Author + information
- Received November 8, 2017
- Revision received December 15, 2017
- Accepted January 8, 2018
- Published online March 12, 2018.
- David J. Tester, BSa,
- Leonie C.H. Wong, MBBChirb,c,
- Pritha Chanana, MSa,
- Amie Jaye, MScd,
- Jared M. Evans, MSa,
- David R. FitzPatrick, MDe,
- Margaret J. Evans, MBChBf,
- Peter Fleming, PhDg,
- Iona Jeffrey, MBChBh,i,
- Marta C. Cohen, MDj,k,
- Jacob Tfelt-Hansen, MD, DMScl,m,
- Michael A. Simpson, PhDd,
- Elijah R. Behr, MDb,c,∗ ( and )
- Michael J. Ackerman, MD, PhDa,∗∗ ()
- aDepartments of Cardiovascular Medicine (Division of Heart Rhythm Services), Pediatrics (Division of Pediatric Cardiology), and Molecular Pharmacology & Experimental Therapeutics (Windland Smith Rice Sudden Death Genomics Laboratory), Mayo Clinic, Rochester, Minnesota
- bMolecular and Clinical Sciences Research Institute, St. George’s, University of London, London, United Kingdom
- cCardiology Clinical Academic Group, St. George’s University Hospitals’ NHS Foundation Trust, London, United Kingdom
- dMedical and Molecular Genetics, Guy's Hospital, King’s College London, London, United Kingdom
- eMRC Human Genetics Unit, University of Edinburgh, Edinburgh, United Kingdom
- fRoyal Infirmary of Edinburgh, Edinburgh, United Kingdom
- gCentre for Child and Adolescent Health, Bristol Medical School, University of Bristol, Bristol, United Kingdom
- hDepartment of Cellular Pathology, St George's, University of London, London, United Kingdom
- iDepartment of Cellular Pathology', St. George's University Hospitals' NHS Foundation Trust, London, United Kingdom
- jHistopathology Department, Sheffield Children’s Hospital, Sheffield, United Kingdom
- kHonorary Senior Lecturer, University of Sheffield, Sheffield, United Kingdom
- lDepartment of Cardiology, The Heart Centre, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
- mDepartment of Forensic Medicine, Faculty of Medical Sciences, University of Copenhagen, Copenhagen, Denmark
Background Sudden infant death syndrome (SIDS) is a leading cause of postneonatal mortality. Genetic heart diseases (GHDs) underlie some cases of SIDS.
Objectives This study aimed to determine the spectrum and prevalence of GHD-associated mutations as a potential monogenic basis for SIDS.
Methods A cohort of 419 unrelated SIDS cases (257 male; average age 2.7 ± 1.9 months) underwent whole exome sequencing and a targeted analysis of 90 GHD-susceptibility genes. The yield of “potentially informative,” ultra-rare variants (minor allele frequency <0.00005) in GHD-associated genes was assessed.
Results Overall, 53 of 419 (12.6%) SIDS cases had ≥1 “potentially informative,” GHD-associated variant. The yield was 14.9% (21 of 141) for mixed-European ancestry cases and 11.5% (32 of 278) for European ancestry SIDS cases. Infants older than 4 months were more likely to host a “potentially informative” GHD-associated variant. There was significant overrepresentation of ultra-rare nonsynonymous variants in European SIDS cases (18 of 278 [6.5%]) versus European control subjects (30 of 973 [3.1%]; p = 0.013) when combining all 4 major cardiac channelopathy genes (KCNQ1, KCNH2, SCN5A, and RYR2). According to the American College of Medical Genetics guidelines, only 18 of 419 (4.3%) SIDS cases hosted a “pathogenic” or “likely pathogenic” variant.
Conclusions Less than 15% of more than 400 SIDS cases had a “potentially informative” variant in a GHD-susceptibility gene, predominantly in the 4- to 12-month age group. Only 4.3% of cases possessed immediately clinically actionable variants. Consistent with previous studies, ultra-rare, nonsynonymous variants within the major cardiac channelopathy-associated genes were overrepresented in SIDS cases in infants of European ethnicity. These findings have major implications for the investigation of SIDS cases and families.
Sudden infant death syndrome (SIDS) is the sudden unexpected death of an infant <1 year of age that remains unexplained despite comprehensive clinical and pathological investigations (1). SIDS represents 70% to 80% of all sudden unexpected infant deaths with an incidence of 0.4 in 1,000 live births in the United Kingdom and 0.5 in 1,000 live births in the United States (2,3). The peak incidence occurs between 2 and 4 months of age and is more common in boys. Such infant deaths are commonly associated with environmental risk factors such as co-sleeping or prone sleeping position (4). Despite successful targeted risk reduction campaigns, the number of SIDS cases has plateaued, and SIDS remains the leading cause of postneonatal death (4).
A triple-risk model for SIDS suggests the convergence of the vulnerable infant in the setting of exogenous stressors during a critical development period (5) (Central Illustration). Although many pathophysiological theories have been proposed, decisive pathogenic substrates or mechanisms triggering an infant’s sudden demise remain unclear (6–9). Several studies have implicated both common and rare genetic variants involved in autonomic function, neurotransmission, energy metabolism, response to infection, and cardiac repolarization (10–14). In addition, potentially lethal genetic heart diseases (GHDs) including long QT syndrome (LQTS), Brugada syndrome, catecholaminergic polymorphic ventricular tachycardia, and hypertrophic cardiomyopathy have been implicated as monogenic causes for a small proportion of SIDS cases (10,13,15–27).
However, fewer than 100 investigations of genetic variations in population-based SIDS cohorts have been published to date, largely on the basis of hypothesis-driven, candidate gene– or pathway-based approaches that recognize established pathobiological risk factors for SIDS, with an average cohort size of just 125 SIDS cases (13). In the present study, using whole exome sequencing (WES), we conducted a GHD-associated gene-specific analysis on a cohort of more than 400 unrelated SIDS cases.
The SIDS cohort (N = 427) consisted of 95 coroners’ cases from the United Kingdom (London, Sheffield, Edinburgh, and Bristol) and 332 coroner-, medical examiner–, or forensic pathologist–referred cases collected from 6 ethnically and geographically diverse U.S. population groups. Because of the lack of uniformity in procedures and reporting among medical examiner offices in the United States, minor differences in protocol may exist. Nonetheless, both gross and histological examinations of all major organs were performed, and all cases satisfied our enrollment criteria, which included the following: 1) sudden unexplained death of an infant <1 year of age; 2) European descent; and 3) a comprehensive negative medicolegal autopsy including a negative toxicology screen result and death scene investigation. Infants with asphyxia or a specific disease causing death were excluded. Ethnicity was self-reported by the referring coroner or medical examiner. This anonymous autopsy study had only limited medical information available such as the sex, ethnicity, age at the time of death, and sleep position. This study complies with the Declaration of Helsinki; locally appointed ethics committees including Mayo Clinic’s Institutional Review Board approved the research protocol. Some of the 332 samples from the United States were included in previous publications that involved hypothesis-driven, specific candidate gene mutational analysis (10,18,19,23–28). Of the 332 U.S. cases, 58 had been analyzed previously for variants in SCN5A (10), KCNQ1 (18), KCNH2 (18), RYR2 (19), SNTA1 (23), KCNJ8 (24), Cx43 (25), GPD1L (26), CAV3 (27), SCN1B (28), SCN2B (28), SCN3B (28), and SCN4B (28). An additional 25 of the 332 cases were also analyzed for RYR2 (19), and an additional 145 of the 332 cases were also analyzed for SNTA1 (23), KCNJ8 (24), Cx43 (25), GPD1L (26), CAV3 (27), SCN1B (28), SCN2B (28), SCN3B (28), and SCN4B (28). None of the 95 cases from the United Kingdom have been published previously.
A total of 973 control exomes (509 female, 464 male) from the ICR1000 U.K. exome series and the 1958 Birth Cohort study were included for case-control analysis (29). As previously reported, exome sequencing was performed using the Illumina TruSeq and Illumina instruments (Illumina, San Diego, California) (29).
Whole exome sequencing
Genomic DNA isolated from each SIDS case underwent WES at the KCL-GSTT Biomedical Research Centre Genomics Platform in London, United Kingdom or Mayo Clinic’s Medical Genome Facility in Rochester, Minnesota.
Paired-end libraries were prepared following the manufacturer’s protocol (Agilent Technologies, Inc., Santa Clara, California) using the Bravo Automated Liquid Handling Platform from Agilent. Briefly, 1 to 3 μg of genomic DNA was fragmented to 150 to 200 bp using the Covaris E210 sonicator (Covaris, Inc., Woburn, Massachusetts). The ends were repaired, and an “A” base was added to the 3′ ends. Paired-end index DNA adaptors (Agilent) with a single “T” base overhang at the 3′ end were ligated, and the resulting constructs were purified using AMPure SPRI (solid-phase reversible immobilization) beads (Agencourt Bioscience Corporation, Beverly, Massachusetts). The adapter-modified DNA fragments were enriched by 4 cycles of polymerase chain reaction using SureSelect forward and SureSelect ILM Pre-Capture Indexing reverse (Agilent) primers. The concentration and size distribution of the libraries was determined on an Agilent Bioanalyzer DNA 1000 chip.
Whole exon capture was performed using the protocol for Agilent’s Sure SelectXT Human All Exon V5+UTR kit. Briefly, 750 ng of the prepared library was incubated with whole exon biotinylated RNA capture baits supplied in the kit for 24 h at 65°C. The captured DNA:RNA hybrids were recovered using Dynabeads MyOne Streptavidin T1 (Invitrogen Dynal, Thermo Fisher Scientific, Waltham, Massachusetts). The DNA was eluted from the beads and purified using Ampure XP beads (Agencourt). The purified capture products were then amplified using the SureSelect Post-Capture Indexing forward and index polymerase chain reaction reverse primers (Agilent) for 12 cycles.
Libraries were pooled at equimolar concentrations and loaded onto paired-end flow cells at concentrations of 7 to 8 pM to generate cluster densities of 600,000 to 800,000/mm2 following Illumina’s standard protocol using the Illumina cBot and HiSeq paired-end cluster kit version 3. Each lane of a HiSeq flow cell produced 21 to 39 Gb of sequence. The level of sample pooling was controlled by the size of the capture region and the desired depth of coverage.
The flow cells were sequenced as 101 × 2 paired-end reads on an Illumina HiSeq 2000 using TruSeq SBS sequencing kit version 3 and HiSeq data collection version 220.127.116.11 software. Base-calling was performed using Illumina’s RTA version 18.104.22.168.
The FASTQ files underwent quality control checks using FASTQC (Babraham Bioinformatics, Babraham Institute, Cambridge, United Kingdom). The Illumina paired-end reads were aligned to the GRCh37 (hg19) human reference genome using Novoalign (Novocraft Technologies, Selandor, Malaysia). Single-sample variant calling with the Genome Analysis Toolkit (GATK version 3.2-2, Broad Institute, Cambridge, Massachusetts) (30), and the resulting gVCFs (genomic variant call format) subsequently underwent multisample genotyping and variant quality score recalibration. Genotypes were excluded if the quality control was <15 or there were fewer than 4 reads supporting the call. Further filtering of variant sites was performed to exclude sites with missingness >0.1 in cases or control subjects. Variants were annotated with respect to the genes in which they reside with Annovar, allele frequencies were obtained from the Exome Aggregation Consortium database, and Combined Annotation Dependent Depletion (CADD) scores were derived from the CADD server (31).
Quality control coverage analysis and principal component analysis for relatedness and ethnicity
Coverage across the exome was assessed using the bedtools software package (Quinlan Laboratory, University of Utah, Salt Lake City, Utah), and cases were excluded from further analysis if <75% of the GENCODE (GENCODE Project, an international collaboration)–defined protein coding exome was covered by <20 reads. A set of 3,847 common variants located outside of regions of the genome where there is extensive linkage disequilibrium was used to estimate relatedness within the study cohort and ethnic ancestry alongside the control group (32). Estimation was undertaken using the first 2 dimensions of a multidimensional scaling using Euclidean distance undertaken with the King software package.
To avoid potential confounding secondary to population stratification resulting from genetic admixture, a principal component analysis (PCA) was performed. The PCA served only for the rare variant analysis between European SIDS cases and European control subjects. The PCA data were not used for attributing causality to identified variants where ethnically matched control subjects would not be necessary for variant adjudication. SIDS cases and control subjects forming a homogeneous cluster on the first 2 components were included in the case-control rare variant analysis.
Genetic heart disease gene-specific variant analysis
Genes known to be associated with cardiac channelopathy (LQTS, catecholaminergic polymorphic ventricular tachycardia, Brugada syndrome) susceptibility and cardiomyopathy (hypertrophic cardiomyopathy, dilated cardiomyopathy, arrhythmogenic cardiomyopathy) susceptibility (N = 90) (Online Table 1) were evaluated for the presence of “ultra-rare” nonsynonymous variants (NSVs) with a minor allele frequency (MAF) <0.00005 (1 in 20,000 alleles) derived from the Genome Aggregation Database (33). A comparison of yield was undertaken for ultra-rare NSVs in SIDS cases of PCA-determined European ancestry versus European control subjects across all 90 GHD-susceptibility genes and the 4 “major” channelopathy genes (KCNQ1, KCNH2, SCN5A, and RYR2).
All putative loss of function variants (i.e., a “radical” variant: frameshift, nonsense, and essential splice-site variants) or missense variants with a previously established abnormal in vitro function characterization that resided within any of the 90 GHD-associated genes and all ultra-rare, missense variants residing in any of the 4 “major” channelopathy genes were considered to be “potentially informative” variants that would be appropriate for investigation of their significance in a family. Such variants were confirmed using standard Sanger sequencing techniques. The American College of Medical Genetics and Genomics standards and guidelines for the interpretation of sequence variants were used to assist in the classification of our genetic findings further among all ultra-rare (MAF <0.00005) NSVs identified across the 90 GHD-associated genes (34).
Categorical variables were expressed as absolute numbers and percentages and were compared with the Fisher exact or chi-square tests. Probability values were determined on the basis of 2-sided tests considered significant at p < 0.05. Analysis was conducted with SPSS version 18.0 software (SPSS, Chicago, Illinois).
WES was performed in 427 SIDS cases. However, quality control metrics excluded 7 cases because of insufficient exome coverage and 1 individual from a half-sibling pair (Online Figure 1). The cohort therefore consisted of 419 cases (257 male, 162 female; average age 2.7 ± 1.9 months) with a skewed bell-shaped distribution of age (Online Figure 2). The epidemiologically higher-risk age group of 2 to 4 months (58.9%) and male sex (61%) accounted for the majority of the cases. The PCA demonstrated a wide distribution of ancestral origins with 278 cases (173 male, 105 female) considered to be of European ancestry and 141 cases (84 male, 57 female) considered to be of mixed-European ancestry (Online Figure 3, Table 1). Sleep characteristics were known in 54% of the cohort (Table 1). There were no significant differences in demographics and sleep characteristics between the European and mixed-European ancestry cases.
Genetic heart disease gene-specific analysis
Overall, a total of 285 unique, ultra-rare NSVs (256 missense, 23 putative loss of function [12 frameshift, 8 splice errors, 3 nonsense], and 6 in-frame indels) were identified in 194 of 419 (46.3%) SIDS cases overall (Figure 1). Further, 45 of 278 (16.2%) European ancestry cases and 25 of 141 (17.7%) mixed-European ancestry cases hosted >1, ultra-rare NSVs.
These ultra-rare NSVs resided in 68 of the 90 GHD-associated genes (21 of 31 channelopathy-associated genes and 47 of 59 cardiomyopathy-associated genes). The gene-specific yields for the overall cohort and the European and mixed European subsets are shown in Online Table 2.
Of the 285 unique, ultra-rare NSVs identified, 57 (20%) were considered “potentially informative.” A total of 25 of 57 (43.9%) were missense variants in the 4 major channelopathy genes, 23 of 57 (40.3%) were putative loss of function variants, and 10 of 57 (17.5%) were variants previously reported in published papers as having an abnormal in vitro functional phenotype (Table 2). Overall, 53 of 419 (12.6%) SIDS cases hosted at least 1 “potentially informative” variant (Figure 1). Four of 419 cases (0.95%) had 2 “potentially informative” variants. The yield was 14.9% (21 of 141) for mixed-European ancestry cases and 11.5% (32 of 278) for European ancestry SIDS cases (Figure 1).
There were no significant differences in the yield of either ultra-rare NSVs among all 90 GHD genes or “potentially informative” variants when comparing sex, sleep position (supine vs. prone), or co-sleeping (yes vs. no) in either the overall or stratified study groups (Table 3). However, there was a significantly higher yield of “potentially informative” GHD-associated genetic variants in those infants who died at >4 months of age (15 of 65 [23.1%]) compared with those younger than 4 months of age (37 of 354 [10.4%]; p = 0.0075) (Figure 2).
Following further vetting using the strict American College of Medical Genetics and Genomics guidelines, only 18 of the 285 ultra-rare NSVs achieved a “pathogenic” or “likely pathogenic” designation and were identified in 18 (4.3%) of the 419 SIDS cases (Table 2, Figure 1). There was no difference in yield of “pathogenic” or “likely pathogenic” variants between the European (12 of 278 [4.3%]) and the mixed-European cohorts (6 of 141 [4.3%]).
Consistent with previous studies, there was significant overrepresentation of ultra-rare NSVs in European SIDS cases (18 of 278 [6.5%]) versus European control subjects (30 of 973 [3.1%]; p = 0.013) when combining all 4 major cardiac channelopathy genes (KCNQ1, KCNH2, SCN5A, and RYR2). (Figure 3). However, there was no significant difference in yield between cases and control subjects for any specific gene.
This paper reports results derived from a whole exome molecular autopsy with GHD gene-specific analysis of a large cohort of unrelated SIDS cases. Previous postmortem genetic studies implicated pathogenic mutations in cardiac channelopathy-associated genes as a monogenic cause for up to 15% of SIDS (13,16–19,23). Furthermore, rare hypertrophic cardiomyopathy–associated sarcomere gene variants were implicated recently in 3.5% of SIDS cases (21). However, on the basis of their prevalence in the Genome Aggregation Database (gnomAD), only 1.4% of these SIDS cases hosted variants rare enough to be considered pathogenic.
Because of the rarity of SIDS and any likely causative disorders, we used a strict minor allele frequency cutoff equivalent to a heterozygous frequency of <1 in 10,000 individuals in gnomAD. Despite using this very conservative rarity filter, 46% of our SIDS cohort harbored novel or ultra-rare, protein-altering genetic variants within 68 of the 90 GHD-susceptibility genes. Unfortunately, despite their rarity, most of these ultra-rare variants still remain variants of uncertain significance stuck in genetic purgatory (35,36). In fact, about 25% of these 90 GHD-associated genes have a negative z-score, suggesting that they tolerate variation (33,37).
Because of ambiguity surrounding the pathogenic nature of many GHD genes that play a “minor” role in their respective diseases, we examined the yield of ultra-rare NSVs with the highest likelihood of being true pathogenic mutations. Overall, about 13% of our SIDS cases hosted “potentially informative” variants regarded as having the greatest probability of being causative for the infant’s sudden death and having potential clinical utility for assessing a family for genetic risk. Unfortunately, the anonymous nature of the cohort prevents us from verifying the presence of these variants among family members for the purpose of potential genotype-phenotype co-segregation analysis or to determine the frequency of de novo status of the variants of interest. Importantly, not all of these variants have been characterized functionally, and great caution must still be exercised, even when interpreting ultra-rare variants residing within the major channelopathy genes.
Recently, Hertz et al. (22) reported a 34% yield of “variants with likely functional effects” following a genetic analysis of GHD-associated genes in only 47 sudden unexpected deaths in infancy cases. However, given the rarity of GHDs in the general population, we believe that their definition of rarity (MAF <1%) was unacceptably and erroneously high, thus causing an overestimated burden of potentially pathogenic variants in their SIDS cohort. In fact, of their 16 “pathogenic” variants, only 1 novel RYR2 variant would have been deemed “potentially informative” by our robust criteria.
In 2017, Neubauer et al. (38) reported a yield of “potentially causative” variants in 20% of their 155 European SIDS cases following WES and genetic interrogation of their 192-gene focused panel that comprised both cardiovascular-associated and metabolic disorder–associated genes. The majority of their seemingly genotype-positive infants had a variant with “likely functional effects” in genes associated with a cardiac channelopathy (9%) or cardiomyopathy (7%). However, most of these variants represent missense variants within “minor” genes (38). In fact, only 2.6% of their cases hosted what we would consider a “potentially informative” variant by our strict definition.
Although our study supports the utility of WES to identify potential sudden death-causing variants within established or potential sudden death–susceptibility genes, the challenge of the WES-based molecular autopsy does not lie in the identification of variants, but rather in the adjudication of their potential pathogenicity. Accurate variant classification is crucial to enable proper counseling of surviving family members. Erroneously or prematurely adjudicating ambiguous variants as pathogenic has the potential to harm patients and their families. Tragically, this became a reality for 1 family described by Ackerman et al. (36) recently, as they dealt with the disastrous consequences of unnecessary treatment on the basis of an erroneously interpreted variant in KCNQ1. Thus, overattribution of SIDS deaths to GHDs has significant implications for the immediate family, and we urge extreme caution in variant interpretation. When such cautionary advice was heeded, <5% of more than 400 SIDS cases had either a “pathogenic” or “likely pathogenic” variant in 1 of 90 GHD-susceptibility genes, a percentage that is substantially lower than in previous extrapolations of the prevalence of either channelopathic or cardiomyopathic SIDS. This parallels our experience of the “molecular autopsy” in unexplained sudden death where stringent variant evaluation results in a significant reduction of numbers of “likely pathogenic” and “pathogenic” variants of clinical utility (39).
Using a similarly stringent variant analysis, we observed previously a 13% (40 of 302) yield of ultra-rare “pathogenic” or “likely pathogenic” variants within sudden death–susceptibility genes among 302 autopsy-negative cases of sudden arrhythmic death syndrome in persons who died at an age >1 year (median age 24 years), compared with a significantly (p = 0.00002) lower yield of 4.3% (18 of 419) in our SIDS cohort (39). These data suggest that most SIDS cases stem from pathobiological bases that are largely different genetically and mechanistically from sudden death occurring after the age of 1 year.
Several risk factors for SIDS have been established. One could hypothesize that vulnerable infants dying of SIDS without the presence of additional risk factors are more likely to host a highly penetrant monogenic cause for their death compared with infants exposed to additional environmental risk factors. Yet no significant differences in the yield of “potentially informative” GHD gene variants associated with sex, sleep position, or bed sharing were observed. However, a significant age effect on the yield, where 23% of those infants older than 4 months of age hosted a “potentially informative” GHD-associated variant compared with only 10% of the infants younger than 4 months of age, was observed. These data support the potential stratification of those SIDS cases that may benefit most from postmortem genetic testing of the major channelopathy- or cardiomyopathy-associated genes.
The significant overrepresentation of ultra-rare NSVs within the 4 major channelopathy genes associated with either inheritable LQTS (KCNQ1, KCNH2, SCN5A) or catecholaminergic polymorphic ventricular tachycardia (RYR2) observed in our European Caucasian SIDS cases compared with ethnically matched control subjects (6.5% vs. 3.1%; p = 0.013) supports that cardiac channelopathies may represent the underlying pathogenic basis for some SIDS cases and that post-mortem genetic testing of the 4 major channelopathy-associated genes may be warranted in cases of SIDS.
In our study, we used a strict MAF cutoff of 0.005% (i.e., 1 in 20,000 alleles or 1 in 10,000 individuals). Although using a stringent threshold could reduce the possibility of identifying variants that would be deemed too common in the population to cause a rare disease such as LQTS, it could also prevent the identification of potentially important functionally significant variants that could play a role in SIDS pathogenesis. For example, the p.R176W-KCNH2 variant was identified in a single European SIDS case in our cohort. This variant has been associated with LQTS previously, it has been demonstrated to have a functional effect by in vitro assays, and it has been considered a founder LQT2 mutation in the Finnish population (40,41). Although this variant meets the current American College of Medical Genetics and Genomics guideline classification of “pathogenic” variant, its heterozygous frequency in gnomAD (44 of 53,551 overall, 31 of 22,031 Europeans, and 10 of 4,022 Finnish European individuals) exceeds our stringent cutoff and was therefore not included in our analysis.
A whole exome molecular autopsy followed by a cardiac gene–specific focus reveals that <15% of more than 400 SIDS cases had a “potentially informative” variant in 1 of 90 GHD-susceptibility genes. Furthermore, <5% of these infants who died possessed variants that are immediately useful in a family for further cascade testing. This finding represents a substantial reduction of the perceived importance of monogenic cardiac genetic disease in SIDS compared with previous studies. Interpretation of GHD-associated rare variants must therefore be stringent and careful given the implications of misattribution in families. This has important clinical implications for the community managing SIDS cases and their relatives.
COMPETENCY IN MEDICAL KNOWLEDGE: Postmortem genetic testing using WES can identify rare NSVs in heart disease susceptibility genes that may underlie some cases of SIDS, predominantly in the 4- to 12-month age group. The yield of these variants involving all but the principal channelopathy genes is similar to that in a healthy group, thus suggesting that single-gene disorders are not the major cause of SIDS. Clinically relevant variants useful for predictive testing are identified in about 4% of infant deaths.
TRANSLATIONAL OUTLOOK: Because the pathogenesis of SIDS is complex and likely involves more than 1 genetic pathway, future research should entail functional studies to identify truly pathogenic variants and assess the degree to which each variant contributes to these cases of unexpected early death.
The authors gratefully acknowledge both the medical examiners and coroners for referring the sudden death victims to our program in an effort to find an explanation for the infant’s sudden death.
This work was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health (grant no. R01HD042569 to Dr. Ackerman) and by the British Heart Foundation (BHF Clinical Research Training Fellowship FS/13/78/30520 to Drs. Wong and Behr). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Mr. Tester is supported by the Mayo Clinic Windland Smith Rice Comprehensive Sudden Cardiac Death Program. Dr. Wong is supported by additional funds from Biotronik and Cardiac Risk in the Young. Dr. Behr is supported by the Higher Education Funding Council for England; is a consultant for Medtronic; has received research funding from Biotronik; and has received funds from The Robert Lancaster Memorial Fund sponsored by McColl's RG Ltd. Dr. Ackerman is a consultant for Audentes Therapeutics, Boston Scientific, Gilead Sciences, Invitae, Medtronic, MyoKardia, and St. Jude Medical; is supported by the Mayo Clinic Windland Smith Rice Comprehensive Sudden Cardiac Death Program; and the Mayo Clinic have an equity- or royalty-based relationship with AliveCor, Blue Ox Health, and StemoniX, although none of these entities were involved in this study in any way. Dr. Simpson has a part-time contract of service with Genomics. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Mr. Tester and Dr. Wong contributed equally to this work and are joint first authors. Drs. Simpson, Behr, and Ackerman contributed equally to this work and are joint senior authors.
- Abbreviations and Acronyms
- genetic heart disease
- Genome Aggregation Database
- long QT syndrome
- minor allele frequency
- nonsynonymous variant
- principal component analysis
- sudden infant death syndrome
- whole exome sequencing
- Received November 8, 2017.
- Revision received December 15, 2017.
- Accepted January 8, 2018.
- 2018 American College of Cardiology Foundation
- Krous H.F.,
- Beckwith J.B.,
- Byard R.W.,
- et al.
- Matthews T.,
- MacDorman M.F.
- ↵Unexplained deaths in infancy: England and Wales: 2009. Office for National Statistics 2011. Available at: http://www.ons.gov.uk/ons/dcp171778_227450.pdf. Accessed January 24, 2018.
- Guntheroth W.G.
- Weese-Mayer D.E.,
- Ackerman M.J.,
- Marazita M.L.,
- Berry-Kravis E.M.
- Salomonis N.
- Glengarry J.M.,
- Crawford J.,
- Morrow P.L.,
- Stables S.R.,
- Love D.R.,
- Skinner J.R.
- Arnestad M.,
- Crotti L.,
- Rognum T.O.,
- et al.
- Hertz C.L.,
- Christiansen S.L.,
- Larsen M.K.,
- et al.
- Cheng J.,
- Van Norstrand D.W.,
- Medeiros-Domingo A.,
- et al.
- Tester D.J.,
- Tan B.-H.,
- Medeiros-Domingo A.,
- Song C.,
- Makielski J.C.,
- Ackerman M.J.
- Van Norstrand D.W.,
- Asimaki A.,
- Rubinos C.,
- et al.
- Van Norstrand D.W.,
- Valdivia C.R.,
- Tester D.J.,
- et al.
- Ruark E.,
- Münz M.,
- Renwick A.,
- et al.
- McKenna A.,
- Hanna M.,
- Banks E.,
- et al.
- ↵Combined Annotation Dependent Depletion (CADD) scores. Available at: http://cadd.gs.washington.edu. Accessed January 24, 2018.
- ↵Exome Aggregation Consortium (ExAC), Cambridge, MA. Available at: http://exac.broadinstitute.org. Accessed December 30, 2017.
- Neubauer J.,
- Lecca M.R.,
- Russo G.,
- et al.
- Lahrouchi N.,
- Raju H.,
- Lodder E.M.,
- et al.
- Lahti A.L.,
- Kujala V.J.,
- Chapman H.,
- et al.
- Crotti L.,
- Lundquist A.L.,
- Insolia R.,
- et al.
- Haghighi K.,
- Kolokathis F.,
- Gramolini A.O.,
- et al.
- Medeiros-Domingo A.,
- Bhuiyan Z.A.,
- Tester D.J.,
- et al.