Mediastinitis after aorto-coronary bypass surgery

OBJECTIVES: To identify risk factors in 60 cases of mediastinitis amongst 2512 patients (2.3%) subjected to isolated coronary bypass surgery from March 1988 through December 1995, treated by a closed irrigation/drainage system. PATIENTS AND METHODS: The mean age of the 60 patients was 56.9 +/- 6.8 years (45-81 years) and 55 (91.6%) were male. Early mediastinal reexploration was performed in all cases immediately after the diagnosis of mediastinitis, with debridement of necrosed tissues, followed by implantation of a closed-circuit irrigation system of the mediastinum constituted by irrigation catheter and drain, closure of the sternum and skin, and specific systemic antibiotic therapy. The mean interval between the original surgery and reexploration was 9.4 days (range 6-14 days). No patient required more extensive procedures, namely omental or muscular flaps. Twenty potential risk factors in patients with mediastinitis, including diabetes mellitus, obesity, coexistence of peripheral vascular disease, decreased LV function, use of inotropes, mediastinal blood drainage and utilization of double IMA, were compared with the group without mediastinitis. RESULTS: Mean cardiopulmonary bypass time was 74.1 +/- 8.1 min, anesthetic time 3.5 +/- 0.8 h and postoperative mechanical ventilation 18 +/- 3 h. A total of 23 patients (38.3%) received one IMA and 35 (58.3%) two IMAs. In the postoperative period, 7 of the 60 patients (11.6%) had required inotropes because of low output. Mediastinal blood loss was 1112cc +/- 452cc and 9 patients (15%) were transfused. Cultures were positive in 40 cases (66.6%) and the most frequent infecting agent was the Staph. epidermidis in 25 cases (62.5%), followed by Candida albicans and Enterobacter and Serratia species (7.5% each); 1 patient (1.7%) died and 9 (15%) had renal failure. The irrigation/drainage was maintained for a mean of 9.1 days (5-83 days). Patients with mediastinitis had a significantly higher prevalence of diabetes (41.6% vs. 18.8%; P < 0.01), obesity (48.3% vs. 15.2%; P < 0.001), peripheral vascular disease (11.6% vs. 4.0%; P < 0.05), but a lower incidence of poor LV function (18.3% vs. 32.7%; P < 0.05). A double IMA was used more frequently in patients who had mediastinitis (58.3% vs. 23.5%; P < 0.001) CONCLUSIONS: Diabetes mellitus, obesity, co-existence of peripheral vascular disease and use of double IMA are risk factors for mediastinitis after coronary artery surgery. The efficacy of the closed method of treatment with a mediastinal irrigation/drainage system was increased with early diagnosis and reintervention