Matrix metalloproteinase-9 and neutrophil to lymphocyte ratio predict delayed perihematomal edema in patients with spontaneous intracerebral hemorrhage

王军,顾双双,高凤娟,蒋云飞,李瑾,沙杜鹃

DOI: https://doi.org/10.3760/cma.j.issn.1673-4165.2021.02.006

2021-01-01

Abstract:Objective:To investigate the predictive value of matrix metalloproteinase-9 (MMP-9) and neutrophil to lymphocyte ratio (NLR) in delayed perihematomal edema (dPHE) after spontaneous intracerebral hemorrhage (sICH).Methods:Patients with sICH admitted to Nanjing Drum Tower Hospital Affiliated to Nanjing University Medical School within 24 h of onset from January 2018 to June 2020 were enrolled retrospectively. Serum MMP-9 levels and peripheral blood cell counts were detected, and NLR were calculated within 24 h of onset. dPHE was defined as an increase of 3 ml in absolute edema volume at 10-21 d after onset of sICH compared with that at 5-9 d. The demographic and baseline clinical and imaging data of the dPHE group and the non-dPHE group were compared. Multivariate logistic regression analysis was used to identify the independent predictors of dPHE. The receiver operating characteristic (ROC) curve was used to evaluate the predictive values of MMP-9 and NLR for dPHE. Results:A total of 195 patients with sICH (61.88±10.60 years old) were enrolled in the study. One hundred and forty-eight patients were males (75.9%). There were 53 patients (27.2%) in the dPHE group and 142 (72.8%) in the non-dPHE group. Univariate analysis showed that age, baseline hematoma volume, baseline National Institutes of Health Stroke Scale score, fasting blood glucose, high-sensitivity C-reactive protein, MMP-9, neutrophil count, NLR and the proportion of irregular hematoma in the dPHE group were significantly higher than those in the non-dPHE group (all P<0.05). Multivariate logistic regression analysis showed that after adjusting for confounding factors, higher MMP-9 (odds ratio [ OR] 4.291, 95% confidence interval [ CI] 2.041-6.590; P=0.007) and higher NLR ( OR 2.530, 95% CI 1.157-4.022; P=0.011) were all the independent predictors of dPHE. ROC curve analysis showed that the area under the curve of MMP-9 for predicting dPHE was 0.819 (95% CI 0.756-0.884; P<0.001), the optimal cut-off value was 164.0 μg/L, and the sensitivity and specificity were 86.79% and 66.90% respectively. The area under the curve of NLR for predicting dPHE was 0.788 (95% CI 0.719-0.856; P<0.001), the optimal cut-off value was 5.683, and the corresponding sensitivity and specificity were 77.36% and 71.13% respectively. Conclusions:sICH patients with higher baseline MMP-9 and NLR are more likely to develop dPHE. Early detection of MMP-9 and NLR in peripheral blood after admission can predict dPHE.