CT Angiography in the Assessment of Saphenous Vein Graft Disease
Editor-In-Chief: C. Michael Gibson, M.S., M.D. [1]
There is interest in the non-invasive assessment of SVG patency using CT imaging. In a recent study that used SVG patency as the primary endpoint, 224/249 (90.3%) of patients returned for follow-up CT at 3 months. There were 704 grafts in these 224 patients. Only 11 segments were non-diagnostic (1.6%). In this relatively modest sized study, there was a significant improvement in SVG patency to 91.6% (219/239) among patients treated with both aspirin and clopidogrel versus 85.7% (198/231) in those patients treated with aspirin alone (relative risk: 1.707; 95% confidence interval: 1.010 to 2.886; p = 0.043)[1].
Early Studies Using 16 Slice CT
Early studies demonstrated a sensitivity of 92% to 100% and a specificity of 89% to 100% in the evaluation of SVG patency with invasive angiography as the gold standard. In the same studies, in the detection of IMA patency, the sensitivity has ranged from 80% to 100% and the specificity has ranged from 82% to 100% [2][3][4].
Methodologic Issues
There are methodologic issues that should be highlighted when evaluating comparisons of CT to angiography as the gold standard. In the above studies, between 4% and 29% of the SVGs were unevaluable[2][3][4]. In the study by Anand, only those patients with disease on CT underwent angiography [5]. More than half of the patients in the study did not undergo invasive angiography, so statements regarding sensitivity and specificity cannot be made. 6 out of 102 SVGs could not be evaluated. Among those selected patients who underwent angiography, there were two false positive diagnoses of a stenosis, and there was "100% predictice accuracy in detecting graft occlusion, and 85% predictive accuracy in detecting graft stenosis". Again, this is an overstatement given that not all patients underwent protocol mandated angiography.
Contemporary Studies Using 64 Slice CT
The resolution of CT angiography improved with the introduction of 64 slice CT. The assessment of the presence of either a stenosis or occlusion of SVG has been associated with high rates of sensitivity of (97%), specificity (97%), and positive and negative predictive values of 93% and of 99%, respectively [6]. Arrhythmias and heart rates > 65 beats/minute during scanning were associated with errors in assessment. It should be noted that in modern studies such as the one above, 12 segments were excluded from analysis because of the presence of stents which are associated with blooming artifact. IT should also be noted that 9 out of 406 grafts demonstrated insufficient image quality for the assessment of the SVGs due either to motion artifacts (8 SVGs) or numerous metallic clips adjacent to the bypass graft (1 SVG). When the data is analyzed on a per patient basis rather than a per SVG basis, among evaluable patients the following were observed [6]:
- Sensitivity: 100% (95% CI 94% to 100%)
- Specificity: 92% (95% CI 82% to 97%)
- Positive predictive value (PPV): 93% (95% CI 85% to 97%)
- Negative predictive value (NPV): 100% (95% CI 93% to 100%)
When all patients were included in the analysis (including those in whom the CT was unevaluable) the rates were as follows [6]:
- Sensitivity: 100% (95% CI 94% to 100%)
- Specificity: 87% (95% CI 76% to 93%)
- Positive predictive value (PPV): 89% (95% CI 79% to 94%)
- Negative predictive value (NPV): 100% (95% CI 93% to 100%)
It should also be noted that these patients were symptomatic and the pre-test probability was high. These rates may be poorer in an asymptomatic population with a low pre-test probability.
Radiation Dose with CT Scans
It should be noted that the radiation dose is more than twice as high for CT scans as it was for invasive angiography (17.8 +/- 5.4 mSv and 8.8 +/- 4.5 mSv, respectively, p < 0.05)[6].
Simultaneous Assessment of Native and SVG Disease
The assessment of SVGs is simplified by the fact that the SVGs are stationary while there is motion artifact in the assessment of native coronary arteries. The assessment of native coronary artery disease in addition to SVG conduits was investigated by Ropers et al [7]. Fifty patients with a total of 138 arterial and venous conduits were assessed a mean of 106 months after CABG. All the arterial and the venous conduits were both evaluable and were correctly classified as either being occluded (n=38) or patent (n=100) when compared with angiography as the "gold standard". With respect to SVG stenosis severity, Sensitivity for stenosis detection in patent grafts was 100% (16/16) with a specificity of 94% (79/84). For the per-segment evaluation of native coronary arteries and distal runoff vessels, sensitivity in evaluable segments (91%) was 86% (87/101) with a specificity of 76% (354/465). If evaluation was restricted to nongrafted arteries and distal runoff vessels, sensitivity was 86% (38/44) with a specificity of 90% (302/334). On a per-patient basis, classifying patients with at least 1 detected stenosis in a CABG, a distal runoff vessel, or a nongrafted artery or with at least 1 unevaluable segment as "positive," MDCT yielded a sensitivity of 97% (35/36) and specificity of 86% (12/14).
Incidental Findings on CT
One study indicated that about 20% of patients undergoing cardiac CT have an incidental finding at the time of the imaging. 9.3% of patients had a cardiac finding such as a pseudoaneurysm or intracardiac thrombus, and 13.1% of patients had a noncardiac finding including pulmonary embolism, lung cancer, or pneumonia[8].
References
- ↑ Gao G et al. Aspirin plus clopidogrel therapy increases early venous graft patency after coronary artery bypass surgery. J Am Coll Cardiol 2010;56:1639–43.
- ↑ 2.0 2.1 Achenbach S, Moshage W, Ropers D, Nossen J, Bachmann K (1997). "Noninvasive, three-dimensional visualization of coronary artery bypass grafts by electron beam tomography". The American Journal of Cardiology. 79 (7): 856–61. PMID 9104894. Unknown parameter
|month=ignored (help) - ↑ 3.0 3.1 Lu B, Dai RP, Zhuang N, Budoff MJ (2002). "Noninvasive assessment of coronary artery bypass graft patency and flow characteristics by electron-beam tomography". The Journal of Invasive Cardiology. 14 (1): 19–24. PMID 11773691. Unknown parameter
|month=ignored (help) - ↑ 4.0 4.1 Stanford W, Brundage BH, MacMillan R, Chomka EV, Bateman TM, Eldredge WJ, Lipton MJ, White CW, Wilson RF, Johnson MR (1988). "Sensitivity and specificity of assessing coronary bypass graft patency with ultrafast computed tomography: results of a multicenter study". Journal of the American College of Cardiology. 12 (1): 1–7. PMID 3288675. Unknown parameter
|month=ignored (help) - ↑ Anand DV, Lim E, Lipkin D, Lahiri A (2008). "Evaluation of graft patency by computed tomographic angiography in symptom-free post-coronary artery bypass surgery patients". Journal of Nuclear Cardiology : Official Publication of the American Society of Nuclear Cardiology. 15 (2): 201–8. doi:10.1016/j.nuclcard.2007.10.007. PMID 18371591.
- ↑ 6.0 6.1 6.2 6.3 Meyer TS, Martinoff S, Hadamitzky M, Will A, Kastrati A, Schömig A, Hausleiter J (2007). "Improved noninvasive assessment of coronary artery bypass grafts with 64-slice computed tomographic angiography in an unselected patient population". Journal of the American College of Cardiology. 49 (9): 946–50. doi:10.1016/j.jacc.2006.10.066. PMID 17336717. Unknown parameter
|month=ignored (help) - ↑ Ropers D, Pohle FK, Kuettner A, Pflederer T, Anders K, Daniel WG, Bautz W, Baum U, Achenbach S (2006). "Diagnostic accuracy of noninvasive coronary angiography in patients after bypass surgery using 64-slice spiral computed tomography with 330-ms gantry rotation". Circulation. 114 (22): 2334–41, quiz 2334. doi:10.1161/CIRCULATIONAHA.106.631051. PMID 17088461. Retrieved 2010-10-31. Unknown parameter
|month=ignored (help) - ↑ Mueller J, Jeudy J, Poston R, White CS (2007). "Cardiac CT angiography after coronary bypass surgery: prevalence of incidental findings". AJR. American Journal of Roentgenology. 189 (2): 414–9. doi:10.2214/AJR.06.0736. PMID 17646469. Unknown parameter
|month=ignored (help)