Predicting Outcomes Following Lower Extremity Open Revascularization Using Machine Learning

Abstract

Lower extremity open revascularization carries significant peri-operative risks; however, outcome prediction tools remain limited. Using machine learning (ML), we developed automated algorithms that predict 30-day outcomes following lower extremity open revascularization. In this multicenter retrospective cohort study, the American College of Surgeons National Surgical Quality Improvement Program-targeted database was used to identify patients who underwent lower extremity infrainguinal open revascularization for atherosclerotic disease between 2011 and 2021. Patients treated for lower extremity aneurysmal disease, acute limb ischemia, trauma, dissection, or malignancy were excluded. Input features included 37 preoperative demographic and clinical variables. The primary outcome was 30-day major adverse limb event (MALE; composite of untreated loss of patency, major reintervention, or major amputation) or death. The 30-day secondary outcomes were individual components of the primary outcome, major adverse cardiovascular event, wound complication, bleeding, other morbidity, nonhome discharge, and unplanned readmission. Our data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, we trained six ML models using preoperative features. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was evaluated with calibration plot and Brier score. Variable importance scores were calculated to determine the top 10 predictive features. Performance was assessed on subgroups based on age, sex, race, ethnicity, symptom status, procedure type, and urgency. Overall, 24,309 patients were included. Thirty-day MALE or death occurred in 2349 (9.3%) patients. Those who developed the primary outcome were older with more comorbidities, had poorer functional status, and were more likely to have high risk physiologic and anatomical features. Our best performing prediction model for 30-day MALE or death was XGBoost, achieving an AUROC of 0.93 (95% confidence interval [CI], 0.92-0.94) (Fig 1). In comparison, logistic regression had an area under the receiver operating characteristic curve of 0.63 (95% CI, 0.61-0.65). For secondary outcomes, XGBoost achieved area under the receiver operating characteristic curves between 0.87 and 0.96. The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.08 (Fig 2). The strongest predictive feature in our algorithm was symptom status (chronic limb-threatening ischemia). Model performance remained robust on all subgroup analyses of specific demographic and clinical populations. Our ML models accurately predict 30-day outcomes following lower extremity open revascularization using preoperative data, performing better than logistic regression. They have potential for important utility in guiding risk mitigation strategies for patients being considered for lower extremity open revascularization to improve outcomes and reduce costs.Fig 2Calibration plot with Brier score for predicting 30-day major adverse limb event (MAKE) or death following lower extremity open revascularization using Extreme Gradient Boosting (XGBoost) model.View Large Image Figure ViewerDownload Hi-res image Download (PPT)