The Geriatric Nutritional Risk Index as an Alternative to the EuroSCORE II in Estimating Cardiac Surgery Risk.

OBJECTIVES: The Geriatric Nutritional Risk Index (GNRI) was introduced to predict mortality in elderly people (age ≥ 65 years). In contrast to common risk scores for cardiac surgery, the GNRI relies on just 6 clinical parameters: age, height, weight, sex, albumin, and sodium. METHODS: This study presents a large single-centre, retrospective analysis of all consecutive patients ≥ 65 years who underwent cardiac surgery using cardiopulmonary bypass (CPB) between 2010 and 2023. Patients were stratified according to GNRI risk classes: major risk (<82), moderate risk (82 to <92), low risk (92 to ≤98), and minimal risk (>98). Receiver operating risk analyses were performed to compare the predictive power of 30-day mortality of GNRI versus the EuroSCORE II. RESULTS: In total, 6712 patients were included: 52 (0.8%) patients were at major risk, 233 (3.5%) patients were at moderate risk, 506 (7.5%) patients were at low risk, and 5921 (88.2%) patients were at minimal risk. The GNRI had the same predictive power for 30-day mortality as the EuroSCORE II (AUC = 0.75 vs AUC = 0.71, P = .051). Notably, when adjusted for the type of surgery, GNRI reached an AUC = 0.77 and did statistically not differ from the EuroSCORE II (P = .6). The Hosmer-Lemeshow test indicated good calibration for both the GNRI model (P = .301) and the adjusted GNRI model (P = .619) and revealed P < .0001 for the EuroSCORE II. CONCLUSIONS: The GNRI demonstrates predictive power for all-cause 30-day mortality comparable to the EuroSCORE II in a real-world patient cohort. GNRI offers a substantial simplification of risk estimation in patients undergoing cardiac surgery with CPB.