Clinical Features and Natural History of Preadolescent Nonsyndromic Hypertrophic Cardiomyopathy

BACKGROUND Up to one-half of childhood sarcomeric hypertrophic cardiomyopathy (HCM) presents before the age of 12 years, but this patient group has not been systematically characterized. OBJECTIVES The aim of this study was to describe the clinical presentation and natural history of patients presenting with nonsyndromic HCM before the age of 12 years. METHODS Data from the International Paediatric Hypertrophic Cardiomyopathy Consortium on 639 children diagnosed with HCM younger than 12 years were collected and compared with those from 568 children diagnosed between 12 and 16 years. RESULTS

A lthough variants in cardiac sarcomeric protein genes are responsible for the majority of childhood-onset hypertrophic cardiomyopathy (HCM), 1,2 because of variable and incomplete age-related penetrance, sarcomeric HCM has historically been considered a disease of adolescence or adulthood. 35][6][7][8] Despite this, the clinical presentation, natural history, and outcomes of preadolescent sarcomeric HCM have not been systematically characterized to date.
For decades, the management of HCM in both children and adults has focused on symptom palliation, family screening, and prevention of diseaserelated complications.With the advent of novel disease-modifying therapies such as myosin inhibitors 9 and gene therapy programs, 10  and 2019. 11HCM was defined as a maximal left ventricular (LV) wall thickness >2 SDs higher than the body surface area-corrected population mean (z-score $þ2). 12pplemental Figure 1 11 are expressed in millimeters and z-scores relative to the distribution of measurements for body surface area in healthy children. 15LV outflow tract (LVOT) gradient was measured at rest.LVOT obstruction (LVOTO) was defined as an instantaneous peak Doppler LVOT pressure gradient $30 mm Hg. 12 Moderate LVOTO was defined as a pressure gradient of 50-90 mm Hg and severe as a pressure gradient $90 mm Hg.Nonsustained VT was defined as $3 consecutive ventricular beats at a rate of >120 beats/min lasting <30 seconds on ambulatory electrocardiography. 12 The indication for ICD implantation was defined as primary prevention in patients considered to be at high risk for life-threatening ven-   Eleven patients (5.9%) were compound heterozygous or homozygous for disease-causing variants (Table 2, Supplemental Table 3).Eleven patients had variants in nonsarcomeric genes that were considered disease causing (P/LP): Patients with P/LP variants were more likely to have family histories of HCM (n ¼ 131 [71.2%] vs n ¼ 54 [40.9%];P < 0.001) but did not otherwise differ in baseline characteristics (Supplemental Tables 4 and 5).2A).The overall annual incidence of mortality or cardiac transplantation was 1.41% (95% CI: 1.10%-1.81%)(Figure 1A).
On multivariable analysis, the presence of heart failure symptoms, increasing left atrial diameter z-score, and the absence of a disease-causing variant on genetic testing were associated with reaching the mortality or transplantation endpoint (Supplemental  2B), and no patients experienced events before 5 years of age.Of these, 23 patients (33.3%) had previously undergone implantation of primary prevention ICDs.Three patients experienced sudden death with ICDs in situ, but electrograms and information on cardiac rhythm at the time of death were not available for review.The overall annual incidence of life-threatening arrhythmic events was 1.52% (95% CI: 1.20%-1.93%)(Figure 1B).

COMPARISON OF HCM PHENOTYPE AND OUTCOMES
BY AGE OF PRESENTATION.

DISCUSSION
To our knowledge, this study is the first systematic description of early childhood-onset HCM.The findings highlight that young patients have a comparable symptom burden and long-term outcomes as those presenting later in childhood (Central Illustration), with the implication that even the youngest children with HCM may benefit from similar management strategies and novel therapeutic approaches as used in older patients with HCM.
PHENOTYPE OF PREADOLESCENT HCM.4][25] The phenotype of nonsyndromic childhood HCM is recognized to be highly variable, but whether age at presentation influences disease severity has not been previously investigated. 4,6,8,26In this study, one-fifth of children with preadolescent childhood HCM were symptomatic at baseline, and the cardiac phenotype was variable, including severe hypertrophy, LVOTO, and lifethreatening ventricular arrhythmias occurring during follow-up.This is comparable with previous population studies with unselected childhood disease. 7,27,28Although the proportion of patients with heart failure symptoms did not differ by age, younger patients were less likely to report unexplained syncope.This could be due to difficulties in distinguishing the cause of syncopal events in young children.However, it may also suggest age-related differences in patient-reported symptoms not explained by differences in phenotype.The degree of hypertrophy was not significantly different between the 2 groups, but a higher proportion of preadolescent patients had LVOTO, and an important minority (6%) had severe obstruction (gradient $90 mm Hg).We speculate that this could be explained by undiagnosed nonsarcomeric disease, which is more commonly associated with LVOTO and typically presents at a younger age. 29However, as the difference persisted in subgroup analysis of patients with identified sarcomeric disease-causing variants, it is more likely that age-related differences in phenotype expression of sarcomeric disease exist.The progression or development of LVOTO in childhood disease is poorly understood, and future studies using serial clinical data are required to investigate the changing burden of LVOTO during childhood, as well as the effect of surgical and pharmacologic septal reduction strategies on prognosis.
GENETICS OF PREADOLESCENT HCM.Recent data have suggested a higher yield of genetic testing for sarcomere protein gene mutations in childhood-onset HCM (1-18 years of age) compared with adult cohorts. 5netic testing was not performed systematically in our cohort, and it was therefore beyond the scope of this study to investigate the yield of genetic testing.
Nonetheless, one strength of this study is that genetic testing information was available for more than onehalf of the cohort, identifying disease-causing variants in more than 50%.This suggests that the yield of genetic testing in preadolescent disease is similarly high and confirms the contribution of sarcomeric disease to early childhood-onset disease. 1,2,302][33] In agreement with previous childhood reports, P/LP variants were most commonly identified Values are n (%) or n.LP ¼ likely pathogenic; P ¼ pathogenic.
in MYH7 and MYBPC3. 5Although compound heterozygote or homozygote sarcomeric variants were more common in the preadolescent group, this remained relatively rare (6%), which suggests that the majority of HCM presenting in childhood is caused by single high-impact gene variants.A small number of patients had disease-causing variants identified in nonsarcomeric protein genes.This could suggest undiagnosed syndromic disease but may also support the role of nonsarcomeric variants contributing to disease phenotype in children with clinically nonsyndromic disease. 34PACT OF AGE ON THE OUTCOMES OF PREADOLESCENT HCM.6,35 In this study, the incidence of death or cardiac transplantation was similar for those presenting before 12 years of age and in later childhood, but for those presenting earlier in childhood, these events occurred at a younger age, and three-quarters took place before 20 years of age.
Increasing age at presentation was associated with a higher risk for mortality or cardiac transplantation on univariate analysis, but this did not remain significant on multivariate analysis.The causes of death were similar in both groups, with SCD remaining the most frequent cause regardless of age The cumulative incidence was similar in those presenting at 1-<12 years of age and $12 years of age for (A) death or cardiac transplantation (P ¼ 0.447) and (B) life-threatening arrhythmic events (P ¼ 0.104).
J A C C V O L .7 9 , N O . 2 0 , 2 0 2 2 of presentation, but a higher proportion of preadolescent patients underwent cardiac transplantation.This is in contrast to a recent North American study, which found that non-SCD death and transplantation occurred more frequently in those diagnosed before 5 years of age, although the majority of these events occurred in infancy, an age range excluded from this study. 27ildren are recognized to have a higher incidence of arrhythmic events compared with adult patients, but the relationship between age and risk remains unclear. 8,11,27,36,37In this study, the overall incidence of arrhythmic events was similar for those presenting prior to or during adolescence, but there was a trend toward a higher prevalence of SCD and resuscitated cardiac arrest rather than appropriate ICD therapies in the younger age group.Patients presenting before 12 years of age were less likely to undergo ICD implantation during follow-up but were more likely to require ICD implantation for secondary prophylaxis following a life-threatening arrhythmia.This could suggest a difference in clinicians' perception of risk and subsequent management for those presenting at a younger age, as well as reflecting technical difficulties in implanting ICDs in small patients. 38Of note, no patients experienced events before 5 years of age, which could suggest that the very young patients, in whom device implantation may represent a technical challenge, may be less likely to benefit from ICD implantation.The time to a life-threatening arrhythmic event did not differ by age of presentation, and arrhythmic events occurred at a younger age in those presenting in preadolescence.This is in contrast to findings reported by Miron et al, 27 in which the frequency of arrhythmic events was highest during adolescent years regardless of age of diagnosis.Our understanding of the interaction among age at presentation, current age, and risk stratification remains incomplete, and recently published pediatric risk models have differed in their treatment of age at presentation as a risk factor for arrhythmic events. 11,27In previous analysis of this dataset, including age as a predictor variable did not improve the ability to predict arrhythmic events (HCM Risk-Kids). 11In contrast, an alternative North American risk model reported that increasing age was associated with an increasing risk for arrhythmic events. 27rther exploration of the relationship is needed, but the data from this study highlight that all childhood patients are at risk for arrhythmic events and stress the importance of risk stratification as a cornerstone of clinical care.
STUDY LIMITATIONS.This study had inherent limitations due to its multicenter and retrospective design, including missing data and incomplete recruitment of eligible patients.Variations in clinical assessment and patient management are inevitable, as patients were recruited from multiple centers and different geographic locations.Previous analysis of this cohort, however, has shown no difference in outcomes by era of presentation, which may be explained by no significant change in management strategies over time, with the exception of ICD implantation. 4,11A diagnosis of HCM may be made following presentation with symptoms, incidentally or through family screening, with the implication that the timing of HCM diagnosis may be significantly determined by local health care screening patterns. 4 is plausible that symptomatic patients, who may have phenotypically more severe disease and worse outcomes, could present at a younger age, introducing bias to the analysis.Data on reason for presentation were not collected in this cohort, although a similar proportion were symptomatic (heart failure symptoms or syncope) at baseline and had family histories of HCM in both age groups.The exclusion of patients presenting with out-of-hospital ventricular fibrillation arrest could have further biased the results toward patients with a less severe disease phenotype.However, a previous national cohort study of childhood HCM from United Kingdom has reported that this is a rare (<4%) reason for presentation during childhood. 4As sudden death is a rare event, a composite definition of life-threatening arrhythmias was used for this study, which includes appropriate ICD therapies.This is in keeping with previous studies in adults and children with HCM, 11,27,36,39   TRANSLATIONAL OUTLOOK: Future studies address the impact of patient age on the risk for arrhythmias and other adverse events associated with HCM and evaluate the benefit of disease-modifying therapies in children.
an improved understanding of age-specific differences would assist clinical care and could help guide the future use of disease-specific therapies in young children with HCM.The aim of this study was to describe the clinical characteristics and outcomes of a large, international, multicenter cohort of children with nonsyndromic HCM presenting before 12 years of age.METHODS STUDY POPULATION.Children meeting diagnostic criteria for HCM between 1 and <12 years of age were identified from the International Paediatric Hypertrophic Cardiomyopathy Consortium (IPHCC), which contains a total cohort of 1,207 children with nonsyndromic disease diagnosed between 1970 describes participation, recruitment, and retention in the consortium.As the IPHCC cohort of patients was previously used to develop and validate the HCM Risk-Kids pediatric HCM risk prediction model for sudden cardiac death (SCD), 11 it does not include patients with histories of resuscitated cardiac arrest, ventricular fibrillation, or sustained ventricular tachycardia (VT) prior to diagnosis.Patients with diagnoses of underlying inborn SCD = sudden cardiac death VT = ventricular tachycardia

FIGURE 1 B 2 Preadolescent
FIGURE 1 Age at Time of Death or Transplantation and Life-Threatening Arrhythmic Events

FIGURE 2
FIGURE 2 Cumulative Incidence of Death or Transplantation and Life-Threatening Arrhythmic Events Young children with HCM have a phenotype similar to patients presenting later and are often symptomatic, warranting early therapeutic intervention.

TABLE 1
Clinical Characteristics and Natural History of Preadolescent and Adolescent Nonsyndromic Hypertrophic Cardiomyopathy left ventricular maximal wall thickness; LVOT ¼ left ventricular outflow tract; NSVT ¼ nonsustained ventricular tachycardia; NYHA ¼ New York Heart Association; SCD ¼ sudden cardiac death; VT ¼ ventricular tachycardia.
meaning that those presenting before 12 years of age experienced events at a younger median age (13.9 years vs 18.7 years; P < 0.001).

TABLE 2
Disease-Causing Variants Identified on Genetic Testing

Cause of Death or Transplant (%) Annual Incidence of Death or Cardiac Transplant Did Not Differ by Age
CENTRAL ILLUSTRATION Clinical Characteristics and Outcomes of Early-Onset Childhood

of Life-Threatening Arrhythmia (%) Annual Incidence of Life-Threatening Arrhythmic Events Did Not Differ by Age
Children presenting at 1-<12 years of age have comparable outcomes as those presenting later in childhood but experience adverse events (death, cardiac transplantation, or life-threatening arrhythmias) at a younger age.CV ¼ cardiovascular; LVOT ¼ left ventricular outflow tract; MLVWT ¼ maximal left ventricular wall thickness.
but it is plausible that not all appropriate therapies would have necessarily resulted in sudden death untreated.Genetic testing was not performed systematically in this cohort, therefore it is beyond the scope of this study to investigate the yield as used in older patients with HCM should be used.FUNDING SUPPORT AND AUTHOR DISCLOSURESThis work was supported by the British Heart Foundation (grant FS/ 16/72/32270) to Drs Norrish and Kaski.This work is (partly) funded by