Coronary Heart Disease
Distinct Heritable Patterns of Angiographic Coronary
Artery Disease in Families With Myocardial Infarction
Marcus Fischer, MD; Ulrich Broeckel, MD; Stephan Holmer, MD; Andrea Baessler, MD;
Christian Hengstenberg, MD; Bjoern Mayer, MD; Jeanette Erdmann, PhD; Gernot Klein, MD;
Guenter Riegger, MD; Howard J. Jacob, PhD; Heribert Schunkert, MD
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Background—Coronary artery disease (CAD) and myocardial infarction (MI) are significantly determined by genetic
background. Whether distinct angiographic features of CAD are affected by inherited factors has never been
investigated. Thus, we analyzed comprehensively the extent to which various aspects of CAD, including disease
severity, distribution of lesions, presence of coronary calcification, morphology of stenoses, and anatomic characteristics, are under genetic control.
Methods and Results—We retrospectively studied the coronary angiograms of 882 siblings with CAD from 401 families.
These families were ascertained through index patients defined by MI before the age of 60 years and at least 1 sibling
with MI or coronary revascularization procedures. Heritability calculations were performed with variance-component
analysis. Additionally, recurrence risks to siblings were analyzed. Traditional cardiovascular risk factors and age at the
first coronary event displayed significant heritable components. After adjustment for age and sex, significant
heritabilities were identified for proximal stenoses, in particular, left main CAD (h2⫽0.49⫾0.12; P⫽0.01), coronary
calcification (h2⫽0.51⫾0.17; P⫽0.001), and ectatic coronary lesions (h2⫽0.52⫾0.07; P⫽0.001). In contrast, no
heritability was found for distal disease (h2⫽0.05⫾0.19; NS), the pattern of coronary arterial blood supply, or the
number of diseased vessels. Calculation of recurrence risks in siblings largely confirmed the heritability estimates.
Conclusions—Distinct morphological characteristics associated with CAD show different degrees of heritability. Notably,
the most hazardous localizations, like left main or proximal disease, display a high heritability. In contrast, some features
of coronary morphology, such as distal disease, do not appear to be markedly influenced by heritable factors.
(Circulation. 2005;111:855-862.)
Key Words: myocardial infarction 䡲 coronary disease 䡲 angiography 䡲 genetics
T
he manifestation of coronary artery disease (CAD) is
influenced by a complex interplay of numerous environmental and genetic factors. With regard to the genetic contribution, a positive family history for myocardial infarction (MI) is
considered to be a strong cardiovascular risk factor.1,2 Despite
extensive molecular genetic investigations, neither CAD nor MI
has been reproducibly associated with specific genetic variants.
This shortcoming has largely been explained by the complex
pattern of the disease, suggesting different disease mechanisms.3
A more comprehensive clinical characterization could give
additional discrimination of the disease phenotype and thus,
narrow the heterogeneity.4 In general, invasive cardiologists
judge the extent (number of vessels involved, location of
lesions), the severity (percentage of diameter narrowing, length
of lesions), and the morphology of coronary lesions of CAD to
characterize each patient’s angiographic pattern of CAD. Accordingly, coronary angiography can provide usable heritable
surrogates of CAD. Until now, only a few case reports on
monozygotic or dizygotic twins have investigated the hereditary
aspects of coronary morphology, but even these have produced
inconsistent results.5–13 Therefore, a large angiographic database
for patients with familial MI was established to investigate the
hypothesis that specific coronary morphologies have a genetic
component. Detailed heritable disease patterns could enhance
clinical risk characterization and offer new opportunities for
clinical follow-up investigations in relatives of affected patients.
Methods
Ascertainment Strategy
CAD families were ascertained through index patients at 15 cardiac
rehabilitation centers distributed throughout Germany. All index
Received June 18, 2004; revision received November 3, 2004; accepted November 10, 2004.
From the Human and Molecular Genetics Center (M.F., U.B., A.B., H.J.J.), Medical College of Wisconsin, Milwaukee, Wis; the Clinic for Internal
Medicine II (M.F., S.H., A.B., C.H., G.R.), University of Regensburg, Regensburg, Germany; the Clinic for Internal Medicine II (B.M., J.E., H.S.),
University of Luebeck, Luebeck, Germany; and the LVA Cardiac Rehabilitation Clinic Höhenried (G.K.), Bernried, Germany.
The online-only Data Supplement, which contains a table, can be found with this article at http://www.circulationaha.org.
Correspondence to Dr Ulrich Broeckel, Human and Molecular Genetics Center, Medical College of Wisconsin, 8701 Watertown Plank Rd, Milwaukee,
WI 53226 (e-mail [email protected]); or Prof Dr Heribert Schunkert, Clinic for Internal Medicine II, University of Luebeck, Ratzeburger Allee 160,
23538 Luebeck, Germany (e-mail [email protected]).
© 2005 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org
DOI: 10.1161/01.CIR.0000155611.41961.BB
855
856
Circulation
February 22, 2005
patients had suffered from an MI before 60 years of age. If at least
1 sibling presented with MI or severe CAD before 70 years of age,
defined by percutaneous coronary intervention or coronary artery
bypass grafting, the nuclear family (index patient, available parents,
and all affected and unaffected siblings) was contacted and invited to
participate in the study. In total, 401 families had coronary angiograms available for retrospective review in at least 2 siblings (882
individuals). Angiograms of index cases and at least 1 sibling were
available for 309 families. In the remaining 92 families, 2 or more
siblings with a coronary angiogram were identified apart from the
index patient. The experimental design and technical operations have
been defined in advance. The Ethics Committees of the University of
Regensburg and the Medical College of Wisconsin approved the
study protocol, and all participants gave written, informed consent.
Subjects and Phenotyping
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
The analyses included 369 pedigrees with 2 affected siblings and 32
pedigrees with ⬎2 affected siblings (whites of northern European
origin). From all study participants, answers to a standardized
questionnaire were obtained by specially trained telephone interviewers; questions related to medical history, presence of coronary
risk factors, clinical events, medication, anthropometric data, and
socioeconomic background. This information was validated by
retrospective analyses of medical records from hospital stays and
primary physicians. Additionally, all patients underwent a medical
examination during a visit scheduled at their primary care physician’s office. If a relative had suffered from an MI or had undergone
a coronary revascularization procedure, the respective information
was verified in hospital discharge summaries, which were independently reviewed by 2 investigators. The diagnosis of MI was
documented as described previously.14
Risk Factors
All cardiovascular risk factors were ascertained at the time of the
first interview in a standardized fashion. We defined hypertension as
a blood pressure ⬎140/90 mm Hg. Hypercholesterolemia was defined as an LDL cholesterol value ⬎130 mg/dL or by the use of
lipid-lowering agents. Current or former cigarette smoking (cessation at time of the index event) was used to define smoking status.
Diabetes mellitus was defined by the use of antidiabetic medication
or by elevated glycosylated hemoglobin (ⱖ6.5%). All biochemical
measurements were carried out in a single laboratory as previously
reported.14
Angiographic Evaluation
Coronary angiograms were scored systematically and in random
order by a single, experienced, interventional cardiologist. Much
effort was expended to ensure the objectivity of evaluation of the
angiograms. For this purpose, the reader was blinded with regard to
family structure, and angiograms of family members were read on
different days separated by the reading of other angiograms. For the
evaluation of angiographic phenotypes, we categorized the coronary
vasculature into the right and left coronary artery, as well as 18
coronary segments according to the classification of the Coronary
Artery Surgery Study.15 On a standardized evaluation form, each of
these coronary segments was characterized for several anatomic and
morphological criteria: (1) The degree of stenosis was derived by
relation of the stenotic segment to the diameter of the nearest
reference segment, and the percent diameter stenosis for each lesion
was graded on a scale of 0 to 4, where 0⫽no stenosis, 1⫽diameter
stenosis ⬍50%, 2⫽diameter stenosis 50% to 69%, 3⫽diameter
stenosis 70% to 99%, and 4⫽total occlusion. A lesion compromising
the lumen by ⬎50% was considered to represent a significant
stenosis. (2) Left main CAD was defined as a lesion proximal to the
bifurcation, including ostial stenosis. (3) “Diffuse CAD” was defined
in cases where more than two thirds of the left or right coronary
vessel tree was affected by wall irregularities or stenoses. (4) The
morphology of stenoses was characterized in terms of length of the
stenotic area (“long”⫽stenosis length ⬎10 mm, “short”⫽stenosis
length ⱕ10 mm) and in terms of ectatic (aneurysmal) coronary
lesions (defined by a vessel diameter ⬎125% of the healthy
reference vessel). (5) A semiquantitative scoring system was used to
grade the presence of calcifications of the coronary vessels (0⫽no
calcification, 1⫽minimal calcifications [visible in 1 or 2 angulations], 2⫽moderate calcifications [visible in all angulations], 3⫽severe [⬎1 cm] or circular calcifications). The calcification phenotype
was analyzed as a dichotomous variable: yes (score 1 to 3) or no
(score 0).
Coronary blood supply was divided into right-dominant, balanced,
and left-dominant.16 Morphological CAD was classified as 1-, 2- or
3-vessel disease. Furthermore, because a number of “jeopardy
scores” have been formerly used to quantify plaque burden and
predict patient-based clinical outcomes, the data on grades of
stenosis in individual segments for each patient were also used to
compute a score for the extent of CAD. (Modified Gensini-index;
segments were weighted by a value from 0.5 to 5.0. Percent diameter
stenosis is weighted from 1 to 16. The product of these 2 weights is
the total weight for each arterial segment. The severity score is the
sum of all total weights.)15 With this modified Gensini index, a
heavier weight was assigned to the more severe luminal narrowings.
Statistical Analysis
Because the sampling of multiple individuals from the same family
does create within-family correlation, the comparison of clinical and
angiographic characteristics between index cases and their siblings
(1 or more) was analyzed by generalized estimating equations, which
are an extension of the generalized linear models methodology
applied to correlated data.17
Heritability estimates (h2) were obtained by variance-component
methodology as implemented in the SOLAR (version 2.1.2) package.18 This approach has been expanded to allow for pedigrees of
arbitrary size and complexity as well as for dichotomous traits by
assuming that an individual belongs to a specific affected status if an
underlying genetically determined risk (ie, liability) exceeds a
certain threshold.19 The latent liability is assumed to have an
underlying multivariate normal distribution. In addition to the
variance components, covariate effects and environmental factors
were included in the model: In all of the analyses, age and sex were
included as covariates; additional covariates included the cardiovascular risk factors body mass index, mean arterial blood pressure,
LDL-HDL ratio, diabetes, and cigarette smoking. Both variance
components and covariate effects were estimated simultaneously by
maximum-likelihood techniques. The null hypothesis of no genetic
effect (h2⫽0) on phenotype was tested with a likelihood-ratio test by
comparing the likelihood of a restricted model in which the parameter h2 was constrained to a value of 0 with that for a general model
in which the same parameter was estimated. Because our samples
were ascertained via index patients, we conditioned the likelihood of
a family on the phenotype of the initial proband to account partially
for the nonrandom sampling.20
As an additional measure of familial aggregation, we performed
logistic regression analyses of selected phenotypes to assess the ratio
of sibling risk for affected individuals to sibling risk for unaffected
individuals (conditional on the index case’s affected status), adjusted
for age and sex.
Because a high degree of correlation between stenoses at different
coronary sites was observed (all Spearman’s rho coefficients
P⬍0.001), principal-component factor analysis was performed to
determine subsets of clusters of correlated stenoses (“factors”). For
the determination of the optimal number of factors being extracted,
the Kaiser-Guttman rule was used.21 This analysis resulted in 5
noncorrelated factors, which were interpreted on the basis of their
factor loadings (correlation between the generated factor and the
original trait, Data Supplement Table and Figure 2). Factor loadings
with coefficient absolute values ⬎0.25 were considered significant.
After excluding subjects with missing information, factor analysis
was carried out for 693 individuals. This subset had features similar
to those of the entire data set. To estimate the actual values of the
factors for individual cases (observations), factor scores were derived and used as quantitative measurements to further estimate the
heritability of the significant principal components. Bivariate genetic
Fischer et al
TABLE 1. Clinical Characteristics Among Index MI Cases and
Their Siblings
Index MI Cases Affected Siblings
(n⫽309)
(n⫽573)
Heritable Patterns of Angiographic CAD
857
TABLE 2. Angiographic Characteristics Among Index MI Cases
and Their Siblings
P
Index MI Cases
(n⫽309)
Affected Siblings
(n⫽572)
P
Age at interview, y
58.2⫾0.4
62.1⫾0.3
⬍0.001
Age at index event, y
50.6⫾0.4
56.6⫾0.3
⬍0.001
Right-dominant, %
75.3
76.8
NS
87.1
72.4
⬍0.001
Balanced, %
14.5
12.8
NS
309/135/162
304/217/345
⬍0.001
Left-dominant, %
10.2
10.4
NS
Sex, % male
MI/PTCA/CABG, n
Risk factors
BMI, kg/m2
Circulation, %
Extent of CAD
27.1⫾0.2
27.0⫾0.2
NS
1-Vessel disease, %
23.1
19.9
NS
NS
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Systolic BP, mm Hg
140⫾1
145⫾1
0.002
2-Vessel disease, %
33.1
31.5
Diastolic BP, mm Hg
84⫾1
84⫾1
NS
3-Vessel disease, %
40.3
43.8
NS
Cholesterol, mg/dL
229⫾3
233⫾2
NS
Severity score, weight points
39.9⫾1.8
44.5⫾1.4
0.052
LDL/HDL
3.5⫾0.1
3.4⫾0.1
NS
No. of stenoses, LCA
2.4⫾0.1
2.7⫾0.1
0.023
Arterial hypertension, %
56.6
63.0
NS
No. of stenoses, RCA
1.1⫾0.1
1.2⫾0.1
NS
Hypercholesterolemia, %
79.6
77.8
NS
LCA diffusely affected, %
39.9
46.8
0.048
RCA diffusely affected, %
46.8
44.8
NS
NS
Diabetes, %
14.2
15.7
NS
Current or former smoker, %
77.2
66.7
0.006
Antihypertensive drugs, %
80.9
79.4
NS
LCA, %
90.6
90.2
Lipid-lowering drugs, %
57.0
54.1
NS
RCA, %
71.9
67.8
NS
Left main CAD,* %
13.6
19.3
NS
Proximal,† %
38.3
45.6
0.035
Distal,‡ %
45.7
37.2
0.016
Bifurcational, %
20.4
21.8
NS
28.6
29.6
NS
PTCA indicates percutaneous transluminal coronary angioplasty; CABG,
coronary artery bypass graft; BMI, body mass index; and BP, blood pressure.
Other abbreviations are as defined in text. Values are mean⫾SE or %. P values
were derived from GEE analysis.
Localization of stenoses
Morphology of stenoses
analysis was used to partition the phenotypic correlation between a
given pair of these quantitative traits into their additive genetic and
random environmental components by means of information on
genetic relationships between relatives and the maximum-likelihood
technique. By using likelihood-ratio tests, the significance of the
additive genetic correlation was determined.
Results
Clinical and Angiographic Characteristics
Clinical and angiographic characteristics of index MI cases
and their siblings are presented in Tables 1 and 2. As
expected, index cases experienced their index event at significantly younger age and were more often male. With the
exception of systolic blood pressure, risk factor distribution
was very comparable between index patients and their
siblings.
Siblings of MI cases presented with a slightly more
severe morphological manifestation of CAD than did index
cases, as indicated by a higher severity score, a higher
number of stenoses, and more diffuse CAD. Moreover, the
prevalence of proximal disease was higher in siblings than
in index MI cases. With regard to morphological characteristics, no differences between the 2 groups could be
observed.
Heritability Analysis
Results of the heritability analyses are listed in Table 3. A
significant degree of inherited influence on age- and
sex-adjusted phenotypic variance could be detected for
proximal localizations of stenoses, in particular, stenoses
located at the left main coronary artery (h2⫽0.49⫾0.12;
Stenosis length ⬎10 mm, %
Ectatic lesions, %
22.6
21.4
NS
Calcification LCA or RCA, %
67.2
65.6
NS
Calcification of LCA, %
63.6
63.8
NS
Calcification of RCA, %
23.8
23.9
NS
LCA indicates left coronary artery; RCA, right coronary artery. Other
abbreviations are as defined in text. P values were derived from GEE analysis.
*Mainstem stenoses and ostial lesions.
†Stenoses located at the ostia, left mainstem, proximal left anterior
descending artery, proximal circumflex artery, and proximal RCA.
‡Distal disease without evidence of proximal disease.
P⫽0.01). In contrast, no heritability was found for distal
disease (h2⫽0.05⫾0.19; NS).
Moreover, a high proportion of the variation in coronary
calcification was due to inherited effects (left coronary
artery: h2⫽0.42⫾0.12; P⫽0.002; right coronary artery:
h2⫽0.84⫾0.12; P⫽0.0000001). Furthermore, genetic effects markedly influenced the presence of ectatic coronary
segments (h2⫽0.52⫾0.07; P⫽0.001). Finally, age of onset
and all cardiovascular risk factors were highly heritable.
With regard to the dichotomous parameter “focal versus
diffuse disease,” no consistent genetic contribution could be
observed; however, focusing the comparison on the left and
the right coronary artery separately, we realized that these
severity parameters showed higher heritabilities in the left
coronary artery and reached significance. The number of
diseased vessels and the length of lesions did not show
significant heritable components.
Moreover, a multivariate analysis was carried out to test
the contribution of risk factors in addition to age and sex.
858
Circulation
February 22, 2005
TABLE 3. Age- and Sex-Adjusted and Multivariable-Adjusted Estimates of Heritability for Selected CAD Phenotypes and
CAD Risk Factors
Age- and
Sex-Adjusted
h 2⫾SE
P
Proportion of
Variance due to
Covariates/KL⫺R 2储
䡠䡠䡠
MultivariableAdjusted
h 2⫾SE
P
Proportion of
Variance due to
Covariates/KL⫺R 2储
⬍0.10
NS
䡠䡠䡠
䡠䡠䡠
0.04
⬍0.10
NS
䡠䡠䡠
⬍0.10
NS
䡠䡠䡠
䡠䡠䡠
0.04
⬍0.10
NS
䡠䡠䡠
0.09
Distribution of CAD
1-Vessel disease, yes/no
⬍0.10
NS
2-Vessel disease, yes/no
⬍0.10
NS
3-Vessel disease, yes/no
0.11⫾0.10
NS
⬍0.10
NS
Severity score, modified Gensini
No. of stenoses, LCA
0.22⫾0.12
No. of stenoses, RCA
⬍0.10
No. of stenoses, proximal
No. of stenoses, distal
0.28⫾0.10
⬍0.10
LCA diffusely affected, yes/no
0.25⫾0.12
RCA diffusely affected, yes/no
⬍0.10
0.03
NS
0.003
NS
0.04
NS
0.19⫾0.12
⬍0.10
䡠䡠䡠
0.03
0.22⫾0.11
䡠䡠䡠
0.05
0.41⫾0.17
䡠䡠䡠
⬍0.10
⬍0.10
0.06
NS
0.02
NS
䡠䡠䡠
0.05
0.005
䡠䡠䡠
0.07
NS
䡠䡠䡠
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Localization of stenoses
Left main disease,* yes/no
0.49⫾0.12
0.01
0.03
0.47⫾0.09
0.01
0.05
Proximal disease,† yes/no
0.32⫾0.05
0.02
0.02
0.32⫾0.06
0.02
0.02
䡠䡠䡠
0.02
⬍0.10
NS
0.25⫾0.21
NS
Distal disease,‡ yes/no
Bifurcational, yes/no
⬍0.10
NS
0.23⫾0.17
NS
䡠䡠䡠
0.02
Morphology of stenoses
Length ⬎10 mm, yes/no
⬍0.10
NS
⬍0.10
NS
0.52⫾0.07
0.001
䡠䡠䡠
0.01
0.54⫾0.08
0.001
䡠䡠䡠
0.01
LCA or RCA, yes/no
0.51⫾0.17
0.001
0.02
0.51⫾0.07
0.001
0.02
LCA, yes/no
0.42⫾0.12
0.002
0.02
0.36⫾0.12
0.01
0.02
RCA, yes/no
0.84⫾0.12
0.0000001
0.02
0.81⫾0.18
0.000005
0.03
⬍0.10
NS
Ectatic lesions, yes/no
Calcification
⬍0.10
NS
0.55⫾0.08
⬍0.0000001
0.02
0.53⫾0.08
⬍0.0000001
䡠䡠䡠
0.06
Diabetes, yes/no
0.69⫾0.13
⬍0.0000001
0.02
0.81⫾0.20
0.0000004
0.02
Systolic BP, mm Hg
0.24⫾0.06
0.00001
0.04
0.19⫾0.06
0.0008
0.10
Diastolic BP, mm Hg
0.23⫾0.06
0.0001
0.02
0.18⫾0.06
0.0009
0.10
LDL, mg/dL
0.37⫾0.07
⬍0.0000001
0.01
0.47⫾0.07
⬍0.0000001
0.05
HDL, mg/dL
0.49⫾0.07
⬍0.0000001
0.09
0.48⫾0.07
⬍0.0000001
0.14
BMI, kg/m2
0.44⫾0.07
⬍0.0000001
0.01
0.46⫾0.07
⬍0.0000001
0.10
Circulation, right/balanced/left
Age of onset, y§
Risk factors
See text and footnotes to Tables 1 and 2 for explanation of abbreviations.
*Mainstem stenoses and ostial lesions.
†Stenoses located at the ostia, left mainstem, proximal left anterior descending artery, proximal circumflex artery, and proximal RCA.
‡Distal disease without evidence of proximal disease.
§Age was not considered as a covariate.
储For continuous traits, the proportion of variance due to all covariates and for discrete traits the Kullback-Leibler R2 (⫽KL⫺R2, which measures
the proportionate reduction in uncertainty due to the inclusion of covariates), are given. This analysis was performed for significant traits only.
Incorporating additional cardiovascular risk factors into the
variance-component analysis did not significantly influence
the heritability estimates of coronary morphological features
(Table 3).
Additionally, age- and sex-adjusted odds ratios and 95%
confidence intervals of second siblings’ affected status conditional on the index patients’ affected status for the specific
phenotype are shown (Figure 1). These risks largely confirmed our heritability estimates.
Principal-Component Factor Analysis
Using the detailed angiographic information, we established
whether certain localizations of coronary lesions are preferentially influenced genetically. Because a high degree of
correlation between stenoses at different coronary sites was
observed (all Spearman’s rho coefficients P⬍0.001; data not
shown), principal-component factor analysis was performed
to reduce the number of variables and to determine subsets of
clusters of correlated coronary lesions. Subsequently, we
Fischer et al
Heritable Patterns of Angiographic CAD
859
Discussion
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Figure 1. Age- and sex-adjusted odds ratios with 95% confidence intervals as derived from logistic regression analyses of
selected phenotypes assessing ratio of sibling risk for affected
individuals to sibling risk for unaffected individuals (conditional
on index patients’ affected status). In total, 309 index MI cases
were included in analysis. Numbers of affected and unaffected
index patients for each trait are given in parentheses. *Mainstem
stenoses without evidence of ostial lesions; †stenoses located
at ostia, left mainstem, proximal left anterior descending artery,
proximal circumflex artery, and proximal right coronary artery;
‡distal disease without evidence of proximal disease. LCA indicates left coronary artery; RCA, right coronary artery.
determined whether such clustered manifestations of CAD
are heritable. Five distinct factors (ie, regions) could be
extracted as described in Figure 2 and in the Data Supplement
Table. In summary, on the basis of this factor analysis, a
reduction of the initial 18 coronary segments to these 5
independent coronary regions appeared appropriate for further analyses. Heritabilities were estimated for all 5 factors
(Figure 2). Whereas heritability estimates for factor 1 (ostia
and left mainstem; h2⫽0.32⫾0.13; P⫽0.008) and factor 2
(proximal left anterior descending, proximal circumflex, and
proximal right coronary artery; h2⫽0.30⫾0.13; P⫽0.01)
were significant, no genetic contribution could be detected for
factors 3 through 5 that summarized the peripheral localizations of coronary arteries. Using bivariate analysis, we
estimated the genetic correlation between each of these
factors. The genetic correlation between each of these traits
was weak and did not reach statistical significance (data not
shown). Thus, different genetic mechanisms rather than a
pleiotropic action of one gene or a group of genes can be
assumed for the development of CAD.
We investigated the hypothesis that various morphological
characteristics of CAD have a genetic component. To
dissect the complex diagnosis of CAD into heritable
components, variance-component analysis on specific anatomic and morphological features was performed on
coronary angiograms of 882 individuals from 401 families
with CAD. The data demonstrate that several angiographically distinct aspects of CAD are heritable: however, the
degree of the genetic contribution varies in different areas
of the coronary arteries. Particularly, a highly heritable
component was observed for hazardous manifestations of
coronary atherosclerosis, namely, ostial lesions and left
main CAD. To our knowledge, this is the first analysis to
dissect the complex phenotype of coronary atherosclerosis
and to explore the genetic contribution to the anatomic
localization of coronary lesions.
So far, only a few case reports have described angiographic findings in twin pairs with CAD. Our review of the
available literature shows that 9 of 12 monozygotic twin
pairs displayed concordance for proximal lesions, whereas
only 3 of 12 were concordant for distal lesions.5–13 This
finding is congruent with our observation that stenoses are
particularly heritable at proximal localizations. The reason
for different heritability estimates of proximal and distal
lesions remains speculative. One possibility might relate to
the distinct ontogenetic determination of both sites. Particularly, the proximal parts of the coronary arteries
develop from origins distinct from the peripheral parts.22
Whereas the proximal portion of the coronary arteries
develops as buds on the walls of the truncus arteriosus, the
distal portion is the forerunner of the subepicardial
branches of the coronary arteries and develops as a
subepicardial vascular network.22 Alternatively, constellations of risk factors or local hemodynamic factors, especially alterations of arterial shear stress, may account for
different angiographic appearances, as suggested by dissimilar findings of angiographic CAD.23,24 However, the
distinction between proximal and distal disease with respect to the degree of heritability was consistent in
multiple models that adjusted for measurable risk factors
or anatomic features.
Consistent with other reports based on fast-gated helical
and electron-beam computed tomography,25,26 a major
genetic contribution in the variation of angiographically
derived vascular calcification was found. Furthermore,
heritability estimates for calcification at other anatomic
sites, such as abdominal aortic and carotid calcific deposits, were found to be of the same magnitude.27,28 In
conjunction, these studies provide strong and consistent
evidence that vascular calcification is influenced by sofar-unknown genetic factors.
It has been shown previously that positive remodeling
and aneurysmal coronary lesions of human coronary arteries occur in response to the development of intimal
plaques29,30 and are modulated by diabetes mellitus, smoking, and hypercholesterolemia.31–33 Our findings suggest
that coronary remodeling is also influenced by genetic
factors. Interestingly, Nishioko et al34 found that proximal
860
Circulation
February 22, 2005
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Figure 2. Schematic overview of factors
resulting from principal-component factor analysis with heritability values. This
analysis was performed because high
degree of correlation between stenoses
at different coronary sites was observed
to determine subsets of clusters of correlated stenoses (“factors”) and to
reduce number of variables. Heritability
estimates were based on 251 families
with at least 2 siblings (n⫽578 individual
siblings). Factor 1: right ostium, left
ostium, and left mainstem. Factor 2:
proximal left anterior descending coronary artery (LAD), proximal circumflex
(CX), and proximal right coronary artery
(RCA). Factor 3: “peripheral LAD region.”
Factor 4: “peripheral CX region.” Factor
5: “peripheral RCA region.” Mid-RCA
segment and marginal I segment did not
reach level of significance.
coronary segments show more prominent compensatory
enlargement than do distal segments of the same patients,
albeit a similar degree of luminal narrowing. These results
support our findings, because proximal stenoses and arterial enlargements are influenced to a higher degree by
genetic factors.
Further observations of the present study include the
similarity between sibling pairs with respect to the initial
age at clinical manifestation of CAD and the coronary risk
profile. Indeed, the data suggest that the impact of genetic
factors in the development of CAD may be complex and is
modulated by age and coronary risk factors. By contrast,
the absence of concordance for the dominant coronary
vessel implies involvement of nonhereditary factors in the
modulation of coronary anatomy. This finding is in agreement with other observational studies, which found considerable variability in the coronary blood supply in
monozygotic twins.5,7–10
With regard to future research, our findings may open a
new window of opportunity for screening strategies in
relatives of CAD patients. Focusing on the most heritable
manifestations of CAD might be an effective strategy to
increase the power to detect susceptibility loci for such
complex diseases. In fact, the high heritability of left main
CAD may have important clinical implications, including a
refined risk prediction in relatives of patients affected with
such particularly severe manifestations. In this context,
noninvasive imaging techniques might become established
methods for routine cardiac diagnostics.35 The high negative predictive value of magnetic resonance imaging or
multislice computed tomography scans suggests that these
methods may have a role in ruling out clinically significant
coronary disease in subgroups of patients with medium- to
high-risk scores, ie, in relatives of patients with left main
CAD. The clinical value and cost-effectiveness of this
approach remain to be determined.
The high variability of the morphological aspects and
related degrees of heritability suggest a diversity of genetic
mechanisms that may require consideration when genetic
variants are being associated with CAD. Currently, most
genetic investigations of CAD use a dichotomous affected/
unaffected phenotypic specification, irrespective of morphological characteristics of the coronary arteries. Our
results indicate, however, that a narrowly defined phenotype can offer advantages over broad definitions.
Some limitations of the present study should be mentioned. Our heritability estimates may overestimate the
true genetic contribution because we did not account for
shared environmental exposures among family members in
our model. Although most siblings (⬎90%) reported living
in separate households from one another and their parents,
it might be possible that a shared environment early in life
contributes to specific phenotypes seen in adulthood.
Furthermore, in a traditional sib-pair design, the heritability is estimated by sibling correlations. With this approach,
already deceased siblings could not be included in the
analyses. Our study sample was ascertained via index
cases for MI before 60 years of age, and a selection bias
might therefore exist in that the heritability of coronary
morphology may decrease with advancing age. Nevertheless, sibling recurrence risks conditioned on the index
patients’ affected status revealed congruent results to our
heritability estimates. Another form of selection bias might
exist in that siblings of MI patients seek medical help more
quickly when a family member experiences an MI. As a
result, siblings with advanced or peripheral CAD might be
underrepresented in our study sample. However, two thirds
of second siblings underwent angiography before the index
cases did. In fact, a secondary analysis in which the index
population consisted of those who underwent angiography
after their siblings resulted in similar heritability estimates
(data not shown).
Fischer et al
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Some of the heritability estimates, particularly lowfrequency traits, should be interpreted cautiously because
of the fact that a considerable number of siblings were
concordant for not expressing the phenotype of interest. In
this regard, the low-frequency traits ostial stenoses and left
mainstem stenoses have been summarized as “left main
CAD” to increase the frequency for estimating heritability.
Notably, recurrence risks to siblings were highly significant for both ostial and left mainstem stenoses. Moreover,
by means of principal-component factor analysis, we
determined factor scores that quantified individual cases
on a latent continuum. Thereby, factor analysis provides a
basis for quantitative phenotypes for each individual for
further analysis. We demonstrated that the heritability
estimates obtained for these quantitative traits were also
congruent with our initial heritability estimates. In our
retrospective study, cardiovascular risk factors were ascertained after index cardiac catheterization. We preferred to
ascertain valid risk factors systematically from all individuals homogeneously at the same time, although misclassifications due to initiated therapy might occur. However,
prospective ascertainment of risk factors before MI was
not feasible, and retrospective survey of risk factors before
the index coronary event would be flawed, too. The issue
of multiple testing should be mentioned. In this regard, we
report alternative measures of familial aggregation by use
of different analytical approaches: Heritability estimates,
sibling recurrence risks, and results from factor analysis
essentially conveyed the same trend. Thus, it is unlikely
that our findings are solely explained by chance. Finally,
our cohort was restricted to whites, such that our findings
may not be generalizable to other ethnicities. Altogether, a
limited generalizability of our heritability estimates might
be considered, and independent confirmation is warranted.
With the ultimate goal of understanding the genetic
basis of CAD, we have demonstrated that multiple distinct
features of coronary atherosclerosis display a heritable
component. The high variability of morphological aspects
and related degrees of heritability suggest a diversity of
genetic mechanisms that may require consideration when
genetic variants are being associated with CAD. Moreover,
the high heritability of particularly severe manifestations
of CAD, including ostial lesions and left main disease, may
have important clinical implications for screening strategies in asymptomatic relatives.
Acknowledgments
Dr Fischer was supported by the Deutsche Akademie für Naturforscher Leopoldina carried by allowances of the Bundesministerium
für Bildung und Forschung (BMBF-LPD 9901/8-72). Drs Jacob and
Broeckel are supported in part by grants from the National Heart,
Lung, and Blood Institute (hypertension SCOR program (HL54998)
and the SCOR program on ischemic heart disease (HL65203 and
HL074321). Furthermore, the study was supported by the Deutsche
Forschungsgemeinschaft (Schu672/10-1, Schu672/12-1, Schu672/
14-1, Ho1073/8-1, He1921/9-1, and BA2227/1-1), the Federal Ministry for Research (KBF-FKZ 01GB9403, to Dr Schunkert), the
National Genome Network (01GS0418 to Drs Schunkert and Hengstenberg), the Ernst- und Berta-Grimmke-Stiftung (Drs Hengstenberg and Schunkert), the Wilhelm-Vaillant-Stiftung (Drs Hengsten-
Heritable Patterns of Angiographic CAD
861
berg, Schunkert, and Holmer), and the Deutsche Stiftung für
Herzforschung (Drs Hengstenberg and Schunkert), Germany.
References
1. Lloyd-Jones DM, Nam BH, D’Agostino RB Sr, Levy D, Murabito JM,
Wang TJ, Wilson PW, O’Donnell CJ. Parental cardiovascular disease as
a risk factor for cardiovascular disease in middle-aged adults: a prospective study of parents and offspring. JAMA. 2004;291:2204 –2211.
2. Marenberg ME, Risch N, Berkman LF, Floderus B, de Faire U. Genetic
susceptibility to death from coronary heart disease in a study of twins.
N Engl J Med. 1994;330:1041–1046.
3. Schunkert H, Fischer M. Old and simple tools may do better—sometimes.
Eur Heart J. 2002;23:1900 –1902.
4. Rice JP, Saccone NL, Rasmussen E. Definition of the phenotype. Adv
Genet. 2001;42:69 –76.
5. Holmes DR Jr, Kennel AJ, Smith HC, Gordon H, Moore SB. Coronary
artery disease in twins. Br Heart J. 1981;45:193–197.
6. Kreulen TH, Cohn PF, Gorlin R. Premature coronary artery disease in
identical male twins studied by selective coronary arteriography. Cathet
Cardiovasc Diagn. 1975;1:91–96.
7. Schilling RJ, Norell MS. Coronary artery disease in identical twins. Eur
Heart J. 1995;16:134 –135.
8. Russo P, Steffenino G, Dellavalle A, Ribichini F. Concomitant myocardial infarction in identical twins with similar coronary lesions. G Ital
Cardiol. 1995;25:341–343.
9. Frings AM, Mayer B, Bocker W, Hengstenberg C, Willemsen D, Riegger
GA, Schunkert H. Comparative coronary anatomy in six twin pairs with
coronary artery disease. Heart. 2000;83:47–50.
10. Nathoe HM, Stella PR, Eefting FD, de Jaegere PP. Angiographic findings
in monozygotic twins with coronary artery disease. Am J Cardiol. 2002;
89:1006 –1009.
11. Samuels LE, Samuels FS, Thomas MP, Morris RJ, Wechsler AS.
Coronary artery disease in identical twins. Ann Thorac Surg. 1999;68:
594 – 600.
12. Kaluza G, Abukhalil JM, Raizner AE. Identical atherosclerotic lesions in
identical twins. Circulation. 2000;101:e63– e64.
13. Herrington DM, Pearson TA. Clinical and angiographic similarities in
twins with coronary artery disease. Am J Cardiol. 1987;59:366 –367.
14. Broeckel U, Hengstenberg C, Mayer B, Holmer S, Martin LJ, Comuzzie
AG, Blangero J, Nurnberg P, Reis A, Riegger GA, Jacob HJ, Schunkert
H. A comprehensive linkage analysis for myocardial infarction and its
related risk factors. Nat Genet. 2002;30:210 –214.
15. Ringqvist I, Fisher LD, Mock M, Davis KB, Wedel H, Chaitman BR,
Passamani E, Russell RO Jr, Alderman EL, Kouchoukos NT, Kaiser GC,
Ryan TJ, Killip T, Fray D. Prognostic value of angiographic indices of
coronary artery disease from the Coronary Artery Surgery Study (CASS).
J Clin Invest. 1983;71:1854 –1866.
16. Baim D, Grossman W. Grossman’s Cardiac Catheterization,
Angiography, and Intervention. 6th ed. Baltimore, Md: Lippincott
Williams & Wilkins; 2000.
17. Liang KY, Zeger SL. Regression analysis for correlated data. Annu Rev
Public Health. 1993;14:43– 68.
18. Almasy L, Blangero J. Multipoint quantitative-trait linkage analysis in
general pedigrees. Am J Hum Genet. 1998;62:1198 –1211.
19. Duggirala R, Williams JT, Williams-Blangero S, Blangero J. A variance
component approach to dichotomous trait linkage analysis using a
threshold model. Genet Epidemiol. 1997;14:987–992.
20. Comuzzie AG, Williams JT. Correcting for ascertainment bias in the
COGA data set. Genet Epidemiol. 1999;17(suppl 1):S109 –S114.
21. Stevens J. Applied multivariate statistics for the social sciences. In:
Applied Multivariate Statistics for the Social Sciences. Mahwah, NJ:
Lawrence Erlbaum Associates; 1996:362–392.
22. Luedinghausen Mv. The Clinical Anatomy of the Coronary Arteries.
Berlin/Heidelberg/New York: Springer; 2003.
23. Bogaty P, Brecker SJ, White SE, Stevenson RN, el-Tamimi H, Balcon R,
Maseri A. Comparison of coronary angiographic findings in acute and
chronic first presentation of ischemic heart disease. Circulation. 1993;
87:1938 –1946.
24. Sharp SD, Williams RR, Hunt SC, Schumacher MC. Coronary risk
factors and the severity of angiographic coronary artery disease in
members of high-risk pedigrees. Am Heart J. 1992;123:279 –285.
25. Wagenknecht LE, Bowden DW, Carr JJ, Langefeld CD, Freedman BI,
Rich SS. Familial aggregation of coronary artery calcium in families with
type 2 diabetes. Diabetes. 2001;50:861– 866.
862
Circulation
February 22, 2005
26. Peyser PA, Bielak LF, Chu JS, Turner ST, Ellsworth DL, Boerwinkle E,
Sheedy PF 2nd. Heritability of coronary artery calcium quantity measured
by electron beam computed tomography in asymptomatic adults. Circulation. 2002;106:304 –308.
27. O’Donnell CJ, Chazaro I, Wilson PW, Fox C, Hannan MT, Kiel DP,
Cupples LA. Evidence for heritability of abdominal aortic calcific
deposits in the Framingham Heart Study. Circulation. 2002;106:337–341.
28. Hunt KJ, Duggirala R, Goring HH, Williams JT, Almasy L, Blangero J,
O’Leary DH, Stern MP. Genetic basis of variation in carotid artery plaque
in the San Antonio Family Heart Study. Stroke. 2002;33:2775–2780.
29. Glagov S, Weisenberg E, Zarins CK, Stankunavicius R, Kolettis
GJ. Compensatory enlargement of human atherosclerotic coronary
arteries. N Engl J Med. 1987;316:1371–1375.
30. McPherson DD, Sirna SJ, Hiratzka LF, Thorpe L, Armstrong ML, Marcus
ML, Kerber RE. Coronary arterial remodeling studied by high-frequency
epicardial echocardiography: an early compensatory mechanism in
patients with obstructive coronary atherosclerosis. J Am Coll Cardiol.
1991;17:79 – 86.
31. Kornowski R. Impact of smoking on coronary atherosclerosis and
remodeling as determined by intravascular ultrasonic imaging. Am J
Cardiol. 1999;83:443– 445, A9.
32. Kornowski R, Mintz GS, Lansky AJ, Hong MK, Kent KM, Pichard AD,
Satler LF, Popma JJ, Bucher TA, Leon MB. Paradoxic decreases in
atherosclerotic plaque mass in insulin-treated diabetic patients. Am J
Cardiol. 1998;81:1298 –304.
33. Tauth J, Pinnow E, Sullebarger JT, Basta L, Gursoy S, Lindsay J Jr, Matar
F. Predictors of coronary arterial remodeling patterns in patients with
myocardial ischemia. Am J Cardiol. 1997;80:1352–1355.
34. Nishioka T, Nagai T, Luo H, Kitamura K, Hakamata N, Akanuma M,
Katsushika S, Uehata A, Takase B, Isojima K, Ohtomi S, Siegel RJ.
Coronary remodeling of proximal and distal stenotic atherosclerotic
plaques within the same artery by intravascular ultrasound study. Am J
Cardiol. 2001;87:387–391.
35. Kim WY, Danias PG, Stuber M, Flamm SD, Plein S, Nagel E, Langerak
SE, Weber OM, Pedersen EM, Schmidt M, Botnar RM, Manning WJ.
Coronary magnetic resonance angiography for the detection of coronary
stenoses. N Engl J Med. 2001;345:1863–1869.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Distinct Heritable Patterns of Angiographic Coronary Artery Disease in Families With
Myocardial Infarction
Marcus Fischer, Ulrich Broeckel, Stephan Holmer, Andrea Baessler, Christian Hengstenberg,
Bjoern Mayer, Jeanette Erdmann, Gernot Klein, Guenter Riegger, Howard J. Jacob and Heribert
Schunkert
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Circulation. published online February 14, 2005;
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2005 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/early/2005/02/14/01.CIR.0000155611.41961.BB.citation
Data Supplement (unedited) at:
http://circ.ahajournals.org/content/suppl/2005/02/18/01.CIR.0000155611.41961.BB.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/