The trusted source for
healthcare information and
Synopsis: Analysis of five easily measurable variables can effectively predict sudden death in patients with hypertrophic cardiomyopathy. Patients with multiple risk factors may have a sudden death probability high enough to justify prophylactic therapy.
Source: Elliott PM, et al. J Am Coll Cardiol 2000;36: 2212-2218.
Elliott and colleagues at St. George’s Hospital in London report on predictors of outcome in a cohort of 368 patients with hypertrophic cardiomyopathy. The influence of five variables on survival was analyzed. These five clinical variables were: nonsustained ventricular tachycardia on a 48-hour ambulatory ECG, history of syncope, exercise blood pressure response, family history of sudden death, and left ventricular wall thickness. Patients in the study were between the ages of 14 and 65. Patients were excluded if they had a documented sustained ventricular arrhythmia or out-of-hospital cardiac arrest, or if they received amiodarone during more than 50% of their follow-up. Other forms of therapy including dual chamber pacing, calcium channel blockers, beta blockers, and other antiarrhythmic drugs were permitted. Follow-up data were collected using a questionnaire sent to all patients’ general practitioners. Additional information, if required, was obtained by direct communication with the patients. An abnormal blood pressure response was measured during either treadmill exercise testing or bicycle ergometry and was defined as a failure of the blood pressure to rise or an actual fall in blood pressure during exercise. Based on prior observations, blood pressure response was classified as a risk factor only in patients less than or equal to 40 years of age.
The final group included 368 patients of whom 239 were men. The mean age at diagnosis of hypertrophic cardiomyopathy was 33 ± 14 years. In this group, 57 patients (16%) had a history of syncope, 90 (25%) had a family history of sudden death, 138 (38%) had an abnormal blood pressure response with exercise, 64 (17%) had nonsustained ventricular tachycardia during Holter monitoring, and 44 (12%) had a left ventricular wall thickness of 30 mm or more. There was no relationship between the pattern of hypertrophy (asymmetric, concentric, apical or eccentric) and survival.
The mean follow-up was 3.6 ± 2.5 years. There were 36 (9.7%) deaths among the 368 patients. Twenty-two deaths (61%) were sudden, five (14%) were from progressive heart failure, four (11%) were from other cardiovascular causes, and five (14%) were from noncardiac causes. The mean age at death or transplantation was 39 ± 16 years. Of the twenty-two patients who died suddenly, 15 were younger than 40 years old.
All five risk factors analyzed were univariate predictors for mortality. However, there was a significant interaction between family history of sudden death and syncope. In the absence of the other factor, each of these factors was associated with a relative risk of less than one. When they were combined however, the univariate risk for the combined factor of family history of sudden death and syncope was 8.2. Multivariate survival analysis was also performed. The multivariate sudden death risk ratios for the four risk factors were: 1.8 for an abnormal exercise blood pressure response, 5.3 for a family history of sudden death and syncope, 1.9 for nonsustained ventricular tachycardia, and 2.9 for left ventricular wall thickness greater than 30 mm. When family history of sudden death and syncope were considered as a single risk factor, 203 (55%) patients had no risk factors. One, two, and three risk factors were seen in 122 (33%), 36 (10%), and 7 (2%) patients, respectively. No patient had four risk factors. If a patient had no risk factor, the estimated six-year survival rate without sudden death was 95%. The survival estimates for patients with one, two, and three risk factors were 93%, 82%, and 36%. Thus, the 43 patients with two or three risk factors had a relative risk of 5.6 for sudden death compared to the 325 patients with zero or one risk factor.
Elliott et al conclude that analysis of five easily measurable variables can effectively predict sudden death in patients with hypertrophic cardiomyopathy. The presence of two or more risk factors was associated with a 4-5% estimated annual sudden death risk. Therefore, patients with multiple risk factors may have a sudden death probability high enough to justify prophylactic therapy. Recommendations about the most appropriate prophylactic therapy, amiodarone, or implantable defibrillators, cannot be made from this study.
Comment by John P. DiMarco, MD, PhD
In this paper, Elliott et al describe easily identifiable clinical variables for sudden death in a large group of patients with hypertrophic cardiomyopathy. They identified a subgroup of 12% of the patients with two or more risk factors who had an annual sudden death risk of about 5%. Patients with none or only one risk factor had a much lower risk of sudden death.
Patients with hypertrophic cardiomyopathy are often difficult to manage. Even in the absence of significant hemodynamic symptoms, sudden death may occur. It is particularly tragic that sudden death is more common in young individuals who may have few other symptoms. Among prophylactic therapies, only amiodarone and defibrillator implantation are thought to be effective. Unfortunately, the value of invasive studies including electrophysiologic testing and hemodynamic measurements for predicting sudden death has been controversial and an easy method to identify a group that justifies these interventions has not been widely accepted. Elliot et al describe a fairly simple system to identify a group with a 5% annual sudden death risk and this value approaches the level at which the hazards and costs of possible prophylactic therapy would be acceptable.
A major limitation of this study is the fact that about half of all the patients seen with hypertrophic cardiomyopathy at Elliott et al’s institution were excluded for one reason or another. Many of these were excluded because of age or a history of amiodarone use. However, the data should apply to patients seen for initial diagnostic evaluation by clinical cardiologists. Data such as these could be used to initiate a comparative trial of interventions to prevent sudden death.