Letter by Boriani et al Regarding Article, “Death Without Prior Appropriate Implantable Cardioverter-Defibrillator Therapy: A Competing Risk Study”

To the Editor:

We were interested to read the article by Koller et al, published in the April 15, 2008, issue of Circulation.¹ The authors address a topical question in the field of sudden cardiac death prevention: the imbalance between the number of patients who receive an implantable cardioverter-defibrillator (ICD) and the relatively small proportion of subjects who receive appropriate ICD therapy. More precisely, they focus on the patients who die before receiving appropriate ICD therapy.

We are facing a progressive broadening of indications for ICD implantation, with an increasing economic burden coupled with the impact of adverse events such as inappropriate ICD therapies. Despite the extensive research, none of the proposed predictors for sudden cardiac death risk stratification has been shown to grant the desired performance. The approach adopted by Koller et al¹ provides new insights for risk-benefit evaluation on this debated subject. They used the competitive risk regression analysis to evaluate the probability of appropriate ICD therapy or death before appropriate ICD intervention in a population sample of 442 ICD carriers for primary/secondary sudden cardiac death prevention. The cumulative incidence of appropriate ICD therapy for ventricular fibrillation (VF) was 13% whereas 23% died without prior therapy for VF. These results are of interest in the complex process of application of the current indications for ICD therapy to “real world” clinical practice.²

However, from a clinical point of view, we think that this analysis should also take into account heart transplantation as a further competing risk factor, because heart transplantation was adopted in other studies as an end point combined with death.^3,4 In fact, many ICD carriers present or will develop heart failure and might eventually be candidates for heart transplantation, and ICD implantation has sometimes been used as a bridge to heart transplantation.⁵ In a study population with half of the subjects aged <63 years, like that reported by Koller et al, heart transplantation may represent a therapeutic option for at least some patients, and we think it should be considered in a competitive risk analysis.

The authors conclude that patients implanted with ICDs for secondary sudden cardiac death prevention experienced appropriate ICD interventions more frequently. It has to be stressed that, in this study, the ICD programming was not predefined and uniform but varied according to clinical judgment, and it is common practice to program automatic ventricular tachycardia/ventricular fibrillation detection, focusing on sensitivity (over specificity) for patients in secondary prevention (eg, treating slower arrhythmias). This can be another possible explanation for this phenomenon (analyzing the different cycle length of the recorded arrhythmias could give a clue on this).

Finally, we think that the reported data on a decreasing incidence of the first appropriate ICD intervention during the follow-up should not be considered as a finding in favor of reconsideration of ICD replacement in patients who did not experience any intervention. A median follow-up of 3.6 years is not enough to explore this topic, and we think that a period lasting at least 2 device lives (≈8 to 10 years) would be more appropriate.