Clinical profile and outcomes of pediatric hypertrophic cardiomyopathy in a South Indian tertiary care cardiac center: a three decade experience

Although much research has been done on adult HCM, research on pediatric cardiomyopathy, is surprisingly scarce. Pediatric onset HCM has diverse etiologies with their variable respective outcomes. Predictors of death or adverse outcomes in pediatric-onset HCM remain unresolved, challenging risk stratification. Nugent et al. reported on 80 children with hypertrophic cardiomyopathy from the National Australian Childhood Cardiomyopathy Study [1]. Yetman et al. published a study on long-term outcome and prognostic determinants in 99 children With hypertrophic Cardiomyopathy in 1998 [2]. Maruizi et al. assessed 100 patients with pediatric-onset hypertrophic cardiomyopathy diagnosed in national referral centers for cardiomyopathies in Italy [3]. Since data on pediatric cardiomyopathy from our part of the world is still limited, this study aimed at describing the clinical profile and outcomes of pediatric hypertrophic cardiomyopathy in our population as well as to analyse the predictors of outcome.

The study enrolled all patients with cardiomyopathy who presented to us between 1990 to 2020 and were younger than 18 yrs. After taking informed consent, we reviewed the available medical records of each enrolled patient. This retrospective cohort of children was prospectively followed up to study their outcome and the possible predictors of outcome. Exclusion criteria were congenital heart defects not associated with malformation syndromes, abnormal ventricular size or function ascribed to intense physical training, chronic hypertension and maternal gestational diabetes mellitus. Routine genotyping for sarcomeric protein mutations and Noonan syndrome was unavailable during most of the study period. After directly reviewing all available cardiac information, we assigned each patient to a diagnostic category according to the phenotypic characteristics following the European Society of Cardiology Classification [4]. Cardiomyopathies were defined as abnormalities of the ventricular myocardium unexplained by abnormal loading conditions or congenital heart disease [5]. Hypertrophic cardiomyopathy was characterized by otherwise unexplained septal hypertrophy, left ventricular free-wall hypertrophy, or both (wall thickness more than 2 SD above the normal of mean for BSA) [6]. The need for ICD was based on one or more of these major risk factors: Family history of HCM and/or SCD, NSVT on an ambulatory monitor, massive LVH, or unexplained syncope [7]. Right ventricular involvement was defined by a right ventricular free wall thickness > 4 mm in the absence of pulmonary valve stenosis [8]. The morphology of left ventricular involvement was classified as either asymmetric septal hypertrophy if the interventricular septum was predominantly involved or concentric left ventricular hypertrophy if both the left ventricular free wall and interventricular septum were affected to a similar degree [1]. Left ventricular outflow tract obstruction was defined as measured resting or provoked systolic peak instantaneous gradient of > 30 mm Hg on echocardiography [7]. Noonan syndrome was diagnosed if characteristic phenotypic features were identified on clinical examination. Familial hypertrophic cardiomyopathy was defined as an affected first- or second-degree relative in the absence of Noonan syndrome, a mitochondrial disorder, or metabolic condition [1]. Massive left ventricular hypertrophy (LVH) was defined for patients younger than 12 years as the left ventricular maximal wall thickness of more than 20 SD; the adult cutoff value of 30 mm was used for all participants older than 12 years [7]. Massive LVH have been variously defined in pediatric HCM literature with some studies considering z score of more than 6 as massive LVH and others defining it as that more than 20. The end-stage phase of HCM was defined as an LV ejection fraction less than 50% [9].

Collected data include demographic descriptors, relevant history, examination findings, and laboratory data. Patients were considered on follow-up if they were seen in the out-patient department in the previous six months. If not seen within the last six months, patients or their guardians were reached out by telephonic interview. A single observer read the earliest available ECG and converted the measurements to age-appropriate Z scores [10]. QT interval (QTc) dispersion was calculated manually from a 12-lead ECG as the largest difference in QTc intervals between leads. Echocardiographic measurements of left ventricular dimensions, diastolic free wall, and septal thickness were expressed as Z scores based on body surface area [11]. The outcome parameters to be studied were Age at Death, Circumstances of death (SCD, Death from Ventricular Arrhythmia, or Death from Heart failure), AV block, BBB, and Requirement of ICD/transplant.

During the thirty-year study period (1990–2020), we identified 233 cases of pediatric cardiomyopathy, out of which sixty-three subjects had hypertrophic cardiomyopathy (27 percent). The demographic characteristics of the patients are described in Table 1. The median age of presentation of HCM patients was 7 yrs (Range 0.1–18 yrs).

This is the most extensive study on pediatric cardiomyopathy from South Asia to our knowledge. We studied the clinical profile and outcomes of cardiomyopathy among children less than 18 years of age. Among the 233 cases identified, 63 cases had hypertrophic cardiomyopathy (27 percent). Although, classically, HCM was thought not to present in infancy, 12% of HCM patients presented in the neonatal period and 37% presented in the first year of life. Earlier studies did not specifically mention the percentage of cases diagnosed in the neonatal period. The median age of presentation of HCM patients in this study was seven years which is almost similar to the study by Lipschultz et al., who also reported a high incidence of HCM in infancy, and the median age of presentation of HCM patients was reported to be 5.9 years [12]. The high percentage of HCM cases presenting in infancy in our study may represent metabolic causes of HCM. Even though Nugent et al. excluded patients with multisystem metabolic causes of cardiomyopathy, they found more than 50% of HCM patients diagnosed in infancy and the median age of presentation of HCM patients to be unusually early, around 5.7 months [6]. This may represent the cases that have been diagnosed as a result of screening. The inclusion of metabolic causes of HCM may have underestimated survival as IEM is associated with worse survival among all etiological subtypes.

Among the 63 HCM patients, only sixteen patients had proven syndromic, metabolic, or genetic disease (25%). Nugent et al. found an underlying syndromal, genetic, or metabolic condition in 57.5% of subjects [1]. This may be because testing was not performed in most of the subjects because of non-availability or financial issues. In our study, LV outflow obstruction was present in 30 patients (47%), almost similar to the pediatric studies conducted so far. In the study by Nugent et al. [1], 40% of patients had LV outflow obstruction, and in the study by Yetman et al. [2], 59% had LV outflow obstruction. In adult HCM literature, LVOTO has been described in 75% either at rest or provocation [13]. Whether LVOT obstruction should be considered as a risk factor for sudden cardiac death remains debatable with some studies reporting that it is associated with a higher risk for SCD while as others have observed that it has a high negative predictive value for SCD [14, 15]. Only four subjects progressed to the burned-out phase in this series. Interestingly, three out of the four had non-obstructive HCM. This observation is similar to that found in other studies. Biagini et al. did not find LVOT obstruction in any of his cohort of burned out HCM [16]. The youngest age at which end stage phase was recognized was 8 yrs (Table 4). Though classically unfavorable adverse remodeling occurs after decades, it has been documented in patients as young as 5 yrs [9, 16].

One of our subjects with infantile-onset of obstructive HCM had infective endocarditis of the mitral valve at eight years of age. Fortunately, he responded to medical treatment. The patient had severe LA dilatation with an LA diameter z score of 5. Infective endocarditis is rare in adults with HCM and is limited to case reports in pediatric HCM [17]. Spirinto et al. found that endocarditis in HCM is virtually confined to the mitral valve and patients with outflow obstruction and is more common in those with both obstruction and atrial dilatation [18]. However, subsequent studies found similar rates of mitral and aortic valve involvement regardless of the presence of left ventricular outflow tract obstruction [19].

Noonan syndrome was present in 9 of the 63 patients with HCM (14%). Out of the nine patients, three presented in infancy (33%). Six of the nine patients were females. Two patients had associated pulmonary stenosis, and one underwent pulmonary balloon valvuloplasty for severe valvular pulmonary stenosis. The outcome was known in 5 out of the nine patients, and all are alive over a mean follow-up of 7 years. One underwent septal myectomy and is doing well after 24 years of follow-up. We have not compared the HCM patients with NS with those without, as our numbers are small. Wilkinson et al. compared data in 74 children with NS and HCM and 792 with idiopathic or familial isolated HCM [20]. Children with NS were diagnosed with HCM before six months old more often (51%) than children with HCM without NS (28%) and were more likely to present with congestive heart failure (24% vs. 9%). Patients with NS with HCM have a worse risk profile at presentation than other children with HCM, resulting in significant early mortality (22% at one year). One patient with MYBPC3 mutation-positive was a 15-year-old boy, a case of myoclonus dystonia with a family history of cardiomyopathy and sudden cardiac death who had diffuse subendocardial enhancement on C MRI. Myoclonus dystonia has never been previously described in MYBPC3 positive HCM patients. Preexcitation was present in two of our patients. First-degree AV block was present in 15% of HCM patients. Only three patients had sustained ventricular arrhythmias. Out of the 17 patients with LGE, four patients had LGE extent > 15%.

Over a mean follow-up period of 5.6 years (0.1–30 years), twenty-one were lost to follow-up (33%). Among the patients whose outcome was known, eleven died (26%), and thirty-one (73%) were alive. This gives 5-year survival rates of 82% and 10-year survival of 78%, which is comparable to other studies. Nugent et al. reported a 5-year survival of 83% [1], while Maruizi et al. reported a 5 yr survival of 95% [3]. Maruizi et al. excluded both syndromal and metabolic causes of HCM in their study, which might be the reason for improved survival in that cohort. Among the eleven patients who died, seven faced sudden cardiac death, three died from heart failure, and one died after recurrent ventricular arrhythmias. Sudden cardiac death has been reported to be the most common cause of death in adult HCM literature as well [21]. We acknowledge that fewer patients underwent SM than in other studies, despite 23 having a gradient > 50 mmHg on echocardiography, probably because most were not severely symptomatic after GDMT. Only one child was NYHA FC III despite GDMT and had a gradient of 100 mmHg on echocardiography and 60 mmHg on cardiac cath but refused invasive treatment. Although our numbers are small, we tried to analyze various postulated predictors of death or ventricular arrhythmias. None of the variables studied had a significant association with death or ventricular arrhythmias (Table 7). This may be due to the high percentage of patients lost to follow-up. Only Consanguinity showed a trend towards higher events of death or ventricular arrhythmias (P-value 0.08), though it did not reach statistical significance. This may reflect underlying inborn errors of metabolism which may have gone unrecognized. Norrish et al. [22] also noted that among the pediatric HCM patients, infantile onset and inborn errors of metabolism had the worst survival among all other etiological groups. Maurizi et al. [3] found that limiting symptoms at diagnosis and Troponin I and T gene mutations were associated with higher risk of lethal arrhythmic events. Interestingly, in his study, likelihood of lethal arrhythmic events was poorly related to classic adult risk factors like extreme LVH and syncope.

The study's primary limitation is the high attrition rate, which is not surprising in a retrospective study conducted over three decades. The difficulty of contacting patients who visited the hospital over the past three decades is obvious. Another limitation of the study is the lesser number of genetic and metabolic abnormalities identified as the testing was not available for most of the study period, and even now, few parents can afford the costs involved. Though it is a single-center study, it may be close to a population-based study as it is from a major referral center of the region. We also acknowledge that this cohort comprises comparatively lesser number of HCM patients than other studies, albeit the study has been conducted over three decades. This could be due to lack of systematic family screening in the early periods of the study.

More than one-third of our HCM cohort presented in infancy. LV outflow tract obstruction is common (47%). Mid myocardial enhancement was the most common pattern of late gadolinium enhancement. SCD was the most common cause of death. The outcome in our HCM cohort is good and similar to other population cohorts. Only Consanguinity showed a trend towards higher events of death or ventricular arrhythmias, though it did not reach statistical significance.