| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 1995;91:532-540.)
© 1995 American Heart Association, Inc.
Articles |
From the Unité de Recherches (K.S., L.C., P.G.), 153 de l'INSERM, and the Service de Cardiologie (M.K.), Groupe Hospitalier Pitié-Salpêtrière, Paris, France.
Correspondence to Ketty Schwartz, PhD, INSERM UR 153, Pavillon Rambuteau Groupe Hospitalier Pitié-Salpêtrière, 47 Boulevard de l'Hôpital, 75651 Paris Cedex 13, France.
Key Words: hypertrophy cardiomyopathy genes
| Introduction |
|---|
|
|
|---|
Although apparently clear, this clinical classification presents major limitations: specific cardiac diseases such as hypertension or ischemic heart disease, as well as general disorders with cardiac involvement, can mimic the clinical presentation of idiopathic cardiomyopathies. Moreover, an overlap exists between these categories. For instance, in end-stage HCM, a marked dilation of both ventricles, similar to that observed in DCM, can be present. Most important, this classification does not address the underlying molecular disorders responsible for the development of the "clinical" cardiomyopathy.
During the past few years, new and unexpected insights into the pathogenesis and classification of cardiomyopathies have emerged from the localization and the identification of disease genes of several inherited forms. The process of disease gene identification used to be laborious and time consuming. Because of the striking development of the resources provided by the Human Genome project, and more specifically by the generation of highly resolutive genetic maps, this process has been greatly facilitated, and the strategy of positional cloning now allows one to map any mendelian trait and, in particular, monogenic human diseases. Curiously, genetics has been relatively late in being applied to cardiac muscle diseases but has been used extensively in the investigation of atherosclerosis and lipoprotein metabolism diseases for many years. The past 2 to 3 years have marked the dawn of a new era for the genetics of cardiac muscle diseases, and the cardiomyopathies have opened this fascinating route. This review focuses on the relatively few inherited HCMs and DCMs for which an abnormal gene or defective protein has been identified. Particular emphasis is placed on the genetic bases of the diseases and on the correlations between the clinical manifestations (phenotype) and a mutant genotype, and the potential clinical implications are discussed. By definition, diseases in which the myocardial pathology is part of a known systemic disorder have not been included.
| Evidence for the Existence of Familial Forms of Cardiomyopathies |
|---|
|
|
|---|
Careful screening of families has provided the evidence that, as in HCM, there is a strong genetic component in DCM, although it was generally considered to be a multifactorial disease related to various toxic environmental factors: in a panel of 59 patients, DCM was familial in approximately 20% of the cases.8 In the present study, complex segregation analysis of the pedigrees supported the evidence for a single dominant locus with incomplete penetrance. However, in most instances, the size of the family and the number of affected subjects do not allow an accurate analysis of the mode of inheritance. An autosomal dominant transmission was nevertheless most frequently suggested,9 10 11 but other modes of inheritance, including autosomal recessive9 11 and X-linked,9 12 have been found. The apparent heterogeneity in the patterns of inheritance raises the possibility that DCM is a polygenic disease with multiple genetic factors being involved or is even multifactorial with the intervention of polygenic and environmental factors to various degrees from one individual to the other.11 In addition to the small size of most of the families with DCM and the potential genetic heterogeneity, another difficulty in linkage analysis is the presence of middle-aged adults with minor or mild cardiac dilation and/or reduced ejection function, a potentially misleading pattern of ischemic heart disease that makes hazardous an accurate classification of these subjects as unaffected or carriers.
| Disease Genes for FHC |
|---|
|
|
|---|
-tropomyosin. Indeed, none of the previous hypotheses of the
pathophysiological mechanisms of the disease would have led us to
suspect that one of the molecular bases could be a defect in a
contractile protein; nevertheless, certain forms of the disease clearly
involve mutations in sarcomeric protein genes.
|
ß-Myosin Heavy Chain, Cardiac Troponin T, and
-Tropomyosin
The genes encoding for the two cardiac myosin heavy chain
(MHC) isoforms (
and ß) are located in tandem on chromosome
14q11-q12 and were therefore unexpected candidate genes when this locus
was found. Subsequently, with the use of genetic mapping and DNA
sequencing, the ß-MHC gene (MYH7) was identified as the
morbid gene carrying a point mutation in the original family described
by Paré.18 Following this initial description,
several other families have shown evidence for linkage to
CMH1.19 20 21 22 At the
present time, 29 missense mutations
have been found in the ß-MHC gene, among which 8 were reported on in
abstract form
(Table
).18 21 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
All are
localized either to the head region or to the junction of the head/rod
domains. Almost simultaneously, our team and one from South Africa
found that codon 403 is a hot spot for mutations, which suggested a
major role of this arginine residue in the maintenance of a normal
function of the MHC protein.31 34 In one family, a
deletion involving the carboxyl-terminal region of ß-MHC has also
been identified,45 but it is not clear in this family
whether this is the cause of the disease.
|
Of course, neither linkage analyses nor identification of the mutations in the ß-MHC gene is sufficient to prove that these mutations are the primary cause of FHC. Some clues have nevertheless been provided over the past 2 years.
First, the substitution of an adenine for a guanine, resulting in the Arg403Gln mutation, was found to be expressed in mRNA of both myocardial and bicep muscle of the same proband.46 47 Very recently, the mutant Arg719Trp ß-MHC protein was found to be present in the myocardium and in the skeletal muscle of two unrelated patients.48 49 However, even though in two subjects demonstrating the same missense mutation total myosin and immunoreactive ß-myosin myocardial levels were similar to those found in various disease control subjects,50 the respective levels of expression of the normal and mutated genes remain to be determined. It is also not known if the other types of mutant MYH7 alleles are transcribed into the corresponding proteins.
Second, myofibrillar organization appears to be intact in a patient with the Arg403Gln mutation,50 and protein stability does not appear to be markedly affected in mammalian nonmuscle cells transfected with expression constructs encoding seven different FHC mutants.51 However, the ability to form filaments appears to be impaired since as many as one third of the transfected cells fail to form filamentous structures.51 It should, however, be pointed out that the latter studies have been carried out in nonmuscle cell types and did not include the coexpression of myosin light chains, which are critical for the normal assembly and function of MHC. It is thus not completely clear as to whether the mutant MHC proteins are indeed assembled abnormally in the muscle context, and it is equally unclear whether this would explain myofibrillar disarray in FHC.
Third, crystallographic data suggest that several mutations are in or near functional sites of the myosin molecule.52 53 For example, three mutated amino acids are located in the nucleotide-binding pocket; four others, including Arg403, are in the actin-binding site; and Leu908 is located in the hinge, where motive force is generated. All of this suggests that the mutations could have a deleterious effect on the various functions of the molecule.
Fourth, in vitro analyses have shown that some of the described mutations induce a decrease in both the actin-activated ATPase activity of myosin fragments54 and the actin translocation rate on the mutated myosin bound to a coverslip surface.30 54
Fifth, the last line of evidence comes from patients with sporadic HCM; de novo myosin mutations have been found in individuals with HCM but whose parents are clinically and genetically unaffected, and one of these mutations was transmitted to an affected child.55
After analysis of the clinical, echocardiographic, and pathological
findings, it became clear that FHC is a heterogeneous disease (see
Reference 56 for review). It was shown in 1990 that FHC is also
genetically heterogeneous since in two of four families, no linkage to
CMH1 was found.57 The next major problem in continuing the
genetic analysis was to obtain access to large informative
pedigrees (more than 50 members). To circumvent this problem and to
analyze small families, we have used two highly informative
microsatellite markers contained in the ß-MHC gene that we named MYOI
and MYOII.58 59 60 61
Analysis of these two microsatellites
enabled us to exclude linkage to CMH1 in eight unrelated medium-size
families.60 Almost simultaneously, other families in which
the disease was not linked to CMH1 were
identified.20 22 25 61 62 63
Last year, we and others found
in unrelated families three other disease loci: one on chromosome 1q3
(CMH2),14 one on chromosome 15q2 (CMH3),16
and one on chromosome 11p13-q13 (CMH4).15 Moreover, we
have evidence that a fifth locus exists.17 By synteny with
the murine genome and by precise genetic analysis, the disease
genes contained in CMH2 and CMH3 were identified very recently; they
encode for two other sarcomeric proteins: cardiac troponin T on CMH2
and
-tropomyosin on CMH3.64 Two missense mutations were
found in the
-tropomyosin gene in exon 5, which encodes part of a
putative binding domain for troponin T. As for the cardiac troponin T
gene, two missense mutations in the putative exons 8 and 9 and a
mutation in the splice donor site of intron 15 were found in three
unrelated families (see the Table
). The missense mutations
affect
nucleotides encoding a region involved in calcium-insensitive binding
to
-tropomyosin, and the mutation in the donor site produces
markedly aberrant cardiac troponin T mRNA transcripts that alter the
carboxyl terminus of troponin T, a region contributing to
calcium-dependent binding to tropomyosin. These observations that
-tropomyosin and cardiac troponin Tas ß-MHChave central
roles
in the structure and function of the sarcomere in striated muscle
confirmed the first hypothesis and suggest that one of the
pathophysiological mechanisms of FHC could be an impaired function of
the contractile apparatus.
What Are the Other Disease Genes?
These findings suggest that
mutations in other contractile protein
genes from either the thick or the thin filament, or in any protein
implicated in filament assembly, may account for FHC at other loci,
including CMH4 on chromosome 11p13-q13. At the present time, only
one family shows linkage to CMH4.15 Enlargement of this
family and the analysis of 14 other new, informative microsatellite
markers allowed us recently to identify new recombinant individuals and
to reduce the candidate interval from 23 centimorgan to 12 centimorgan.
From the 60 genes already located in this interval, none encode for
another sarcomeric protein. Cardiac troponin C gene, which was not
mapped to the genome, was excluded from chromosome 11 by specific
analysis of DNA isolated from human x hamster cell lines (data not
shown). As for cardiac troponin I gene, it was mapped to chromosome
19p13.2-q13.265 and therefore is a candidate for families
in whom the disease is not linked to any known locus. It is clear that
much more work is needed to identify the other genes that cause FHC and
to be able to propose a general and unifying hypothesis for the
pathogenesis of this disease.
Link Between Molecular and Organ Abnormalities
If the
hypothesis that FHC is a disease of the sarcomere is valid,
one has to explain the paradox of the coexistence of genetic
abnormalities that alter sarcomeric function and result in
depressed contractility at the molecular level and the reported
maintained and even increased systolic function in patients with
HCM.66 Indeed, pump function indexes such as ejection
fraction, cardiac output, or left ventricular systolic pressure are
normal or "supernormal" in a majority of patients, and these
observations led to the concept of a "hyperdynamic" or a
"hypercontractile" state in this disease, particularly in the
presence of left ventricular
obstruction.66 67 68 Because
cardiac performance results from preload, afterload, contractility,
relaxation, and compliance, normal systolic indexes do not necessarily
indicate a normal contractile
state.69 70 71 Indeed, in a
group of patients with HCM, isovolumic phase and ejection phase indexes
of contractility (+dp/dt, dp/dt/DP40, ejection fraction) were found to
be normal, whereas afterload was significantly
decreased.72 However, an index of contractility supposed
to be independent of loading conditions, ie, the end-systolic
stresstovolume ratio, was reduced in patients to half of the
normal
value, as was the unit muscle performance (minute work/mass). These
results suggest that intrinsic contractility is actually decreased in
HCM, and this has been confirmed by other studies when left ventricular
hypertrophy was related to HCM.73 These findings at the
organ level are therefore in keeping with the in vitro analyses and
support the idea that in HCM, hypertrophy would be a compensatory
mechanism to maintain normal systolic function. Understanding why this
hypertrophy is, in most instances, eccentric and not concentric as it
is following a chronic increase in afterload is the next challenge.
| Disease Genes for DCM |
|---|
|
|
|---|
In three pedigrees with X-linked DCM, defects in the dystrophin gene and low abundance of cardiac dystrophin but not of skeletal muscle dystrophin were found, and the authors demonstrated that this disease is associated with a deletion of the muscle-promoter region and the first exon of the gene.74 75 A reduced abundance of dystrophin was also found in the BIO 14.6 hamster, which is a widely recognized animal model of DCM.76 The pathogenetic hypothesis for this type of DCM is a disruption of the membrane cytoskeleton of the myocyte due to the reduced dystrophin content, and this hypothesis has been reinforced by the recent finding of a deficiency of a dystrophin-associated glycoprotein in this same strain of hamster.77 Because Duchenne and Becker muscular dystrophies are due to dystrophin gene abnormalities and because there have been some reports of Becker dystrophy with predominant or even exclusive cardiac involvement and no or only minor skeletal muscle involvement,78 79 the possibility that some patients with DCM are in fact carriers of a dystrophin gene defect was raised. In a series of patients with familial and nonfamilial DCM, screening of the dystrophin gene defects did not reveal any of the known deletions observed in Duchenne and Becker muscular dystrophy.80 However, it should be pointed out that only 14 exons were studied and that the promoter region of the gene was not analyzed. It is possible that the proportion of cases of DCM related to dystrophin gene defects is small and that screening the dystrophin gene in individuals with DCM should be limited to cases of X-linked cardiomyopathy and/or cardiomyopathy with muscle abnormalities.
Most recently, a morbid gene that causes both an atrioventricular conduction defect and DCM in a large kindred with an autosomal dominant inheritance has been mapped to chromosome 1p1-1q1.81 Based on syntenic mapping studies, the authors speculate that the gap junction protein connexin 40 is a candidate gene for this particular disease.
| Phenotype/Genotype Relation |
|---|
|
|
|---|
Prognosis
Most known information on differences in the
prognosis of HCM
according to different mutations comes from the analysis of
ß-MHCrelated families,21 28 and the main
clinical
consequences of the published mutations of this gene are listed in the
Table
. Until now, the clinical aspect that has been the most
extensively studied is the severity of the disease. For example,
several studies report that the Arg403Gln mutation is
associated with a high incidence of disease-related deaths, whereas the
Leu908Arg or the Val606Met mutations appear to
be more benign.21 28 Based on these observations, it
has
been speculated that mutations that involve a change in the charge of
the amino acid residue such as the Arg249Gln,
Arg403Gln, or Arg453Cys mutations would be
associated with a poor prognosis, whereas neutral mutations such as the
Leu908Val and the Val606Met would be
benign.21 28 However, a large number of premature
sudden
cardiac deaths or disease-related deaths were observed in a kindred
with the neutral Val606Met mutation, whereas in a small
Korean kindred with six individuals bearing the Arg403Gln
"malignant mutation," no sudden deaths have
occurred.40 Similarly, in a large kindred, the
Gly256Glu mutation was associated with a low incidence of
sudden deaths, although this mutation induces a change in the charge of
the molecule.40 These discrepancies underline the need for
large kindreds for the purpose of prognostic analyses and the need for
the creation of an international survey on the HCM data base to
establish whether a given mutation is benign or malignant. At
present, most kindreds are too small to allow any definitive
conclusions to be drawn, and most reported data should be considered
preliminary.
Penetrance and Left Ventricular Hypertrophy
Correlations
between the different mutations observed in the
ß-MHC gene and the degree or pattern of ventricular hypertrophy or
the penetrance are even more confusing. As observed in the
Table
, there
is no clear relation between the fact that a given mutation is neutral
or not and the fact that penetrance is full or partial. Few studies
have addressed the relation between missense mutations and the degree
or the pattern of hypertrophy. An echocardiographic study of 39 adults
carrying six distinct mutations concluded that there were no overt
differences in the echocardiographic characteristics, but this study
lacked the precision to detect significant differences.83
Furthermore, this kind of analysis is difficult for three main
reasons.
First, the number of affected individuals in a given family is in many instances too small to give conclusive evidence that the degree of hypertrophy is influenced by a given mutation. The possibility that distinct ß-MHC mutations are associated with different clinical presentations therefore remains open.
Second, parent mutations
occurring within the same codon may
result in differences of penetrance and clinical presentation. In
one of our families (kindred 720), a G
T transversion in codon 403
resulted in the mutation Arg403Leu. This large family is
associated with a partial penetrance and a high incidence of end-stage
HCM characterized by dilation and pump failure, whereas other families
with a G
A transition in the same codon, resulting in the mutation of
Arg403Gln, have a full
penetrance.21 28 31
Moreover, in family 720, left ventricular hypertrophy was severe,
whereas in another kindred, family 730, another mutation in codon 403
(Arg403Trp) was associated with mild left ventricular
hypertrophy.31
Third, the degree and pattern of hypertrophy may be different in unrelated families carrying the same mutation or even in relatives of the same family. For example, it was observed that the Val606Met mutation is associated with various degrees of hypertrophy in three unrelated families, but this finding did not influence the outcome.28 In three unrelated families carrying the Arg403Gln mutations, variations have been reported with regard to the presence or absence of right ventricular hypertrophy and left ventricular obstruction.18 40
These findings emphasize the role of other factors, including environmental differences, acquired traits (eg, differences in lifestyle, risk factors, and exercise), or modifier genes, that could modulate the phenotypic expression of the disease. Modifier genes are apparent in all genetically mixed populations, and they are commonly referred to as the "genetic background" in which the mutant gene finds itself. Various observations support this idea: (1) first-degree relatives with FHC may exhibit markedly distinct expression of the disease occurring at very different ages84 ; (2) monozygotic twins may develop a different expression of the disease, in particular with regard to the extent of hypertrophy and outflow obstruction.85 A possible gene modifying the pattern of hypertrophy could be the angiotensin-converting enzyme gene, which contains an insertion/deletion (I/D) polymorphism. This polymorphism is associated with the level of circulating enzyme, the D/D genotype being associated with higher levels of circulating angiotensin-converting enzyme than I/D and I/I genotypes.86 The D/D genotype is also associated with an increased risk of left ventricular hypertrophy detected by electrocardiography in middle-aged men recruited in a general population (odds ratio, 2.64).87 The frequency of allele D was reported to be higher in FHC families with a high incidence of sudden cardiac deaths than in those with a low incidence.88 However, the comparison was made with healthy relatives and not with the general population, and the conclusions of this study, although very interesting, should be considered with caution and deserve further confirmatory results.
Phenotype and
genotype analysis of other affected genes, including
the
-tropomyosin gene on chromosome 15 and the cardiac troponin T
gene on chromosome 1, are in an early stage and do not allow any
definite conclusions to be made. Preliminary studies suggest
differences in phenotypes and prognosis according to the type of
mutation.14 16 64
| Clinical Implications |
|---|
|
|
|---|
Revisiting Diagnosis Criteria
Determination of the genotype
in FHC provides the
opportunity of reassessing major criteria for the diagnosis of HCM by
ECG and by Doppler echocardiography. One might expect reexamination of
diagnostic criteria in genotyped individuals to be fruitful for
clinical cardiology. Assessment of the sensitivity, the specificity,
and the predictive values of major abnormalities observed on ECGs and
by Doppler echocardiography, including ventricular morphology, systolic
anterior motion, and Doppler functional indices such as outflow
gradient and diastolic abnormalities, can now be carried out for all of
the commonly reported mutations.40 83 This analysis
may enable one to find new, subtle indexes of the disease and to define
more flexible and accurate diagnostic criteria in the highly selected
population of families at risk for the disease.
Phenotype/Genotype Relation and Prognosis Stratification
Genetic studies have provided conclusive evidence that the
clinical presentation of FHC covers a broad spectrum from
apparently normal ventricular function and morphology to severe
hypertrophy. The implementation of a large-scale database reporting the
different mutations in the different morbid genes associated with the
disease in relation to clinical presentations should provide the
opportunity to establish a new classification based on the most common
genetic defects or on the underlying molecular mechanism. It is
therefore likely that the comparison of the clinical data of many
unrelated families bearing the same genetic defect will allow the
identification of specific clinical subgroups, although this
analysis may be obscured by other factors, including ethnic origin
or modifier genes.
The analysis of mutation-specific natural history is another major issue. The identification of malignant mutations has important clinical implications with regard to genetic counseling and identification of individuals at high risk of disease-related death, regardless of the mechanism involved (ventricular arrhythmias, atrial arrhythmias, bradyarrhythmias, hypotension, myocardial ischemia, left ventricular outflow obstruction).56 Labeling FHC individuals who are prone to sudden death or disease-related death is a difficult clinical issue since clinical and morphological presentations of HCM, including the severity of the hypertrophy, are not accurate prognostic markers,89 90 91 and factors recognized as indicating an increased risk of sudden death (ventricular tachycardia on Holter, young age, familial cardiac death, history of cardiac arrest or syncope) are not highly sensitive or specific. Reports of family history and careful follow-up of patients carrying identical mutations will therefore be useful for this purpose, although the question of how to treat these individuals remains open.
Clearly, more studies that would improve our understanding of the relation between phenotype and genotype are warranted, and the ultimate value of genotyping in FHC may be primarily to define an "at-risk" group. This is an area that deserves further careful clinical research.
Identification of Healthy Carriers
Genetic analysis of FHC
kindreds has revealed the presence of
clinically healthy individuals carrying the mutant allele, which is
associated in first-degree relatives with a typical phenotype of the
disease.21 31 Although it is known that children may
develop clinical symptoms during adolescence and that there are reports
of selected kindreds with myocardial disarray but no overt
hypertrophy,56 92 identification of
"asymptomatic
ill" individuals bearing malignant mutations raises new, important
clinical questions, particularly in young adults. Again, exchange of
scientific information and careful follow-up of these selected
individuals are necessary to assess whether the mutation remains a
curiosity without clinical relevance (as in obligate carriers) or
whether these individuals will develop the disease later and require
early medical management.
Routine Genotyping
Molecular biology applied to routine
genotyping offers promising
perspectives in genetic counseling, provided that sufficient scientific
information is available to establish a genetic-based prognostic
database. The main potential application in families with malignant
mutations is individual information given to adolescents or young
adults for professional purposes (eg, athletes or air crew members).
However, evaluation and treatment of each individual should be
considered on a case-by-case basis since a widespread difference in
phenotype can be seen in patients harboring similar genotypes. Until we
define the other contributing modifiers, we will not know what
recommendations to make to minimize the risks of the "at-risk"
group. Clearly, there is a need to develop animal model systems in
which to evaluate these modifier effects on an experimental basis.
Moreover, concerns that arose about the psychological, clinical, and
discriminative aspects of large-scale genetic risk testing in other
diseases such as Huntington's disease also apply to cardiomyopathies,
and at present, widespread testing is not indicated in these
diseases.
Therapy
Preliminary reports of conventional medical
strategies based on
specific mutations have been published. In two kindreds with the
Arg403Gln mutation, affected individuals have been treated
with ß-blockers to prevent excessive tachycardia and with verapamil
to prevent myocardial ischemia.40 These reports indicate
that medical testing (including invasive testing) and medical
management of HCM (ß-blockers, verapamil, amiodarone,
implantable defibrillator) might be tailored in the future according to
the underlying genetic defect. Although research in DCM is preliminary,
it is likely that identification of morbid genes in familial forms of
the disease will allow early identification and medical management of
affected individuals to prevent progressive pump failure, sudden death,
or both.
To envisage at this time new gene-based therapeutic strategies to cure or prevent the development of hypertrophic cardiomyopathy is risky, since it is an autosomal dominant disease that involves structural proteins that are expressed at relatively high levels throughout the heart muscle. Once the other determinants (which may very well be acquired) that induce the pathological phenotype in certain patients that harbor the genotype are identified, it should be possible to suggest new preventive or curative strategies. To achieve these important goals, continuous collaboration between geneticists and cardiologists appears to be of high priority for a careful, detailed, large-scale analysis of phenotype/genotype relations of patients that harbor mutations in one of the known loci.
Continued
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
ur Vaisseaux. In press.
Val
mutation and a 403 Arg
Gln mutation. Circulation. 1992;86:345-352.
-Tropomyosin and cardiac
troponin T mutations cause familial hypertrophic
cardiomyopathy: a disease of the sarcomere.
Cell. 1994;77:701-712. [Medline]
[Order article via Infotrieve]This article has been cited by other articles:
![]() |
A. Bye, M. Langaas, M. A. Hoydal, O. J. Kemi, G. Heinrich, L. G. Koch, S. L. Britton, S. M. Najjar, O. Ellingsen, and U. Wisloff Aerobic capacity-dependent differences in cardiac gene expression Physiol Genomics, March 14, 2008; 33(1): 100 - 109. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. S. Adabag, S. A. Casey, M. A. Kuskowski, A. G. Zenovich, and B. J. Maron Spectrum and prognostic significance of arrhythmias on ambulatory Holter electrocardiogram in hypertrophic cardiomyopathy J. Am. Coll. Cardiol., March 1, 2005; 45(5): 697 - 704. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. J. Maron, J.G. Seidman, and C. E. Seidman Proposal for contemporary screening strategies in families with hypertrophic cardiomyopathy J. Am. Coll. Cardiol., December 7, 2004; 44(11): 2125 - 2132. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Gilbert-Barness Metabolic Cardiomyopathy and Conduction System Defects in Children Ann. Clin. Lab. Sci., January 1, 2004; 34(1): 15 - 34. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Kulikovskaya, G. McClellan, J. Flavigny, L. Carrier, and S. Winegrad Effect of MyBP-C Binding to Actin on Contractility in Heart Muscle J. Gen. Physiol., November 24, 2003; 122(6): 761 - 774. [Abstract] [Full Text] [PDF] |
||||