Predictive Performance of Routine Blood Test Parameters for the Severity of Brain Damage in Patients With Acute Cerebral Hemorrhage in High-Altitude Regions

Objective To investigate the predictive value of changes in various indicators of routine blood tests for the severity of brain damage in patients with acute cerebral hemorrhage (ACH) in high-altitude regions.

Methods A retrospective analysis was conducted using data from 249 inpatients diagnosed with ACH at a hospital in Xizang. Patient data, including demographic information and results of routine blood tests, were collected. Based on their Glasgow Coma Scale (GCS) scores, the patients were divided into 2 groups, a mild brain damage group (GCS ≥ 13 points) and a moderate-to-severe brain damage group (GCS ≤ 12 points). Demographic information and laboratory test results were compared between the two groups. Binary logistic regression analysis was performed to examine the relationship between indicators of routine blood tests and the severity of brain damage, and key indicators were identified. The receiver operating characteristic (ROC) curve analysis was used to evaluate the predictive performance of key indicators in various combinations for the severity of brain injury in ACH patients.

Results The moderate-to-severe group had a higher proportion of high-altitude residents, as well as elevated levels of white blood cells (WBC), neutrophils (N), neutrophil-to-lymphocyte ratio (NLR), systemic inflammatory index (SII), neutrophil-to-platelet ratio (NPR), and C-reactive protein (CRP)-to-lymphocyte ratio (CLR), compared to the mild group. On the other hand, the levels of lymphocytes (L), platelets (PLT), and platelet-to-albumin ratio (PAR) in the moderate-to-severe group were lower than those in the mild group. Binary logistic regression analysis showed that WBC (OR = 1.221, 95% CI: 1.127-1.322), CRP (OR = 1.019, 95% CI: 1.004-1.033), CRP-to-albumin ratio (CAR) (OR = 1.845, 95% CI: 1.137-2.996), and CLR (OR = 1.018, 95% CI: 1.005-1.030) were positively associated with moderate-to-severe brain damage, and that their elevated levels were associated with an increase in risk by the corresponding folds. PAR (OR = 0.845, 95% CI: 0.721-0.990) was negatively correlated with moderate-to-severe brain damage, with an increase resulting in a reduction of risk by 0.845 times. ROC curve analysis showed that, in the mild group, the combined AUC for WBC + CRP was 0.689 (optimal cutoff value 0.19, specificity 0.776, 95% CI: 0.624-0.755) ; the PAR + CAR + CLR combination (CAR was not significant, P > 0.05) had an AUC of 0.675 (optimal cutoff value 0.19, specificity 0.760, 95% CI: 0.609-0.741); the key indicator combination of WBC + PAR + CLR (CRP was not significant, P > 0.05) demonstrating the strong predictive performance moderate-to-severe brain damage, yielding an AUC of 0.737 (optimal cutoff value 0.08, specificity 0.624, 95% CI: 0.676-0.798).

Conclusion The indicator combinations of WBC + CRP, PAR + CLR, and WBC + PAR + CLR exhibit significant value for predicting the severity of brain injury in ACH patients and may serve as potential predictive tools for the severity of brain damage in patients with acute cerebral hemorrhage in high-altitude regions.