Cardiopulmonary Exercise Testing in Pediatric Patients With Hypertrophic Cardiomyopathy

Exercise stress testing (EST) is clinically useful in select pediatric patients with hypertrophic cardiomyopathy (HCM). Ectopy on EST is a risk factor for cardiac death, cardiac transplant, arrhythmias requiring implantable cardioverter-defibrillator

Background

Exercise stress testing (EST) in pediatric hypertrophic cardiomyopathy (HCM) patients has not well described in a large heterogenous cohort.

Objectives

The objective of the study was to determine the clinical utility of EST in pediatric HCM.

Methods

This was a retrospective single-center analysis of HCM patients younger than 21 years who had EST between January 1, 2000, and January 1, 2019. Clinical, demographic characteristics, and EST data were analyzed, using the last EST during the study or prior to the event in subjects with a primary outcome. The primary composite endpoint included cardiac death, transplant, or arrhythmia requiring implantable cardioverter-defibrillator placement. Outcome analysis was performed using Cox proportional hazard modeling.

Results

The study cohort included 140 patients, 52% with a recognized genetic variant. There were 2 tests aborted due to safety concerns (ST-segment changes, ventricular ectopy). The median age at first EST was 13.6 years. Ninety percent of patients were tested using cycle ergometry, and 44% were on a beta-blocker. The median peak oxygen consumption was 37.1 mL/kg/min (IQR: 12.5 mL/kg/min) or 81.2% predicted, the mean anaerobic threshold was 21.8 Ml (IQR: 8.3 mL), and the median peak power was 2.6 ± 1.1 W/kg or 73.7% predicted. Ectopy during EST was seen in 44% of patients, and 8% had an abnormal blood pressure response to exercise. The endpoint was reached in 12 patients. The presence of any degree of ectopy was a predictor of the composite endpoint (hazard ratio: 5.8; 95% CI: 1.3-26.7).

Conclusions

EST is clinically useful in select pediatric patients with HCM. Ectopy on EST is a risk factor for cardiac death, cardiac transplant, and arrhythmias requiring implantable cardioverter-defibrillator.

Introduction

Hypertrophic cardiomyopathy (HCM) is the most common inherited form of pediatric cardiomyopathy with an incidence of 0.47 per 100,000 children¹^,² and is characterized by a heterogenous collection of clinical phenotypes.³ While the majority of patients with HCM have a relatively benign course⁴, the disease can also be characterized by malignant arrythmias, progressive heart failure, and sudden cardiac death (SCD).⁵ As such, there have been several efforts to evaluate risk factors for mortality in children with HCM, with degree of hypertrophy, unexplained syncope, and nonsustained ventricular tachycardia all shown to be predictive of SCD.^6-9

In the adult population, there has been a recent focus on the utility of serial exercise stress testing (EST) in patients with HCM, with studies showing that subnormal measures of gas exchange, including maximal oxygen consumption (VO₂) and ventilatory inefficiency, are associated with reduced survival,^10-13 and the recently updated American Heart Association guidelines recommend EST as a component of the clinical evaluation.¹⁴ In children, however, EST has not been universally incorporated into the standard of care¹⁵, and with the exception of noting the negative prognostic implications of a failure in the ability to augment blood pressure response,⁷ the utility of EST in pediatric patients with HCM is largely unknown.

Thus, the purpose of this study is to describe EST in pediatric HCM patients performed at a single center over the course of an 18-year period, with a focus on exercise performance compared to normative values, changes in exercise performance with increasing age, and findings on EST predictive of adverse cardiac outcomes.

Methods

This study was performed with approval from the Children’s Hospital of Philadelphia Institutional Review Board. Written informed consent was waived owing to the retrospective nature of the study.

Cohort selection

This was a single-center retrospective observational cohort study. Participants were included in our study if they had a diagnosis of HCM, underwent EST at the Children's Hospital of Philadelphia between January 1, 2000, and January 1, 2019, and were younger than 21 years at the time of testing. The standard of care at our center is to refer all patients with a diagnosis of HCM for EST, unless they are physically unable to participate in testing. Potential cases were identified by querying the Exercise Physiology Laboratory Database for patients coded as HCM. Each participant's medical record was reviewed to confirm the diagnosis of HCM. All patients included in this study were classified as phenotype positive by their primary cardiologist. Patients who were described as possible HCM, or who were genotype positive/phenotype negative, were not included in the study cohort. Subjects meeting the primary endpoint (defined below) prior to their first EST were excluded.

Data collection

Demographic and clinical information was collected by manual review of the electronic medical record, including age at HCM diagnosis and first EST, history of genetic diagnosis, cardiac procedures and medications, arrhythmia leading to intervention, and SCD.

Data collected from review of EST reports included maximal VO₂ and VO₂ at the anaerobic threshold, heart rate response, blood pressure response, and presence of exercise-associated ectopy or ST-segment changes during EST. The presence of any ectopic beats, either atrial or ventricular, during warmup, exercise, or recovery was defined as ectopy. An abnormal blood pressure response (ABPR) was defined as the failure to augment the systolic blood pressure during exercise by at least 20 mm Hg. Data from echocardiograms, including left ventricular ejection fraction (EF), septal wall thickness, and left ventricular outflow tract obstruction, were collected from clinical reports. Imaging studies performed within 3 months prior to or 1 month after EST were included. For those with multiple tests, descriptive statistics were captured from either the most recent EST within the study window or the most recent EST prior to reaching the primary outcome.

Follow-up information through April 1, 2019, was included, as well as clinical history preceding initial EST. Data from both inpatient and outpatient clinical records were included.

Clinical outcome

The primary study endpoint was a composite outcome of SCD, aborted SCD, transplant, or ventricular arrhythmia prompting implantable cardioverter-defibrillator (ICD) placement, with each subject limited to a single outcome. If a patient had 2 outcomes separated in time (ie, an aborted SCD followed by transplant), the initial event was selected as the outcome. The components of the composite outcome were chosen based on their similar underlying biologic mechanisms. Patients with no follow-up visits after their first EST during the study period were defined as lost to follow-up.

Statistical analysis

Clinical and demographic characteristics were summarized using median (IQR) for continuous variables and percentage for categorical variables. EST data analysis was performed using the last EST completed within the study window or the last test prior to the event in subjects with a primary outcome. Percent predictive values were determined based on previously published normative values.¹⁶ Demographics and clinical characteristics were compared between those patients without end point reached vs those patients with end point using t-test, Mann-Whitney U test, chi-square test, and Fisher exact test. To describe the EST characteristic change across age, mixed-effects regression models with fractional polynomial functions were used adjusting for the presence of a maximum effort test, to describe either linear or curvilinear associations.¹⁷ A total of 164 models with the dimension of the fractional polynomial up to 3 (m = 3, chosen by the plotted data and sensitivity analysis up to 4 dimension) were tested for each EST characteristic, and the parsimonious polynomial models were determined by the function selection procedure in STATA (StataCorp LLC), that is, selecting the simpler model with the lowest Bayesian information criterion, which indicated a better fit than other models.¹⁸^,¹⁹ LOWESS (Locally Weighted Scatterplot Smoothing) plots of predicted mean and 95% CIs across age were presented with scatter plots.

To examine the demographic and EST characteristics associated with the primary outcome, separate Cox proportional hazards models were used for each characteristic. Separate models were used for EST variables due to collinearity. Selection of candidate factors associated with the primary outcome was determined based on clinical suspicion. Beta-blocker use was adjusted in the Cox model of peak heart rate. The assumption of proportional hazards over time was satisfied by testing the statistical significance of predictors by time interaction effects in the model and plotting the Schoenfeld residuals. Hazard ratios and 95% CIs are presented. Kaplan-Meier analysis was used to examine the survival rates stratified by the presence or absence of ectopy on EST. The log-rank test was used to test the difference between survival curves. Statistical analysis was performed using SAS version 9.4 (SAS Institute) and STATA version 15.

Results

Demographics and clinical characteristics

Over the 18 years studied, 140 patients with HCM who underwent at least 1 EST were identified. Seventy percent of the cohort was white (n = 98), 80% were male (n = 112), and slightly over one-half (52.1%, n = 73) had an identified pathogenic or potentially pathogenic mutation at the time of testing (Supplemental Table 1). The median age of HCM diagnosis was 12.0 years, and the median age at first EST was 13.6 years

Exercise performance in HCM

This study shows that, overall, cardiovascular fitness was impaired in this cohort of pediatric patients with HCM and that it declines with age. Peak VO₂ represents the maximum volume of oxygen consumed by the body and is a useful measure of cardiorespiratory fitness in a healthy child.¹⁰^,¹²^,¹³^,^21-23 Previous studies have demonstrated impaired exercise function in adults with HCM and have shown that reduced peak VO₂ is predictive of total mortality and progression to advanced heart failure or heart transplant. In our study population, peak VO₂ and power were reduced when compared to normative values, and the inverse correlation of peak VO₂ and age suggests these patients have worsening exercise function as they age. While several studies have shown that heart failure symptoms are predictive of adverse outcomes in children with HCM,¹^,²⁴ a recent risk prediction model for SCD in children with HCM did not include heart failure symptoms in their determination of risk.²⁵ This discrepancy may be due to the difficulty in eliciting and identifying heart failure symptoms in children, and EST may be an alternative method to identify previously unrecognized functional limitations. While impaired performance on EST was not predictive of adverse outcomes in our cohort, this may be because the study was underpowered for relatively rare outcomes and deserves further evaluation in a larger cohort. Aside from potential implications on outcomes, this finding is further concerning in light of the known correlation between reduced activity and psychosocial wellbeing²⁶^,²⁷ and the possibility that activity restrictions have come with the unintended consequence of increasing the cardiovascular risk profile in a population already with risk factors in place.^28-30

Risk factors identified by EST

Our study demonstrates that EST may provide information helpful for cardiac risk stratification for children with HCM. Ectopy of any kind, including isolated premature atrial or ventricular contractions, correlated with the composite outcome of SCD, transplant, or ventricular arrhythmia prompting ICD placement; 10 of 12 children who reached the primary outcome had ectopy on EST. While previous studies have shown nonsustained ventricular arrhythmia as a risk factor for SCD in young people with HCM⁶^,²⁵ and that, in adults, atrial fibrillation during EST predicts poor outcomes,³¹ our finding that exercise-induced atrial or ventricular ectopy is associated with outcomes is a novel finding and further highlights the utility of EST in this population. Left atrial (LA) size, a marker of impaired diastolic function, is a known risk factor for SCD in both children and adults. We hypothesize that atrial ectopy, which was found to be associated with adverse outcomes, may be more commonly seen in children with LA dilation and LA hypertension, a physiology which can be provoked during exercise.

It is also important to note that of the 80 patients who had no ectopy during testing, none of them died, had SCD, or were transplanted, and only 2 of that group ultimately had ventricular tachycardia which led to ICD placement. Thus, the absence of ectopy on EST may help in identifying a group of patients at lower risk.

These findings have several important clinical implications. First, they suggest that patients with ectopy on EST may deserve closer monitoring with more frequent outpatient follow-up and ambulatory rhythm monitoring. Second, they suggest that patients with no ectopy on EST may be a lower risk cohort and, therefore, might be reasonable candidates for a shared decision-making model for athletic participation. However, other clinical variables must also be integrated into risk stratification in these patients.

ABPR has previously been cited as a risk factor for SCD in children⁷; no such predictive association was seen in our study group. Decker et al⁷ used treadmill testing and the modified Bruce protocol, a technique that can make accurate assessment of blood pressure difficult. The majority of our cohort was tested using cycle ergometry with direct auscultation of blood pressure, which produces less artifact and, we believe, provides a more reliable result. The findings in this study further support the recently updated HCM guidelines which have removed ABPR as part of the routine evaluation of SCD risk.¹⁴

Further work evaluating the relationship between exercise capacity and symptoms, quality of life, and outcomes in children with HCM is needed. A better understanding could aid in the difficult process of forming exercise guidelines, which traditionally have taken a conservative approach despite consensus that exercise positively impacts cardiovascular and overall health and evidence that inactivity is prevalent in patients with HCM.²⁷^,³² If decreased aerobic capacity portends worse outcomes in these patients, the balance of risk vs benefit regarding exercise may change. Thus, EST may serve as an integral tool in the process of “shared decision-making” around exercise introduced in the most recent American Heart Association guidelines.¹⁴

Study Limitations

This study has limitations intrinsic to retrospective analysis, and the findings may differ from a prospectively enrolled cohort. Our study did not include a control group. Our center is a tertiary referral center that may see a greater portion of severe HCM, and some patients were referred from smaller centers and thus had limited follow-up. Given that the focus of this study was on exercise testing, not all conventional risk factors for SCD were evaluated. The composite outcome was rare, and the study may not be powered to detect all risk factors. As a consequence, a multivariable analysis of risk factors for the primary outcome could not be performed. While this study adds to the literature on the value of EST in children with HCM, limitations exist due to the small numbers of patients and events and the possible selection bias of patients undergoing EST.

Conclusions

Exercise capacity may be impaired in pediatric patients with HCM and can worsen with age. EST is a valuable clinical tool to assess exercise capacity in this population. Ectopy on EST is associated with the composite outcome of SCD, cardiac transplant, and arrhythmias requiring ICD. These findings suggest that EST is useful in the care of pediatric patient with HCM.