OBJECTIVE: Thoracic empyema in cirrhotic patients is a challenging situation, and the clinical characteristics are rarely reported. The objective of this study was to report the clinical characteristics among this group and to evaluate whether thoracoscopic intervention would affect clinical outcomes. METHODS: Between 2001 and 2010, we retrospectively reviewed the clinical characteristics, bacteriologic studies, and treatment outcomes of 63 cirrhotic patients with thoracic empyema. A propensity-score based process, matched on age, sex, diabetes mellitus, malignancy, cause, and Child-Pugh classification (A, B, or C), was performed to equalize potential prognostic factors in thoracoscopy and nonthoracoscopy groups. The Kaplan-Meier curve and log-rank test were applied to compare the survival to discharge between the 2 matched groups. RESULTS: The median patient age was 61 years. Thirty-two patients (51%) underwent thoracoscopic management, and the remaining patients underwent thoracocentesis or tube thoracostomy. The median hospital stay was 28 days, and 19 patients (30%) had in-hospital mortality. Multivariate analysis showed that Child-Pugh C disease and positive blood cultures were risk factors for in-hospital mortality (P = .016 and .027, respectively), whereas thoracoscopic management may be favorable for survival (P = .041). The propensity score-matched analysis showed a significant reduction in intensive care unit stay (P = .044) in the thoracoscopy group. Kaplan-Meier survival analysis revealed a higher survival to discharge, favoring thoracoscopy over non-thoracoscopy treatment (P = .035). CONCLUSIONS: Management of thoracic empyema in cirrhotic patients is complicated and associated with a high mortality. With proper patient selection, thoracoscopic management is feasible and may provide a better chance of survival.
OBJECTIVE:Thoracic empyema in cirrhoticpatients is a challenging situation, and the clinical characteristics are rarely reported. The objective of this study was to report the clinical characteristics among this group and to evaluate whether thoracoscopic intervention would affect clinical outcomes. METHODS: Between 2001 and 2010, we retrospectively reviewed the clinical characteristics, bacteriologic studies, and treatment outcomes of 63 cirrhotic patients with thoracic empyema. A propensity-score based process, matched on age, sex, diabetes mellitus, malignancy, cause, and Child-Pugh classification (A, B, or C), was performed to equalize potential prognostic factors in thoracoscopy and nonthoracoscopy groups. The Kaplan-Meier curve and log-rank test were applied to compare the survival to discharge between the 2 matched groups. RESULTS: The median patient age was 61 years. Thirty-two patients (51%) underwent thoracoscopic management, and the remaining patients underwent thoracocentesis or tube thoracostomy. The median hospital stay was 28 days, and 19 patients (30%) had in-hospital mortality. Multivariate analysis showed that Child-Pugh C disease and positive blood cultures were risk factors for in-hospital mortality (P = .016 and .027, respectively), whereas thoracoscopic management may be favorable for survival (P = .041). The propensity score-matched analysis showed a significant reduction in intensive care unit stay (P = .044) in the thoracoscopy group. Kaplan-Meier survival analysis revealed a higher survival to discharge, favoring thoracoscopy over non-thoracoscopy treatment (P = .035). CONCLUSIONS: Management of thoracic empyema in cirrhoticpatients is complicated and associated with a high mortality. With proper patient selection, thoracoscopic management is feasible and may provide a better chance of survival.