Is mediastinitis a preventable complication? A 10-year review

Background. The incidence of mediastinitis after cardiac surgical intervention is reported to be between 0.15% and 5% and is a major cause of postoperative morbidity. A number of risk factors have been identified, most of which are not modifiable. It is our contention that this complication can be reduced to a minimum by the consistent application of good operative technique and postoperative management. Methods. We reviewed the records of all 9,771 patients who underwent cardiac surgical procedures between 1987 and 1997. All operations were performed using a common skin preparation, draping, and antibiotic prophylaxis. Cases of mediastinitis were defined according to Centers for Disease Control and Prevention criteria and were identified from three sources: medical records database, hospital infection control, and the Society of Thoracic Surgeons database. Risk factors were assessed using χ2 and Fisher’s exact tests. Results. Of 24 patients identified as having deep sternal wound infection (incidence, 0.25%), 2 died (mortality rate, 8.3%), 18 required reoperation (75%), and only 4 needed pectoral muscle flaps. Statistical analysis revealed only the presence of chronic obstructive pulmonary disease as a significant risk factor (p< 0.01). Other factors, including diabetes, renal failure, smoking, sex, age, reoperation, morbid obesity, and steroid use, were not significant. The use of internal mammary arteries (single or bilateral) was not associated with mediastinitis. Postoperative complications, including prolonged ventilation, inotropic support, and the need for blood products, were not significant risk factors. The patients who developed mediastinitis were more likely to be readmitted to the hospital (p< 0.005) and more likely to require reoperation (p< 0.005). Conclusions. In a large patient series we found a low incidence of mediastinitis (0.25%) and an even lower incidence of required reoperation (0.19%). Except for the use of bone wax and the use of bilateral mammary arteries in diabetic patients, none of the previously identified risk factors are modifiable. We believe that with strict adherence to perioperative aseptic technique, attention to hemostasis, and precise sternal closure, a very low incidence of mediastinitis can be achieved.