Bariatric surgery: trends in utilization, complications, conversions and revisions.

BACKGROUND: Sleeve gastrectomy (SG) increased in popularity after 2010 but recent data suggest it has concerning rates of gastroesophageal reflux and need for conversions. This study aims to evaluate recent trends in the utilization of bariatric procedures, associated complications, and conversions using an administrative claims database in the United States. METHODS: We included adults who had bariatric procedures from 2000 to 2020 with continuous enrollment for at least 6 months in the MarketScan Commercial Claims and Encounters database. Index bariatric procedures and subsequent revisions or conversions were identified using CPT codes. Baseline comorbidities and postoperative complications were identified with ICD-9-CM and ICD-10 codes. Cumulative incidences of complications were estimated at 30-days, 6-months, and 1-year and compared with stabilized inverse probability of treatment weighted Kaplan-Meier analysis. RESULTS: We identified 349,411 bariatric procedures and 5521 conversions or revisions. The sampled SG volume appeared to begin declining in 2018 while Roux-en-Y gastric bypass (RYGB) remained steady. Compared to RYGB, SG was associated with lower 1-year incidence [aHR, (95% CIs)] for 30-days readmission [0.65, (0.64-0.68)], dehydration [0.75, (0.73-0.78)], nausea or vomiting [0.70, (0.69-0.72)], dysphagia [0.55, (0.53-0.57)], and gastrointestinal hemorrhage [0.43, (0.40-0.46)]. Compared to RYGB, SG was associated with higher 1-year incidence [aHR, (95% CIs)] of esophagogastroduodenoscopy [1.13, (1.11-1.15)], heartburn [1.38, (1.28-1.49)], gastritis [4.28, (4.14-4.44)], portal vein thrombosis [3.93, (2.82-5.48)], and hernias of all types [1.36, (1.34-1.39)]. There were more conversions from SG to RYGB than re-sleeving procedures. SG had a significantly lower 1-year incidence of other non-revisional surgical interventions when compared to RYGB. CONCLUSIONS: The overall volume of bariatric procedures within the claims database appeared to be declining over the last 10 years. The decreasing proportion of SG and the increasing proportion of RYGB suggest the specific complications of SG may be driving this trend. Clearly, RYGB should remain an important tool in the bariatric surgeon's armamentarium.

Updates in Bariatric Surgery.

Bariatric surgery is currently the most effective long-term treatment for morbid obesity as well as type-2 diabetes mellitus. The field of metabolic and bariatric surgery has seen tremendous growth over the past decade with dramatically reduced risks. This article aims to provide an update on bariatric surgery, highlighting the latest outcomes, improvements, and challenges in the field. Recently, the American Society of Metabolic and Bariatric Surgery (ASMBS) and the International Federation for the Surgery of Obesity and Metabolic Disorders (IFSO) released a major update to the indications for bariatric surgery at BMI ≥35 kg/m2 regardless of co-morbidities and 30-34.9 kg/m2 with obesity-related comorbidities. Sleeve gastrectomy has emerged as the most popular bariatric procedure in the last 10 years with its remarkable efficacy and safety profile. The implementation of the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) and Enhanced Recovery After Surgery (ERAS) protocols have significantly improved the quality of care for all bariatric patients. The recent introduction and FDA approval of Glucagon-Like Peptide-1 (GLP-1) agonists for chronic obesity has garnered significant media coverage and popularity, but no guidelines exist regarding its use in relation to bariatric surgery. This update underscores the need for tailored approaches, ongoing research, and the integration of evidence-based medicine and innovations to enhance patient care.

Application of machine learning to predict postoperative gastrointestinal bleed in bariatric surgery.

BACKGROUND: Postoperative gastrointestinal bleeding (GIB) is a rare but serious complication of bariatric surgery. The recent rise in extended venous thromboembolism regimens as well as outpatient bariatric surgery may increase the risk of postoperative GIB or lead to delay in diagnosis. This study seeks to use machine learning (ML) to create a model that predicts postoperative GIB to aid surgeon decision-making and improve patient counseling for postoperative bleeds. METHODS: The Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) database was used to train and validate three types of ML methods: random forest (RF), gradient boosting (XGB), and deep neural networks (NN), and compare them with logistic regression (LR) regarding postoperative GIB. The dataset was split using fivefold cross-validation into training and validation sets, in an 80/20 ratio. The performance of the models was assessed using area under the receiver operating characteristic curve (AUROC) and compared with the DeLong test. Variables with the strongest effect were identified using Shapley additive explanations (SHAP). RESULTS: The study included 159,959 patients. Postoperative GIB was identified in 632 (0.4%) patients. The three ML methods, RF (AUROC 0.764), XGB (AUROC 0.746), and NN (AUROC 0.741) all outperformed LR (AUROC 0.709). The best ML method, RF, was able to predict postoperative GIB with a specificity and sensitivity of 70.0% and 75.4%, respectively. Using DeLong testing, the difference between RF and LR was determined to be significant with p < 0.01. Type of bariatric surgery, pre-op hematocrit, age, duration of procedure, and pre-op creatinine were the 5 most important features identified by ML retrospectively. CONCLUSIONS: We have developed a ML model that outperformed LR in predicting postoperative GIB. Using ML models for risk prediction can be a helpful tool for both surgeons and patients undergoing bariatric procedures but more interpretable models are needed.

Comparison of Robotic, Laparoscopic, and Open Unilateral Repair of Non-recurrent Inguinal Hernia.

BACKGROUND: There is a paucity of data comparing open, robotic, and laparoscopic approaches on unilateral, non-recurrent inguinal hernias. Our study presents a large, retrospective triple-arm outcome analysis between robotic, laparoscopic, and open unilateral, non-recurrent inguinal hernia repairs at a single institution. METHODS: 706 patients who underwent elective, non-recurrent inguinal hernia repair performed by 8 general surgeons at a single institution from 2016 to 2019 were reviewed retrospectively. Patient baseline characteristics, operative times, resident involvement, and postoperative outcomes were analyzed for all repair types. A cost analysis of the different procedures was performed. RESULTS: There were 305 laparoscopic repairs, 207 robotic repairs, and 194 open repairs. Open and laparoscopic repairs were performed on patients who were older (p =< .001) and with a higher Charlson Comorbidity Index (p =< .001). Patient BMI was higher in minimally invasive repair than open repair (P = .021). There were no significant differences in complication rates on pairwise analysis. Robotic and open repairs had significantly longer operative times than laparoscopic repairs (P < .001). There was less resident involvement in robotic repair than with the other approaches (P < .001). Resident involvement was associated with shorter OR times (P = .001) and no significant difference in postoperative complications. There was a trend over the study period toward faster operative times and more robotic repair. Robotic repair is the most expensive repair, followed by laparoscopic and open repairs. CONCLUSION: All 3 repair techniques can be performed without significant differences in outcomes. The technique utilized should be based on surgeon preference and patient characteristics.

Force feedback vessel ligation simulator in knot-tying proficiency training.

BACKGROUND: Tying gentle secure knots is an important skill. We have developed a force feedback simulator that measures force exerted during knot tying. This pilot study examines the benefits of this simulator in a deliberate practice curriculum. METHODS: The simulator consists of silastic tubing with a force sensor. Knot quality was assessed using digital caliper measurement. Participants performed 10 vessel ligations as a pretest, then were shown force readings and tied knots until reaching proficiency targets. Average peak forces precurriculum and postcurriculum were compared using Student t test. RESULTS: Participants exerted significantly less force after completing the curriculum (.61 N ± .22 vs 1.42 N ± .53, P < .001), and had fewer air knots (10% vs 27%). The curriculum was completed in an average of 19.4 ± 6.27 minutes and required an average of 11.7 ± 4.03 knots to reach proficiency. CONCLUSIONS: This study demonstrates the feasibility of real-time feedback in learning to tie delicate knots. The curriculum can be completed in a reasonable amount of time, and may also work as a warm-up exercise before a surgical case.

Design of vessel ligation simulator for deliberate practice.

BACKGROUND: Surgical residents develop technical skills at variable rates, often based on random chance of cases encountered. One such skill is tying secure knots without exerting excessive force. This study describes the design of a simulator using a force sensor to measure instantaneous forces exerted on a blood vessel analog during vessel ligation and the development of expert-derived performance goals. MATERIALS AND METHODS: Vessel ligations were performed on Silastic tubing at an offset from a Vernier Force Sensor. Nine experts (surgical faculty and senior residents) and 10 novices (junior residents) were recruited to each perform 10 vessel ligations (two square knots each) with two-handed and one-handed techniques. Internal consistency for the series of vessel ligations was tested with Cronbach alpha. Maximum forces exerted by novices and experts were compared using Student t-test. RESULTS: Internal consistency across the 10 ligations on the simulator was excellent (Cronbach alpha = 0.91). The expert group on average exerted a significantly lower maximum force when compared with novices while performing two-handed (0.76 ± 0.39 N versus 1.12 ± 0.49 N, P < 0.01) and one-handed (0.84 ± 0.32 N versus 1.36 ± 0.44 N, P < 0.01) vessel ligations. CONCLUSIONS: Although the expert group performed vessel ligations with significantly lower peak force than the novice group, there were novices who performed at the expert level. This is consistent with the conceptual framework of milestones and suggests that the skill of gentle knot-tying can be measured and develops at different chronologic levels of training in different individuals. This simulator can be used as part of a deliberate practice curriculum with instantaneous visual feedback.