Factors associated with recurrent bacterial empyema thoracis

• Main Authors: Tantraworasin A., Thepbunchonchai A., Siwachat S., Ruengorn C., Khunyotying D., Kaufman A., Taioli E., Saeteng S.
• Format: Journal
• Published: 2017
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85016404728&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/41201
• Institution: Chiang Mai University

© 2017. Background: Empyema thoracis (ET) is one of the common thoracic diseases frequently found in developing countries. In the past, only a few studies have analyzed recurrent ET, and none had clearly identified the prognostic factors for recurrence. The aim of this study is to identify the prognostic factors of recurrent ET in Northern Thailand. Methods: A retrospective cohort study was conducted. All patients diagnosed with Stage II and III ET at Maharaj Nakorn Chiang Mai Hospital (a tertiary-care hospital in northern Thailand) between January 1, 2007 and November 31, 2012 were enrolled in this study. All clinical data were extracted from the medical recording system. The primary outcome was recurrent disease. Multivariable Cox's proportion hazard model was used to identify the independent prognostic factors for recurrence. Results: There were 382 patients enrolled in this study, and 34 patients (8.9%) had recurrent disease. The most common gram-positive and gram-negative pathogens found were Staphylococcus aureus and Acinetobacter baumannii, respectively. Three independent prognostic factors for recurrent disease were nonsepsis status prior to the surgery [hazard ratio (HR) = 12.3; 95% confidence interval (CI), 4.25-35.43], nonperforming decortication (HR = 5.4; 95% CI, 1.82-15.92), and persistent pleural spaces (HR = 4.1; 95% CI, 1.93-8.68). Conclusions: Clinical characteristics, surgical procedure, and persistent pleural spaces were independent prognostic factors for ET recurrence in this study. Decortication and early thoracoplasty in patients who had persistent pleural space should be considered. Large cohort studies are warranted to support these findings.