Author + information
- Received September 11, 2002
- Revision received May 8, 2003
- Accepted May 13, 2003
- Published online September 3, 2003.
- Abdou Elhendy, MD, PhD*,
- Douglas W Mahoney, MSc†,
- Bijoy K Khandheria, MD, FACC*,
- Kelli Burger, BSc† and
- Patricia A Pellikka, MD, FACC*,* ()
- ↵*Reprint requests and correspondence:
Dr. Patricia A. Pellikka, Division of Cardiovascular Diseases, Mayo Clinic and Mayo Foundation, 200 1st Street SW, Rochester, Minnesota 55905, USA.
Objectives The goal of this research was to study the association between heart rate (HR) response to exercise and the risk of death and myocardial infarction (MI) after adjustment for left ventricular (LV) function and myocardial ischemia.
Background Chronotropic incompetence during exercise testing is associated with increased mortality. It is unknown whether LV dysfunction or ischemia accounts for this.
Methods We studied 3,221 patients (age 59 ± 12 years; 1,701 men) who underwent treadmill exercise echocardiography. We considered two markers of chronotropic incompetence: 1) failure to achieve 85% of the maximal predicted HR, and 2) low (<0.8) chronotropic index. The independent association between HR response and end points was evaluated by an adjusted risk (AR) model, which included clinical parameters, ejection fraction, and the severity of ischemic wall motion abnormalities.
Results Target HR was not achieved in 495 (15%) patients. Low chronotropic index was observed in 793 (25%) patients. There were 129 deaths (41 cardiac) during a median follow-up of 3.2 years. Myocardial infarction occurred in 65 patients. Low chronotropic index was associated with cardiac death (AR, 1.54; 95% confidence interval [CI], 1.18 to 2.04; p = 0.002) and MI (AR, 1.37; 95% CI, 1.09 to 1.69; p = 0.007). Failure to achieve 85% of maximal predicted HR was associated with increased mortality (AR, 1.49; 95% CI, 1.02 to 2.22; p = 0.04) and cardiac death (AR, 2.13; 95% CI, 1.10 to 4.17; p = 0.03).
Conclusions Impaired chronotropic response to exercise is associated with increased mortality and cardiac events even after adjusting for LV function and the severity of exercise-induced myocardial ischemia.
Chronotropic incompetence during an exercise stress test has been associated with increased mortality (1–7). Many mechanisms for this association have been suggested such as advanced age, a higher prevalence and severity of coronary artery disease (CAD), left ventricular (LV) dilation, modulation of autonomic tone, sinus or atrioventricular node dysfunction, and ischemia (8–12). Previous studies have shown that the increased mortality associated with chronotropic incompetence is persistent after adjusting for risk factors and ST-segment changes with exercise (9). Chronotropic incompetence is associated with LV dysfunction and myocardial ischemia (9–12). Left ventricular function and myocardial ischemia have been shown to be predictive of cardiac events. Therefore, it is not known whether the increased mortality in patients with chronotropic incompetence is persistent after adjustment for the severity of these abnormalities.
Risk stratification of patients with known or suspected CAD by an exercise stress imaging test requires careful recognition of the independent prognostic significance of exercise hemodynamic parameters as well as the extent of functional abnormalities. Exercise echocardiography is a well-established technique for the diagnosis and prognostic stratification of CAD. This technique allows evaluation of resting LV function and the extent of myocardial ischemia manifested as exercise-induced wall motion abnormalities (13–16). The aim of this study was to determine whether chronotropic incompetence is predictive of death and nonfatal myocardial infarction (MI), after adjustment for the severity of LV dysfunction and the extent of myocardial ischemia with exercise echocardiography.
Patients underwent a symptom-limited exercise echocardiography test, using the standard Bruce protocol at Mayo Clinic, Rochester Minnesota, between January 1990 to December 1995. Exclusion criteria were refusal to participate in research (145 patients), poor imaging quality (257 patients), beta-blocker therapy (1,201 patients), previous revascularization (622 patients), exercise-induced arrhythmias necessitating termination of the exercise stress test (31 patients), left bundle branch block (131 patients), atrial fibrillation (5 patients), and significant valvular heart disease (125 patients). Selection criteria were met in 3,362 patients. Follow-up was available in 3,221 (96%); these patients constituted the study population. The Institutional Review Board of the Mayo Foundation approved the study.
Exercise echocardiography protocol
Patients underwent symptom-limited treadmill exercise testing according to the Bruce protocol. Workload was expressed in metabolic equivalents (METS). Two-dimensional echocardiographic images were obtained from the parasternal and apical windows at rest and immediately after exercise. Both digitized and videotape-recorded images were used for interpretation of the studies (17). Ejection fraction at rest was measured using a previously validated modification of the method of Quinones et al. (18)or by visual estimation (19). Assessment of ejection fraction using this method has been validated in our laboratory (16). Regional wall motion was assessed semiquantitatively by an experienced echocardiographer, blinded to clinical information and unaware of the purpose of this study. Wall motion at rest and with exercise was scored 1 through 5 (1 = normal) according to a 16-segment model (20). Wall motion score index (WMSI) was determined at rest and peak exercise as the sum of the segmental scores divided by the number of visualized segments. The development of new or worsening wall motion was considered indicative of myocardial ischemia. A wall motion abnormality present at rest and unchanged with exercise was classified as “fixed.” Exercise echocardiography results were defined as abnormal if there was ischemia or fixed wall motion abnormalities (21). The exercise electrocardiogram (ECG) was considered positive for ischemia if there was horizontal or down-sloping ST-segment depression of ≥1 mm at 80 ms after the J point, nondiagnostic if the baseline ST-segment was abnormal, and negative for ischemia in the absence of these criteria.
Chronotropic incompetence was assessed as failure to achieve 85% of the age-predicted heart rate (HR). The age-predicted maximum HR was determined as 220 − the patient's age. Because of the possible confounding effects of age, physical fitness, and resting HR on this method, chronotropic response was also assessed by calculating the ratio of HR reserve expended to metabolic reserve expended at peak exercise as previously described (9,10). For any given stage of exercise, the percent metabolic reserve expended is:
In an analogous fashion, the percent HR reserve expended is:
In a group of healthy adults, the ratio of percent HR reserve expended to percent metabolic reserve expended during exercise was approximately one (95% confidence interval [CI], 0.8 to 1.3). Thus, chronotropic incompetence can be defined as a percent HR reserve expended to percent metabolic reserve expended ratio of <0.8; this will be referred to as a low chronotropic index. We considered the ratio of HR reserve expended to metabolic reserve expended at peak exercise, when, by definition, the proportion of metabolic reserve expended has a value of one. Using this approach, the chronotropic index was based entirely on directly measured variables, namely resting HR, peak HR, and age (22,23)Chronotropic index <0.8 was considered indicative of chronotropic incompetence (9,10).
Follow-up and end points
Follow-up was obtained by mailed questionnaires and scripted telephone interviews. Events were verified by contacting the patients' primary physician and reviewing medical records and death certificates. The end points considered were total mortality and hard cardiac events defined as nonfatal MI and cardiac death. Sudden unexpected death occurring without another explanation was included as cardiac death. Myocardial infarction was defined according to usual clinical, ECG, and enzymatic criteria. Coronary revascularization procedures during the follow-up period were also noted. Patients who underwent revascularization were censored at the time of revascularization for the end point of hard cardiac events. For analysis of all-cause mortality, patients were not censored if they had revascularization or MI before death.
Continuous variables were reported as mean ± SD, and comparisons between groups were based on the Wilcoxon rank-sum test. Categorical variables are summarized as percentages, and group comparisons were based on the chi-square test. Survival free of the end point of interest was estimated by the Kaplan-Meier method. Univariable and multivariable association of clinical and exercise echocardiographic variables with the end points were assessed in the Cox proportional hazards framework. Variables were selected in a stepwise forward selection manner with entry and retention set at a significance level of 0.05. The results of these analyses were summarized as risk ratios with corresponding 95% CI. To assess the association between exercise HR parameters and outcome, independent of other clinical and echocardiographic parameters, a two-step modeling approach was used. In the first step, clinical, exercise stress variables (excluding the HR parameters), and echocardiographic parameters were fit in a multivariate analysis model. In the next step, exercise HR parameters were added to the model to obtain adjusted risk (AR) ratios and to evaluate potential interaction effects between HR parameters with variables obtained in the first step. The proportional hazards assumptions were evaluated using previously described methods (24)and were not rejected for the variables in the final modeling step. All analyses were performed using SAS Version 8 (SAS Institute, Cary, North Carolina).
The mean age was 59 ± 12 years (range 20 to 88 years). There were 1,701 (53%) men. A total of 187 (6%) patients had previous MI. Hypertension was present in 1,203 (37%) patients, hypercholesterolemia in 1,679 (52%) patients, and 1,567 (49%) patients were smokers. Reasons for termination of the exercise stress test were fatigue in 2,001 (62%) patients, dyspnea in 884 (27%) patients, angina in 123 (4%) patients, ST-segment changes in 65 (2%) patients, and leg distress in 300 (9%) patients.
A total of 495 (15%) patients failed to achieve the target HR, and 793 (25%) patients had a low chronotropic index. The mean percentage of the age-predicted maximum HR that was achieved was 96 ± 11%, and the mean chronotropic index was 0.93 ± 0.23. Table 1demonstrates clinical data in patients that did and did not reach 85% of the age-predicted HR. Hemodynamic data are presented in Table 2.
Resting wall motion abnormalities were detected in 760 (24%) patients. Ischemia (new or worsening wall motion abnormalities with exercise) was detected in 789 (25%) patients. Of these patients, 388 (49%) had resting wall motion abnormalities as well. The exercise echocardiogram was considered abnormal (rest and/or exercise-induced wall motion abnormalities) in 1,161 (36%) patients. Echocardiographic data in patients that did and did not reach 85% of the age-predicted HR are shown in Table 2. Among patients with ischemia by exercise echocardiography, ischemic ECG changes were detected in 77 of 168 (46%) patients with chronotropic incompetence and 225 of 621 (36%) patients without chronotropic incompetence.
Association of HR response with wall motion abnormalities
In a multivariate analysis model of clinical variables, parameters independently associated with a larger percentage of ischemic segments were age (p = 0.0001), male gender (p = 0.001), diabetes mellitus (p =0.006), prior MI (p = 0.0001), and the chronotropic index (p = 0.002). Parameters independently associated with a larger percentage of segments with abnormal wall motion at rest were age (p = 0.0001), male gender (p = 0.0001), diabetes mellitus (p =0.006), hypertension (p =0.009), prior MI (p = 0.0001), and chronotropic index (p = 0.01).
A total of 129 deaths occurred during a median follow-up of 3.2 years. Deaths were from cardiac causes in 41 (32%) patients; 174 patients underwent myocardial revascularization before other events. Revascularization was more frequently performed in patients who did not achieve 85% of the age-predicted HR (13% vs. 4%, p = 0.001). Sixty-five patients had MI during follow-up. Among these patients, 21 underwent revascularization before the occurrence of MI. There were a total of 80 hard cardiac events (cardiac death or MI) without intervening revascularization.
Both total mortality and hard cardiac events occurred more frequently in patients who did not achieve 85% of the age-predicted HR. This difference was evident in the presence and absence of exercise wall motion abnormalities (Figs. 1 and 2). ⇓⇓
Independent predictors of outcome
In the first step of the multivariate analysis model, before consideration of chronotropic response variables, independent predictors of all-cause mortality were age (chi-square = 7.8, p = 0.005), male gender (chi-square = 11.3, p = 0.0008), workload (chi-square = 29.6, p = 0.0001), and exercise WMSI (chi-square = 20.3, p = 0.0001). Independent predictors of hard cardiac events were age (chi-square = 3.3, p = 0.07), male gender (chi-square = 2.8, p = 0.09), smoking (chi-square = 7.7, p = 0.006), history of MI (chi-square = 7.1, p = 0.008), workload (chi-square = 6.3, p = 0.01), resting ejection fraction (chi-square = 14.2, p = 0.0002), difference between exercise and rest WMSI (chi-square = 5.7, p = 0.02). When hard cardiac events were studied by type, independent predictors of cardiac death were history of MI (chi-square = 4.6, p = 0.03), workload (chi-square = 11, p = 0.0009), resting ejection fraction (chi-square = 6.4, p = 0.01), and the presence of exercise wall motion abnormalities in multivessel distribution (chi-square = 6.7, p = 0.01). Independent predictors of nonfatal MI were age (chi-square = 13, p = 0.0002), smoking (chi-square = 7, p = 0.008), history of MI (chi-square = 7, p = 0.0006), and exercise WMSI (chi-square = 7, p = 0.008).
Table 3demonstrates the absolute risk and the AR of mortality, cardiac death, and MI associated with various measures of chronotropic incompetence. Parameters considered in the AR model for each end point were those that were independent predictors of that particular end point. This analysis represents the second step in the multivariate analysis. One or more of the parameters of the chronotropic response to exercise was incremental to clinical, exercise test, and echocardiographic parameters in the prediction of each end point of interest. We found no significant interactions between each of the parameters of chronotropic incompetence with the variables in the first step or with calcium channel blocker therapy with regard to prediction of hard cardiac events.
In this study, we assessed the independent prognostic significance of impairment of HR response to exercise in 3,221 patients with known or suspected CAD who underwent a symptom-limited exercise stress echocardiogram. We considered two measures of chronotropic incompetence: inability to achieve 85% of the maximal HR predicted for age or a chronotropic index <0.8. A total of 15% of patients failed to achieve the target HR, and 25% had a low chronotropic index. Chronotropic incompetence was associated with the major risk factors of CAD and with a higher prevalence and severity of resting LV dysfunction and myocardial ischemia, demonstrated as transient wall motion abnormalities. The association between chronotropic incompetence and the severity of resting and exercise-induced wall motion abnormalities persisted after controlling for other clinical risk factors and exercise ECG parameters.
Association of chronotropic incompetence with prognosis
During a median follow-up period of 3.2 years, there were 129 deaths, of which 41 were due to cardiac causes. Myocardial infarction occurred in 65 patients. Patients with chronotropic incompetence had a higher incidence of all-cause mortality, cardiac death, and nonfatal MI. The risk of death and MI associated with chronotropic incompetence persisted after controlling for clinical characteristics, exercise tolerance, the severity of LV dysfunction at rest, and the extent of exercise-induced myocardial ischemia. In a multivariate model, predictors of all-cause mortality were an older age, male gender, lower workload, and exercise WMSI. The addition of parameters of HR response to that model demonstrated that failure to achieve 85% of the predicted HR imposed a significant risk of death, additional to those identified by clinical, stress test, and echocardiographic parameters. When cardiac death was studied as a separate end point, independent predictors were age, male gender, smoking, workload, resting ejection fraction, and the presence of wall motion abnormalities in multivessel distribution. A lower chronotropic index and a low exercise HR were predictive of the occurrence of cardiac death and MI after adjustment for clinical and echocardiographic parameters.
The use of 85% of predicted HR was superior to chronotropic index as an independent predictor of all-cause mortality; chronotropic index was superior in predicting cardiac events. Therefore, these parameters should be considered to provide complementary information.
Patients with chronotropic incompetence were more frequently treated with calcium channel blockers, which might have an effect on HR response to exercise. However, the association between chronotropic incompetence and the risk of death and MI persisted after controlling for clinical parameters, including medications.
Contribution of myocardial ischemia
Chronotropic incompetence has been associated with a higher prevalence of reversible perfusion abnormalities in patients undergoing exercise thallium-201 imaging as reported by Lauer et al. (9). In these patients, chronotropic incompetence predicted mortality independent of the association with reversible abnormalities. However, the impact of the severity of myocardial ischemia and of resting LV function was not investigated. Because patients with the most severe perfusion abnormalities are known to have the worst prognosis (25), it cannot be determined from the previous studies whether the increased risk associated with chronotropic incompetence was related to its association with the severity of myocardial ischemia.
A comprehensive risk assessment of patients undergoing stress-imaging techniques requires accurate evaluation of the extent and severity of inducible abnormalities as well as the total extent of fixed and inducible abnormalities (25). Therefore, additional risk assessment using other stress testing parameters should include variables that provide incremental prognostic information after accounting for the extent and severity of inducible abnormalities. In this study, we demonstrated that patients with chronotropic incompetence had a higher prevalence and severity of exercise-induced myocardial ischemia. The difference between rest and exercise WMSI was an independent variable associated with increased cardiac event rate. However, in the multivariate analysis, chronotropic incompetence was associated with adverse outcome after adjustment for the severity of myocardial ischemia. Among patients with an abnormal exercise echocardiogram, impaired chronotropic response to exercise stratified the patients into high and low/intermediate risk groups. Therefore, the chronotropic response to exercise identifies patients in whom an abnormal exercise echocardiogram is an indicator of poor outcome.
One explanation for the association of chronotropic incompetence with adverse outcome after controlling for the presence and severity of myocardial ischemia is that some patients with CAD fail to demonstrate ischemia or have underestimation of extent of ischemia because of the low exercise HR. Low exercise HR may limit the sensitivity of the exercise echocardiogram.
Myocardial function and exercise-induced wall motion abnormalities
The independent association between chronotropic incompetence and mortality after adjustment for resting ejection fraction has not been previously reported. Lauer et al. (10)studied 231 patients who underwent exercise echocardiography and were followed for a mean of 41 months. The composite end point of death, nonfatal MI, unstable angina, and late myocardial revascularization occurred in 41 patients. Chronotropic incompetence was predictive of an adverse cardiovascular prognosis after adjusting for the presence of myocardial ischemia. The numbers of patients and events in their study were too small to demonstrate an association with the end point of death and nonfatal MI. Additionally, the impact of the severity of resting LV dysfunction and the extent of myocardial ischemia were not studied.
This study reports an intermediate-term follow-up after exercise echocardiography. Although the study demonstrated an independent association between HR response to exercise and outcome, it is not known whether this association is maintained at a longer term follow-up.
Clinical implications and conclusions
Impaired chronotropic response to exercise is associated with an increased risk of mortality and MI among patients with known or suspected CAD. This risk is persistent after adjusting for resting ejection fraction and the severity of exercise-induced myocardial ischemia. Therefore, interpretation of the results of an exercise stress echocardiogram should include the evaluation of the HR response to exercise, in addition to the severity of resting and exercise-induced wall motion abnormalities, as all these parameters have independent prognostic value.
- adjusted risk
- coronary artery disease
- confidence interval
- heart rate
- left ventricular
- metabolic equivalents
- myocardial infarction
- wall motion score index
- Received September 11, 2002.
- Revision received May 8, 2003.
- Accepted May 13, 2003.
- American College of Cardiology Foundation
- Lauer M.S,
- Okin P.M,
- Larson M.G,
- Evans J.C,
- Levy D
- Ellestad M.H
- Ellestad M.H,
- Wan M.K
- Lauer M.S,
- Pashkow F.J,
- Larson M.G,
- Levy D
- Lauer M.S,
- Mehta R,
- Pashkow F.J,
- Okin P.M,
- Lee K,
- Marwick T.H
- Arruda A.M,
- Das M.K,
- Roger V.L,
- Klarich K.W,
- Mahoney D.W,
- Pellikka P.A
- Dujardin K.S,
- Enriquez-Sarano M,
- Rossi A,
- Bailey K.R,
- Seward J.B
- Schiller N.B,
- Shah P.M,
- Crawford M,
- et al.
- Armstrong W.F,
- Pellikka P.A,
- Ryan T,
- Crouse L,
- Zoghbi W.A
- Fletcher G.F,
- Balady G,
- Froelicher V.F,
- Hartley L.H,
- Haskell W.L,
- Pollock M.L
- Grambsch P.M,
- Therneau T.M
- Iskander S,
- Iskandrian A.E