Risk Stratification in Hypertrophic Cardiomyopathy: Is Two-Dimensional Echocardiographic Strain Ready for Prime Time?

Since the initial clinical description 50 years ago, hypertrophic cardiomyopathy (HCM) still remains the most common cause of sudden cardiac death in young patients. Currently, 5 conventional risk markers (ie, family history of sudden death due to HCM, syncope, abnormal blood pressure response to exercise, nonsustained ventricular tachycardia [NSVT] on Holter monitoring, and maximal left ventricular [LV] wall thickness ≥ 30 mm) can be used to select high-risk patients who will most likely benefit from an implantable cardioverter-defibrillator (ICD) for primary prevention of sudden death. For patients deemed to be high risk, the ICD has been shown to be effective in terminating lethal ventricular tachyarrythmias. As a result, the translation of these primary prevention risk factors into clinical practice has affected the lives of a substantial number of patients with HCM by providing them with necessary lifesaving therapy. Nevertheless, the current risk-stratification strategy has a number of limitations, including the low positive predictive values for the individual risk factors (15%-30%) and the difficulties inherent in determining how many risk factors are enough to constitute high-risk status in an individual patient. As a result of these imperfections, sudden death occasionally occurs among patients with clinically identified HCM with no evidence of increased risk. Thus, novel clinical markers are necessary to help enhance risk-stratification decision making, particularly with regard to recommendations for ICD therapy among patients in whom risk may still remain ambiguous even after assessment with the conventional markers.

So what does the future hold for improving on our current risk-stratification strategy in patients with HCM? An answer to this question first requires a closer look into the possible mechanisms responsible for promoting adverse events in this disease, in particular a greater appreciation for the abnormal myocardial substrate that defines HCM. In this regard, the histopathologic hallmarks of HCM include myocyte hypertrophy and disarray, and an expanded extracellular collagen matrix, which is primarily the result of structurally abnormal intramyocardial arteries that lead to microvascular ischemia, myocyte death, and repair in the form of replacement fibrosis. This abnormal myocardial substrate is thought to provide a nidus for the generation of potentially lethal ventricular arrhythmias—the predominant mechanism of sudden death in this disease. Indeed, postmortem studies in patients with HCM who have died suddenly have shown that fibrosis comprises an average of 15% of the hypertrophied ventricular septum, 8-fold greater than in patients with other forms of heart disease, suggesting a possible causal association between fibrosis and potentially lethal arrhythmias in this disease. In addition, cross-sectional studies in patients with HCM have demonstrated increased values of collagen synthesis, which correlate with abnormalities of diastolic function. Therefore, the expanded collagen matrix in HCM likely represents an important contributor to risk for sudden death and heart failure symptoms.

However, it is only recently that novel cardiovascular imaging techniques have emerged to permit characterization of the myocardium in patients with HCM. This can be accomplished through direct visualization of fibrosis using contrast-enhanced cardiovascular magnetic resonance (CMR) or quantitative measures of myocardial function assessed using newer two-dimensional (2D) echocardiographic techniques or tagged CMR (presaturation grids are applied to the myocardium that deform during the cardiac cycle, providing quantitative measures of regional function with offline analysis software). Therefore, these cardiovascular imaging methods provide an opportunity for a greater understanding of the complex relationship between fibrosis and important HCM-related disease variables, including increased arrhythmic risk.

Contrast-enhanced CMR with late gadolinium enhancement (LGE) has the capability of identifying myocardial fibrosis in vivo in patients with structural heart disease. The prevalence of LGE (ie, fibrosis) in HCM is 50% to 60% and can often occupy substantial portions of the LV myocardium. LGE is thought to represent myocardial fibrosis on the basis of selected reports in native hearts after transplantation in patients with end-stage HCM (ejection fraction < 50%) that demonstrated concordance between in vivo LGE-CMR images and gross and histopathologic evidence of fibrosis. Although LGE can be observed in LV segments with normal or mildly increased wall thickness, there is a strong association between maximal wall thickness and LGE, with LGE more common in thick LV segments. This finding suggests that fibrosis is part of the same disease process that results in hypertrophy in patients with HCM.

However, what are the clinical consequences of fibrosis detected by CMR? In this regard, LGE is associated with a number of unfavorable disease consequences in patients with HCM. An inverse relation is evident between the extent of LGE and LV ejection fraction, with substantially more fibrosis in patients in the end-stage phase of HCM than in patients with normal (or hyperdynamic) systolic function. Therefore, LGE seems to be associated with adverse LV remodeling with systolic dysfunction. In addition, a number of cross-sectional clinical studies have demonstrated that patients with HCM with LGE are more likely to have ventricular tachyarrhythmias on ambulatory Holter monitoring compared with patients without LGE, with up to a 7-fold increase in risk of NSVT. The totality of these CMR-based observations provide the first compelling evidence linking the in vivo identification of myocardial fibrosis to a number of adverse disease complications in patients with HCM, including ventricular tachyarrhythmias and systolic dysfunction.

Although CMR represents a powerful and exciting imaging technique, which will undoubtedly continue to contribute to our understanding the relation between HCM disease pathophysiology and clinical outcome, a number of important practical issues prohibit its widespread use in the assessment of all patients with HCM in routine clinical practice. A CMR unit is associated with substantial cost, requires highly skilled personal for both image acquisition and interpretation, and is not portable. In addition, select patients have contraindications to CMR imaging, such as those with previous device implantation and claustrophobia, and the administration of gadolinium is not recommended in patients with advanced renal failure. As a result, CMR will remain largely confined to select centers and unlikely to undergo widespread dissemination in the offices of most practicing cardiologists in North America and the rest of the world.

In addition, certain technical limitations may also affect the ability of CMR to provide an assessment of the underlying abnormal myocardial substrate in all patients. For example, the limited spatial resolution of CMR may not allow for the identification of small, diffuse amounts of fibrosis, particularly among patients with HCM in whom phenotypic expression is limited to minimal hypertrophy. The clinical significance of fibrosis not seen by contrast-enhanced CMR currently remains uncertain. However, a recent case report of a young patient with HCM who died suddenly with no visible LGE on contrast-enhanced CMR, but with a small, focal area of replacement scarring present in the apical portions of both the anterolateral and posteromedial papillary muscles on postmortem examination, raises the possibility that even a small amount of scar may provide a nidus for the generation of lethal arrhythmias in HCM. These limitations to CMR underscore the need for higher resolution scar imaging or additional imaging techniques that could provide a more sensitive assessment of the presence of underlying abnormalities in myocardial function in patients with HCM.

In this regard, over the last decade a number of echocardiographic-Doppler methods have been used in patients with HCM to assess for evidence of global and regional myocardial dysfunction despite the presence of normal (or hyperdynamic) systolic function. Tissue Doppler imaging measures the velocity of myocardial tissue motion in systole and diastole. Abnormalities in tissue Doppler velocities are observed frequently in HCM and even seen in those with HCM before the development of hypertrophy. However, strain imaging has technical limitations because of its angle dependence and need for optimal alignment for Doppler measurements.

A novel, non-Doppler technique of speckle-tracking imaging recently has emerged, which directly assesses myocardial motion by tracking speckles in the ultrasonic image from 2 dimensions. This technique is easier to perform compared with tissue Doppler, because images are angle independent and require only 1 cardiac cycle (with processing and interpretation done with offline analysis software) with good reproducibility. Previous observations using 2D strain have demonstrated reductions in global and regional systolic function (particularly in the longitudinal orientation) in patients with HCM, despite a normal ejection fraction as assessed by standard methods of LV quantification. These observations seem to suggest that strain imaging may be detecting subtle abnormalities in myocardial function because of structural changes in the underlying myocardial substrate. However, what are alterations in strain imaging really representing and is there a role for strain imaging in the prognostic assessment of patients with HCM?

In this issue of the Journal of the American Society of Echocardiography , Di Salvo et al apply 2D strain to evaluate regional systolic functional abnormalities to identify patients with HCM at increased risk of ventricular tachyarrhythmia. Among a sizeable population of patients with HCM with normal systolic function, ambulatory evidence of NSVT occurred more frequently in those with significantly reduced peak longitudinal systolic strain. In addition, multivariate analysis demonstrated that > 3 LV segments with a reduction in longitudinal systolic strain of ≥ −10% represented an independent predictor of NSVT with a high sensitivity and specificity for detection of ambulatory ventricular tachyarrhythmias demonstrated by receiver operating characteristic analysis. As a result, the authors conclude that these findings may have implications for the risk stratification of patients with HCM, because NSVT is generally regarded to be an independent determinant of increased risk of sudden death in this disease.

This is an important investigation that expands our understanding of the clinical significance of echocardiographically derived measures of myocardial function. Indeed, the results of the present investigation parallel those of the previously mentioned CMR studies demonstrating a similar association between LGE (ie, fibrosis) and ventricular tachyarrythmias. This observation seems to suggest that abnormalities in strain may represent a surrogate of myocardial fibrosis. In this regard, previous investigations using strain imaging have demonstrated greater reductions in global strain among HCM patients with fibrosis (as detected by CMR) compared to those without. As well, LV segmental strain seems reduced in myocardial segments in which fibrosis is present compared with segments without. Tagged CMR studies have also confirmed substantially impaired myocardial function in segments with LGE compared with regions without, suggesting that myocardial fibrosis may be a significant contributor to the pathophysiology of regional dysfunction in patients with HCM.

Undoubtedly, in a heterogeneous disease such as HCM, a number of additional factors beyond just fibrosis likely contribute to the complex interactions that can result in abnormal strain. For example, the interplay between other HCM disease features, such as LV outflow tract obstruction and microvascular ischemia, as well as those of modifiable factors such as hypertension, obesity, and epicardial coronary artery disease, may also be expected to have an impact on myocardial strain. Ultimately, it may not be critical to completely understand the underlying mechanisms responsible for these reductions in myocardial strain, but rather to demonstrate, with well-designed studies in large cohorts with HCM, that these abnormalities are independently associated with an increased risk in this disease. In this regard, Di Salvo et al’s study has brought us one step closer to realizing this goal with 2-dimensional strain imaging.

However, if we are to begin to consider measures of myocardial strain as an emerging marker of risk based on the association with NSVT, a number of important issues remain. Although there was a statistically significant difference in peak systolic longitudinal strain in patients with HCM with NSVT compared with those without, the individual values demonstrated significant overlap. This issue makes it less clear how well these measures can be translated easily to clinical practice to identify patients at risk for potentially lethal ventricular tachyarrhythmias. For this reason, it was necessary for the authors to express the relationship between strain and NSVT with an index that included more than 3 LV segments with abnormal peak systolic longitudinal strain. In addition, the cutoff values used to define abnormal strain were arbitrary, based on mean values within the HCM study cohort. Therefore, as the authors suggest, more work is necessary to clarify the precise number of LV segments necessary to define increased risk and more robust cutoff values for abnormal strain that will ultimately perform best in predicting outcome.

In addition, in the current study a significant relationship was also evident between maximal LV wall thickness and NSVT, with an area under the receiver operating characteristic curve similar to that of longitudinal systolic strain and NSVT. This observation is not necessarily surprising, particularly because strain and myocardial fibrosis have been associated with extent of LV hypertrophy. Therefore, these results raise the important issue of whether measures of longitudinal strain in fact provide greater accuracy in identifying patients with HCM at risk compared with maximal LV wall thickness alone. Previous echocardiographic studies in patients with HCM have demonstrated maximal LV wall thickness (≥30 mm) to be an independent risk factor for sudden death in this disease, and therefore management decisions with regard to ICD implantation are often decided on the basis of measures of wall thickness. Therefore, before we can move ahead to consider strain as a marker of risk, future studies are required to help clarify whether wall thickness and fibrosis are truly independent markers with regard to predicting future risk of serious ventricular tachyarrhythmias. Finally, 2D strain imaging is technically limited by issues related to through-plane motion in which certain LV segments are difficult to track from diastole to systole. In this regard, it is conceivable that even newer speckle-based echocardiographic techniques using three-dimensional strain may overcome this limitation and represent an even more robust technique for assessment of myocardial function.

Nevertheless, can 2D strain imaging be directly applied to the management strategies of patients with HCM? Whether the identification of abnormalities in myocardial function by this technique can identify patients with HCM at risk of adverse consequences, such as sudden death, remains uncertain. In this regard, the current study design used NSVT as a surrogate marker for sudden death. This is a common strategy for most short-term clinical studies in HCM because the event rate for this disease is low, often necessitating the use of already established risk markers to suggest a possible link to increased risk of sudden death. Therefore, it would be premature at this early juncture to make management decisions about ICD therapy for primary prevention solely on the basis of the results of echocardiographic strain studies. However, these data should be a call to arms to motivate further investigation with a larger, more diverse patient population with HCM studied over a longer follow-up period to determine the precise association between abnormalities in 2D echocardiographic strain and sudden death. The results of such a study will help assure us of how best to extrapolate strain data to the broader clinically identified population with HCM in an effort to improve risk stratification. Because of the low event rate, a large number of patients are necessary to achieve the statistical robustness necessary; therefore, a focused effort directed at organizing a multicenter, prospective study to evaluate the independent prognostic value of echocardiographic strain imaging would seem to be a reasonable next step.

During the last half decade, cardiovascular imaging techniques have contributed to many of the great advances in our understanding of HCM. The data by Di Salvo et al provide additional compelling support for a continued march forward in a quest to understand how best to integrate and apply various imaging techniques to better serve the HCM patient community. In fact, HCM may represent one of the strongest cases today for an increasing role for multimodality imaging in providing a comprehensive approach to the assessment of patients with this disease. Indeed, with the emergence of a number of advancements in cardiovascular imaging techniques, the future is bright for the cardiovascular imager in HCM.

Conflicts of interest: none.

Editorial Comments published in the Journal of the American Society of Echocardiography (JASE) reflect the opinions of their author(s), and do not necessarily represent the views of JASE, its editors, or the American Society of Echocardiography.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jun 16, 2018 | Posted by in CARDIOLOGY | Comments Off on Risk Stratification in Hypertrophic Cardiomyopathy: Is Two-Dimensional Echocardiographic Strain Ready for Prime Time?

Full access? Get Clinical Tree

Get Clinical Tree app for offline access