Article Text
Abstract
In 1958, the British forensic pathologist, Donald Teare, reported a family in which eight young people had died suddenly from asymmetrical hypertrophy of the left ventricle. Five decades on, the prevention of premature death from ventricular tachyarrhythmia, heart failure and stroke remains a major aim of clinical management in what is now called hypertrophic cardiomyopathy. In this paper, we review the underlying mechanisms of death and discuss the strengths and weaknesses of current international guidelines for the identification and treatment of high-risk patients.
Statistics from Altmetric.com
In the opening article of this anniversary issue of the journal, Caroline Coats and Arthur Hollman elegantly illustrate the impact that individual scientists and doctors can have on prevailing concepts of disease.1 While Donald Teare’s paper on familial ventricular hypertrophy was not the first to describe what we now call hypertrophic cardiomyopathy,2 its publication did mark a turning point in the story of the condition by defining the disease and stimulating an intense clinical investigation which continues to the present day. As is often the case with newly recognised disorders, the most severe examples were the first to be studied by clinical investigators in Europe and North America. Work by Goodwin, Braunwald, Wigle and others led to the concept that hypertrophic cardiomyopathy was a disease predominantly of young people with a high incidence of sudden cardiac death and severe heart failure. In subsequent decades, greater awareness of the condition, the emergence of family screening and the introduction of ever more sophisticated imaging technologies revealed a much larger and hitherto overlooked population of individuals of all ages with ventricular hypertrophy, few if any symptoms, and little risk of premature death. Indeed, as Michael Davies and Donald Teare recognised in 1974: “It is, however, not generally appreciated that hypertrophic obstructive cardiomyopathy is not necessarily asymmetrical or fatal or that it also occurs in older subjects”.3 As we pause for reflection 50 years after Teare’s paper, the major clinical challenge is the identification of the small number of individuals who are prone to serious complications and rapid disease progression.
PATHOPHYSIOLOGY OF SUDDEN CARDIAC DEATH IN HYPERTROPHIC CARDIOMYOPATHY
In many respects, hypertrophic cardiomyopathy is the prototype substrate for ventricular arrhythmia. At a whole organ level, hypertrophy causes dispersion of repolarisation and refractoriness, thereby increasing the vulnerability of the myocardium to triggered arrhythmia (fig 1). The characteristic disruption of cell-to-cell alignment (myocyte disarray), together with expansion of the interstitial compartment and replacement fibrosis, creates areas of conduction block that predispose to re-entry arrhythmia4 (fig 2) and, at a single cell level, abnormalities in ion fluxes (in particular calcium) during repolarisation of the cardiomyocyte cell membrane cause after-depolarisations and triggered activity.5 Finally, this complex arrhythmogenic substrate is modulated by maladaptive autonomic responses, myocardial ischaemia and left ventricular outflow tract obstruction.6–8 And yet, in spite of this severe derangement of cardiac structure and function, contemporary studies report low annual mortality rates and even suggest that the overall survival of some patient cohorts is very similar to that of age-matched controls.9 Analysis of studies published over the last 50 years shows that mortality figures have fallen substantially (fig 3), with annual sudden death rates in the order of 1% or less.10 There are many possible explanations for this downward trend including a less severe clinical profile of populations reported in the literature (table 1), as well as the impact of modern clinical management strategies, although the latter hypothesis remains unproved in the absence of large-scale randomised trials. Notwithstanding these reassuring data, it is clear that the risk of sudden death varies between individual patients and within patients as they age. Thus, while sudden cardiac death can occur throughout life, there is a peak in incidence during late adolescence and young adulthood, whereas death from heart failure and stroke are of greater importance in middle-aged and elderly patients as progressive myocyte loss and myocardial fibrosis result in diastolic dysfunction and reduced contractile performance.9–12
In the years following Teare’s description of multiple cases of asymmetrical septal hypertrophy in a single family, it has been established that most patients with hypertrophic cardiomyopathy have familial disease with autosomal dominant inheritance. In the majority, the disease is caused by mutations in one of 12 genes that encode proteins of the cardiac sarcomere.13 Animal models of sarcomeric protein gene mutations suggest that male gender is associated with earlier disease expression and more severe morphological and functional abnormalities, including greater left ventricular dilatation and systolic impairment.14 15 Postmortem examinations of human hearts show that male patients with hypertrophic cardiomyopathy have more extensive fibrosis,16 and a retrospective analysis of older men reported larger cavity dimensions and lower ejection fractions compared to females of the same age.17 Nevertheless, a detrimental effect of male gender on natural history has yet to be demonstrated; indeed, in a multicentre study of approximately 1000 patients, female patients had a greater risk of progression to heart failure, death from heart failure or stroke.18 Although the rate of hypertrophic cardiomyopathy-related death was similar in men and women, there was a trend towards increased sudden death risk in women aged 40 or less.
In the United States, hypertrophic cardiomyopathy has been reported as the commonest cause of sudden unexpected cardiac death in young people and in competitive athletes.19 In Europe, it is much less common as a cause of sports-related death, possibly reflecting different national policies on pre-participation screening and other factors such as racial origin.20 In young patients known to have hypertrophic cardiomyopathy, sudden death may occur during or immediately after strenuous exertion, but the majority of such events happen during mild exertion or sedentary activities.21 In the past, it was assumed that many episodes of cardiac arrest with ventricular fibrillation were precipitated by brief runs of ventricular tachycardia or the sudden onset of atrial fibrillation, but data obtained from implantable cardioverter-defibrillators have shown that ventricular fibrillation often arises directly from sinus rhythm,22 illustrating the vulnerability of the myocardium in hypertrophic cardiomyopathy and the powerful influence of ill-understood physiological modulators of the arrhythmogenic substrate.
RECOMMENDATIONS FOR THE ASSESSMENT OF SUDDEN DEATH RISK
In the 50 years since Donald Teare’s paper, numerous clinical features have been suggested as markers of increased risk for sudden cardiac death and other disease-related complications.23–25 The contemporary approach to risk stratification is based on an assessment of a small number of readily determined clinical parameters that reflect the severity of the underlying myocardial disease (table 2).25 Consensus guidelines rank these risk factors into “major” and “possible” but, with the exception of an aborted cardiac arrest, there is very little evidence to suggest that any one risk factor is more predictive than another.
The two major anamnestic variables are a history of unexplained syncope (that is, excluding vasovagal faints that are not associated with an increased risk of sudden death) and a family history of premature and sudden cardiac death, both of which present their own unique problems to cardiologists managing individuals or families with the disease. The assessment of “unexplained” syncope needs to consider many potential mechanisms for loss of consciousness including left ventricular outflow tract obstruction, arrhythmia and peripheral vasodilatation caused by abnormal vascular reflexes. The circumstances of a syncopal event may provide a clue for the potential mechanism—for example, repeated episodes during exercise or on rapid standing might be caused by provocable left ventricular outflow tract obstruction, whereas unheralded syncope at rest or during mild exertion might suggest an arrhythmia as the cause. However, even when an extensive clinical assessment is performed, it remains difficult to identify retrospectively the mechanism responsible for the syncopal event.
The interpretation of sudden deaths in a family can also be a major challenge. The cause of death, particularly in distant relatives, is frequently unknown and can be impossible to determine retrospectively if a postmortem examination was not performed. In other cases, death occurs in the presence of other co-morbidities that may have been contributory factors. These uncertainties mean that in the individual patient, a family history of sudden cardiac death is no more predictive (and possibly less so) than other risk markers.
The remaining markers of increased sudden death risk are obtained from standard investigations, including two-dimensional echocardiography, ambulatory electrocardiographic recording and upright symptom-limited exercise testing. In several studies, echocardiographic measurements of severity and distribution of left ventricular hypertrophy have been shown to be associated with the risk of sudden cardiac death.26–28 While risk would appear to increase in a linear fashion with maximum left ventricular wall thickness, the available data are based on a small number of events and current guidelines report a value ⩾30 mm as the most predictive cut-off for sudden death risk. The prognostic power of this risk marker is greater in young patients and is increased by the presence of other risk factors.26 29 However, failure to visualise all myocardial segments and inter-observer error in echocardiographic wall thickness measurements have to be taken into account when making clinical decisions. Echocardiography is also important in determining the presence of dynamic left ventricular outflow tract obstruction. A number of studies have shown that outflow tract obstruction is associated with an increased risk of sudden death, although the relation between the severity of the gradient and risk varies.8 30–32 This observation raises difficult questions with respect to risk management because, unlike most other risk markers, outflow tract obstruction can be reduced with drugs or invasive procedures. However, interventions such as surgical myectomy or alcohol septal ablation are associated with a risk of death and complications, and there are no prospective data showing that relief of obstruction improves survival. At present, the consensus is that all patients with symptomatic outflow obstruction should initially receive medical therapy (β-blockers, verapamil, disopyramide) and only when severe symptoms persist should patients be considered for invasive gradient reduction.24
Several studies have shown that the presence of one or more runs of non-sustained ventricular tachycardia (NSVT) at a rate of 120 beats/minute or more on the ambulatory electrocardiogram is associated with a relative risk for sudden cardiac death of around 2–2.5.31–39 There is very little evidence that the duration, frequency or rate of runs influence this association, but age is an important modifier, with a fourfold relative risk in patients younger than 30 years of age.38 NSVT becomes increasingly common above this age, but the relative risk for sudden cardiac death decreases. The explanation for the complex relation between age, NSVT and sudden death risk is unknown, but it is possible that the substrate for NSVT is age dependent—for example, progressive myocyte loss and fibrosis may account for the increased incidence of NSVT in older patients, whereas myocyte disarray, myocardial ischaemia and abnormal autonomic function might be more important in the young.
In patients with hypertrophic cardiomyopathy, upright exercise testing is used to assess vascular behaviour during exercise. Several studies have shown that a failure of systolic blood pressure to rise by more than 20–30 mm Hg from baseline is associated with an increased cardiovascular mortality.23 31 40 41 Many mechanisms may account for this abnormal response, including inappropriate vasodilatation in non-exercising muscles, splanchnic venous pooling, poor augmentation of cardiac output and sub-endocardial myocardial ischaemia due to microvascular dysfunction. As with the severity of hypertrophy and NSVT, the clinical significance of an abnormal blood pressure response during exercise varies with age, being higher in younger (age 40–50 years or less) than in older patients. However, the blood pressure response is unhelpful in children and young adolescents, as it is normal to have a flat blood pressure response during exercise in this age group. Early interruption of the test because of inadequate effort, as well as the effects of medications, should also be considered when interpreting blood pressure responses to exercise.
PREVENTION OF SUDDEN DEATH
Even a casual reading of the literature on hypertrophic cardiomyopathy soon reveals the lack of prospective randomised trials on pharmacological and interventional treatments. This limitation is explained, in part, by the relative infrequency of the disease in clinical practice and the low rate of sudden death, which would dictate the need for prospective trials requiring many years of follow-up. The consequence is that the major sources of evidence for most treatments in patients with hypertrophic cardiomyopathy are observational non-randomised series. This lack of prospective trials in hypertrophic cardiomyopathy has become particularly relevant since the development of the implantable cardioverter-defibrillator (ICD), a device that has proved to be one of the most important therapeutic innovations of the past 50 years. In patients with coronary artery disease and dilated cardiomyopathy, large randomised trials have compared ICDs with antiarrhythmic drugs (predominantly amiodarone) and have shown that only the ICD improves survival. In patients with hypertrophic cardiomyopathy similar comparative data are not available. However, multicentre registries report a 10% per year ICD intervention rate for ventricular fibrillation or rapid ventricular tachycardia in patients with a history of aborted sudden death, and approximately 4% per year in high risk patients (some of whom were taking amiodarone) without previous cardiac arrest.42 43 Based on the assumption that these observational data translate into a similar survival benefit to that observed in other diseases, consensus guidelines recommend ICD implantation in patients with hypertrophic cardiomyopathy and a sustained ventricular arrhythmia, in those with multiple risk factors and in selected patients with a single risk factor judged to be at high risk.25 However, the notion that implantation of an ICD may be justified in patients with a single marker of increased risk raises some important clinical questions, given that approximately 25% of patients with hypertrophic cardiomyopathy have only one risk factor.23
There is general agreement that the greater the number of clinical risk factors a patient has, the higher is their risk of sudden cardiac death.24 25 Few studies have tried to determine the magnitude of risk in relation to the number of risk factors, but in one the presence of two or more risk markers identified a cohort of patients with an annual sudden death risk of approximately 3–6%, whereas the presence of any single risk factor was associated with an annual risk of approximately 1%.23 Given that ICDs are associated with significant lifelong morbidity and limitations in quality of life, particularly in young patients, it is crucially important to identify those patients with a single risk factor who may be at high risk of sudden death and candidates for an ICD. The current guidelines deal with this dilemma by using the ambiguous sentence “ICD implantation can be effective for primary prophylaxis against sudden cardiac death in patients with hypertrophic cardiomyopathy who have one or more major risk factors”.25 Indeed, the guidelines reflect a genuine concern that implantation of ICDs exclusively in patients with multiple risk factors could result in a failure to treat individuals who have only a single conventional risk factor (for example, multiple sudden and unexpected deaths in young relatives) and yet may still be at high risk of a fatal event. A single sudden death in a family including other young affected individuals without additional risk factors may represent an even more difficult clinical setting. For example, should prophylactic ICD implantation be recommended for all the other affected family members. Therefore, the primary question remains, which patients with a single risk factor are at high risk? For the moment, the answer is non-evidence based and relies to a large extent on the experience of the individual cardiologist. In selected patients, the presence of other clinical features such as left ventricular outflow tract obstruction, myocardial ischaemia (albeit very difficult to assess) or extensive myocardial scar detected by cardiovascular magnetic resonance imaging might tip the balance in favour of more aggressive management. However, all patients perceived to be at a potentially increased risk for sudden death should be fully informed of the advantages and possible complications of the ICD, as well as the unresolved issues regarding risk stratification in hypertrophic cardiomyopathy.
OTHER CAUSES OF DEATH IN HYPERTROPHIC CARDIOMYOPATHY
Discussions about risk stratification and indications for ICDs in patients with hypertrophic cardiomyopathy sometimes overshadow other important aspects of disease management that can impact on quality of life and longevity. In particular, stroke secondary to atrial fibrillation represents a significant risk in older patients.44–46 Left atrial size is the strongest predictor of atrial fibrillation, the risk rising exponentially in patients with a left atrial diameter of more than 45 mm or an indexed maximum left atrial volume of more than 34 ml/cm2. In addition to stroke risk, the development of atrial fibrillation is associated with reduced survival and progression to severe heart failure.47 Based on data from other cardiac disorders, anticoagulant therapy with warfarin is indicated in patients with either paroxysmal or chronic atrial fibrillation. Prophylactic anticoagulation in patients with marked left atrial dilatation who have not yet developed atrial fibrillation remains an unresolved issue.25
Approximately 1% of patients with hypertrophic cardiomyopathy develop end-stage disease with severe systolic dysfunction every year. This unfavourable evolution of the disease is associated with an overall mortality rate up to 11% per year from progressive heart failure and sudden ventricular arrhythmia.48–54 Patients who progress to end-stage hypertrophic cardiomyopathy with severe heart failure tend to have more marked hypertrophy, lower fractional shortening and larger left ventricular end-diastolic dimensions at initial evaluation than individuals who do not develop end-stage disease.53 54 In some series, age is also an important prognostic factor, with a twofold relative risk of death from end-stage disease or transplantation for patients aged 40 years or more compared to patients aged 15–40 years.55 The benefits of administering inhibitors of the renin-angiotensin-aldosterone system and β-blockers in patients with systolic heart failure caused by dilated cardiomyopathy or coronary artery disease are well established. However, the only data in patients with hypertrophic cardiomyopathy come from two small trials evaluating the effect of angiotensin-receptor blockers on left ventricular function. A prospective randomised study of 24 patients with non-obstructive hypertrophic cardiomyopathy followed up for 12 months showed that valsartan suppressed the synthesis of type I collagen and was associated with lower aldosterone levels.56 A second prospective, but non-randomised trial of losartan in 20 patients with non-obstructive hypertrophic cardiomyopathy followed for 6 months showed a significant improvement in diastolic performance and a reduction in left atrial diameter in the treatment group.57 These data provide some evidence that inhibition of the renin-angiotensin-aldosterone system might be beneficial in patients with hypertrophic cardiomyopathy, but there are important limitations to the use of vasodilators in this disease, not least their potential to provoke left ventricular outflow tract obstruction by reducing peripheral resistance. This problem is not, however, shared by aldosterone antagonists such as spironolactone, and while there are no data in humans, animal studies have suggested that these medications may inhibit hypertrophic and profibrotic signals, reduce myocyte disarray and normalise collagen volume fraction.58 59 A number of trials of various interventions designed to treat left ventricular systolic impairment in patients with hypertrophic cardiomyopathy are in progress.
FUTURE APPROACHES TO RISK ASSESSMENT?
One approach to patients who fall into the grey zone for risk of hypertrophic cardiomyopathy-related death is to seek additional clinical parameters that are associated with a poor outcome. For example, particular sarcomeric protein gene mutations,60 61 impaired coronary flow reserve7 and extensive areas of hyperenhancement on gadolinium-enhanced cardiac magnetic resonance imaging62 could prove useful. However, an approach that simply bolts on additional parameters to the current model is unlikely to resolve the many different scenarios that cardiologists encounter when managing such an aetiologically diverse and clinically heterogeneous disorder. It is also unsatisfactory to apply dichotomous criteria for risk assessment to continuous variables such as wall thickness or blood pressure response. A more effective approach might be to increase the sophistication and power of the risk stratification model by considering the interaction of a broader range of parameters that more accurately describe the pathophysiological substrate of the disease. This could then be used to construct a risk algorithm analogous to that used in patients with diabetes or vascular disease. The adoption of a standardised approach to risk estimation evaluation would assist in disease management and help cardiologists to better understand individual patients’ risk. Development of such a tool is a formidable task because of the heterogeneous clinical course of the condition and relatively low annual event rate, but the growth of interest in cardiomyopathies and the support for international collaborative research from bodies such as the European Commission offer cautious optimism that new approaches to risk stratification will emerge.
CONCLUSIONS
It is impossible to anticipate every clinical scenario in any disease, and so it is not surprising that current guidelines on the identification and treatment of patients with hypertrophic cardiomyopathy who are at risk of premature death are imperfect. Nevertheless systematic approaches to clinical assessment and careful monitoring of patients with hypertrophic cardiomyopathy do identify subgroups that benefit from medical intervention and almost certainly improve long-term outcomes. The agenda for the future is to build on the work of the past five decades since Donald Teare’s classic paper in order to develop international collaborations with the power to address some of the many questions that remain unanswered.
REFERENCES
Footnotes
Funding: None.
Competing interests: None.