Predictive value of the risk analysis index for 30-day mortality following surgical management of thoracolumbar vertebral body fractures.

Purpose: Frailty has emerged as a key predictor of postoperative outcomes, with indices like the revised Risk Analysis Index (RAI) and the modified Frailty Index (mFI-5) used for assessment. The purpose of this study was to compare the predictive value and discriminatory accuracy of the mFI-5 and RAI for adverse outcomes following surgical management of thoracolumbar (TL) spine fractures.

Methods: This was a retrospective cohort study utilizing data from the 2015-2019 years of the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database. Patients were categorized using the RAI and mFI-5 frailty indices. The primary outcome was 30-day mortality and secondary outcomes included non-home discharge, extended hospital stays, and postoperative complications. Logistic regression was used to analyze the predictive value of frailty and discriminatory accuracy was assessed using receiver operating curve analysis.

Results: There were 42,230 patients (median age = 64 years) included in this study. Increasing frailty was a significant predictor of greater odds for 30-day mortality. The RAI demonstrated superior discriminatory accuracy for 30-day mortality (C-statistic = 0.82) compared to both mFI-5 (C-statistic = 0.63) and age alone (C-statistic = 0.66).

Conclusion: The RAI is a more effective tool than the mFI-5 and age alone for preoperative frailty-based risk stratification in TL fracture patients, particularly in predicting 30-day mortality. Incorporating the RAI into clinical practice may improve preoperative informed consent, surgical candidate selection, and patient outcomes.