Listen to this manuscript’s

audio summary by

JACC Editor-in-Chief

Dr. Valentin Fuster.

J O U R N A L O F T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y V O L . 6 9 , N O . 7 , 2 0 1 7

ª 2 0 1 7 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N .

P U B L I S H E D B Y E L S E V I E R . A L L R I G H T S R E S E R V E D .

I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 6 . 1 1 . 0 6 0

REVIEW TOPIC OF THE WEEK

Advances in the Genetics ofCongenital Heart Disease

A Clinician’s Guide

Gillian M. Blue, PHD,a,b,c Edwin P. Kirk, MBBS, PHD,d,e Eleni Giannoulatou, DPHIL,f,g,h Gary F. Sholler, MBBS,a,b,c

Sally L. Dunwoodie, PHD,f,g,h Richard P. Harvey, PHD,f,g,h David S. Winlaw, MBBS, MDa,b,c

JACC JOURNAL CME

This article has been selected as the month’s JACC Journal CME activity,

available online at http://www.acc.org/jacc-journals-cme by selecting the

CME tab on the top navigation bar.

Accreditation and Designation Statement

The American College of Cardiology Foundation (ACCF) is accredited by

the Accreditation Council for Continuing Medical Education (ACCME) to

provide continuing medical education for physicians.

The ACCF designates this Journal-based CME activity for a maximum

of 1 AMA PRA Category 1 Credit(s). Physicians should only claim

credit commensurate with the extent of their participation in the activity.

Method of Participation and Receipt of CME Certificate

To obtain credit for JACC CME, you must:

1. Be an ACC member or JACC subscriber.

2. Carefully read the CME-designated article available online and in this

issue of the journal.

3. Answer the post-test questions. At least 2 out of the 3 questions

provided must be answered correctly to obtain CME credit.

4. Complete a brief evaluation.

From the aKids Heart Research, The Children’s Hospital at Westmead, Sydn

dren’s Hospital at Westmead, Sydney, Australia; cSydney Medical School, Uni

Genetics, Sydney Children’s Hospital, Sydney, Australia; eSchool of Women

versity of New South Wales, Sydney, Australia; fVictor Chang Cardiac Res

Vincent’s Clinical School, Faculty of Medicine, University of New South W

nology and Biomolecular Sciences, University of New South Wales, Sydney

lationships relevant to the contents of this paper to disclose.

Manuscript received September 26, 2016; revised manuscript received Nove

5. Claim your CME credit and receive your certificate electronically by

following the instructions given at the conclusion of the activity.

CME Objective for This Article: After reading this article, the reader

should be able to: 1) identify patients with congenital heart disease who

will benefit from referral to a clinical genetics service; 2) explain the

genetic basis of congenital heart disease to affected patients and their

families; and 3) compare available genetic technologies and their

respective applications.

CME Editor Disclosure: JACC CME Editor Ragavendra R. Baliga, MD,

FACC, has reported that he has no financial relationships or interests

to disclose.

Author Disclosures: The authors have reported that they have no

relationships relevant to the contents of this paper to disclose.

Medium of Participation: Print (article only); online (article and quiz).

CME Term of Approval

Issue Date: February 21, 2017

Expiration Date: February 20, 2018

ey, Australia; bHeart Centre for Children, The Chil-

versity of Sydney, Australia; dDepartment of Medical

’s and Children’s Health, Faculty of Medicine, Uni-

earch Institute, Darlinghurst, Sydney, Australia; gSt

ales, Sydney, Australia; and the hSchool of Biotech-

. The authors have reported that they have no re-

mber 15, 2016, accepted November 17, 2016.

Blue et al. J A C C V O L . 6 9 , N O . 7 , 2 0 1 7

A Clinician’s Guide to CHD Genetics F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0

860

Advances in the Genetics ofCongenital Heart Disease

A Clinician’s Guide

Gillian M. Blue, PHD,a,b,c Edwin P. Kirk, MBBS, PHD,d,e Eleni Giannoulatou, DPHIL,f,g,h Gary F. Sholler, MBBS,a,b,c

Sally L. Dunwoodie, PHD,f,g,h Richard P. Harvey, PHD,f,g,h David S. Winlaw, MBBS, MDa,b,c

ABSTRACT

Our understanding of the genetics of congenital heart disease (CHD) is rapidly expanding; however, many questions,

particularly those relating to sporadic forms of disease, remain unanswered. Massively parallel sequencing technology

has made significant contributions to the field, both from a diagnostic perspective for patients and, importantly,

also from the perspective of disease mechanism. The importance of de novo variation in sporadic disease is a recent

highlight, and the genetic link between heart and brain development has been established. Furthermore, evidence of

an underlying burden of genetic variation contributing to sporadic and familial forms of CHD has been identified.

Although we are still unable to identify the cause of CHD for most patients, recent findings have provided us with a much

clearer understanding of the types of variants and their individual contributions and collectively mark an important

milestone in our understanding of both familial and sporadic forms of disease. (J Am Coll Cardiol 2017;69:859–70)

© 2017 by the American College of Cardiology Foundation. Published by Elsevier. All rights reserved.

CURRENT STATUS OF GENETIC RESEARCH IN

CONGENITAL HEART DISEASE

Since the first report specifically addressing thegenetics of congenital heart disease (CHD) in the1950s (1), much research has been devoted to under-standing the heritable nature of this condition,including the notable work of James Nora on multi-factorial inheritance in the 1960s (2), as well as thelandmark Baltimore-Washington Infant Studyassessing the epidemiology of CHD in the 1980s (3)(Figure 1). Familial forms of CHD have providedmost of the genetic information on structural heartdisease to date because of their suitability to earlyresearch techniques, including linkage analysis andcandidate gene approaches. Indeed, many of thegenes associated with CHD, including NKX2-5,GATA4, TBX5, NOTCH1, and TBX20, were identifiedusing these early genetic techniques (4–8); however,a prerequisite to the use of linkage analysis is theexistence of large families with multiple affected in-dividuals segregating disease according to Mendelianprinciples, which is rare in CHD (9).

The development of chromosomal microarray(CMA) technology, including array comparativegenome hybridization and single-nucleotide poly-morphism (SNP) arrays, in the early 2000s provided anew tool for research in the field (Figure 1). UsingCMA, a novel candidate gene, TAB2, was identified

after the identification of an 850-kb deletion onchromosome 6q that was shared among 12 patientswith CHD (10). Since then, a number of studies haveused CMA to locate novel candidate genes involved inheterotaxy (10), isolated tetralogy of Fallot (TOF) (11),and left-sided CHD (12), as well as novel genomic re-gions of interest (13). Comparative genome hybridi-zation has largely replaced routine karyotyping inclinical practice as part of the initial assessment ofnewborns with important CHD. Although CMA hasmany uses, particularly in the clinical and diagnosticsetting, such as excluding diagnoses of trisomy 21,22q11 deletion syndrome, and other major chromo-somal abnormalities, it is fairly limited from aresearch perspective.

The contribution of somatic mutations as a poten-tial genetic mechanism emerged after the discoveriesof the Reamon-Buettner and Borlak research groupand the Leipzig heart collection (14,15). This was ahighly attractive hypothesis that could explain theclinical presentation of many of the isolated, sporadicforms of CHD. A number of other research teamsattempted to replicate these findings using fresh-frozen tissue (as opposed to formalin-fixed hearts)but did not identify any important somatic mutations(16,17). Although subsequent studies suggest thatsomatic mutations are not a common cause of CHD,there is a possibility that they may play an as yetundetermined role in disease development in a

AB BR E V I A T I O N S

AND ACRONYM S

AVSD = atrioventricular septal

defect

CHD = congenital heart disease

CMA = chromosomal

microarray

CNV = copy number variations

ECA = extracardiac congenital

anomaly

GWAS = genome-wide

association studies

MPS = massively parallel

sequencing

NDD = neurodevelopmental

disabilities

SNP = single-nucleotide

polymorphism

TOF = tetralogy of Fallot

WGS = whole-genome

sequencing

J A C C V O L . 6 9 , N O . 7 , 2 0 1 7 Blue et al.F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0 A Clinician’s Guide to CHD Genetics

861

polygenic or multifactorial setting. Other novelmechanisms of disease have been proposed recently,including the concept of “off targets” of mutanttranscription factors, relevant in experimentalmodels of CHD, as well as cancer (18,19).

It was realized, through application of linkageanalysis and CMA, that most of the identified muta-tions were family specific and could not account forthe majority of presenting CHD cases, and so thefocus of CHD genetic research shifted to commonvariants. At the time, it was postulated that multiplecommon variants with small effects most likelyunderlie common diseases such as CHD, and thisbecame known as the common disease–commonvariant hypothesis (20). This theory was adopted intoour understanding of complex CHD and, after theapplication of a more advanced study design knownas genome-wide association studies (GWAS), led tothe identification of associations between a numberof genomic regions and CHD risk, including for atrialseptal defects (21), TOF (22), and left-sided CHD (23).Associations between variants in specific genes (ISL1)and CHD risk were also observed (24). In GWAS, sig-nificant associations are identified between a markerSNP in a genomic region and the CHD trait, yet inmost cases, we have little understanding of whichvariants in close proximity to the marker SNP that

FIGURE 1 Timeline of CHD Genetic Discoveries and the Genetic Tech

First report onCHD genetics

(Campbell, 1949)

Multifactorialinheritancehypothesis

(Nora, 1968)

Empiric RR inCHD

(Nora, 1970)Epidem

CHD –first r

(Ferenc

1950 1960 1970 1980

Genetic technologies/study designs are indicated by blue arrows and mar

septal defect; BWIS ¼ Baltimore-Washington Infant Study; CHD ¼ cong

exome sequencing; GWAS ¼ genome-wide association studies; MPS ¼ m

CHD; RR ¼ recurrence risk; sCHD ¼ syndromic CHD.

segregate with a haplotype have a causal role.Furthermore, the identification of a statisticalassociation between a genetic variation anddisease does not explain the underlyingbiology, which leaves many questionsregarding the mechanisms involved unan-swered. These associations could be highlysignificant from a statistical perspective buthave limited clinical relevance, with low oddsratios. Another limitation of GWAS is theinability to transfer findings to other pop-ulations because of allele frequency differ-ences between ethnic groups. One of thebiggest limitations, however, is that the SNPgenotypes identified by GWAS have onlyexplained a small percentage of thepopulation-attributable risk, for example,only 9% for atrial septal defects (21). As aresult, although they contribute to our overallunderstanding of CHD development, GWAShave failed to explain the majority of genetic

variation observed in complex disease, giving rise tothe term missing heritability (25). After this, the the-ory emerged that the missing heritability might beexplained by rare variants with larger effects acting inconjunction with common variants and environ-mental factors (26).

nologies and Study Designs Used

iology of BWISeportz, 1985)

First single-genemutation

associated withsyndromic CHD

(TBX5)(Basson, 1997)

First single-genemutation associatedwith nonsyndromic

CHD (NKX2-5)(Schott, 1998)

GWAS in CHD(Cordell, 2013)

Somatic mutations(Reamon-Buettner, 2004)

De novo variantsin sporadic CHD

(Zaidi, 2013)CNV analysis in CHD(Thienpont, 2007) CHD gene

panel(Blue, 2014)

ES in CHD & NDD(Homsy, 2015)ES in AVSD

(Priest, 2016)Clinical ES

in familial CHD(LaHaye, 2016)

ES in sCHD vs nsCHD(Sifrim, 2016)

ES in familial CHD(Arrington, 2012)

GATA4 geneassociated with

CHD(Garg, 2003)

1990

Linkage analysis

CMA

GWAS

MPS

2000 2010 2020

k the approximate time when the technology was developed and used. AVSD ¼ atrioventricular

enital heart disease; CMA ¼ chromosomal microarray; CNV ¼ copy-number variation; ES ¼assively parallel sequencing; NDD ¼ neurodevelopmental disabilities; nsCHD ¼ nonsyndromic

TABLE 1 Summary of Recent MPS Study Findings on Sporadic CHD

First Author(Ref. #) Yr Cohort Primary Findings Comments

Zaidi et al. (34) 2013 Sporadic nonsyndromic CHD (includingcases with NDD/ECA)*

� Significant excess of damaging de novo variantsin HHE and chromatin-modifying genes

� De novo variants implicated in 10% of sporadicnonsyndromic CHD

� Cohort combines cases withisolated CHD and CHD þ NDD/ECA

� Chromatin-modifying pathwayimplicated

Homsy et al.(35)

2015 Sporadic nonsyndromic isolated CHD vs.CHD þ NDD vs. CHD þ NDD þ ECA*

� Significant excess of damaging de novo variantsin HHE and across all genes

� Significant excess of damaging de novo variantsin CHD þ NDD þ ECA and CHD þ NDD but notisolated CHD

� Shared genetic contributions to CHD, NDDand ECA

� Cohort incorporates cohort fromstudy by Zaidi et al. (34)

� Replicated findings from Zaidiet al. (34)

� Chromatin-modifying pathwayimplicated

Sifrim et al. (37) 2016 CHD þ ECA/facial gestalt (referred to assyndromic CHD) vs. nonsyndromicisolated CHD

� Significant increase in damaging de novo variantsin CHD þ ECA/facial gestalt in HHE genes

� Significant increase in damaging inherited variantsin nonsyndromic CHD in HHE genes and acrossall genes

� Different cohort with differentparticipant definitions

� Confirms findings from Homsyet al. (35) re CHD þ NDD/ECA

� Additional pathways implicated

*Excluding patients with an established genetic diagnosis/syndrome.

CHD ¼ congenital heart disease; ECA ¼ extracardiac congenital anomaly; HHE ¼ high heart expression; MPS ¼ massively parallel sequencing; NDD ¼ neurodevelopmental disabilities.

Blue et al. J A C C V O L . 6 9 , N O . 7 , 2 0 1 7

A Clinician’s Guide to CHD Genetics F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0

862

Advances in technology provided a potential tool toexplain the missing heritability. Collectively termedmassively parallel sequencing (MPS), this tool caneither be targeted (exome sequencing and disease-specific gene panels) or nontargeted (whole-genomesequencing [WGS]). Through its unbiased approach,many sources of variation, including common andrare variants, indels (insertions and deletions), andcopy-number variations (CNVs), can be identifiedwith MPS, providing a detailed picture of genomicvariation compared with SNP-only data in GWAS (27).Furthermore, by supporting the analysis of individualvariants, MPS has enabled research to be directed backto affected individuals and their families, as opposedto the population-applicable results of GWAS. Theadvent of this technology marked a change in focusfrom large cohorts with similar phenotypes and ethnicbackground to studies of well-phenotyped families,more akin to the clinical workflow.

Since its development, MPS technology, specif-ically exome sequencing, has been applied to bothfamilial and sporadic forms of CHD. Initially, familialforms were expected to be easily “solved” on the basisof the presumed Mendelian inheritance of the heartdefects. Although causal variants have been identifiedin some families (28–32), in practice, there areconsiderable challenges in applying this approach.Variant interpretation can be challenging, and it is notalways possible to firmly establish the pathogenicityof a variant. This is particularly so in small, dominantfamilies, in which the number of candidate variantscan be such that it is impossible to filter down to asingle, strong candidate. In a recent example, no high-effect coding variants were identified in a multigen-erational family with bicuspid aortic valve and other

forms of CHD, despite the use of 3 commonly usedvariant selection strategies and linkage analysis incombination with exome sequencing (33). In somesettings, yield can be improved by combining exomesequencing with other techniques, such as homozy-gosity mapping in consanguineous families (31,32).Together, these studies highlight the difficulties withMPS technology, even in cases with seemingly Men-delian inheritance, in which affected/nonaffectedstatus is easily discerned.

Exome sequencing has also been extended to spo-radic forms of CHD (Table 1). The first study to applythis tool to sporadic CHD using parent-offspring triosidentified a significant increase (odds ratio: 2.53) in denovo coding variants in w4,000 genes highlyexpressed in the developing heart in more than 300cases of severe CHD (34). Enriching for deleterious denovo mutations by first removing missense mutationsat weakly conserved sites, followed by removing thoseat highly conserved sites, thereby leaving onlynonsense, splicing, and frameshift mutations, signifi-cantly increased the association with disease to oddsratios of 3.60 and 7.50, respectively (p ¼ 0.001).A subsequent study in which a significant excess ofdamaging de novo variants was identified across allgenes (enrichment ¼ 1.4), as well as in genes highlyexpressed in the developing heart (enrichment ¼ 2.4)(35), replicated these findings. The subsequent studyalso presented important genetic evidence to supportthe previously suggested genetic link between heartand brain development in nonsyndromic CHD (36).Amore recent study applied exome sequencing to bothsyndromic and nonsyndromic forms of CHD (37). Inmany of these studies, the term syndromicwas used todescribe clinical perceptions rather than clinical

FIGURE 2 The Number and Effect Sizes of the Contributing

Genetic Variants for the Different Inheritance Modes

Observed in CHD in Conjunction With Disease Prevalence

Increasing prevalence

Mendelian CHD Oligogenic CHD

Number and effect sizes of contributing variants

Complex CHD

Each spot represents a variant contributing to the phenotype

and the size of the circle is representative of the effect of the

variant on the phenotype. Adapted with permission to CHD

from Marian (20). CHD ¼ congenital heart disease.

J A C C V O L . 6 9 , N O . 7 , 2 0 1 7 Blue et al.F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0 A Clinician’s Guide to CHD Genetics

863

diagnoses. As such, in this study, syndromic caseswere defined as patients with CHD presenting with anextracardiac congenital anomaly (ECA) or a distinctfacial gestalt, and not people with an establishedgenetic diagnosis or syndrome. A significant excess ofde novo protein-truncating variants was identifiedin the syndromic cases, whereas the nonsyndromiccases exhibited a significant excess of inheritedprotein-truncating variants. Exome sequencing hasalso been applied to specific types of CHD, namely,atrioventricular septal defects (AVSD) (38,39).

Other targeted MPS methodologies, such as tar-geted gene panels, have been applied to CHD. Thefirst CHD-specific gene panel identified the cause forthe heart defects in 31% of the cohort comprising fa-milial, nonsyndromic CHD (40). Interestingly, in justover one-half of those diagnosed, the causal gene wasassociated with a specific syndrome; however, thephenotypic presentation was primarily cardiac based.In a subsequent similar study, likely causal variantswere identified in 46% of the cohort, also comprisingfamilial, nonsyndromic CHD (41). Although genepanels are an attractive alternative in that they miti-gate some of the issues associated with exomesequencing, such as inconsistent coverage, lengthyanalyses, and large data storage requirements, thiscomes at the expense of novel gene discovery. “CHDpanels” are now available commercially alongsidecardiomyopathy and arrhythmia panels. With thedeclining costs of exome and genome sequencing,opportunities to create “virtual panels” focusing ongenes of interest, without specific analysis of unre-lated genes, is being realized (42).

Taken together, the studies discussed in thepreceding text demonstrate the success of MPSapplication in CHD; however, they also highlight thatfor many of these individuals and families, theanswer is likely to lie beyond the exome and in thegenome, such as in the noncoding regulome (43). Italso raises the possibility that some familial forms ofCHD, particularly in small families, are oligogenic/polygenic, requiring the additive effects of 2 or morevariants for disease manifestation. To date, no pub-lished studies have applied WGS to CHD, and it islikely that this more comprehensive analysis willprovide further insights into the genetic architectureof both familial and sporadic forms of disease.

CONTRIBUTIONS FROM MPS STUDIES

Aside from the potential personal benefits to affectedindividuals and their families, MPS studies have alsosignificantly contributed to our understanding of thegenetic architecture of CHD. First, they suggest thatnot all seemingly Mendelian forms of CHD are due tosingle gene mutations, but could be oligogenic orpolygenic. This is evident in families such as the largefamily segregating bicuspid aortic valve in a seem-ingly dominant inheritance pattern, noted previ-ously, in which no likely causal variants wereidentified (33). Although this could be due, amongother reasons, to the causal variant residing in thenoncoding region of the genome, in which case itcould potentially be identified through the applica-tion of WGS, the idea that not all seemingly Mende-lian forms of CHD are due to a single genetic mutationis an attractive explanation for the reduced pene-trance and variable expressivity that often accom-pany familial CHD.

Second, a novel and common theme among someof the more recent studies is the suggestion of anunderlying burden of genetic variation in CHDdevelopment. As demonstrated by these studies, in-dividuals with CHD have identifiable additional ge-netic variation in genes expressed during heartformation, irrespective of whether they are consid-ered complex or Mendelian (34,35,37,39,40). Thedistinction between Mendelian and complex formsof disease is becoming increasingly blurred andeventually might be viewed as a continuum apartfrom single-gene disorders (Figure 2). As mentioned,this is evident clinically in the reduced penetranceand variable expression often present in familial CHD(8), and it is also highlighted in the many familieswith enough affected family members for them not tobe considered sporadic, but with too few to beconsidered familial.

Blue et al. J A C C V O L . 6 9 , N O . 7 , 2 0 1 7

A Clinician’s Guide to CHD Genetics F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0

864

Another interesting discovery arising from thesestudies is the contribution of de novo variants in spo-radic CHD development (Table 1) (34,35,37). The find-ings by Zaidi et al. (34) implicate de novo variants inkey heart-expressing genes in the pathogenesis ofw10% of sporadic cases. However, although thisfinding marks a novel and significant contribution toour overall understanding of disease mechanisms forthe majority of presenting cases in the clinical world, itfocuses attention on the remaining 90% of sporadicCHD that arises from causes other than or in addition tode novo variation in known pathogenic genes. Simi-larly, Homsy et al. (35) and Sifrim et al. (37) identified asignificant excess of damaging de novo variants in CHDcases with neurodevelopmental disabilities (NDDs) orECA in genes highly expressed in the developing heartand brain. These findings, therefore, implicate de novovariations not only in isolated CHD but in develop-mental disorders in general.

Finally, the previously suggested genetic linkbetween heart and brain development was demon-strated (35). This is a timely and important discovery,because CHD physicians and surgeons have demon-strated that the likelihood of neurodevelopmentalissues in association with complex neonatal CHDdepends more on pre-existing factors and less on thepatient’s clinical course, including specific operativetechniques (such as deep hypothermic circulatoryarrest) that were previously thought to be highlycontributory (44,45). The idea that common geneticfactors contribute to both CHD and NDD (36) is alsosupported by the many genetic syndromes in whichboth cardiac and NDD occur, such as Williams syn-drome, Alagille syndrome, Noonan syndrome, and22q11 deletion syndrome, among others (46). How-ever, the findings from Homsy et al. (35) provided theimportant evidence that such a link exists, throughthe identification of a significant enrichment ofdamaging de novo variants among published NDDgenes with high heart expression in CHD cases withNDD. The marked difference in de novo mutationburden between syndromic and nonsyndromic CHDidentified by Sifrim et al. (37) further confirms thisfinding, because NDD is the most commonly associ-ated extracardiac malformation. Not all cases withCHD have NDD and vice versa, which suggests thatin addition to exhibiting pleiotropic effects, thesegenetic variants also display variable expressivity.

BURDEN OF GENETIC VARIATION MODEL

The concept of additional genetic burden in specificgenes is evident in a number of diseases, includingautism and peripheral neuropathy. Autism is

considered a multigenic disorder, with current esti-mates implicating w1,000 genes in disease causation,including both rare and common genetic variants(47). In peripheral neuropathy, a significant increasein rare variants in 58 relevant genes in case versuscontrol subjects was demonstrated, with the com-bined effect of rare variants contributing to mutationburden and phenotypic variability, including diseaseseverity (48). The first evidence of genetic burden inCHD was presented by Zaidi et al. (34). In their studyon sporadic CHD, they identified a significant excessof de novo variants in genes involved in histone3 lysine 4 methylation. Subsequently, an excess ofdamaging de novo variants was identified in heart-expressing genes, as well as across all genes (35).Since then, a number of other studies have reportedadditional burden of de novo or rare variants in spe-cific genes in patients with CHD. Gene burden testingon patients with nonsyndromic AVSD identified asignificant enrichment of rare variants in the geneNR2F2 after the analysis of 9 genes with verified denovo variants in 13 trios (38). Using a rare-diseaseinheritance model in genes previously associatedwith CHD, Priest et al. (39) identified an increase ofinherited and de novo variants in 16 genes in patientswith AVSD compared with control subjects. Sifrimet al. (37) proposed a similar model for nonsyndromicCHD, identifying a significant excess of inherited andde novo protein-truncating variants in CHD-associated genes. Interestingly, the significantexcess of rare inherited variants was still apparentafter the removal of genes associated with CHD andother developmental disorders, which suggests con-tributions from novel CHD-associated genes. Thefindings of these studies therefore suggest a combi-nation of de novo and inherited rare variants in dis-ease causation.

Although the focus of the studies mentionedpreviously was on sporadic forms of disease, geneticburden has also been reported in syndromic CHDpatients, specifically trisomy 21 and 22q11 deletionsyndrome (49,50). These studies identified an in-crease in additional variation in syndromic patientspresenting with a heart defect compared with thosewith no heart defects. Surprisingly, additional geneticvariation has also been reported in familial forms ofCHD. In addition to achieving a clinically actionablemolecular diagnosis in 31% of the cohort comprisingfamilial forms of CHD, Blue et al. (40) identifieda significant increase in rare and low-frequencyvariants (minor allele frequency <0.05) in familialCHD case subjects compared with control subjects.Importantly, this difference was observed after thecausal variants in 31% of the CHD cases were

J A C C V O L . 6 9 , N O . 7 , 2 0 1 7 Blue et al.F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0 A Clinician’s Guide to CHD Genetics

865

removed, demonstrating the presence of additionalgenetic variation, even in families in which a pre-sumed single causal variant was identified. This sur-prising finding supports the hypothesis regarding theputative burden of this variation; however, the rela-tive contribution of these additional variants to thedevelopment of the heart defect, especially in thoseindividuals or families in which the pathogenicvariant was identified as being the sole cause, isunclear at present. One possibility is that theseadditional variants contribute to the final presenta-tion of the heart defect and could therefore explainin part the variable expression and penetrance oftenevidenced in families affected by CHD. Furthermore,the burden of genetic variation model for CHDrepresents a plausible explanation for the occurrenceof CHD through the existence of a threshold levelof variation, above which other nongenetic factors,such as teratogens and other environmental factors,including stochastic effects, might exert a significanteffect.

Additional genetic burden has also been identifiedin CHD cases with NDD or ECA (35). As noted pre-viously, exome sequencing in 559 CHD cases withNDD or ECA identified a 3-fold enrichment ofdamaging de novo variants of genes with high heartexpression compared with controls and a 4.7-foldenrichment in genes with high heart expression in138 CHD cases with both NDD and ECA. Importantly,genes with high heart expression were identified inembryonic mouse hearts, not humans, in this study.Although additional genetic burden has been re-ported in other neurological disorders, such asautism and peripheral neuropathy (47,48), the sig-nificant enrichment identified in this study wasobserved in genes highly expressed in the heart,which suggests that these de novo variants havepleiotropic effects, as well as the potential to affectdevelopment of the heart, brain, and other organs.The concept of burden of genetic variation is there-fore likely to be a common theme for congenitaldisorders in general, and further research is requiredto understand these complex developmental pro-cesses and interactions.

PATIENT GENOTYPES AND

CLINICAL OUTCOMES

Understanding the genetic basis of CHD is importantfor affected individuals and their families in regardto family planning and further clinical management;however, there may be other, wider implicationsfor the affected individual. With the majority ofpatients born with CHD surviving to adulthood, the

focus of research has shifted to improving outcomespost-surgery, treatment, and quality of life ofpatients. Much focus has been directed to the bestways of providing high quality and consistent carethrough improving surgical techniques and periop-erative care, but variations in patient outcomesremain poorly understood. Chromosomal abnormal-ities and syndromes are well-described risk factors formortality and morbidity in neonatal cardiac surgery(51), but subtler genetic variation is also likely tobe relevant.

Among the first studies highlighting the impor-tance of patient genotype on post-surgical outcomesin CHD were studies investigating polymorphisms inapolipoprotein E. These studies identified an associ-ation between the apolipoprotein E ε2 allele andadverse neurodevelopmental outcomes (52), as wellas impaired weight gain (53) post-cardiac surgeryin infancy. Since then, a number of other genotypeshave been associated with various surgical outcomes.Examples include the carbamoyl-phosphate synthe-tase I Thr1405NAsn genotype and its associationwith increased post-operative pulmonary arterypressure (54) and the angiotensin-converting enzymeinsertion/deletion polymorphism and the associatedincreased risk of post-operative tachycardia (55).Cardioprotective effects have also been described,such as the common mitochondrial aldehydedehydrogenase-2 Glu504Lys genotype, which resultsin increased tolerance to ischemic and reperfusioninjury (56). Similarly, protective variants in hypoxia-inducible factor 1a have been associated with pres-ervation of right ventricular function, less rightventricular dilation, and fewer reinterventions inpatients with TOF, which suggests that pre-operativeadaption to hypoxia may influence post-operativeright ventricular phenotype (57).

Other forms of genetic variation have been asso-ciated with post-surgical outcomes. The study byCarey et al. (58) identified a significant associationbetween pathogenic CNVs and poor linear growth, aswell as poor neurocognitive outcomes in patientswith single-ventricle heart defects. More recently,Kim et al. (59) found that CNVs present in childrenwith isolated CHD were associated with a 2.55-foldincreased risk of death or transplantation after sur-gery. An earlier study by the same group identifieda 16-fold increased risk of death or heart trans-plantation post-surgery in infants homozygous forboth the VEGFA and SOD2 SNPs, rs833069 andrs2758331, respectively (60). Collectively, thesestudies suggest that CNVs, as well as specific geno-types, are important modifiers of post-surgery sur-vival and heart transplantation.

CENTRAL ILLUSTRATION Genetics of CHD: Percentages of Known and Unknown Causes of theDifferent Forms of Presenting Nonsyndromic CHD Patients

Familial CHD

Variants inknown genes

31%-46%(Gene panel)

Unknowncause

54%-69%(WGS)

CHD + ECA

CNV25%

(CMA)

Single genevariants

26%(ES)

Unknowncause49%

(WGS)

Sporadic CHD

De Novo~10% (ES)

CNV3%-10%

(CMA)Unknown

cause[Includesinherited

rare, high-riskvariants]

~80%(WGS)

Blue, G.M. et al. J Am Coll Cardiol. 2017;69(7):859–70.

Percentages are based on current published reports (34,35,37,40–42,64,66). The molecular technologies commonly used or recommended

to identify the various genetic causes are indicated in parentheses. Pedigrees below provide examples of the inheritance modes expected for

the different forms of presenting nonsyndromic CHD patients. Blue circles/squares represent individuals affected by CHD, and light blue

represents ECA. CHD ¼ congenital heart disease; CMA ¼ chromosomal microarray; CNV ¼ copy-number variation; ECA ¼ extracardiac

congenital anomalies; ES ¼ exome sequencing; WGS ¼ whole-genome sequencing.

Blue et al. J A C C V O L . 6 9 , N O . 7 , 2 0 1 7

A Clinician’s Guide to CHD Genetics F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0

866

The identification of specific genotypes or allelesassociated with improved long-term surgical care is agrowing field of research, not to mention a field withsignificant implications for direct clinical manage-ment. The ability of MPS to identify both rare andcommon variants is likely to uncover many moreassociations between genotype and long-term out-comes, thereby enabling the provision of individual-ized care and ensuring optimal treatment.

IMPLICATIONS FOR PATIENTS

On the basis of everything we have learned fromlinkage analysis, GWAS, and more recently, MPStechnology, we have a much clearer picture of CHDcausation than ever before. Although we cannot yetpinpoint the genetic variant(s) causing heart defectsin every patient, we have a much better under-standing of the types of variants involved and theroles they play (Central Illustration).

In patients with accurately phenotyped familialCHD, we know that there is a 31% to 46% chance ofidentifying the causal genetic variants and that thesevery often reside in known CHD genes (40,41). Fa-milial CHD is therefore well suited to targeted genepanels comprising genes commonly associated withCHD as an initial test. Although genetic testing infamilial forms of CHD is currently not routine, thechance of achieving a molecular diagnosis with MPStechnology is now comparable to that of other geneticconditions (61). Providing a molecular diagnosis notonly has implications for psychosocial well-being andfuture family planning but can also significantly in-fluence the clinical management of some patientswith variants in specific genes known to be associatedwith the future development of conduction defects orcardiomyopathies, such as NKX2-5, TBX5, and TBX20(8,62,63). Although novel or rare variants with largeeffects are regarded as the “major players” in familialCHD, the final presentation of disease (including the

J A C C V O L . 6 9 , N O . 7 , 2 0 1 7 Blue et al.F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0 A Clinician’s Guide to CHD Genetics

867

presence, severity, or type of heart defect) may bedetermined by additional genetic variation, such asother rare and common variants with smaller effects(33,40). However, we also need to be aware of thepossibility that some familial forms with a seeminglyMendelian inheritance might in fact be oligogenic orpolygenic in nature.

In patients with sporadic CHD, we know that denovo coding variants in known and novel CHD genesaccount for a small proportion of cases (w10%)(34,35). Many of these de novo changes were identi-fied in genes involved in the production, reading, orremoval of histone 3 lysine 4 methylation, therebyimplicating this chromatin-modifying pathway as animportant contributor to CHD development (34,35).There is an over-representation of genes associatedwith gene ontology terms relating to protein phos-phorylation, as well as neural tube and cardiacdevelopment, in addition to chromatin modification,thereby implicating these as additional pathwaysof interest (37).

We also know that approximately 3% to 10% ofisolated CHD cases are caused by CNVs (64). MPStechnologies and chromosomal microarrays aretherefore appropriate tools for initial assessment ofisolated CHD cases; however, it is important toremember that compared with familial forms of CHD,the chance of identifying a causal variation (includinga gene mutation or CNV) is low (up to 10%).

For the remaining w80% of isolated CHD cases, theinheritance is presumed to be multifactorial, impli-cating many genetic and environmental factors indisease causation. We know that this geneticcomponent includes a significant excess of inheritedrare, damaging variants (37) and that these mostlikely act in conjunction with other more commonvariants conferring a small effect, which cumulativelygives rise to the heart defect (Figure 2). Environ-mental and other factors will, of course, also make acontribution. Although WGS is the most suitable toolfor assessment, because it covers the entire spectrumof genetic variation, interpretation of findings,particularly in noncoding regions, remains a signifi-cant obstacle. For this group of patients, participationin research studies to identify the types of variantsand pathways involved in disease development iscurrently the best approach.

It is well known that patients with CHD, includingthose with isolated CHD, are at an increased risk ofECA, such as neurological, genitourinary, and cranio-facial malformations, among others (65). In patientspresenting with ECA in addition to the heart defects,the diagnostic rate of exome sequencing is increasedsignificantly (26%) and comparable to that of other

genetic conditions (29.5%) (66). The detection ofcausal CNVs is also significantly increased (up to 25%)in patients with CHD and ECA (64). For this group ofpatients, CMA and MPS can therefore be considered asgood options from a diagnostic perspective (10,67).

As demonstrated by Homsy et al. (35), we knowthat some genetic variants can predispose an in-dividual to the development of both heart andbrain abnormalities. The significant enrichment indamaging de novo mutations in genes highlyexpressed in the heart and brain in the CHD plus NDDcohort and the lack thereof in the CHD-only cohortare strongly suggestive of the development of NDDin CHD infants comprising this group of variants.Specifically, this study showed that patients withdamaging de novo mutations in 69 genes implicatedin both CHD and NDD had a significantly increasedrisk of NDD compared with control subjects. Inparticular, damaging mutations in genes involved inchromatin modifiers conferred the highest risk forNDD. Collectively, this information is beginning topinpoint distinct genes and genetic variants associ-ated with an increased risk of NDD development inCHD patients. The ability to conduct a risk analysis toidentify which CHD patients are at increased risk ofdeveloping NDD on the basis of their genotypes is fastbecoming a reality.

In the study by Homsy et al. (35), the investigatorsalso identified a significant enrichment of geneshighly expressed in the heart in CHD cases presentingwith ECA other than NDD, including craniofacial,skeletal, and genitourinary malformations. Morerecently, a genetic link between congenital renal de-fects and nonsyndromic CHD has been identified (68),thereby providing further evidence for the geneticmechanisms underlying the observed increase in ECAin children with nonsyndromic CHD. Although thiswas not clarified by the investigators, the renal ab-normalities identified in this study could be indica-tive of syndromic CHD.

The studies discussed describe a new understand-ing of the role of genetic variation; however, we mustbe cautious not to over-call the likelihood of a geneticdiagnosis on the basis of bioinformatics analysesalone, because there is often little evidence that avariant identified is the cause of disease. By way ofexample, of the first 15 genes identified in Table 2 byZaidi et al. (34) as candidate CHD risk genes, 4 haveheart defects in null mice, 5 have no heart defect in nullmice, and 5 have no supporting mouse data. Bioinfor-matics predictions of deleterious effects could bewrong in up to 40% of cases (69), particularly whenfunctional analyses or animal models have beenestablished. Requirements for interpretation of

PERSPECTIVES

COMPETENCY IN PATIENT CARE: With advances

in genetic technology, molecular diagnostic yield in

patients with nonsyndromic familial CHD, as well as

CHD and extracardiac congenital anomalies, is com-

parable to that of other genetic diseases and should

be considered. The establishment of a molecular

diagnosis has important implications for patient

management and family planning.

TRANSLATIONAL OUTLOOK: A genetic and mo-

lecular diagnosis is becoming a reality for many pa-

tients, particularly those with known syndromes,

additional birth abnormalities, or neurodevelopmental

delay. Defining the burden of variation underlying

complex, sporadic forms of disease is the next horizon.

Blue et al. J A C C V O L . 6 9 , N O . 7 , 2 0 1 7

A Clinician’s Guide to CHD Genetics F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0

868

research-generated data, a belief in “big genomics,”and affected families’ thirst for information are placingincreasing demands on clinical genetics services.

CONCLUSIONS

There are likely to be w400 genes involved in thecausation of CHD, many of which are yet to be iden-tified (34,35). Rare inherited variants in genes notpreviously associated with CHD or associated extrac-ardiac manifestations, including NDD, are likely tobe highly relevant (37). MPS technology is thereforelikely to reveal many more coding genes associatedwith and contributing to the intricate developmentalnetwork of cardiogenesis. Although WGS is currentlystill in its infancy, and further refinements in infor-mation analysis are required, it is likely to identifycontributions of the noncoding regions of the genometo CHD causation, including from regulatory elementsand microRNAs (43,70). Furthermore, advances inbioinformatics tools will enable more sophisticatedanalyses and, together with larger collaborative ef-forts, provide an opportunity to assess the burden ofvariation predisposing toward defects in heart devel-opment in complex forms of disease.

Epigenetic factors, as well as the interplay betweengenetic and environmental contributions, such ashypoxia during embryogenesis, are likely to beimportant contributors (71); however, studies toinvestigate these factors, particularly the interactionbetween genetic and environmental contributors, arecomplex and could require large-scale collaborativeefforts. The greatest bottleneck, however, will be inmaking sense of the information and effects of indi-vidual mutations, as well as how they result in orpredispose an individual to CHD.

The translation of this information into the clinicalworkflow and patient care will need to be promotedand refined. A recent scientific statement addressesthe requirements for the effective integration of core

competencies in genetics and genomic knowledgerelevant to specialists in the field (72). For an indi-vidual with isolated, sporadic CHD, genetic testing is,at present, a low-yield exercise, but major advancesare expected in coming years. However, we can nowprovide a genetic diagnosis to many more patientswith familial and syndromic CHD than ever before,and we have a framework and the technology toevaluate the polygenic burden in common forms ofCHD. We now have a unifying hypothesis that mayaccount for many instances of NDD in children withCHD, further establishing the role of genetics in CHDas a relevant clinical focus and not just a scientificwork in progress.

ADDRESS FOR CORRESPONDENCE: Professor DavidWinlaw, Heart Centre for Children, The Children’sHospital at Westmead, Locked Bag 4001, WestmeadNSW 2145, Australia. E-mail: david.winlaw@health.nsw.gov.au.

RE F E RENCE S

1. Campbell M. Genetic and environmental factorsin congenital heart disease. QJM 1949;18:379–91.

2. Nora JJ. Multifactorial inheritance hypothesisfor the etiology of congenital heart diseases: thegenetic-environmental interaction. Circulation1968;38:604–17.

3. Ferencz C, Rubin JD, McCarter RJ, et al.Congenital heart disease: prevalence at livebirth:the Baltimore-Washington Infant Study. Am JEpidemiol 1985;121:31–6.

4. Schott JJ, Benson DW, Basson CT, et al.Congenital heart disease caused by mutations in

the transcription factor NKX2-5. Science 1998;281:108–11.

5. Garg V, Kathiriya IS, Barnes R, et al. GATA4mutations cause human congenital heart defectsand reveal an interaction with TBX5. Nature 2003;424:443–7.

6. Basson CT, Bachinsky DR, Lin RC, et al.Mutations in human TBX5 [corrected] causelimb and cardiac malformation in Holt-Oramsyndrome [published correction appears inNat Genet 1997;15:411]. Nat Genet 1997;15:30–5.

7. Garg V, Muth AN, Ransom JF, et al. Mutations inNOTCH1 cause aortic valve disease. Nature 2005;437:270–4.

8. Kirk EP, Sunde M, Costa MW, et al. Mutations incardiac T-box factor gene TBX20 are associatedwith diverse cardiac pathologies, including defectsof septation and valvulogenesis and cardiomyop-athy. Am J Hum Genet 2007;81:280–91.

9. Calcagni G, Digilio M, Sarkozy A, Dallapiccola B,Marino B. Familial recurrence of congenital heartdisease: an overview and review of the literature.Eur J Pediatr 2007;166:111–6.

J A C C V O L . 6 9 , N O . 7 , 2 0 1 7 Blue et al.F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0 A Clinician’s Guide to CHD Genetics

869

10. Thienpont B, Mertens L, de Ravel T, et al.Submicroscopic chromosomal imbalances detec-ted by array-CGH are a frequent cause ofcongenital heart defects in selected patients. EurHeart J 2007;28:2778–84.

11. Breckpot J, Tranchevent LC, Thienpont B, et al.BMPR1A is a candidate gene for congenital heartdefects associated with the recurrent 10q22q23deletion syndrome. Eur J Med Genet 2012;55:12–6.

12. Hitz MP, Lemieux-Perreault LP, Marshall C,et al. Rare copy number variants contribute tocongenital left-sided heart disease [publishedcorrection appears in PLoS Genet 2013;9]. PLoSGenet 2012;8:e1002903.

13. Geng J, Picker J, Zheng Z, et al. Chromosomemicroarray testing for patients with congenitalheart defects reveals novel disease causing lociand high diagnostic yield. BMC Genomics 2014;15:1127.

14. Reamon-Buettner SM, Borlak J. TBX5 muta-tions in non-Holt-Oram syndrome (HOS) mal-formed hearts. Hum Mutat 2004;24:104.

15. Reamon-Buettner SM, Borlak J. Somatic NKX2-5 mutations as a novel mechanism of disease incomplex congenital heart disease. J Med Genet2004;41:684–90.

16. Draus JM Jr., Hauck MA, Goetsch M,Austin EH 3rd, Tomita-Mitchell A, Mitchell ME.Investigation of somatic NKX2-5 mutations incongenital heart disease. J Med Genet 2009;46:115–22.

17. Esposito G, Butler TL, Blue GM, et al. Somaticmutations in NKX2–5, GATA4, and HAND1 are nota common cause of tetralogy of Fallot or hypo-plastic left heart. Am J Med Genet A 2011;155:2416–21.

18. Luna-Zurita L, Stirnimann CU, Glatt S, et al.Complex interdependence regulates heterotypictranscription factor distribution and coordinatescardiogenesis. Cell 2016;164:999–1014.

19. Bouveret R, Waardenberg AJ, Schonrock N,et al. NKX2-5 mutations causative for congenitalheart disease retain functionality and are directedto hundreds of targets. Elife 2015;4:e06942.

20. Marian AJ. Molecular genetic studies of com-plex phenotypes. Transl Res 2012;159:64–79.

21. Cordell HJ, Bentham J, Topf A, et al. Genome-wide association study of multiple congenitalheart disease phenotypes identifies a susceptibilitylocus for atrial septal defect at chromosome 4p16.Nat Genet 2013;45:822–4.

22. Cordell HJ, Töpf A, Mamasoula C, et al.Genome-wide association study identifies loci on12q24 and 13q32 associated with tetralogy ofFallot. Hum Mol Genet 2013;22:1473–81.

23. Hanchard NA, Swaminathan S, Bucasas K, et al.A genome-wide association study of congenitalcardiovascular left-sided lesions shows associationwith a locus on chromosome 20. Hum Mol Genet2016;25:2331–41.

24. Stevens KN, Hakonarson H, Kim CE, et al.Common variation in ISL1 confers genetic suscep-tibility for human congenital heart disease. PLoSOne 2010;5:e10855.

25. Manolio TA, Collins FS, Cox NJ, et al. Findingthe missing heritability of complex diseases.Nature 2009;461:747–53.

26. Eichler EE, Flint J, Gibson G, et al. Missingheritability and strategies for finding the under-lying causes of complex disease. Nat Rev Genet2010;11:446–50.

27. Sadee W, Hartmann K, Seweryn M, Pietrzak M,Handelman SK, Rempala GA. Missing heritability ofcommon diseases and treatments outside theprotein-coding exome [published correction ap-pears in Hum Genet 2015;134:509]. Hum Genet2014;133:1199–215.

28. Arrington CB, Bleyl SB, Matsunami N, et al.Exome analysis of a family with pleiotropiccongenital heart disease. Circ Cardiovasc Genet2012;5:175–82.

29. Liu JJ, Fan LL, Chen JL, Tan ZP, Yang YF.A novel variant in TBX20 (p.D176N) identified bywhole-exome sequencing in combination with acongenital heart disease related gene filter isassociated with familial atrial septal defect.J Zhejiang Univ Sci B 2014;15:830–7.

30. Greenway SC, McLeod R, Hume S, et al. Exomesequencing identifies a novel variant in ACTC1associated with familial atrial septal defect. Can JCardiol 2014;30:181–7.

31. Ta-Shma A, Pierri CL, Stepensky P, et al. Iso-lated truncus arteriosus associated with a muta-tion in the plexin-D1 gene. Am J Med Genet A2013;161A:3115–20.

32. Tariq M, Belmont JW, Lalani S, Smolarek T,Ware SM. SHROOM3 is a novel candidate for het-erotaxy identified by whole exome sequencing.Genome Biol 2011;12:R91.

33. Martin LJ, Pilipenko V, Kaufman KMP, et al.Whole exome sequencing for familial bicuspidaortic valve identifies putative variants. Circ Car-diovasc Genet 2014;7:677–83.

34. Zaidi S, Choi M, Wakimoto H, et al. De novomutations in histone-modifying genes in congen-ital heart disease. Nature 2013;498:220–3.

35. Homsy J, Zaidi S, Shen Y, et al. De novo mu-tations in congenital heart disease with neuro-developmental and other congenital anomalies.Science 2015;350:1262–6.

36. Gaynor JW, Stopp C, Wypij D, et al. Interna-tional Cardiac Collaborative on Neurodevelopment(ICCON) Investigators. Neurodevelopmental out-comes after cardiac surgery in infancy. Pediatrics2015;135:816–25.

37. Sifrim A, Hitz MP, Wilsdon A, et al. Distinctgenetic architectures for syndromic and non-syndromic congenital heart defects identified byexome sequencing. Nat Genet 2016;48:1060–5.

38. Al Turki S, Manickaraj AK, Mercer CL, et al.Rare variants in NR2F2 cause congenital heartdefects in humans [published corrections appearin Am J Hum Genet 2016;98:592 and Am J HumGenet 2014;95:126]. Am J Hum Genet 2014;94:574–85.

39. Priest JR, Osoegawa K, Mohammed N, et al. Denovo and rare variants at multiple loci support theoligogenic origins of atrioventricular septal heartdefects. PLoS Genet 2016;12:e1005963.

40. Blue GM, Kirk EP, Giannoulatou E, et al. Tar-geted next-generation sequencing identifiespathogenic variants in familial congenital heartdisease. J Am Coll Cardiol 2014;64:2498–506.

41. Jia Y, Louw JJ, Breckpot J, et al. The diagnosticvalue of next generation sequencing in familialnonsyndromic congenital heart defects. Am J MedGenet A 2015;167A:1822–9.

42. LaHaye S, Corsmeier D, Basu M, et al. Utiliza-tion of whole exome sequencing to identifycausative mutations in familial congenital heartdisease. Circ Cardiovasc Genet 2016;9:320–9.

43. Postma AV, Bezzina CR, Christoffels VM. Ge-netics of congenital heart disease: the contributionof the noncoding regulatory genome. J HumGenet 2016;61:13–9.

44. Gaynor JW. The encephalopathy of congenitalheart disease. J Thorac Cardiovasc Surg 2014;148:1790–1.

45. Marelli A, Miller SP, Marino BS, Jefferson Al,Newburger JW. Brain in congenital heart diseaseacross the lifespan: the cumulative burden ofinjury. Circulation 2016;133:1951–62.

46. Fahed AC, Nemer GM. Genetic causes of syn-dromic and non-syndromic congenital heart disease.In: Cooper DN, Chen J-M, editors. Mutations inHuman Genetic Disease. Rijeka, Croatia: InTechOpen,2012:chap 6. Available at: http://www.intechopen.com/books/mutations-in-human-genetic-disease/genetic-causes-of-syndromic-and-non-syndromic-congenital-heart-disease. Accessed December 17,2016.

47. Sanders SJ, He X, Willsey AJ, et al. Insights intoautism spectrum disorder genomic architecture andbiology from 71 risk loci. Neuron 2015;87:1215–33.

48. Gonzaga-Jauregui C, Harel T, Gambin T, et al.Exome sequence analysis suggests that geneticburden contributes to phenotypic variability andcomplex neuropathy. Cell Rep 2015;12:1169–83.

49. Ackerman C, Locke Adam E, Feingold E, et al.An excess of deleterious variants in VEGF-Apathway genes in Down-syndrome-associatedatrioventricular septal defects. Am J Hum Genet2012;91:646–59.

50. Mlynarski EE, Sheridan MB, Xie M, et al. Copy-number variation of the glucose transporter geneSLC2A3 and congenital heart defects in the 22q11.2deletionsyndrome.AmJHumGenet2015;96:753–64.

51. Landis BJ, Cooper DS, Hinton RB. CHD associ-ated with syndromic diagnoses: peri-operative riskfactors and early outcomes. Cardiol Young 2016;26:30–52.

52. Gaynor JW, Nord AS, Wernovsky G, et al.Apolipoprotein E genotype modifies the risk ofbehavior problems after infant cardiac surgery.Pediatrics 2009;124:241–50.

53. Burnham N, Ittenbach RF, Stallings VA, et al.Genetic factors are important determinants ofimpaired growth after infant cardiac surgery.J Thorac Cardiovasc Surg 2010;140:144–9.

54. Canter JA, Summar ML, Smith HB, et al. Ge-netic variation in the mitochondrial enzymecarbamyl-phosphate synthetase I predisposeschildren to increased pulmonary artery pressurefollowing surgical repair of congenital heart

Blue et al. J A C C V O L . 6 9 , N O . 7 , 2 0 1 7

A Clinician’s Guide to CHD Genetics F E B R U A R Y 2 1 , 2 0 1 7 : 8 5 9 – 7 0

870

defects: a validated genetic association study.Mitochondrion 2007;7:204–10.

55. Smith AH, Flack EC, Borgman KY, et al.A common angiotensin-converting enzyme poly-morphism and preoperative angiotensin-converting enzyme inhibition modify risk oftachyarrhythmias after congenital heart surgery.Heart Rhythm 2014;11:637–43.

56. Zhang H, Gong DX, Zhang YJ, Li SJ, Hu S. Ef-fect of mitochondrial aldehyde dehydrogenase-2genotype on cardioprotection in patients withcongenital heart disease. Eur Heart J 2012;33:1606–14.

57. Jeewa A, Manickaraj AK, Mertens L, et al. Ge-netic determinants of right-ventricular remodelingafter tetralogy of Fallot repair. Pediatr Res 2012;72:407–13.

58. Carey AS, Liang L, Edwards J, et al. Effect ofcopy number variants on outcomes for infantswith single ventricle heart defects. Circ CardiovascGenet 2013;6:444–51.

59. Kim DS, Kim JH, Burt AA, et al. Burden ofpotentially pathologic copy number variants ishigher in children with isolated congenital heartdisease and significantly impairs covariate-adjusted transplant-free survival. J Thorac Car-diovasc Surg 2016;151:1147–51.e4.

60. Kim DS, Kim JH, Burt AA, et al. Patient ge-notypes impact survival after surgery for isolatedcongenital heart disease. Ann Thorac Surg 2014;98:104–10; discussion 110–1.

61. Yang Y, Muzny DM, Reid JG, et al. Clinicalwhole-exome sequencing for the diagnosis of

mendelian disorders. N Engl J Med 2013;369:1502–11.

62. Costa MW, Guo G, Wolstein O, et al. Functionalcharacterization of a novel mutation in NKX2-5associated with congenital heart disease andadult-onset cardiomyopathy. Circ CardiovascGenet 2013;6:238–47.

63. Zhang XL, Qiu XB, Yuan F, et al. TBX5 loss-of-function mutation contributes to familial dilatedcardiomyopathy. Biochem Biophys Res Commun2015;459:166–71.

64. Lander J, Ware SM. Copy number variation incongenital heart defects. Curr Genet Med Rep2014;2:168–78.

65. Egbe A, Uppu S, Lee S, Ho D, Srivastava S.Prevalence of associated extracardiac malforma-tions in the congenital heart disease population.Pediatr Cardiol 2014;35:1239–45.

66. Powis Z, Thrush D, Davis BT, Dolinsky JS.Diagnostic exome sequencing in pediatric patientswith congenital heart disease (abstr). J Am CollCardiol 2016;67:991.

67. Richards AA, Santos LJ, Nichols HA, et al.Cryptic chromosomal abnormalities identified inchildren with congenital heart disease. Pediatr Res2008;64:358–63.

68. San Agustin JT, Klena N, Granath K, et al.Genetic link between renal birth defects andcongenital heart disease. Nat Commun 2016;7:11103.

69. Miosge LA, Field MA, Sontani Y, et al.Comparison of predicted and actual consequences

of missense mutations. Proc Natl Acad Sci U S A2015;112:E5189–98.

70. Porrello ER. microRNAs in cardiac develop-ment and regeneration. Clin Sci (Lond.) 2013;125:151–66.

71. Shi H, O’Reilly VC, Moreau JL, et al. Gestationalstress induces the unfolded protein response,resulting in heart defects. Development 2016;143:2561–72.

72. Mital S, Musunuru K, Garg V, et al., AmericanHeart Association Council on Functional Genomicsand Translational Biology; Council on Cardiovas-cular Disease in the Young; Council on Cardiovas-cular and Stroke Nursing; Stroke Council; Councilon Lifestyle and Cardiometabolic Health; andCouncil on Quality of Care and OutcomesResearch. Enhancing literacy in cardiovasculargenetics: a scientific statement from the AmericanHeart Association. Circ Cardiovasc Genet 2016;9:448–67.

KEY WORDS chromosome aberrations,comparative genomic hybridization, genome-wide association study, high-throughputnucleotide sequencing, molecular diagnosis,patient care

Go to http://www.acc.org/jacc-journals-cme totake the CME quiz forthis article.