The purpose of this study was to evaluate the prognostic value of stress echocardiography in patients with angiographically significant coronary artery disease (CAD). Two hundred sixty patients (mean age 63 ± 10 years, 58% men) who underwent stress echocardiography (41% treadmill, 59% dobutamine) and coronary angiography within 3 months and without intervening coronary revascularization were evaluated. All patients had significant CAD as defined by coronary stenosis ≥70% in major epicardial vessels or branches (45% had single-vessel disease, and 55% had multivessel disease). The left ventricle was divided into 16 segments and scored on a 5-point scale of wall motion. Patients with abnormal results on stress echocardiography were defined as those with stress-induced ischemia (increase in wall motion score of ≥1 grade). Follow-up (3.1 ± 1.2 years) for nonfatal myocardial infarction (n = 23) and cardiac death (n = 6) was obtained. In patients with angiographically significant CAD, stress echocardiography effectively risk stratified normal (no ischemia, n = 91) in contrast to abnormal (ischemia, n = 169) groups for cardiac events (event rate 1.0%/year vs 4.9%/year, p = 0.01). Multivariate logistic regression analysis identified multivessel CAD (hazard ratio 2.53, 95% confidence interval 1.16 to 5.51, p = 0.02) and number of segments in which ischemia was present (hazard ratio 4.31, 95% confidence interval 1.29 to 14.38, p = 0.01) as predictors of cardiac events. A Cox proportional-hazards model for cardiac events showed small, significant incremental value of stress echocardiography over coronary angiography (p = 0.02) and the highest global chi-square value for both (p = 0.004). In conclusion, in patients with angiographically significant CAD, (1) normal results on stress echocardiography conferred a benign prognosis (event rate 1.0%/year), and (2) stress echocardiographic results (no ischemia vs ischemia) added incremental prognostic value to coronary angiographic results, and (3) stress echocardiography and coronary angiography together provided additive prognostic value, with the highest global chi-square value.
The prognostic value of stress echocardiography is routinely incorporated into clinical practice. Survival rates for patients with stable obstructive coronary artery disease (CAD) are correlated with the extent of disease. The objectives of the present stress echocardiographic study were twofold: (1) to define the prognostic value of stress echocardiography in patients with angiographic CAD and (2) to evaluate the incremental value of clinical findings, stress electrocardiography, stress echocardiography, and coronary angiography separately and together in predicting cardiac events.
Methods
We identified 260 nonconsecutive patients who were referred for exercise or pharmacologic stress echocardiography from March 21, 2000, to December 31, 2008. Successful follow-up (100%) for cardiac events ≥1 year after testing was obtained.
Maximal symptom-limited treadmill exercise (electrocardiographic [ECG]) testing was performed using a standard Bruce protocol. Patients exercised to general fatigue, with premature termination for severe angina, ventricular tachycardia, hemodynamically significant arrhythmias, or hemodynamic instability. Postexercise echocardiographic images were acquired <30 to 60 seconds after the termination of treadmill exercise. In pharmacologic stress, dobutamine was administered intravenously beginning at a dose of 5 to 10 μg/kg/min and increased by 10 μg/kg/min every 3 minutes to a maximum of 40 μg/kg/min or until a study end point was achieved. The end points for termination of the dobutamine infusion included the development of new segmental wall motion abnormalities, attainment of >85% of age-predicted maximum heart rate, or development of significant adverse effects related to dobutamine infusion.
The left ventricle was divided into 16 segments as recommended by the American Society of Echocardiography, and a score was assigned to each segment at baseline, with each stage of stress (dobutamine only) and during recovery. Each segment was scored as 1 = normal, 2 = mild to moderate hypokinesia (reduced wall thickening and excursion), 3 = severe hypokinesia (markedly reduced wall thickening and excursion), 4 = akinesia (no wall thickening and excursion), or 5 = dyskinesia (paradoxical wall motion away from the center of the left ventricle during systole). A normal response to stress was defined as normal wall motion at rest with an increase in wall thickening and excursion during stress. An abnormal (ischemic) response to stress was defined as (1) a left ventricular (LV) wall segment that did not increase in thickening and excursion during stress (lack of a hyperdynamic wall motion response) or (2) a deterioration in LV wall segment thickening and excursion during stress (increase in wall motion score of ≥1 grade) and (3) a biphasic response with dobutamine stress. Maximal severity was the score of the LV wall segments with the greatest value (worst wall motion grade) at postexercise stress (range 0 to 5). Peak wall motion score index after stress was derived from the cumulative sum score of 16 LV wall segments divided by the number of visualized segments. The rest ejection fraction used in the study analysis was an average visual estimation from 2 experienced echocardiographers.
All cardiac catheterizations were performed using a standard Judkins technique. Significant coronary stenosis was defined as a luminal diameter narrowing ≥70% in either a main epicardial artery or a major branch.
Follow-up was obtained in all patients by means of physician-directed telephone interviews using a standardized questionnaire. The hard end points of the study were nonfatal myocardial infarction or cardiac death. Nonfatal myocardial infarction was documented when diagnostic changes in cardiac enzymes (troponin) were accompanied by appropriate clinical symptoms, ECG findings, or both. Cardiac death was confirmed by review of hospital medical records, death certificate, and autopsy records when available.
All analyses were performed using commercially available statistical software (SPSS for Windows version 10.0.5; SPSS, Inc., Chicago, Illinois). Continuous variables are expressed as mean ± SD. Patient groups were compared using Student’s t tests. Differences in categorical variables among groups were assessed using chi-square analysis. Univariate analysis was performed to determine the relation between clinical and echocardiographic variables and cardiac events. Univariate variables that were predictive of cardiac events were considered in multivariate logistic regression analysis. Kaplan-Meier cumulative survival analysis with stratification by normal or abnormal stress echocardiographic results was performed. The comparison of survival between groups was made using the Mantel-Cox test. Statistical significance was defined as p <0.05.
A forward conditional (Wald) Cox proportional-hazards model with all assumptions tested was used to determine the incremental prognostic value of stress echocardiographic variables over clinical, stress ECG, and coronary angiographic variables. The stepwise selection or removal of variables for inclusion was based on clinical judgment and univariate statistical significance.
Results
In the study cohort of 260 patients, 107 (41%) underwent treadmill exercise and 153 (59%) underwent pharmacologic stress. The patient characteristics, stress echocardiographic, and coronary angiographic results are listed in Table 1 . Patients were followed for up to 5 years (mean 3.1 ± 1.2), and all patients were followed for ≥1 year. Among the study cohort of 260 patients, 29 cardiac events (11%) occurred during the follow-up period. These included 18 nonfatal myocardial infarctions (7%) and 11 cardiac deaths (4%). There were 7 cardiac events in patients who underwent treadmill stress and 22 events in those who underwent dobutamine stress (6.5%/year vs 14.4%/year, p = 0.11). Forty-eight patients underwent early coronary revascularization <60 days after stress echocardiography (30 percutaneous coronary interventions and 18 bypass procedures). There were no hard cardiac events in these patients up to 1 year after revascularization.
| Variable | Stress Echocardiographic Results | p Value | |
|---|---|---|---|
| Normal | Abnormal | ||
| (n = 91) | (n = 169) | ||
| Age (years) | 62 ± 11 | 65 ± 10 | 0.02 |
| Men | 46 (50%) | 106 (63%) | 0.05 |
| Abnormal rest ECG results | 55 (60%) | 84 (50%) | 0.10 |
| Previous myocardial infarction | 35 (38%) | 56 (33%) | 0.39 |
| Previous percutaneous coronary intervention | 27 (30%) | 37 (22%) | 0.16 |
| Previous bypass surgery | 19 (21%) | 27 (16%) | 0.32 |
| History of hypertension | 64 (70%) | 126 (74%) | 0.36 |
| History of diabetes | 32 (35%) | 67 (40%) | 0.48 |
| Number of cardiac risk factors | 2.4 ± 1.1 | 2.38 ± 1.13 | 0.91 |
| Aspirin | 63 (69%) | 95 (56%) | 0.04 |
| β blockers | 45 (49%) | 80 (47%) | 0.74 |
| % maximum age-predicted heart rate | 86 ± 16 | 88 ± 11 | 0.37 |
| Treadmill exercise | 31 (34%) | 76 (45%) | 0.12 |
| Abnormal stress ECG results | 14 (15%) | 45 (26%) | 0.03 |
| Rest wall motion score index | 1.3 ± 0.4 | 1.2 ± 0.3 | 0.19 |
| Number of new ischemic wall motion abnormalities | — | 4.5 ± 3.1 | <0.0001 |
| Peak wall motion score index | 1.2 ± 0.3 | 1.6 ± 0.4 | <0.0001 |
| Ejection fraction (%) | 56 ± 6 | 55 ± 5 | 0.36 |
| 1-vessel disease | 45 (50%) | 71 (42%) | 0.25 |
| 2- or 3-vessel disease | 46 (50%) | 98 (58%) | 0.25 |
| Cardiac events | |||
| Myocardial infarction | 3 (3.3%) | 15 (8.8%) | 0.09 |
| Cardiac death | 0 (0.0%) | 11 (6.5%) | 0.01 |
Descriptive patient characteristics and exercise and stress echocardiographic variables in patients with and without cardiac events on follow-up are listed in Table 2 .
| Variable | Cardiac Events | No Events | p Value |
|---|---|---|---|
| (n = 29) | (n = 231) | ||
| Age (years) | 66 ± 12 | 63 ± 10 | 0.38 |
| Men | 18 (62%) | 134 (58%) | 0.69 |
| Abnormal rest ECG results | 19 (66%) | 120 (55%) | 0.16 |
| Previous myocardial infarction | 16 (55%) | 75 (32%) | 0.01 |
| Previous percutaneous coronary intervention | 4 (14%) | 60 (26%) | 0.16 |
| Previous bypass surgery | 6 (21%) | 40 (17%) | 0.62 |
| History of hypertension | 25 (86%) | 165 (71%) | 0.12 |
| History of diabetes | 14 (48%) | 85 (36%) | 0.28 |
| Number of cardiac risk factors | 2.2 ± 1.01 | 2.4 ± 1.13 | 0.37 |
| Aspirin | 17 (59%) | 141 (61%) | 0.78 |
| β blockers | 17 (59%) | 108 (47%) | 0.22 |
| % maximum age-predicted heart rate | 87 ± 9 | 87 ± 13 | 0.94 |
| Treadmill exercise | 7 (24%) | 100 (43%) | 0.08 |
| Abnormal stress ECG results | 6 (21%) | 53 (23%) | 0.77 |
| Rest wall motion score index | 1.1 ± 0.3 | 1.2 ± 0.4 | 0.20 |
| Number of new ischemic wall motion abnormalities | 3.6 ± 2.6 | 2.9 ± 3.4 | 0.005 |
| Peak wall motion score index | 1.4 ± 0.4 | 1.4 ± 0.5 | 0.51 |
| Ejection fraction (%) | 56 ± 5 | 55 ± 5 | 0.67 |
| 1-vessel disease | 8 (28%) | 108 (47%) | 0.04 |
| 2- or 3-vessel disease | 21 (72%) | 123 (53%) | 0.04 |
All variables listed in Table 1 were considered in the univariate analysis. Significant univariate predictors of cardiac events are listed in Table 3 . Clinical and echocardiographic variables significant on univariate analysis were considered in multivariate analysis. On multivariate logistic regression analysis, multivessel CAD (hazard ratio 2.53, 95% confidence interval 1.16 to 5.51, p = 0.02) and the number of segments in which ischemia was present (hazard ratio 4.31, 95% confidence interval 1.29 to 14.38, p = 0.01) were predictors of cardiac events.
