Journal of Clinical Neuroscience · May 1, 2026
Assessing the Assessors: Comparative performance of frailty and nutritional metrics in predicting 30-Day morbidity and mortality after cranial neurosurgery
By Farid M, Wang J, Perdikis B et al.
INTRODUCTION: Frailty and nutritional status are increasingly recognized as key determinants of perioperative outcomes in cranial neurosurgery. Numerous indices, including the American College of Surgeons Surgical Risk Calculator (ACS-SRC), are used for risk stratification, yet their comparative accuracy across complication domains and neurosurgical indications remains uncertain. OBJECTIVE: To compare the predictive performance of five preoperative risk metrics, ACS-SRC, serum albumin, Risk Analysis Index (RAI), Geriatric Nutritional Risk Index (GNRI), and Modified Frailty Index-5 (mFI-5), for forecasting postoperative morbidity and mortality after cranial neurosurgery performed for trauma, tumor, vascular, or functional indications. METHODS: Adults (>18 years) undergoing cranial neurosurgical procedures in the ACS NSQIP database (2017-2022) were analyzed. Multivariable logistic regression generated adjusted odds ratios (aORs). Discriminative performance was measured using ROC analysis and C-statistics (DeLong test). Subgroup analyses were stratified by surgical indication. RESULTS: Among 77,514 patients, the ACS-SRC demonstrated the highest accuracy across outcomes, with C-statistics of 0.87 (95% CI: 0.87-0.88) for mortality, 0.72 (95% CI: 0.72-0.73) for overall complications, and 0.75 (95% CI: 0.74-0.75) for major complications. Serum albumin was the next strongest predictor (mortality C-statistic 0.68, 95% CI: 0.67-0.69), outperforming or equaling frailty metrics on most endpoints. RAI exceeded mFI-5 in predictive value (mortality C-statistic 0.70, 95% CI: 0.69-0.71), while GNRI performed similarly to mFI-5 but below RAI. Higher RAI, mFI-5, and ACS-SRC scores, along with lower albumin levels, independently predicted increased 30-day mortality and complications. Subgroup analyses showed that ACS-SRC and serum albumin maintained strong performance across trauma, tumor, and vascular cohorts. CONCLUSION: ACS-SRC remains among the most accurate preoperative tools for predicting adverse outcomes after cranial neurosurgery. Serum albumin offers strong predictive capability and may serve as a practical alternative when calculator inputs are incomplete. These findings reinforce the value of structured risk stratification for surgical counseling and outcome optimization.