An Autopsy Case of Sudden Unexpected Death of a Young Adult in a Hot Bath: Molecular Analysis Using Next-Generation DNA Sequencing

Abstract

We report a case of sudden unexpected death of a young woman who was found in a bathtub of hot water. The autopsy concluded that all possible causes of sudden loss of consciousness, except cardiac origin, could be excluded. However, the heart did not show any obvious pathological changes. We used next-generation DNA sequencing (NGS) to examine 73 genes and detected 3 rare, potentially pathogenic variants with minor allele frequencies ⩽1.0%. The pathogenicity of these variants was evaluated using 8 in silico predictive algorithms, and SCN5A_p.Gly289Ser, CACNB2_p.Ser502Leu, and MYH11_p.Lys1573Glu were detected as possible pathogenic variants. Inherited heart disease is a likely cause of sudden unexpected deaths of young people in hot baths, even before the clinical manifestation of the disease. In the future, molecular analysis by NGS may help to predict young to early middle-aged people who could be at risk of sudden arrhythmogenic fatality in hot baths.

Introduction

Mortality rates due to accidental drowning are higher in Japan than in Western countries,¹ largely resulting from a higher incidence of sudden unexpected death (SUD) of Japanese people in hot baths.^2–4 Although the precise mortality rates are unknown, approximately 10% of SUDs confirmed by the Tokyo Medical Examiner’s Office occur at home, involving individuals who took deep hot baths.²

A few studies have examined the autopsies of victims of SUD associated with taking hot baths. Satoh et al⁴ summarized the findings of 268 autopsy cases of SUD in hot baths. Most of the subjects were above 70 years of age. Pathological and serological examination of the 173 subjects who did not show decomposition revealed a high incidence of structural cardiac disorders, such as coronary artery disease and cardiomegaly. Only 7 of the subjects were below 50 years of age, but all 7 had a history of epilepsy.⁴ Here, we present a rare autopsy case of SUD of a young subject found dead in a hot bath. We attempted genetic screening using next-generation DNA sequencing (NGS), which allows large numbers of samples to be sequenced simultaneously. It can be used for the comprehensive analysis of panels of 20 to 80 genes associated with inherited arrhythmia or cardiomyopathy⁵ to detect arrhythmogenic potential in the victims whose hearts have no significant structural disorders.

Case Report

A 28-year-old female beauty therapist was found dead in a bathtub with her face submerged. Resuscitation was not successful. There was no clinical history of significant organ or functional disease, such as epilepsy, that could have caused SUD or syncope in any of the cases. There was no family history of heart disease, and no electrocardiography had been performed within the past 10 years.

During medicolegal autopsy, no traumatic injury was found, but signs of drowning, specifically froth in the upper airway and pulmonary edema, were evident. Low levels of ethanol (1.1 mg/mL) were detected in the blood, but the full toxicological examination was negative. We concluded that all possible causes of sudden loss of consciousness, other than those of cardiac origin, were excluded by the full autopsy examination as well as the investigation of the scene of death.

The heart weighed 200 g and was examined as described in a previous report,⁶ but it did not show any significant pathological changes. Under microscopic examination, ischemic necrosis of myocytes, substantial coronary artery atherosclerosis with luminal narrowing greater than 50%, and myocardial disarray were not evident. Diffuse but very mild interstitial fibrosis of the left ventricle was found (Figure 1).

Figure 1.

Gross and microscopic appearance of the deceased victim’s heart (after fixation with formalin): (A) Horizontal section of both ventricles (scale bar = 1 cm) and (B) mild interstitial fibrosis of the left ventricle, visualized using Elastica-Masson staining (scale bar = 100 µm).

Molecular Testing

The ethical committee of Toyama University approved this study, which was performed in accordance with the ethical standards established in the 1964 Declaration of Helsinki.

The genetic analysis using NGS was conducted as described in a previous report.⁷ Genomic DNA samples of the case were extracted directly from whole blood using the QIAamp DNA Mini Kit (Qiagen Sciences Inc., Germantown, MD, USA). We designed a custom AmpliSeq panel using Ion AmpliSeq Designer software (http://www.ampliseq.com) to target all exons of 73 cardiac disorder–related genes associated with cardiomyopathy and channelopathy (Table 1). This custom panel, which consisted of 2 separate polymerase chain reaction (PCR) primer pools and produced a total of 1870 amplicons, was used to generate the target amplicon libraries. Genomic DNA samples were PCR-amplified using the designed custom panel and the Ion AmpliSeq Library Kit v2.0 (Life Technologies, Carlsbad, CA, USA). Prepared libraries were pooled in equimolar concentrations for multiplexing. Emulsion PCR and Ion Sphere Particle enrichment were conducted with the Ion PGM Template OT2 200 Kit (Life Technologies). Ion Sphere Particles were loaded on an Ion 314 Chip Kit v2 and sequenced using an Ion PGM Sequencing 200 Kit (Life Technologies).

Table 1.

List of the 73 analyzed genes associated with inherited cardiac diseases.

ABCC9, ACTC1, ACTN2, AKAP9, ANK2, BAG3, BMPR1A, CACNA1C, CACNB2, CALR3, CAPN3, CAV3, COL4A1, DES, DMD, DSC2, DSG2, DSP, ELN, EMD, GAA, GATA4, GLA, GPD1L, HCN4, JUP, KCNE1, KCNE2, KCNE3, KCNH2, KCNJ2, KCNQ1, KRAS, LAMP2, LDB3, LMNA, MYBPC3, MYH11, MYH6, MYH7, MYL2, MYL3, MYLK, MYOZ2, NKX2-5, NRAS, PKP2, PLN, PRKAG2, PTPN11, RAF1, RPS7, RYR2, SCN1B, SCN3B, SCN4B, SCN5A, SGCD, SLC25A4, SMAD3, SNTA1, SOS1, STARD3, TAZ, TBX5, TGFBR1, TGFBR2, TMEM43, TNNC1, TNNI3, TNNT2, TPM1, VCL

The Torrent Suite and Ion Reporter Software 5.0 (Life Technologies) were used to perform primary to tertiary analyses, including optimized signal processing, base calling, sequence alignment with the hg19 human genome reference (http://genome.ucsc.edu/), and variant analysis. For all variants detected, we consulted the East Asian (EAS) population database of 4327 individuals from the Exome Aggregation Consortium (http://exac.broadinstitute.org) to filter out those variants for which the minor allele frequency (MAF) was ⩾1.0% or undetermined in the EAS population.

For each genetic variation identified, we applied the Single Nucleotide Polymorphism Database (dbSNP) as a population database and the Human Gene Mutation database (HGMD) and ClinVar as reported disease-causing mutation databases. We also included 8 types of in silico predictive algorithms to evaluate the pathogenicity of identified variants. The URL for each database, in silico algorithms, and conditions used to evaluate pathogenicity are listed in Table 2.

Table 2.

Databases used for variant interpretation.

Name	Web site	Condition of pathogenicity
Population database
Single Nucleotide Polymorphism Database (dbSNP)	http://www.ncbi.nlm.nih.gov/SNP
Reported disease-causing mutation database
Human Gene Mutation database (HGMD)	http://www.hgmd.cf.ac.uk
ClinVar	http://www.ncbi.nlm.nih.gov/clinvar
In silico prediction algorithm
Functional Analysis Through Hidden Markov Models (FATHMM)	http://fathmm.biocompute.org.uk	Damaging
Mutation Assessor	http://mutationassessor.org	Medium, High
SIFT Sequence (SIFT)	http://sift.jcvi.org	Damaging
Align GVGD	http://agvgd.iarc.fr/index.php	⩾C15
MutationTaster	http://www.mutationtaster.org	Disease causing
PolyPhen-2	http://genetics.bwh.harvard.edu/pph2	Probably damaging, Possibly damaging
Protein Variation Effect Analyzer (PROVEAN)	http://provean.jcvi.org/index.php	Deleterious
Combined Annotation-Dependent Depletion (CADD)	http://cadd.gs.washington.edu	Score >10

Results of Molecular Testing

From the NGS analysis, SCN5A_p.Gly289Ser, CACNB2_p.Ser502Leu, and MYH11_p.Lys1573Glu were detected as rare variants in EAS, and the MAFs were 0%, 0.95%, and 0.035%, respectively. The sequences of SCN5A_p.Gly289Ser and CACNB2_p.Ser502Leu found in this case study are depicted in Figure 2.

Figure 2.

Sequences for possible channelopathy-related pathogenic variants from the deceased: (A) SCN5A, (B) CACNB2, and (C) MYH11.

SCN5A_p.Gly289Ser was previously reported as possibly pathogenic in an earlier study⁸ and was evaluated as “conflicting interpretations of pathogenicity” in ClinVar. The other 2 variants were evaluated as having “uncertain significance” in ClinVar and are not noted in HGMD. After using our in silico predictive algorithm analyses, SCN5A_p.Gly289Ser was evaluated as possibly pathogenic twice, CACNB2_p.Ser502Leu was evaluated as possibly pathogenic 5 times, and MYH11_p.Lys1573Glu was evaluated as possibly pathogenic 6 times (Table 3).

Table 3.

Detected variants and results of in silico analysis.

	SCN5A p.Gly289Ser	CACNB2 p.Ser502Leu	MYH11 P. Lys1573Glu
Transcript	NM_198056.2	NM_000724.3	NM_002474.2
MAF(%)	0.0	0.95	0.035
dbSNP	rs199473084	rs137886839	rs151101824
Disease-causing mutation database
ClinVar	Uncertain significance	Conflicting interpretations of pathogenicity	Uncertain significance
HGMD	Disease-causing mutation (long QT syndrome) CM097628	None	None
In silico prediction algorithm
FATHMM	Damaging	Damaging	Tolerated
Mutation Assessor	Neutral	Low	Medium
SIFT	Tolerated	Damaging	Damaging
Align GVGD	C0 (the lowest risk grade)	C0 (the lowest risk grade)	C0 (the lowest risk grade)
MutationTaster	Polymorphism	Disease causing	Disease causing
Polyphen-2	Benign	Possibly damaging	Probably damaging
PROVEAN	Neutral	Neutral	Deleterious
CADD	11.76	25.4	26.4

Abbreviations: CADD, combined annotation dependent depletion; FATHMM, Functional Analysis Through Hidden Markov Models; MAF, minor allele frequency; PROVEAN, Protein Variation Effect Analyzer; SIFT, SIFT Sequence.

Bold type shows the pathogenic condition in each in silico algorithm.

Discussion

Previous reports indicate a number of heart conditions that may cause SUD in young adults, including structural heart disease such as coronary anomaly, hypertrophic cardiomyopathy, and arrhythmogenic right ventricular cardiomyopathy.^9,10 No evidence of these conditions were found in this particular case. Our study supports the findings of others with evidence that concealed cardiomyopathy and channelopathy may also result in SUD of young adults.^7,11,12

Arrhythmogenic events and related SUD usually require both an abnormal myocardial substrate and an inciting trigger, such as exercise, being asleep, or emotional stress as seen in many cases.¹³ Nagasawa et al¹⁴ found that in the elderly, blood pressure and heart rate begin to rise immediately on immersion in a hot bath. These changes were associated with a temporary decrease in sympathetic activity without the compensatory parasympathetic suppression, resulting in hypotension and bradycardia. Chiba et al³ showed that increased peripheral blood pressure and cardiac output occurred after bathing in both young and elderly subjects. In addition, asymptomatic ventricular tachycardia occurred in elderly individuals while sitting in hot water, and this arrhythmia developed within 5 minutes of immersion.³ Many Japanese people love to take bath in water that reaches shoulder depth and soak in a sitting position, and a higher water temperature than in Western countries is generally used (approximately 40-42°C).⁴ These studies, along with the case described here, show that immersion in deep hot water can trigger an arrhythmogenic event which could lead to drowning in the bath water in individuals with an inherited or acquired heart disease.

We recently reported the significance of postmortem genetic analysis using NGS for both SUD syndrome⁷ and SUD with epilepsy cases¹⁵ in young people to explore the cause of such death if pathological change of the heart is not evident. The present case is notable in that detection of rare gene variants suggested that the deceased might have had undiagnosed arrhythmogenic potential which could cause sudden loss of consciousness during bathing.

However, we should note the limitations involved in interpretation of the detected variants found by NGS analysis.^16,17 We frequently depend on population databases and in silico analyses to evaluate the pathogenicity of detected variants because functional or genetic analysis of family members, the criterion standard method for evaluating the pathogenicity of genetic variants, can be difficult in some cases. However, the evaluations obtained from different in silico analyses do not always correspond, as shown in our case. Guidelines for interpreting sequence variants recommend that several in silico analyses should be used to evaluate the pathogenicity of arrhythmia-related gene variants because most algorithms used for missense variant prediction are only 65% to 80% accurate when examining known disease variants.¹⁸ In addition, Le Scouarnec et al¹⁹ and Kapplinger et al²⁰ indicated that identification of a variant does not confirm the presence of the disease because many of the variants found in Brugada syndrome patients were also identified in the control population. Genetic analysis using NGS may provide significantly useful information about the mechanism of SUD in some conditions, but careful and comprehensive evaluation of the detected variants is needed when the evaluation for pathogenicity differs across a range of predictive procedures.

SCN5A_p.Gln289Ser is a very rare variant, not only in EAS but across the world. Most of the in silico tools evaluated this variant as “negative.” However, 1 patient with long QT syndrome has been reported to have this variant.⁸ Therefore, we have evaluated this variant as possibly pathogenic in our case. In addition, although the relevance of drinking alcohol to SUD occurring in hot baths is not fully understood, some researchers consider that ethanol intake may increase the chance of developing atrial fibrillation, a prolonged QT interval, and SUD²¹ as the combination of pathogenic genetic variants and alcohol intake might increase the risk of sudden arrhythmogenic events occurring in hot baths.

The CACNB2_p.Ser502Leu variant is rare but has a relatively high incidence in EAS, yet 5 of 8 in silico tools evaluated this variant as pathogenic. The L-type calcium channel is composed of 4 subunits, and CACNB2 codes one of 3 ancillary subunits. CACNB2 is the dominant isoform known to play an essential role in the voltage dependence of the L-type calcium channel. Accelerated inactivation of the calcium current was found in 1 person who had a mutation in CACNB2, and the variant is also associated with Brugada syndrome, short QT, long QT 8 (Timothy syndrome), J wave syndrome, and sudden death.²² There are very few autopsy reports of SUD with CACNB2 variants detected. Our results show that the variants of CACNB2 may also have the potential to cause arrhythmogenic SUD in hot baths.

We should note that we cannot evaluate the pathogenic significance of the combined effect of the variants seen in present case. Currently available in silico tools can only indicate the pathogenicity of single genes; thus, the pathogenic significance of interactions between different gene variants cannot be fully evaluated. This victim’s heart might have concealed an arrhythmogenic potential from the combination of 2 channelopathy-related pathogenic variants, even if structural abnormality was not evident.

The role of the variant of MYH11 (which encodes a smooth muscle myosin heavy chain) is not well established. This gene belongs to the myosin heavy chain family and is a major contractile protein in smooth muscle cells.¹⁹ Whereas mutations in MYH11 have been identified in families with inherited patent ductus arteriosus and thoracic aortic aneurysms and dissections,^23,24 cases of arrhythmia or sudden death associated with this MYH11 variant have not been reported. Examination of further cases involving this variant will be useful to determine how significant this variant is.

Given the high cost of genetic analysis, it is not feasible to conduct this analysis routinely for every case. In many circumstances, careful toxicological screening and histological examination of the heart and other organs should provide enough evidence to specify cause of death or might at least contribute to narrowing the list of target genes to explore. In particular, detection of minimal cardiac pathology, including necrosis, inflammation, and fatty infiltration into the ventricle, during an examination might be indicative of cardiomyopathy-related genetic variants. Such findings may prompt further genetic analysis, even if the observed pathological changes do not fulfill the commonly used diagnostic criteria of structural heart diseases.^7,15

Conclusions

We report here a rare autopsy case of a young female adult who died suddenly and unexpectedly in her bathtub. Genetic analysis using NGS showed 2 previously unpredicted channelopathy-related variants with possible arrhythmogenic potential. A combination of the possible pathogenic channelopathy-related gene variants might have contributed to this unusual death in the bathtub, and the event may also have been triggered by bathing under the influence of alcohol. Although the evaluation of these detected variants is still complicated by our inability to completely assess pathogenicity, future molecular analysis by NGS may help to predict which young people could be at risk of SUD in hot baths.