Abstract 14698: Bilateral Internal Mammary Artery Grafting Prolongs Survival in Diabetic Patients: A Long-term Follow-up of Propensity Matched Cohorts
Background: The prevalence of diabetes is increasing at an unprecendented rate, affecting nearly 8% of the population. Previous studies have demonstrated a potential benefit for surgical over interventional revascularization in this group of patients. Similarly, studies have shown the superiority of bilateral internal mammary artery (BIMA) grafting over single internal mammary artery (SIMA) grafting in select populations. However, concerns regarding sternal wound infection (SWI) have discouraged use of BIMA grafting in diabetics. Therefore, we studied the long-term results of BIMA vs. SIMA grafting in a large population of diabetic patients in which BIMA grafting was broadly applied.
Methods: Between February 1972 and May 1994, 1107 consecutive diabetic patients underwent coronary artery bypass grafting (CABG) with either SIMA (n=646) or BIMA (n=461) grafting. Optimal matching using propensity scoring was used to create matched SIMA (n=414) and BIMA (n=414) cohorts. Cross-sectional follow-up (6 weeks−30.1 years, mean 8.9years) was 98.5% complete. Multivariable analyses were performed to determine correlates of operative (OM) and late mortality. Clinical functional status and SF-36 of late survivors were compared.
Results: There was no difference in OM, SWI or total complications between matched SIMA and BIMA groups (OM: 10/414; 4.6% vs. 13/414;3.1%, p=0.279; SWI 7/414;1.7% vs. 13/414;3.1%p=0.179; total complications 71/414, 17.1% vs. 71/414, 17.1%, p=1.000). Choice of conduit was not identified as a predictor of OM on logistic regression. Late survival was significantly prolonged with the use of BIMA grafting (median survival SIMA 9.8 years vs. BIMA 13.1 years; p=0.001). Use of BIMA was a found to be a predictor of late survival on Cox regression (p=0.021). At late follow-up of matched groups110 SIMA and 171 BIMA patients completed the SF−36. Physical health summary scores were comparable (42.5 vs. 43.6, p=0.474), while mental health summary scores favored SIMA patients (56.9 vs. 53.9; p=0.001).
Conclusions: BIMA, compared with SIMA grafting in propensity matched patients provides diabetics with prolonged survival and good quality of life without any increase in perioperative morbidity or mortality.
- © 2010 by American Heart Association, Inc.