ABSTRACT
Objective:To investigate the correlation between neutrophil to lymphocyte ratio (NLR) and 30-day clinical outcomes in patients with spontaneous cerebral hemorrhage and whether adding NLR to ICH score improve the accuracy of predicting poor outcomes.Methods:Patients with spontaneous intracerebral hemorrhage admitted to the Department of Neurology, the Second People's Hospital of Hefei from March 2018 to April 2019 were enrolled retrospectively. The demographic and baseline clinical and imaging data were documented. The absolute neutrophil counts and absolute lymphocyte counts within 24 h of onset were obtained and NLR was calculated. At 30 d after the onset of cerebral hemorrhage, the modified Rankin Scale was used to evaluate the outcomes. Good outcome was defined as ≤2, and poor outcome was defined as >2. Multivariate logistic regression analysis was used to determine the independent risk factors for poor outcomes. The receiver operating characteristic (ROC) curve was used to evaluate the predictive value of baseline NLR, ICH score and NLR+ ICH score for poor outcomes in patients with spontaneous cerebral hemorrhage. Results:A total of 159 patients with spontaneous cerebral hemorrhage were enrolled, including 106 males (66.67%), age 62.29±15.10 years. Neutrophil count was (7.30±3.95)×10 9/L, lymphocyte count was (1.41±0.67)×10 9/L, and NLR was 6.94±7.66. Baseline hematoma volume was 17.93±25.87 ml, median ICH score 0 (interquartile range 0-1). The outcomes of 60 patients (37.7%) were poor at 30 d. Univariate analysis showed that coronary heart disease, systolic blood pressure, diastolic blood pressure, high-sensitivity C-reactive protein, fasting blood glucose, white blood cell count, neutrophil count, NLR, hematoma broken into the ventricle, hematoma volume, NIHSS and ICH scores in the poor outcome group were significantly higher than those of the good outcome group (all P<0.05). Multivariate logistic regression analysis showed that NLR was an independent predictor of poor outcomes at 30 d after the onset of spontaneous cerebral hemorrhage (odds ratio 1.135, 95% confidence interval 1.092-2.321; P=0.038). The ROC curve analysis showed that the best cut-off value of NLR was 6.679, and the sensitivity and specificity of predicting poor outcomes were 51.67% and 76.77% respectively; the best cut-off value of ICH score was 1.0, and the sensitivity and specificity of predicting poor outcomes were 69.71% and 89.80% respectively; the sensitivity and specificity of the combined application of NLR + ICH score to predict poor outcomes were 74.58% and 82.65% respectively. Conclusions:NLR was independently associated with poor outcomes at 30 d after the onset of spontaneous cerebral hemorrhage. Adding it to the ICH score could improve the accuracy of predicting poor outcome.