In-Hospital Mortality for Community Acquired Pneumonia and Evaluation According to National Diagnosis and Treatment Guidelines

We retrospectively evaluated 135 patients hospitalized in 2001-2003 with community acquired pneumonia to find out in-hospital mortality, mortality-related factors and also to investigate the adherence to National Diagnosis and Treatment Guidelines for Pneumonia (NDTGP). Demographic data, physical examination and laboratory findings were investigated to evaluate the adherence to NDTGP with respect to hospitalization and treatment criteria. Chi-square (Fisher's exact test), Mann-Whitney U tests and logistic regression were used for statistical analysis. Mean age was 52.7 +/- 21.6 for 87 (64.4%) male and 48 (35.6%) female patients. Predisposing factors were present in 57.8% of the patients, whereas 37.8% had co-morbidity and COPD was the leading one (11.9%). The most frequent symptom was cough (84.4%) and fever was > 40 degrees C in 8 patients. In 49 cases (36.3%) there was hypoxemia, and in 19 cases (14.1%) hyperkapnia. In years 2001, 2002, and 2003 the adherence to NDTGP increased gradually with rates of 66.7%, 83.7%, 94.7% for hospitalization and of 39.6%, 53.1%, 60.5% for treatment regimen, respectively. In-hospital mortality decreased with rates of 10.4%, 8.2% and 5.3% as well. Overall in- hospital mortality rate was 8.1% (11 cases). Age > 65, co-morbidity, confusion, cerebrovascular disease (CVD), bilateral involvement on chest x-ray, signs of sepsis and high blood urea were associated with significantly higher mortality (p<0.05). Multivariate analysis by logistic regression revealed that CVD (p=0.02) and signs of sepsis (p=0.03) were independent risk factors of mortality. As a conclusion, adherence to NDTGP is improving year by year in our department and in- hospital mortality rate decreases; CVD and signs of sepsis were independent predictors of in- hospital mortality.