Prediction of Antibiotic Efficacy in Cerebral Hemorrhage-Associated Pneumonia Using Inflammatory Hematological Indices: A Retrospective Study.

Purpose: The role of inflammatory response in secondary infections following cerebral hemorrhage has attracted considerable attention. We aimed to explore the associations between inflammatory markers and antibiotic efficacy on the cerebral hemorrhage-associated pneumonia. Methods: We conducted a retrospective study including 200 patients with cerebral hemorrhage-associated pneumonia. In the retrospective study, baseline data, blood cell counts, and C-reactive protein (CRP) were collected. The inflammatory markers including neutrophil-lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR), monocyte-lymphocyte ratio (MLR), systemic immune-inflammatory index (SII) were calculated. The difference between the values at admission and at the time of initial efficacy evaluation was denoted by "Δ". These inflammatory markers were examined as potential biomarkers to predict the efficacy of initial antibiotic treatment in finally included patients by univariate analysis and multivariate logistic regression models. Results: In the final analysis, there were 105 cases with effective antibiotic treatment and 95 cases with ineffective antibiotic treatment. At the time of initial efficacy evaluation, there were significant differences in the values of NLR, MLR, PLR, SII, and CRP, P<0.05. Significant differences were also showed in the values of ΔNLR, ΔMLR, and ΔSII, P<0.05. The results of multivariate logistic regression model showed that ΔNLR (OR: 1.260, P=0.016), ΔPLR (OR: 0.985, P=0.008), and CRP (OR: 0.975, P=0.043) at the time of efficacy evaluation were possibly associated with the antibiotic efficacy. Conclusion: ΔNLR, ΔPLR, and CRP at the time of efficacy evaluation appear the promising parameters in guiding antibiotic treatment in patients with cerebral hemorrhage-associated pneumonia. However, large-sample, multicenter external validation studies are still needed.