INTRODUCTION: Legionella pneumophila is the causal agent of 5% to 12% of sporadic community-acquired pneumonia cases, though rates are changing with the use of new diagnostic methods. METHODS: This is a retrospective study of all patients admitted to our hospital with community-acquired pneumonia due to Legionella pneumophila between 1997 and 2001. Diagnostic criteria included either a positive Legionella serogroup 1 urinary antigen test or seroconversion and a chest radiograph consistent with pneumonia. RESULTS: A total of 97 patients were studied. Ninety cases (92.8%) were community-acquired and 7 (7.2%) were associated with travelling. In 82 cases (84.5%) the presentation was sporadic. Seventy-five patients were smokers (77.3%). The most common symptoms were fever in 91 patients (93.8%) and cough in 67 (68.1%). In five patients (5.2%) creatine phosphokinase concentrations were over 5 times their baseline values (in two over 100 times); four of these patients presented acute renal failure. Seroconversion was observed in 23/42 patients (54.8%). There were no statistically significant differences between the administration of erythromycin or clarithromycin in monotherapy, or in combination with rifampin. Nineteen patients (19.6%) presented acute renal failure and mechanical ventilation was necessary in 22 (22.7%). Twelve patients died (12.5%). Independent prognostic factors associated with death included respiratory rate > 30 breaths/min, urea > 60 mg/dL and PaO2 < 60 mmHg. A significant linear association was found between severity scale scores and the presence of complications or mortality. CONCLUSION: The Legionella urinary antigen test permits early diagnosis and treatment of this disease. The severity scale is an indicator of complications or death.
INTRODUCTION:Legionella pneumophila is the causal agent of 5% to 12% of sporadic community-acquired pneumonia cases, though rates are changing with the use of new diagnostic methods. METHODS: This is a retrospective study of all patients admitted to our hospital with community-acquired pneumonia due to Legionella pneumophila between 1997 and 2001. Diagnostic criteria included either a positive Legionella serogroup 1 urinary antigen test or seroconversion and a chest radiograph consistent with pneumonia. RESULTS: A total of 97 patients were studied. Ninety cases (92.8%) were community-acquired and 7 (7.2%) were associated with travelling. In 82 cases (84.5%) the presentation was sporadic. Seventy-five patients were smokers (77.3%). The most common symptoms were fever in 91 patients (93.8%) and cough in 67 (68.1%). In five patients (5.2%) creatine phosphokinase concentrations were over 5 times their baseline values (in two over 100 times); four of these patients presented acute renal failure. Seroconversion was observed in 23/42 patients (54.8%). There were no statistically significant differences between the administration of erythromycin or clarithromycin in monotherapy, or in combination with rifampin. Nineteen patients (19.6%) presented acute renal failure and mechanical ventilation was necessary in 22 (22.7%). Twelve patients died (12.5%). Independent prognostic factors associated with death included respiratory rate > 30 breaths/min, urea > 60 mg/dL and PaO2 < 60 mmHg. A significant linear association was found between severity scale scores and the presence of complications or mortality. CONCLUSION: The Legionella urinary antigen test permits early diagnosis and treatment of this disease. The severity scale is an indicator of complications or death.