Patients with obstructive hypertrophic cardiomyopathy (HC) are at increased risk of death or severe symptoms. Although metabolic exercise testing is routinely used in the evaluation of a variety of cardiac diseases, data on the prognostic information that can be derived from such testing in HC is lacking. We examined 182 patients (mean age 53 ± 15 years; 65% men) with obstructive HC and minimal or no cardiovascular symptoms. Each patient underwent maximum exercise testing with simultaneous metabolic gas exchange measurement. Follow-up (mean 4.0 ± 3.0 years; 100% complete) was performed to determine the vital status and occurrence of severe symptoms, defined as class III or IV symptoms of dyspnea or angina. Multiple parameters of metabolic exercise were associated with a risk of death and the development of severe symptoms. On multivariate analyses, the independent predictors of death and severe symptoms were the severity of the left ventricular outflow tract gradient at rest and the percentage of predicted peak myocardial oxygen consumption achieved during exercise. For patients with a percentage of predicted peak myocardial oxygen consumption of <60%, the 4-year survival rate free of death and severe symptoms was only 59%. In conclusion, among patients with obstructive HC and mild or no symptoms, a low metabolic exercise capacity is associated with an increased risk of death and the subsequent development of severe symptoms.
Metabolic exercise testing is an established clinical tool that is routinely used in the management of a variety of cardiac and pulmonary disorders. For patients with heart failure, metabolic parameters, such as exercise ventilation and peak myocardial oxygen consumption (VO 2 ), identify high-risk patients who warrant close clinical observation and treatment. Among patients with hypertrophic cardiomyopathy (HC), these parameters can also assist in the diagnosis of the disease in equivocal patients, facilitate the clinical assessment of functional capacity, and have been found to be reflective of the ventricular filling pressure. Nonetheless, data on the prognostic utility of metabolic exercise testing in HC are presently lacking. In the present study, we examined the prognostic utility of metabolic exercise testing for identifying patients with obstructive HC at increased risk of death or the development of severe cardiac symptoms.
Methods
The Mayo institutional review board approved the present study. Consecutive patients with HC seen at Mayo Clinic (Rochester, Minnesota) from November 1, 1991 to December 31, 2008 with the following criteria were enrolled in the present investigation: (1) age ≥18 years; (2) the presence of obstructive HC; (3) mild or no cardiac symptoms; (4) no immediate need for cardiac surgery that would lead to concomitant myectomy and relief of left ventricular outflow tract (LVOT) obstruction (e.g., endocarditis, aortic root surgery); (5) the absence of previous cardiac arrest; and (6) informed consent for study participation. For the definition of mild or no cardiac symptoms, the presence of both New York Heart Association functional class II or less and Canadian Cardiac Society angina class II or less were required, in the absence of syncope or other cardiac functional impairment. This entry date was selected based on the initial availability of metabolic exercise testing for routine patient evaluation at our institution. The diagnosis of HC was determined by typical clinical, electrocardiographic, and echocardiographic features, with ventricular myocardial hypertrophy occurring in the absence of any other cardiac or systemic disease that could have been responsible for the hypertrophy. The magnitude of myocardial hypertrophy was assessed with M-mode and 2-dimensional transthoracic echocardiography using standard techniques. Patients with a peak LVOT velocity of ≤30 mm Hg at rest underwent provocation with either Valsalva strain or amyl nitrite inhalation per standard protocol in the echocardiographic laboratory. LVOT obstruction was defined as a velocity of >30 mm Hg at rest or >50 mm Hg with provocation using Doppler echocardiography. In all cases, particular consideration was given to distinguishing the Doppler signal of LVOT obstruction from that of mitral regurgitation.
Each patient underwent maximum, symptom-limited exercise testing with gas exchange measurement (Medical Graphics, St. Paul, Minnesota) under close supervision by experienced personnel. Blood pressure measurements by sphygmomanometry and standard 12-lead electrocardiography were performed with the patient at rest, at 1-minute intervals, and for ≥6 minutes in the recovery period. Breath-by-breath analysis was used to determine the minute ventilation (V E ), VO 2 , and carbon dioxide output (VCO 2 ). These variables were averaged over 30 seconds, with rolling averages examined every 10 seconds. The respiratory exchange ratio was calculated as the VCO 2 /VO 2 ratio. The anaerobic threshold was calculated using the V-slope method and expressed as a percentage of the peak VO 2 . An abnormal blood pressure response to exercise was defined as the occurrence of hypotension or a failure of the systolic blood pressure to increase by ≥20 mm Hg at peak exercise. In patients who underwent multiple exercise testing, only the data from the study performed at the initial evaluation were considered.
Vital status, cardiac symptoms, and the need for septal reduction therapy (e.g., surgical myectomy, percutaneous alcohol septal ablation) were ascertained by a follow-up evaluation consisting of telephone interview, mailed questionnaires, and interrogation of the Social Security Death Index. For deceased patients, procurement of death certificates and interviews with the next-of-kin were performed to determine the cause of death. Death from congestive heart failure was defined as death occurring in the context of long-standing cardiac decompensation with progression of the disease during the preceding year with the development of pulmonary edema or cardiogenic shock. Sudden cardiac death was defined as instantaneous and unexpected death with or without documented ventricular fibrillation within 1 hour after a witnessed collapse, in patients who previously were in stable clinical condition, nocturnal death with no antecedent history of worsening symptoms, or appropriate discharge of an implanted internal cardioverter-defibrillator device for treatment of a lethal arrhythmia (i.e., sustained ventricular tachycardia or fibrillation).
The primary end point of interest was the occurrence of death or severe symptoms (New York Heart Association class III or greater or Canadian Cardiac Society angina class III or greater). The survival estimates with 95% confidence intervals were calculated using the Kaplan-Meier method. For purposes of data analysis, the patients were grouped according to the severity of the LVOT gradient (<60 vs ≥60 mm Hg), peak VO 2 (<18 vs ≥18 ml/min/kg), V E /VCO 2 (≤33 vs >33), and percentage of predicted peak VO 2 (<60%, 60% to 80%, and >80%). The categories for LVOT gradient severity were chosen on the basis of previous studies that demonstrated the prognostic utility of these cutoffs in minimally symptomatic or asymptomatic obstructive HC. The categories of the percentage of predicted peak VO 2 were chosen on the basis of previous studies of metabolic exercise in HC and other heart failure studies.
Univariate models were constructed to determine the variables associated with the end point of death and the occurrence of severe symptoms. The variables considered were age, male gender, coronary artery disease, atrial fibrillation, previous stroke, diabetes mellitus, hypertension, LVOT gradient at rest, maximum left ventricular wall thickness, β-receptor antagonist use, left atrial volume index, posterior wall thickness, left ventricular end-diastolic dimension, left ventricular end-systolic dimension, peak heart rate, peak systolic blood pressure, abnormal blood pressure response to exercise, V E /VCO 2 , peak VO 2 , respiratory exchange ratio, anaerobic threshold, electrocardiographic evidence of ischemia, and percentage of predicted peak VO 2 . Variables that were significant at the 0.05 level were then considered for a multivariate model. Stepwise techniques were used to identify which variables were independently associated with each end point and thus incorporated into a final multivariate model. Comparisons of continuous variables were made with the appropriate 2-sample test: a 2-sample t test in cases in which the variable distributions were symmetric, and a Wilcoxon rank sum test otherwise. The variables are reported as the mean ± SD.
All authors had full access to, and take full responsibility for, the integrity of the data, and all have read and agreed to the report as written.
Results
The baseline characteristics of the study population are listed in Table 1 . Mild dyspnea or angina was present in a slight majority of patients (n = 99 [54%]), and 83 patients (46%) were completely asymptomatic. LVOT obstruction at rest was present in 96 patients (53%), and latent in 86 patients. A total of 103 patients (57%) were receiving medical therapy with 1 or a combination of negative inotropic agents.
| Variable | Value |
|---|---|
| Age (years) | 53 ± 15 |
| Men | 119 (65%) |
| New York Heart Association class | |
| I | 92 (51%) |
| II | 90 (49%) |
| Canadian Cardiac Society angina class | |
| I | 132 (73%) |
| II | 50 (37%) |
| End-diastolic diameter (mm) | 44.9 ± 7.5 |
| End-systolic diameter (mm) | 25.1 ± 6.1 |
| Peak left ventricular outflow tract gradient | |
| At rest | 46.3 ± 38.5 |
| At ≥30 mm Hg | 96 (53%) |
| Maximal septal thickness (mm) | 19.9 ± 5.2 |
| Posterior wall thickness (mm) | 13.3 ± 2.8 |
| Left atrial volume index (ml/m 2 ) | 46.6 ± 8.1 |
| Family history of hypertrophic cardiomyopathy | 43 (24%) |
| Family history of sudden death from hypertrophic cardiomyopathy | 21 (12%) |
| Atrial fibrillation | 24 (13%) |
| Previous stroke | 5 (3%) |
| Hypertension | 49 (27%) |
| Diabetes mellitus | 6 (3%) |
| Coronary artery disease | 7 (4%) |
| Medications | |
| β-Receptor antagonist | 89 (49%) |
| Calcium channel blocker | 41 (23%) |
| Disopyramide | 15 (8%) |
| Amiodarone | 9 (5%) |
A treadmill ramp protocol was used for nearly all patients (98%), and a cycle protocol was used for the remainder ( Table 2 ). A total of 59 patients (31%) had an abnormal blood pressure response to exercise. Electrocardiographic evidence of myocardial ischemia (≥1 mm horizontal ST-segment depression in leads V 5 or V 6 ) occurred in 5 patients, although 26 patients had significant baseline abnormalities (e.g., severe left ventricular hypertrophy with strain) that precluded the accurate detection of ischemia. For the entire study population, the peak VO 2 was 22.7 ± 7.6 ml/min/kg with a mean percentage of predicted of 75 ± 20%. The peak VO 2 was <60% of predicted in 45 patients (25%), 60% to 80% of predicted in 73 patients (40%), and >80% of predicted in 64 patients (35%).
| Variable | At Rest | Exercise |
|---|---|---|
| Heart rate (beats/min) | 71 ± 14 | 135 ± 27 |
| Systolic blood pressure (mm Hg) | 120 ± 18 | 153 ± 38 |
| Oxygen pulse (ml/min) | 5.1 ± 1.7 | 14.5 ± 4.3 |
| Ventilation (l/min) | 12.7 ± 3.6 | 69.4 ± 25.5 |
| Peak oxygen consumption (ml/min/kg) | — | 22.7 ± 7.6 |
| Percentage of predicted | — | 75 ± 21 |
| Anaerobic threshold (ml/min) | — | 1,360 ± 476 |
| Percentage of peak oxygen consumption | — | 70 ± 9 |
| Maximum ventilation, percentage of predicted | — | 51 ± 13 |
| Minute ventilation/carbon dioxide output | — | 31.9 ± 4.7 |
| Respiratory exchange ratio | — | 1.13 ± 0.12 |
| Metabolic equivalents | — | 6.8 ± 2.4 |
The mean follow-up for the study was 4.0 ± 3.2 years (99% complete). There were 18 deaths (10%), including 8 from noncardiac causes. Of the remaining patients, 2 had sudden cardiac death and 1 died from heart failure. Seven patients died from unknown etiologies in the absence of overt premortal causative conditions. In the entire population, 46 patients (25%) developed severe cardiac symptoms (New York Heart Association class III-IV dyspnea or Canadian Cardiac Society class III-IV angina). Septal reduction therapy for the onset of severe symptoms (surgical myectomy in 22; alcohol septal ablation in 3) was undertaken in 25 patients (14%). Overall, the 4-year survival rate free of death or severe symptoms was 79.9% (95% confidence interval 73.3% to 86.5%, Figure 1 ) .
Multiple metabolic exercise parameters, in addition to several echocardiographic variables, were predictive of death or the onset of severe symptoms ( Table 3 and Figures 2 and 3 ) . Of these predictors, only the severity of the LVOT gradient and the percentage of predicted VO 2 were independent on multivariate analyses.
