Bilateral internal thoracic artery grafting: midterm results of composite versus in situ crossover graft.

Background: Two common techniques of bilateral internal thoracic artery grafting are the composite T graft and in situ crossover graft. The superiority of one method over the other has not yet been established.
Methods: From April 1996 to July 1999, bilateral skeletonized internal thoracic arteries were used as T grafts (composite group, n = 649) and in situ grafts (cross group, n = 351) in 1,000 consecutive patients. In the cross group, in situ right internal thoracic artery was routed anterior to the aorta across the midline for grafting to the left anterior descending artery, and the left internal thoracic artery was used for the circumflex branches.
Results: The two groups had comparable preoperative risk profiles. Bypass time and aortic cross-clamping time were longer in the composite group (80 +/- 38 and 67 +/- 29 minutes versus 66 +/- 43 and 55 +/- 34 minutes, respectively). Number of anastomoses per patient was similar (3.1 versus 3.2). However, more sequential anastomoses were performed in the composite group (62% versus 53%), and the gastroepiploic artery was used more often in the cross group (30% versus 19%). Thirty-day mortality was 3.9% in the composite and 2.3% in the cross group (not significant). Occurrence of postoperative complications (sternal infection, myocardial infarction, cerebrovascular accident, and bleeding) was similar. Late follow-up (2 to 56 months) showed increased return of angina (6% versus 3.1%; p = 0.046) and decreased 4-year survival (Kaplan-Meier; 86% +/- 2.7% versus 92.4% +/- 1.5%; p = 0.07) in composite patients.
Conclusions: Early results of bilateral internal thoracic artery grafting with composite T graft are comparable with those of in situ grafts. However, increased angina return and decreased midterm survival led us to recommend in situ grafting whenever technically possible.