Background
An important goal of noninvasive stress testing is the identification of patients with left main coronary artery or three-vessel disease, because coronary artery disease extension and severity are major prognostic factors in ischemic heart disease. Wall motion abnormalities during vasodilator stress echocardiography become apparent in more than one coronary territory only in a small number of patients with multivessel disease. The aim of this study was to assess the value of change in left ventricular ejection fraction change (ΔLVEF) to identify patients with multivessel obstructive coronary artery disease during dipyridamole stress echocardiography.
Methods
All dipyridamole stress echocardiographic studies performed at the authors’ institution from October 2007 through March 2010 were retrospectively reviewed, and 150 patients who underwent coronary angiography within the next 60 days were selected. Left ventricular end-diastolic volume and end-systolic volume were measured at baseline and at the end of high-dose dipyridamole; ΔLVEF was calculated as stress ejection fraction minus rest ejection fraction. Patients were divided into four groups (controls and patients with single-vessel, two-vessel, and three-vessel disease) on the basis of coronary angiographic results.
Results
The mean LVEF increased significantly from rest to peak stress in all groups except the three-vessel disease group. Mean ΔLVEF was negative in patients with three-vessel or left main coronary artery disease (−2.8 ± 5.1%) and significantly lower compared with all other angiographic groups (10.2 ± 5.1% and 6.2 ± 4.1%, respectively, for single-vessel and two-vessel disease). The negative value of ΔLVEF for three-vessel disease was due mainly to increased end-systolic volume at peak stress. Receiver operating characteristic curves demonstrated excellent accuracy of ΔLVEF compared with change in wall motion score index in identifying patients with multivessel disease, with areas under the curves of 0.96 and 0.62, respectively.
Conclusions
ΔLVEF is significantly lower in patients with severe coronary artery disease compared with those with single-vessel or two-vessel disease; reduced ΔLVEF identifies high-risk patients, who are likely to benefit from a more aggressive therapeutic strategy.
An important goal of noninvasive stress testing is the identification of patients with left main coronary artery or three-vessel disease, because coronary artery disease (CAD) extent and severity are major prognostic factors in ischemic heart disease. Patients with multivessel or left main CAD in particular can obtain the highest prognostic benefit from coronary revascularization.
Vasodilator stress echocardiography is a well-established technique for the detection of obstructive CAD, and its feasibility and accuracy have recently been improved by the introduction of ultrasound contrast agents and myocardial perfusion imaging.
Although vasodilator stress echocardiography is certainly more sensitive in patients with multivessel disease (compared with single-vessel disease), its sensitivity remains suboptimal in these patients; inducible wall motion abnormalities in a multivessel distribution become apparent only in a small proportion of such patients, not different from the behavior of perfusion defects during vasodilator single-photon emission computed tomography (SPECT) or 82 Rb positron emission tomography (PET).
A recent clinical study demonstrated an inverse relationship between CAD severity and change in left ventricular ejection fraction (ΔLVEF), calculated as stress ejection fraction minus rest ejection fraction, when assessed in patients using vasodilator 82 Rb PET.
The aim of this study was to assess the value of ΔLVEF during high-dose dipyridamole stress echocardiography in identifying patients with multivessel CAD.
Methods
Study Population
We retrospectively reviewed all dipyridamole stress echocardiographic studies performed at our institution from October 2007 through March 2010 and analyzed those who underwent coronary angiography within the next 30 days. Patients with prior revascularization (both surgical and percutaneous), prior myocardial infarction, or LVEFs < 50% were excluded. Patients with poor acoustic windows or more than mild valvular disease were excluded as well.
Echocardiography
Dipyridamole stress echocardiography was performed in all patients using a Philips iE33 ultrasound unit and S5 probe (Philips Medical Systems, Andover, MA) using a high-dose protocol (0.84 mg/kg/6 min). We routinely used SonoVue (Bracco Imaging Italia srl, Milan, Italy) as a contrast media at rest and at peak stress. Rest and peak images were reviewed offline by two experienced echocardiographers blinded to the results of coronary angiography.
Regional wall motion analysis was evaluated at baseline and at peak stress by a semiquantitative assessment of wall motion score index using a 17-segment model of the left ventricle.
Left ventricular end-diastolic volume (EDV) and end-systolic volume (ESV) were measured at baseline and at the end of high-dose dipyridamole by manually tracing the endocardial border and using the modified biplane Simpson’s method. LVEF and ΔLVEF (stress ejection fraction − rest ejection fraction) were then calculated.
End-Systolic Pressure-Volume Determination
Blood pressure was recorded using a sphygmomanometer and a standard stethoscope. Systolic blood pressure (SP) and diastolic blood pressure were measured in the right arm with the patient lying in the left lateral position.
Assessment of the end-systolic pressure-volume relationship was made using an echocardiographic, noninvasive method to evaluate changes in the inotropic state.
Force was determined as the ratio of SP to ESV at rest and at peak stress. The end-systolic pressure-volume relationship was defined as up-sloping when SP/ESV increased during stress and down-sloping when peak exercise SP/ESV index was equal to or lower than rest values.
Coronary Angiography
Quantitative coronary angiography was performed by an experienced cardiologist unaware of the results of echocardiography. Any visually evident stenosis was measured using a computer-based system dedicated to quantitative analysis (Qangio XA version 7.0; Medis Medical Imaging, Leiden, The Netherlands) and expressed as percentage narrowing using the nearest normal-appearing region as the reference.
Patients were divided into four groups (controls, with no significant CAD, and patients with single-vessel, two-vessel, and three-vessel or left main CAD) on the basis of coronary angiographic results. Significant coronary artery stenosis was defined as ≥70% diameter stenosis present in the proximal or mid portion of one major epicardial coronary vessel or as ≥50% diameter stenosis present in the left main coronary artery.
Statistical Analysis
Categorical variables are expressed as frequencies and percentages and continuous variables as mean ± SD. Intergroup comparisons for continuous variables were performed using one-way analysis of variance with Bonferroni’s correction. Categorical variables were compared using χ 2 tests. Twenty-five patients were randomly selected to test intraobserver and interobserver variability for EDV, ESV, and LVEF measurements using Bland-Altman analysis.
Results
From October 2007 through March 2010, 1,315 patients underwent dipyridamole-atropine stress myocardial contrast echocardiography at our institution for evaluation of suspected CAD, of whom 239 underwent coronary angiography within 60 days. Sixty-one patients were excluded because of previous revascularization procedures or ST-segment elevation myocardial infarction, 26 were excluded because of baseline ejection fractions < 50%, and two were excluded for poor acoustic windows. Consequently, 150 patients were included in the present study.
Angiographic Characteristics
CAD was present in 108 patients: 41 patients had multivessel CAD (left main or three-vessel CAD), 29 had two-vessel CAD, 38 had single-vessel CAD, and 42 had no significant coronary artery stenosis. The distribution of lesions was as follows: left main trunk in 13 patients, left anterior descending coronary artery in 71 patients, left circumflex coronary artery in 73 patients, and right coronary artery in 65 patients.
Baseline Clinical Characteristics of Patients
A total of 150 patients were included in the present study. The mean age was 66.5 ± 9.2 years, and 36.7% were women.
Patients were divided into four groups depending on the presence and extent of CAD. Table 1 summarizes baseline clinical characteristics.
| Characteristic | Control ( n = 42) | Single-vessel disease ( n = 38) | Two-vessel disease ( n = 29) | Three-vessel disease ( n = 41) | P |
|---|---|---|---|---|---|
| Men | 13 (31.0%) | 24 (63.2%) | 24 (82.8%) | 34 (82.9%) | <.05 |
| Age (y) | 63.0 ± 10.7 | 67.2 ± 7.9 | 67.6 ± 7.9 | 68.8 ± 8.7 | <.05 |
| Hypertension | 18 (42.9%) | 28 (73.7%) | 22 (75.9%) | 17 (41.5%) | <.05 |
| Diabetes | 6 (14.3%) | 8 (28.1%) | 9 (31.0%) | 11 (26.8%) | NS |
| Hypercholesterolemia | 15 (35.7%) | 18 (47.4%) | 21 (72.4%) | 17 (41.5%) | <.05 |
The prevalence of male gender (82.9%) was higher in the three-vessel disease group than among controls; patients with three-vessel disease were also older than patients in the control group.
Hemodynamics during Stress
Hemodynamic profiles at rest and at peak stress were similar across all groups. There were no differences regarding heart rate, SP, and the rate-pressure product achieved during stress among the four groups ( Table 2 ). The mean peak heart rate during stress was 115 ± 27 beats/min for patients with severe CAD.
| Characteristic | Control ( n = 42) | Single-vessel disease ( n = 38) | Two-vessel disease ( n = 29) | Three-vessel disease ( n = 41) | P ∗ |
|---|---|---|---|---|---|
| HR at rest (beats/min) | 70.0 ± 11.5 | 68.8 ± 12.1 | 69.6 ± 10.1 | 70.4 ± 12.4 | NS |
| HR at peak stress (beats/min) | 120.3 ± 27.0 | 125.4 ± 30.6 | 115.1 ± 26.1 | 115.0 ± 27.0 | NS |
| SP at rest (mm Hg) | 120.8 ± 23.0 | 117.4 ± 24.6 | 120.9 ± 28.1 | 129.0 ± 31.8 | NS |
| SP at peak stress (mm Hg) | 139.0 ± 23.8 | 143.5 ± 19.9 | 138.6 ± 14.8 | 138.7 ± 20.3 | NS |
| RPP at peak stress | 16,983.9 ± 5,650.6 | 18,181.6 ± 5,643.4 | 15,804.8 ± 3,330.1 | 16,109.0 ± 4,927.6 | NS |
Relationship between ΔLVEF and the Extent of CAD
One hundred fifty stress echocardiograms were reviewed, and left ventricular volume, LVEF, and ΔLVEF were calculated ( Table 3 ). EDV and ESV at rest were significantly higher in patients with three-vessel disease (101.1 ± 27.5 vs 76.5 ± 31.4 mL, P < .05, and 38.7 ± 13.4 vs 28.2 ± 17.6 mL, P < .05, respectively) compared with controls, while LVEF at rest did not show a statistical difference between the two groups. Patients with three-vessel disease showed the same results compared with those with single-vessel disease. Patients with two-vessel disease showed the same tendency without reaching statistical significance. The mean LVEF increased from rest to peak stress in all groups except the three-vessel disease group. Mean ΔLVEF was negative (−2.8 ± 5.1%) in patients with three-vessel and left main CAD and was significantly lower than in patients in the other angiographic groups ( P < .001; Figure 1 ). On the contrary, patients with less severe CAD showed significant increases in LVEF during vasodilator stress of 10.2 ± 5.1% and 6.2 ± 4.1%, respectively, for the single-vessel and two-vessel disease groups. The reduced ΔLVEF in patients with three-vessel disease was due mainly to increased ESV at peak stress. Receiver operating characteristic curves demonstrated excellent accuracy of ΔLVEF compared with change in wall motion score index in identifying patients with three-vessel or left main CAD, with areas under the curves of 0.96 and 0.62, respectively ( Figure 2 ).
| Characteristic | Control ( n = 42) | Single-vessel disease ( n = 38) | Two-vessel disease ( n = 29) | Three-vessel disease ( n = 41) | P ∗ |
|---|---|---|---|---|---|
| EDV at rest (mL) | 76.5 ± 31.4 | 86.2 ± 26.4 | 95.7 ± 31.1 | 101.1 ± 27.5 | <.05 |
| EDV at peak stress (mL) | 52.0 ± 24.8 | 58.4 ± 19.3 | 68.0 ± 24.6 | 70.3 ± 19.4 | <.05 |
| ESV at rest (mL) | 28.2 ± 17.6 | 31.3 ± 12.5 | 38.0 ± 17.3 | 38.7 ± 13.4 | <.05 |
| ESV at peak stress (mL) | 19.1 ± 15.6 | 21.7 ± 10.6 | 32.4 ± 15.9 | 40.5 ± 15.0 | <.001 |
| LVEF at rest (%) | 64.7 ± 6.5 | 63.9 ± 7.3 | 61.2 ± 8.1 | 61.9 ± 8.2 | NS |
| LVEF at peak stress (%) | 75.4 ± 6.8 | 74.1 ± 7.3 | 67.4 ± 7.9 | 59.1 ± 9.7 | <.001 |
| ΔLVEF (%) | 10.6 ± 5.6 | 10.2 ± 5.1 | 6.2 ± 4.1 | −2.8 ± 5.1 | <.001 |
| ΔWMSI | 0.05 ± 0.16 | 0.12 ± 0.14 | 0.14 ± 0.15 | 0.16 ± 0.19 | <.05 |
| SP/ESV at rest | 5.67 ± 3.67 | 4.49 ± 2.93 | 4.19 ± 2.11 | 4.24 ± 3.04 | NS |
| SP/ESV at peak stress | 9.73 ± 5.99 | 8.46 ± 4.50 | 5.46 ± 2.87 | 4.05 ± 2.34 | <.001 |
| ΔSP/ESV (stress − rest) | 4.06 ± 3.70 | 3.97 ± 3.47 | 1.27 ± 1.36 | −0.20 ± 1.29 | <.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
