results to improvement with time of BS outcomes. RESULTS: The numbers of primary procedure and reoperations increased similarly before and after 2013 within OSEAN and in Rest of France. The 90-day mortality rate became significantly lower within OSEAN than in the rest of France after 2013 (0.03% vs 0.08%, P < 0.01). This difference was confirmed in multivariate analysis after adjustment to the procedure specific mortality (P < 0.04). The reduction of 90-day mortality was most visible for sleeve gastrectomy. CONCLUSION: The implementation of centralized care for early postoperative complications after BS in OSEAN was associated with reduced 90-day mortality. Our results indicate that this reduction was not due to a lower incidence of complications but to the improvement of their management.

Impact of Surgical Technique on Long-term Complication Rate After Laparoscopic Adjustable Gastric Banding (LAGB): Results of a Single-blinded Randomized Controlled trial (ANOSEAN Study).

OBJECTIVE: The aim of the study was to explore the impact of the absence of band fixation on the reoperation rate and to identify other risk factors for long-term complications. BACKGROUND: Laparoscopic adjustable gastric banding has been demonstrated to permit important weight loss and comorbidity improvement, but some bands will have to be removed mainly for failure or in case of planned 2-step surgery. Then, the absence of a gastro-gastric suture (GGS) would allow easier band removal. There are insufficient data to conclude that GGS should be abandoned, as the associated risk of band slippage has not been prospectively assessed. METHODS: The ANOSEAN study was a randomized controlled single-blind trial (CPP 2009-A00346-51). Primary outcome was reintervention rate for band removal or repositioning at 3 years. It included 706 patients in 17 bariatric centers. Patients in group 1 received a gastric band with GGS. Inclusion criteria were adapted from National Institutes of Health recommendations. Surgical technique was standardized among all surgeons. RESULTS: At 3 years, the reintervention rate for band retrieval or repositioning was significantly higher in the absence of band fixation (19.4% vs11.3%; P = 0.013), partly because of the slippage rate (10.3% vs 3.6%; P = 0.005). Body mass index <40 kg/m at baseline was also an independent risk factor of slippage (odds ratio 2.769, 95% confidence interval 1.373, 5.581). CONCLUSIONS: GGS prevents band slippage and lower reintervention rate at 3 years. Fixation could be discussed for patients with high BMI who are scheduled to undergo 2-step surgery, but it needs to be specifically assessed.

Preoperative weight loss with intragastric balloon decreases the risk of significant adverse outcomes of laparoscopic gastric bypass in super-super obese patients.

BACKGROUND: Extreme obesity increases morbi-mortality in bariatric surgery. We previously showed that significant weight loss can be obtained within 3 months in super-super obese patients (BMI > 60 kg/m(2)) with an intragastric balloon (IGB). Here, we explored the potential benefit of preoperative IGB on the outcome of laparoscopic gastric bypass (LGBP) in super-super obese patients. METHODS: In this case-control study, we compared the records of 60 consecutive super-super obese patients (66.5 ± 3.4 kg/m(2)) submitted to a LGBP between 2004 and 2009, with preoperative IGB (n = 23, cases) or without (n = 37, controls). We analyzed the clinical outcome of surgery and a composite end point of significant adverse events defined as the presence of at least one of the following conditions: conversion to laparotomy, intensive care unit stay for more than 2 days, and overall hospital stay superior to 2 weeks. RESULTS: All baseline clinical and biological characteristics were homogenous between both groups. IGB was maintained during 155 ± 62 days and induced a loss of 5.5 ± 1.3 kg/m(2) (11.2 ± 3.2% of excess body mass index) at the time of LGBP, associated with a decrease in systolic blood pressure and gamma-glutamyl transpeptidase level (p < 0.05 vs. baseline). Operative time was lower in the IGB group (146 ± 47 vs. 201 ± 81 min in controls; p < 0.01). Significant adverse events occurred less frequently after LGBP in the IGB group (2 vs. 13 in controls; p < 0.05). All patients were alive at 1 year and overall weight loss was similar in both groups (52.4 ± 17.3 vs. 50.3 ± 12.7 percent of excess BMI loss in controls; NS). CONCLUSIONS: IGB prior to LGBP in super-super obese patients significantly reduced excess BMI. It was associated with a shorter operative time and a lower overall risk of significant adverse outcomes.

Chylothorax: unusual complication after laparoscopic gastric banding.

Laparoscopic adjustable gastric banding is one of the most frequently performed bariatric procedures because of its low operative risk and morbidity. Postoperative chylothorax has never been reported following bariatric surgery. We present the case of a 41-year-old woman who developed a massive right chylothorax after a laparoscopic gastric banding, whose lymphogram showed thoracic duct disruption. Good outcome was achieved after thoracoscopy approach with duct ligation. Although rare, chylothorax is a severe complication, and surgeons must be aware of anatomic landmarks. Chyle leak could be under-diagnosed in postoperative uncomplicated pleural effusions.