Published online before print April 10, 2008, 10.1148/radiol.2473071132
(Radiology 2008;247:679-686.)
© RSNA, 2008
Diagnostic Performance of 16- and 64-Section Spiral CT for Coronary Artery Bypass Graft Assessment: Meta-Analysis1
Michèle Hamon, MD,
Olivier Lepage, MD,
Patrizia Malagutti, MD,
John W. Riddell, MD,
Rémy Morello, MD,
Denis Agostini, MD, PhD, and
Martial Hamon, MD
1 From the Departments of Radiology (Michèle Hamon), Thoracic and Cardiovascular Surgery (O.L.), Nuclear Medicine (D.A.), Cardiology (J.W.R., Martial H.) and Statistics (R.M.), University Hospital of Caen, Avenue Côte de Nacre 14033 Caen, Normandy, France; INSERM 744, Institut Pasteur de Lille, Lille, France (Martial H.); and Department of Cardiology, S. Anna Hospital, Ferrara, Italy (P.M.). Received June 28, 2007; revision requested August 30; revision received October 29; accepted January 4, 2008; final version accepted January 16.
Address correspondence to Michèle H. (e-mail: hamon-mi{at}chu-caen.fr).
 |
ABSTRACT
|
|---|
Purpose: To perform a meta-analysis to evaluate the accuracy of 16- and 64-section spiral computed tomography (CT) to help assess coronary artery bypass grafts (CABGs).
Materials and Methods: The MEDLINE, Cochrane library, and BioMed Central databases were searched for relevant original articles published up to May 2007. Major criteria for article inclusion were that it (a) used multisection CT as a diagnostic test for the assessment of significant lesions (occlusion or >50% stenosis) of CABG, (b) used a 16- or 64-section scanner, and (c) used coronary angiography as the reference standard. After data extraction, the analysis was performed according to a random-effects model. Between-study statistical heterogeneity was also assessed by using the Cochran Q 
2 test.
Results: Of 158 screened articles, 15 fulfilled all inclusion criteria. Graft assessability (including distal anastomosis) ranged from 78%–100% among all included studies (mean, 92.4%; 90% with 16- and 96% with 64-section CT; P < .001). Statistical heterogeneity was observed for specificity and positive likelihood ratio (LR), justifying the use of the random-effects model. The analysis, pooled from 15 studies (723 patients, 2023 CABGs), provided the following results for the assessment of graft obstruction (occlusion and >50% stenosis): sensitivity, 97.6% (95% confidence interval [CI]: 96%, 98.6%); specificity, 96.7% (95% CI: 95.6%, 97.5%); positive predictive value, 92.7% (95% CI: 90.5%, 94.6%); negative predictive value, 98.9% (95% CI: 98.2%, 99.4%); positive LR, 23.42 (95% CI: 13.69, 40.07); negative LR, 0.045 (95% CI: 0.028, 0.071); and diagnostic odds ratio, 780.32 (95% CI: 379.12, 1606.1).
Conclusion: Multisection CT provided high accuracy for the evaluation of CABG obstruction in assessable conduits, and might be used as a noninvasive tool for the evaluation of suspected graft dysfunction in patients who are at high risk for complications from coronary angiography.
© RSNA, 2008
|
INTRODUCTION
|
|---|
Although selective conventional coronary angiography is considered as the reference standard for the assessment of coronary artery bypass graft (CABG), it is an invasive and potentially harmful procedure, which carries a small but definite risk of major complications (1). Recent advances in multisection computed tomography (CT), which might represent a minimally invasive alternative to coronary angiography, have allowed analysis of coronary arteries and CABG. Promising results for bypass patency evaluation were reported with the first-generation (four- and eight-section) multisection CT scanners, but there was a major limitation related to the presence of a large number of unassessable grafts because of motion artifact, presence of surgical wires or clips, and heavy calcification (2). The introduction of 16- and, more recently, 64-section scanners promises to reduce the number of unassessable grafts, improving the diagnostic accuracy of multisection CT for helping detect graft disease.
The purpose of our study was to perform a meta-analysis to evaluate the accuracy of 16- or 64-section spiral CT for the assessment of CABGs.
|
MATERIALS AND METHODS
|
|---|
Search Strategy
Database searches for articles in all languages up to May 2007 were performed in MEDLINE, Cochrane Library, and BioMed Central databases independently by two investigators (Martial H., Michèle H.). We combined the medical subject headings for "computed tomography," "multisection computed tomography" (multisection CT), and "coronary angiography," with the expanded term "coronary artery bypass graft disease." We also scanned references in the articles and reviews retrieved. The studies were carefully examined by the same two investigators to exclude potentially duplicate or overlapping data. Meeting abstracts were excluded, as they could not provide adequately detailed data and their results might not be final. Only studies evaluating the presence of significant obstructive coronary artery bypass disease (>50% lumen reduction) as a result of coronary angiography and multisection CT in the same patients were included. Disagreements were resolved by consensus.
Study Selection and Data Extraction
We included a study if its researchers (a) used multisection CT as a diagnostic test for the existence of occlusion or substantial stenosis (defined as a lumen reduction of >50%) in bypass grafts; (b) used the newest generation of CT scanners (16-section at the least); (c) reported cases in absolute numbers of true-positive, false-positive, true-negative, and false-negative results or presented sufficiently detailed data for deriving these figures; and (d) used coronary angiography as the reference standard for diagnosing obstructive CABG disease.
The same investigators performed the data extraction independently and discrepancies were resolved by consensus. The following information was extracted from each study: first author; year of publication; journal name; study population characteristics including sample size (number of subjects evaluated with both tests without minimal sample size, number of patients excluded), mean age, mean heart rate (and standard deviation), sex, relative timing of the two imaging procedures, blinded evaluation of one test to the result of the other and to the clinical condition of the tested subject, and mean time since surgery; and technical characteristics of the CT scanner including type and brand of machine used, rotation time (in milliseconds), and collimation (in millimeters). Technical parameters for multisection CT acquisition were amount of iodine contrast-enhanced material (in grams), number and type of grafts, and rate of fully assessable grafts. Each graft was analyzed separately. For sequential grafts in which the comparative study considered each distal anastomosis as a separate conduit in its analysis, this procedure was repeated in the analysis. The main analysis was that the evaluation of multisection CT accuracy compared with angiography in the diagnosis of graft obstruction (defined as occlusion and a >50% stenosis) after CABG. When sufficient data were available, separate analysis was performed for occlusion and substantial stenosis. The study quality conformed to the Quality Assessment of Diagnostic Accuracy Studies guidelines (3,4).
Data Synthesis and Statistical Analysis
Categorical variables from individual studies are presented as percentages and continuous variables are presented as median values. Measures of diagnostic accuracy are reported as point estimates with 95% CIs.
By means of true-positive, true-negative, false-positive, and false-negative rates, we computed sensitivity, specificity, positive and negative likelihood ratios (LRs) and diagnostic odds ratios (ORs) as previously described (5).
We computed all statistics for individual studies and then combined them by using a random-effects model, weighting each point estimate by the inverse of the sum of its variance and the between-study variance. Between-study statistical heterogeneity was also assessed by using the Cochran Q 2 test. Weighted symmetric summary receiver operating characteristic curves, with pertinent areas under the curve, were computed by using the Moses-Shapiro-Littenberg method (6–8). The standard error of the log diagnostic OR was plotted against the log diagnostic OR.
Statistical computations were performed with software (SPSS, version 11.0, SPSS, Chicago, Ill) (Meta-DiSc [9]), and alpha significance testing was performed at the two-tailed .05 level (10).
|
RESULTS
|
|---|
Database searches identified 158 potentially relevant citations (Fig 1). After title and abstract assessment, we retrieved, as complete reports, 34 studies, from which 19 were excluded because (a) they did not employ at least a 16-section scanner, (b) they had overlapping data, (c) it was impossible to find or calculate absolute figures from presented data, or (d) no systematic angiographic examination control was performed. We thus included 15 studies in our review (11–25). All studies were published between June 2004 and May 2007 (Table 1).
Diagnostic Performance of Multisection CT
The 15 reports eligible for analysis included 723 patients with bypass grafts. The mean age was 65 years. Of the 640 patients where the sex was specified, there were 543 men. Graft assessability, including evaluation of distal anastomosis, ranged from 78% to 100% for all studies (78%–100% and 87%–100% for 16- and 64-section CT, respectively). Pooled diagnostic performances of these 15 reports (2023 grafts) for the existence of occlusion or substantial stenosis provided the following results: a sensitivity of 97.6% (95% CI: 96%, 98.6%), a specificity of 96.7% (95% CI: 95.6%, 97.5%), a positive predictive value of 92.7% (95% CI: 90.5%, 94.6%), a negative predictive value of 98.9% (95% CI: 98.2%, 99.4%), a positive LR of 23.42 (95% CI: 13.69, 40.07), a negative LR of 0.04 (95% CI: 0.03, 0.07), a diagnostic OR of 780.32 (95% CI: 379.12, 1606.10), and 0.99 area under the curve (95% CI: 0.9882, 0.9968) (Figs 2–6). The studies included in the funnel plot of standard error by using log ORs were distributed symmetrically around the vertical line represented by the pooled log OR and between the guidelines for the 95% CI (Fig 7).
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6: Plot of symmetric summary receiver operating characteristic (SROC) of multisection CT in comparison with coronary angiography. Receiver operating characteristic curve shows diagnostic accuracy by plotting specificity in horizontal axis and sensitivity in vertical axis. Pertinent area under the curve (AUC) and Q* statistic (where sensitivity and specificity are maximal) are also included, both with standard errors (SE).
|
|
Ten studies, including 1308 grafts, provided information for occlusion analysis with a sensitivity of 99.3% (95% CI: 97.3%, 99.9%) and a specificity of 98.7% (95% CI: 97.9%, 99.3%). Nine studies, including 871 nonoccluded grafts, provided information for stenosis analysis with a sensitivity of 94.4% (95% CI: 87.5%, 98.2%) and a specificity of 98.0% (95% CI: 96.7%, 98.8%).
Statistical heterogeneity was observed for specificity and positive LR, justifying the use of the random-effects model (Tables 2, 3).
|
DISCUSSION
|
|---|
The results of our meta-analysis demonstrate that obstructive bypass graft disease can be detected by using at least a 16-section CT with a high diagnostic accuracy with a sensitivity of 98%, a specificity of 97%, a positive predictive value of 93% and a negative predictive value of 99%, albeit with only 92.4% of scanned grafts being fully assessable. This performance is better than for nongraft coronary arteries as shown in a recent meta-analysis, on a per-vessel basis where the sensitivity was 82% and specificity was 91%, for the diagnosis of significant coronary artery stenosis (5). This improvement in performance may be explained by a number of differences between bypass grafts and native coronary arteries. Graft conduits often have a larger diameter, are less frequently calcified and are relatively immobile, making CT imaging less challenging. Moreover, they are more often occluded than stenosed (26), which may make the diagnosis of obstruction more clear cut. Indeed, in the studies where information about occlusion rate was given in addition to the overall rate of substantial stenosis, there was much better detection of occlusion than for substantial stenosis.
There are a number of potential advantages for multisection CT over coronary angiography in the assessment of CABG. Multisection CT is a less invasive technique than coronary angiography. In addition, coronary angiography for CABG is technically more difficult than for native vessels, with a longer duration of procedure, more contrast agent used, and an increased complication rate (19). With coronary angiography, it is sometimes difficult to locate the origin of the grafts and explore them selectively. In one study, 93% of grafts were visualized by using CT, compared with only 86% by using coronary angiography (15). In the case of patients who need repeat cardiac bypass surgery, multisection CT offers the surgeon precise information about the position of the existing grafts and the existence of a calcification of the aorta. In patients who require additional aortic valve surgery, an assessment of aortic root, including size and calcification, is also possible (27).
There are a number of limitations of multisection CT that need to be considered. There are a substantial number of scanned grafts that are not fully assessable (up to 22% in one study that used a 16-section scanner). Reasons for nonassessability are cardiac motion, respiratory artifacts, poor opacification, and the presence surgical clips. However, with a 64-section scanner, with increased temporal and spatial resolution and therefore shorter acquisition time, there is an improvement in the rate of fully assessable scanned grafts compared with a 16-section scanner. Up to 40 seconds were needed for a complete assessment of CABG with a four-section scanner. This has been reduced to less than 30 seconds with a 16-section scanner and to less than 20 seconds with the latest 64-section scanner. However, even with the latest scanners, surgical metal clips and evaluation of distal anastomoses remain challenging.
Besides general limitations of cardiac multisection CT (eg, use of potentially nephrotoxic agents, radiation exposure, and necessary selection of patients, with a priori exclusion of patients with atrial fibrillation and severe respiratory disease), some limitations are specific to the evaluation of patients with CABG. First, data acquisition does not provide information about flow characteristics and the functional state of the graft. Second, severe calcifications and extensive atherosclerotic disease of the native coronary arteries, frequently observed in patients with previous CABG, impair the assessability of the native coronary segments. Third, the increased acquisition volume increases the radiation exposure, and the increased amount of contrast media that is required further increases the risk of nephrotoxicity.
It should be noted that not all of the studies provided sufficient data to evaluate certain important issues. Insufficient information was provided to assess radiation exposure. Also, it was not possible to perform a patient-based analysis, which is more clinically relevant but was only provided in four studies (20,21,23,24). In addition, native vessel analysis was not systematically reported. Only a few studies have investigated both bypass and natives vessels (14,18,20,22,23), providing insufficient data to assess multisection CT performance for coexistent native coronary artery disease in this meta-analysis. However, from a clinical point of view, complete evaluation should include visualization of the native coronary arteries in addition to the grafts. Information about scan volume acquisition was incomplete in the included studies. To limit radiation exposure, some studies did not perform a complete acquisition of the grafts (origin and proximal part of the internal mammary artery grafts were not covered by the scan volume) (11,16,17,19,23). Thus, a relevant stenosis in this segment might remain undetected.
As mentioned previously, statistical heterogeneity has been documented for specificity and positive LR, although the use of the random-effects model should still provide relatively robust results. The well-known tendency toward publication bias favoring studies with positive and encouraging results also complicates comprehensive evaluation, even if, as indicated in the present analysis, the symmetric distribution of the studies included in the funnel plot would suggest that there is not important publication bias.
In summary, the results of our meta-analysis demonstrate a high accuracy of noninvasive multisection CT for the assessment of CABG compared with coronary angiography, although 7.6% of scanned grafts were not fully assessable. Detection of graft occlusion was better than detection of substantial stenosis. The latest multisection CT scanner shows potential to become a first-line tool for the noninvasive evaluation of patients with suspected graft dysfunction.
|
ADVANCES IN KNOWLEDGE
|
|---|
- The results of our meta-analysis by using a random-effects model demonstrate a high accuracy (96.9%) of noninvasive multisection CT for the assessment of coronary artery bypass graft (CABG) compared with coronary angiography, although 7.6% of scanned grafts were not fully assessable.
- Sensitivity for detection of graft occlusion (99.3%) was better than for detection of substantial stenosis (94.4%).
|
IMPLICATION FOR PATIENT CARE
|
|---|
- Multisection CT provided high accuracy (96.9%) for the evaluation of CABG obstruction in assessable conduits and might be used as a noninvasive tool for the evaluation of patients suspected of having graft dysfunction.
|
FOOTNOTES
|
|---|
Abbreviations: CABG = coronary artery bypass graft • CAD = coronary artery disease • LR = likelihood ratio • OR = odds ratio
Author contributions: Guarantors of integrity of entire study, Michèle H., Martial H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, Michèle H., Martial H.; statistical analysis, Michèle H., R.M., Martial H.; and manuscript editing, all authors
Authors stated no financial relationship to disclose.
|
References
|
|---|
- Gobel FL, Stewart WJ, Campeau L, et al. Safety of coronary arteriography in clinically stable patients following coronary bypass surgery: Post CABG Trial Investigators. Cathet Cardiovasc Diagn 1998;45:376–381.
- Stein PD, Beemath A, Skaf E, et al. Usefulness of 4-,8- and 16-slice computed tomography for detection of graft occlusion or patency after coronary artery bypass grafting. Am J Cardiol 2005;96:1669–1673.[CrossRef][Medline]
- Whiting P, Harbord R, Kleijnen J. No role for quality scores in systematic reviews of diagnostic accuracy studies. BMC Med Res Methodol 2005;5:19.[CrossRef][Medline]
- Deville WL, Buntinx F, Bouter LM, et al. Conducting systematic reviews of diagnostic studies: didactic guidelines. BMC Med Res Methodol 2002;2:9.[CrossRef][Medline]
- Hamon M, Biondi-Zoccai G, Malagutti P, et al. Diagnostic performance of multislice spiral computed tomography of coronary arteries as compared against conventional invasive coronary angiography: a meta-analysis. J Am Coll Cardiol 2006;48:1896–1910.[Abstract/Free Full Text]
- Moses LE, Shapiro D, Littenberg B. Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. Stat Med 1993;12:1293–1316.[Medline]
- Walter SD. Properties of the summary receiver operating characteristic (SROC) curve for diagnostic test data. Stat Med 2002;21:1237–1256.[CrossRef][Medline]
- Lijmer JG, Bossuyt PM, Heisterkamp SH. Exploring sources of heterogeneity in systematic reviews of diagnostic tests. Stat Med 2002;21:1525–1537.[CrossRef][Medline]
- Zamora J, Abraira V, Muriel A, Khan K, Coomarasamy A. Meta-DiSc: a software for meta-analysis of test accuracy data. BMC Med Res Methodol 2006;6:31.[CrossRef][Medline]
- Halligan S, Altman DG. Evidence-based practice in radiology: step 3 and 4—appraise and apply systematic reviews and meta-analysis. Radiology 2007;243:13–27.[Abstract/Free Full Text]
- Martuscelli E, Romagnoli A, D'Eliseo A, et al. Evaluation of venous and arterial conduit patency by 16-slice spiral computed tomography. Circulation 2004;110:3234–3238.[Abstract/Free Full Text]
- Schlosser T, Konorza T, Hunold P, Kühl H, Schmermund A, Barkhausen J. Noninvasive visualization of coronary artery bypass grafts using 16-detector row computed tomography. J Am Coll Cardiol 2004;44:1224–1229.[Abstract/Free Full Text]
- Chiurlia E, Menozzi M, Ratti C, Romagnoli R, Modena MG. Follow-up of coronary artery bypass graft patency by multislice computed tomography. Am J Cardiol 2005;95:1094–1097.[CrossRef][Medline]
- Salm LP, Bax JJ, Jukema W, et al. Comprehensive assessment of patients after coronary artery bypass grafting by 16-detector-row computed tomography. Am Heart J 2005;150:775–781.[CrossRef][Medline]
- Trigo Bautista A, Estornell J, Ridocci F, et al. Non-invasive assessment of coronary artery bypass grafts by computed tomography: comparison with conventional coronary angiography [in Spanish]. Rev Esp Cardiol 2005;58:807–814. [Published correction appears in Rev Esp Cardiol 2005;58(10):1248.][CrossRef][Medline]
- Vernhet-Kovacsik H, Battistella P, Demaria R, et al. Early postoperative assessment of coronary artery bypass graft patency and anatomy: value of contrast-enhanced 16-MDCT with retrospectively ECG-gated reconstructions. AJR Am J Roentgenol 2006;186(6 suppl 2):S395–S400.
- Anders K, Baum U, Schmid M, et al. Coronary artery bypass graft (CABG) patency: assessment with high resolution submillimeter 16-slice multidetector-row computed tomography (MDCT) versus coronary angiography. Eur J Radiol 2006;57:336–344.[CrossRef][Medline]
- Burgstahler C, Beck T, Kuettner A, et al. Non-invasive evaluation of coronary artery bypass grafts using 16-row multi-slice computed tomography with 188 ms temporal resolution. Int J Cardiol 2006;106:244–249.[CrossRef][Medline]
- Yamamoto M, Kimura F, Niinami H, Suda Y, Ueno E, Takeuchi Y. Noninvasive assessment of off-pump coronary artery bypass surgery by 16-channel multidetector-row computed tomography. Ann Thorac Surg 2006;81:820–827.[Abstract/Free Full Text]
- Malagutti P, Nieman K, Meijboom WB, et al. Use of 64-slice CT in symptomatic patients after coronary bypass surgery: evaluation of grafts and coronary arteries. Eur Heart J 2007;28:1879–1885.[Abstract/Free Full Text]
- Pache G, Saueressig U, Frydrychowicz A, et al. initial experience with 64-slice cardiac CT: non-invasive visualization of coronary artery bypass grafts. Eur Heart J 2006;27:976–980.[Abstract/Free Full Text]
- Dikkers R, Willems TP, Tio RA, Anthonio RL, Zijlstra F, Oudkerk M. The benefit of 64-MDCT prior to invasive coronary angiography in symptomatic post-CABG patients. Int J Cardiovasc Imaging 2006;23:369–377.[Medline]
- Ropers D, Pohle FK, Kuettner A, et al. Diagnostic accuracy of noninvasive coronary angiography in patients after bypass surgery using 64-slice spiral computed tomography with 330-ms gantry rotation. Circulation 2006;114:2334–2341.[Abstract/Free Full Text]
- Meyer TS, Martinoff S, Hadamitzky M, et al. Improved non-invasive assessment of coronary artery bypass grafts with 64-slice computed tomographic angiography in an unselected patient population. J Am Coll Cardiol 2007;49:946–950.[Abstract/Free Full Text]
- Jabara R, Chronos N, Klein L, et al. Comparison of multidetector 64-slice computed tomographic angiography to coronary angiography to assess the patency of coronary artery bypass grafts. Am J Cardiol 2007;99:1529–1534.[CrossRef][Medline]
- Bourassa MG, Fisher LD, Campeau L, Gillespie MJ, McConney M, Lesperance J. Long-term fate of bypass grafts: the coronary artery surgery study (CASS) and Montreal Heart Institute experiences. Circulation 1985;72(6 pt 2):V71–V78.[Medline]
- Feuchtner GM, Schachner T, Bonaros N, et al. Evaluation of ascending aortic atherosclerosis with 16-multidetector computed tomography is useful before total endoscopic coronary bypass surgery. Heart Surg Forum 2006;9:E754–E758.[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
S. Plein
Noninvasive Assessment of Coronary Artery Bypass Graft Disease: The Potential Role of Contrast-Enhanced Cardiac Magnetic Resonance
J. Am. Coll. Cardiol. Img.,
April 1, 2009;
2(4):
446 - 448.
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
