Samuel B Tomlinson1, Keaton Piper2, Kristopher T Kimmell2, G Edward Vates2. 1. Department of Neurosurgery, University of Rochester Medical Center, Rochester, New York, USA. Electronic address: samuel_Tomlinson@urmc.rochester.edu. 2. Department of Neurosurgery, University of Rochester Medical Center, Rochester, New York, USA.
Abstract
BACKGROUND: Evaluating preoperative frailty is critical for guiding shared surgical decision-making. The purpose of this study was to develop a novel preoperative frailty index for classification of adverse outcomes following cranial neurosurgery procedures. METHODS: The American College of Surgeons National Surgical Quality Improvement Program database was queried for all cranial neurosurgery cases from 2006 to 2014. Sequential univariate and multivariate testing was used to identify significant independent predictors of 30-day mortality. Frailty scores were computed by summating across weighted predictors. Receiver operating characteristic curve analysis quantified the discriminative capacity of the frailty score for classifying mortality and other major adverse outcomes. RESULTS: List-wise exclusion of patients with incomplete datasets yielded a final sample of 27,098 patients (mortality rate = 3.9%). Multivariate regression testing identified 19 independent predictors of 30-day mortality. Receiver operating characteristic curve analysis revealed impressive outcome discrimination (area under the curve = 0.87, P < 0.001, optimal classification accuracy = 83.0%). Patients in the "high-risk" group (score ≥4, n = 5155) had significantly increased risk for mortality (15.4%) and major adverse outcomes (32.0%) compared with patients in the "low-risk" group (n = 21,943, mortality = 1.2%, major adverse outcomes = 4.0%). The frailty score remained highly discriminative across all age groups examined. CONCLUSIONS: Neurosurgical patients undergo extensive preoperative evaluation, but the field currently lacks a robust bedside scoring system for quantifying patient frailty. In this study, we introduced a novel preoperative frailty index capable of classifying 30-day morbidity and mortality outcomes following cranial neurosurgeries.
BACKGROUND: Evaluating preoperative frailty is critical for guiding shared surgical decision-making. The purpose of this study was to develop a novel preoperative frailty index for classification of adverse outcomes following cranial neurosurgery procedures. METHODS: The American College of Surgeons National Surgical Quality Improvement Program database was queried for all cranial neurosurgery cases from 2006 to 2014. Sequential univariate and multivariate testing was used to identify significant independent predictors of 30-day mortality. Frailty scores were computed by summating across weighted predictors. Receiver operating characteristic curve analysis quantified the discriminative capacity of the frailty score for classifying mortality and other major adverse outcomes. RESULTS: List-wise exclusion of patients with incomplete datasets yielded a final sample of 27,098 patients (mortality rate = 3.9%). Multivariate regression testing identified 19 independent predictors of 30-day mortality. Receiver operating characteristic curve analysis revealed impressive outcome discrimination (area under the curve = 0.87, P < 0.001, optimal classification accuracy = 83.0%). Patients in the "high-risk" group (score ≥4, n = 5155) had significantly increased risk for mortality (15.4%) and major adverse outcomes (32.0%) compared with patients in the "low-risk" group (n = 21,943, mortality = 1.2%, major adverse outcomes = 4.0%). The frailty score remained highly discriminative across all age groups examined. CONCLUSIONS: Neurosurgical patients undergo extensive preoperative evaluation, but the field currently lacks a robust bedside scoring system for quantifying patient frailty. In this study, we introduced a novel preoperative frailty index capable of classifying 30-day morbidity and mortality outcomes following cranial neurosurgeries.
Authors: Hoi-Ying H Chan; Frank A Segreto; Samantha R Horn; Cole Bortz; Godwin G Choy; Peter G Passias; Hamish H Deverall; Joseph F Baker Journal: Asian Spine J Date: 2019-05-14