Abstract
Mediastinitis is treated with either vacuum-assisted closure (VAC) or traditional closed drainage (TCD) with irrigation. The aim of the study was to determine the effect of the two treatments on mortality and re-infection rate in a source population, using 21 314 consecutive patients undergoing isolated coronary artery bypass grafting (CABG) from January 1997 to October 2010. Median observation time was 2·9 years in the VAC group and 8·0 years in the TCD group. The epidemiological design was of an exposed (VAC, n = 64) versus non-exposed (TCD, n = 66) cohort with two endpoints: (1) mortality and (2) failure of sternal wound healing or re-infection. The crude effect of treatment technique versus endpoint was estimated by univariate analysis. Stratification analysis by the Mantel-Haenszel method was performed to quantify confounders and to pinpoint effect modifiers. Adjustment for confounders was performed using Cox regression analysis. Mediastinitis was diagnosed 6-105 (median 14) days after primary operation in the VAC group and 13 (5-29) days in the TCD group. There was no difference between groups in long-term survival. Failure of sternal wound healing or re-infection occurred less frequently in the VAC group (6%) than in the TCD group (21%; relative risk = 0·29, 95% CI = 0·06-0·88, P = 0·01). There are concerns for increase in right ventricle rupture in VAC compared with TCD. There was no difference in survival after VAC therapy and TCD therapy of post-CABG mediastinitis. Failure of sternal wound healing or re-infection was more common after TCD therapy.