2020 年 84 巻 8 号 p. 1218-1226
Hypertrophic cardiomyopathy (HCM) is the most frequent hereditary cardiomyopathy, showing an autosomal-dominant f inheritance. A great deal of attention has been paid to genetics, left ventricular tract obstruction and the prediction and prevention of sudden cardiac death in HCM. Needless to say, these are very important, but we should recognize the heterogeneity in etiology, morphology, clinical course and management of this unique cardiomyopathy. Another important perspective is that HCM causes left ventricular remodeling over time and is a disease that requires lifelong management in the real world.
Hypertrophic cardiomyopathy (HCM) is the most frequent hereditary cardiomyopathy, showing an autosomal-dominant inheritance.1,2 At least 1 in 500 people suffer from this condition, and a recent report taking into account genetic background and new information from cardiac magnetic resonance imaging (CMR) gave a figure of about 1 in 200,1 making HCM a common disease that will be often encountered in daily clinical practice. The understanding of HCM has progressed with advances in genetic testing, imaging and management. This review focuses on those advances and the issues that remain, incorporating recent knowledge with our experience.3
Morphological definitions of HCM are broadly similar between the European Society of Cardiology (ESC),4 the American College of Cardiology (ACC)/American Heart Association (AHA),5 and the Japanese Circulation Society (JCS)/Japanese Heart Failure Society.6 That is, HCM is defined by a maximum left ventricular (LV) wall thickness ≥15 mm (≥13 mm in the presence of a family history of HCM) by echocardiography or CMR. However, etiological definitions differ among these groups. The ACC/AHA defines HCM as “a disease state characterized by unexplained LV hypertrophy associated with nondilated ventricular chambers in the absence of another cardiac or systemic disease.”5 In contrast, the ESC defines it as “the presence of increased LV wall thickness that is not solely explained by abnormal loading conditions.”4
There are 2 problems with the definition and diagnosis of HCM. As described later, HCM has been considered a “sarcomere disease” based on a sarcomere genetic mutation.1,2,7,8 However, even when clinically diagnosed with HCM, mutations were identified in 32% of 2,912 probands.7 This gap represents a problem that must be solved when considering the heterogeneity of HCM.
Another problem is the HCM phenocopies (i.e., secondary cardiomyopathy). The natural history and treatment differ completely between HCM caused by a sarcomere mutation and secondary cardiomyopathies such as Anderson-Fabry disease9 and cardiac amyloidosis.10 Disease-specific treatments, such as enzyme replacement therapy for Anderson-Fabry disease9 and tafamidis and patisiran for cardiac amyloidosis,10 are progressing rapidly. Therefore, it is vitally important to accurately differentiate secondary cardiomyopathy similar in form to HCM in order to provide appropriate treatment.
The Japanese guidelines emphasize that after careful consideration of the genetic background, a diagnosis of HCM can be made after carefully excluding secondary cardiomyopathy as much as possible (Figure 1).6
Definition of Cardiomyopathies by JCS/JHFA Guidelines. ARVC, arrhythmogenic right ventricular cardiomyopathy; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; HHD, hypertensive heart disease; IHF, ischemic heart disease; PPCM, peripartum cardiomyopathy; RCM, restrictive cardiomyopathy. (From Tsutsui et al6 with translation and permission.)
Since the first report of a β-myosin heavy chain (MYH7) mutation in a large HCM family, more than 2,000 mutations encoding proteins of thick and thin myofilament contractile components of the cardiac sarcomere or Z disk have been reported. Among these, 8 genetic mutations encoding the constituent proteins of the sarcomere are the most common and the concept of HCM as a “sarcomere disease” has thus become established.1,2,7,8 In particular, mutations of MYH7 and myosin binding protein C (MYBPC3) account for approximately 70% of total mutations. Interestingly, Inagaki et al reported that in their analysis of 34 cases with mid-ventricular obstruction (MVO), 21% displayed HCM-related sarcomere mutations and 18% had dilated cardiomyopathy (DCM)/arrhythmogenic right ventricular cardiomyopathy (ARVC)-related mutations.11
Whether or not the type of genetic mutation relates to phenotype and prognosis has been studied with great interest. Early reports found that some mutations of MYH7 were associated with sudden cardiac death (SCD) and MYBPC3 mutations in older patients had a relatively good clinical course, whereas TNNT2 mutations, despite associations with a mild degree of hypertrophy, tended to evolve into systolic dysfunction and poor prognosis. However, even with the same gene among family members, different morphologies, clinical symptoms and prognoses have been observed. Because the phenotype of HCM involves not only genetic mutations, but also many other factors such as modifier genes and environmental factors,2 estimating the phenotype from the genetic mutation is currently considered difficult.12 Similarly, it is generally thought that prognosis (especially SCD) is difficult to predict based solely on the results of genetic testing.4–6
It is necessary to the understanding and compliance with related laws and guidelines, such as thosefor genetic test and genetic counselling in cardiovascular disease (JCS 2011), when performing genetic testing.13
Clinical Markers of Positive Genetic MutationsThe Hypertrophic Cardiomyopathy Registry is a registry of 2,755 patients with HCM who systematically underwent CMR, genetic testing, and biomarker measurements.14 LV morphology by CMR was divided into 6 subtypes as shown in Figure 2. For these subtypes A–F, the frequencies were 46%, 38%, 8%, 1%, 3% and 9%, respectively. Among these morphological subtypes, reverse curvature septal hypertrophy is more often seen in patients with sarcomere mutation,14,15 even in elderly patients.16
Steady-state free precession 4-chamber long-axis cine images from individual patients with the 6 different morphological subtypes of hypertrophic cardiomyopathy (HCM). The 6 subtypes are: (A) localized basal septal hypertrophy; (B) reverse curvature septal hypertrophy; (C) apical HCM; (D) concentric HCM; (E) mid-cavity obstruction with apical aneurysm; or (F) other. (From Neubauer et al14 with permission.)
The Mayo clinical phenotype-based genotype predictor score was devised to evaluate positive mutation by 6 clinical findings (Figure 3).17 Validation studies, including our patient cohort, show this score is useful for predicting the presence of genetic mutation (Figure 3).18,19
Clinical phenotype-based genotype predictor score. Dx, diagnosis; HCM, hypertrophic cardiomyopathy; Hx, history; MLVWT, maximum left ventricular wall thickness; SCD, sudden cardiac death.
The prevalence of secondary cardiomyopathy is thought to be 5–10% in patients with clinically suspected HCM.4,6 Glycogen storage disease such as Pompe disease and Danon disease, lysosomal diseases such as Anderson-Fabry disease, mitochondrial diseases, and cardiac amyloidosis are important differential diagnoses.
Anderson-Fabry disease has an X-linked inheritance caused by genetic deficiency or decreased activity of α-galactosidase (α-Gal A), one of the lysosomal hydrolases, based on mutations in GLA on the long arm of the X chromosome (Xq22.1). Currently, 1–3% of patients with clinically suspected HCM are considered to have Fabry disease.9
Amyloidosis is a disease in which amyloid fibrils are deposited extracellularly. The onset of cardiac amyloidosis mostly involves immunoglobulin light chain (AL) amyloidosis and transthyretin (TTR) amyloidosis. Moreover, ATTR is caused by hereditary ATTR amyloidosis (ATTRv, traditionally known as familial amyloid polyneuropathy) caused by TTR mutation and wild-type ATTR amyloidosis (ATTRwt, traditionally known as senile systemic amyloidosis) caused by the deposition of wild-type TTR without TTR mutation.10 In recent years, advances in noninvasive testing modalities such as99 m Tc pyrophosphate scintigraphy and CMR have revealed that ATTRwt is more frequent than previously thought and some patients were previously misdiagnosed as HCM.20
HCM With Unidentifiable Sarcomere Genetic MutationsOnly 30–50% of patients with suspected HCM have been identified as showing a sarcomere genetic mutation, with the cause unknown in the remaining cases. Histologic findings in patients with and without myofilament mutations have been reported as similar, except for severity of myocyte hypertrophy, but those patients may show different characteristics in some respects such as age, sex, comorbidities, morphology and disease severity compared with HCM patients with sarcomere genetic mutation.21 The etiology of HCM in patients without a sarcomere genetic mutation thus remains unresolved and is an important issue.
Distribution Asymmetric septal hypertrophy has been reported as characteristic of HCM. However, as mentioned earlier, recent advances in diagnostic imaging, especially with CMR, have provided new information on morphology. Today, HCM is thought to exhibit various patterns of hypertrophy affecting ≥1 LV segments.1,2,14
Apical HCM was first described in Japan by Sakamoto and Yamaguchi and its prevalence in Japan is high compared with other Western countries.22 Apical HCM is generally classified as having hypertrophy localized to the apex, but should be classified into 1 subtypes: 1 in which the hypertrophy is limited only to the apex without midsegment involvement (pure apical type), and the other in which hypertrophy extends from the apex to the midsegment of the LV (mixed type/distal dominant type).23,24 These subtypes have different genetic background and prognosis (Figure 4).23
Two distinct types of “apical” hypertrophic cardiomyopathy. IVS, interventricular septum; PM, papillary muscle. (From Kubo et al23 with permission.)
Thickness Maximal LV wall thickness as assessed by echocardiography and CMR is related to adverse cardiac events. Extreme LV hypertrophy (wall thickness >30 mm) is an independent predictor for SCD in HCM and is an indication for implantable cardioverter defibrillator (ICD) to prevent SCD.1,5,25 Patients with maximal LV wall thickness >30 mm show a substantial long-term risk for SCD, with rates of 20% at 10 years and 40% at 20 years.26 However, it should be noted that the negative predictive value of LV wall thickness is limited because most SCD cases occur in patients with LV wall thickness <30 mm. Conversely, a decrease in wall thickness over time suggests an evolution of LV systolic dysfunction (so-called “dilated phase of HCM [D-HCM]” in Japan or the endstage phase of HCM), and this phenomenon also represents a risk factor for poor outcome in HCM.27,28
LV Outflow Tract (LVOT) Obstruction and Mitral Valve ApparatusLVOT obstruction is present at rest in 20–25% of HCM patients, but it is a dynamic obstruction and the pressure gradient is easily changed by various factors. In fact, 70% of HCM patients show LVOT obstruction at rest or with provocation such as the Valsalva maneuver or exercise.29 In most cases, LVOT obstruction causes systolic anterior motion of the mitral valve, with the anteriorly moving leaflets contacting the septum during systole. This causes the LV ejection flow stream to push the mitral valve leaflets upward, in effect thickening the interventricular septum via a drag effect, with abnormalities of the mitral valve complex such as anteriorly displaced papillary muscles and excess leaflets (particularly the anterior leaflet), and abnormal attachment of the chordae tendineae to the mitral leaflets.30 Recognition of abnormalities of the mitral complex is important when planning invasive treatments for LVOT obstruction.
Heart failure (HF) symptoms and clinical outcomes for patients with obstructive HCM, even when provoked by stress, are worse than for those with non-obstructive HCM.29,31 An LVOT gradient >30 mmHg at rest has been associated with increased risk of SCD. However, the absolute difference in the annual rate of SCD between patients with and without LVOT is small and LVOT obstruction shows a high negative-predictive value for SCD of 95%, but a low positive-predictive value of only 7%.32
LV Diastolic DysfunctionLV diastolic dysfunction is a hallmark of HCM, with the LV end-diastolic pressure easily elevated at rest and on exertion. However, noninvasive evaluation of diastolic function and filling pressure in HCM is difficult. The American Society of Echocardiography recommends a comprehensive approach using the E/e’ ratio, left atrial (LA) volume index, pulmonary vein atrial reversal velocity, and peak velocity of the tricuspid regurgitation jet by continuous-wave Doppler.33 Evaluation of diastolic dysfunction adds additional information for prognostication. Assessment using E/e’ has been reported as useful in predicting adverse outcomes, particularly HF in HCM patients.34
LV Systolic DysfunctionThe LV ejection fraction (LVEF) is generally used as parameter of LV systolic function. LVEF is a basically preserved in HCM and has limitations as a parameter of LV systolic function in HCM because LVEF is overestimated by a smaller LV volume, which directly affects the EF equation. Indeed, systolic velocity by tissue Doppler imaging and LV global longitudinal stain (GLS) are abnormal despite preserved EF, meaning that myocardial function is impaired despite preservation of LVEF in HCM.35,36 A systematic review of more than 3,000 HCM cases suggested associations between abnormal GLS and both adverse composite cardiac outcomes and ventricular arrhythmias.37
A significant proportion of HCM patients evolve to progressive LV systolic dysfunction, LV enlargement and thinning of the LV wall and suffering from refractory severe HF such as DCM.27,28 D-HCM, or endstage phase of HCM, is defined by LVEF <50%, but myocardial degeneration should be recognized as progressing even with preserved EF in the low-normal range. The exact prevalence of this condition is unclear, but 4 of 137 patients (3%) were already in LV systolic dysfunction at the time of diagnosis, and 10% developed LV systolic dysfunction during follow-up (11.4±5.7 years).38 Young age at diagnosis, severe LV wall hypertrophy, family history and genetic mutations have been reported as predictors of evolution to LV systolic dysfunction.
Late Gadolinium Enhancement (LGE) by CMRLGE on CMR is thought to indicate fibrosis in HCM. LGE is found in approximately 70% of patients with HCM, mainly in the thickened part of the LV wall and at the junction between the ventricular septum and the free wall of the right ventricle.
The relationship between LGE and prognosis has been attracting great interest. A study using quantitative contrast-enhanced MRI in 1,300 cases of HCM identified percentage LGE (%LGE) as an independent predictor of SCD, and %LGE ≥15% of total LV mass was associated with a 2-fold increase in SCD at 5 years compared with patients with no LGE.39 In 1423 consecutive low- or intermediate-risk patients with HCM and preserved LVEF, %LGE ≥15% was associated with a significantly higher rate of the composite endpoint of SCD and appropriate ICD discharge.40 Therefore, extensive LGE is considered a major risk factor for SCD in HCM.1,25
LV Apical AneurysmAs recognition of apical aneurysm is often difficult without CMR, this finding is under-recognized. Its prevalence by CMR has been reported as a few percent for overall HCM cases.41 Aneurysm formation is caused by increased cavity pressure, wall stress and ischemia, and has been thought of as a risk factor for SCD and thromboembolism. A study of apical HCM showed apical aneurysm was observed in 8.5% and the event rate was 10.1%/year in patients with LV apical aneurysm.42 The paradoxical diastolic flow from the apex towards the base on echocardiography or the disappearance of a giant negative T-wave on ECG suggests degeneration of the apical myocardium.
LA DiameterThe LA diameter as determined by M-mode or 2D echocardiography in the parasternal long-axis plane is one component of the ESC HCM Risk-SCD score.4 However, the results of other studies are not in agreement regarding LA diameter as a risk factor for SCD in HCM, because the LA diameter is affected by factors such as LV diastolic function, mitral regurgitation and atrial fibrillation.41
Relatively few biomarkers have been discussed, but can provide important information regarding diagnosis, management and prognosis in HCM.43
The pathophysiological roles and mechanisms of release of N-terminal pro-B-type natriuretic peptide (NT-proBNP) and BNP are not fully understood in HCM, but the levels of NT-proBNP/BNP have been associated with many clinical parameters such as LV hypertrophy, LVOT obstruction and diastolic dysfunction. The NT-proBNP/BNP level is not strongly related to NYHA class or exercise capacity. On the other hand, the association between NT-proBNP/BNP level and adverse cardiovascular events appears certain to some extent. Two large clinical studies have shown NT-proBNP/BNP as an independent predictor of morbidity and mortality. In one study, in comparison with patients with BNP level ≤98 pg/mL, the hazard ratio for death was 6.98 in patients with BNP level ≥298 pg/mL.44 Another study showed an overall event rate of 6.1% per year in patients with NT-proBNP ≥135 pmol/L and 0.4% per year in those with NT-proBNP ≤44 pmol/L, concluding that elevated NT-proBNP level represents a strong predictor of overall prognosis, particularly for HF-related death and transplantation in patients with HCM.45
Elevated troponin is seen in patients with HCM and the level of high-sensitivity cardiac troponin T (hs-cTnT) was elevated in 99 of 183 patients (54%) in our study.46 Although the exact mechanism remains unclear, elevation of troponin may result from microvascular ischemia or replacement fibrosis and the level of troponin relates to increased cardiovascular events in HCM. Patients with hs-cTnT elevated to ≥0.014 ng/mL have shown a 5-fold increase in the risk of combined adverse events and an 11-fold increase in the risk of HF at 4-year follow-up, and this risk increased with higher levels of hs-cTnT.46 BNP and troponin may have different prognostic significance, and a combined assessment of BNP and troponin may thus provide more useful information in the management of HCM (Figure 5).47
Utility of combined measurements of cTnI and BNP for predicting adverse outcomes in HCM. BNP, B-type natriuretic peptide; CI, confidence interval); cTnI, cardiac troponin I. (From Kubo et al47 with permission.)
HCM-related cardiac events include (1) SCD, (2) HF, and (3) cerebral embolism due to atrial fibrillation (AF).1–6 HCM-related mortality has been reported to be around 0.5–1.5%/year with contemporary management.1 However, HCM-related morbidity rates are never low48 (Figure 6) and HF in particular is becoming important as a residual unresolved issue.28,49
Rate of HCM-related deaths and adverse cardiovascular events in the Kochi cohort. HCM, hypertrophic cardiomyopathy. (From Kubo et al48 with permission.)
SCD is the most significant catastrophic form of HCM-related cardiovascular events. Pharmacotherapy is largely ineffective in preventing SCD in HCM, and therefore, implantation of an ICD is considered important as a preventative measure.1–6,25
The 2011ACC/AHA guidelines recommend consideration of ICD implantation based on the presence of ≥1 major risk factors (family history of SCD, extreme LV hypertrophy ≥30 mm, or unexplained syncope) or potential risk mediators.5 The 2014 ESC guidelines recommend ICD implantation using the Risk-SCD risk calculator based on 7 indices.4 The utility of the HCM Risk-SCD score underwent international validation testing in 3,703 HCM patients, including Asian patients, and a meta-analysis.50,51 On the other hand, the HCM Risk-SCD score has been reported as much less sensitive than an enhanced ACC/AHA strategy.25,52 Indeed, sensitivity is important to prevent SCD, but specificity is also important because it avoids overuse of ICDs. Therefore, a more accurate prediction method should continue to be under consideration. The 2018 JCS/JHFA guidelines proposed a new recommendation taking into account as many Japanese studies as possible while referring to the 2011ACC/AHA and 2014ESC guidelines.6 However, validation of this recommendation in Japanese cohorts is necessary.
AFAF is a frequently complicating arrhythmia that occurs in approximately 25% of HCM patients, causing cerebral embolism and triggering HF.53 As AF in HCM risks development of cerebral embolism, even with low CHADS254 and CHA2DS2-VASc scores,55 anticoagulant therapy should be considered in all patients unless contraindicated. Although no randomized control study has been conducted comparing warfarin and direct oral anticoagulants (DOACs) in HCM, some observational studies and meta-analyses have shown that DOACs offer similar or lower rates of thromboembolic and bleeding events in patients with HCM compared with warfarin.56,57 The recent developments of catheter ablation or Maze operation have proven effective in maintaining sinus rhythm. Although AF has been previously cited as a prognostic predictor for HCM, the effect of AF on prognosis has diminished under intensive management.58
HFUnlike the management strategies for SCD and AF, which have reduced mortality and mobility rates under contemporary strategies, HF remains a challenge in HCM.
Although HF in HCM has been caused by various factors, 3 clinical pictures appear: (1) HF with preserved EF (HFpEF) not involving LV obstruction (LVOT or mid-ventricle); (2) HFpEF involving LVOT or MVO; and (3) HF with reduced EF (HFrEF) due to evolution to LV systolic dysfunction (D-HCM or endstage phase).28,49 In addition, AF, myocardial ischemia and mitral regurgitation modify the clinical presentation. Female patients show more severe HF symptoms than male patients. Patients with any sarcomere mutations (not specific mutations) also show a higher frequency of HF than those without mutations.59
LVOT obstruction plays a major role in the HF symptoms in HCM patients. It is reported that progression to NYHA class III/IV HF is 7.4% per year in patients with obstruction at rest, 3.2% per year in patients with obstruction at rest or provoked by exercise and 1.6% per year in patients showing no obstruction with exercise.31 Therefore, LVOT obstruction is an important therapeutic target in HCM patients with HF symptoms. Contemporary management of LVOT obstruction promises to improve symptoms in obstructive HCM. First, medical treatment (β-blockers, calcium-channel blockers (verapamil and diltiazem) and sodium-channel blockers (disopyramide and cibenzolin in Japan)) is administered to relieve exertional symptoms. If sufficient symptom improvement is not achieved with pharmacotherapy, septal reduction therapies (percutaneous transluminal septal myocardial ablation (PTSMA)/alcohol septal ablation or surgical myectomy) are considered based on the anatomic features of the type of HCM and the patient’s background.4–6
Most patients without LVOT obstruction show a relatively benign clinical course with no or mild HF symptoms. However, the prognosis of advanced HF remains poor in patients with severe LV diastolic dysfunction or progression to LV systolic dysfunction among patients without LVOT obstruction. LV systolic dysfunction is managed according to the HFrEF guidelines, but treatment is not effective in many cases. In accordance with the progress of LV remodeling, HF become refractory and mortality remains high. At present, heart transplantation is the only definitive treatment option. Therefore, the need for improvement in diastolic function and prevention of progression to LV systole dysfunction is currently unmet, and the development of new approaches such as drugs or a gene-based therapeutic technology to prevent myocardial damage is urgently needed for HCM management.60 Among these, it was recently reported that mavacamten, which is a allosteric inhibitor of cardiac-specific myosin adenosine triphosphatase, showed significant reductions in NT-proBNP and cTI levels, suggesting improvement in myocardial wall stress in patients with non-obstructive HCM61 and is expected for future development.
Since the reports by Teare and Brock, traditionally a great deal of attention has been paid to genetics, LVOT obstruction and the prediction and prevention of SCD in HCM,1,2,7,8,25,29 all of which are very important issues. In addition, it is important to have a perspective of HCM as a heterogeneous and lifelong disease in the real world. HCM is a disease that is extremely diverse in etiology, morphology, and prognosis (Figure 7). As lifelong LV remodeling with associated complications occurs (Figure 8), long-term follow-up and appropriate therapeutic interventions are important in HCM.3,62 Again, genetics, imaging modalities and management and observation reports from regional institutes and registries have substantially taught us that: (1) HCM is a common disease in the real world; (2) the etiology cannot be explained by sarcomere genetic mutation alone; (3) the morphological and pathological conditions are highly variable; (4) these conditions change via lifelong LV remodeling; (5) accordingly, many patients suffer from AF and HF related to disease progression, in addition to SCD; and (6) better understanding of the heterogeneity and lifelong LV remodeling in HCM and continued progress in treatments will save many patients.
Heterogeneity in morphology, prognosis and management of HCM. AF, atrial fibrillation; ICD, implantable cardioverter defibrillator; HCM, hypertrophic cardiomyopathy; LVOT, left ventricular outflow tract; SCD, sudden cardiac death; SRT, septal reduction therapy.
Lifelong LV remodeling and cardiovascular events in hypertrophic cardiomyopathy (HCM). D-HCM, dilated phase of HCM; LVH, left ventricular hypertrophy; SCD, sudden cardiac death. (From Tsutsui et al6 with translation and permission.)
Most of our experience described in this review has been supported by the Kochi Cardiomyopathy Network since 2004.3 Thanks to all the participating investigators (Naohisa Hamashige, MD, Kazuya Kawai, MD, Masahiko Fukatani, MD, Shoichi Kubokawa, MD, Yoko Nakaoka, MD (Chikamori Hospital), Takashi Yamasaki, MD (National Hospital Organization Kochi National Hospital), Yoko Hirakawa, MD (Tosa Municipal Hospital), Masanori Kuwabara, MD, Takashi Furuno, MD (Kochi Prefectural Aki General Hospital), Toshikazu Yabe, MD (Kochi Prefectural Hata Kenmin Hospital) and the many staff in the Department of Cardiology and Geriatrics, Kochi University).
H.K. (Daiichi Sankyo Company, Takeda Pharmaceutical Company, Mitsubishi Tanabe Pharma Corporation, Bayer Yakuhin, Ono Pharmaceutical Company). H.K. is a member of Circulation Journal ’ Editorial Team.