Log in to MyACC
Cardiac Resynchronization Therapy Is More Effective in Women Than Men: The MADIT-CRT (Multicenter Automatic Defibrillator Implantation Trial With Cardiac Resynchronization Therapy) Trial
Study Questions:
In patients with mild systolic heart failure (HF), does sex impact the response to cardiac resynchronization therapy with defibrillator (CRT-D)?
Methods:
Patients enrolled into the MADIT-CRT trial were dichotomized by sex (n = 453 female and 1,367 male) and the impact of CRT-D versus defibrillators (ICD) alone were compared. The primary outcome of interest was death or HF hospitalization (whichever came first) with secondary outcomes of HF admission and death at any time (assessed individually) in the CRT-D and ICD groups. Per MADIT-CRT entry criteria, all patients had New York Heart Association class I or II HF, a left ventricular ejection fraction <30%, and were in sinus rhythm with a QRS ≥130 ms.
Results:
At baseline, females were more likely than males to have nonischemic etiologies for HF, a left bundle branch block (LBBB), and less renal dysfunction. The primary endpoint occurred in 376 patients over 2.4 years of follow-up, of which 11% (n = 29 of 275) were females with CRT-D, 29% (n = 51 of 178) were females with an ICD, 20% (n = 159 of 814) were males with CRT-D, and 25% (n = 137 of 553) were males with an ICD (Kaplan-Meier p < 0.001). The hazard ratio (HR) for death/HF in females with CRT-D versus ICD was 0.31 (95% confidence interval [CI], 0.19-0.50 ) compared with 0.72 (95% CI, 0.57-0.92) in men. The HR for HF admission alone in females with CRT-D versus ICD was 0.30 (95% CI, 0.18-0.50) compared with 0.65 (95% CI, 0.50-0.84) in men. Finally, females receiving CRT-D had a significantly lower HR for death at any time (HR, 0.28; 95% CI, 0.10-0.79), whereas CRT-D was not significantly better than ICD in men (HR, 1.05; 95% CI, 0.70-1.57). All three analyses yielded a significant difference in outcomes between females and males (all interactions, p < 0.03). The interaction for sex was also significant when assessing outcomes for those with LBBB and QRS ≥150 ms.
Conclusions:
Outcomes following CRT were better in females compared with males enrolled into the MADIT-CRT trial.
Perspective:
Females are often under-represented in clinical trials, but results are extrapolated to them. In this carefully selected MADIT-CRT cohort, there appeared to be a difference in the magnitude of benefit received from CRT-D between the sexes. It is possible that the differences are related to patient selection and the fact that the females in the study were truly different at baseline from males. No information is provided on medical therapies administered at follow-up either. For most of the subanalyses, confidence intervals were also very wide due to limited power. In fact, there was a nonsignificant trend toward harm with CRT-D in women without an LBBB. Thus, despite the high-quality statistical techniques employed, the only way to truly parse this out is to do a prospective, matched study (LBBB, HF etiology) using gender stratification.
Keywords: Follow-Up Studies, Cardiology, Bundle-Branch Block, Stroke Volume, New York, Heart Failure, Systolic, Hospitalization, Defibrillators, Implantable, Cardiac Resynchronization Therapy
< Back to Listings
