An increase in the diastolic to systolic flow velocity ratio (D/S) in the proximal left internal thoracic artery (ITA) after coronary artery bypass grafting (CABG) enables noninvasive assessment of graft patency by transthoracic Doppler echocardiography (TTDE). The increase in the D/S can be less pronounced at a site distant from the anastomosis. We postulated that proximal ITA flow patterns differ between the left and right ITAs and that the increase in D/S is less pronounced in the right than in the left proximal ITA.
Proximal ITA flow was examined by TTDE in 129 consecutive patients after CABG of the left (75) or right (69) ITA to the left coronary artery. The mean D/S of the ITAs was compared with coronary angiography.
The D/S was lower in the group with a patent right ITA than in the group with a patent left ITA ( P < .05). The D/S of both the left and right ITAs negatively correlated with angiographic stenosis ( r = 0.56 or 0.67, P < .001, respectively). The regression line was significantly shifted downward in the right ITA compared with the left ITA, according to analysis of covariance ( P = .01). Graft stenosis was predicted by a D/S of <0.57 and <0.28 with an accuracy of 91% and 97% in the left and right ITAs, respectively.
The patency of both left and right ITA grafts to the left coronary artery can be assessed using TTDE, but different cutoff values of D/S are required to diagnose severe ITA stenosis.
The internal thoracic arteries (ITAs) are the first choice of conduits for grafting to the left coronary artery (LCA) in patients with ischemic heart disease. The long-term patency and functional effectiveness of the ITAs are superior to those of vein grafts. Postoperative clinical benefits are improved using the bilateral ITA for patients with multivessel diseases compared with a single ITA and other bypass conduits. Therefore, the right ITA is now frequently used as a bypass conduit to the LCA. Coronary angiography remains the conventional gold standard for assessing conduit patency despite being invasive and associated with risks of various intraoperative complications, such as arterial dissection and stroke.
Flow in the ITA has been investigated using transthoracic Doppler echocardiography (TTDE) since the 1990s. The diastolic flow velocity of the ITA increases after coronary artery bypass grafting (CABG) as the result of physiologically decreased resistance in the coronary circulation, and an increase in the diastolic to systolic flow velocity ratio (D/S) has been used as an index of bypass graft patency. Several reports have shown that a decrease in D/S <0.5 or 0.6 is a powerful predictive parameter of ITA graft stenosis. On the other hand, studies using Doppler flow wires have shown variations in Doppler flow velocity patterns at the proximal, mid, and distal portions of ITA grafts. These studies demonstrated that the increase in D/S can be less pronounced at a site distant from the anastomosis. The proximal right ITA is further from the anastomosis than the proximal left ITA.
We postulated that proximal ITA flow patterns differ between left and right ITAs and that the increase in D/S is less pronounced in the right than in the left proximal ITA, resulting in a need for different D/S cutoff values for assessing severe ITA stenosis. The present study compares the D/S for evaluating the severe graft stenosis between the right and left ITAs using TTDE.
Materials and Methods
We studied 146 consecutive patients (124 men and 22 women, age 66 ± 11 years) who underwent CABG using the left or right ITA; 84 patients had left ITA grafts, and 82 patients had right ITA grafts. Twenty patients received grafts using both ITAs. They underwent TTDE and coronary angiography to treat recurrent angina and atypical chest pain, or postoperatively to confirm graft patency. The Institutional Committee of Kagoshima University Hospital approved the study protocol.
Detection of Proximal Left and Right ITAs by TTDE
Standard two-dimensional and Doppler echocardiography proceeded using a digital ultrasound system (HDI-5000, Philips Medical Systems, Bothell, WA; Acuson Sequoia, Siemens, Mountain View, CA; Vivid 7, GE Healthcare, Milwaukee, WI) with 2- to 4-MHz transducers. To detect ITA flow, the initial velocity range of the color Doppler was set at approximately ±21 cm/sec and subsequently adjusted to optimize visualization of color signals in the bypass graft. All patients were examined in the supine position. A transducer was placed on the left or right supraclavicular fossa, and the first task in detecting ITA flow was to identify the left or right subclavian artery (SCA). The proximal left ITA was detected by rotating the transducer slightly clockwise and inclining it toward the anterior chest wall ( Figure 1 ). The proximal right ITA was detected by rotating the transducer clockwise 90 degrees at the level of the thyrocervical trunk bifurcation of the right SCA and inclining the transducer toward the anterior chest wall ( Figure 2 ). Flow in both ITAs was determined by recognizing downward flow in the tubular structures branching from the subclavian arteries. Transducer position and Doppler beam direction were adjusted to minimize the Doppler angle to the ITA flow direction by color flow mapping. Flow velocity in the ITA was examined 1.5 to 2.0 cm away from the SCA by pulsed-wave Doppler echocardiography with a sample volume of 2 mm. The sample volume direction was adjusted along the long axis of the ITA flow for angle correction to determine flow velocity more accurately. The ITA flow was recorded as a biphasic flow pattern by pulsed-wave Doppler echocardiography ( Figure 3 ).
Measurements of ITA Flow by TTDE
The peak systolic and diastolic velocity and the mean systolic and diastolic velocity were measured by tracing the outer borderline of the systolic and diastolic Doppler velocity profiles, respectively. Each parameter was obtained by averaging measurements from three to five consecutive cardiac cycles, and the peak and mean D/S were calculated.
Angiographic Analysis of ITA Grafts
All patients underwent coronary angiography within 1 month after the TTDE study using a standard femoral approach. Nitroglycerin (0.3 mg) was selectively injected into each bypass graft. We quantified the percentage of stenosis in the ITA and the anastomotic site between the ITA and the LCA using a computer-assisted method (Cardiovascular Measurement System version 2.0, Medical Imaging System, Nuenen, The Netherlands). A selected cinematic frame was digitized, and an automatic edge-detection program determined the graft and the coronary artery contours by assessing brightness along scan lines that ran perpendicular to the center lines of the graft. The image was calibrated on the basis of the known size of the catheter, and the vessel diameters were displayed for the length of the segment analyzed. The computer program automatically calculated the minimum lumen diameter and the percentage of diameter stenosis. An ITA with a diameter stenosis of ≥75% was considered severely stenotic.
Reproducibility of Measurements
Inter- and intraobserver variability were determined as the difference between repeated mean systolic velocity and diastolic velocity values in 10 patients by 2 independent observers and by the same observer, respectively.
All continuous data are expressed as means ± SD. The significance of differences between two groups was tested using the unpaired Student t test, and the significance of differences among groups was tested using one-way analysis of variance with subgroup analysis using Scheffe’s post hoc test. The D/S and percentage of diameter stenosis in the left and right ITAs were compared using regression analysis and analysis of covariance (ANCOVA). Receiver operating characteristics curves were constructed, and areas under the curves were measured to determine cutoff values for optimal sensitivity and specificity. P values < .05 were considered statistically significant.
Detection of Flow in the ITA Using TTDE
Flow in the left ITA was detected in 84 patients (100%) and in the right ITA was detected in 75 of 82 patients (91%) by TTDE. All seven patients in whom right ITA flow was undetectable had grafts to the left circumflex artery (LCX). Three of these seven patients had grafts that used both ITAs, and left ITA flow was detected by TTDE.
Angiographic Assessment of ITA and LCA
Fifteen patients had ≥75% stenosis in the LCA distal to the anastomosis. Two of them had grafts using both ITAs, and another ITA had no distal coronary artery stenosis. We excluded these 15 ITAs in 15 patients from this study. Therefore, we finally investigated 144 ITAs in 129 patients, of whom 15 received grafts using both ITAs. Table 1 summarizes the clinical characteristics of the patients. The study patients were divided into the following groups: patent left ITA with <75% diameter stenosis ( n = 66), stenotic left ITA with ≥75% diameter stenosis ( n = 9), patent right ITA ( n = 50), and stenotic right ITA ( n = 19).
|No. of patients||129|
|Age (y)||66 ± 10|
|Months after CABG||21 ± 13.1 (0.5–120)|
|Left ITA–LCX||21 (sequential 2)|
|Right ITA–LCX||38 (sequential 3)|
|Symptomatic angina||58/ 129|
|Congestive heart failure||20/129|
|Previous myocardial infarction||49/129|
|Ejection fraction by Simpson’s method (%)||48 ± 10 (35–78)|
Doppler Measurements of ITA Flow in Four Groups
Table 2 shows a comparison of Doppler values among the four groups. The peak systolic velocity did not significantly differ among the groups. The peak diastolic velocity was significantly lower in a stenotic left ( P < . 05) or right ( P < . 01) ITA than in a patent left ITA. The peak D/S was significantly smaller in a stenotic right ITA than in a patent left ITA. However, peak D/S did not significantly differ between patent and stenotic left or right ITAs. The mean systolic velocity did not significantly differ among the groups, whereas mean diastolic velocity and mean D/S were significantly lower in stenotic ITAs ( P < . 01, respectively) than in patent left or right ITAs. The mean D/S was significantly lower in patent right ITAs than in patent left ITAs ( P < . 05).
|Patent left ITA |
n = 66
|Stenotic left ITA |
n = 9
|Patent right ITA |
n = 50
|Stenotic right ITA |
n = 19
|Peak systolic velocity (cm/sec)||39.2 ± 14.7||35.5 ± 24.1||44.1 ± 22.2||37.1 ± 22.5|
|Peak diastolic velocity (cm/sec)||37.8 ± 36.4||9.9 ± 4.3 ∗||30.2 ± 17.0||11.3 ± 12.3 †|
|Peak D/S||1.05 ± 1.1||0.37 ± 0.24||0.75 ± 0.48||0.29 ± 0.32 †|
|Mean systolic velocity (cm/sec)||26.1 ± 9.3||23.0 ± 12.9||29.2 ± 17.6||25.4 ± 14.7|
|Mean diastolic velocity (cm/sec)||25.3 ± 9.8||6.3 ± 4.0 †||23.5 ± 15.7 ‡||9.9 ± 10.1 † §|
|Mean D/S||1.01 ± 0.35||0.31 ± 0.20 †||0.83 ± 0.35 ∗ ‡||0.32 ± 0.24 † §|
Table 3 compares Doppler measurements between left anterior descending (LAD) and LCX anastomoses in patent left and right ITAs, respectively. Mean systolic and diastolic velocities and mean D/S did not significantly differ between LAD and LCX anastomoses in left or right ITAs.
|Left ITA||LAD anastomosis |
n = 47
|LCX anastomosis |
n = 19
|Mean systolic velocity (cm/sec)||25.6 ± 9.4||27.4 ± 9.0||NS|
|Mean diastolic velocity (cm/sec)||24.4 ± 9.8||27.2 ± 10.0||NS|
|Mean D/S||0.99 ± 0.32||1.06 ± 0.42||NS|