Children with heart disease are at risk for sudden death during exercise, yet decisions regarding sports participation are often based on resting data. Acceleration across the left ventricular outflow tract (LVOT) assessed on stress echocardiography may suggest a diagnosis of hypertrophic cardiomyopathy in patients in whom it is not otherwise obvious. However, the range of peak velocities across the LVOT in healthy youth is unknown. The aim of this study was to describe LVOT velocities with maximal exercise in this age group.
Subjects up to 18 years old were prospectively enrolled if they had normal results on resting echocardiography and were undergoing exercise testing for other reasons. Subjects with significant comorbidities, suspected cardiomyopathy, or family histories of cardiomyopathy were excluded. Peak LVOT velocities were measured in the upright position using continuous-wave Doppler immediately after maximal exercise.
Fifty subjects (mean age, 13.8 ± 2.8 years) were included. Twenty-eight (56%) were male, and 40 (80%) were Caucasian. The median peak LVOT velocity measured immediately after exercise was 2.5 m/sec (range, 1.3–5.9 m/sec). Sixteen subjects (32%) developed peak LVOT velocities of ≥3 m/sec. Twelve of the 16 (75%) with elevated velocities had a dynamic outflow tract Doppler pattern, of whom eight had evidence of intracavitary narrowing on two-dimensional echocardiography.
The development of significant exercise-induced LVOT velocities may be a normal physiologic finding in healthy youth. The measurement of LVOT velocities alone with maximal exercise may not help distinguish patients with hypertrophic cardiomyopathy from healthy children.
Stress echocardiography (SE) is presently used as a measure of assessment of cardiac physiology during activity with the potential for detecting subclinical or occult pathology. In the adult population, SE has been used to determine the severity of hypertrophic cardiomyopathy (HCM), and it has been used to help predict outcomes in this population. Recent American College of Cardiology and American Heart Association guidelines for the diagnosis and management of HCM have supported using SE to determine the presence of physiologically provocable left ventricular outflow tract (LVOT) obstruction in patients with symptoms during routine physical activities who do not manifest outflow obstruction at rest.
HCM is the most common cause of sudden cardiac death in young athletes in the United States. Unfortunately, in many cases, the disease is not diagnosed until autopsy. Screening for HCM in family members of affected patients or in children who have concerning symptoms with exercise often results in uncertainty about the diagnosis because of the lack of significant left ventricular outflow tract obstruction on resting echocardiography. If the development of significant LVOT velocities during or immediately after maximal exercise is pathologic, SE could then be used as a screening tool to identify patients with milder forms of HCM. However, normal values for healthy pediatric patients without suspected disease have not been reported. With this in mind, we sought to describe peak LVOT velocities immediately after maximal exercise in a cohort of healthy youth.
The internal review board at The Children’s Hospital of Philadelphia approved the study, and all study participants gave informed consent (if they were 18 years of age or older) or assent (if they were <18 years of age). For subjects <18 years of age, a parent or legal guardian was approached to obtain informed consent. Pediatric subjects, who were undergoing exercise stress testing as part of their cardiology outpatient evaluation, were approached in the Exercise Physiology Laboratory at The Children’s Hospital of Philadelphia between October 2011 and August 2012 for enrollment. The majority of subjects were referred for exercise testing to evaluate the conduction system or to look for arrhythmias ( Figure 1 ). Patients were included if they were in the age range of 8 to 18 years, completed maximal exercise tests (defined by a respiratory exchange ratio [RER] ≥ 1.1 and/or maximum heart rate ≥ 85% of predicted), and had normal results on resting echocardiography at The Children’s Hospital of Philadelphia within 9 months of their exercise stress tests. Normal resting echocardiographic results were defined as a structurally normal heart with normal ventricular performance and normal left ventricular wall measurements ( Z score < 2). Exclusion criteria were syncope on exertion, family history of cardiomyopathy, family history of sudden cardiac death in a first-degree relative (age < 40 years), use of a daily β-blocker or calcium channel blocker, presence or suspicion of a myopathy, diagnosis of a metabolic disorder, hypertension requiring medication, diabetes mellitus, and sickle cell disease.
We theorized that pubertal issues such as hormonal alterations and increased muscle mass may have an effect on exercise-induced LVOT velocities in the general pediatric population. Thus, once enrolled, subjects were asked to complete a validated pubertal screening questionnaire to determine their stage of pubertal development (Tanner stage).
Exercise modality was chosen on the basis of the referring physician’s clinical question and included a cycle ergometer using a ramp cycle protocol or a 1-min incremental treadmill protocol, changing both speed and grade. Complete metabolic assessment of expired gases was performed on a breath-by-breath basis using a metabolic cart (Vmax 29; CareFusion, Yorba Linda, CA). Subject age, height, weight, blood pressure and heart rate at rest and at maximal exercise, and RER data were obtained from the exercise laboratory report.
Echocardiography with standard imaging planes was performed using a Philips iE33 ultrasound system (Philips Medical Systems, Andover, Massachusetts) coupled with an appropriately sized transducer. Three experienced sonographers acquired all of the images for this study (R.M., M.W., Y.W.). Continuous-wave (CW) Doppler was used to measure LVOT velocities from the apical four-chamber view with the cursor at the level just below the aortic valve. CW Doppler was chosen over pulsed-wave Doppler to measure the highest velocity jet immediately after exercise (healthy youth recover rapidly and have a rapid decrease in heart rate immediately after exercise). Moreover, CW Doppler was used to model our protocol after several other studies evaluating LVOT velocities in healthy adult subjects and adult subjects with HCM.
At rest, images were obtained in both the supine and upright positions. First, subjects laid in the left lateral decubitus position for assessment of the peak LVOT velocity by CW Doppler and evaluation of mitral regurgitation (MR), both in the apical view. Next, the parasternal long-axis view was obtained to evaluate for systolic anterior motion (SAM) of the mitral valve. Subjects were then asked to stand upright for approximately 1 to 2 min, after which the peak LVOT velocity (CW Doppler) was again measured in the apical view. SE was performed immediately after the maximal exercise test (the first image was acquired within 60–90 sec of exercise termination) with the subject upright either on the cycle ergometer or treadmill. The first measurement recorded was the peak LVOT velocity (CW Doppler) in the apical five-chamber view. Subjects were then transitioned immediately to an examination table adjacent to the exercise equipment for repeat measurement of the peak LVOT velocity in the left lateral decubitus position (CW Doppler) and for evaluation of MR, both done using the apical view. Finally, the parasternal long-axis view was obtained to assess for SAM.
We evaluated for MR both at rest and immediately after exercise to be sure we were not contaminating our measurement of the peak LVOT velocity by capturing the MR regurgitant jet. MR was graded subjectively, using color flow mapping in the apical four-chamber view, as none or trivial, mild, moderate, or severe.
Two echocardiographers (M.S.C., P.S.) independently interpreted each of the stress echocardiographic studies. The highest velocity was measured in each of the four positions (rest supine, rest upright, maximal exercise upright, and maximal exercise supine), and the average of the two observers’ data was used for analysis. Echocardiographic clips were preselected for each subject from each of the four positions listed above. Each reader was asked to interpret these clips on two separate occasions, approximately 3 months apart in time. The readers were blinded to their initial measurements and the measurements of the other reader. Interrater and intrarater reliability, the degree of agreement among raters, was assessed by comparing the upright velocities recorded immediately after exercise and measured by the two readers for 25 subjects.
Continuous variables are expressed as mean ± SD and qualitative variables as percentages. Interrater and intrarater reliability was evaluated using intraclass correlation coefficients and reported along with 95% confidence intervals (CIs). Medians and interquartile ranges (IQRs) for LVOT velocities and heart rate at rest and immediately after maximal exercise were constructed separately for supine and upright positions. The signed rank test was performed to determine if there was a significant difference in heart rate with positional change at rest. Wilcoxon’s rank-sum test was used to assess if there were significant differences in LVOT velocities induced by maximal exercise in the upright position between male and female subjects, between Caucasian and non-Caucasian subjects, between pubertal subjects and prepubertal subjects, and between subjects who exercised on the bike and those who exercised on the treadmill. Pearson’s or Spearman’s correlation coefficients between LVOT velocities measured immediately after exercise in the upright position and age, body mass index, body surface area, time nil per os, maximal oxygen consumption, peak heart rate, peak blood pressure, and Tanner stage were calculated and significant results reported. Finally, the median LVOT velocity and IQR of subjects with a dynamic outflow Doppler pattern after exercise and those without were calculated separately and Wilcoxon’s rank-sum test were used to evaluate if there was a difference between these two groups of subjects.
Fifty subjects met the inclusion criteria and consented to the study. Demographic and baseline characteristics for the 50 subjects evaluated with SE can be seen in Table 1 , and reasons for referral for exercise testing can be seen in Figure 1 . Twenty-eight subjects (56%) were male, with a mean age of 13.8 ± 2.8 years; 41 subjects (82%) had entered Tanner stages II to V of puberty. Forty-five (90%) subjects underwent baseline echocardiography within 1 month of the exercise test and five within 9 months of the exercise test. A graphical representation of the echocardiographic data can be seen in Figure 2 .
|Age (y), mean ± SD||13.8 ± 2.8|
|African American||6 (12%)|
|Asian/Pacific Islander||2 (4%)|
All subjects underwent resting electrocardiography as part of their routine outpatient evaluation. Significant findings included a borderline prolonged corrected QT interval ( n = 9), a short PR interval and a δ wave consistent with Wolff-Parkinson-White syndrome ( n = 5), premature ventricular contractions ( n = 6), voltage criteria for left ventricular hypertrophy ( n = 4), nonspecific T-wave changes ( n = 2), and supraventricular couplets ( n = 1). No subject with voltage criteria for left ventricular hypertrophy had abormal ST-segment or T-wave changes.
Exercise Testing Results
All subjects completed a maximal exercise test as defined above. The mean RER was 1.22 ± 0.09. RER data were not available for three subjects, and three subjects had RERs ≤ 1.1 (all of whom had maximum heart rates ≥ 85% of predicted). Of the 50 subjects, 37 (74%) performed the study on a cycle ergometer and 13 on a treadmill. No study was terminated because of dysrhythmia or patient symptoms. All subjects were in sinus rhythm with no ventricular ectopy during image acquisition. No subject developed significant ST-segment or T-wave changes, suggestive of ischemia, during the exercise stress test or during the recovery period. Forty-nine subjects (98%) had documentation of a normal blood pressure response to exercise. For one subject, it was reported that the peak blood pressure was difficult to auscultate because of background noise from the treadmill.
Resting Echocardiographic Measurements
The median LVOT velocity in the supine position before exercise was 1.3 m/sec (range, 1.0–1.9 m/sec). In the upright position, the median LVOT velocity was 1.2 m/sec (range, 0.8–1.7 m/sec). The median heart rate in the supine position was 68 beats/min (IQR, 59–78 beats/min), with a significant increase when the subject stood upright (median, 86 beats/min; IQR, 71–94 beats/min) ( P < .001). One subject had mild MR before exercise; the remainder had no or trivial MR. No subject had evidence of SAM.
Exercise Echocardiographic Measurements
The median decrease in heart rate from maximal exercise to the first image acquired was 13 beats/min (range, 0–35 beats/min). The median heart rate during image acquisition in the upright position was 183 beats/min (IQR, 176–193 beats/min). The median peak LVOT velocity measured immediately after maximal exercise in the upright position was 2.5 m/sec (range, 1.3–5.9 m/sec). Sixteen subjects (32%) developed elevated peak LVOT velocities (defined as ≥3 m/sec); six of these had peak LVOT velocities ≥ 4 m/sec. Twelve of the 16 subjects (75%) with elevated peak LVOT velocities had evidence of a dynamic outflow tract Doppler pattern (characterized by a late-peaking, dagger-shaped Doppler signal), of whom eight had evidence of intracavitary narrowing on two-dimensional echocardiography ( Figure 3 , Video 1 [available at www.onlinejase.com ]). Five of the six subjects with LVOT velocities ≥ 4 m/sec had both of these findings. Subjects with a dynamic outflow tract Doppler pattern immediately after maximal exercise had higher peak LVOT velocities than subjects without this pattern (median, 3.9 m/sec [IQR, 3.4–5.0 m/sec] vs 2.3 m/sec [IQR, 2.1–2.6 m/sec]; P < .001). No subject had more than trivial MR after exercise. One subject had mild MR at rest but had no MR after exercise. MR was not evaluated for in six subjects after maximal exercise. No subject had evidence of SAM immediately after maximal exercise.