Author + information
- ↵*Reprint requests and correspondence:
Dr. William J. McKenna, Department of Cardiological Sciences, St. George’s Hospital Medical School, Cramer Terrace, London SW17 0RE, UK.
Hypertrophic cardiomyopathy (HCM) is a major cause of premature sudden death in the young, a fact made all the more unsettling to affected families and physicians by the frequent absence of symptoms prior to death (1). Published annual sudden death rates have varied as a function of the patient selection and management, with values prior to the use of prophylactic implantable cardioverter-defibrillators (ICD) for children and adolescents of 3% to 6%, and 0.5% to 1.0% in adults. The motivation to increase efforts to identify high-risk patients has been greatly stimulated by the growing recognition that the ICD is highly effective in primary and secondary prevention of sudden cardiac death in HCM (2).
Clinically, HCM is defined and diagnosed by the demonstration of unexplained left ventricular hypertrophy (LVH). Mutations in eight different sarcomeric contractile protein genes (beta myosin heavy chain [MYH7], troponin T [TNNT2], alpha tropomyosin [TPM1], myosin binding protein C [MYBPC3], regulatory myosin light chain [MYL2], essential myosin light chain [MYL3], troponin I [TNNI3] and alpha cardiac actin [ACTC] have been shown to cause disease; mutations in the titin, alpha heavy chain myosin and troponin C genes have also been described (3). Preliminary genotype-phenotype studies have suggested that some mutations may be associated with a high incidence of disease-related mortality, whereas others appear to have a more “benign” course (4). This has led to the attractive hypothesis that genotyping may facilitate identification of individuals at high risk of sudden cardiac death (5).
In this issue of the Journal, Ackerman et al. (5)from the Mayo Clinic examine the prevalence in a population of 293 patients of several mutations (four MYH7: R403Q [exon 13], R453C [exon 14], G716R and R719W [exon 19], and one TNNT2 mutation: R92W [exon 9]), which have previously been associated with premature sudden cardiac death. Fifty-three percent of the study cohort, approximately twice the proportion of patients seen in most referral centers, had symptoms associated with left ventricular (LV) outflow tract obstruction, reflecting the well-recognized surgical expertise of the Mayo Clinic. Although the risk profile of the cohort was not presented in detail with regard to syncopal episodes, nonsustained ventricular tachycardia (NSVT) on Holter, or abnormal blood pressure response during upright exercise, 24% were recorded to have a family history of at least one sudden cardiac death prior to age 40, 6% had extreme hypertrophy (maximum LV wall thickness ≥30 mm) and 8.5% had received an ICD. These proportions and other demographics such as age, symptoms and LV morphology are similar to those seen in other referral centers.
Ackerman et al. (5)found a malignant mutation in only 3 (1%) of the 293 patients. They conclude that given the low prevalence of “malignant” MYH7 and TNNT2 mutations in a tertiary center, genetic testing is unlikely to contribute significantly to risk assessment. This view is supported by a number of additional considerations, in particular the observation that even within so-called high risk families, there is variable disease expression and prognosis, probably reflecting the influence of other genetic and environmental factors on gene expression. For example, a large Scottish family with an exon 15 splice site mutation in TNNT2 in which eight individuals died suddenly age <30 years (six as teenagers) whereas eight other affected individuals survived into their seventh or eighth decade (4). This variability is also illustrated by a number of reported exceptions to many of the proposed genotype-phenotype associations, although these may reflect factors such as the size of the families studied, the failure to consider the influence of effective prophylactic treatment and the confounding presence of a second disease causing mutation. A limitation of the approach taken in the Ackerman et al. (5)study was that there must have been other “malignant” mutations that were impossible to assess because of our current limited state of knowledge and genotyping capacity. The fact that for several of the disease genes (MYH7, MYBPC3), individual families have their own private mutations will also continue to represent an obstacle.
If genotyping is currently problematic in prognostic assessment, can clinical evaluation identify the high-risk patient? Over the past four decades, a number of clinical features have been proposed as markers of sudden death risk in patients with hypertrophic cardiomyopathy. Their application in clinical practice has, however, remained unsystematic and haphazard, reflecting not only perceived doubts about the predictive value of individual risk markers, but also a number of misconceptions about the natural history of the disease. The latter have been exacerbated by what seem to be contradictory survival studies demonstrating both prognosis that differs little from that seen in age-matched controls and others that report increases in risk sufficient to warrant aggressive prophylactic antiarrhythmic treatment in anyone with one of a number of “adverse” clinical manifestations. This apparent paradox is exemplified by the recent debate on the clinical significance of extreme LVH in HCM (6,7). Several studies have shown that patients with a maximum LV wall thickness of ≥30 mm have an increased risk of dying suddenly, a finding that some authorities suggest is sufficient to warrant consideration of an ICD in any patient with “extreme” hypertrophy. Unfortunately, at least 10% of patients in most of the published survival studies have an LV wall thickness of ≥30 mm. Moreover, the majority of sudden deaths occur in patients with a wall thickness of <30 mm. Reliance on hypertrophy alone would, therefore, fail to prevent the majority of HCM-related sudden deaths.
The fundamental problem with most proposed clinical risk markers is that, individually, they are at best only modestly predictive of short- to medium-term risk of sudden death (8). Many cardiologists have, as a consequence, treated patients in reaction to events such as the sudden death of a sibling or an episode of unexplained syncope. Although this policy is understandable given the often emotive context, clinical and genetic studies suggest that it may not always be necessary. Despite clinical experience suggesting that adverse family history and syncope are harbingers of future events in some individual circumstances, most statistical analyses of outcome in large series indicate that they are no more predictive of sudden death risk than other less routinely determined and less dramatic clinical markers (e.g., NSVT on Holter, abnormal exercise blood pressure response). In the case of syncope, this is not particularly surprising given the many different mechanisms that may cause a patient with HCM to faint. Similarly, the highly variable natural history in patients with the same mutation indicates that family history by itself is unlikely to be a reliable guide to an individual patient’s risk of sudden death.
Clinically, the easiest high-risk group to define are those patients who have already declared their propensity for fatal ventricular arrhythmia by surviving an episode of sustained ventricular tachycardia or ventricular fibrillation. These patients have a 5% to 11% chance of a further event in five years; given the proven efficacy of ICDs in this group, they should be treated aggressively (2). At the other extreme, most experts would agree that asymptomatic patients with mild hypertrophy (≤2 cm), a normal exercise blood pressure response, an absence of arrhythmia on Holter and exercise testing and with no family history of premature sudden death have a very low risk of sudden death and can, in the main, be reassured.
To guide decision making in the remainder of patients, we have recently proposed that patients should be “stratified” using a careful but relatively simple assessment that in addition to clinical and family history includes exercise testing, Holter monitoring and echocardiography (8). In a prospective study of a consecutively referred population, patients with two or more risk markers (a history of unexplained syncope, a family history of premature sudden deaths, an abnormal exercise blood pressure response, NSVT, or massive LVH) had an annual sudden death risk of at least 3%, suggesting that noninvasive clinical assessment alone can be used to identify a cohort of patients who might benefit from ICD implantation (8). It is clear that risk stratification remains a challenge in particular subgroups of patients—for example, the young and patients who have only a single “risk factor.” When advising individual patients on their risk, the clinical assessment has also to take into account factors such as age, evidence of ischemia and additional details related to the risk marker, which, though important, have been difficult to factor in to statistical analysis. The complexity of the risk assessment does not, however, negate the process itself, and in experienced hands most high-risk patients can be identified prospectively.
Does genotyping have a future role in guiding risk-factor stratification? The concept of a malignant versus benign mutation, which underpinned the Ackerman et al. study (5), assumes some degree of homogeneity of patients with the same mutation. Similarly, the current approach to clinical risk-factor assessment assumes patients with the same risk markers have a similar risk of sudden death. Nevertheless it is probable that future studies will confirm the original hypothesis in Ackerman et al. (5)that selected MYH7 and TNNT2 mutations do confer an increased risk of disease-related complications. At present, the clinician does not have genotyping available as a routine clinical test and must consider an HCM patient under the broad umbrella of “unexplained LVH.”
Finally, preliminary genotype-phenotype studies suggest differences between the disease-causing genes with respect to penetrance, age of expression, as well as morphology and prognostic severity (3,4). In the future, HCM may be considered as eight or more different but related heart-muscle disorders with a differing pathogenesis and, importantly, a different risk-factor profile. Whether the attractive concept of benign versus malignant mutations is sustainable is uncertain, but knowledge of the underlying genotype should set the framework for both symptomatic and risk management strategies.
↵* Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- McKenna W.J.,
- Behr E.R.
- Ackerman M.J.,
- VanDriest S.L.,
- Ommen S.R.,
- et al.
- Elliott P.M.,
- Poloniecki J.,
- Dickie S.,
- et al.