Hospital-acquired pneumonia in Japan may have a better mortality profile than HAP in the United States: a retrospective study.

The characteristics of hospital-acquired pneumonia (HAP) are not well documented. In the present study we investigated the severity and mortality, microbiological profile, and the value of Gram staining in culture-confirmed HAP in a Japanese hospital by retrospective review conducted at a Japanese university hospital. Only culture-confirmed cases with good specimen quality were included in the analysis. The clinical characteristics of HAP, as well as the causative organisms, were investigated. Furthermore, the prognostic ability of existing prediction rules were evaluated for prediction of overall mortality. Forty-two cases were enrolled in this analysis. The majority of patients were admitted to the ICU (61.9 %), and 40.5 % had ventilator-associated pneumonia (VAP). The 30-day mortality was 23.8 %, which is less than that reported in the United States. Factors commonly known to be associated with worse outcome in the USA did not appear to influence the mortality from HAP in Japan. The most frequent causative organisms were methicillin-resistant Staphylococcus aureus (MRSA), followed by Pseudomonas spp. Sensitivity and negative predictive value of Gram staining were 89.4 and 85.7 %, respectively. SMART-COP predicted 30-day mortality with an area under the ROC curve (AUC) >0.7. The characteristics of HAP in Japan differ from HAP reported in the USA. In addition to lower mortality, we found both fewer ICU cases and VAP. Gram staining of good-quality specimens demonstrated promising sensitivity to predict the causative organisms. SMART-COP predicted mortality with appropriate ROC curve (AUC).