Long-term mortality with cardiac resynchronization therapy in the Cardiac Resynchronization-Heart Failure (CARE-HF) trial

Eur J Heart Fail. 2012 Jun;14(6):628-34. doi: 10.1093/eurjhf/hfs055. Epub 2012 May 2.


Aims: The Cardiac Resynchronization-Heart Failure (CARE-HF) study helped establish an important role for cardiac resynchronization therapy (CRT) in the management of selected patients with heart failure. We now report the long-term outcome during and subsequent to the randomized trial.

Methods and results: Enrollment was completed in March 2003. After reporting the main study results in 2005, investigators were asked to inform patients of the results and implant a CRT device if still appropriate. Subsequently, investigators were asked to consent patients for long-term follow-up until 30 September 2009. Of 813 patients originally enrolled, 343 (42%) died prior to re-consent, 111 patients (14%) were not or could not be contacted, 50 (6%) were alive but declined to participate, and 309 (38%) consented to long-term follow-up. Of patients originally assigned to the control group, > 95% of survivors had received CRT by the time of re-consent. From the time of randomization, 222 patients originally assigned to pharmacological therapy and 192 originally assigned to CRT were known to have died. The hazard ratio for mortality in patients originally assigned to CRT compared with those originally assigned to the control group was 0.77 (95% confidence interval 0.63-0.93; P = 0.007). No subgroup interactions were observed.

Conclusion: The effect of CRT on mortality observed during the randomized CARE-HF trial persisted during long-term follow-up. A high rate of CRT device implantation in the control group after completion of the randomized phase of the study may have prevented further divergence of the survival curves.

Publication types

  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Cardiac Pacing, Artificial
  • Cardiac Resynchronization Therapy / methods*
  • Confidence Intervals
  • Female
  • France
  • Heart Failure / drug therapy
  • Heart Failure / mortality*
  • Heart Failure / therapy
  • Humans
  • Kaplan-Meier Estimate
  • Male
  • Middle Aged
  • Prospective Studies
  • Statistics as Topic
  • Stroke Volume
  • Time Factors
  • Ventricular Function, Left