|
|
||||||||
Ann Thorac Surg 2006;82:44-50
© 2006 The Society of Thoracic Surgeons
a Department of Cardiac Surgery, Azienda Ospedaliera di Perugia, Perugia, Italy
b Department of Imaging Diagnostics, AUSL 2, Assisi, Italy
c Radiology Institute, Università Politecnica delle Marche, Ancona, Italy
Accepted for publication March 7, 2006.
* Address correspondence to Dr Di Lazzaro, S. C. di Cardiochirurgia, Ospedale R. Silvestrini, S. Andrea delle fratte 06156 Perugia, Italy (Email: davide.dilazzaro{at}ospedale.perugia.it).
| Abstract |
|---|
|
|
|---|
METHODS: Between April and October 2005, we reviewed all patients operated on by coronary artery bypass grafting at our institution between January and December 2002. A total of 62 patients received a radial artery graft. Of these, 22 were lost at the time of follow-up. The other 40 patients were enrolled for a multidetector row computed tomography study. Demographic and instrumental data were collected for all the patients. A total number of 145 grafts were studied, with complete and excellent visualization.
RESULTS: The scans revealed a 97.77% (44 of 45) patency rate for left internal mammary arteries, 90.57% (48 of 53) for vein grafts, and 73.91% (34 of 46) for radial arteries (mammary artery plus vein grafts versus radial artery patency, p < 0.001). If analyzed for target vessel, we found the poorest result of radial grafts when placed on the right coronary artery (40% [2 of 5] patency rate).
CONCLUSIONS: Noninvasive control of previously bypassed patients is feasible, with no discomfort for them and excellent visualization of grafts. The use of the radial artery as a conduit for bypass graft can be achieved with good results, after a careful choice of the target vessel.
| Introduction |
|---|
|
|
|---|
Our aim was to evaluate whether the use of MDCT in a follow-up study was feasible and easy to set up, and to review our experience with RA grafts.
| Material and Methods |
|---|
|
|
|---|
The other 40 patients were enrolled for an MDCT angiography. Exclusion criteria were allergies to iodinated contrast media, renal (creatinine level > 200 mg/dL), respiratory or cardiac failure, cardiac arrhythmias, or heart rate greater than 70 beats/min. We collected demographic and instrumental data (echocardiography, exercise stress test) before MDCT for all the patients. None of them was excluded from the study. Baseline demographic data are shown in Table 1. The study was approved by the local ethics committee, and informed consent was obtained from all patients.
|
Harvesting Technique
The RA was harvested with an open method from the nondominant forearm, along with both satellite veins, using low-strength electrocautery. The RA was treated with topical papaverine, eventually injected intraluminally at the end of the harvesting.
Vasodilator Therapy
Intraoperative vasodilator therapy was intravenous nitroglycerin, followed postoperatively by oral nitrates, which were discontinued after 6 months.
Computed Tomography Scanning Protocol
Contrast-enhanced MDCT was performed by administering 110 mL of low osmolar nonionic high-concentration contrast medium (iomeprol, Iomeron 400; Bracco, Milan, Italy) through a dual-head power injector (Stellant, Medrad, Indianola, PA) through an 18- to 20-gauge needle inserted into an antecubital vein at a rate of 4 mL/s, followed immediately by 40 mL of saline flush at a rate of 4 mL/s. For optimal contrast enhancement, the scan delay was determined by the bolus-tracking technique, which positioned the region of interest in the ascending aorta. Electrocardiographic-gated computed tomographic data were collected with a 16 x 0.75 collimation, a 420-ms rotation time, a pitch of 0.3 with a tube current of 500 mA at 120 KV. The field of view was 150 to 200 mm with an image matrix of 512 x 512 pixels. Scanning times varied from 15 to 20 seconds. All scans were reconstructed by using retrospective gating (62.5% to 75% of the R-R interval) and multisegment reconstruction, with 1-mm thick images reconstructed every 0.5 mm. The cardiac phase with the least motion was chosen for further evaluation.
Computed tomographic scans were obtained by using a 16-MDCT scanner (Philips Brilliance 16; Philips Medical Systems, Best, the Netherlands). After a scout image, the scan coverage was selected from the carina to the apex of the heart in the cephalic to caudal direction to include the maximum extent of the grafts. To maintain heart rate less than 60 beats/min, all patients received oral ß-blockers (atenolol 50 to 100 mg/day, according to body weight and basal heart rate) 2 to 3 days before examination; the mean heart rate in our cohort was 54 ± 11.3 beats/min.
Image Processing and Analysis
Data sets were reviewed using a separate PC-based workstation (MX View; Philips Medical Systems), and image interpretation was based on evaluation of axial source, multiplanar reconstructions, maximum-intensity projections, and volume-rendering images. Advanced software tools (arterial tree, curved multiplanar reformation along centerline of the vessel, vascular two-dimensional or three-dimensional map, virtual angioscopy) were used in single cases. Images were evaluated independently by three radiologists (A.G., F.C., W.P.) who were aware of the previous CABG surgery type. A consensus opinion was reached among all three in discordant cases. Each graft was screened separately.
Concordance among readers for the detection of occlusion or significant conduit stenosis by MDCT was calculated by the Cohen
value, according to the formula
= Io Ie/1 Ie, where Io is the observed concordance and Ie the expected concordance. The
value for interobserver variation in the detection of patency and significant stenoses of conduits was 0.86 for reader 1 and 0.83 for reader 2.
Data Collection and Statistical Analysis
All the data were collected prospectively. Percentages are given to describe categorical variables. Statistical analysis of the data was performed with the Statistical Analysis Software (SAS) version 8.2 for Windows system. Fisher's exact test was used to analyze categorical variables. A probability value of less than 0.05 was considered significant. In the absence of conclusive MDCT variables to define CABG patency, we used the following three-point scale, according to angiographic studies presented in the literature: grade 0 was defined as stenosis
70% of the lumen (Fig 1), grade 1 was defined as a stenosis of 70% to 99% (Fig 2), and grade 2 corresponds to complete occlusion of the graft (Fig 3). Both grade 0 and grade 1 were considered patent.
|
|
|
| Results |
|---|
|
|
|---|
Angiographic Results
A total of 40 patients were studied, for a total number of 145 grafts. Of these, 45 (39 single, 6 sequential anastomoses) were LIMAs, 53 (45 single, 8 sequential) were SVGs, 46 (41 single, 5 sequential) were RAs, and 1 was a right gastroepiploic artery. The best overall patency was for LIMA (97.77% patency, 44 of 45), followed by SVG (90.57%, 48 of 53), and by RA (73.91%, 34 of 46); the only right gastroepiploic artery we studied was patent. Table 2
reports the patency and occlusion in the entire cohort, by type of segment.
|
|
| Comment |
|---|
|
|
|---|
Potential advantages of MDCT compared with SCA in postoperative follow-up of CABG patients are many, ranging from the possibility of performing a noninvasive procedure without hospitalization to discerning ancillary findings (a case of early-stage pulmonary neoplasm in our series) to diagnosing early and late extracardiac complications presenting with chest pain, such as pericardial or pleural effusion, sternal infection, or pulmonary embolism [22]. In addition, MDCT is useful in preoperative planning before redo CABG surgery because of its superiority compared with SCA in defining the position of patent grafts and vital mediastinal structures [22, 25].
All segments were depicted with adequate diagnostic quality and were capable of being evaluated. From our experience, three-dimensional volume-rendered images, thanks to their panoramic overview, are very useful in the depiction of complex anatomy of the grafts and, used as an initial tool together with two-dimensional axial images, facilitate subsequent vessel interpretation. Thin-slab maximum-intensity projection images best perform visualization of the entire graft course and anastomoses. We did not consider the coronary artery collaterals in the MDCT analysis for two reasons: first, there are not data about the utility of MDCT studies at this regard; second, we believe that to assess viable myocardium there are more reliable methods other than MDCT.
Regarding the use of the RA, all the anastomoses were visible. Evaluation of distal anastomosis still represents a challenge owing to artifacts from vascular metal clips with beam hardening and partial volume effect. In the series by Khan and associates [18], assessment of distal anastomosis yielded no difference between 4-slice and 16-slice MDCT technology. Anyway, our data confirm those of others [23] with a reported eligibility for distal bypass anastomoses ranging from 74% to 99%. To maximize the visualization of distal anastomoses we reduced motion artifacts by using recommended R-R reconstruction intervals at 62.5% to 75% to evaluate vessels grafted to the left anterior descending coronary artery or circumflex territory and at 50% of the cardiac cycle to visualize vessels grafted to the right coronary artery territory. When necessary, we shifted the reconstruction intervals.
Although Zacharias and coworkers [13] and Tatoulis and colleagues [12] considered string-sign as occlusions, in our series, according to the Radial Artery Patency Study Group [1], we considered grade 1 grafts as patent. Concerning this, a potential source of misinterpretation of graft patency with SCA (mechanical "catheter-tip spasm") can be minimized with MDCT. At any rate, the RA, although narrowed, is still capable of responding to increased myocardial oxygen demand; the corresponding string-sign seems to be an active process of vascular remodeling related to an adaptive response to low flow demand. Furthermore, an increased myocardial oxygen demand is associated with a positive remodeling characterized by luminal enlargement with secondary increased wall shear stress (according to Poiseuille's law) and increased deterioration of the graft in the long-term, as previously stated [9]. The considerations about graft remodeling are supported by the evidence that ischemia is uncommon in territories supplied by grafts with angiographic string-sign and, in our case, even in regions supplied by an occluded arterial RA graft, thanks to the establishment of collateral pathways. This consideration can explain the absence of signs or symptoms of ischemia in the vast majority of our patients with occluded RA [3].
Angiographic Results
Starting from the new evidence provided by Acar and colleagues [4] about the patency of RA grafts, a great number of papers appeared among international literature. Almost all of the most relevant papers [1, 413], all angiography-based, reported good short-term, midterm, and long-term patency rates for the RA, except for one by Khot and colleagues [14], which reported, in patients with sign or symptoms of myocardial ischemia, poorer results in terms of patency rate of RAs. The examination of angiographic results of symptomatic patients [15] did not recognize a superiority in use of the RA in terms of patency compared with both the LIMA and the right internal mammary artery.
Even if the majority of studies tends toward a general agreement about good patency rates of the RA, there is not a general consensus about its real role as a second-choice arterial conduit after the LIMA. We eagerly await the 10-year results of the Radial Artery Patency and Clinical Outcome Study, and of the Radial Artery Patency Study Group, whose interim results were recently published [10, 14]; both studies, at the end of their 10-year follow-up in symptom-free patients, should be able to say a definitive word about long-term patency rate of RA grafts.
Apart from this, a general consensus on use of RA grafts was reached about the quality, grade of stenosis, and location of target vessels. The RA should be used to bypass coronary arteries carrying critical stenoses (>70%), avoiding, if possible, the right coronary artery, as it carries worse patency rates than other arteries [3, 5, 10, 26]. Also the type of proximal anastomosis seems to affect patency rates of the RA even if, in this case, a more widespread consensus is lacking; three studies [3, 5, 16] did not find any difference in patency rates regardless of the type of proximal anastomosis (free graft to ascending aorta, or composite graft to LIMA or right internal mammary artery), whereas an angiographic, nonrandomized retrospective study found differences in patency [26].
From the point of view of results, our study, conducted in all patients operated on with an RA graft regardless of symptoms (thus avoiding the bias presented by those studies that were ischemia-directed), showed, as expected, the best patency rates for use of the LIMA, but also showed patency rates for SVG being better than RA (90.57% versus 73.91%). Moreover, the majority of our patients were asymptomatic at the time of study, and only 2 of the 7 symptomatic patients had one graft closed. The causes of such results can be many: first, the time of follow-up may be too short to expect atherosclerosis to begin affecting SVG grafts, which could be affected later in their postoperative life, 5 to 10 years out. Second, the choice of target vessel may affect patency; even if fewer in number compared with the grafts on left coronary arteries, the RA grafts on the right coronary artery and its branches had a worse performance in terms of patency rate (40% versus a cumulative 78.1%), confirming data from previous studies. Moreover, as we were at our initial experience with use of the RA, we frequently considered it as third conduit of choice after the LIMA and SVG, placing it on less important vessels.
This fact could also explain why 9 patients of the group of 33 asymptomatic patients had closed RA grafts. Another cause could be related to the grade of collateralization of target vessels. A high grade of collateral flow could determine a highly competitive situation, even in the presence of total occlusion [15]; on the other side, a closure of the RA could be compensated by the collateral circulation, avoiding the development of myocardial ischemia.
Female sex is another factor limiting patency rates of the RA [14]. In our study the number of women is too small to allow any consideration (3 of 40) but, also in this small group, 1 patient had a patent RA, and 2 had a closed RA.
Limitations of This Study
A limitation of our study is the lack of validation of CABG patency with the gold standard of coronary angiography. However, considering the larger size, the relative independence of heart rate compared with native coronary arteries, and the high diagnostic quality of a state-of-art MDCT scanner, already validated with coronary angiography [24], we thought that correlation with this technique was not necessary. Thus, our focus was based on the reliability of the 16-slice MDCT in evaluating midterm status of RAs used as grafts. Another limitation is the relatively short follow-up period, compared with other, larger studies in terms of number of patients. The last limitation could be the relatively high number of patients lost to follow-up. From a speculative point of view, these patients, which in the majority of cases left as their only contact information cell phone numbers which were easily changed during the years, may have died as a result of RA occlusion. But, considering that among the patients studied the occlusion of the RA did not lead to such a dramatic event, and that the RA was not our first-choice graft, this hypothesis could be considered as very remote.
Future of Multidetector Row Computed Tomography
A new generation of MDCT scanners has recently been introduced, allowing the simultaneous acquisition of 64 slices per rotation with improved spatial and temporal resolution with reduced scanning time. The ability to visualize smaller vessels with reduced blooming effects by metal clips and to freeze images within the cardiac cycle provides encouraging results [27]; however, there are no definitive results regarding CABG assessment compared with 16-slice MDCT.
Conclusions
The 16-slice MDCT scanner, allowing accurate analysis of bypass grafts, represents a useful noninvasive diagnostic tool in clinically routine follow-up of CABG patients. Maximum-intensity projection images provide the best visualization of the graft and anastomoses during follow-up, and multiplanar reconstructions are superior to maximum-intensity projections, especially if surgical clips are present. The RA, considering only the patients investigated, is a good choice as an alternative conduit, but requires the right selection of patients and target coronary arteries.
| Notice From the American Board of Thoracic Surgery Regarding Trainees and Candidates for Certification Who Are Called to Military Service Related to the War on Terrorism |
|---|
|
|
|---|
Carolyn E. Reed, MD Chair
The American Board of Thoracic Surgery
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Collins, C. M. Webb, C. F. Chong, N. E. Moat, and for the Radial Artery Versus Saphenous Vein Patenc Radial Artery Versus Saphenous Vein Patency Randomized Trial: Five-Year Angiographic Follow-Up Circulation, June 3, 2008; 117(22): 2859 - 2864. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J. Gibbons, P. A. Araoz, and E. E. Williamson The Year in Cardiac Imaging J. Am. Coll. Cardiol., September 4, 2007; 50(10): 988 - 1003. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |