30-day Major Adverse Cardiac Events (MACE) after elective bariatric surgery: the development and validation of the Bari-MACE clinical predictive tool

Machine learning prediction of major adverse cardiac events after elective bariatric surgery

BACKGROUND

Machine learning (ML) is an emerging technology with the potential to predict and improve clinical outcomes including adverse events, based on complex pattern recognition. Major adverse cardiac events (MACE) after bariatric surgery have an incidence of 0.1% but carry significant morbidity and mortality. Prior studies have investigated these events using traditional statistical methods, however, studies reporting ML for MACE prediction in bariatric surgery remain limited. As such, the objective of this study was to evaluate and compare MACE prediction models in bariatric surgery using traditional statistical methods and ML.

METHODS

Cross-sectional study of the MBSAQIP database, from 2015 to 2019. A binary-outcome MACE prediction model was generated using three different modeling methods: (1) main-effects-only logistic regression, (2) neural network with a single hidden layer, and (3) XGBoost model with a max depth of 3. The same set of predictor variables and random split of the total data (50/50) were used to train and validate each model. Overall performance was compared based on the area under the receiver operating curve (AUC).

RESULTS

A total of 755,506 patients were included, of which 0.1% experienced MACE. Of the total sample, 79.6% were female, 47.8% had hypertension, 26.2% had diabetes, 23.7% had hyperlipidemia, 8.4% used tobacco within 1 year, 1.9% had previous percutaneous cardiac intervention, 1.2% had a history of myocardial infarction, 1.1% had previous cardiac surgery, and 0.6% had renal insufficiency. The AUC for the three different MACE prediction models was: 0.790 for logistic regression, 0.798 for neural network and 0.787 for XGBoost. While the AUC implies similar discriminant function, the risk prediction histogram for the neural network shifted in a smoother fashion.

CONCLUSION

The ML models developed achieved good discriminant function in predicting MACE. ML can help clinicians with patient selection and identify individuals who may be at elevated risk for MACE after bariatric surgery.

How Good is Stratification and Prediction Model Analysis Between Primary and Revisional Roux-en-Y Gastric Bypass Surgery? A Multi-center Study and Narrative Review

INTRODUCTIONS

Revision surgery because of weight recurrence is performed in 2.5-33% of primary vertical banded gastroplasty (VBG), laparoscopic sleeve gastrectomy (LSG), and gastric band (GB) cases. These cases qualify for revisional Roux-en-Y gastric bypass (RRYGB).

METHODS

This retrospective cohort study analyzed data from 2008 to 2019. A stratification analysis and multivariate logistic regression for prediction modeling compared the possibility of sufficient % excess weight loss (%EWL) ≥ 50 or insufficient %EWL < 50 between three different RRYGB procedures, with primary Roux-en-Y gastric bypass (PRYGB) as the control during 2 years of follow-up. A narrative review was conducted to test the presence of prediction models in the literature and their internal and external validity.

RESULTS

A total of 558 patients underwent PRYGB, and 338 underwent RRYGB after VBG, LSG, and GB, and completed 2 years of follow-up. Overall, 32.2% of patients after RRYGB had a sufficient %EWL ≥ 50 after 2 years, compared to 71.3% after PRYGB (p ≤ 0.001). The total %EWL after the revision surgeries for VBG, LSG, and GB was 68.5%, 74.2%, and 64.1%, respectively (p ≤ 0.001). After correcting for confounding factors, the baseline odds ratio (OR) or sufficient %EWL ≥ 50 after PRYGB, LSG, VBG, and GB was 2.4, 1.45, 0.29, and 0.32, respectively (p ≤ 0.001). Age was the only significant variable in the prediction model (p = 0.0016). It was impossible to develop a validated model after revision surgery because of the differences between stratification and the prediction model. The narrative review showed only 10.2% presence of validation in the prediction models, and 52.5% had external validation.

CONCLUSION

Overall, 32.2% of all patients after revisional surgery had a sufficient %EWL ≥ 50 after 2 years, compared to PRYGB. LSG had the best outcome in the revisional surgery group in the sufficient %EWL group and the best outcome in the insufficient %EWL group. The skewness between the prediction model and stratification resulted in a partially non-functional prediction model.

Prediction of thirty-day morbidity and mortality after duodenal switch using an artificial neural network

BACKGROUND

Understanding factors that increase risk of both mortality and specific measures of morbidity after duodenal switch (DS) is important in deciding to offer this weight loss operation. Artificial neural networks (ANN) are computational deep learning approaches that model complex interactions among input factors to optimally predict an outcome. Here, a comprehensive national database is examined for patient factors associated with poor outcomes, while comparing the performance of multivariate logistic regression and ANN models in predicting these outcomes.

METHODS

2907 DS patients from the 2019 Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database were assessed for patient factors associated with the previously validated composite endpoint of 30-day postoperative reintervention, reoperation, readmission, or mortality using bivariate analysis. Variables associated (P ≤ 0.05) with the endpoint were imputed in a multivariate logistic regression model and a three-node ANN with 20% holdback for validation. Goodness-of-fit was assessed using area under receiver operating curves (AUROC).

RESULTS

There were 229 DS patients with the composite endpoint (7.9%), and 12 mortalities (0.4%). Associated patient factors on bivariate analysis included advanced age, non-white race, cardiac history, hypertension requiring 3 + medications (HTN), previous foregut/obesity surgery, obstructive sleep apnea (OSA), and higher creatinine (P ≤ 0.05). Upon multivariate analysis, independently associated factors were non-white race (odds ratio 1.40; P = 0.075), HTN (1.55; P = 0.038), previous foregut/bariatric surgery (1.43; P = 0.041), and OSA (1.46; P = 0.018). The nominal logistic regression multivariate analysis (n = 2330; R² = 0.02, P < 0.001) and ANN (R² = 0.06; n = 1863 [training set], n = 467 [validation]) models generated AUROCs of 0.619, 0.656 (training set) and 0.685 (validation set), respectively.

CONCLUSION

Readily obtainable patient factors were identified that confer increased risk of the 30-day composite endpoint after DS. Moreover, use of an ANN to model these factors may optimize prediction of this outcome. This information provides useful guidance to bariatricians and surgical candidates alike.

Effect of high-risk factors on postoperative major adverse cardiovascular and cerebrovascular events trends following bariatric surgery in the United States from 2012 to 2019

BACKGROUND

Recent examination of trends in postoperative major adverse cardiovascular and cerebrovascular events (MACE) following bariatric surgery, including accredited and nonaccredited centers, and the factors affecting those trends, is lacking.

OBJECTIVES

The objective of this study was to evaluate current trends for postoperative MACE after bariatric surgery in both accredited and nonaccredited centers and the factors affecting these trends.

SETTING

This retrospective study was conducted using National Inpatient Sample database from 2012 to 2019.

METHODS

All patients who underwent inpatient laparoscopic sleeve gastrectomy (LSG), open sleeve gastrectomy (SG), laparoscopic Roux-en-Y gastric bypass (LRYGB), and open Roux-en-Y gastric bypass (RYGB) were examined. Composite MACE (acute myocardial infarction, cardiac arrest, acute stroke, and in-hospital death during bariatric surgery hospitalization) was calculated and analyzed over time along with patient demographic and co-morbid diseases using survey-weighted logistic regression.

RESULTS

MACE incidence was lowest for LSG (0.07%), followed by LRYGB (0.16%), SG (3.47%), and RYBG (3.51%). Open procedure, increasing age, male sex, body mass index ≥50, coronary artery disease, congestive heart failure, and chronic kidney disease were independent predictors for increased MACE risk. MACE incidence increased over time for SG (odds ratio [OR] 1.25 [1.16, 1.34]; P < .0001) and RYGB (OR 1.14 [1.06, 1.22]; P = .0004) but decreased for LRYGB (OR 0.93 [0.87, 1] P = .06). After adjustment for high-risk covariates, increased MACE trend seen over time was attenuated in SG (OR 1.13 [1.04-1.22]; P = .005) and RYGB (OR 1.04 [0.96-1.12]; P = .36), while there was minimal effect of these high-risk covariates on MACE trend over time in LSG and LRYGB.

CONCLUSIONS

MACE following LSG and LRYGB is rare, occurring in 0.1% of patients. Persistently increasing high-risk conditions and demographics has had minimal effect on MACE over time for LSG and LRYGB but has had significant effect on MACE trend over time in SG and RYGB.

Patients with sleep-disordered breathing for bariatric surgery

The prevalence of patients with obesity continues to rise worldwide and has reached epidemic proportions. There is a strong correlation between obesity and sleep-disordered breathing (SDB), and, in particular, obstructive sleep apnea (OSA). OSA is often undiagnosed in the surgical population. Bariatric surgery has been recognized as an effective treatment option for both obesity and OSA. Laparoscopic bariatric procedures, particularly laparoscopic sleeve gastrectomy (LSG), have become the most frequently performed procedures. OSA has been identified as an independent risk factor for perioperative complications and failure to recognize and prepare for patients with OSA is a major cause of postoperative adverse events, suggesting that all patients undergoing bariatric surgery should be screened preoperatively for OSA. These patients should be treated with an opioid-sparing analgesic plan and continuous positive airway pressure (CPAP) perioperatively to minimize respiratory complications. With the number of bariatric surgical patients with SDB likely to continue rising, it is critical to understand the best practices to manage this patient population.