Objectives: The prevalence of ventricular dysfunction in patients undergoing coronary operations, as well as the prevalence of other risk factors in these patients, has been increasing. We identified the predictors of mortality and morbidity in patients with ventricular dysfunction to permit more accurate evaluation of risk and to direct future strategies to improve outcomes.
Methods: Demographic, intraoperative, and outcome data were collected prospectively on 20,614 patients undergoing isolated coronary operations at our institution from 1982-1997. Multivariable regression analyses were used to identify the independent predictors of mortality and low-output syndrome.
Results: Moderate ventricular dysfunction (ejection fraction, 20%-40%) was noted in 4107 (19.9%) patients, and severe dysfunction (ejection fraction, <20%) was noted in 680 (3.3%) patients. Patients with worse ventricular function had an increasing prevalence of other risk factors with time. Mortality decreased between the 1982-1986 and 1987-1991 cohorts but did not decrease further. Low-output syndrome was less common in the 1992-1997 cohort than in previous years. The predictors of mortality were ventricular dysfunction, age, reoperation, year of operation, urgency, female sex, and left main stenosis. Low-output syndrome was predicted by ventricular dysfunction, reoperation, year of operation, female sex, urgency, extensive coronary disease, age, left main stenosis, and symptom class.
Conclusions: Despite the increasing prevalence and risk profile of patients with ventricular dysfunction, mortality rates and incidence of low-output syndrome declined with time. Patients with severe dysfunction were at greatest risk when facing reoperation or urgent operation. Earlier intervention and more aggressive preoperative optimization may improve outcomes in these high-risk patients.