Malnutrition is highly prevalent among patients with central nervous system (CNS) injuries. Although nutrition-related blood biomarkers help evaluate malnutrition, their prognostic value during CNS injury rehabilitation remains unclear. This study aimed to evaluate the impact of nutrition-related blood biomarkers on rehabilitation outcomes in CNS injury patients. A total of 669 patients with CNS injuries admitted to a Chinese tertiary teaching hospital from December 2017 to December 2025 were enrolled. Functional outcomes were assessed using the mean relative functional gain (mRFG) of the modified Barthel index. Independent prognostic factors were identified using both univariate and multivariate analyses, including demographic and clinical characteristics, complications, and blood biomarkers. The receiver operating characteristic (ROC) curve analysis was employed to assess the optimal cutoff values for independent predictors. Findings were validated using restricted cubic spline (RCS) models, subgroup analyses, and sensitivity analyses. Multivariate analysis demonstrated that albumin (odds ratio OR = 0.934, P = 0.029), total cholesterol (TC, OR = 1.816, P < 0.001), and low-density lipoprotein cholesterol (LDL-C, OR = 1.580, P = 0.003) were independent predictors of functional prognosis. When grouped by cutoff values, significant differences in mRFG were observed for albumin (cutoff value = 36.35 g/L, P = 0.002), TC (cutoff value = 3.24 mmol/L, P < 0.001), and LDL-C (cutoff value = 1.89 mmol/L, P < 0.001). RCS analyses demonstrated no significant nonlinear relationships. Subgroup analyses revealed no significant effect modification by injury type, although TC showed significant positive associations in spinal cord injury and hemorrhagic stroke. A significant gender interaction was found for TC (P for interaction = 0.021), with a stronger association in males (OR = 2.43, 95% CI: 1.70–3.48). Albumin, TC, and LDL-C are independent predictors of functional prognosis in patients with CNS injury. Albumin is protective, while TC and LDL-C correlate with poorer outcomes. The identified cutoff values may help clinicians guide patient management and prognostic assessment.
Cheng et al. (Tue,) studied this question.