Author + information
- Mark V. Sherrid, MD⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Mark V. Sherrid, Hypertrophic Cardiomyopathy Program, St. Luke's-Roosevelt Hospital Center, 1000 10th Avenue, 3B30, New York, New York 10019
“Case 1. In September, 1950, a boy of 14, had a “blackout” and fallen from his bicycle; two months later he again became dizzy and fell to the ground. … On February 20, 1951, he was being chased around the playground of his school, when he suddenly collapsed and was found to be dead on arrival at hospital twenty minutes later.” And “Case 9 … aged 16, a brother of Case No. 5, collapsed and died while riding his bicycle. … Post mortem he was found to be a well nourished and well developed young boy whose heart was virtually identical in appearance with that of his sister.”
Thus begins Donald Teare's (1) 1958 report, the first modern clinical and pathological description of 9 patients with hypertrophic cardiomyopathy (HCM). From the beginning, HCM has been associated with sudden cardiac death (SCD) in adolescents. Also, one notes the familial occurrence of HCM, the familial predisposition for SCD, the association of death with exertion, and the premonitory syncope.
The clinical diagnosis of HCM rests on the detection of ventricular hypertrophy in the absence of a hemodynamic or other cause for the hypertrophy observed. In the pediatric group, besides hemodynamic causes, other causes of hypertrophy such as inborn errors of metabolism, malformation syndromes, and neuromuscular disorders must be excluded before the diagnosis of primary HCM is made. In the Pediatric Cardiomyopathy Registry, these other causes of hypertrophy were detected in nearly 30% of patients, usually diagnosed in the first 2 years of life; the remaining 70% have primary HCM (2). In primary HCM, mutations of genes coding sarcomeric proteins are found in 50% of adult cases and are the cause of hypertrophy.
HCM is the most common cause of SCD in the young (3). In unselected patients surviving >1 year, annual mortality was 1% per year, similar to that observed in adults, and the mode of death was most often sudden (2). However, the cumulative incidence of SCD in HCM is not well defined in pediatric cohorts, because the follow-up durations are relatively short. Follow-up through the pediatric age range may underestimate the cumulative magnitude of the problem because the disease, and its associated risk factors, persists into adult life. Risk factors of massive hypertrophy and family history of SCD in a first-degree relative do not disappear when a patient achieves maturity. The strategy of risk stratification indicating implantable cardioverter-defibrillator (ICD) therapy depends on finding selected patients at higher risk for SCD (2% to 4% per year).
In this issue of the Journal, Maron et al. (4) report on the efficacy and complications of ICD implantation in 224 highly selected patients with HCM (mean age at implantation 14.5 years), followed for 4.3 years. For clinical indications, patients were selected to undergo ICD implantation on the basis of the SCD risk factors that have been advanced in adults. The best established are survival after resuscitated SCD or sustained ventricular tachycardia, massive hypertrophy, unexplained syncope, a family history of SCD in a close relative, and left ventricular systolic dysfunction (5,6). These risk factors have low positive predictive value in adults but are the best available clinical markers. Until this report, their predictive utility has not been established in children.
The SCD rate in the 224 patients who underwent implantation was very low; SCD occurred once in 963 patient-years (0.1 SCDs per 100 patient-years). The 1 patient with SCD died suddenly because of the electrical failure of his defective ICD. In the patients who received ICDs as primary prevention solely because of risk factors, there were appropriate discharges in 26 of 188 (14%), an appropriate intervention rate of 3.1% per year. Appropriate discharge rates were 4 times higher in secondary prevention patients who had already been resuscitated from SCD or sustained ventricular tachycardia. Complications were frequent: 28% of patients experienced inappropriate ICD shocks from sinus tachycardia, supraventricular tachycardia, T-wave oversensing, or lead problems, and 12% had other significant complications inherent to transvenous ICD implantation. Thus, 40% experienced device-related complications by age 17 (9.5% per year).
In considering ICD implantation, physicians weigh benefit versus risks. The risks are reported here and previously (4,5,7). But what is the precise benefit of ICD implantation in this age range? In large ICD studies of dilated cardiomyopathy or ischemic left ventricular dysfunction, appropriate discharges in ICD groups exceed SCDs in control groups by a ratio of 2:1. With conventional ICD settings, appropriate therapies may be delivered to ventricular tachycardia that would have self-terminated before the therapy was delivered. Thus, not all appropriate ICD discharges occur in patients who would otherwise have died suddenly (8–10). The validity of these concepts is demonstrated in the recent randomized MADIT-RIT (Multicenter Automatic Defibrillator Implantation Trial: Reduce Inappropriate Therapy) comparing conventional ICD programming with more restrictive criteria, requiring either a heart rate >200 beats/min or adding delay before discharges. Nearly half the patients in the trial had nonischemic heart disease. Episodes of antitachycardia pacing were >60% lower with restrictive programming because of the phenomenon of self-termination. Moss et al. (11) commented that some antitachycardia interventions for ventricular tachycardia with conventional programming may thus be considered unnecessary. We do not know how the devices were programmed for the pediatric HCM ICD cohort, but half of the episodes requiring therapy were for ventricular tachycardia.
We do not know whether the 3% intervention rate for appropriate primary prevention therapy reported here in young patients with HCM is equivalent to lives saved; given the data mentioned previously, it is likely there is some overestimation. The application of more restrictive programming criteria for therapy would allow an important and more precise recount of the incidence of appropriate therapy in patients with HCM. There is no doubt that ICD therapy is lifesaving in patients with HCM at high risk, including the pediatric cohort. What is in doubt is the percentage of patients who benefit from primary prevention, and this may figure into the risk-benefit equation when there is ambiguity, as there so often is. A count of appropriate discharges may overestimate benefit, and recognition of this may temper judgment when the indication is marginal or in doubt.
ICD implantation is a lifelong decision with profound consequences for adolescents. How should families be counseled about benefit and risks? The risks of ICD implantation may be effectively communicated as 6 I's: implantation risk, infection, inappropriate shock, device imperfection, insufficiency (heart failure caused by tricuspid regurgitation), and insurance risk (never using the device, analogous to outliving term life insurance) (7). Despite this litany of complications, the benefits of ICD implantation in carefully selected patients outweigh the risks. Complications are virtually always manageable, while SCD is irretrievable.
Inappropriate shocks are reduced by correct ICD programming. In the MADIT-RIT trial, inappropriate shocks were strikingly reduced by 79% when high-rate programming was applied that restricted device therapy to heart rates > 200 beats/min. If these settings were adopted in the pediatric HCM group, it is likely that the high incidence of inappropriate shocks observed here (28%; 6.5% per year) would be lowered, without adverse risk for syncope or death.
Fewer than 1% of all ICDs are placed in children, and no system has been designed specifically for this population. Pediatric patients who are candidates for ICD present unique choices (12). The weak link in ICD therapy is lead failure, and problems related long term to venous access (7). These issues are compounded by the expectation that fully functioning leads will be required for many decades in pediatric patients, longer than in adults. But existing transvenous leads have been plagued by fractures and insulation failures, frequently requiring extraction, with its own significant risks. Because short-term and long-term risks rise with the insertion of multiple transvenous leads, a single lead often is the preferred choice for children. Single-coil leads, with active fixation, have been recommended, as well as allowance for lead redundancy to accommodate growth (12).
The approval by the U.S. Food and Drug Administration of the completely subcutaneous ICD may offer a choice for patients who do not require ventricular pacing for persistent bradycardia or antitachycardia pacing (13). This device may provide a safer option for adolescents, but long-term experience is not yet accumulated. A trial comparing conventional ICDs with the subcutaneous ICD in adults has been launched, with endpoints to assess noninferiority with respect to complications, shock efficacy, and mortality (14). A drawback to its use in very small children is that the generator is twice the weight and volume of current systems because of the requirement of a higher capacity battery. However, its benefits long term (i.e., no intravascular leads and lower radiation exposure) may prove to outweigh this disadvantage.
Heart failure symptoms and left ventricular outflow tract (LVOT) obstruction are frequent in patients with HCM. In the present study, 43 patients (19%) underwent surgical septal myectomy before ICD implantation. In adults, survival is excellent after successful surgical relief of LVOT obstruction (15,16). Moreover, in patients who undergo ICD implantation, appropriate discharges are very uncommon (17). McLeod et al. (17) found an annualized ICD event rate of only 0.24% per year in patients after surgery. A randomized trial of myectomy will never be done because of its benefit for symptom relief. Nevertheless, mortality data suggest that myectomy reduces the risk for death by removing the underlying substrate for SCD and heart failure. This raises a question as yet unanswered by the data of Maron et al. (4): should pediatric patients with a risk factor for SCD and other symptoms receive ICDs even if they have complete surgical relief of obstruction with its expected excellent long-term prognosis? Syncope, a potential indication for ICD, in patients with rest or provocable LVOT obstruction is often caused by obstruction, not by arrhythmias. The history or workup might support one or the other etiology. Obstructed patients might continue to experience syncope if ICDs are implanted, when the loss of consciousness is actually because of gradient. These patients might be better off treated with surgical myectomy, saving them lifelong leads. Similarly, inadequate blood pressure rise with exercise or hypotension can be caused by LVOT obstruction. It is frequently reversible after surgical septal myectomy. A prospective long-term registry of survival of adolescent myectomy patients might answer these questions.
In another era, soon after the introduction of prosthetic valve replacement, it became clear that the heart disease had been replaced by the serious syndromes of valve replacement. So too, patients who receive ICDs have their own particular morbidity and management problems. Patients with SCD risk present evolving choice and long-term management issues that require judgment and expert care. Nowhere is this as clear as in children and adolescents with HCM and high risk for SCD.
Dr. Sherrid has reported that he has no relationships relevant to the contents of this paper to disclose.
↵⁎ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
- American College of Cardiology Foundation
- Teare D.
- Colan S.D.,
- Lipshultz S.E.,
- Lowe A.M.,
- et al.
- Maron B.J.
- Maron B.J.,
- Spirito P.,
- Ackerman M.J.,
- et al.
- Elliott P.M.,
- Poloniecki J.,
- Dickie S.,
- et al.
- Tung R.,
- Zimetbaum P.,
- Josephson M.E.
- Ellenbogen K.A.,
- Levine J.H.,
- Berger R.D.,
- et al.
- Connolly S.J.
- Hill A.C.,
- Dubin A.M.
- Ommen S.R.,
- Maron B.J.,
- Olivotto I.,
- et al.
- McLeod C.J.,
- Ommen S.R.,
- Ackerman M.J.,
- et al.