Background: The internal mammary artery is used as coronary artery graft conduit because of its superior patency. According to some authors, the bilateral IMA can increase perioperative morbidity. The aim of this study was to determine the risk factors increasing perioperative mortality and morbidity in the use of bilateral IMA.
Methods: We analyzed the data of 474 patients operated consecutively with the use of bilateral IMA between January 1987 and December 1995 at the Department of Cardiac Surgery of the Varese Hospital. The univariate analysis was done on 17 ordinal variables using a "Fisher exact test" and on 4 continuous variables by "pooled-variance t-test" to investigate risk factors for mortality, mediastinitis, superficial wound infection and aseptic dehiscence of the sternum; a p-value lower than 0.1 was used as cut-off point to introduce the variables into a stepwise multiple logistic regression analysis.
Results: From the univariate analysis are: postoperative low-output syndrome (p = 0.01), LVEF (p = 0.02) and number of grafts (p = 0.04) are correlated to hospital mortality (1.5%); obesity (p < 0.001) and peripheral arteriopathy (p = 0.009) are correlated to postoperative mediastinitis (5%); obesity (p < 0.001), peripheral arteriopathy (p = 0.009), surgeon (p = 0.001), year of operation (p < 0.001), reoperation for bleeding (p = 0.004) and length of extracorporeal circulation (p = 0.02) are correlated to superficial wound infection (7%); obesity (p = 0.002) and COPD (p = 0.05) are correlated to aseptic dehiscence of the sternum (2%). The multivariate analysis identified low LVEF as the only independent risk factor for hospital mortality (p = 0.03), whereas obesity (p = 0.01) and peripheral vasculopathy (p = 0.03) proved to be correlated to postoperative mediastinitis; obesity (p < 0.001), year of the operation (p < 0.001), low LVEF (p = 0.007) and reoperation for bleeding (p = 0.01) were correlated to superficial infection of the wound and obesity turned out to be the only risk factor for aseptic dehiscence for the sternum (p = 0.003). The infection of the wound did not increase mortality, but it did increase the mean postoperative length of hospital stay (6 days for patients free of any complications of the wound versus 29.7 days for patients with complications of the wound).
Conclusions: In patients with bilateral mammary grafts, obesity is the main risk factor for complications of the wound and this event greatly increases the length of the patient's hospital stay. Consequently, we suggest that bilateral mammary artery grafts be used carefully in this subset of patients.