Reverse Remodeling and the Risk of Ventricular Tachyarrhythmias in the MADIT-CRT (Multicenter Automatic Defibrillator Implantation Trial–Cardiac Resynchronization Therapy)
Is the cumulative incidence of ventricular tachyarrhythmias (VTAs) different between responders and nonresponders with cardiac resynchronization-defibrillator (CRT-D) therapy devices?
This was a substudy of the MADIT-CRT trial, which randomized patients with cardiomyopathy (ejection fraction <30%) and a QRS >130 msec on electrocardiogram to CRT-D versus isolated implantable cardioverter-defibrillators (ICDs). Echocardiograms were obtained at baseline and 12 months following device implant. A CRT-D “high responder” was defined as a patient with a ≥25% reduction in left ventricular end-systolic volume (LVESV) after therapy; all others were termed “low responders.” The primary endpoint of interest was the first occurrence of VTA event (including ventricular tachycardia, fibrillation, and flutter) in the responders versus nonresponders 1 year beyond device implant. Hazard ratios (HRs) for outcomes were derived using Cox proportional hazards modeling.
The cohort was comprised of 1,372 patients. After 12 months of follow-up, 529 (71%) of 749 CRT-D patients were high responders. Of the 623 ICD only patients, 30 (4.8%) were high responders. High responders in the CRT-D group were more likely than low CRT-D responders to be female, with a nonischemic etiology for their cardiomyopathy, and a longer QRS interval. While there were no differences in baseline LV volumes on echocardiogram, high responders had greater reductions in LVESV and end-diastolic volumes (LVEDV) than low CRT-D responders at follow-up. Likewise, low responders had greater LVESV and LVEDV reductions than ICD-only patients. There were 184 (13%) appropriate ICD therapies at a median 1.2 years following the 1-year post-enrollment echocardiogram. There were 55 (4%) deaths. At 2 years following response assessment, the probability of a VTA was 12% in CRT-D responders, 21% in ICD-only patients, and 28% in low CRT-D responders (p < 0.001). Similar trends were noted for the combined occurrence of VTA/death. On multivariable analysis, the risk of VTA was not significantly higher among CRT-D nonresponders compared with ICD-only patients (HR, 1.26; p = 0.20), but high responders had a 66% lower likelihood of VTA than the two aforementioned groups combined. Every 10% reduction in LVESV led to a 20% adjusted lower risk of a VTA.
The authors concluded that reverse remodeling is associated with fewer VTAs.
While CRT has been clearly shown to improve survival and functional capacity in heart failure, the association between CRT and VTAs is less concrete. This MADIT-CRT subanalysis provides strong support for benefits of reverse remodeling from CRT in reducing VTAs. A concern raised by this study is the trend toward increased events in those who do not respond to CRT-D compared with ICD alone. Patients with New York Heart Association class III/IV heart failure within 90 days were excluded, but that does not rule out the possibility that the CRT-D nonresponders were just “sicker end-stage heart failure patients” at baseline than either the responders or the ICD-only patients. Providing preimplant estimated mortality risk based on Seattle Heart Failure Model score may have been helpful. Of greater concern is the possibility that failure to respond to CRT-D could correlate with poor LV lead placement or other unknown CRT-specific factors, actually increasing events through adverse modulation of electrical activity. While only trends are noted, further investigation is warranted. It would be easy to turn the lead “off” if no response is seen after 1 year. Finally, lack of CRT-D response should trigger consideration for referral to a heart failure/transplant specialist.
Keywords: Tachycardia, Ventricular, Follow-Up Studies, Cardiomyopathies, Heart Conduction System, Heart Failure, Electrocardiography, New York, Defibrillators, Implantable, Cardiac Resynchronization Therapy
< Back to Listings