RESUMO
<p><b>BACKGROUND</b>Community-acquired pneumonia (CAP) remains one of the leading causes of death from infectious diseases around the world. Most severe CAP patients are admitted to the intensive care unit (ICU), and receive intense treatment. The present study aimed to evaluate the role of the pneumonia severity index (PSI), CURB-65, and sepsis score in the management of hospitalized CAP patients and explore the effect of ICU treatment on prognosis of severe cases.</p><p><b>METHODS</b>A total of 675 CAP patients hospitalized in the Second Affiliated Hospital of Zhejiang University School of Medicine were retrospectively investigated. The ability of different pneumonia severity scores to predict mortality was compared for effectiveness, while the risk factors associated with 30-day mortality rates and hospital length of stay (LOS) were evaluated. The effect of ICU treatment on the outcomes of severe CAP patients was also investigated.</p><p><b>RESULTS</b>All three scoring systems revealed that the mortality associated with the low-risk or intermediate-risk group was significantly lower than with the high-risk group. As the risk level increased, the frequency of ICU admission rose in tandem and LOS in the hospital was prolonged. The areas under the receiver operating characteristic curve in the prediction of mortality were 0.94, 0.91 and 0.89 for the PSI, CURB-65 and sepsis score, respectively. Compared with the corresponding control groups, the mortality was markedly increased in patients with a history of smoking, prior admission to ICU, respiratory failure, or co-morbidity of heart disease. The differences were also identified in LOS between control groups and patients with ICU treatment, heart, or cerebrovascular disease. Logistic regression analysis showed that age over 65 years, a history of smoking, and respiratory failure were closely related to mortality in the overall CAP cohort, whereas age, ICU admission, respiratory failure, and LOS at home between disease attack and hospital admission were identified as independent risk factors for mortality in the high-risk CAP sub-group. The 30-day mortality of patients who underwent ICU treatment on admission was also higher than for non-ICU treatment, but much lower than for those patients who took ICU treatment subsequent to the failure of non-ICU treatment.</p><p><b>CONCLUSIONS</b>Each severity score system, CURB-65, sepsis severity score and especially PSI, was capable of effectively predicting CAP mortality. Delayed ICU admission was related to higher mortality rates in severe CAP patients.</p>