Effect of Elapsed Time From Coronary Revascularization to Implantation of a Cardioverter Defibrillator on Long-Term Survival in the MADIT-II Trial


  • MADIT-II was supported by a research grant from Boston Scientific Corp., St. Paul, Minnesota, to the University of Rochester School of Medicine and Dentistry. This long-term study was not funded by Boston Scientific Corp.

  • This research was carried out while Dr. Alon Barsheshet was a Mirowski-Moss Career Development Awardee at the University of Rochester Medical Center, Rochester, NY.

  • Dr. Moss reports honoraria for Grand Round lectures on this topic. He and Drs. Zareba, Klein, and Huang were participants on the MADIT research grants. Other authors: No disclosures.

Alon Barsheshet, M.D., Heart Research Follow-up Program, Box 653, University of Rochester Medical Center, Rochester, NY 14642, USA. Fax: 585-273-5283; E-mail: alon.barsheshet@heart.rochester.edu


Coronary Revascularization and Long-Term Mortality in MADIT-II. Introduction: Coronary revascularization (CR) may reduce arrhythmia risk and improve long-term outcome in patients with left ventricular dysfunction. This study was designed to evaluate the effect of elapsed time from CR on long-term mortality and arrhythmic risk among patients who receive an implantable cardioverter defibrillator (ICD).

Methods and Results: We evaluated the risk of 8-year mortality by elapsed time from CR to ICD implantation (categorized as: no CR; recent CR [<2 years]; or nonrecent CR [≥2 years], and assessed as a continuous measure) among 720 ICD recipients enrolled in the Multicenter Automatic Defibrillator Trial-II. At 8years of follow-up, patients who did not undergo CR and those who underwent nonrecent CR had significantly higher mortality rates than patients who underwent recent CR (54%, 54%, and 36%, respectively; P < 0.001). Multivariate analysis demonstrated that no- and nonrecent CR were associated with respective 48% (P = 0.022) and 67% (P < 0.001) increases in mortality risk compared with recent CR. Assessment of time from CR as a continuous measure showed that every year elapsed from CR was associated with an adjusted 6% increase in 8-year mortality (P < 0.001), and in respective 6% (P < 0.001) and 6% (P = 0.003) increased risk for in-trial appropriate ICD therapy of ventricular tachyarrhythmias and appropriate ICD shocks.

Conclusions: We observed a direct relationship between elapsed time from CR and long-term mortality following ICD implantation. The favorable long-term effect on outcome of recent CR may be related to a time-dependent effect of CR on ventricular arrhythmic burden and the need for appropriate ICD shocks. (J Cardiovasc Electrophysiol, Vol. pp. 1-6)