Identifying Clinically Meaningful Subgroups following Open Reduction and Internal Fixation for Proximal Humerus Fractures: A Risk Stratification Analysis for Mortality and 30-day Complications Using Machine Learning

Background Identification of prognostic variables for poor outcomes following open reduction internal fixation (ORIF) of displaced proximal humerus fractures have been limited to singular, linear factors and subjective clinical intuition. Machine learning has the capability to objectively segregate patients based on various outcome metrics and report the connectivity of variables resulting in the optimal outcome. Therefore, the purpose of this study was to (1) utilize unsupervised machine learning to stratify patients to high risk and low risk clusters based on postoperative events, (2) compare the machine learning clusters to the American Society of Anesthesiologists (ASA) classification for assessment of risk, and (3) determine the variables that were associated with high-risk patients after proximal humerus ORIF. Methods The American College of Surgeons – National Surgical Quality Improvement Program (ACS-NSQIP) database was retrospectively queried for patients undergoing ORIF for proximal humerus fractures between 2005-2018. Four unsupervised machine learning clustering algorithms were evaluated to partition subjects into “high risk” and “low risk” subgroups based on combinations of observed outcomes. Demographic, clinical, and treatment variables were compared between these groups using descriptive statistics. A supervised machine learning algorithm was generated to identify patients who were likely to be “high-risk” and were compared to ASA classification. A game-theory based explanation algorithm was used to illustrate predictors of “high-risk” status. Results Overall, 4,670 patients were included, of which 202 were partitioned into the “high-risk” cluster while the remaining (4,468 patients) were partitioned into the “low risk” cluster. Patients in the “high-risk” cluster demonstrated significantly increased rates of the following complications: 30-day mortality, 30-day readmission rates, 30-day reoperation rates, nonroutine discharge rates, length of stay, and rates of all surgical and medical complications assessed with the exception of urinary tract infection (P < .001). The best performing SML algorithm for preoperatively identifying “high-risk” patients was the extreme-gradient boost (XGBoost), which achieved an area under the receiver operating characteristics curve (AUROC) of 76.8%, while ASA classification had an AUROC of 61.7%. Shapley values identified the following predictors of “high-risk” status: greater BMI, increasing age, ASA class 3, increased operative time, male gender, diabetes, and smoking history. Conclusion Unsupervised machine learning identified that “high-risk” patients have a higher risk of complications (8.9%) than “low-risk” groups (0.4%) with respect to 30-day complication rate. A supervised machine learning model selected greater BMI, increasing age, ASA class 3, increased operative time, male gender, diabetes, and smoking history to effectively predict “high-risk” patients.