ABSTRACT
BACKGROUND: Diabetes is commonly regarded as a risk factor for mortality and morbidity after coronary artery bypass surgery. METHODS: Between April 1997 and December 2002, 6,033 consecutive patients underwent isolated coronary artery bypass surgery. Eight hundred and fourteen (13.5%) patients had diabetes (530 oral-dependent, 284 insulin-dependent). Patients with diet-controlled diabetes were classified as nondiabetics. Deaths occurring over time were described using Kaplan-Meier techniques. To control for differences in patient characteristics, we constructed a propensity score (for diabetes) and this was included along with the comparison variable in multivariate logistic regression and Cox proportional hazards analyses. RESULTS: In-hospital mortality was significantly higher for diabetic patients in the univariate analyses; however, this association disappeared after adjusting for the propensity score. Further analyses found that insulin-dependent diabetes was associated with an increased incidence of acute renal failure (adjusted odds ratio 4.15; p = 0.002), deep sternal wound infection (adjusted odds ratio 2.96; p = 0.039), and prolonged postoperative stay (adjusted odds ratio 1.60; p = 0.017). Oral-controlled diabetes was not associated with any of these outcomes. Four hundred and ninety-eight (8.3%) deaths occurred during the study follow-up. After adjusting for patient characteristics, the adjusted hazard ratio of midterm mortality for diabetes was 1.35; p = 0.013. CONCLUSIONS: Insulin-dependent diabetes has a significant impact on in-hospital morbidity. Although diabetic patients are not at increased risk of in-hospital mortality, longevity is significantly decreased during a five-year follow-up period.
