Abstract Rationale Managing thoracic empyema in the current era involves clinical decisions around which patients will most benefit from a medical, potentially including intrapleural fibrinolysis, versus surgical approach. Although heterogeneous practices across health systems have been described, evidence remains limited by small, single center cohorts. Therefore, our multicenter study characterizes patient, clinical, and microbiological differences in the management of patients with culture-positive empyema. Method Using a federated approach across five U.S. health systems participating in the Common Longitudinal ICU Format (CLIF), we conducted a retrospective cohort study of hospitalized adults with culture-positive empyema (defined as pleural fluid culture positivity) between 2018-2024. Patients were excluded if they received less than five days of antibiotics or had tuberculous or mycobacterial empyema. We categorized patients into: (1) conservative medical management with antibiotics alone, (2) aggressive medical management (antibiotics with intrapleural fibrinolysis), or (3) surgical management (thoracoscopy or video-assisted thoracic surgery, irrespective of intrapleural fibrinolysis use). The primary outcome was the proportion of patients treated under the three management strategies. Secondary outcomes included differences in patient comorbidities, culture results, antibiotic administration, alteplase and dornase instillations, and clinical outcomes. Result Of 1,318 patient hospitalizations with culture-positive empyema (mean age 61, 35% female, 70.4% non-Hispanic white), management strategies included conservative medical management in 591 (45%), aggressive medical management in 494 (38%) and surgical management in 233 (18%) hospitalizations (Table 1). Almost one-third of cultures were polymicrobial, and almost one-fifth grew fungi. Vancomycin, piperacillin-tazobactam, cefepime, metronidazole, and meropenem were the most commonly prescribed antibiotics. The proportions of patients in each treatment strategy varied across sites, ranging from 29.0-59.1% for conservative medical, 25.6-52.0% for aggressive medical, and 4.9-34.0% for surgical management. Patients treated surgically had lower Charlson Comorbidity Index and Elixhauser scores (p 0.01) and similar length of stay (26.6 vs 25.6 vs 29.6 days, p = 0.14). Among surgical treated patients, 93 (40%) patients received alteplase and 32 (13.7%) patients received dornase alpha. The unadjusted in-hospital mortality rate was 14.9% for conservative medical management, 7.1% for aggressive medical management, and 5.6% for surgical management. Conclusion This large, multicenter retrospective cohort study using CLIF data emphasizes the heterogeneous approaches to managing thoracic empyema across U.S. health systems. The novel use of CLIF enables robust, real-world insights to inform future standardization efforts and outcome studies. This abstract is funded by: none
Sura et al. (Fri,) studied this question.