Background
Wall motion (WM), Doppler-derived measurement of the coronary flow reserve (CFR) in the left anterior descending coronary artery (LAD), and myocardial perfusion imaging (MPI) can be sequentially assessed during dipyridamole stress echocardiography. Data regarding the relative diagnostic value of each of these parameters when assessed during the same examination in patients with suspected coronary artery disease (CAD) are lacking.
Methods
Dipyridamole stress echocardiography was performed in 400 patients at two centers, before the performance of clinically indicated coronary angiography. The diagnostic accuracy of WM, CFR-LAD, combined WM and CFR-LAD, and MPI was measured in comparison with quantitative angiographic results.
Results
For CAD defined as ≥1 stenosis >50%, MPI had the highest sensitivity (96%), lowest specificity (66%), and highest accuracy (86%); WM and CFR-LAD had the highest specificities (85% and 80%), lowest sensitivities (63% and 66%), and lowest overall accuracies (70% and 71%). Combined WM and CFR-LAD obtained intermediate values for both sensitivity (84%) and specificity (71%) and the second best accuracy (80%). For CAD defined as >70% stenosis, MPI, combined WM and CFR-LAD, and WM obtained similar accuracies ( P = NS), but WM showed the best balance of sensitivity (73%) and specificity (73%), with the highest Youden index.
Conclusions
MPI had the highest sensitivity and accuracy for the detection of CAD > 50% during dipyridamole stress echocardiography, despite showing the lowest specificity among tested parameters. Standalone WM and combined WM and CFR-LAD were not significantly inferior in terms of overall accuracy when CAD > 70% was the diagnostic end point. The addition of MPI or CFR-LAD to standard WM assessment allows the detection of milder CAD.
During the past decade, stress echocardiography (SE) has moved beyond the standalone analysis of wall motion (WM), thanks to the clinical feasibility of both Doppler-derived measurement of coronary flow reserve (CFR) in the left anterior descending coronary artery (LAD) and contrast myocardial perfusion imaging (MPI).
Both CFR-LAD and MPI have been independently applied in addition to WM analysis during SE, and each has demonstrated incremental sensitivity for the detection of coronary artery disease (CAD), although to date, only MPI has demonstrated an increase in overall diagnostic accuracy.
There is a lack of data regarding diagnostic comparisons and the potential synergistic value of these 3 parameters for the detection of CAD, when they are sequentially assessed during SE. Such a multiparametric stress echocardiographic protocol can be clinically implemented using commercially available equipment (using a single multifrequency probe) thanks to contemporary low-power, real-time contrast echocardiography, which permits the simultaneous (or sequential) assessment of WM, MPI, and CFR-LAD.
The use of ultrasonographic contrast media during SE has the triple advantage of higher feasibility and accuracy of WM assessment, adjunctive MPI analysis, and higher feasibility of CFR-LAD measurement. Dipyridamole is the ideal stressor for the sequential assessment of WM, CFR-LAD, and MPI within the same examination, because each of the 3 parameters has largely proven its diagnostic value for CAD detection using high-dose dipyridamole stress echocardiography.
We sought to prospectively compare the diagnostic value of combined WM, CFR-LAD, and MPI assessment during dipyridamole SE in patients undergoing clinically indicated coronary angiography.
Methods
Patients were enrolled from January 2009 to March 2010 from cardiology departments in Parma and Venice, Italy, with a predetermined target number of 400 patients. Four hundred fifty-four patients with chest pain syndromes already scheduled for clinically indicated coronary angiography were ultimately considered; 32 patients refused consent for enrollment.
Exclusion criteria were (1) a poor acoustic window precluding satisfactory imaging despite contrast administration, (2) severe valvular heart disease, (3) sustained ventricular arrhythmias or hemodynamic instability, (4) active chest pain within the past 24 hours, (5) known allergy to sulfonamides, (6) pregnancy or lactation, (7) severe chronic obstructive pulmonary disease, and (8) previous coronary artery bypass surgery. Of the 422 patients who consented to participate, 22 (5.2%) met exclusion criteria (10 with inadequate baseline echocardiographic image quality, 3 with sulfonamide allergy, and 9 with severe chronic obstructive pulmonary disease). Four hundred patients (263 men; mean age, 66 ± 11 years) represented the final study group. All patients were evaluated after the discontinuation of antianginal drugs (nitrates, calcium antagonists, β-blockers) for ≥24 hours. The study was approved by the institutional review boards. All patients gave written informed consent when they underwent SE. When patients provided written informed consent, they also authorized physicians to use their clinical data. The study complied with the Declaration of Helsinki.
Resting Echocardiography and SE
The contrast stress echocardiographic protocol is summarized in Figure 1 . Two-dimensional echocardiography, 12-lead electrocardiography, and blood pressure monitoring were performed in combination with high-dose dipyridamole (0.84 mg/kg over 6 min) in accordance with a well-established protocol. Transthoracic stress echocardiographic studies were performed using commercially available ultrasound machines (iE33; Philips Medical Systems, Andover, MA) equipped with a multifrequency phased-array probe (S5), with second-harmonic and low–mechanical index (MI) power modulation technology.
Apical two-chamber, three-chamber, and four-chamber views were always obtained. A modified three-chamber view for distal LAD imaging was integrated into the cardiac imaging sequence. Contrast WM, CFR-LAD, and MPI were sequentially assessed during SE using the same probe (S5) by activation of the appropriate preset. The left ventricle was divided into 17 segments, as suggested by the American Society of Echocardiography and the European Association of Echocardiography.
Contrast WM and MPI
A 0.5-mL bolus of SonoVue (Bracco Imaging Italia srl, Milan, Italy) was administered at rest and at peak stress (repeatable if contrast imaging time was not sufficient to acquire all required cine loops), followed by low-power (MI = 0.10) continuous imaging for both WM and MPI assessment; after the disappearance of microbubbles, standard (high-MI) imaging was resumed for WM monitoring. For MPI assessment, flash-replenishment sequences were used both in the continuous (40 frames/sec) and triggered end-systolic mode. A few seconds before administration of the contrast bolus, the low-MI setting was activated and fine-tuned for gain and MI so that no signal was detectable from the myocardium. After the SonoVue bolus, the ideal timing to start the flash-replenishment sequences for MPI is approximately 10 to 20 sec after peak video intensity is reached, as soon as it starts to decrease, so that it does not exceed the dynamic range of the system anymore. Normal perfusion after dipyridamole was assigned if myocardium was fully replenished 1.5 to 2 sec after the end of flash impulse (after dipyridamole usually corresponding to three cycles); perfusion was defined as abnormal if myocardium was not fully replenished after this time in one or more segments. The cutoff for normal replenishment at rest was considered to be 4 sec after the flash impulse. A perfusion defect was scored as fixed or reversible on the basis of its presence at rest. Only the presence of a reversible perfusion defect classified the results of MPI as positive or abnormal for the diagnostic end point of the study, whereas the presence of a fixed perfusion defect was considered to indicate negative or normal results of MPI. Basal segments were assessed for perfusion only when clearly visualized; otherwise (because of shading artifacts or low ultrasound penetration) they were not scored.
Segmental WM was graded as follows: 1 = normal, 2 = hypokinetic, 3 = akinetic, and 4 = dyskinetic. Inducible ischemia was defined as the occurrence of a stress-induced new dyssynergy or worsening of rest hypokinesia in one ore more segments.
CFR-LAD
Coronary flow in the mid-distal portion of the LAD was sought in the low parasternal long-axis cross-section or modified two-chamber view under the guidance of color Doppler flow mapping. Color-coded blood flow from the LAD was visualized both at rest and during peak stress using contrast enhancement (SonoVue, 0.2-mL intravenous bolus) in all patients; flow velocities were measured at baseline and at peak stress (before aminophylline injection). For both color Doppler flow mapping and pulsed-wave velocity measurements, the standard high-MI setting for CFR was adjusted after contrast bolus by lowering the MI to 0.1 to 0.3. At each time point, the three best profiles of peak diastolic Doppler flow velocities were measured, and the results were averaged. Coronary blood flow velocity reserve was defined as the ratio between hyperemic and basal peak diastolic coronary flow. CFR-LAD was considered normal when it was ≥1.9; this is one of the best diagnostic cutoffs found in the literature, and it is specifically validated for contrast-enhanced CFR measurements.
Image Analysis
In our study, to minimize the subjectivity of analysis, the assessment of all tested parameters (WM, MPI, and CFR-LAD) was performed by the consensus of two investigators. Readers were forced per the protocol to score as normal or abnormal all three parameters in each patient, whatever the image quality. Intraobserver and interobserver agreement was determined for WM, MPI, and CFR-LAD in 40 randomly selected patients, independently assessed by two investigators for each parameter. Results are presented as percentages with corresponding κ values.
Quantitative Coronary Angiography
All patients underwent coronary x-ray angiography within 1 week after SE. Conventional coronary x-ray angiography was performed using the Judkins approach, with selective catheterization of the left and right coronary artery system. The angiograms were evaluated for the presence of significant stenosis (ie, >50% and >70% luminal diameter reduction) in major epicardial coronary arteries and their branches (vessel diameter > 2.0 mm) by an experienced cardiologist blinded to the stress echocardiographic data.
The offline maximum percentage luminal reduction was determined for any visually evident stenosis using standard quantitative coronary angiographic software (Medis Medical Imaging Systems, Leiden, The Netherlands). The angiographic results were then classified as one-vessel, two-vessel, or three-vessel disease or exclusion of significant obstructive CAD for both stenosis cutoffs (50% and 70%).
Statistical Analysis
Continuous variables are presented as mean ± SD. Categorical variables were assessed using χ 2 tests. Sensitivity, specificity, and accuracy were calculated using standard definitions and are presented with 95% confidence intervals. Differences between sensitivity and specificity for WM, MPI, CFR-LAD, and the combination of WM and CFR-LAD were analyzed using McNemar’s test; accuracy was compared using the χ 2 test and the Youden index. A P value < .05 (two sided) was considered significant. The Youden index was calculated as [1 − (1 − sensitivity) + (1 − specificity)] and summarizes both sensitivity and specificity in one number between 0 and 1. The higher the Youden index, the better the diagnostic accuracy of the test.
Results
The main clinical, angiographic, and echocardiographic data are reported in Table 1 . Significant CAD (>50% stenosis) was detected in 268 patients (67%). One hundred thirty-four patients had single-vessel disease, and 134 had multivessel disease. In the 268 patients with angiographically significant CAD (stenosis > 50%), the LAD was involved in 202 patients (75%), in 82 (61%) of those with single-vessel disease and in 120 (90%) of those with multivessel disease.
| Characteristic | Value |
|---|---|
| Age (years) | 66 ± 11 (29–91) |
| Men/women | 263/137 |
| Risk factors and history | |
| Hypertension | 288 (72%) |
| Hypercholesterolemia | 256 (64%) |
| Smoking | 92 (23%) |
| Diabetes mellitus | 112 (28%) |
| Family history of CAD | 84 (21%) |
| Ejection fraction | 55 ± 7 |
| Reduced ejection fraction (<50%) | 128 (32%) |
| Prior CAD ∗ | 132 (33%) |
| Previous myocardial infarction | 91 (23%) |
| Previous revascularization | 104 (26%) |
| Coronary angiographic results | |
| Patients with CAD > 50% | 268 (67%) |
| One-vessel disease | 134 (34%) |
| Two-vessel disease | 82 (21%) |
| Three-vessel disease | 52 (13%) |
| LAD disease (>50%) | 202 (47%) |
| Stress echocardiographic results | |
| Patients with reversible WM abnormalities | 188 (47%) |
| Patients with CFR-LAD < 1.9 | 200 (50%) |
| Patients with reversible WM abnormalities or CFR-LAD < 1.9 | 264 (66%) |
| Patients with reversible MPI abnormalities | 302 (76%) |
∗ Defined as either known coronary artery stenosis > 50%, prior myocardial infarction, or prior myocardial revascularization.
Stress Echocardiographic Findings
The mean CFR-LAD value was 1.93 ± 0.4. Two hundred patients had normal CFR (≥1.9) and 200 had abnormal CFR (<1.9) in the LAD. On individual patient analysis, the test was positive for reversible WM abnormalities in 188 patients (47%), for combined WM and CFR-LAD in 264 (66%), and for reversible MPI abnormalities in 302 (76%). Of the 268 patients with significant CAD at angiography, 116 (43%) were in the MPI+/WM+/CFR+ class, 50 (19%) in the MPI+/WM+/CFR− class, 56 (21%) in the MPI+/WM−/CFR+ class, 35 (13%) in the MPI+/WM−/CFR− class, and 2 (0.7%) in the MPI−/WM+/CFR− class.
Only 7 patients (2.6%) with obstructive CAD had normal results for all three parameters (MPI−/WM−/CFR−), and all had single-vessel disease and 50% to 70% stenosis.
Diagnostic Accuracy
Sensitivity, specificity, and accuracy for WM, CFR-LAD, combined WM and CFR-LAD, and MPI for the presence of CAD > 50% or CAD > 70% in one or more coronary territories are reported in detail in Table 2 and Figures 2 and 3 . Ninety-five percent confidence intervals for all accuracy data are reported in Figures 3 and 4 . When CAD was defined as the presence of at least one >50% stenosis, MPI had the highest sensitivity (96%), lowest specificity (66%), and highest accuracy (86%) (Youden index, 0.62), whereas standalone WM and standalone CFR-LAD had the highest specificities (85% and 80%), lowest sensitivities (63% and 66%), and lowest overall accuracies (70% and 71%).
| Sensitivity (%) | Specificity (%) | Accuracy (%) and Youden Index | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Variable | WM | CFR | WM + CFR | MPI | WM | CFR | WM + CFR | MPI | WM | YI | CFR | YI | WM + CFR | YI | MPI | YI |
| CAD > 50% | ||||||||||||||||
| All patients (rank ∗ ) | 168/268 (63) | 176/268 (66) | 226/268 (84) | 257/268 (96) | 112/132 (85) | 106/132 (80) | 94/132 (71) | 87/132 (66) | 280/400 (70) | 0.39 | 282/400 (71) | 0.46 | 320/400 (80) | 0.55 | 344/400 (86) | 0.62 |
| Prior CAD | 55/96 (57) | 66/96 (69) | 83/96 (86) | 90/96 (94) | 33/36 (92) | 25/36 (69) | 23/36 (64) | 28/36 (77) | 88/132 (67) | 0.49 | 91/132 (69) | 0.38 | 108/132 (82) | 0.53 | 118/132 (89) | 0.71 |
| No prior CAD | 113/172 (66) | 110/172 (64) | 143/172 (83) | 167/172 (97) | 79/96 (82) | 81/96 (84) | 71/96 (74) | 59/96 (61) | 192/268 (72) | 0.48 | 191/268 (71) | 0.48 | 214/268 (80) | 0.57 | 226/268 (84) | 0.58 |
| LAD disease | 134/202 (66) | 156/202 (77) | 182/202 (90) | 193/202 (96) | ||||||||||||
| Anterior | 38/82 (46) | 64/82 (78) | 68/82 (83) | 75/82 (91) | ||||||||||||
| Posterior | 34/66 (52) | 20/66 (30) | 44/66 (67) | 64/66 (97) | ||||||||||||
| Anterior and posterior | 96/120 (80) | 92/120 (77) | 114/120 (95) | 118/120 (98) | ||||||||||||
| One-vessel | 62/134 (46) | 78/134 (58) | 96/134 (72) | 125/134 (93) | ||||||||||||
| Two-vessel | 60/82 (73) | 50/82 (61) | 78/82 (95) | 82/82 (100) | ||||||||||||
| Three-vessel | 46/52 (88) | 46/52 (88) | 52/52 (100) | 50/52 (96) | ||||||||||||
| CAD >70% | ||||||||||||||||
| All patients (rank ∗ ) | 126/172 (73) | 112/172 (65) | 154/172 (90) | 170/172 (99) | 166/228 (73) | 140/228 (61) | 118/228 (52) | 96/228 (42) | 292/400 (73) | 0.46 | 252/400 (63) | 0.26 | 272/400 (68) | 0.42 | 266/400 (67) | 0.41 |
| Prior CAD | 39/54 (72) | 36/54 (67) | 53/54 (98) | 54/54 (100) | 59/78 (76) | 37/78 (47) | 31/78 (40) | 34/78 (44) | 98/132 (74) | 0.48 | 73/132 (55) | 0.14 | 84/132 (64) | 0.38 | 88/132 (67) | 0.44 |
| No prior CAD | 87/118 (74) | 76/118 (64) | 101/118 (86) | 116/118 (98) | 107/150 (71) | 103/150 (69) | 85/150 (57) | 62/150 (41) | 204/400 (51) | 0.45 | 179/400 (45) | 0.33 | 186/400 (47) | 0.43 | 178/400 (45) | 0.39 |
| LAD disease | 78/98 (80) | 80/98 (82) | 94/98 (96) | 98/98 (100) | ||||||||||||
| Anterior | 22/30 (73) | 28/30 (93) | 28/30 (93) | 30/30 (100) | ||||||||||||
| Posterior | 24/40 (60) | 8/40 (20) | 28/40 (70) | 40/40 (100) | ||||||||||||
| Anterior and posterior | 80/102 (78) | 76/102 (75) | 98/102 (96) | 100/102 (98) | ||||||||||||
| One-vessel | 36/58 (62) | 32/58 (55) | 44/58 (76) | 58/58 (100) | ||||||||||||
| Two-vessel | 48/66 (73) | 38/66 (58) | 62/66 (94) | 66/66 (100) | ||||||||||||
| Three-vessel | 42/48 (88) | 42/48 (88) | 48/48 (100) | 46/48 (96) | ||||||||||||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
