BACKGROUND CONTEXT Preoperative prognostication of 30-day mortality in patients with metastatic spine tumors can optimize surgical risk stratification and guide the decision-making process to improve survival. PURPOSE To develop and validate a set of predictive variables of 30-day mortality following surgery for metastatic spine tumors. STUDY DESIGN/SETTING The patient cohort was identified from the American College of Surgeons National Surgical Quality Improvement Program (2005-2016). PATIENT SAMPLE Statistical analysis included 3,566 patients with 30-day mortality in 2.8% (n=100) patients. OUTCOME MEASURES Machine Learning Algorithm METHODS We performed logistic regression (enter, stepwise and forward) and least absolute shrinkage and selection operator (LASSO) method for selection of variables, which resulted in 18-candidate models. The final model was selected based upon clinical knowledge and numerical results. RESULTS The model with 10-predictive factors which included: gender, American Society of Anesthesiologiss Grade, body mass index, location of the tumor, type of surgery, functional health status, chronic obstructive pulmonary disorder, preoperative serum albumin, preoperative hematocrit and preoperative white blood cell count performed best on discrimination, calibration, Brier score and decision analysis to develop a machine learning algorithm. Logistic regression showed higher AUCs than LASSO across these different models. The predictive probability derived from the best model was uploaded on an open access web application which can be found at: https://spine.massgeneral.org/drupal/Mortality-MetastaticSpineTumor CONCLUSIONS Machine learning algorithms show promising results for predicting 30-day mortality following surgery for metastatic spine tumors. These algorithms can be useful aids for counseling patients, assessing pre-operative medical risks, and predicting survival after surgery. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.
Read full abstract