Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/wolters-kluwer-health/heritability-of-coronary-artery-disease-insights-from-a-classical-twin-iCXtxcNsnm
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- Wolters Kluwer Health
- Copyright
- © 2022 The Authors.
- ISSN
- 1941-9651
- eISSN
- 1942-0080
- DOI
- 10.1161/circimaging.121.013348
- Publisher site
- See Article on Publisher Site
Abstract
Circulation: Cardiovascular Imaging ORIGINAL ARTICLE Heritability of Coronary Artery Disease: Insights From a Classical Twin Study Zsofia D. Drobni , MD, PhD; Marton Kolossvary , MD, PhD; Julia Karady , MD; Adam L. Jermendy, MD, PhD; Adam D. Tarnoki, MD, PhD; David L. Tarnoki, MD, PhD; Judit Simon , MD; Balint Szilveszter, MD, PhD; Levente Littvay , PhD; Szilard Voros, MD, PhD; Gyorgy Jermendy, MD, PhD; Bela Merkely , MD, PhD; Pal Maurovich-Horvat , MD, PhD, MPH BACKGROUND: Genetics have a strong influence on calcified atherosclerotic plaques; however, data regarding the heritability of noncalcified plaque volume are scarce. We aimed to evaluate genetic versus environmental influences on calcium (coronary artery calcification) score, noncalcified and calcified plaque volumes by coronary computed tomography angiography in adult twin pairs without known coronary artery disease. METHODS: In the prospective BUDAPEST-GLOBAL (Burden of Atherosclerotic Plaques Study in Twins—Genetic Loci and the Burden of Atherosclerotic Lesions) classical twin study, we analyzed twin pairs without known coronary artery disease. All twins underwent coronary computed tomography angiography to assess coronary atherosclerotic plaque volumes. Structural equation models were used to quantify the contribution of additive genetic, common environmental, and unique environmental components to plaque volumes adjusted for age, gender, or atherosclerotic cardiovascular disease risk estimate and statin use. RESULTS: We included 196 twins (mean age±SD, 56±9 years, 63.3% females), 120 monozygotic and 76 same-gender dizygotic pairs. Using structural equation models, noncalcified plaque volume was predominantly determined by environmental factors (common environment, 63% [95% CI, 56%–67%], unique environment, 37% [95% CI, 33%–44%]), while coronary artery calcification score and calcified plaque volumes had a relatively strong genetic heritability (additive genetic, 58% [95% CI, 50%–66%]; unique environmental, 42% [95% CI, 34%–50%] and additive genetic, 78% [95% CI, 73%–80%]; unique environmental, 22% [95% CI, 20%–27%]), respectively. CONCLUSIONS: Noncalcified plaque volume is mainly influenced by shared environmental factors, whereas coronary artery calcification score and calcified plaque volume are more determined by genetics. These findings emphasize the importance of early lifestyle interventions in preventing coronary plaque formation. REGISTRATION: URL: https://www.clinicaltrials.gov; Unique identifier: NCT01738828. Key Words: atherosclerosis ◼ computed tomography angiography ◼ coronary angiography ◼ coronary artery disease ◼ prevalence oronary artery disease (CAD) is a multifactorial dis- computed tomography (CT) has a substantial genetic ease, influenced by the interplay of environmental component, ranging from 30% to 45% in the litera- 1 3–7 Cand genetic factors. Heritability of CAD has been ture. However, the heritability of noncalcified plaque estimated to be 40% to 60%, suggesting that genet- (NCP) volume has only been addressed in a small num- 2 8–10 ics play an important role in its development. Coronary ber of family studies. Healthy individuals with a fam- artery calcification (CAC) as assessed by noncontrast ily history of early onset CAD have a higher prevalence Correspondence to: Pal Maurovich-Horvat, MD, PhD, MPH, Department of Radiology, Medical Imaging Centre, Semmelweis University, Budapest, Hungary. Email maurovich-horvat.pal@med.semmelweis-univ.hu Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/CIRCIMAGING.121.013348. For Sources of Funding and Disclosures, see page 140. © 2022 The Authors. Circulation: Cardiovascular Imaging is published on behalf of the American Heart Association, Inc., by Wolters Kluwer Health, Inc. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution, and reproduction in any medium, provided that the original work is properly cited. Circulation: Cardiovascular Imaging is available at www.ahajournals.org/journal/circimaging Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022 133 Drobni et al Heritability of Coronary Artery Disease (CPs) are less prone to cause events as compared with 11,12 CLINICAL PERSPECTIVE NCPs. In a prospective twin study of 196 subjects, coronary Investigating familial aggregation of complex traits is computed tomography angiography was performed. limited. Shared environment confounds efforts to char- Using structural equation models, the heritability of acterize genetic heritability of complex traits. How- coronary artery calcification score, calcified and non- ever, twins with their precisely matching age represent calcified plaque volumes were estimated. Noncalci- a unique cohort among family studies, and they also fied plaque volume was predominantly determined share a wide range of environmental and socioeconomic by environmental factors (common environment: variables that may influence the expression of com- 63%, unique environment: 37%), while coronary plex traits, such as CAD. Twin studies by modeling the artery calcification score and calcified plaque vol- shared environment, provide a powerful tool to investi- umes had a relatively strong genetic heritability gate the genetic and environmental factors contributing (additive genetic: 58%, unique environmental: 42% 14,15 to a multifactorial disease. and additive genetic: 78%, unique environmen- tal: 22%, respectively). Based on our findings, the Therefore, our aim was to assess the relative contribu- genetic background of noncalcified plaque is smaller tion of genetic and environmental factors on NCP, CAC than that of coronary artery calcification score and score, and CP volumes using coronary CT angiography calcified plaque. Our results are hypothesis generat- (CTA) in adult twin pairs without known CAD. ing but could also imply that socioeconomic status and lifestyle parameters have bigger impact on non- calcified plaque volumes as compared with coronary METHODS artery calcification score and calcified plaque vol- The data will be made available from the corresponding author umes. Based on these, there is an urgent need to on reasonable request after institutional approval. start primary prevention of coronary artery disease as early in life as possible. Study Population This prospective, single-center, classical twin study was con- ducted under the name of BUDAPEST-GLOBAL (Burden of Atherosclerotic Plaques Study in Twins—Genetic Loci and the Nonstandard Abbreviations and Acronyms Burden of Atherosclerotic Lesions) study. Participants had been co-enrolled within an international, multicenter clinical A additive genetic factors study; Genetic Loci and the Burden of Atherosclerotic Lesions ASCVD atherosclerotic cardiovascu- 16,17 (http//:www.ClinicalTrials.gov: NCT01738828). lar disease Detailed study description and enrollment criteria were BUDAPEST-GLOBAL Burden of Atherosclerotic published previously. Monozygotic and same-gender dizygotic Plaques Study in Twins— twins were recruited from the Hungarian Twin Registry, all Genetic Loci and the participants provided informed consent. We assessed zygos- Burden of Atherosclerotic 19,20 ity using a multiple self-reported questionnaire. The study Lesions was approved by the National Scientific and Ethics Committee C common environmental (institutional review board number: ETT TUKEB 58401/2012/ factors EKU [828/PI/12]) and was performed according to the prin- CAC coronary artery calcification ciples stated in the Declaration of Helsinki. We excluded 3 twin pairs due to insufficient image quality, resulting in 196 twin CAD coronary artery disease subjects as our final cohort. CP calcified plaque CT computed tomography Anthropometric Data, Medical History, and CTA computed tomography angiography Laboratory Analysis E Unique environmental Information regarding past medical history and current medical factors therapy were recorded by self-reported questionnaires. Fasting peripheral blood draw and brachial blood pressure measure- NCP noncalcified plaque ments as well as anthropometric data were recorded before the coronary CTA examination. Fasting blood samples were ana- 8 lyzed by standard methods in a certified laboratory. of NCP. NCPs are also more prevalent in younger patients with a family history of CAD, compared with patients with no family history or even to symptomatic Coronary CTA Image Acquisition and patients. It has been suggested that NCPs represent Assessment an earlier stage in atherosclerotic plaque develop- Coronary CTA was performed with a 256-slice multidetector ment. On the contrary, calcification of atherosclerotic CT (Brilliance iCT; Philips HealthTech, Best, the Netherlands). plaque may occur in later stages, and calcified plaques Prospective ECG triggered noncontrast and contrast-enhanced Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022 134 Drobni et al Heritability of Coronary Artery Disease scans were acquired. Beta-blockers were administered as follows: any visible structure that could be assigned to the coro- needed to reach a target heart rate below 60/min. Sublingual nary artery wall in at least 2 independent planes and had a den- nitroglycerin (0.8 mg) was administered on the table. Four- sity below the contrast-enhanced coronary lumen but above phasic contrast injection protocol was used (Iomeprol 400 g/ the surrounding connective tissue. Furthermore, NCP compo- 3 22 cm ; Iomeron, Bracco Imaging S.p.A., Milan, Italy). All image nents were defined as a CT density between (−100) HU and analyses were performed offline on a dedicated workstation 350 HU and CP components as a density between 351 and (Intellispace portal, Philips Healthcare). The coronary CTA 2000 HU. Detailed plaque analysis method is reported in the data sets were analyzed quantitatively to determine coronary Supplemental Material. plaque volumes. CAC score was quantified on the noncontrast- enhanced images using the Agatston method. Statistical Analysis Continuous variables are presented as means and SD, while Semiautomated Plaque Quantification categorical parameters are shown as numbers and percent- Quantitative CT analysis was performed by an experienced ages. Continuous variables were compared using Student reader (Dr Drobni) using a dedicated software tool for plaque t test, whereas binary variables were compared using Fisher assessment (QAngio CT; Medis BV, Leiden, the Netherlands; exact-test and categorical parameters using the χ test Figure 1). Coronary atherosclerotic plaques were defined as between monozygotic and dizygotic individuals. Reproducibility Figure 1. Demonstration of the plaque analysis method. Stretched multiplanar reconstruction of the left anterior descending (LAD) artery and quantification graph of the LAD. Yellow shaded regions correspond to analyzed segments and blue shaded regions correspond to plaques. Components of coronary plaque are shown as follows: lipid-rich (low-computed tomographic attenuation) plaque components are shown in red; fibro-fatty tissue is shown in light green; fibrous tissue is shown in dark green; calcium is shown in white. Example 1: One of the siblings had plaques; plaques were analyzed, and for the other sibling, segments in the same location and same length were analyzed. Example 2: Both siblings had plaques; all the segments were analyzed which included any plaque in either of the siblings. In case neither of the siblings had coronary artery disease (CAD), a plaque volume of 0-0 was used. Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022 135 Drobni et al Heritability of Coronary Artery Disease of plaque quantification was tested by 2 independent observ- CAC and plaque volume (NCP and CP) analysis and on sub- ers (Dr Drobni: 5 years of experience, Dr Simon: 2 years of sets, where siblings with no disease were excluded. Siblings experience) based on 10 randomly selected monozygotic twin with no disease are referenced as 0-0 pairs. For the 2 differ- pair and 10 randomly selected dizygotic twin pair images using ent analysis, 2 0-0 subsets were created. For the CAC analysis the intraclass correlation (ICC) coefficient. Coefficient values one subset with those who had a CAC >0 was created, and for are interpreted as 1.00 to 0.81: excellent; 0.80 to 0.61: good; the plaque volume analysis another subset, with those who had 0.60 to 0.41: moderate; 0.40 to 0.21: fair; 0.20 to 0.00: poor. coronary plaque (either NCP or CP) on the CTA images. Descriptive statistics, correlations and reproducibility measure- Our raw data on CAC score, NCP and CP were trans- ments were calculated using IBM SPSS Statistics version 25 formed to a log scale, and an inverse-normal transformation (IBM, Armonk, NY). on age-sex categories was also performed for the full cohort Genetic structural equation models were built to quantify and for the subsets. Log likelihood-based 95% CI were cal- the proportion of genetic and environmental factors contribut- culated for all estimated parameters. All analyses were per- ing to CAC score, CP, and NCP. These models are based on formed using R version 3.5.2. Twin modeling was performed variation between the twins, which can be broken down to using OpenMx version 2.12.2. A P value of <0.05 was con- additive genetic factors (A), common environmental factors sidered statistically significant. (C) and unique environmental factors (E), therefore the acro- nym ACE model. Additive genetic factors describe the effect of multiple RESULTS genetic alleles that exert influence additively. Common environ- Study Population mental factors are shared by the twin pairs during their life- time, such as the same early childhood, education in the same Baseline demographics and clinical characteristics of the school, living in the same town, sharing similar socioeconomic overall population (n=196) are summarized in Table 1. status even in adulthood, etc. Unique environmental factors The mean±SD age of the cohort was 56±9 years (63.3% are conditions to which only one of the siblings was exposed. female), and dizygotic subjects were older than the mono- Sub models of the full models (full ACE model) were cal- zygotic subjects (58±8 versus 55±10 years, P=0.005). culated, and the more parsimonious model with the best fit The most prevalent cardiovascular risk factor in the total was selected. All calculations were adjusted for age and sex or cohort was hypertension (83/196, 42.3%), and 16.8% atherosclerotic cardiovascular disease (ASCVD) risk estimate and statin use. Analysis was performed on the full cohort for (33/196) of the participants were on primary preventive Table 1. Basic Demographics and Clinical-Laboratory Characteristics of Twin Subjects Monozygotic Dizygotic twins, Parameters Total, n=196 twins, n=120 n=76 P value Demographics and clinical data Female, n (%) 124 (63.3) 72 (60.0) 52 (68.4) 0.29 Age, mean±SD, y 56±9 55±10 58±8 0.005 Height, mean±SD, cm 166.6±9.7 166.3±10.1 166.9±9.0 0.68 Weight, mean±SD, kg 77.1±16.9 77.3±17.9 76.8±15.4 0.82 Body mass index, mean±SD, kg/m 27.7±5.1 27.7±5.0 27.6±5.3 0.82 Systolic blood pressure, mean±SD, mm Hg 139.7±20.2 138.6±19.4 141.5±21.5 0.34 Hypertension, n (%) 83 (42.3) 48 (40.0) 35 (46.1) 0.46 Diabetes, n (%) 17 (8.7) 12 (10.0) 5 (6.6) 0.45 Current smoker, n (%) 31 (15.8) 19 (15.8) 12 (15.8) 1.00 Statin use, n (%) 33 (16.8) 17 (14.2) 16 (21.1) 0.24 Stroke, n (%) 2 (1.0) 2 (1.7) 0 (0) 0.52 10-year ASCVD risk, mean±SD, % 7.9±7.7 7.7±7.6 8.3±7.8 0.62 Laboratory parameters Fasting blood glucose, mean±SD, mg/dL 96.3±23.3 96.9±27.4 95.4±14.9 0.63 Hemoglobin A1c, mean±SD, % 5.5±0.9 5.6±1.0 5.3±0.8 0.01 Serum total cholesterol, mean±SD, mg/dL 214.8±42.2 217.6±43.2 210.5±40.5 0.25 Serum LDL-cholesterol, mean±SD mg/dL 134.7±38.4 136.7±40.8 131.6±34.1 0.35 Serum HDL-cholesterol, mean±SD, mg/dL 62.1±14.9 61.7±15.9 62.9±13.0 0.61 Triglycerides, mean±SD, mg/dL 137.7±91.2 142.6±104.1 129.9±66.0 0.30 Serum C-reactive protein, mean±SD, mg/L 2.9±4.4 2.7±2.9 3.2±6.2 0.51 <0.05 was considered significant. ASCVD indicates atherosclerotic cardiovascular disease; LDL, low-density lipoprotein; and HDL, high-density lipoprotein. Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022 136 Drobni et al Heritability of Coronary Artery Disease statin therapy (Table 1). Both total cholesterol (214.8±42.2 (ICC for NCP and CP was 0.99 and 0.98, intra-reader mg/dL) and LDL (low-density lipoprotein)-cholesterol lev- and ICC for NCP and CP was 0.97 and 0.99, inter-reader els (134.7±38.4 mg/dL) were slightly elevated with no respectively). difference between monozygotic and dizygotic groups Out of the 196 twins, 74 (37.8 %) had coronary cal- (P=0.25 and P=0.35, respectively). The 10-year ASCVD cification on noncontrast enhanced images, 34 were risk estimate was 7.9±7.7% for the total cohort, with 83 dizygotic and 40 were monozygotic. The prevalence of a subjects as low-risk ( 5.0%), 34 subjects as borderline CAC score higher than zero was not different among the risk (5.0%–7.4%), 63 subjects as intermediate risk (7.5%– monozygotic and dizygotic twins (44.7% [34/76] versus 19.9%), and 16 high risk ( 20.0%) subjects. Significant 33.3% [40/120] P=0.13). The prevalence of discordant difference was observed in the HbA1c levels between twin pairs, meaning one sibling had a CAC score higher the monozygotic and dizygotic group (5.6±1.0% versus than zero, whereas the other did not, was significantly 5.3±0.8%, P=0.01). Otherwise, there were no significant higher in the dizygotic group (Table 3). The prevalence of differences between the groups (Table 1). concordant twin pairs with a CAC score of 0 was higher Data regarding family history, socioeconomic sta- in the monozygotic group (Table 3). Comparing mono- tus, and lifestyle parameters were available in all sub- zygotic and dizygotic twins, we found no differences jects, and these results are summarized in Table 2. regarding the median CAC score values in those who Dizygotic twins had a stronger family history for dys- had a CAC score higher than 0 (115 [8–336] for the lipidemia (65.8% [50/76] versus 36.7% [44/120], monozygotic group versus 97 [47–208] for the dizygotic < > P 0.001) and for coronary revascularization (15.8% group; P 0.9). [12/76] versus 6.7% [8/120], P=0.04) as compared Out of the 196 twins, 102 (52.0%) had coronary with monozygotic twins. plaques, 42 were dizygotic and 60 were monozygotic. The prevalence of any CAD was not different among the groups. (dizygotic, 55.3% [42/76] versus monozygotic, CAD Characteristics 50.0% [60/120] group P=0.56). In those, who had CAD, Intrareader and interreader agreement showed excel- the median segment involvement score was 3 (2–5), and lent reproducibility for all plaque volume measurements the median segment stenosis score was 4 (2–8), with no Table 2. Data on Family History and Lifestyle Parameters Monozygotic Dizygotic twins, Parameters Total, n=196 twins, n=120 n=76 P value Positive family history of Smoking, n (%) 160 (81.6) 98 (81.7) 62 (81.6) 0.99 Diabetes, n (%) 92 (46.9) 52 (43.3) 40 (52.6) 0.20 Hypertension, n (%) 150 (76.6) 88 (73.3) 62 (81.6) 0.18 Dyslipidemia, n (%) 94 (48.0) 44 (36.7) 50 (65.8) <0.001 Cerebrovascular event, n (%) 88 (44.9) 48 (40.0) 40 (52.6) 0.083 Myocardial infarction, n (%) 100 (51.0) 60 (50.0) 40 (52.6) 0.72 Coronary revascularization, n (%) 20 (10.2) 8 (6.7) 12 (15.8) 0.040 Socioeconomic and lifestyle parameters Place of living City, n (%) 113 (57.7) 65 (54.2) 48 (63.2) Town, n (%) 64 (32.7) 45 (37.5) 19 (25.0) 0.031 Village, n (%) 19 (9.7) 10 (8.3) 9 (11.8) Daily meat consumption, n (%) 78 (39.8) 49 (40.8) 29 (38.2) 0.71 Daily fruit and vegetables consumption, n (%) 183 (93.4) 112 (93.3) 71 (93.4) 0.98 Daily alcohol intake, median (IQR), unit 0.0 (0.0–6.0) 0.0 (0.0–5.1) 1.5 (0.0–6.0) 0.18 Daily coffee intake, median (IQR), unit 2.0 (1.0–2.0) 1.75 (1.0–2.0) 2.0 (1.0–2.0) 0.64 Daily physical activity Sedentary, median (IQR), h/d 8.0 (5.0–11.0) 8.0 (5.0–12.0) 8.0 (5.0–11.0) 0.84 Light, median (IQR), h/d 4.0 (2.0–8.0) 4.5 (2.9–8.0) 4.0 (2.0–6.0) 0.32 Moderate, median (IQR), h/d 2.0 (1.0–4.0) 1.0 (1.0–3.0) 2.0 (1.0–4.0) 0.093 Hard, median (IQR), h/d 0.0 (0.0–1.0) 0.0 (0.0–1.0) 0.0 (0.0–0.0) 0.20 <0.05 was considered significant. IQR indicates interquartile range. Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022 137 Drobni et al Heritability of Coronary Artery Disease Table 3. Data on Coronary Artery Disease Characteristics Monozygotic twins Dizygotic twins Parameters n=120 n=76 P value Calcium score Discordant pairs, n (%) 20 (16.6) 28 (36.8) Concordant negative, n (%) 70 (58.3) 28 (36.8) 0.002 Concordant positive, n (%) 30 (25.0) 20 (26.3) Coronary atherosclerosis Discordant pairs, n (%) 24 (20.0) 20 (26.3) Concordant negative, n (%) 48 (40.0) 24 (31.6) 0.41 Concordant positive, n (%) 48 (40.0) 32 (42.1) Discordant pairs: only one sibling had disease. Concordant pairs: both siblings or neither of the siblings had the disease. A positive subject for calcium score was defined as having a calcium score >0 and for coronary atherosclerosis as any coronary plaque on contrast enhanced images. P<0.05 was considered significant. significant difference among the dizygotic and monozy- NCP was similar. In both cases the best fitting model was gotic groups (P=0.2 for both). AE with a modest genetic influence (CP; A: 48% [95% Comparing monozygotic and dizygotic twins with CAD, CI, 41%–55%], E: 52% [95% CI, 44%–59%], Table S4 we found no differences regarding NCP volume (dizy- and NCP; 45% [95% CI, 36%–52%], E: 55% [95% CI, gotic, 107 [52–178] mm versus monozygotic, 79 [36– 48%–64%], Table S6). 175] mm ; P=0.5) and CP volume (dizygotic, 43 [7–65] 3 3 mm versus monozygotic, 18 [5–84] mm , P=0.4). DISCUSSION Using coronary CTA in adult twins without known CAD, we Heritability of CAC Score, CP, and NCP found that NCP volume was mostly influenced by envi- Using genetic structural equation models, adjusted for ronmental, rather than genetic factors, while CAC score age and gender, ASCVD risk estimate and statin use, and CP volume had a relatively strong genetic heritability. the best fitting models were selected both for the full Our results remained similar, irrespective of the adjusted cohort and for the 2 subsets (Figure 2). Twin statistics covariates. We repeated our analysis in 2 cohorts where were performed on the raw data, log transformed and on the 0-0 pairs were excluded. The contribution of genetic the inverse normal transformed data. Here, we report the factors to phenotypic variance of CAC score disappeared results from the inverse normal transformed data. All the and diminished for CP from 78% to 48%. The contri- other results can be found in the Supplemental Material. bution of common, shared environmental factors to the In the full cohort, CAC score had a strong genetic her- phenotypic variance of NCP disappeared, and a moder- itability (adjusted for ASCVD risk and statin use; A: 58% ate genetic background of 45% was found when the 0-0 [95% CI, 50%–66%], E: 42% [95% CI, 34%–50%], pairs were excluded. The loss of a genetic contribution in Table S1 and S2) similarly to CP (adjusted for ASCVD the subgroup analysis for CAC score could also suggest risk and statin use; A: 78% [95% CI, 73%–80%], E: that the overwhelming majority of genetic contributions 22% [95% CI, 20%–27%], Tables S3 and S4; Figure 2). are protective against coronary calcification. On the contrary, NCP quantities were influenced pre- Positive family history of CAD is considered to be dominantly by common, shared environmental factors an independent risk factor for future cardiovascular (adjusted for ASCVD risk and statin use; C: 63% [95% events. In the Framingham Heart Study, a positive fam- 29,30 CI, 56%–67%], E: 37% [95% CI, 33%–44%], Table S5 ily history of CAD was a strong predictor of CAD and and S6; Figure 2). Similar results were found when the was the strongest clinical predictor in younger patients of genetic structural equation models were adjusted for age future myocardial infarction. Healthy first-degree rela- and gender (Tables S1 through S6). tives of patients with early onset CAD have an ≈5-fold We repeated the analysis in the subsets of siblings, increase of total coronary plaque volume compared with where both twins without disease were excluded. This symptomatic patients. It has been demonstrated that 3–7 resulted in 98 subjects for CAC score, and in 124 sub- CAC has a substantial genetic component. In line with jects for plaque volume (CP and NCP) analysis. Adjust- these findings, the investigators of a community-based ing for ASCVD risk and statin use, the best fitting model study from Rochester reported that 40% of the interin- was model E, for CAC score showing that 100% of the dividual variation in CAC quantity is attributed to genetic phenotypic variance in CAC score was attributable to factors. These findings suggest a strong hereditary unique environmental factors (E: 100% [95% CI, 100%– component of coronary atherosclerosis. In our study, we 100%]). In the subset analysis, the heritability of CP and found a slightly stronger genetic background both for Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022 138 Drobni et al Heritability of Coronary Artery Disease Figure 2. Twin statistics results. Bar graphs demonstrate the results of the genetic structural equation model of calcium score, calcified, and noncalcified plaque volumes in both the total cohort and in the 2 subsets of siblings with 0-0 pairs excluded. ASCVD indicates atherosclerotic cardiovascular disease. 8,10 CAC score (58%) and CP (78%) as previously reported importance of genetics in CAD development. How- in the above-mentioned studies. ever, the fact that a trait runs in family does not mean that The Swedish twin trial enrolled over 20 000 twins, and it has a strong genetic background, since families may the genetic basis of CAD-related mortality was evaluated. share not only genes but common environmental factors, An increased risk for CAD was found among close rela- for example socioeconomic status, dietary, and exercise 1,2 13,24 tives. If a twin sibling had died from early onset CAD, the habits. In a family study, distinction between genetic 13,31,32 sibling’s relative hazard of death due to CAD was roughly and common environmental effects may be difficult, double in male monozygotic twins as compared with male but twin studies can overcome this limitation, since twin dizygotic twins, and nearly 6× higher in females. Among studies can estimate the magnitude of genetic and com- our twin subjects, there were more discordant pairs for mon environmental components to a specific trait, by 13,15,33 CAC score among the dizygotic group and more concor- modeling the shared environment. This may be a dant pairs among the monozygotic groups. These results reason why we only found a strong genetic component also imply a strong genetic background of coronary cal- behind NCP volumes after adjusting for age and gen- cification as also shown in more detail with the genetic der or ASCVD risk estimate and statin use and only in a structural models in our main analysis. subset of twins when we excluded the 0-0 pairs. Impor- In our classical twin study, we found a relatively strong tantly, twin studies, compared with family studies, can be genetic dependence of CAC score and CP volumes, cor- more powerful to examine heritability of a complex trait 3,7 roborating former observations. On the contrary, we because they can take advantage of a unique character- observed that environment, especially common environ- istics of twins: their shared genetics and matching age. mental factors play a more important role than genetic Furthermore, a twin design naturally accounts for mater- effects in determining NCP volumes in individuals with- nal factors and a range of early environmental factors, out known CAD. These findings are seemingly in con- which might potentially bias the associations. Twins are tradiction to former suggestions, which describe the exposed to higher degree of shared family environment Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022 139 Drobni et al Heritability of Coronary Artery Disease compared with nontwin siblings, and they also share a of CAD. Measurement error may appear as part of the range of environmental variables and socioeconomic unique environmental component as it is uncorrelated status even in adulthood, which do contribute to the across measurements. Due to the cross-sectional nature expression of complex traits. Therefore, we believe that of our study, we had no data about the plaque devel- our findings are complimentary and additive to previous opment over time. The study was a single-center inves- observations regarding the importance of genetic and tigation with twins from a White population, which may environmental factors influencing plaque volumes. limit the generalizability of our findings. A potential bias Our results did change when we analyzed CAD heritabil- could be the age difference between the monozygotic ity in a subgroup, excluding the 0-0 plaque pairs. Since this and dizygotic groups, which could inflate the correlation study enrolled asymptomatic subjects with no history of CAD, among the dizygotic group, especially for CP volumes. excluding the 0-0 pairs did significantly reduce our sample However, calcified and NCP volumes were not differ- size. Performing twin statistics on a phenotype which can ent among the monozygotic and dizygotic groups. We be 0 is extremely challenging, and we approached it from believe that by performing the heritability analysis on the several angles. In cases where one sibling had CAD and the inverse-normal transformed data on age-sex categories other did not, we co-registered the plaque localization and and adjusting for age and gender and for ASCVD risk used that instead of 0 for the nondiseased sibling. Moreover, estimate, we were able to address this issue. we performed a rank based inverse normal transformation, to balance out the 0 tail in the data distribution caused by Conclusions the nondiseased subjects. The inverse normal transforma- We have observed that NCP volume is predominantly tion removed the heavy 0 tail, and we further adjusted it for determined by common environmental factors while CAC age, sex, and ASCVD risk estimate and statin use. score and CP volume are influenced mainly by genetics. It is important to note that our findings do not explain These findings suggest that lifestyle may have an important the underlying pathophysiological background of CAD role in the initiation of CAD (NCP), and genetics may have a and plaque formation. Nevertheless, it is generally strong effect on CP formation. These results underline the accepted that during coronary plaque formation, NCPs importance of optimal risk factor management early in life. are present at early stage while CPs represent a later stage of plaque development. Accordingly, our results indicate that early development of coronary plaques is ARTICLE INFORMATION mainly influenced by common environmental factors, Received July 22, 2021; accepted January 24, 2022. such as dietary and exercise habits and socioeconomic Affiliations status, in contrast to plaque calcification which is more MTA-SE Cardiovascular Imaging Research Group, (Z.D.D., M.K., J.K., A.L.J., J.S., dependent on genetics. B.S., P.M.-H.), and Department of Radiology, Medical Imaging Centre (A.D.T., D.L.T., An important clinical implication of our findings is P.M.-H.), Semmelweis University, Budapest, Hungary. Department of Political Sci- within primary prevention of CAD, especially of NCP. ence, Central European University, Budapest, Hungary (L.L.). Global Genomics Group, Richmond, VA (S.V.). Bajcsy-Zsilinszky Hospital, Budapest, Hungary (G.J.). Based on our findings, the genetic background of NCP Heart and Vascular Center (B.M.), Cardiovascular Imaging Research Center, De- is smaller than the CAC score and CP, and our data also partment of Radiology, Massachusetts General Hospital, Harvard Medical School, imply that socioeconomic status and lifestyle parameters Boston, MA, USA (M.K., J.K.). have bigger impact on NCP as compared with CAC score Acknowledgments and CP. Based on these, there is an urgent need to start We thank the team at the Heart and Vascular Center for their help in performing primary prevention of CAD as early in life as possible. the study. Starting with lifestyle changes could be the foundation Sources of Funding of preventative measures for NCP. The data and analysis This study was supported by a New Horizons Grant from the EASD to Dr Jer- presented in this paper should be considered hypothe- mendy. Dr Drobni was supported by the ÚNKP-21-4-I-SE new national excel- lence program of the ministry for innovation and technology from the source of sis-generating and may need further confirmation. the national research, development, and innovation fund. Dr Merkely was funded by Project no. NVKP_16-1–2016-0017 (National Heart Program) with the sup- port provided from the National Research, Development and Innovation Fund of Limitations Hungary, financed under the NVKP_16 funding scheme. The research was fi- nanced by the Thematic Excellence Programme (2020-4.1.1.-TKP2020) of the Our sample size is relatively modest and comparable with Ministry for Innovation and Technology in Hungary, within the framework of the other twin studies. The aim was to balance the overall Therapeutic Development and Bioimaging thematic programmes of the Semmel- weis University. population for 50% females and ≥50% dizygotic twins; however, 63% of the twins are female and 39% are dizy- Disclosures gotic twins. This might be due to the fact that females Dr Voros is a shareholder in Global Genomics Group, LLC, and receives salary and monozygotic twins are more willing to participate in from Global Genomics Group, LLC. The other authors report no conflicts. research than males. In the present study, statins were Supplemental Materials used in 16.8% of the patients, which is relatively low; Supplemental Document however, it can still influence the presence and phenotype Tables S1–S6 Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022 140 Drobni et al Heritability of Coronary Artery Disease 19. Heath AC, Nyholt DR, Neuman R, Madden PA, Bucholz KK, Todd RD, REFERENCES Nelson EC, Montgomery GW, Martin NG. Zygosity diagnosis in the absence 1. Marenberg ME, Risch N, Berkman LF, Floderus B, de Faire U. Genetic sus- of genotypic data: an approach using latent class analysis. Twin Res. ceptibility to death from coronary heart disease in a study of twins. N Engl J 2003;6:22–26. doi: 10.1375/136905203762687861 Med. 1994;330:1041–1046. doi: 10.1056/NEJM199404143301503 20. Christiansen L, Frederiksen H, Schousboe K, Skytthe A, von Wurmb- 2. Zdravkovic S, Wienke A, Pedersen NL, Marenberg ME, Yashin AI, Schwark N, Christensen K, Kyvik K. Age- and sex-differences in the validity De Faire U. Heritability of death from coronary heart disease: a 36-year of questionnaire-based zygosity in twins. Twin Res. 2003;6:275–278. doi: follow-up of 20 966 Swedish twins. J Intern Med. 2002;252:247–254. doi: 10.1375/136905203322296610 10.1046/j.1365-2796.2002.01029.x 21. Rickham PP. Human experimentation. Code of ethics of the world medi- 3. Peyser PA, Bielak LF, Chu JS, Turner ST, Ellsworth DL, Boerwinkle E, cal association. Declaration of helsinki. Br Med J. 1964;2:177. doi: nd Sheedy PF 2 . Heritability of coronary artery calcium quantity measured by 10.1136/bmj.2.5402.177 electron beam computed tomography in asymptomatic adults. Circulation. 22. Karády J, Panajotu A, Kolossváry M, Szilveszter B, Jermendy ÁL, 2002;106:304–308. doi: 10.1161/01.cir.0000022664.21832.5d Bartykowszki A, Károlyi M, Celeng C, Merkely B, Maurovich-Horvat P. The 4. Paixao AR, Berry JD, Neeland IJ, Ayers CR, Rohatgi A, de Lemos JA, effect of four-phasic versus three-phasic contrast media injection protocols Khera A. Coronary artery calcification and family history of myocardial on extravasation rate in coronary CT angiography: a randomized controlled infarction in the Dallas heart study. JACC Cardiovasc Imaging. 2014;7:679– trial. Eur Radiol. 2017;27:4538–4543. doi: 10.1007/s00330-017-4866-0 686. doi: 10.1016/j.jcmg.2014.04.004 23. Achenbach S, Moselewski F, Ropers D, Ferencik M, Hoffmann U, 5. Khera A, Joshi P. What’s a Malignant Family History?: you’ll know it when MacNeill B, Pohle K, Baum U, Anders K, Jang IK, et al. Detection of calci- you see it. JACC Cardiovasc Imaging. 2017;10(10 Pt A):1136–1138. doi: fied and noncalcified coronary atherosclerotic plaque by contrast-enhanced, 10.1016/j.jcmg.2016.11.014 submillimeter multidetector spiral computed tomography: a segment-based 6. Rampersaud E, Bielak LF, Parsa A, Shen H, Post W, Ryan KA, Donnelly P, comparison with intravascular ultrasound. Circulation. 2004;109:14–17. doi: nd Rumberger JA, Sheedy PF 2 , Peyser PA, et al. The association of coronary 10.1161/01.CIR.0000111517.69230.0F artery calcification and carotid artery intima-media thickness with distinct, 24. Zyphur MJ, Zhang Z, Barsky AP, Li W-D. An ACE in the hole: twin fam- traditional coronary artery disease risk factors in asymptomatic adults. Am J ily models for applied behavioral genetics research. Leadersh Q. Epidemiol. 2008;168:1016–1023. doi: 10.1093/aje/kwn211 2013;24:572–594. 7. Wagenknecht LE, Bowden DW, Carr JJ, Langefeld CD, Freedman BI, 25. Franić S, Dolan CV, Borsboom D, Boomsma DI. Structural Equation Modeling Rich SS. Familial aggregation of coronary artery calcium in families with type in Genetics. Guilford Press; 2012:617–635. 2 diabetes. Diabetes. 2001;50:861–866. doi: 10.2337/diabetes.50.4.861 26. Goff DC Jr, Lloyd-Jones DM, Bennett G, Coady S, D’Agostino RB, 8. Kral BG, Becker LC, Vaidya D, Yanek LR, Qayyum R, Zimmerman SL, Gibbons R, Greenland P, Lackland DT, Levy D, O’Donnell CJ, et al; American Dey D, Berman DS, Moy TF, Fishman EK, et al. Noncalcified coronary College of Cardiology/American Heart Association Task Force on Practice plaque volumes in healthy people with a family history of early onset Guidelines. 2013 ACC/AHA guideline on the assessment of cardiovascular coronary artery disease. Circ Cardiovasc Imaging. 2014;7:446–453. doi: risk: a report of the American College of Cardiology/American Heart Asso- 10.1161/CIRCIMAGING.113.000980 ciation Task Force on Practice Guidelines. Circulation. 2014;129(25 Suppl 9. Otaki Y, Gransar H, Berman DS, Cheng VY, Dey D, Lin FY, Achenbach S, 2):S49–S73. doi: 10.1161/01.cir.0000437741.48606.98 Al-Mallah M, Budoff MJ, Cademartiri F, et al. Impact of family history of coro- 27. Team RC. R: A language and environment for statistical computing. Founda- nary artery disease in young individuals (from the CONFIRM registry). Am J tion for Statistical Computing, Vienna, Austria. 2017. Cardiol. 2013;111:1081–1086. doi: 10.1016/j.amjcard.2012.12.042 28. Colditz GA, Stampfer MJ, Willett WC, Rosner B, Speizer FE, Hennekens CH. 10. Christiansen MK, Jensen JM, Nørgaard BL, Dey D, Bøtker HE, Jensen HK. A prospective study of parental history of myocardial infarction and coro- Coronary plaque burden and adverse plaque characteristics are increased nary heart disease in women. Am J Epidemiol. 1986;123:48–58. doi: in healthy relatives of patients with early onset coronary artery disease. 10.1093/oxfordjournals.aje.a114223 JACC Cardiovasc Imaging. 2017;10(10 Pt A):1128–1135. doi: 10.1016/j. 29. Lloyd-Jones DM, Nam BH, D’Agostino RB Sr, Levy D, Murabito JM, jcmg.2016.10.014 Wang TJ, Wilson PW, O’Donnell CJ. Parental cardiovascular disease as a 11. Maurovich-Horvat P, Ferencik M, Voros S, Merkely B, Hoffmann U. Compre- risk factor for cardiovascular disease in middle-aged adults: a prospec- hensive plaque assessment by coronary CT angiography. Nat Rev Cardiol. tive study of parents and offspring. JAMA. 2004;291:2204–2211. doi: 2014;11:390–402. doi: 10.1038/nrcardio.2014.60 10.1001/jama.291.18.2204 12. Hoffmann U, Moselewski F, Nieman K, Jang IK, Ferencik M, Rahman AM, 30. Murabito JM, Pencina MJ, Nam BH, D’Agostino RB Sr, Wang TJ, Cury RC, Abbara S, Joneidi-Jafari H, Achenbach S, et al. Noninvasive Lloyd-Jones D, Wilson PW, O’Donnell CJ. Sibling cardiovascular disease assessment of plaque morphology and composition in culprit and stable as a risk factor for cardiovascular disease in middle-aged adults. JAMA. lesions in acute coronary syndrome and stable lesions in stable angina by 2005;294:3117–3123. doi: 10.1001/jama.294.24.3117 multidetector computed tomography. J Am Coll Cardiol. 2006;47:1655– 31. MacGregor AJ, Snieder H, Schork NJ, Spector TD. Twins. Novel uses to 1662. doi: 10.1016/j.jacc.2006.01.041 study complex traits and genetic diseases. Trends Genet. 2000;16:131– 13. Martin N, Boomsma D, Machin G. A twin-pronged attack on complex traits. 134. doi: 10.1016/s0168-9525(99)01946-0 Nat Genet. 1997;17:387–392. doi: 10.1038/ng1297-387 32. Krieger N, Chen JT, Coull BA, Selby JV. Lifetime socioeconomic position and 14. Mangino M, Spector T. Understanding coronary artery disease using twin twins’ health: an analysis of 308 pairs of United States women twins. PLoS studies. Heart. 2013;99:373–375. doi: 10.1136/heartjnl-2012-303001 Med. 2005;2:e162. doi: 10.1371/journal.pmed.0020162 15. Craig JM, Calais-Ferreira L, Umstad MP, Buchwald D. The value of twins for 33. Schultz WM, Kelli HM, Lisko JC, Varghese T, Shen J, Sandesara P, health and medical research: a third of a century of progress. Twin Res Hum Quyyumi AA, Taylor HA, Gulati M, Harold JG, et al. Socioeconomic status Genet. 2020;23:8–15. doi: 10.1017/thg.2020.4 and cardiovascular outcomes: challenges and interventions. Circulation. 16. Voros S, Maurovich-Horvat P, Marvasty IB, Bansal AT, Barnes MR, Vazquez G, 2018;137:2166–2178. doi: 10.1161/CIRCULATIONAHA.117.029652 Murray SS, Voros V, Merkely B, Brown BO, et al. Precision phenotyping, 34. Elder SJ, Lichtenstein AH, Pittas AG, Roberts SB, Fuss PJ, Greenberg AS, panomics, and system-level bioinformatics to delineate complex biologies of McCrory MA, Bouchard TJ Jr, Saltzman E, Neale MC. Genetic and envi- atherosclerosis: rationale and design of the “Genetic Loci and the Burden of ronmental influences on factors associated with cardiovascular disease Atherosclerotic Lesions” study. J Cardiovasc Comput Tomogr. 2014;8:442– and the metabolic syndrome. J Lipid Res. 2009;50:1917–1926. doi: 10.1194/jlr.P900033-JLR200 451. doi: 10.1016/j.jcct.2014.08.006 35. Coccaro EF, Jacobson KC. PennTwins: a population-based cohort for 17. Maurovich-Horvat P, Tárnoki DL, Tárnoki ÁD, Horváth T, Jermendy ÁL, twin studies. Twin Res Hum Genet. 2006;9:998–1005. doi: 10.1375/ Kolossváry M, Szilveszter B, Voros V, Kovács A, Molnár AÁ, et al. Rationale, design, and methodological aspects of the BUDAPEST-GLOBAL Study 36. Achenbach S, Ropers D, Pohle K, Leber A, Thilo C, Knez A, Menendez (Burden of Atherosclerotic Plaques Study in Twins-Genetic Loci and the T, Maeffert R, Kusus M, Regenfus M, et al. Influence of lipid-lowering Burden of Atherosclerotic Lesions). Clin Cardiol. 2015;38:699–707. doi: therapy on the progression of coronary artery calcification: a prospec- 10.1002/clc.22482 tive evaluation. Circulation. 2002;106:1077–1082. doi: 10.1161/01.cir. 18. Littvay L, Métneki J, Tárnoki AD, Tárnoki DL. The hungarian twin registry. 0000027567.49283.ff Twin Res Hum Genet. 2013;16:185–189. doi: 10.1017/thg.2012.76 Circ Cardiovasc Imaging. 2022;15:e013348. DOI: 10.1161/CIRCIMAGING.121.013348 March 2022
Journal
Circulation Cardiovascular Imaging – Wolters Kluwer Health
Published: Mar 24, 2022