Several lines of evidence support a genetic cause for HLHS. First, there are numerous reports linking occurrence of HLHS to chromosomal abnormalities, e.g., Turner syndrome (monosomy X) and Jacobsen syndrome (chromosome 11q deletion) [20, 21]. Less frequently, trisomy 18, Smith-Lemli-Opitz syndrome, VACTERL association, CHARGE syndrome, Wolf-Hirschhorn syndrome, Rubinstein-Taybi syndrome, Noonan syndrome, and Holt-Oram syndrome have been reported [22–25]. However, these reports need to be analyzed cautiously as lack of phenotype detail and variation in CHD phenotype criteria blurs the genotype-phenotype correlations. For example, LV hypoplasia observed in Noonan syndrome or Holt-Oram syndrome may be associated with unbalanced atrioventricular septal defect rather than HLHS.

A second line of evidence supporting genetic origins of HLHS is that heritability (h², a statistical measure of genetic effect size) indicates that HLHS is determined largely by genetic factors. In a family-based study where all participants were screened for CHD by echocardiography, Hinton et al. [17] determined HLHS heritability was very high. In addition, the recurrence risks for HLHS and any CHD were 8 % and 22 %, respectively. Further, a tenfold increase in bicuspid aortic valve was noted. McBride et al. [19] also found high heritability using a family-based analysis of phenotypically related left-sided heart malformations.

Further evidence comes from use of linkage analysis to identify genomic regions that encode genes influencing the inheritance of HLHS. Hinton et al. [26] used nonparametric linkage analysis and identified two significant loci on chromosomes 10q22 and 6q23. These findings confirmed that nonsyndromic HLHS is genetically heterogeneous. Interestingly, ~21 % of kindreds contributed to linkage at each locus, suggesting these loci account for a substantial number of HLHS cases. Further, a suggestive HLHS locus on 11q22, previously identified in a case of HLHS with a balanced translocation t [10;11] (q24;q23), validated these analyses [27]. When the linkage approach was extended to a family-based cohort ascertained by either an HLHS or BAV proband, subset linkage analysis showed a significant improvement in the logarithm of odds (LOD) score at chromosome 14q23, providing evidence that some HLHS and BAV cases are genetically related [26]. In a family-based cohort ascertained by probands with left-sided malformations that included aortic valve stenosis, aortic coarctation, and HLHS, McBride et al. [28] found evidence for suggestive linkage to chromosomes 2p23, 10q21, and 16p12. They concluded that overlapping linkage peaks provide evidence for a common genetic etiology. Mitchell et al. [29] performed a genome-wide association study (GWAS) using a trio design where probands were ascertained by varied left-sided malformations including HLHS. Major findings included association at a chromosome 16 locus and suggestive association at loci on chromosome 3 and 10.

Several investigators have reported findings of genetic variants in single genes using mutation analysis of candidate genes in series of cases or small families. Genetic variants in connexin 43 (GJA1/Cx43) [30], NK2 homeobox 5 (NKX2-5) [31–33], Notch 1 (NOTCH1) [34, 35], and V-erb-B2 avian erythroblastic leukemia viral oncogene homolog 4 (ERBB4) [36] have been reported in HLHS. Ware et al. [37] identified patients with HLHS who had mutations in Zic family member 3 (ZIC3), a transcription factor associated with heterotaxy syndrome. More recently [38], patients with left-sided heart defects were found to be enriched for de novo variants in histone-modifying genes (H3K4me-H3K27me pathway); interestingly, patients with conotruncal defects and heterotaxy demonstrated similar enrichment. In addition to these reports of germinal mutations, there is a single report that identifies somatic mutations in heart and neural crest derivatives expressed 1 (HAND1) [39].

Genetic studies of HLHS patients have also evaluated copy number variations (CNVs), which are genomic regions of DNA gains or losses >1000 base pairs [40–45]. CNVs are detectable as a result of recent advances in molecular cytogenetics, particularly using microarray-based methods. Utilizing such methods to scan the genome, it has become evident that a significant proportion of the normal healthy human genome harbors CNV of unknown medical significance. However, other CNVs, usually large and often de novo, are considered pathogenic. Several investigators have examined the role of CNV in CHD, and 3 recent family-based studies examine the role of CNV in HLHS. Hitz et al. [40] sought to determine the impact of structural genomic variation in multiplex families ascertained by a proband with a left-sided heart malformation including some cases with HLHS. They searched for unique or rare copy number variations present only in affected members. Their findings indicate that unique CNVs contribute to at least 10 % of left-sided heart malformations cases. Carey et al. [41] evaluated patients with single ventricle, many of whom had HLHS. Putatively pathogenic CNVs had a prevalence of 13.9 %, which was significantly greater than the 4.4 % rate of such CNVs among controls. The authors concluded that pathogenic CNVs seem to contribute to the cause of single ventricle forms of CHD in ≥10 % of cases and are clinically subtle but adversely affect outcomes in children harboring them. Warburton et al. [42] compared CNV rates in two types of CHD, HLHS versus conotruncal defects, and found no significant difference. However, they found a substantially higher rate of de novo CNVs in probands with CHD than in control families (9 vs. 2 %). Among the de ovo or rare inherited CNVs, there were 12 CNVs that the authors considered likely to be causally related to CHD.

Taken together, these diverse studies provide strong evidence for genetic origins of HLHS and related cardiac phenotypes. Initially, analyses of pedigrees ascertained by an HLHS proband were interpreted as indicating simple Mendelian inheritance of HLHS [46]. However, review of literature indicates that despite considerable effort, identification of single genetic variants that “cause” HLHS has remained elusive. These results may indicate that HLHS inheritance is complex rather than simple, a finding concluded by several studies [26, 40]. The implication of this conclusion is that researchers must move beyond the expectation that a single disease-causing variant can be found. Although analytical methods to evaluate the effects of 2 or more genetic variants are well established, the ability to apply these methods in a high-throughput manner must be developed. Furthermore, researchers must realize that utilization of these more complex models requires careful consideration of study design and statistical power. Optimal approaches to discovering pathogenetic variants in complex diseases remain unclear [47].

What types of genes might be involved in the genetic underpinnings of HLHS? Concepts of cardiac development have greatly influenced our understanding of the formation of the mesoderm derived, 4-chambered vertebrate heart. Human genetic studies have identified mutations in genes important for early heart formation that cause CHD, supporting the idea that these birth defects are caused by alterations during cardiogenesis [48–52]. There has been considerable interest in CHD such as HLHS, in which individual chambers or valves of the developing heart are selectively impaired [48]. A widely accepted hypothesis is that HLHS develops as a result of embryonic alterations in blood flow, such as premature narrowing of the foramen ovale [1] or aortic valve obstruction [53]. In this light, it is noteworthy that valve malformation is a prominent part of the HLHS phenotype as evidenced by the frequent occurrence of left- and right-sided valve dysplasia in HLHS probands and the presence of BAV in family members [17]. The “no-flow, no-grow” hypothesis of heart maldevelopment also is supported by studies in the embryonic chick heart, which alter blood flow during cardiac development [54]. An alternative hypothesis for HLHS etiology focuses on the summation of separate genetic modules. Recent studies have investigated chamber-specific regulatory mechanisms, e.g., TBX5 (T-box 5) and IRX1 (iroquois homeobox 1), leading to formation of morphologically, functionally, and molecularly distinct cardiac chambers [48, 49]. In this context, it has been suggested that LV hypoplasia may result from a primary defect in myocardial growth during development. Unfortunately, there are presently no experimental models of HLHS to elucidate the relative contribution of these two hypotheses, but defining the genetic underpinnings of HLHS should enlighten the debate.