Risk of Ventricular Tachyarrhythmias after Cardiac Surgery
Abstract & Commentary
By John P. DiMarco, MD, PhD, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville. Dr. DiMarco does research for Medtronic, is a consultant for Medtronic, Novartis, and St. Jude, and is a speaker for Boston Scientific.
Source: El-Chami MF, et al. Ventricular arrhythmia after cardiac surgery: Incidence, predictors, and outcomes. J Am Coll Cardiol 2012;60:2664-2671.
This paper reviews the experience at a large cardiac surgery center with ventricular arrhythmias in the postoperative period. The authors searched the Society of Thoracic Surgeons’ (STS) adult cardiac database for all patients undergoing cardiac surgery at Emory University between January 2004 and July 2010. Patients who had a postoperative ventricular arrhythmia described as either sustained ventricular tachycardia (VT) or ventricular fibrillation (VF) were included. In addition, six other patients with frequent salvos of nonsustained VT who did not fulfill the 30-second criterion, but required intravenous anti-arrhythmic drugs, were also included. The charts of patients with postoperative ventricular arrhythmias were then reviewed to gain additional data concerning mechanism (VT vs VF) and timing (≤ 48 hours vs > than 48 hours after surgery), which were not included in the STS database. Mortality was obtained from hospital follow-up and the national Social Security Index.
There were 14,720 patients included in the analysis. These patients underwent either isolated coronary artery bypass grafting (60.2%), isolated valve surgery (22%), or combined coronary artery and valve surgery (9.6%). Half of all surgeries and almost 80% of coronary resvacularization only surgeries were performed off pump. Postoperative ventricular arrhythmias were reported in 248 patients (1.7%). Patients with postoperative ventricular arrhythmias were older, had lower mean ejection fractions, and were more likely to have had congestive heart failure and emergency surgery. By multivariate logistic regression analysis, five risk factors for postoperative arrhythmias were identified. These included older age, emergent surgery, lower ejection fraction, use of cardiopulmonary bypass, and peripheral vascular disease.
Postoperative ventricular arrhythmias were associated with increased long-term mortality. Mortality was highest in the first six postoperative months, with 6-month survival for patients with postoperative arrhythmias of 59.8% vs 93.8% for those who were ventricular arrhythmia free. Data about the mechanism and timing of the arrhythmias were only available in about one-half of the patients. VF had a higher early and late hazard for death. Ventricular arrhythmias that occurred more than 48 hours after surgery demonstrated worse early mortality risk compared to those that occurred within the first 48 hours. Interestingly, despite the occurrence of sustained ventricular arrhythmias, only 85 patients (34%) were treated with amiodarone and only 20 (8%) received an implantable cardioverter defibrillator before discharge. The authors conclude that risk factors for postoperative ventricular arrhythmias can be identified, but the overall risk for ventricular arrhythmias is low. Further data about the mechanism and timing of arrhythmia are needed for conclusions to be made about these findings.
This paper describes the pattern of postoperative ventricular arrhythmias in a very large group of patients undergoing cardiac surgery in a single institution. However, the primary tool used (the STS Adult Cardiac Database) is not ideally designed to help us understand these arrhythmias. It is striking to me that the incidence of postoperative ventricular arrhythmias has been relatively standard over many years. In 1984, our group at the University of Virginia reported a 1.4% incidence of unanticipated postoperative ventricular arrhythmias over the preceding 3 years.1 This is very similar to the 1.7% incidence listed here. We and other authors had noted that arrhythmias occurring on the first postoperative day when the patients were most likely to be unstable were often fatal, but if the patient survived the initial event, they often did well. Arrhythmias that occurred after the first postoperative day tended to be recurrent and more often associated with adverse long-term prognosis. Unfortunately, the STS database does not routinely collect the critical data about timing, mechanism, and the early postoperative setting that might help us better understand these arrhythmias. If we had a better understanding, we might be able to better define optimal long-term therapy.
1. Kron IL, et al. Unanticipated postoperative ventricular tachyarrhythmias. Ann Thorac Surg 1984;38:317-322.