Systematic review and meta‐analysis of antibiotic prophylaxis to prevent infections from chest drains in blunt and penetrating thoracic injuries

2012 
Background: No consensus exists as to whether antibiotic prophylaxis in tube thoracostomy as primary treatment for traumatic chest injuries reduces the incidence of surgical-site and pleural cavity infections. Methods: A systematic literature search was performed according to PRISMA guidelines to identify randomized clinical trials on antibiotic prophylaxis in tube thoracostomy for traumatic chest injuries. Data were extracted by two reviewers using piloted forms. Mantel–Haenszel pooled odds ratios (ORs) were calculated with 95 per cent confidence intervals (c.i.). Results: Eleven articles were included, encompassing 1241 chest drains in 1234 patients. Most patients (84·7 per cent) were men, and a penetrating injury mechanism was most common (856, 69·4 per cent). A favourable effect of antibiotic prophylaxis on the incidence of pulmonary infection was found, with an OR for the overall infectious complication rate of 0·24 (95 per cent c.i. 0·12 to 0·49). Patients who received antibiotic prophylaxis had an almost three times lower risk of empyema than those who did not receive antibiotic treatment (OR 0·32, 0·17 to 0·61). A subgroup analysis in patients with penetrating chest injuries showed that antibiotic prophylaxis in these patients reduced the risk of infection after tube thoracostomy (OR 0·28, 0·14 to 0·57), whereas in a relatively small blunt trauma subgroup no effect of antibiotic prophylaxis after blunt thoracic injury was found. Conclusion: Infectious complications are less likely to develop when antibiotic prophylaxis is administered to patients with thoracic injuries requiring chest drains after penetrating injury. Copyright © 2011 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    28
    References
    48
    Citations
    NaN
    KQI
    []