Clinical characteristics of the population

In line with data published in paediatric registries from North America^{Reference Lipshultz, Sleeper and Towbin1} and Australia,^{Reference Nugent, Daubeney and Chondros2} the leading cause of cardiomyopathy in this study was dilated cardiomyopathy (48.6%). In contrast, the European registry reported hypertrophic cardiomyopathy as the most common cause. This difference may be explained by the exclusion of patients younger than 1 year of age and the high referral rate of patients with hypertrophic cardiomyopathy to specialised centres, given the perceived risk of genetic disease.^{Reference Kaski, Norrish and Gimeno Blanes10}

The rates of presentation for other cardiomyopathy subtypes, such as restrictive cardiomyopathy, left ventricular non-compaction cardiomyopathy, and arrhythmogenic cardiomyopathy, were consistent with those reported by Ware et al. ^{Reference Ware, Wilkinson and Tariq18} In the present study, a female predominance was observed across all cardiomyopathy subtypes, except for arrhythmogenic cardiomyopathy, which aligns with the findings of Ware et al. ^{Reference Ware, Wilkinson and Tariq18} Other North American and Australian registries, however, reported a higher incidence in males.^{Reference Lipshultz, Sleeper and Towbin1,Reference Nugent, Daubeney and Chondros2} Recent studies on adults have determined that sex influences the prevalence, morpho-functional manifestations, and clinical course of cardiomyopathies. Although the underlying mechanisms remain poorly understood, genes involved in the expression of various ion channels, ventricular remodelling, and disease modifiers may influence outcomes, alongside epigenetic factors (social and environmental complexities between the sexes), potentially leading to greater disadvantages for women with this condition^{Reference Argiró, Ho and Day19,Reference Xu, Chen and Tillman20} These findings have not been extensively studied in the paediatric population, presenting an opportunity for future research.

Familial disease and genetics

The proportion of patients with a familial history was estimated at 31.6%, with variations observed according to the subtype of cardiomyopathy. This was most common in patients with arrhythmogenic cardiomyopathy, where the reported cases represented a familial cluster, followed by those with dilated cardiomyopathy. In contrast, the rate of family history was lowest in patients with hypertrophic cardiomyopathy, which contrasts with previous studies where familial disease was more prevalent in this cardiomyopathy subtype.^{Reference Kaski, Norrish and Gimeno Blanes10} This discrepancy may be explained by the fact that family screening in our healthcare system is complex. There are a few cardiogenetic programmes available, and screening is sometimes incomplete or performed only once without follow-up. In addition, administrative limitations prevent the direct issuance of referrals or screening medical orders for the family members of the index case.

Genetic diagnosis and variants

The overall genetic diagnosis rate was 25.6%, with the highest rate observed in restrictive cardiomyopathy (50%), hypertrophic cardiomyopathy (25%), left ventricular non-compaction cardiomyopathy (25%), and dilated cardiomyopathy (24.3%). The diagnostic yield of exome sequencing was 71.4%. The rate of genetic diagnosis with pathogenic and likely pathogenic variants was similar to that reported by Ware et al., ^{Reference Ware, Wilkinson and Tariq18} which found a rate of 32%. However, the yield was higher in other registries, such as the European Cardiomyopathy and Myocarditis Registry^{Reference Kaski, Norrish and Gimeno Blanes10} (60%) and the North American Paediatric Cardiomyopathy Registry^{Reference Ware, Wilkinson and Tariq18} (48%). This discrepancy may reflect selection bias from expert centres with easier access to genetic testing. This contrasts with the population in the present study, where the retrospective nature of the research limited the ability to perform additional genetic studies or reanalyse previously conducted genetic tests in a controlled manner.

As determined in both adult and paediatric studies, sarcomeric genes are predominantly responsible for cardiomyopathy across different subtypes.^{Reference Kaski, Syrris and Esteban21–Reference Ware, Bhatnagar and Dexheimer24} In this study, the most common pathogenic or likely pathogenic variants across the cardiomyopathy subtypes were in MYH7, TTNI3, TNNT2, MYBPC3, and FLNC. The MYH7 gene was the most frequent cause in dilated cardiomyopathy, while MYBPC3 was most commonly seen in hypertrophic cardiomyopathy, consistent with international^{Reference Kaski, Syrris and Esteban21–Reference Bagnall, Singer and Wacker23} and national^{Reference Castro-Oróstegui and Beltrán-Durán25} reports on genotyping in paediatric cardiomyopathies. However, these findings suggest that genotype-phenotype correlations specific to each cardiomyopathy subtype may provide a more refined understanding of the molecular underpinnings of the disease and its association with presentation, clinical progression, and outcomes. Therefore, when an index case is identified, family cascade screening should be initiated due to the importance of early detection, especially in patients associated with the autosomal dominant inheritance pattern of these cardiomyopathies.

Although in recent years—particularly in adult populations—excessive trabeculations have been increasingly recognised as a morphological phenotype within dilated and hypertrophic cardiomyopathies, this concept has not yet gained wide acceptance in paediatric cardiology. The main reasons include genetic heterogeneity, overlap between genotype and phenotype, the lack of diagnostic imaging criteria validated for children, and heterogeneous findings observed in longitudinal follow-up of paediatric patients with left ventricular non-compaction cardiomyopathy. Moreover, the prognostic significance of excessive trabeculation in this group remains uncertain. Therefore, multicentre studies and an international consensus specifically focused on the paediatric population are needed to refine diagnostic accuracy and improve risk stratification and outcomes in these patients.^{Reference Petersen, Friedrich and Ho26,Reference van Waning, Caliskan and Hoedemaekers27}

The potential impact of the number of variants on clinical symptoms and the outcomes of different cardiomyopathies has been explored.^{Reference Xie, Yang and Gong28,Reference Walsh, Thomson and Ware29} In this study, six patients presented pathogenic variants associated with variant of uncertain significance in other proteins. One patient, with a pathogenic mutation in MYH6 and variants of uncertain significance in SCN5A, developed complete atrioventricular block after the cardiomyopathy diagnosis. However, no differences in mortality, transplantation, or use of ventricular assist devices were observed when compared to those with a single variant. Another aspect to consider is the role of digenic inheritance, which suggests that mutations in more than one gene may contribute to disease development. This complicates the diagnosis and understanding of cardiomyopathy pathogenesis and remains an area of active research.^{Reference Parikh, Day and Lakdawala30,Reference Liu, Yang, Gong, Wang, Yao and Xie31}

Treatment

A high proportion of patients in this study received cardiovascular medications, which may indicate bias due to the severity of their presentation upon admission, as also noted in the European Registry of Cardiomyopathy and Myocarditis. One in nineteen patients received implantable cardioverter defibrillator therapy, which is consistent with other paediatric studies.^{Reference Maron, Spirito and Ackerman32} The proportion was higher in patients with arrhythmogenic cardiomyopathy (33% vs. 60%) compared to other registries of both adults and children.^{Reference Kaski, Norrish and Gimeno Blanes10–Reference Marino, Digilio, Toscano, Giannotti and Dallapiccola14,Reference Wilkinson, Lowe and Salbert16–Reference Charron, Elliott and Gimeno33} In patients with hypertrophic cardiomyopathy, the proportion requiring septal myomectomy (16.6% vs.7.5%) was higher than in other studies despite the fact that in our cohort, we did not include patients with phenocopies, in whom the prevalence of left ventricular outflow tract obstruction is higher.^{Reference Wilkinson, Lowe and Salbert34–Reference Jia, Meng, Tang, Sun, Peng and Zhang36}

Outcomes

The overall mortality rate in the present registry was 22.3%, occurring within the first 10 years after diagnosis. This rate varied substantially when analysing the subtypes separately, being higher in our patients with arrhythmogenic cardiomyopathy (100%), which was explained by a particular severe form of the disease in a familial presentation.^{Reference Kaski, Norrish and Gimeno Blanes10,Reference Erolu, Akalın, Saylan Çevik and Yücelten37}

Heart transplantation was a common treatment option in some of the cardiomyopathy subtypes, and it was more frequent within the first 10 years after diagnosis. One in two patients with restrictive cardiomyopathy required a heart transplant, followed by one in five patients with dilated cardiomyopathy, consistent with previous studies reported in this patient subpopulation.^{Reference Towbin, Lowe and Colan38} A ventricular assist device was required in two patients in the entire cohort (1 dilated cardiomyopathy, 1 hypertrophic cardiomyopathy) as a bridge to heart transplantation.