Treatment and outcome of 104 hospitalized patients with legionnaires' disease

Background: Large outbreaks of Legionella pneumonia are rare, but when they occur provide an opportunity to assess predictors of mortality and efficacy of drug therapy. Although erythromycin has been the treatment of choice for many years, newer antimicrobials with increased activity against Legionella are available. A large outbreak of legionnaires' disease associated with the Melbourne Aquarium occurred in April 2000. Aim: To describe the patterns and impact of Legionella therapy, and predictors of outcome in a large group of hospitalized patients with legionnaires' disease. Methods: A 6-month retrospective audit of hospitalized patients with proven legionnaires' disease around the time of the Melbourne Aquarium outbreak was conducted. Statistical analysis was performed using SAS version 8.0 (SAS Institute Inc., NC, USA). Results: Data were obtained on 104 patients (71 aquarium related, 33 not related). There were six deaths (mortality rate 5.8%), three of which were attributable directly to progressive legionnaires' disease. The major predictors of death were pre-existing cardiac failure (P = 0.0035) and renal disease (P = 0.026). Erythro-mycin is still the most commonly used antibiotic (80% received i.v. erythromycin) with clinicians prescribing more than one active Legionella drug in the majority of cases (76%). Choice of initial antibiotic therapy did not statistically affect outcome as measured by death, length of hospital stay or time to defervescence, although there was a trend towards improved survival with i.v. erythromycin (P = 0.063). Intravenous erythromycin was associated with a 19% rate of phlebitis, whereas side-effects from other antibiotics were uncommon. Conclusion: he most commonly used Legionella therapy in Australia remains erythromycin. This continues to be an effective agent, however, side-effects are common.