Appropriate implantable cardioverter defibrillator therapy in hypertrophic cardiomyopathy: What happens on Sunday afternoons in May?

This editorial refers to ‘The relation of ventricular arrhythmia electrophysiological characteristics to cardiac phenotype and circadian patterns in hypertrophic cardiomyopathy’ by C. O'Mahony et al., on page 724 Hypertrophic cardiomyopathy (HCM) is a clinically and genetically heterogeneous disease. Sudden cardiac death (SCD) is the most devastating complication of HCM. The substrate for potentially life-threatening ventricular arrhythmias (VA) in HCM is complex. The hypertrophy causes dispersion of repolarization and refractoriness, which increases the vulnerability to triggered arrhythmia. Myocardial disarray, the histological hallmark of the disease leads to a disruption of cell-to-cell alignment and together with the expansion of the interstitial compartment and replacement fibrosis creates unidirectional conduction block that predispose to re-entry arrhythmia. Sarcomeric protein gene mutations influence calcium homeostasis, and abnormalities in ion fluxes during repolarization can cause after-depolarization and triggered activity. This unpredictable, complex arrhythmogenic substrate is further influenced by maladaptive autonomic responses, myocardial ischaemia, left ventricular outflow tract obstruction, and currently unknown factors.1 Although these characteristics are to some extent present in most HCM patients, risk of SCD in the general HCM population is low.2 The risk of sudden death increases in the presence of the following clinical risk factors: positive family history of SCD, unexplained syncope, abnormal blood pressure response to exercise, left ventricular wall thickness ≥30 mm, and non-sustained ventricular tachycardia, which form the basis of the guidelines for therapy with an implantable cardioverter defibrillator (ICD) in HCM.3 According to the guidelines, an ICD should be implanted in patients with HCM and a history of sustained VA (Class I, level of evidence B), and in patients with one or more known major clinical risk factor (Class IIa, level of evidence C).4 The patient population described in the article by O'Mahony et al.5 is a subset of the study by the same group that has recently been published in Heart.6 In this paper, 334 HCM patients with an ICD were included and the results clearly illustrate the limitations of ICDs in HCM patients. Although ICDs were very effective in preventing SCD, the rate of inappropriate shocks (4.6%/year) doubles those of appropriate shocks (2.3%/year) and there is lack of the relationship between the number of risk factors and the incidence of appropriate shocks. Ventricular fibrillation was the cause of appropriate ICD shocks in half the cases and fractional shortening (as a measure of left ventricular systolic function) was the only independent predictor for receiving an appropriate ICD shock. Inappropriate shocks were caused by atrial tachyarrhythmias in half the patients; the use of atrial leads did not reduce the inappropriate shock rates. ICD implant-related complication rates were 5.1%/year and were higher in patients with complex devices. In order to gain further insight into the arrhythmogenic substrate in HCM, the authors examined electrical triggers of VA in HCM by studying the stored intracardiac electrograms and the relation of the ventricular electrophysiological characteristics with cardiac phenotype and circadian patterns. A total of 230 HCM patients followed from 2003 onwards were included. The fast majority (224; 97%) received an ICD for primary prevention based on the presence of ≥1 risk factors. Most patients received a two-chamber system and in half of the patients, the device was programmed to deliver antitachycardia pacing (ATP). In total, 56 VA episodes (39 treated with ATP and 17 with shock) in 29 patients were included in analysis. The study illustrates the different patterns of initiation of VA even in a single HCM patient. Late coupled premature ventricular complexes (PVC) initiated the majority (72%) of VA and in most patients led to monomorphic ventricular tachycardia (86%). Since most HCM patients have PVCs, it remains unclear as to why and which PVC will lead to VA. The degree of hypertrophy, left ventricular outflow tract obstruction, systolic function, and left atrial diameter did not independently predict initiation of VA by PVC. The PVC-dependent initiation, the monomorphic nature and the effective ATP are characteristics of re-entry; promoted by the presence of myocardial disarray and fibrosis. In clinical practice, it is not feasible to look for myocardial disarray, but cardiac magnetic resonance imaging with late gadolinium enhancement enables the non-invasive detection and quantification of focal fibrosis, and is known to be an independent predictor of adverse outcome; it, therefore, would be interesting to correlate the current findings with imaging data.7 Abnormal automaticity or triggered activity forms another underlying mechanism for VA in HCM as a quarter of the analysed VA was sudden-onset without preceding PVC. Sarcomeric protein gene mutations cause abnormal intracellular calcium homeostasis which may lead to triggered activity; however, this study does not include genetic data of the studied patients.8 In contrast to the previous publication,6 ventricular fibrillation was the culprit arrhythmia in only 9% of treated VA. This difference is explained by the fact that only shocks are reported and not VA terminated by ATP.6 In the current study, ATP seems to terminate arrhythmias efficiently but did not significantly reduce occurrence of appropriate ICD shocks when comparing the groups with and without ATP with similar clinical characteristics. This can be explained by self-termination of most ATP-treated VA without haemodynamic compromise and the occurrence of VA immediately entering the ventricular fibrillation zone. However, patients were not randomized to undergo ATP and not all clinical data and ICD programming data are available for analysis. The VA cycle length was only determined by age, with older patients having slower rates, which according to the authors, may explain the peak of SCD in early adulthood.9 Interestingly in the larger cohort in the recent paper, age was not a predictor for appropriate ICD discharge.6 Circadian variations in VA events were different from those observed in ischaemic cardiomyopathies, with most VA occurring during daytime, on Sundays and in May. Sudden death in HCM has been associated with physical activity and in the current study 25% had baseline tachycardia, with a heart rate >100/min prior to the VA.10 However, data on the level of activity are lacking and 20% of VA occurred during potentially sleeping hours. The article nicely illustrates the heterogeneous aspects of HCM and the need for further studies including other clinical data in order to gain further insight in the modulators of the arrhythmogenic substrate in HCM. Conflict of interest: none declared.