The diagnosis of lesions with severe calcium or in-stent stenosis using coronary computed tomography angiography (CCTA) is still difficult. The aim of the present study was to evaluate the accuracy of transthoracic Doppler echocardiography (TTDE) in patients with suspected angina pectoris, who had a proximal left coronary artery (LCA) site that could not be evaluated by CCTA. Fifty-eight patients were evaluated. The proximal LCA was defined as the left main coronary artery and proximal left anterior descending coronary artery. All patients underwent TTDE and had coronary angiography performed as a reference method. We measured the proximal left coronary flow velocity (CFV) by both color and pulse Doppler methods. Proximal coronary flow was detected in 45 (78%) of 58 patients. CFVs measured by both methods were significantly greater in the group with severe stenosis (percent diameter stenosis >70%) than in the groups with moderate stenosis (percent diameter stenosis 40% to 70%) or without stenosis (color Doppler: 148 ± 42 cm/s, 89 ± 40 cm/s, and 41 ± 22 cm/s, respectively, p <0.05; pulse Doppler: 143 ± 61 cm/s, 82 ± 33 cm/s, and 39 ± 17 cm/s, respectively, p <0.05). Receiver operating characteristic curve analysis showed that the optimal CFV cut-off values obtained by color and pulse Doppler to diagnose severe stenosis were 92 cm/s (sensitivity, 100%; specificity, 90%) and 81 cm/s (sensitivity, 100%; specificity, 85%), respectively. In conclusion, TTDE could diagnose proximal LCA stenosis with good accuracy in patients who could not be evaluated by CCTA.
Screening examinations have been performed in patients with suspected coronary artery disease using coronary computed tomography angiography (CCTA). However, it is still difficult to examine lesions with severe calcium or in-stent stenosis using CCTA because of blooming artifacts and beam-hardening artifacts. In a previous study, we reported that the use of Doppler flow velocity in the continuity equation and a mosaic flow pattern on color Doppler obtained by transthoracic Doppler echocardiography (TTDE) could diagnose proximal left coronary artery (LCA) stenosis in patients with angina pectoris. Therefore, we hypothesized that TTDE would also be useful to diagnose a proximal LCA stenosis that was unable to be evaluated by CCTA. The aim of this study was to evaluate the accuracy of TTDE in patients with suspected angina pectoris, who had a proximal LCA site that could not be evaluated by CCTA.
Methods
We enrolled 574 consecutive patients with suspected angina pectoris, who had CCTA performed from December 2010 to February 2012. There were 187 patients excluded because of the presence of at least 1 of the following criteria: myocardial infarction, congestive heart failure, hypertrophic cardiomyopathy, left ventricular hypertrophy with posterior wall thickness >11 mm, past history of coronary artery bypass surgery, ejection fraction <60%, atrial fibrillation, significant valvular disease, and total occlusion of the left anterior descending coronary artery (LAD). These conditions were considered to influence coronary flow velocity (CFV) at rest. Of the remaining 387 patients, 58 had a proximal LCA site that was difficult to evaluate by CCTA because of severe calcium or stent implantation, and these patients were evaluated by TTDE in this study. The proximal LCA was defined as the left main coronary artery (LMCA) and proximal LAD. All patients underwent TTDE and had coronary angiography (CAG) performed as a reference method. All patients continued taking anti-ischemic medications. The protocol was approved by the Institutional Review Board of our hospital, and informed consent was obtained at the time of echocardiography.
The CCTA examinations were performed using a dual-source scanner (Somatom Definition Flash, Siemens Medical Solutions, Forchheim, Germany) in all patients. Scan parameters were 120 kV tube voltage, 280 ms gantry rotation time, and 2 × 128 × 0.6 mm collimation with z-Sharp. A test injection of 15 ml of contrast medium (370 mg/ml iopamidol, Iopamiron, Bayer HealthCare, Osaka, Japan) followed by 30 ml of saline solution was given to investigate the optimal timing for data acquisition. Actual contrast enhancement was achieved by injecting 30 to 60 ml of contrast medium, followed by 30 ml of saline solution. All injections were delivered at 5 ml/s. We routinely performed reconstructions at 70% of the RR interval.
Longitudinal and cross-sectional images were manually reconstructed using a commercially available external workstation (Ziostation2, Ziosoft, Tokyo, Japan). The lumen of the proximal LCA was evaluated using transaxial, longitudinal, and cross-sectional images by 2 experienced radiologists, who were blinded to the subject’s identity and clinical profile. Differences in interpretation were resolved by consensus or a third investigator if necessary. A calcium score was calculated according to the Agatston method on the same workstation using a previously reported method.
Echocardiographic examinations were performed using a Vivid 7 dimension ultrasound machine (GE Healthcare, Milwaukee, Wisconsin) with an M4S probe (frequency, 1.8 to 4.0 MHz). Patients were examined in the left lateral decubitus position. After routine examination, the root of the aorta was imaged in the parasternal short-axis view, and proximal coronary flow was sought under color Doppler guidance. If proximal coronary flow could not be visualized after 10 minutes of examination, the measurement was judged a failure based on the echocardiographer’s decision. The measurement of CFV using color Doppler was performed as previously described. In brief, the initial color Doppler velocity range was set to ±19.0 cm/s. When proximal LCA flow was detected and showed flow aliasing, the color velocity range was gradually increased until color aliasing nearly disappeared (i.e., the entire signal area of coronary flow became homogenous). When color aliasing did not disappear in the image at the highest velocity of the color Doppler range, the color baseline was shifted until color aliasing nearly disappeared. The velocity range of color Doppler or the color baseline velocity when color aliasing nearly disappeared was defined as “isovelocity.” In patients with coronary flow detection but no color aliasing, the value of isovelocity was defined as 19 cm/s. The values of isovelocity were compared with the results of CAG. We also measured CFV in the proximal LCA and the distal LAD using pulse Doppler, as previously reported. The sample volume was placed at the same aliasing site that was used to measure isovelocity, and CFV was recorded. An effort was made to set the angle correction as small as possible. The mean angle correction for the proximal average diastolic peak velocity (ADPV) measurement was 31 ± 10° (range, 9° to 46°), and that for the distal ADPV measurement was 18 ± 11° (range, 0 to 39°). Using the analysis program in the ultrasound system, the ADPV was determined from velocities of >3 cardiac cycles for both the proximal LCA and distal LAD. The distal ADPV to proximal ADPV ratio (DAPAR) was calculated with the following formula: DAPAR = distal ADPV/proximal ADPV. The values of proximal ADPV and DAPAR were compared with the CAG findings.
Diagnostic CAG was carried out within 24 hours after the echocardiographic examination. Quantitative analysis of the percent diameter stenosis was performed with CMS analysis software (Medical Image Systems, Leiden, Netherlands). Severe and moderate stenoses were defined as a percent diameter stenosis >70% and 40% to 70%, respectively. Significant stenosis was defined as moderate or more stenosis in the present study.
Data are expressed as mean ± SD. The CFV data from the group without stenosis and the 2 groups with stenosis were compared using a 1-way analysis of variance followed by the Scheffe multiple comparison test. The area under a receiver operating characteristic curve was used to determine the optimal cut-off values of isovelocity, proximal ADPV, and DAPAR for the prediction of severe or significant stenosis. Intraobserver and interobserver reproducibility was assessed using the intraclass correlation coefficient. All statistical analyses were performed using SPSS 15.0J software for Windows (SPSS Inc., Chicago, Illinois). Values of p <0.05 were considered statistically significant.
Results
Proximal left coronary flow in the short-axis view was detected in 45 (78%) of 58 patients. Distal LAD coronary flow was detected in all patients. The baseline characteristics of these 45 patients are listed in Table 1 . The mean Agatston score (excluding 2 patients with stent in the proximal LCA) was 1,110 ± 1,360. There were 5 patients with severe stenosis (mean percent diameter stenosis, 81 ± 8%) and 14 patients with moderate stenosis (mean percent diameter stenosis, 54 ± 8%).
Variable | Value |
---|---|
Men/Women | 30/15 |
Age (yrs) | 73 ± 9 |
Body mass index (kg/m 2 ) | 23.8 ± 3.1 |
Unstable angina pectoris | 3 (7) |
Hypertension | 37 (82) |
Dyslipidemia | 20 (44) |
Diabetes mellitus | 22 (49) |
Current smoker | 21 (47) |
Coronary angiographic results | |
Number of coronary arteries narrowed | |
1 | 14 (31) |
2 | 14 (31) |
3 | 10 (22) |
Stenting in the proximal LCA | 2 (4) |
Table 2 lists the coronary flow parameters in the groups with different degrees of coronary stenosis as assessed by CAG. Isovelocity and proximal ADPV in the group with severe stenosis were significantly greater than those in the other groups. DAPAR in the groups with moderate and severe stenoses was significantly less than that in the group without stenosis. Isovelocity and proximal ADPV in the group with moderate stenosis were significantly greater than those in the group without stenosis. There were no significant differences in the distal ADPV among the 3 groups.
Variable | Coronary Stenosis Assessed by CAG | ||
---|---|---|---|
None (n = 26) | Moderate (n = 14) | Severe (n = 5) | |
Isovelocity (cm/s) | 41 ± 22 | 89 ± 40 ∗ | 148 ± 42 ∗ , † |
Proximal ADPV (cm/s) | 39 ± 17 | 82 ± 33 ∗ | 143 ± 61 ∗ , † |
Distal ADPV (cm/s) | 21 ± 7 | 24 ± 9 | 23 ± 10 |
DAPAR | 0.60 ± 0.25 | 0.34 ± 0.17 ∗ | 0.19 ± 0.09 ∗ |