Relationship between rehospitalization and future death in patients treated for potentially lethal arrhythmia.

INTRODUCTION: It is generally considered that death is the only appropriate endpoint to evaluate interventions for preventing death; however, this belief may be based on the previous use of inappropriate or inadequate surrogates for death. The aim of this study was to evaluate whether rehospitalization following implementation of an intervention is a reasonable surrogate for death. METHODS AND RESULTS: The time from discharge following intervention to rehospitalization was evaluated for 997 patients discharged after baseline hospitalization in the Antiarrhythmics Versus Implantable Defibrillators Trial. The relationship between rehospitalization for various reasons and subsequent death was compared in the two treatment arms to assess the adequacy of rehospitalization as a surrogate for death. Included were rehospitalization for: any reason, a cardiac problem, a noncardiac problem, new or worsened congestive heart failure (CHF), an acute coronary syndrome, and a cardiac procedure. For all of the reasons except cardiac procedure, rehospitalization was associated with a substantially increased hazard for subsequent death. Rehospitalization for new or worsened CHF was most closely (that is, temporally) related to subsequent death and was the only reason for rehospitalization, which fully explained the treatment effect of implantable cardiac defibrillators compared with antiarrhythmic drugs on death. CONCLUSION: Rehospitalization is a significant risk factor for subsequent death. However, only rehospitalization for new or worsened CHF appears to be a potential surrogate for death in the setting of antiarrhythmic interventions.