A retrospective analysis of the management of parapneumonic empyemas in a county teaching facility from 1992 to 2004.

OBJECTIVES: To characterize how patients with empyemas are managed initially at our facility and to determine how "less aggressive" treatments (eg, no drainage, repeat thoracentesis, or tube thoracostomy) affect short-term outcomes (ie, inpatient mortality and the need for a second intervention) compared to "more aggressive" treatments (eg, intrapleural fibrinolytic agents, video-assisted thoracoscopic surgery, or other surgery). We will also assess whether earlier diagnosis, earlier antibiotic treatment, fewer patient comorbidities, and consulting appropriate services improve mortality.
DESIGN: Retrospective chart analysis.
SETTING: County teaching hospital in Los Angeles, CA. PATIENTS: Seventy-two adult inpatients with parapneumonic empyemas.
INTERVENTIONS: Mortality and the need for second intervention rates were calculated and compared with data published in the 2000 American College of Chest Physicians consensus statement on the management of parapneumonic effusions using the Fisher exact test. Comparisons were made between empyema survivors and nonsurvivors using t tests and chi(2) tests.
RESULTS: All 72 patients were managed with less aggressive initial treatments. There were no differences in mortality when our patients were compared to the less aggressive group from the literature (6% vs 9%, respectively; p = 0.40; relative risk, 0.6; 95% confidence interval [CI], 0.23 to 1.62) or the more aggressive group from the literature (6% vs 3%, respectively; p = 0.29; relative risk, 1.8; 95% CI, 0.64 to 5.23). There was no difference between the second intervention rate of our patients and that of the less aggressive group from the literature (47% vs 43%, respectively; p = 0.47; relative risk, 1.1; 95% CI, 0.86 to 1.42), although there was a difference when compared to the more aggressive group (47% vs 11%, respectively; p < 0.0001; relative risk, 4.5; 95% CI, 3.20 to 6.31). There were no statistically significant differences in time of diagnosis, the timing of antibiotic treatment, the number of patient comorbidities, or the number of services consulted when survivors and nonsurvivors from the study were compared.
CONCLUSIONS: Patients with empyemas at our hospital are treated with less aggressive initial treatments and have a higher second intervention rate when compared to patients described in the literature who were initially managed with more aggressive treatments.