Hypertrophic cardiomyopathy (HCM) is an inherited condition with a diverse clinical course, ranging from stable active individuals to patients incapacitated by symptoms and/or arrhythmias. Early reports and investigations on HCM focused on the presence of subaortic obstruction, characterized by a systolic murmur that changed with bedside maneuvers and by a pressure gradient in the left ventricular (LV) outflow tract (LVOT) noted in the cardiac catheterization laboratory that varied with pharmacologic or physiologic provocation. Advances in echocardiography have greatly enhanced both our recognition and our understanding of the different manifestations of HCM. The pathophysiology of exercise intolerance includes dynamic LVOT obstruction, mitral regurgitation, diastolic dysfunction, myocardial ischemia, and autonomic dysfunction. Echocardiography has helped elucidate numerous mechanisms for these abnormalities and is the primary imaging modality used in the screening, diagnosis, and intraprocedural and long-term monitoring of patients with HCM.
Risk Assessment in Patients with Hypertrophic Cardiomyopathy
The greater challenge has been the emerging role of echocardiography in detecting patients with HCM who are at high risk for serious cardiovascular events, including sudden cardiac death (SCD). A great number of studies have examined multiple risk factors for SCD in patients with HCM in an attempt to improve risk assessment for a condition that affects 1 in 500 individuals, has an incidence of SCD of approximately 1% per year, and may occasionally present with SCD in the absence of preceding symptoms. Risk stratification in HCM is not straightforward, because the major risk factors for SCD in HCM generally have low positive predictive value. In the management of individual patients, there can be ambiguity in the interpretation of their family or personal histories and the results of the patients’ investigations. Although implantable cardioverter-defibrillators (ICDs) are effective for the primary prevention of SCD, patients face decades of potential complications from ICDs (4% per year), including a significant risk for inappropriate ICD discharges.
Current guidelines from the American College of Cardiology Foundation and American Heart Association consider prophylactic ICD therapy to be a reasonable recommendation in the presence of one of the following clinical or echocardiographic markers: (1) a family history of SCD in one or more first-degree relatives, (2) recent unexplained syncope, and (3) the presence of a maximal LV wall thickness ≥ 30 mm. Other strategies to alter the natural history of HCM, prevent HCM-related events (heart failure, stroke, atrial and ventricular arrhythmias), and treat symptoms include pharmacotherapy and invasive septal reduction therapy (surgical myectomy or septal ethanol ablation). Contemporary studies with baseline echocardiographic and clinical data have helped identify patients at higher risk for HCM-related events. There are three echocardiographic findings associated with an increased risk for HCM-related death: (1) LV wall thickness ≥ 30 mm, (2) LVOT gradient ≥ 30 mm Hg at rest, and (3) left atrial (LA) enlargement. The presence of massive LV hypertrophy (i.e., wall thickness ≥ 30 mm) is the strongest echocardiographic risk marker associated with SCD. Resting LVOT obstruction, defined by the traditional threshold of an LVOT gradient ≥30 mm Hg at rest, has also been shown to significantly increase the risk for HCM-related death and SCD. Within the subset of patients with resting LVOT obstruction, two large studies have demonstrated that a higher resting LVOT gradient (categorized by an LVOT velocity > 4 m/sec or an LVOT gradient ≥ 64 mm Hg) was an independent predictor of HCM-related death. The development of an anteroposterior LA diameter > 48 mm was associated with an increased risk for all-cause, HCM-related, and heart failure–related death (but not SCD). More recent echocardiographic studies with smaller HCM cohorts and shorter observation periods have explored additional Doppler echocardiographic measures, LA volume indexed to body surface area, and Doppler tissue imaging, to further refine the risk assessment of patients. The detection of an LA volume index > 27 mL/m 2 , a Doppler tissue imaging lateral annular systolic velocity < 4 cm/sec, and a septal E/e′ ratio > 15 impart an increased risk for HCM-related events.
In the present issue of JASE , Reant et al . report on their analysis of a multitude of resting and exercise echocardiographic parameters to better predict HCM-related outcomes. This group of experienced investigators prospectively evaluated 115 patients with HCM. All patients had baseline comprehensive resting Doppler echocardiographic studies, including two-dimensional speckle-tracking echocardiography (STE) for the assessment of global longitudinal strain (GLS). Exercise echocardiography was performed with a bicycle in a semisupine position (50°) with a slight left lateral tilt to allow simultaneous echocardiographic imaging during cycling. Outcomes of interest were a composite cardiac end point of HCM-related death (or an event equivalent to death, such as an appropriate discharge from an ICD) or progression to New York Heart Association class III or IV symptoms. In their regression models, GLS was dichotomized at an absolute GLS level of 15%. This value was selected on the basis of data from Yingchoncharoen et al . that showed worse outcomes in patients with GLS ≤ 15% and asymptomatic aortic stenosis, another condition with pathologic LV hypertrophy. Eighteen patients reached the cardiac end point during the mean follow-up period of 19 months. Cox modeling showed two independent determinants of the composite end point, which conferred a greater than threefold increased risk: GLS ≤ 15% and peak exercise-induced LVOT gradient ≥50 mm Hg. Important additional findings from Kaplan-Meier survival analyses were the following: (1) patients with GLS > 15% and exercise LVOT gradients < 50 mm Hg had the best prognosis, and (2) within the group of patients with GLS > 15%, patients with exercise LVOT gradients ≥ 50 mm Hg did worse than those with exercise LVOT gradients < 50 mm Hg. Finally, the degree of LVOT obstruction significantly influenced outcomes, especially in patients with baseline resting LVOT obstruction: patients with resting LVOT obstruction and higher exercise LVOT gradients (≥50 mm Hg) were significantly more likely to reach the composite end point than those with resting LVOT obstruction and lower exercise LVOT gradients (<50 mm Hg). On the basis of their findings, the authors assert that GLS and exercise echocardiography provide incremental information and should be obtained on a routine basis in patients with HCM.
The major contributions of this study are to underscore the use of longitudinal strain and the degree of exercise-induced LVOT obstruction to predict outcomes in HCM. However, as acknowledged by the authors, the major limitations of this study are the low number of events ( n = 18) and the short follow-up time. The largest proportion of cardiac “events” was the deterioration to New York Heart Association class III or IV symptoms in 10 patients. The remaining eight events were one death due to heart failure, four episodes of sustained ventricular tachycardia, and three hospitalizations for heart failure. In examining the 18 patients in the group with cardiac events (and comparing them with the 97 patients who remained free of events), these 18 patients had features on their baseline echocardiographic evaluations that suggested significant LVOT obstruction: 67% (vs 31% in the event-free group) had resting LVOT gradients ≥ 30 mm Hg, 61% (vs 24%) had exercise LVOT gradients ≥ 50 mm Hg, the average resting LVOT gradient was 52 ± 39 mm Hg (vs 27 ± 31 mm Hg), and the average exercise LVOT gradient was 76 ± 55 mm Hg (vs 40 ± 40 mm Hg). Also, LVOT gradients were recorded while patients were on medications (mainly β-blockers), which suggests suboptimal control of their resting and provocable LVOT gradients despite patients receiving pharmacotherapy. Thus it may not be altogether surprising that these patients experienced declines in functional status during follow-up.
Myocardial Mechanics in Hypertrophic Cardiomyopathy: What Do You Gain with Strain?
Two-dimensional STE is an advanced echocardiographic technique for the quantification of global and regional myocardial function. Multidirectional myocardial deformation (longitudinal, circumferential, and radial strain) and rotational mechanics (apical rotation and torsion) have been reported in multiple cardiac disorders. Strain imaging has provided greater insight into myocardial mechanics, myofiber architecture within the left ventricle, and the determinants of systolic ejection. Longitudinal deformation of the left ventricle is predominantly affected by the subendocardial region and is most sensitive to the presence of myocardial pathology. Midmyocardial and epicardial function contribute to rotational motion and may not be disturbed in early disease states. Consequently, circumferential strain and twist may be normal or may exhibit exaggerated compensation to maintain LV systolic performance.
Interest has focused on the ability of STE to detect early myocardial dysfunction before an overt drop in the LV ejection fraction. The role of STE in recognizing subclinical LV dysfunction has been especially relevant in patients receiving potentially cardiotoxic chemotherapeutic agents but is also pertinent in patients with hypertrophic ventricles because aortic stenosis and HCM can both progress to decompensated LV systolic dysfunction. There is increasing evidence that longitudinal strain is valuable in ascertaining prognosis: a recent systematic review and meta-analysis showed that GLS was a stronger predictor of adverse cardiac events than the LV ejection fraction in multiple cardiac conditions. The most robust and widely studied parameter of myocardial deformation appears to be GLS. Measures of global circumferential and radial strain have been less consistent and have lower reproducibility.
The feasibility of STE to investigate patients with HCM was first reported by the same center as the present study by Reant et al . : the authors evaluated 26 patients with nonobstructive HCM and normal LV systolic function. Compared with healthy subjects, average longitudinal, transverse, circumferential, and radial strain were reduced in patients with HCM. Other studies of myocardial deformation in HCM cohorts have revealed differences in various strain parameters compared with controls. In a cohort of 72 patients with nonobstructive and obstructive HCM, patients with the obstructive form of HCM had lower longitudinal strain and higher circumferential strain than patients with nonobstructive HCM. Furthermore, there is a significant correlation between longitudinal strain and the degree of myocardial fibrosis seen on cardiac magnetic resonance imaging. The available data suggest that lower longitudinal strain reflects a diseased ventricle in HCM. However, our current understanding of the overall myocardial mechanics in HCM is far from complete. In detailing the effects of surgical abolition of LVOT obstruction, data from our center showed that the low preoperative longitudinal strain of patients with obstructive HCM did not normalize after surgical myectomy. Nonetheless, what was observed was a significant diminution in the abnormally high preoperative circumferential strain and apical rotation. The reduction in circumferential strain and apical rotation most likely denote a decrease in compensatory mechanisms that preserve LV ejection. These observations suggest that longitudinal mechanics in patients with HCM are not mainly attributable to pressure overload but are due to the underlying cardiomyopathy, with its intrinsic sarcomere dysfunction, myocyte disarray, and myocardial fibrosis. Finally, the derangements in systolic deformation are likely not responsible for the diastolic heart failure that may occur in HCM. Two-dimensional strain imaging has extended our understanding of diastolic mechanics and has shown a significant delay in untwisting, which has been correlated with LV filling pressure.
Beyond the intricacies of myocardial mechanics that that are specific to HCM, the major issue with the use of GLS as a prognostic marker relates to technical considerations surrounding strain imaging. There are inherent differences in the methodology of the assessment of strain among different vendors and software systems, which produce heterogeneous values for strain. Sources of variation stem from disparities in acquisition (e.g., frame rate), postprocessing, and the hemodynamic load. The results of a meta-analysis of >2,500 normal subjects determined that blood pressure was independently associated with the variation in normal values of GLS. Given the above concerns, there is an ongoing initiative among the American Society of Echocardiography, the European Association of Cardiovascular Imaging, and hardware and software vendors to reduce intervendor variability and to standardize strain measurements.
Myocardial Mechanics in Hypertrophic Cardiomyopathy: What Do You Gain with Strain?
Two-dimensional STE is an advanced echocardiographic technique for the quantification of global and regional myocardial function. Multidirectional myocardial deformation (longitudinal, circumferential, and radial strain) and rotational mechanics (apical rotation and torsion) have been reported in multiple cardiac disorders. Strain imaging has provided greater insight into myocardial mechanics, myofiber architecture within the left ventricle, and the determinants of systolic ejection. Longitudinal deformation of the left ventricle is predominantly affected by the subendocardial region and is most sensitive to the presence of myocardial pathology. Midmyocardial and epicardial function contribute to rotational motion and may not be disturbed in early disease states. Consequently, circumferential strain and twist may be normal or may exhibit exaggerated compensation to maintain LV systolic performance.
Interest has focused on the ability of STE to detect early myocardial dysfunction before an overt drop in the LV ejection fraction. The role of STE in recognizing subclinical LV dysfunction has been especially relevant in patients receiving potentially cardiotoxic chemotherapeutic agents but is also pertinent in patients with hypertrophic ventricles because aortic stenosis and HCM can both progress to decompensated LV systolic dysfunction. There is increasing evidence that longitudinal strain is valuable in ascertaining prognosis: a recent systematic review and meta-analysis showed that GLS was a stronger predictor of adverse cardiac events than the LV ejection fraction in multiple cardiac conditions. The most robust and widely studied parameter of myocardial deformation appears to be GLS. Measures of global circumferential and radial strain have been less consistent and have lower reproducibility.
The feasibility of STE to investigate patients with HCM was first reported by the same center as the present study by Reant et al . : the authors evaluated 26 patients with nonobstructive HCM and normal LV systolic function. Compared with healthy subjects, average longitudinal, transverse, circumferential, and radial strain were reduced in patients with HCM. Other studies of myocardial deformation in HCM cohorts have revealed differences in various strain parameters compared with controls. In a cohort of 72 patients with nonobstructive and obstructive HCM, patients with the obstructive form of HCM had lower longitudinal strain and higher circumferential strain than patients with nonobstructive HCM. Furthermore, there is a significant correlation between longitudinal strain and the degree of myocardial fibrosis seen on cardiac magnetic resonance imaging. The available data suggest that lower longitudinal strain reflects a diseased ventricle in HCM. However, our current understanding of the overall myocardial mechanics in HCM is far from complete. In detailing the effects of surgical abolition of LVOT obstruction, data from our center showed that the low preoperative longitudinal strain of patients with obstructive HCM did not normalize after surgical myectomy. Nonetheless, what was observed was a significant diminution in the abnormally high preoperative circumferential strain and apical rotation. The reduction in circumferential strain and apical rotation most likely denote a decrease in compensatory mechanisms that preserve LV ejection. These observations suggest that longitudinal mechanics in patients with HCM are not mainly attributable to pressure overload but are due to the underlying cardiomyopathy, with its intrinsic sarcomere dysfunction, myocyte disarray, and myocardial fibrosis. Finally, the derangements in systolic deformation are likely not responsible for the diastolic heart failure that may occur in HCM. Two-dimensional strain imaging has extended our understanding of diastolic mechanics and has shown a significant delay in untwisting, which has been correlated with LV filling pressure.
Beyond the intricacies of myocardial mechanics that that are specific to HCM, the major issue with the use of GLS as a prognostic marker relates to technical considerations surrounding strain imaging. There are inherent differences in the methodology of the assessment of strain among different vendors and software systems, which produce heterogeneous values for strain. Sources of variation stem from disparities in acquisition (e.g., frame rate), postprocessing, and the hemodynamic load. The results of a meta-analysis of >2,500 normal subjects determined that blood pressure was independently associated with the variation in normal values of GLS. Given the above concerns, there is an ongoing initiative among the American Society of Echocardiography, the European Association of Cardiovascular Imaging, and hardware and software vendors to reduce intervendor variability and to standardize strain measurements.
Methods of Inducing Left Ventricular Outflow Tract Obstruction in Hypertrophic Cardiomyopathy
The study by Reant et al . also addressed the implications of exercise-induced LVOT obstruction. Dynamic LVOT obstruction is common in patients with HCM: one-third have resting LVOT obstruction (resting LVOT gradient ≥ 30 mm Hg), one-third have provocable/latent LVOT obstruction (resting LVOT gradient < 30 mm Hg, provocable LVOT gradient ≥30 mm Hg), and one-third have nonobstructive HCM (both resting and provocable LVOT gradient < 30 mm Hg). The conventional threshold for consideration of invasive septal reduction therapy has been the presence of class III or IV symptoms and a resting or provocable LVOT gradient ≥50 mm Hg. Echocardiography is the preferred diagnostic test to determine the hemodynamic status of patients, which is essential to differentiate patients with obstructive HCM (resting/provocable) from those with nonobstructive HCM and to guide treatment. The symptoms attributable to LVOT obstruction (dyspnea, angina, exertional presyncope or syncope) can be very successfully abolished, whereas patients with nonobstructive HCM and symptoms secondary to diastolic dysfunction may be very challenging to manage. Long-term data from our own center have demonstrated excellent outcomes in patients treated with pharmacotherapy (β-blockers and/or disopyramide) or invasive septal reduction therapy.
In patients with symptoms and resting LVOT gradients < 30 mm Hg, it is crucial to perform provocative maneuvers to document the presence of a provocable LVOT gradient. Obstruction can be provoked by the following approaches: (1) reduction in preload (strain phase of the Valsalva maneuver, standing), (2) reduction in afterload (inhalation of amyl nitrate, sublingual isosorbide dinitrate), and (3) increase in contractility and heart rate (exercise, isoproterenol). The impetus for developing methods of provocation other than exercise stemmed from concerns regarding the safety of exercise testing in the HCM population. The most common method of provocation used in echocardiography laboratories to evaluate patients with HCM is the Valsalva maneuver. However, the Valsalva maneuver underestimates the magnitude of the provocable LVOT gradient compared with exercise testing. In addition, the Valsalva maneuver has low sensitivity (40%) for unmasking significant LVOT obstruction compared with provocation with exercise. Other drawbacks include the fact that a significant number of patients have difficulty performing a Valsalva maneuver, and it can interfere with recording an adequate LVOT signal, particularly when there is excessive movement of the chest wall with straining. Other methods of provocation (standing, amyl nitrite, sublingual administration of isosorbide dinitrate) have also been found to be inferior to exercise, with lower provocable LVOT gradients achieved and a decreased likelihood of inducing a provocable LVOT gradient than with exercise ( Table 1 ). Amyl nitrite is a very useful alternative to exercise, but its availability is limited. The use of dobutamine is discouraged, except at experienced centers, because this drug can cause a cavity obliteration signal that can be confused with a legitimate increased LVOT velocity from mitral leaflet–septal contact.
