Abstract
Background/Objectives: The geriatric nutrition risk index (GnRI) has been regarded as a useful predictor of morbidity and mortality in elderly patients. This study aimed to determine the use of the GnRI as a prognostic predictor in extremely elderly patients undergoing proximal femur fracture surgery and the usefulness of evaluation tools related to a patient's underlying disease and functional capacity in predicting the prognosis of extremely elderly patients. Methods: We analyzed 548 patients who had undergone surgery for proximal femur fracture caused by trauma, with an age of ≥80 years, without other accompanying trauma. Results: Body mass index (BMI) (OR, 1.077; 95% CI, 1.010-1.149; p = 0.023), serum albumin levels (0.389; 0.223-0.678; p = 0.001), and Charlson comorbidity index (CCI) (1.170; 1.014-1.349; p = 0.031) were determined to be predictors of morbidity in a multivariable regression analysis. The area under the curve (AUC) in the receiver operating characteristic curve of BMI was 0.565 (95% CI, 0.493-0.637; p = 0.065), and the optimal cut-off value could not be determined. The AUC of serum albumin was 0.647 (0.576-0.717; p < 0.001), and the optimal cut-off value was 3.65 g/dL (sensitivity, 72.2%; specificity, 52.7%). The AUC of the CCI was 0.648 (0.580-0.715; p < 0.001), and the optimal cut-off value was 6.5 (sensitivity, 63.3%; specificity, 61.4%). Conclusions: The GnRI was not a predictive factor for patient prognosis after proximal femur fractures in extreme elderly patients. Rather, serum albumin level and CCI, which reflect the patient's underlying comorbid conditions, were more useful in predicting in-hospital morbidity after proximal femur surgery in extremely elderly patients.