BACKGROUND Deep sternal wound infection (DSWI) is a serious postoperative complication of cardiac surgery. In this study we investigated the incidence of DSWI and effect of re-exploration for bleeding on DSWI mortality. METHODS We reviewed 73,700 cases registered in the Japan Adult Cardiovascular Surgery Database (JACVSD) during the period from 2004 to 2009 and divided them into five groups: 26,597 of isolated coronary artery bypass graft (CABG) cases, 23,136 valvular surgery cases, 17,441 thoracic aortic surgery cases, 4,726 valvular surgery plus CABG cases, and 1,800 thoracic aortic surgery plus CABG cases. We calculated the overall incidence of postoperative DSWI, incidence of postoperative DSWI according to operative procedure, 30-day mortality and operative mortality of postoperative DSWI cases according to operative procedure, 30-day mortality and operative mortality of postoperative DSWI according to whether re-exploration for bleeding, and the intervals between the operation and deaths according to whether re-exploration for bleeding were investigated. Operative mortality is defined as in-hospital or 30-day mortality. Risk factors for DSWI were also examined. RESULTS The overall incidence of postoperative DSWI was 1.8%. The incidence of postoperative DSWI was 1.8% after isolated CABG, 1.3% after valve surgery, 2.8% after valve surgery plus CABG, 1.9% after thoracic aortic surgery, and 3.4% after thoracic aortic surgery plus CABG. The 30-day and operative mortality in patients with DSWI was higher after more complicated operative procedures. The incidence of re-exploration for bleeding in DSWI cases was 11.1%. The overall 30-day/operative mortality after DSWI with re-exploration for bleeding was 23.0%/48.0%, and it was significantly higher than in the absence of re-exploration for bleeding (8.1%/22.0%). The difference between the intervals between the operation and death according to whether re-exploration for bleeding had been performed was not significant. Age and cardiogenic shock were significant risk factors related to re-exploration for bleeding, and diabetes control was a significant risk factor related to DSWI for all surgical groups. Previous CABG was a significant risk factor related to both re-exploration for bleeding and DSWI for all surgical groups. CONCLUSIONS The incidence of DSWI after cardiac surgery according to the data entered in the JACVSD registry during the period from 2004 to 2009 was 1.8%, and more complicated procedures were followed by higher incidence and mortality. When re-exploration for bleeding was performed, mortality was significantly higher than when it was not performed. Prevention of DSWI and establishment of an effective appropriate treatment for DSWI may improve the outcome of cardiac surgery.