The changing pattern of reoperative coronary surgery: trends in 1230 consecutive reoperations.

OBJECTIVE: We noted an increasing risk profile of patients undergoing reoperative coronary surgery. We evaluated the risk compared with primary procedures, our results over a 16-year span, and the predictors of hospital outcomes after redo surgery.
METHODS: We analyzed 20,614 patients undergoing isolated coronary surgery at our institution from 1982 to 1997. Of these, 1230 (6.0%) were undergoing reoperation. Independent predictors of outcomes were identified by multivariable regression.
RESULTS: The prevalence of reoperation peaked in 1994 at 8.2%. Patients undergoing reoperation were more likely to be male, to have left ventricular dysfunction and worse symptoms, and to require an urgent operation than patients undergoing a primary operation (P <.0001). Perioperative myocardial infarctions (3.7% vs 7.4%), low-output syndrome (9.0% vs 24.0%), and death (2.4% vs 6.8%) were more common in patients undergoing reoperation (all P <.0001). Over the years, the risk profile of patients undergoing reoperation increased. Age, left ventricular dysfunction, severity of symptoms, extent of coronary artery disease, left main stenosis, and requirement for urgent or emergency operations increased with time (P <.05). However, mortality, myocardial infarction, and low-output syndrome have remained constant. The independent predictors of mortality after reoperative surgery were increased age, greater Canadian Cardiovascular Society symptom class, earlier year of operation, and greater left ventricular dysfunction. After 1990, analysis of an expanded data set also identified peripheral vascular disease and failure to use retrograde cardioplegia as predictors of mortality.
CONCLUSIONS: Improving results of reoperative surgery have been offset by an increasing patient risk profile. Meticulous operative technique and retrograde cardioplegia may permit good results in these high-risk patients.