Assessing the Predictive Utility of the C-Reactive Protein-to-Lymphocyte Ratio for Mortality in Isolated Traumatic Brain Injury: A Single-Center Retrospective Analysis.

Early identification of high-risk traumatic brain injury (TBI) patients is crucial for optimizing treatment strategies and improving outcomes. The C-reactive protein-to-lymphocyte ratio (CLR) reflects systemic immunology and inflammation function and serves as a new biomarker for patient stratification. This study aimed to assess the predictive value of the CLR for mortality in patients with isolated moderate to severe TBI. A retrospective analysis of trauma registry data from 2009 to 2022 was conducted, including 1641 adult patients with isolated moderate to severe TBI. Patient demographics, the CLR, injury characteristics, and outcomes were compared between deceased and surviving patients. Univariate and multivariate analyses were performed to identify mortality risk factors. The optimal CLR cut-off value for predicting mortality was determined using receiver operating characteristic (ROC) curve analysis. The CLR was significantly higher in deceased patients compared to survivors (60.1 vs. 33.9, < 0.001). The optimal CLR cut-off value for predicting mortality was 54.5, with a sensitivity of 0.328 and a specificity of 0.812. The area under the ROC curve was 0.566, indicating poor discriminative ability. In the multivariate analysis, the CLR was not a significant independent predictor of mortality (OR 1.03, = 0.051). After propensity score matching to attenuate the difference in baseline characteristics, including sex, age, comorbidities, conscious level, and injury severity, the high-CLR group (CLR ≥ 54.5) did not have significantly higher mortality compared to the low-CLR group (CLR < 54.5). While the CLR was associated with mortality in TBI patients, it demonstrated poor discriminative ability as a standalone predictor. The association between a high CLR and worse outcomes may be primarily due to other baseline patient and injury characteristics, rather than the CLR itself.