|Home | About | Journals | Submit | Contact Us | Français|
The aim of this study was to evaluate the accuracy of multislice computed tomography in detecting graft stenosis or occlusion after coronary artery bypass grafting, using coronary angiography as the standard.
From January 2005 through May 2006, 25 patients (19 men and 6 women; mean age, 54 ± 11.3 years) underwent diagnostic investigation of their bypass grafts by multislice computed tomography within 1 month of coronary angiography. The mean time elapsed after coronary artery bypass grafting was 6.2 years.
In these 25 patients, we examined 65 bypass conduits (24 arterial and 41 venous) and 171 graft segments (the shaft, proximal anastomosis, and distal anastomosis). Compared with coronary angiography, the segment-based sensitivity, specificity, and positive and negative predictive values of multislice computed tomography in the evaluation of stenosis were 89%, 100%, 100%, and 99%, respectively. The patency rate for multislice compu-ted tomography was 85% (55/65: 3 arterial and 7 venous grafts were occluded), with 100% sensitivity and specificity.
From these data, we conclude that multislice computed tomography can accurately evaluate the patency and stenosis of bypass grafts during outpatient follow-up.
Coronary artery bypass grafting (CABG) is effective and successful in treating diffuse coronary artery disease. However, recurrences of the symptoms after CABG are documented in 4% to 8% of the patients within 1 year.1 These symptoms are most often due to graft occlusions. Arterial grafts have comparatively good patency rates, but among venous grafts 7% are occluded within 1 week, 15% to 20% are occluded within 1 year, and almost half are occluded within 10 years.1 Evaluation of graft patency is crucial in symptomatic patients after CABG.
Coronary angiography, the gold standard in evaluating graft patency after CABG, is a stressful, invasive procedure, with 1.5% morbidity and 0.15% mortality rates.2 Since the 1980s, several other diagnostic methods (most of them noninvasive)—such as electron-beam tomography,3 helical scan computed tomography,4 transcutaneous Doppler echocardiography,5 digital subtraction angiography,6 magnetic resonance imaging,7 and conventional computed tomography8—have been used to evaluate graft patency. However, none of these tests has been widely accepted as accurate in this application. As a consequence of certain characteristics of venous and arterial conduits (their greater diameter, their direction with reference to the plane of the cross beam, and their relative spatial fixation), multislice computed tomography (MSCT) can evaluate them more effectively than it can evaluate native coronary arteries. Although MSCT assessment of vessels can be hindered by calcification or metallic clips, arterial and venous conduits are typically free of calcification; clips remain a problem in institutions where individual surgeons prefer to use them.9
In the present study, we evaluated the diagnostic accuracy of MSCT in detecting stenosis or occlusion of coronary artery bypass grafts, using coronary angiography as the standard.
Study Group. From January 2005 through May 2006, we performed MSCT on the coronary artery bypass grafts of 29 patients within 1 month of their having been checked by coronary angiography. All patients were in sinus rhythm and clinically stable. Exclusion criteria for MSCT were frequent ectopic beats, atrial fibrillation, previous allergic reaction to the contrast agent, renal insufficiency, hyperthyroidism, severe heart or lung failure, unstable angina, and a heart rate >70 beats/min. We excluded 4 patients: 2 for frequent ectopic beats, 1 for severe allergic reaction to the contrast agent, and 1 for having a heart rate of >70 beats/min, despite a high dose of β-blocker.
The remaining 25 patients (19 men and 6 women; mean age, 54 ± 11.3 years) were included in the study. The mean time elapsed between CABG and MSCT was 6.2 years. A total of 65 conduits (24 in situ arterial and 41 venous) had been implanted on the native coronary arteries at the time of CABG.
All patients received intravenous metoprolol (administered in accordance with body mass and basal heart rate) to reduce the heart rate and received a 0.5-mg nitroglycerin tablet sublingually 2 minutes before the beginning of the procedure. The institutional review board approved the research protocols, and all patients gave their informed consent.
Multislice Computed Tomographic Scan. The multidetector computed tomographic data were acquired using a 16-slice multidetector computed tomographic scanner (SOMATOM Sensation™ 16, Siemens Medical Solutions AG; Erlangen, Germany) with 16 × 0.75-mm detector collimation, a Gantry rotation time of 420 msec, tube current of 550 mA-s, and tube voltage of 120 kV. Eighty to 100 mL of a nonionic contrast agent, iopamidol (Iopamiro®, Bracco S.p.A.; Milano, Italy) (370 mg/mL), was injected intravenously, followed by 50 mL of saline via the “care bolus” technique (rate, 4–5 mL/sec). Axial slices were reconstructed by use of an electrocardiographic gated half-scan reconstruction algorithm (temporal resolution of approximately 210 msec in the center of the scan field), with a slice thickness of 1.0 mm and a reconstruction interval of 0.6 mm. We used a dosemodulation technique to reduce the radiation dose, because in CABG imaging the average dose per patient is approximately 13 to 15 mSv. Patients were instructed to hold their breaths during the scan. The total scan time was 20 to 25 seconds. Besides axial images, multiplanar reformation, maximum intensity projection, curved multiplanar reformation, and reconstructed 3-dimensional volume rendering images were analyzed by 2 experienced cardiologists and 2 radiologists, each working independently of the rest.
Conventional Coronary Angiography. Conventional coronary angiography was performed by means of Philips Integris-H5000 equipment (Philips Medical Systems Nederland B.V.; Best, The Netherlands). The grafts were selectively catheterized. Two cardiologists who had no knowledge of the MSCT findings evaluated the angiograms quantitatively. We stratified bypass grafts into the following categories: patent (no stenosis, <50% stenosis, ≥50% stenosis) or occluded. Significant stenosis was ≥50%.
Statistical Analysis. Conventional quantitative coronary angiography was regarded as the standard of reference.
The diagnostic accuracy of MSCT was expressed in terms of its sensitivity, specificity, negative predictive value, and positive predictive value in regard to its overall accuracy and its application to individual coronary segments.
The coronary angiography and MSCT investigations were completed in all patients without complications. Of the 65 conduits (24 in situ arterial and 41 venous) in 25 patients, there were 171 graft segments (shaft of graft, proximal and distal anastomoses for venous grafts, and shaft of graft, distal anastamosis for arterial grafts) available for analysis. Of the 4 distal graft segments that MSCT could not fully evaluate, 3 could not be evaluated because of hemoclips, and 1 could not be evaluated because of a coronary stent. There were no significant stenoses in those 4 distal anastamotic sites on coronary angiography.
When all bypass conduits were analyzed in regard to patency, 3 arterial and 7 venous conduits were judged to be occluded on both MSCT and coronary angiogra-phy (Table I). Therefore, the sensitivity and specificity of MSCT was 100% in evaluating graft patency.
In regard to significant stenosis, angiography found 4 stenoses in the bodies of the conduits (Fig. 1) and 5 stenoses at anastomotic sites (3 distal and 2 proximal) (Table II). All of these stenoses except for 1 at a distal anastamotic site were also detected by MSCT. The sensitivity and specificity of MSCT for significant stenosis (including all of the segments that could not be evaluated) were 89% and 100%, respectively. The positive predictive value of MSCT was 100% and the negative predictive value was 99%, with a diagnostic accuracy of 99%. When segments that could not be evaluated by means of MSCT were excluded, the sensitivity, specificity, and positive predictive value were the same; the negative predictive value was 99%, and the accuracy was 99% (Table III).
The optimal method to evaluate bypass grafts after coronary revascularization is debatable. Although noninvasive tests such as scintigraphy have been used, the gold standard remains post-bypass coronary angiography. However, coronary angiography is invasive and costly, and it carries procedure-related risks.2 Because of this, alternative, less invasive methods have been investigated for imaging venous and arterial conduits.3–8 These tests are less risky than coronary angiography, but they have many drawbacks. For example, their accuracy in evaluating arterial grafts tends to be lower than their accuracy in evaluating venous grafts, because of the arterial conduits' smaller size and the increased likelihood of encountering metal clips. Other drawbacks of noninvasive methods include breathing artifacts, which reduce image quality, and the consequent need for lengthy breath-holds. Technical improvements in MSCT have enabled this technique to be used for monitoring patency after CABG. Multislice computed tomography has good image quality due to short scanning time and slice thickness, and it is faster than other noninvasive methods.
In our study, we demonstrated a close correlation between the qualitative measurements of coronary stenosis by MSCT and those by conventional coronary angiography. We compared our results with those of other studies in the English-language medical literature.10–14 Our data confirm results from previous studies that showed a good correlation between MSCT and coronary angiography for quantification of the dimensions of nonstenosed coronary artery segments10 and stenotic or occluded grafts.11–14 This study demonstrates the ability of MSCT to quantify the degree of coronary stenosis. Categorization by MSCT of the severity of stenosis correlated well with that of coronary angiography, although a small systematic underestimation by MSCT was observed.
However, the applicability and accuracy of 16-slice MSCT remain subject to several limitations: frequent ectopic beats or atrial fibrillation, a heart rate >70 or 80 beats/min despite therapy, and an inability to hold the breath are all patient variables that can reduce image quality. Radiation exposure and the use of a contrast agent may be a further limitation of the method in some patients. Unlike coronary angiography, MSCT cannot provide any information about the flow characteristics of coronary circulation. This is a substantial limitation, if MSCT is to be used in patients with acute coronary syndromes.
In our experience, MSCT is a very useful and noninvasive test of early and late graft patency in outpatients. Moreover, the equipment and methods involved in MSCT are rapidly progressing such that MSCT might, in the near future, challenge the superiority of coronary angiography in the evaluation of patients with coronary artery disease.
Address for reprints: Baris Caynak, MD, Department of Cardiac Surgery, Florence Nightingale Hospital, Abide-i Hurriyet Cad. No:290, Sisli, Istanbul 80220, Turkey. E-mail: moc.liamtoh@sirabkanyac