LAUSR

Background The relationship between biomarkers and hospital-acquired pneumonia (HAP) is understudied, especially in severe cases admitted to the intensive care unit (ICU). Compared with community-acquired pneumonia (CAP), HAP might have different traits regarding biomarkers due to the previous history in hospitals. Methods A total of 593 adult patients were enrolled in this retrospective cohort study to determine the neutrophil/lymphocyte count ratio (NLCR), procalcitonin (PCT), C-reactive protein (CRP) and serum lactate level upon admission to the ICU. According to diagnosis, patients were divided into two groups: non-infection and HAP. Discriminant analysis was performed based on better outcomes of diagnostic performance and severity evaluation. The diagnostic performance of each individual biomarker was assessed by constructing receiver operating characteristic (ROC) curves and calculating the area under each ROC curve (AUROC). Multivariable analysis was also applied to determine the most appropriate prognostic factors. Results NLCR, PCT and CRP were markedly different between the non-infection and HAP groups. NLCR had a worse ability to discriminate severe infection (AUROC 0.626; 95% CI 0.581–0.671) than conventional markers such as CRP (0.685, 95% CI 0.641–0.730) and PCT (0.661, 95% CI 0.615–0.707). In addition, the AUROC of composite biomarkers, especially the combination of NLCR, CRP and WBC, was significantly greater than that of any single biomarker. Conclusions NLCR was not comparable to conventional single biomarkers, such as CRP and PCT, for diagnosing or evaluating the severity of HAP. Composite biomarkers that have good accessibility, especially the combination of NLCR, CRP and WBC, could help with early diagnosis and severity evaluation.