Clinical manifestations, genetic profiles, and sudden cardiac arrest in pediatric hypertrophic cardiomyopathy: Challenges of risk prediction for initial sudden cardiac arrest presentations

Abstract

Background

Sudden cardiac arrest (SCA) is a leading cause of death in pediatric hypertrophic cardiomyopathy (HCM).

Objective

The study sought to analyze the clinical and genetic characteristics of pediatric HCM and assess the applicability of current SCA risk prediction models.

Methods

We enrolled individuals diagnosed as HCM before 20 years of age, between 2000 and 2020, excluding those secondary to hemodynamic causes and those associated with genetic syndromes other than RASopathies.

Results

Among 91 patients (31 female, 60 male), SCA occurred in 13 (14.3%) patients, with 6 (46%) cases presenting as the initial symptom. These 6 patients were older and had lower left ventricular mass z scores compared with those who experienced SCA later during follow-up. Whole exome sequencing in 55 patients identified genetic pathogenic variants in 80% of cases. The most prevalent pathogenic variants were MYH7 (40%) and MYBPC3 (24%) within the sarcomere gene group, and RAF1 (36.8%) and PTPN11 (21.1%) among RASopathies. SCA events occurred mostly between 10 and 18 years of age. The SCA event-free survival rate was 84.2% by 10 years after diagnosis and associated with sarcomere gene pathogenic variants (odds ratio 10.2). Excluding the 6 patients presented as SCA initially, both the HCM Risk-Kids and PRIMaCY (precision medicine in cardiomyopathy) genetic scoring system exhibited strong predictive power for SCA during follow-up.

Conclusion

In pediatric HCM, SCA is notably associated with sarcomere gene pathogenic variants. While newer risk scoring systems, if incorporated with genetic information, effectively predict SCA in this Asia cohort, a challenge remains: nearly half of SCA cases present as the initial clinical manifestation.

Key Findings.

• ▪RASopathy-related hypertrophic cardiomyopathy (HCM) was more frequently associated with heart failure symptoms and related mortality, whereas sarcomere gene–related HCM more often manifested as sudden cardiac arrest (SCA) or syncope.
• ▪SCA occurred in 14.3% of pediatric HCM cases, with 46% presenting as the initial symptom.
• ▪Both the HCM Risk-Kids and PRIMaCY (precision medicine in cardiomyopathy) genetic models showed strong discriminatory ability in predicting SCA during follow-up.
• ▪A considerable proportion of SCA cases present initially, making prevention challenging; thus, familial cascade or school-based electrocardiography screening may be crucial for identifying high-risk patients.

Introduction

Hypertrophic cardiomyopathy (HCM) stands as the second most prevalent form of pediatric cardiomyopathies, typically characterized by ventricular hypertrophy, predominantly in the left ventricle (LV). The incidence rate is approximately 0.47 per 100,000 children. Clinical manifestations often include exercise intolerance, chest pain, syncope, and in severe cases, sudden cardiac arrest (SCA).¹ Genetic studies have advanced our understanding of the pediatric cardiomyopathy landscape, revealing intrinsic genetic pathogenic variants as the primary cause, particularly involving sarcomere genes, syndromic, or metabolic storage diseases.2, 3, 4

Patients with HCM face a heightened risk of SCA, particularly during physical exertion.⁵ Large cohort studies report sudden cardiac death (SCD) incidence rates ranging from 0.8% to 2% annually.⁶ In the pediatric population, this rate is notably higher, and children also exhibit a 36% greater likelihood of experiencing such events compared with adults.⁶ Defining high-risk patients is crucial, but traditional risk stratification algorithms are extrapolated from adult practice, considering factors such as extreme LV hypertrophy (LVH), unexplained syncope, nonsustained ventricular tachycardia (NSVT), LV outflow tract (LVOT) obstruction, left atrial (LA) diameter, LV dysfunction, age, and family history of SCD. The American Heart Association (AHA) recommends individuals with ≥1 risk factor as high risk.⁷

Recently, 2 pediatric specific risk score systems, HCM Risk-Kids (https://hcmriskkids.org/) and PRIMaCY (precision medicine in cardiomyopathy) (https://primacycalculator.com/), have been developed. HCM Risk-Kids employs 5 predictors, including maximal LV wall thickness z score, LA z score, maximal LVOT gradient, NSVT, and unexplained syncope, to calculate the risk of SCA.⁸ The European Society of Cardiology recommends using this risk score system (HCM Risk-Kids) for the indication of implantable cardioverter-defibrillator implantation.⁹ PRIMaCY’s risk score is similar but includes age as an independent predictor and incorporates genetic test results if available.¹⁰ To differentiate between the approaches in our study, if genetic information is not included in the PRIMaCY model, we refer to it as the PRIMaCY clinical model, whereas incorporating genetic data results, we refer to it as the PRIMaCY genetic model.

We hypothesized that because of the potential racial differences, risk score systems for SCA, if incorporated genetic information, may be more applicable to Asia HCM cohort. We analyzed the clinical characteristics and genetic profile in pediatric patients diagnosed with HCM in a tertiary care center in Asia. We also assessed the applicability of the current risk score systems for SCA outlined in the AHA guideline, as well as HCM Risk-Kids and PRIMaCY risk score systems.

Methods

Patients

We identified all cases diagnosed with HCM in individuals diagnosed under 20 years of age, between 2000 and 2020, at National Taiwan University Children Hospital and National Taiwan University Hospital. The research reported in this paper adhered to the 1975 Declaration of Helsinki, as reflected in a priori approval by the Human Research Ethics Committee of our hospital (approval number 201912075RINA). We retrospectively collected demographic data, medical records, and previous genetic study results, with informed consent waived for this retrospective portion. Cardiac dimensions, wall thickness, LV mass, and ejection fraction were assessed using M-mode echocardiography. The Boston Children Hospital z score calculator (https://zscore.chboston.org/) were utilized for z score calculation. The diagnostic criteria for HCM included septal or LV free wall hypertrophy, defined as a z score >2 for patients younger than 18 years of age and a wall thickness of ≥13 to 15 mm for those 18 to 20 years of age, depending on family history, in the absence of secondary causes. Patients with hypertension-related HCM, associated congenital heart disease, storage diseases, or syndromic diseases with multiple anomalies (except RASopathy) were excluded due to differing treatment strategies and clinical courses. Significant LVOT obstruction gradient was defined as a resting pressure gradient over the outflow tract >30 mm Hg. Diagnosis of associated Noonan syndrome was confirmed through genetic study.

Mutational screening and analysis

Patients were invited to undergo genetic testing after providing informed consent. Genomic DNA was extracted from venous blood samples using standard procedures. A comprehensive genetic panel, encompassing over 400 genes associated with hereditary arrhythmias and cardiomyopathies (including RASopathies), was employed for the genetic analysis (as our previous report).¹¹ We included genes recommended by Heart Rhythm Society/European Heart Rhythm Association/Asia Pacific Heart Rhythm Society expert consensus,¹² ClinVar, and the Human Gene Mutation Database (professional version). Next-generation sequencing with MiSeq (Illumina) were performed for detection. The procedures included capture-based exomic sequencing, variant calling, and annotation and were performed as we described previously.¹¹ For candidate variants with allele frequency <0.001 according to the Taiwan Biobank, we followed the definition of the American College of Medical Genetics and Genomics and considered it a pathogenic variant if it was either a null variant (nonsense or frameshift) or classified as pathogenic in ClinVar or Human Gene Mutation Database, and consider it as a likely pathogenic variant if it did not meet the previous criteria but was predicted as deleterious by 6 in silico analyses, along with a high rejected substitutions score in the GERP++, as in our previous work.¹³ Finally, Sanger sequencing was used to confirm the nucleotide changes in the identified variants.

Statistical analysis

Data are expressed as mean ± SD. We compared the basic demographic data, genetic results, clinical symptoms, electrocardiography (ECG) and echocardiographic parameters, and N-terminal pro–brain natriuretic peptide (NT-proBNP) values among patients with aborted SCA, patients who reached the primary endpoint (death or heart transplantation), and those without events. The nonparametric Mann-Whitney U test was used for the numerical data analysis, and the chi-square or Fisher exact test was used for the categorical data analysis. The Kaplan-Meier survival curve was used for the survival analysis. Cox regression including parameters of sex, initial diagnostic age, genetic variants, functional class, LV ejection fraction, LV mass z score and interventricular septum (IVS)/LV posterior wall ratio, initial ST depression at lateral leads, LVOT obstruction, and initial QTc interval, was used for risk factors analysis of SCA and heart transplantation–free survival. A P value <.05 was considered significant. The C-index was used to measure how well the model discriminated between patients with high and low risk of SCA. A value of 0.5 for the C-index indicates no discrimination, and a value equal to 1 indicates perfect discrimination.

Results

Basic demographics

We enrolled 91 (31 female, 60 male) patients, with a male-to-female ratio of 1.94 and a mean age at diagnosis of 8.1 ± 7.1 years (Table 1). The most common initial presentation was identified due to heart murmur evaluations or screenings (3 through school ECG surveys and 2 via family cascade screening) in 52 (57.8%) patients. Aborted SCA was the initial presentation in 6 (6.7%) patients. A positive family history was noted in 21 (23.1%) patients. The majority (n = 75 [82.4%]) initially presented with New York Heart Association functional class I or II symptoms.

Table 1.

Demographic data of the pediatric HCM cohort: Comparison between patients with sarcomere gene pathogenic variants and those with RASopathy syndrome

Variable	All (n = 91)	Sarcomere gene pathogenic variants (n = 25)	RASopathy syndrome (n = 19)	P value
Genetic test		—	—
Sarcomere gene (+)	22 (24.2)
Noonan syndrome spectrum	19 (20.9)
Negative	10 (11)
Not done	40 (44)
Female/male	31/60	6/19	9/10	.123
Initial diagnosis age, y	8.1 ± 7.1	12. ± 5.2	1.3 ± 2.0	<.001∗
Family history				.062
SCD	4 (4.4)	3 (12)	0
HCM	11 (12.1)	5 (20)	1 (5.3)
Both SCD and HCM	6 (6.6)	4 (16)	1 (5.3)
Initial symptoms				.072
SCA	6 (6.7)	4 (16)	0
Syncope	7 (7.7)	3 (12)	0
Chest pain/exercise intolerance	25 (27.5)	7 (28)	5 (26.3)
Murmur/screening	52 (57.8)	11 (44)	14 (73.7)
NYHA functional class (initial)				.144
I	59 (64.8)	19 (76)	10 (52.6)
II	16 (17.6)	4 (16)	3 (15.8)
III	12 (13.2)	2 (8)	3 (15.8)
IV	4 (4.4)	0	3 (15.8)
Syncope	12 (13.3)	6 (24)	1 (5.3)	.099
Initial ECG parameters
ST depression lateral leads	19 (22.9)	11 (45.8)	2 (11.1)	.021∗
T inversion lateral leads	27 (32.5)	15 (62.5)	2 (11.1)	.001∗
QRS duration, ms)	92.0 ± 22.8	99.0 ± 13.2	75.6 ± 13.1	<.001∗
QTc interval, ms	445 ± 38	447 ± 27	446 ± 22	.866
Initial echocardiographic parameters
LVEF, %	76.0 ± 13.0	71.5 ± 13.5	84.0 ± 7.8	.001∗
LVMI, g/m2	159 ± 77	178 ± 67	142 ± 58	.065
LV mass z score	3.2 ± 2.2	3.4 ± 2.0	3.8 ± 2.2	.579
IVS z score	10.0 ± 5.5	9.9 ± 5.9	10.4 ± 5.7	.779
IVS/LVPW	1.8 ± 0.8	2.0 ± 1.0	1.6 ± 0.5	.148
LVOT obstruction (PG >30 mm Hg)	29 (31.9)	4 (16)	8 (42.1)	.057
Initial NT-proBNP, pg/mL	3117 ± 4768	1251 ± 1282	4178 ± 5066	.034
LVOT intervention	20 (22)	5 (20)	6 (31.6)	.489
Outcome
Mortality or heart transplantation	11 (12.1)	2 (8)	4 (21.1)	.378
Aborted SCA	13 (14.3)	10 (40)	1 (5.3)	.013∗
Follow-up duration (y)	8.6 ± 7.4	7.2 ± 4.8	13.1 ± 10.0	.013∗

Values are n (%), n, or mean ± SD.

ECG = electrocardiography; HCM = hypertrophic cardiomyopathy; IVS = interventricular septum; LV = left ventricular; LVEF = left ventricular ejection fraction, LVMI = left ventricular mass index; LVOT = left ventricular outflow tract; LVPW = left ventricular posterior wall; NT-proBNP = N-terminal pro–brain natriuretic peptide; NYHA = New York Heart Association; PG = pressure gradient; SCA = sudden cardiac arrest; SCD = sudden cardiac death.

^∗P value less than .05.

Initial ECG findings indicated LV or right ventricular hypertrophy in 83% of patients, with a strain pattern present in 54.2%. On echocardiography, the initial LV ejection fraction was 76 ± 13%, with LV ejection fraction <50% in 3.4% of patients. The initial LV mass index was 159 ± 77 g/m², the initial LV mass z score was 3.2 ± 2.2, and the initial IVS z score was 10.0 ± 5.5. LVOT obstruction was present in 29 (31.9%) patients, with 20 (22%) undergoing LVOT obstruction intervention.

Genetic data and the association with clinical manifestation

Among the 55 patients who received genetic testing, the yield rate was 80%. In the remaining 36 patients who did not undergo genetic testing, 4 died before the study began, 12 declined genetic testing, and 20 were lost to follow-up. Table 1 presents comparisons between patients with sarcomere gene pathogenic variants and those with RASopathy syndrome. Patients with sarcomere gene pathogenic variants had an older age at initial diagnosis, a more frequent initial presentation of SCA or syncope, and a higher prevalence of ST depression and T-wave inversion in the lateral leads compared with those with RASopathy.

The details of genetic pathogenic variants are summarized in Table 2. Among the sarcomere gene, MYH7 was the most prevalent pathogenic variant in this young population, followed by MYBPC3. Regarding the RAS gene in young patients with RASopathy-associated HCM, RAF1 and PTPN11 were identified as the most common pathogenic variants. Within the sarcomere gene pathogenic variant group, SCA did not show a specific association with particular genetic variants.

Table 2.

Genetic variants and clinical outcome in pediatric HCM patients

Gene group	Gene of P or LP	Patients	Outcome
Sarcomere	MYH7	10 (40)	2 died, 2 SCA
	MYBPC3	6 (24)	2 SCA
	TPM1	4 (16)	All survived
	TNNT2	2 (8)	1 SCA
	TNNI3	1 (4)	1 SCA
	TTN	1 (4)	Survived
	VCL	1 (4)	1 SCA
RASopathy	RAF1	7 (36.8)	1 died,1 SCA
	PTPN11	4 (21.1)	1 died
	SOS1	2 (10.5)	1 died
	HRAS1	1 (5.6)	Survived
	MRAS	1 (5.6)	Survived
	BRAF	1 (5.6)	Survived
	LZTR1	1 (5.6)	Survived

Values are n (%).

HCM = hypertrophic cardiomyopathy; LP = likely pathogenic variant; P = pathogenic variant; SCA = sudden cardiac arrest.

Outcome

The mean follow-up duration of the 91 patients was 8.6 ± 7.4 years. The 5-year and 10-year heart transplantation–free survival rates were 91.8% and 86.3%, respectively, while the SCA-free survival rates were 86.9% and 84.2%, respectively. During the 5- and 10- year follow-up period, the loss to follow-up rate was 7.6% and 12%. The age range for SCA was predominantly between 10 and 18 years. Among the 91 patients, 13 experienced SCA. Notably, of the 13 patients who experienced SCA, 46.3% (6 patients) had it as their initial presentation. A comparison was made among patients without cardiovascular events, those with aborted SCA, and those who reached the primary endpoint (death or heart transplantation) without SCA (Table 3). Patients who died or underwent heart transplantation without SCA had a significantly lower initial diagnostic age, a higher incidence of RASopathy genetic variants, and poorer functional class at initial diagnosis. Patients who experienced SCA had an older diagnostic age, a higher incidence of sarcomere genetic variants, and a notable proportion presented initially with collapse or syncope. Initial echocardiographic parameters, including LV mass, IVS thickness, and LVOT obstruction, showed no significant differences among these 3 outcome groups.

Table 3.

Comparisons among patients with no events, those presenting with aborted SCA, and those who died or underwent heart transplantation

	No event	Death/heart transplantation without SCA	SCA	P value
Female/male	26/44	1/7	4/9	.365
Initial diagnostic age, y	7.9 ± 7.2	2.9 ± 5.6	12.8 ± 4.8	.005∗
HCM group
Sarcomere genetic variants (+)	15 (21.4)	0	10 (76.9)	<.001∗
RASopathy genetic variants (+)	14 (20)	4 (50)	1 (7.7)
Family history				.009∗
SCD	1 (1.4)	0	3 (23.1)
HCM	10 (14.3)	0	1 (7.7)
Both	4 (5.7)	0	2 (15.4)
Initial symptoms				<.001∗
Collapse	0	0	6 (46.2)
Syncope	5 (7.2)	0	2 (15.4)
Chest pain/dyspnea	19 (27.5)	3 (37.5)	3 (23.1)
Incidental	45 (65.2)	6 (62.5)	2 (15.4)
NYHA functional class (initial)				<.001∗
I	50 (71.4)	1 (12.5)	8 (61.5)
II	13 (18.6)	0	3 (23.1)
III	7 (10)	3 (37.5)	2 (15.4)
IV	0	4 (50)	0
Previous syncope	9 (13)	0	3 (23.1)	.657
Initial ECG parameters
ST depression lateral leads	13 (20.3)	1 (14.3)	5 (41.7)	.231
T inversion lateral leads	22 (34.4)	1 (14.3)	4 (33.3)	.559
QRS duration, ms	90.3 ± 20.2	98.3 ± 41.6	96.6 ± 21	.541
QTc interval, ms	442 ± 38	466 ± 37	448 ± 35	.355
Initial echocardiographic parameters
LVEF, %	78.6 ± 10.9	73.8 ± 16.5	63.8 ± 14.6	.007∗
LVMI, g/m2	159 ± 70	158 ± 153	156 ± 59	.383
LV mass z score	3.3 ± 2.2	3.3 ± 3.2	2.6 ± 1.8	.668
IVS z score	10.3 ± 5.8	8.0 ± 3.1	9.6 ± 4.5	.529
IVS/LVPW	1.9 ± 0.9	1.5 ± 0.6	1.9 ± 0.7	.264
LVOT obstruction (PG >30 mm Hg)	24 (34.3)	4 (13.8)	1 (7.7)	.086
Initial NT-proBNP, pg/mL	3322 ± 5275	4798 ± 2379	1226 ± 1207	.112
LVOT intervention	16 (22.9)	1 (12.5)	3 (23.1)	.795

Values are n, mean ± SD, or n (%).

ECG = electrocardiography; HCM = hypertrophic cardiomyopathy; IVS = interventricular septum; LV = left ventricular; LVEF = left ventricular ejection fraction, LVMI = left ventricular mass index; LVOT = left ventricular outflow tract; LVPW = left ventricular posterior wall; NT-proBNP = N-terminal pro–brain natriuretic peptide; NYHA = New York Heart Association; PG = pressure gradient; SCA = sudden cardiac arrest; SCD = sudden cardiac death.

^∗P value less than .05.

Risk analysis and application of risk scoring system

Cox regression analysis were assessed for their impact on the primary endpoint (death or heart transplantation) or SCA. Initial New York Heart Association functional class emerged as the sole significant risk factor for the primary endpoint, with an odds ratio of 5.08 (95% confidence interval 2.1–12.3, P < .001) (Supplemental Table 1). Similarly, in the analysis of the risk of SCA, sarcomere genetic variants were identified as the critical risk factor, with an odds ratio of 10.2 (95% confidence interval 1.29–80.5, P = .011) (Supplemental Table 2).

After excluding the 6 patients with aborted SCA as their initial presentation, we applied the risk factors outlined by the AHA consensus guideline (Figure 1A). We also applied the HCM Risk-Kids score system to individuals 1 to 16 years of age and the PRIMaCY score system (including genetic test results or not) to those under 18 years of age, excluding cases of RASopathy (Figures 1B–1D). Both the HCM Risk-Kids and PRIMaCY genetic models demonstrated perfect discriminatory ability, with events occurring exclusively in the high-risk group. In contrast, the AHA guideline and the PRIMaCY clinical score system did not demonstrate good discriminatory ability (Table 4).

Figure 1.

The sudden cardiac arrest event-free survival curve for sudden cardiac death since initial diagnosis, categorized according to (A) American Heart Association guideline, (B) HCM Risk-Kids risk score, (C) PRIMaCY (precision medicine in cardiomyopathy) genetic risk score, and (D) PRIMaCY clinical risk score. Neither the AHA guideline nor the PRIMaCY clinical risk score perfectly differentiates between low- and high-risk groups (P values by log-rank analysis were .37 and .23, respectively). In contrast, the HCM Risk-Kids and PRIMaCY genetic risk score systems can effectively differentiate between low- and high-risk groups, with all events occurring in the high-risk group (P = .086 and P = .050 by log-rank analysis, respectively).

Table 4.

HR and C-index of the predictivity of SCA among 4 different prediction models

	SCA rate (%)	P value	HR (95% CI)	P value	C-index (95% CI)
AHA model (n = 91)		.360			0.596 (0.375–0.817)
0 risk factor	3.9		Reference
≥1 risk factors	10.3		2.22 (0.37–13.26)	.38
HCM Risk-Kids model (n = 43)		.061
≤4%	0		NA
4%–6%	0		NA
≥6%	29.4		NA
PRiMACY genetic model (n = 60)		.059
<4.7%	0		NA
4.7%–8.3%	0		NA
> 8.3%	21.7		NA
PRiMACY clinical model (n = 60)		.198			0.709 (0.571–0.848)
<4.7%	0		NA
4.7%–8.3%	7.1		Reference
> 8.3%	17.4		2.261 (0.253–20.239)	.47

The AHA guideline was applied to the whole cohort, and HCM Risk-Kids and PRIMaCY (clinical and genetic) models were applied for hypertrophic cardiomyopathy patients excluding RASopathy and at the defined age group.

AHA = American Heart Association; CI = confidence interval; HR = hazard ratio; NA = not applicable; PRIMaCY = precision medicine in cardiomyopathy.

We also compared the 6 patients with SCA as their initial presentation to those with SCA at later follow-up. ECG parameters, family history, genetic background, and sex showed no significant differences between these two groups. However, patients with SCA as their initial presentation were older (15.8 ± 1.9 years of age vs 10.2 ± 5.1 years of age, P = .005), had a smaller LV mass z score (1.2 ± 2.1 vs 3.4 ± 1.1, P = .073), and lower initial NT-proBNP (291 ± 340 pg/mL vs 1786 ± 1203 pg/mL, P = .143). We then calculated the HCM Risk-Kids and PRIMaCY scores for these 6 patients using the parameters immediately after they were successfully resuscitated: 3 (50%) patients were categorized as low risk in the HCM Risk-Kids score system and 1 (16.6%) patient was categorized as low risk in the PRIMaCY genetic score system.

Discussion

In this pediatric HCM cohort, several key findings emerged: (1) RASopathy-related HCM was more commonly associated with heart failure symptoms and heart failure–related mortality, whereas sarcomere gene–related HCM more frequently presented initially with SCA or syncope and exhibited a higher incidence of SCA; (2) SCA is not rare in HCM, and nearly half of the events occur as the initial symptom, making risk prediction of these populations challenging; and (3) excluding the 6 patients who presented initially as SCA, both the HCM Risk-Kids and PRIMaCY genetic models demonstrated strong discriminatory ability for predicting SCA.

Clinical outcome of HCM

HCM is one of the most common inherited heart diseases and a leading cause of sudden cardiac death.¹⁴ Pathogenic variants in sarcomeric/myofilament-related protein genes are the most common causes of HCM, followed by Noonan spectrum syndrome (RASopathy).^3,15 In the present study, the clinical presentation of RASopathy-associated HCM differs significantly from that of sarcomere HCM. First, the age of diagnosis is younger, with a mean age of only 1.3 years in these RASopathy-associated HCM cases compared with 12.1 years for sarcomere-related HCM. Second, ST depression and T-wave inversion in the lateral leads on a 12-lead ECG are less frequently observed in RASopathy-associated HCM compared with sarcomere HCM. Third, a significant portion of the patients with RASopathy-associated HCM experienced progression of heart failure symptoms to New York Heart Association functional class III or IV, a scenario rarely seen in sarcomere-related HCM. Last, most patients with RASopathy-associated HCM were identified through echocardiography screening for heart murmurs or other anomalies, with none presenting initially with syncope or collapse. In contrast, 28% of patients with sarcomere-related HCM initially presented with collapse or syncope. These findings were consistent with the recent study from Lynch and colleagues.¹⁶ These differences assist clinicians in distinguishing between RASopathy-associated and sarcomere-related HCM based on clinical characteristics.

In studies from Western countries, children with nonsyndromic HCM generally have relatively benign outcomes, with a reported 5-year survival rate of 82.2%.¹⁷ In our study, the 5-year heart transplantation-free survival rate for the entire HCM cohort was 91.8%, which is comparable to outcomes reported in Western countries. However, our findings indicate that SCA is more common in the sarcomere group than in the RASopathy group. This contrasts with Lynch and colleagues’ report,¹⁶ which observed a similar incidence of SCA in both groups. However, the number of SCA events in RASopathy group was low in both studies (1 in our study and 4 in theirs), suggesting that larger studies with longer follow-up durations may be necessary to further clarify these issues.¹⁸ Notably, 47% of SCAs occurred as the initial presentation—a rate higher than previously reported.¹⁹ As a tertiary care center, there may be a selection bias due to the higher number of referred patients who were successfully resuscitated from SCA. Racial differences may also play a role, as previous studies have demonstrated variations in incidence and clinical presentation among different ethnic groups.^20,21 Furthermore, a Korean study demonstrated that despite the HCM Risk-SCD calculator exhibiting a high negative predictive value in an Asian population, a significant proportion of SCD events still occurred among patients classified as low to intermediate risk.²² Despite the high rate of SCA events, most patients survived without neurological complications. The low mortality rate may be attributed to the widespread use of automated external defibrillators in schools, which is part of our national policy.

SCA risk factors analysis in HCM

In the AHA/American College of Cardiology guidelines, family history of SCD, massive LVH with IVS thickness >3 cm, unexplained syncope, apical aneurysm, poor ejection fraction, and NSVT are key risk factors for SCD and indications for implantable cardioverter-defibrillator implantation.²³ However, this risk prediction model was derived from adult studies. When applied to our cohort, its discriminatory ability was unsatisfactory, with a C-index of 0.596.²⁴ Recently, a meta-analysis identified previous ventricular fibrillation, NSVT, unexplained syncope, and extreme LVH as risk factors for SCD in childhood HCM.²⁵ Two multicenter studies have since proposed risk score calculators for pediatric populations. The HCM Risk-Kids study included 1,072 nonsyndromic pediatric HCM patients, utilizing LV wall thickness, LA diameter, LVOT obstruction, family history of SCD, NSVT, unexplained syncope, and age as risk factors, each with varying weight. This calculator, available online, allows for the calculation of individualized 5-year SCD risk. A previous validation reported a C-index of 0.69 for HCM Risk-Kids.²⁶ PRIMaCY, which included 572 nonsyndromic pediatric HCM patients, used the same risk factors with different weightings but also incorporated pathogenic gene variants. PRIMaCY has been externally validated, with a C-index of 0.71.¹⁰

In the present study, we applied the HCM Risk-Kids and PRIMaCY calculator in the present cohort after excluding 6 patients with SCA as their initial presentation. We found that both HCM Risk-Kids and PRIMaCY genetic model have good discrimination ability, as all patients that experienced SCA were in the high-risk group (Figure 1). This finding is consistent with a recent report demonstrating comparable C-indices for the PRIMaCY clinical and genetic models.²⁷ Moreover, a study by Fontanges and colleagues²⁸ showed that both the HCM Risk-Kids and PRIMaCY genetic models have a high negative predictive value but a relatively low positive predictive value, which is in line with our findings (Table 4). However, nearly half of our cohort of SCA experienced SCA as their initial presentation. As a major tertiary care center, many pediatric patients with SCA are referred to our hospital, which could lead to an overestimation of SCA incidence in our cohort. Nonetheless, a substantial percentage of patients experienced SCA as their initial presentation, making the prevention of SCA in pediatric HCM challenging.

The mechanisms of SCA in these HCM patients are multifactorial. Potential causes include mutation effects, such as disturbed calcium handling leading to triggered activity by afterdepolarization, and secondary effects from increased automaticity and reentry arrhythmia due to cardiomyocyte hypertrophy, fibrosis, and disarray.^29,30 In our cohort, nearly half of the patients with SCA had aborted SCA as their initial presentation, even during early-stage HCM, with low LV mass and IVS thickness on echocardiography. The early onset of SCD may be attributed to initial myocyte disarray and disturbed calcium handling. Familial cascade or school ECG screening could be key to identifying high-risk patients and preventing SCA in this population.

Genetic variants in HCM

Since the first genetic variant was identified in 1990, multiple pathogenic variants in distinct sarcomere proteins have been reported. The most common causal genes are MYH7 and MYBPC, accounting for 75% to 80% of cases, followed TNNT2 and TNNI3.^6,31 In the present study, when considering only nonsyndromic patients, MYH7 and MYBPC3 were the most frequent genetic variants, representing 40% and 24% of cases, respectively. Additionally, TPM1, TNNT2, TNNI3, TTN, and VCL were identified as possible genetic variants. These findings align with previous studies.^6,31 While previous research has highlighted the role of genetic variants in disease outcomes, the results have been inconsistent.32, 33, 34 Differences in patient age groups and study endpoints may account for the variation in findings. In this study, no single genetic variant was found to pose a significantly higher risk for SCA than others, suggesting that all sarcomere pathogenic variants should be considered equally in pediatric HCM.

Study limitation

This retrospective study has several limitations. First, not all patients underwent genetic testing. Second, due to the study's retrospective design, follow-up data—such as serial echocardiography results and NT-proBNP levels—were not consistently available for all patients. Third, as a single-center study, the sample size is relatively small, which may limit the statistical power. Future multicenter and prospective studies are needed to further validate the prognostic factors identified in this study.

Conclusion

Pediatric HCM, whether associated with sarcomere gene pathogenic variants or RASopathy, shows distinct patterns of disease progression and outcomes. SCA is notably linked to sarcomere gene pathogenic variants. While newer risk scoring systems are effective at predicting SCA during follow-up, we still encounter the difficulty that nearly half of the cases present as the initial symptom. Familial cascade or school ECG screening may be a key method to identifying high-risk patients and preventing SCA in this population.