Effect of negative pressure wound therapy followed by tissue flaps for deep sternal wound infection after cardiovascular surgery: propensity score matching analysis.

OBJECTIVES Deep sternal wound infection (DSWI) after cardiovascular surgery via median sternotomy remains a severe complication associated with a drastic decrease in the quality of life. We assessed the risk factors for in-hospital death caused by DSWI and the available treatments for DSWI. METHODS Between January 1991 and August 2015, we retrospectively reviewed 73 patients (51 males and 22 females, mean age 67.5 ± 10.3 years) who developed DSWI after cardiovascular surgery via median sternotomy. Pathogenic bacteria mainly comprised methicillin-resistant Staphylococcus aureus (MRSA) (49.3%). Fifteen patients (20.5%) died in hospital with DSWI. Treatment of DSWI consisted of open daily irrigation (up to 2006) or negative pressure wound therapy (NPWT) (2007 onwards), followed by primary closure or reconstruction of tissue flaps. We assessed the risk factors for in-hospital mortality from DSWI by comparing data from the 15 patients who died and the 58 survivors using propensity score matching analysis of the treatments used for DSWI. RESULTS Univariate analysis identified age, use of intra-aortic balloon pumping, prolonged mechanical ventilation, tracheotomy, prolonged intensive care unit stay, postoperative low output syndrome, postoperative myocardial infarction, postoperative renal failure, postoperative use of haemodialysis, postoperative pneumonia, postoperative cerebral disorder, MRSA infection, NPWT and tissue flaps as being associated with in-hospital mortality (P < 0.05). Multivariate analysis identified NPWT (odds ratio, 0.062; 95% confidence interval, 0.004-0.897; P = 0.041) and tissue flaps (odds ratio, 0.022; 95% confidence interval, 0.000-0.960; P = 0.048) as independently associated with reduced in-hospital mortality after DSWI. On comparing 22 patients receiving NPWT with 22 not on NPWT using propensity score matching, patients on NPWT had significantly lower in-hospital mortality than those without NPWT (NPWT vs non-NPWT, 5 vs 36%, P = 0.021). In DSWI infected by MRSA, NPWT significantly reduced the in-hospital mortality caused by DSWI (NPWT vs non-NPWT, 0 vs 52%, P = 0.003). CONCLUSIONS NPWT and tissue flaps may be favourable factors associated with reduced in-hospital mortality attributable to DSWI. NPWT as a bridge therapy to tissue flaps may play a major role in treating DSWI and improve the prognosis for patients with MRSA-infected DSWI.