To the Editor:

We have read the article by Marín et al1 with interest and would like to describe the results of a similar study conducted at our hospital.

We studied 20 patients with hypertrophic cardiomyopathy (HCM) and implantable cardioverter defibrillator (ICD) between January 1993 and April 2005. The ICD was implanted for both secondary prevention (SP) and primary prevention (PP), with the latter considered to be the presence of 1 or more risk factors recognized as predictors of sudden death.2 The ICD was indicated for SP in 14 patients (70%), (7 with sudden death and 7 with sustained ventricular tachycardia), and PP in 6 (30%); among the PP group, 33% had a single risk factor. During a median follow-up of 6.5 years (PP 3 years vs SP 7 years; P=.016), 2 patients died, 1 in each prevention group (overall survival, 94% [5]). The percentage of patients free of appropriate shocks was 55% (12%) (PP 66% [19] vs SP 52% [14]; P=.87), with most patients receiving initial therapy in the first year of follow-up. Four (44%) of the patients with appropriate shocks had only 1 risk factor. There were no significant differences in appropriate shocks among those who had 1 or more risk factors (hazard ratio [HR], 1.25; 95% confidence interval [CI], 0.47-3.33), and no factors were significantly associated with a greater percentage of appropriate shocks. Inappropriate shocks were observed in 40%: 1 (16%) during PP and 7 (50%) during SP (P=.4). The main causes were sinus tachycardia, followed by atrial fibrillation, with 1 case due to oversensing; 50% of these also had appropriate shocks during follow-up.

We would like to make several comments about the use of the ICD to prevent sudden death in HCM and to compare our findings with the recent study published in this journal by Marín et al.1 First, our patients presented a high percentage of appropriate shocks (45%), higher than the values reported up to now and probably attributable to the longer follow-up time. The indication of an ICD for PP in these patients is increasingly accepted in light of recently published studies3-5; however, whether or not the presence of a single risk factor justifies implantation is still controversial and the major difference among the various research groups. Our group reflects a less restrictive indication. A third of our patients received an implant for a single risk factor, versus 4.4% in Marín's study.1 In the latter, the significantly lower percentage of appropriate shocks in the PP group makes the predictive value of a single risk factor alone questionable for justifying implantation of an ICD. Unlike the published series, we found a high percentage of appropriate shocks (33%) in this prevention group, but no differences in the percentage of appropriate therapies between the 2 groups, probably because of the smaller number of patients in PP. The low number of patients is an important limitation and, as mentioned by the authors of the cited article,1 more studies and research on new risk markers are needed to assess the efficacy of the ICD in PP. Multicenter studies such as the study underway6 will make a significant contribution in this regard.