OBJECTIVE: The introduction of drug-eluting stents significantly reduced restenosis and reinterventions in patients undergoing percutaneous coronary interventions. This study compares results of Cypher stenting with those of surgical arterial revascularization in patients with diabetes mellitus. METHODS: From May 2002 through May 2005, 523 consecutive diabetic patients underwent myocardial revascularization: 244 underwent percutaneous coronary interventions incorporating drug-eluting stents, and 279 were treated surgically. All single-vessel patients in the surgical group were treated with the left internal thoracic artery, and most multivessel patients were treated with 2 internal thoracic arteries. After propensity score matching, 2 groups (93 patients each) were used to compare the 2 revascularization modalities. RESULTS: The number of coronary vessels treated per patient was higher in the surgical group (2.72 vs 1.75, P < .001). Follow-up ranged between 6 and 42 months (mean, 19 months). Overall mortality (early and late) was 3.2% in the surgical group and 2.2% in the Cypher group (P = .65). Two-year angina-free survival and reintervention-free survival (Kaplan-Meier) of the surgical group were 88% and 95%, respectively, compared with 47.8% (P = .001) and 83.6% (P = .01), respectively, in the percutaneous coronary intervention group. Cox proportional hazards modeling revealed assignment to the Cypher group to be the only predictor of reintervention (odds ratio, 3.86; 95% confidence interval, 1.25-11.9). Assignment to the Cypher group (hazard ratio, 5.92; 95% confidence interval, 2.96-11.87) and insulin treatment (hazard ratio, 2.06; 95% confidence interval, 1.06-4.02) were independent predictors of angina recurrence. CONCLUSIONS: The midterm clinical outcome of diabetic patients who underwent surgical arterial revascularization is better than that of patients undergoing percutaneous coronary intervention treated with drug-eluting stents.
OBJECTIVE: The introduction of drug-eluting stents significantly reduced restenosis and reinterventions in patients undergoing percutaneous coronary interventions. This study compares results of Cypher stenting with those of surgical arterial revascularization in patients with diabetes mellitus. METHODS: From May 2002 through May 2005, 523 consecutive diabeticpatients underwent myocardial revascularization: 244 underwent percutaneous coronary interventions incorporating drug-eluting stents, and 279 were treated surgically. All single-vessel patients in the surgical group were treated with the left internal thoracic artery, and most multivessel patients were treated with 2 internal thoracic arteries. After propensity score matching, 2 groups (93 patients each) were used to compare the 2 revascularization modalities. RESULTS: The number of coronary vessels treated per patient was higher in the surgical group (2.72 vs 1.75, P < .001). Follow-up ranged between 6 and 42 months (mean, 19 months). Overall mortality (early and late) was 3.2% in the surgical group and 2.2% in the Cypher group (P = .65). Two-year angina-free survival and reintervention-free survival (Kaplan-Meier) of the surgical group were 88% and 95%, respectively, compared with 47.8% (P = .001) and 83.6% (P = .01), respectively, in the percutaneous coronary intervention group. Cox proportional hazards modeling revealed assignment to the Cypher group to be the only predictor of reintervention (odds ratio, 3.86; 95% confidence interval, 1.25-11.9). Assignment to the Cypher group (hazard ratio, 5.92; 95% confidence interval, 2.96-11.87) and insulin treatment (hazard ratio, 2.06; 95% confidence interval, 1.06-4.02) were independent predictors of angina recurrence. CONCLUSIONS: The midterm clinical outcome of diabeticpatients who underwent surgical arterial revascularization is better than that of patients undergoing percutaneous coronary intervention treated with drug-eluting stents.