Author + information
- Received September 21, 2000
- Revision received March 19, 2001
- Accepted April 12, 2001
- Published online August 1, 2001.
- Barry J Maron, MD, FACC∗,* (, )
- Hideshi Niimura, MD†,
- Susan A Casey, RN‡,
- Marjorie K Soper, MD‡,
- Gregory B Wright, MD, FACC‡,
- J.G Seidman, PhD§ and
- Christine E Seidman, MD§
- ↵*Reprint requests and correspondence:
Dr. Barry J. Maron, Hypertrophic Cardiomyopathy Center, Minneapolis Heart Institute Foundation, 920 East 28th Street, Suite 40, Minneapolis, Minnesota 55407
We sought to determine whether the development of left ventricular hypertrophy (LVH) can be demonstrated during adulthood in genetically affected relatives with hypertrophic cardiomyopathy (HCM).
Hypertrophic cardiomyopathy is a heterogeneous cardiac disease caused by mutations in nine genes that encode proteins of the sarcomere. Mutations in cardiac myosin-binding protein C (MyBPC) gene have been associated with age-related penetrance.
To further analyze dormancy of LVH in patients with HCM, we studied, using echocardiography and 12-lead electrocardiography, the phenotypic expression caused by MyBPC mutations in seven genotyped pedigrees.
Of 119 family members studied, 61 were identified with a MyBPC mutation, including 21 genetically affected relatives (34%) who did not express the HCM morphologic phenotype (by virtue of showing normal left ventricular wall thickness). Of these 21 phenotype-negative individuals, 9 were children, presumably in the prehypertrophic phase, and 12 were adults. Of the 12 adults with normal wall thickness ≤12 mm (7 also with normal electrocardiograms), 5 subsequently underwent serial echocardiography prospectively over four to six years. Of note, three of these five adults showed development of LVH in mid-life, appearing for the first time at 33, 34 and 42 years of age, respectively, not associated with outflow obstruction or significant symptoms.
In adults with HCM, disease-causing MyBPC mutations are not uncommonly associated with absence of LVH on echocardiogram. Delayed remodeling with the development of LVH appearing de novo in adulthood, demonstrated here for the first time in individual patients with prospectively obtained serial echocardiograms, substantiates the principle of age-related penetrance for MyBPC mutations in HCM. These observations alter prevailing perceptions regarding the HCM clinical spectrum and family screening strategies and further characterize the evolution of LVH in this disease.
Hypertrophic cardiomyopathy (HCM) is a familial cardiac disease with heterogeneous clinical and phenotypic expression (1,2), caused by mutations in nine genes that encode proteins of the cardiac sarcomere (3–9). In genetically affected individuals in the present study group, HCM was due to a mutation on chromosome 11p11.2, where the
cardiac myosin-binding protein C (MyBPC) gene is encoded (7,8). Myosin-binding protein C contributes to the structural integrity of the sarcomere by binding to the myosin heavy chain and the cytoskeleton protein titin (7,8).
A clinical diagnosis of HCM is made by using two-dimensional echocardiography, with the morphologic phenotype characterized by a variety of patterns of left ventricular hypertrophy (LVH) in the presence of a nondilated left ventricular (LV) chamber (1,2,10). Usually, hypertrophy is absent or only mildly expressed until adolescence, at which time LV wall thickness often increases substantially, achieving its mature morphologic state only after full growth and maturation at ∼17 to 18 years of age (11), with little further increase in wall thickness during adulthood (12).
However, for the MyBPC mutant gene, retrospective cross-sectional analysis has suggested that penetrance of the phenotype may be age-related, with appearance often delayed until adulthood (7). We have focused on the phenotypic expression of MyBPC in HCM, and present, for the first time, prospectively obtained, serial echocardiographic evidence for the development of LVH in genetically affected adults.
Selection of pedigrees and clinical investigation
Of the 13 families from the Minneapolis Heart Institute genotyped in collaboration with the Department of Genetics at Brigham and Women’s Hospital, seven proved to have mutations in the MyBPC gene. Closely related relatives of each proband at risk of inheriting the disease gene underwent a cardiovascular examination before genotyping, including 12-lead electrocardiography and two-dimensional echocardiography with Doppler imaging. Serial echocardiograms were obtained in five genetically affected adult family members in whom the initial echocardiographic study at >25 years of age showed normal LV wall thickness; seven other relatives who met these criteria did not agree to have serial echocardiographic studies. Genetic and selected clinical data from some patients were included in a previous report (7).
Identification of mutations in the cardiac MyBPC gene
As previously described (7), we obtained 5 to 30 ml of peripheral blood from each proband, and deoxyribonucleic acid (DNA) was isolated. The entire sequence of cardiac MyBPC is in the GenBank data base (accession no. U91629). Exons 2 through 35, which encode protein sequences, were amplified from genomic DNA by using primers designed from intron sequences. Genomic DNA fragments amplified with the polymerase chain reaction were purified with the QIAquick PCR purification kit (QIAGEN, Santa Clarita, California) to remove residual primers and were sequenced with the ABI PRISM dye-terminator cycle-sequencing kit (Perkin-Elmer, Norwalk, Connecticut).
Diagnostic clinical studies
Echocardiographic studies were performed with Hewlett-Packard (Andover, Massachusetts) nos. 500 and 2000 instruments. Images were obtained in multiple cross-sectional planes using standard transducer positions (10). The extent and distribution of LVH were assessed from echocardiographic images, and the site of maximal wall thickness was identified (10). Other cardiac dimensions were measured from M-mode echocardiograms consistent with previous recommendations (13). Peak instantaneous LV outflow gradient was estimated with continuous wave Doppler echocardiography under basal conditions (10).
Analysis of echocardiograms and electrocardiograms
Echocardiograms and electrocardiograms (ECGs) were interpreted prospectively by one experienced observer (B.J.M.) who had no advance knowledge of the DNA diagnosis made in the laboratory or previous clinical findings. In selected instances, repeat echocardiograms were unavoidably interpreted with some knowledge of the clinical circumstances.
Definition of HCM
The HCM phenotype was defined, using echocardiography, as a hypertrophied nondilated LV in the absence of another cardiac or systemic disease capable of producing a comparable magnitude of LVH (14). For adult family members (≥18 years old), maximal wall thickness ≥13 mm in one or more regions of the LV (in the absence of body surface area correction) was used as an arbitrary but conventional (7,10,12)clinical cut-off value to designate LVH. For children <18 years old, LVH was defined as maximal wall thickness ± 2 SD, relative to body surface area, age and gender (14). Twelve-lead ECG abnormalities were not used as the sole criterion for HCM diagnosis, although ECG alterations may precede the appearance of LVH on the echocardiogram (15,16).
The study group includes 119 surviving family members in seven MyBPC pedigrees, studied with two-dimensional echocardiography, 12-lead electrocardiography and DNA analysis. Before this investigation, 14 other relatives died of HCM (including 2 with the MyBPC mutant gene), and 14 died of other cardiac or noncardiac causes.
The MyBPC mutations were identified in 61 (51%) of the 119 surviving relatives. Specific mutations identified were insertions (Ins18bp1163 and InsG791 , missense mutations (Arg495Gln and Arg326Gln [unpublished data]) and truncations (Int8DSG+1A, IntDSG+1T and Int33DSG+1A). These genetic studies confirmed previously recognized (and clinically evident) HCM in 15 children or adults (including the 7 probands).
Subjects without the HCM phenotype
At the time of genotyping, of the 61 relatives with MyBPC mutations, 21 (34%) did not have morphologic evidence of the HCM phenotype, with normal LV wall thickness in each segment. Nine of these 21 individuals were ≤18 years of age and were judged to probably be in the prehypertrophic phase of their disease (11); each was asymptomatic.
The 12 remaining family members with the mutant MyBPC gene (25% of the 48 genetically affected adults) are particularly noteworthy because they were >25 years old and showed normal LV wall thickness (≤12 mm) at the time of genotyping (mean age 38.9 ± 13 years [range 26 to 62], 8 women and 4 men). Maximal LV thickness in these 12 relatives was 8 to 12 mm. Only one had systolic anterior motion (SAM) of the mitral valve, mild in degree, and with no evidence of outflow obstruction; LV end-diastolic cavity dimensions were within normal limits (mean 44 ± 4 mm [range 38 to 56]).
In 7 of these 12 adults, the ECG was within normal limits, but one or more ECG alterations were evident in the remaining 5 individuals, including T-wave inversion and left-axis deviation (n = 2 each) and abnormal Q-waves, left atrial enlargement and decreased R-waves in leads V1and V2(n = 1 each). Two of the 12 genotype-positive/phenotype-negative relatives represent examples of incomplete penetrance of the HCM phenotype in which that family member was an obligatory gene carrier (Fig. 1).
Development of LVH in adulthood
In 5 of the 12 genetically affected adults with normal LV wall thickness at the time of genotyping, one to four additional echocardiograms were obtained three to six years later (Table 1). Two of these five relatives (Patient nos. 4 and 5 in Table 1) showed no measurable change in LV wall thickness over this period.
However, of note, the remaining three patients (Patient nos. 1 to 3 in Table 1) demonstrated substantial increases in wall thickness of 6 to 7 mm (change of 55% to 64%). These spontaneous hypertrophic changes and the development of the HCM phenotype during adulthood occurred by 33 to 42 years of age in the presence of a nondilated LV chamber and were not associated with the development of outflow obstruction under basal conditions, significant cardiac symptoms or systemic hypertension (Figs. 2 and 3). ⇓⇓
One of the three adult relatives who developed LVH showed marked changes in wall thickness during pregnancy (Patient no. 2 in Table 1and Fig. 2). Another had an enlarged and elongated mitral valve (18)producing a moderate degree of SAM (preferentially of the posterior leaflet) (19), as the initial and sole clinically overt manifestation of the mutant gene at the first echocardiographic evaluation (Patient no. 1 in Table 1and Fig. 3). One of these three adults with morphologic conversion had a normal ECG and two had abnormal ECG patterns at the time of their initial echocardiogram (when hypertrophy was absent) these ECGs showed little change in pattern as LV wall thickness increased, although greater precordial T-wave inversion was evident in one subject.
At genotyping, penetrance of the MyBPC phenotype (i.e., LVH) appeared to be age-related when assessed across all decades of life in a cross-sectional analysis. In genetically affected relatives, identification of the HCM phenotype by echocardiography increasedwith age. Left ventricular hypertrophy was less frequent in relatives <20 years old (4 [31%] of 13) and most common in those >50 years (15 [83%] of 18). Furthermore, genetically affected family members without LVH were relatively common among adults 20 and 39 years old (i.e., 15 [35%] of 23).
Subjects with the HCM phenotype
At genotyping, 40 (66%) of the 61 relatives with MyBPC mutations showed characteristic morphologic evidence of the HCM phenotype, with increased LV wall thickness in one or more wall segments. Most were adults (n = 36; mean age 44 ± 11 years [range 28 to 72]), but four were children (1 to 12 years old). Of these 40 relatives, 26 (65%) were male, 15 (38%) were symptomatic (including 7 with moderate to severe symptoms) and 6 (15%) had SAM of the mitral valve, but without evidence of basal outflow obstruction by Doppler imaging.
Maximal LV wall thickness was 7 to 33 mm (20 ± 5 mm in adults; 12.5 ± 4 mm in children), most prominently involving the anterior ventricular septum (n = 34), posterior septum (n = 3) or anterolateral free wall (n = 3). Left ventricular hypertrophy diffusely involved the septum and LV free wall in 14 subjects, only the anterior and posterior septa in 16, and was confined to the anterior septum in 10. There was no relationship between the particular MyBPC mutation and distribution of LVH.
The 12-lead ECG showed a variety of abnormal patterns associated with the morphologic HCM phenotype in 31 relatives (75%), most commonly T-wave inversion (n = 18), deep Q-waves (n = 9) and increased precordial voltages (n = 8). Normal ECG patterns, however, were present in 10 (25%) of the 40 relatives.
Clinical studies (i.e., without laboratory DNA analysis) have suggested that absence of echocardiographically detectable LVH by the time mature growth and development are achieved (age 17 to 18 years) signifies that a mutation causing HCM is probably absent (11,12). Such clinical judgments have generally proved reasonable in the context of most HCM mutant genes, because virtually complete penetrance of the phenotype (i.e., LVH) is usually evident by the end of the accelerated adolescence growth period (11,12).
However, as present data confirm, this basic clinical tenet does not necessarily apply to MyBPC gene mutations (nor to some troponin T mutations) for which penetrance of the HCM morphologic phenotype may be incomplete in adults (5,7,20,21). Indeed, fully 25% of our genetically affected adult family members had normal echocardiograms at the time of genotyping. This observation, as well as the fact that the MyBPC gene is not uncommonly responsible for HCM (7–9,20), suggests that there may well be many individuals affected by MyBPC (but without LVH) who will escape clinical detection.
De novo remodeling in adults with MyBPC
In this regard, our prospectively obtained, serial echocardiographic observations have documented, for the first time, that de novo LV remodeling may occur during the adult years, with genetically affected individuals (previously without LVH) spontaneously developing increased wall thickness characteristic of HCM. Hence, we have demonstrated that the dormant prehypertrophic phase of HCM can extend well into mid-life in genetically affected individuals who, for some reason, do not experience the substantial increase in LV wall thickness usually associated with adolescence (11). Rapid progression of LVH during young adulthood over relatively short periods also suggests the possibility that an environmental factor may have triggered the hypertrophic response. The phenotypic heterogeneity of HCM is underlined by the fact that one of our young adults who developed LVH had (before developing hypertrophy) an elongated and enlarged mitral valve (18,19)producing moderate SAM of the mitral valve, as the initial and sole clinically overt evidence of HCM.
Development of LVH in adult patients with serial follow-up represents an important clinical observation for three reasons. First, demonstration of the initial development of LVH during adulthood (by prospective serial echocardiographic studies in the same patients) substantiates, in a definitive fashion, the phenomenon of age-related prevalence and late-onset hypertrophy in MyBPC, which had previously been hypothesized indirectly from retrospective, cross-sectional analyses (7). Second, delayed appearance of hypertrophy in adulthood may itself be clinically relevant and have an effect on risk stratification by reassigning selected asymptomatic individuals to higher risk categories, such as those with extreme wall thickening (22)in whom treatment for prevention of sudden death may be considered (23). Third, our observations are particularly relevant to family screening strategies. In the absence of DNA analysis, family members with normal echocardiograms at the time of their full growth and maturation (usually ∼18 years of age) can no longer be reassured unequivocally that they do not carry a mutant gene for HCM. This situation may necessitate further echocardiographic studies in adulthood in selected clinical circumstances. Therefore, we believe that our present observations modify the prevailing concepts about HCM and alter clinical practice.
Although demonstration of spontaneous, de novo development of LVH in adults with HCM is novel, we have presently identified only a small number of such individuals with this form of disease progression. Therefore, these cases do not provide evidence for the frequency of adult morphologic conversions, but nevertheless establish the important principle that this process does in fact occur. Certainly, such structural remodeling throughout adulthood in family members with previously normal echocardiograms has proved extremely difficult to document, given the many years (even decades) over which serial echocardiographic studies would have to be performed (2). Furthermore, we believe that such examinations performed routinely would be unproductive in the vast majority of individuals (most of whom would not carry a mutant gene) and also create the unnecessary psychological impression of cardiac disease in many genetically unaffected individuals (9).
At this preliminary stage of observation, the clinical significance that should be attached to genotype-positive/phenotype-negative individuals with MyBPC in families with HCM is unresolved. Indeed, with the advent of preclinical genetic HCM diagnosis, many more children and young adults will be identified as affected with the disease, based solely on DNA analysis, and in the absence of typical morphologic features. However, it should be emphasized that the vast majority of adverse clinical events occurring in HCM have been associated with penetrance of the morphologic phenotype (i.e., LVH), and usually in the context of substantial wall thickening (1,2,22,23). Although the precise risk level associated with the absence of LVH in genetically affected individuals is largely unresolved, there is little evidence that such affected individuals are predisposed to fatal cardiac events (24,25), with the possible exception of a few selected families with the cardiac troponin T gene (21). Therefore, management of individuals with a mutant HCM gene in the absence of hypertrophy requires restraint and prudence before adversely influencing employment and educational opportunities (9)or restricting lifestyle activities, such as competitive sports (26).
Role of the 12-lead ECG
We found little evidence to support the 12-lead ECG as superior to the echocardiogram for clinical identification of adult family members with a disease-causing mutation in the absence of hypertrophy (27,28). In this regard, normal ECG patterns were present in almost 60% of our genetically affected adults without LVH (and even 25% of those adults with the HCM phenotype). Nevertheless, the ECG has some clinical utility as a supplemental test to raise the suspicion of HCM in selected individuals without LVH (particularly children), i.e., consistent with the previous observation in nongenotyped families that ECG patterns are often abnormal in children with HCM during their prehypertrophic phase (15). Furthermore, two of our three adult relatives with de novo development of LVH had ECG alterations before the appearance of hypertrophy (one with a clearly abnormal ECG pattern), and including a patient with an enlarged and elongated mitral valve as the only morphologic marker of the mutant gene (18).
This report represents the first clinical documentation that the HCM phenotype may develop de novo in adults. These novel observations regarding the spontaneous, rapid onset of LVH delayed to mid-life (i.e., adult morphologic conversions) were derived in the context of individual patient profiles obtained prospectively by serial echocardiography. Such findings expand the HCM disease spectrum and underline potential diagnostic and management dilemmas raised by genetically affected family members without the morphologic HCM phenotype. These data alter our clinical perceptions of HCM and emphasize the power and utility of laboratory DNA diagnosis for patients affected by this complex disorder.
☆ This study was supported by grants from the National Institutes of Health and Howard Hughes Medical Foundation (to Drs. J.G. Seidman and Christine E. Seidman) and from the Paul G. Allen Foundation (to Dr. Maron).
- deoxyribonucleic acid
- electrocardiogram or electrocardiographic
- hypertrophic cardiomyopathy
- left ventricle or left ventricular
- left ventricular hypertrophy
- myosin-binding protein C
- systolic anterior motion
- Received September 21, 2000.
- Revision received March 19, 2001.
- Accepted April 12, 2001.
- American College of Cardiology
- Schwartz K,
- Carrier L,
- Guicheney P,
- et al.
- Marian A.J,
- Roberts R
- Maron B.J,
- Moller J.H,
- Seidman C.E,
- et al.
- Klues H.G,
- Schiffers A,
- Maron B.J
- Spirito P,
- Maron B.J
- Sahn D.J,
- DeMaria A,
- Kisslo J,
- et al.
- Klues H.G,
- Maron B.J,
- Dollar A.L,
- et al.
- Maron B.J,
- Harding A.M,
- Spirito P,
- et al.
- Charron P,
- Dubourg O,
- Desnos M,
- et al.
- Moolman J,
- Corfield V.A,
- Posen B,
- et al.
- Maron B.J,
- Kragel A.H,
- Roberts W.C
- McKenna W.J,
- Stewart J.T,
- Nihoyannopoulos P,
- et al.
- Maron B.J,
- Mitchell J.H
- Charron P,
- Dubourg O,
- Desnos M,
- et al.