|Year : 2017 | Volume
| Issue : 3 | Page : 832-836
Coronary stent patency and in-stent re-stenosis in patients referred to multislice computed tomography coronary angiography
Ahmed A Reda, Ahmed M El-Kersh, Morad B Mena, Hend M Abdo El-Deeb MBBCh
Department of Cardiology, Faculty of Medicine, Menoufia University, Menoufia, Egypt
|Date of Submission||17-Jul-2016|
|Date of Acceptance||19-Jan-2017|
|Date of Web Publication||15-Nov-2017|
Hend M Abdo El-Deeb
Nasr City, Cairo, 11765
Source of Support: None, Conflict of Interest: None
Our aim was to assess the diagnostic accuracy of multislice computed tomography (CT) coronary angiography in the evaluation of coronary stent patency using different methods of image reconstruction.
Over the past 25 years, catheter-based intervention has become the dominant form of coronary revascularization. Although the use of recently introduced drug-eluting stents has resulted in even further reductions in the occurrence of re-stenosis, in-stent thrombosis and neointimal hyperplasia may still occur and cause partial or complete obstruction. Conventional coronary angiography is still the technique of choice for the diagnosis of in-stent re-stenosis, but may involve major complications; therefore, multidetector CT can be useful to assess the condition of the whole coronary tree, as it provides information about the number, severity, and location of coronary lesions.
Patients and methods
We studied 30 patients with past history of percutaneous coronary intervention who were referred for coronary multislice CT coronary angiography using different methods of image reconstruction, either filtered back projection or iterative reconstruction, compared with the standard coronary angiography.
By comparing filtered back projection and iterative reconstruction, we found that I46f showed statistically lower noise than B46f (29.8 ± 3.9 vs. 36.2 ± 3.2) (P < 0.001) regarding signal-to-noise ratio (14.8 ± 2.1 for I46f vs. 12.0 ± 3.7 for B46f) (P < 0.001). The contrast-to-noise ratio was also statistically better with I46f than with B46f (7.3 ± 0.9 for I46f vs. 5.9 ± 0.5 for B46f) (P < 0.001).
We concluded that the sharp kernel for each filter has higher image noise than the medium kernel, and when comparing both filters together we found that iterative reconstruction, sharp kernel (I46f), has the best image quality.
Keywords: filtered back projection, iterative reconstruction, multislice computed tomography, sharp kernel, smooth kernel
|How to cite this article:|
Reda AA, El-Kersh AM, Mena MB, Abdo El-Deeb HM. Coronary stent patency and in-stent re-stenosis in patients referred to multislice computed tomography coronary angiography. Menoufia Med J 2017;30:832-6
|How to cite this URL:|
Reda AA, El-Kersh AM, Mena MB, Abdo El-Deeb HM. Coronary stent patency and in-stent re-stenosis in patients referred to multislice computed tomography coronary angiography. Menoufia Med J [serial online] 2017 [cited 2018 Sep 25];30:832-6. Available from: http://www.mmj.eg.net/text.asp?2017/30/3/832/218271
| Introduction|| |
The most important advance in the field of percutaneous coronary interventions has been the introduction of coronary stent implantation in the 1990s, which led to reductions in both the risk of acute major complications and the incidence of re-stenosis, compared with the risks after balloon angioplasty ,. Although the use of recently introduced drug-eluting stents has resulted in even further reductions in the occurrence of re-stenosis, in-stent thrombosis and neointimal hyperplasia may still occur and cause partial or complete obstruction.
Moreover, although several characteristics of high-risk populations have been described as clinical predictors, the likelihood of re-stenosis in a particular patient remains largely unpredictable ,,,. For these reasons, conventional coronary angiography is still the technique of choice for the diagnosis of in-stent re-stenosis, although cardiac catheterization may involve major complications and is associated with moderate-to-high costs. Hence, the latest generation of multidetector (multisection) computed tomography (CT) scanners, which offer a smaller voxel size, faster gantry rotation speed, and reconstruction of sections per gantry rotation, provide an appealing alternative for noninvasive luminal assessment in patients with chest pain after coronary stent placement .
| Patients and Methods|| |
We studied 30 Egyptian patients with a past history of percutaneous coronary intervention who were referred for coronary multislice CT coronary angiography at Kobry El Quba Military Hospital and As-Salam International Hospital in the period between 25 November 2014 and 25 May 2015.
After informed consent was obtained from all patients, they were subjected to the following.
Full history taking, thorough clinical examination, laboratory investigations, and CT coronary angiography using Siemens Somatom Force (Siemens Healthcare GmbH Henkestr, Erlangen, Germany) 256-slice with the patients in the supine position. Three ECG leads were attached to obtain an adequate ECG tracing. A noise-free ECG signal is important to synchronize the ECG signal to the raw image data. Intravenous access through a large intravenous line (e.g., 18 G) is necessary to ensure easy injection of the viscous contrast agent at a flow rate of 5 ml/s.
Stents were evaluated as patent, stenosed (≥50% narrowing of lumen diameter), or occluded, and coronary angiography was performed as it is the gold standard for diagnosing in-stent re-stenosis.
We used two filters, and each one had two kernels – medium and sharp:
- Filtered back projection (FBP): any image has few frequencies, such as high and low frequencies. The low-frequency component is responsible for the blur of the image, and therefore we divided this low component by a certain factor then rebuilding of the image was done. This is FBP and it includes: medium kernel (B26f) and sharp kernel (B46f)
- Iterative reconstruction in image space (IRIS): it overcame many physical factors that could reduce the sensitivity of FBP, such as beam hardening and metal artifacts, and was of two types: medium kernel (I26f) and sharp kernel (I46f).
Regarding medium and sharp kernels, the medium kernel generates images with lower noise but with reduced spatial resolution, whereas the sharp kernel generates images with higher spatial resolution, but increases image noise.
Quantitative image analysis
Quantitative image analysis was carried out with respect to the following three parameters: (a) noise was measured using a circular region of interest placed in the contrast-enhanced lumen of the ascending aorta on FBP and corresponding IRIS images on both medium and sharp kernels, respectively; (b) the signal-to-noise ratio (SNR) was calculated by dividing the mean density derived from the sinus of Valsalva, the interventricular septum, the left anterior descending coronary artery, the circumflex artery, and the right coronary artery by image noise; and (c) the contrast-to-noise ratio (CNR), which was the difference between the mean density of the septum and the coronary vessel (left anterior descending coronary artery) that was divided by image noise.
Qualitative image analysis
Qualitative image analysis was performed by two independent readers. Each reader rated each data set using a five-point Likert scale according to image noise, coronary wall definition, contrast resolution, and general image impression .
Diagnostic value, through accuracy: where all true positive plus true negative cases were divided by all cases, sensitivity: true positive cases divided by all positive cases and specificity true negative cases divided by all negative cases
Data were statistically analyzed using statistical package for the social sciences program version 13 for Windows (SPSS; SPSS Inc., Chicago, Illinois, USA), and a P value less than 0.05 was considered statistically significant.
Data are shown as mean, ranges, or values and their 95% confidence intervals, frequencies, and percentages.
| Results|| |
The present study included 30 patients – 27 males and three females. Twenty-three of them were smokers, 20 patients had hypertension, 19 patients were diabetic, 20 patients had drug-eluting stent, and 10 had bare metal stent [Table 1].
Regarding FBP, the present study showed that noise was statistically lower in B26f than in B46f (23.6 ± 2.7 vs. 36.2 ± 3.2, respectively) (P < 0.001). Regarding iterative reconstruction, I26f had statistically lower noise than I46f (18.0 ± 2.2 vs. 29.8 ± 3.9, respectively) (P < 0.001). Comparing the two filters together, we found that I46f showed statistically lower noise than B46f (29.8 ± 3.9 vs. 36.2 ± 3.2) (P < 0.001) [Table 2].
|Table 2: Objective image quality parameters in each reconstruction protocol|
Click here to view
Regarding the SNR, it statistically improved with iterative reconstruction than with back projection filter for both kernels (22.6 ± 3.5 for I26f vs. 17.2 ± 2.5 for B26f and 14.8 ± 2.1 for I46f vs. 12.0 ± 3.7 for B46f) with P value less than 0.001 for both [Table 2].
CNR statistically improved with iterative filters than with FBP (12.1 ± 1.5 for I26f vs. 9.3 ± 0.9 for B26f with a P < 0.001 and 7.3 ± 0.9 for I46f vs. 5.9 ± 0.5 for B46f with a P < 0.001) [Table 2].
In addition, the study showed that I46f gave the best image quality interpretation by the two readers – that is, moderate agreement (κ=0.40–0.59) [Table 3].
Finally, I46f had the best sensitivity, accuracy, and negative predictive value in detecting in-stent re-stenosis.
| Discussion|| |
Although improvement in diagnostic accuracy by CT technology has significantly improved, the issue was how to get better image quality. This was achieved by high-radiation dose; however, with the introduction of infrared (IR) technique, attempts to treat noise properly at very low signal levels, and consequently reduce the noise and artifacts presents in the resulting reconstructed images and results, were evaluated.
Regarding noise, our study showed that that IRIS could significantly reduce image noise and improve imaging of coronary stents compared with traditional FBP, where there was a statistically significant difference between IRIS and FBP with use of smooth (18.0 ± 2.2 vs. 23.6 ± 2.7) or sharp (29.8 ± 3.9 vs. 36.2 ± 3.2) kernels (P < 0.001). SNR as well as CNR improved significantly with IRIS, and this was in agreement with Park et al. , who reported image noise to be significantly lower with IR algorithm compared with FBP with different kernels. In additionstudies performed by Bittecourt et al. , showed lower image noise in IRIS than in FBP (22.6 ± 4.5 vs. 28.6 ± 5.1). Moreover, Moscariello et al. , showed that half-dose IRIS compared with full-dose FBP significantly lowered image noise. Finally, the studies performed by Fuchs et al.  and von Spiczak et al. , reported image noise to be significantly lower with IR algorithm compared with FBP with different kernels.
On the contrary, Min et al.  showed that no statistically significant difference was observed for image noise between IR and FBP reconstruction techniques using either smooth (18.43 ± 7.64 vs. 21.85 ± 5.9) or sharp (19.43 ± 7.98 vs. 21.98) kernels (P > 0.05).
Regarding SNR, we found that there was a highly statistically significant difference between IRIS and FBP where SNR was higher in IRIS(22.6 ± 3.5 vs. 17.2 ± 2.5) for smooth and (14.8 ± 2.1 vs. 12.0 ± 3.7) for sharp kernels, respectively (P < 0.001), which was in agreement with Ebersberger et al.  and Wuest et al.  whose studies reported that SNR showed a significant difference between IR and FBP (22.1 ± 8.6 vs. 14.3 ± 6.7) (P < 0.05) in a full-dose reconstruction algorithm for smooth and (12.8 ± 3.5 vs. 7.0 ± 1.0) for sharp kernels, respectively (P < 0.05).
In addition, regarding CNR, we noticed a highly statistically significant difference between IRIS and FBP where it was higher in IRIS (12.1 ± 1.5 vs. 9.3 ± 0.9) for smooth and (7.3 ± 0.9 vs. 5.9 ± 0.5) for sharp kernels, respectively (P < 0.001), which was in agreement with Wuest et al.  and Eisentopf et al. , who reported that CNR significantly improved with IR than with FBP for both filters.
Moreover, our study showed better image quality with IRIS than with FBP, which was in agreement with Renker et al.  and Yang et al. , who reported statistically better qualitative image quality with IR in coronary stenting when compared with FBP. On the contrary, studies carried out by Ebersberger et al. , showed that image quality did not show a statistically significant difference between IRIS and FBP (3.5 ± 1.0 vs. 3.7 ± 1.1, P > 0.05).
Diagnostic value: in terms of sensitivity, specificity, and diagnostic accuracy, we found that iterative reconstruction improved the accuracy of stenosis detection compared with FBP (accuracy 90.0 vs. 76.7%, specificity 100.0 vs. 100.0%, sensitivity 87.5 vs. 70.8%, positive predictive value 100.0 vs. 100.0%, and negative predictive value 66.7 vs. 46.2%), which was in agreement with Moscariello et al. , who showedaccuracy of 96.9 versus 93.8%, specificity of 94.6 versus 89.2%, and positive predictive value of 93.3 versus 87.5% for IRIS versus FBP. Andreini et al. , reported statistical significance in the diagnostic accuracy of IR when compared with FBP, 96 versus 91%, and Oda et al. , showed that the mean sensitivity, specificity, and accuracy were 98, 77.4, and 80.1% for coronary CT angiography using the IR technique and were 90.5, 68.2, and 68.6%, respectively, for coronary CT angiography using the FBP technique with no significant difference in the sensitivity, but significant differences in specificity and accuracy were found (P < 0.05).
| Conclusion|| |
We concluded that the sharp kernel for each filter has higher image noise than the smooth kernel, and when comparing both filters together we found that iterative reconstruction I46f has the best image quality.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Serruys PW, de Jaegere P, Kiemeneij F, Macaya C, Rutsch W, Heyndrickx G, et al.
Comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. Benestent Study Group. N
Engl J Med 1994; 331
Fischman DL, Leon MB, Baim DS, Schatz RA, Savage MP, Penn I, et al
. Randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. Stent Restenosis Study Investigators. N
Engl J Med 1994; 331
de Feyter PJ, Kay P, Disco C, Serruys PW. Reference chart derived from post-stent-implantation intravascular ultrasound predictors of 6-month expected restenosis on quantitative coronary angiography. Circulation 1999; 100
Mercado N, Boersma E, Wijns W, Gersh BJ, Morillo CA, de Valk V, et al
. Clinical and quantitative coronary angiographic predictors of coronary restenosis: a comparative analysis from the balloon-to-stent era. J Am Coll Cardiol 2001; 38
Serruys PW, Kay IP, Disco C, Deshpande NV, deFeyter PJ. Periprocedural quantitative coronary angiography after Palmaz-Schatz stent implantation predicts the restenosis rate at 6 months: results of a meta-analysis of the Belgian Netherlands Stent study (BENESTENT) I, BENESTENT II Pilot, BENESTENT II and MUSIC trials. Multicenter ultrasound stent in coronaries. J Am Coll Cardiol 1999; 34
Jeremias A, Kutscher S, Haude M, Heinen D, Holtmann G, Senf W, et al
. Non ischemic chest pain induced by coronary interventions: a prospective study comparing coronary angioplasty and stent implantation. Circulation 1998; 98
Schuijf JD, Bax JJ, Salm LP, Jukema JW, Lanb HJ, van der Wall EE. Feasibility of assessment of coronary stent patency using 16-slice computed tomography. Am J Cardiol 2004; 94
Park EA, Lee W, Kim KW, Kim KG, Thomas A, Chung JW, Park JH. Iterative reconstruction of dual source coronary CT angiography: assessment of image quality and radiation dose. Int J Cardiovasc Imaging 2012; 28
Bittecourt MS, Schmidt B, Seltman M, Muschiol G, Ropers D, Gunther Daniel W, et al
. Iterative reconstruction in image space (IRIS) in cardiac computed tomography: initial experience. Int J Cardiovasc Imaging 2011; 27
Moscariello A, Takx RA, Schoepf UJ, Renker M, Zwermer PL, O'Brien TX, et al
. Coronary CT angiography: image quality, diagnostic accuracy and potential for radiation dose reduction using a novel iterative image reconstruction technique – comparison with traditional filtered back projection. Eur Radiol 2011; 21
Fuchs T, Stehli J, Fiechter M, Dougound S, Sah B, Gebhard C. First in vivo
head-to-head comparison of high-definition stent imaging with 64-slice computed tomography. Int J Cardiovasc Imaging 2013; 29
Von Spiczak J, Morsbach F, Winklhofer S, Frauenfelder T, Leschka S, Flohr T. Coronary artery stent imaging with CT using an integrated electronics detector and iterative reconstructions: first in vitro
experience. J Cardiovasc Comput Tomogr 2013; 7
Min J, Swaminathan R, Vass M, Gallagher S, Weinsaft W. High definition multidetector computed tomography for evaluation of coronary artery stents: comparison to standard-definition 64-detector row computed tomography. J Cardiovasc Comput Tomogr 2009; 3
Ebersberger U, Tricarico F, Schoepf U, Blanke P, Spears J, Rowe G. CT evaluation of coronary artery stents with iterative image reconstruction: improvements in image quality and potential for radiation dose reduction. Eur Radiol 2013; 23
Wuest W, May M, Scharf M, Layritz C, Eisentopf J, Ropers D. Stent evaluation in low-dose coronary CT angiography: effect of different iterative reconstruction settings. J Cardiovasc Comput Tomogr 2013; 7
Eisentopf J, Achenbach S, Ulzheimer S, Layritz C, Wuest W, May M. Low-dose dual-source CT angiography with iterative reconstruction for coronary artery stent evaluation. J Am Coll Cardiol 2013; 6
Renker M, Ramachandra A, Schoepf UJ, Raupach R, Apfaltrer P, Rowe. GW, et al
. Iterative reconstruction techniques: applications for cardiac CT. J Cardiovasc Comput Tomogr. 2011; 5
Yang WJ, Zhang H, Xiao H, Li JY, Liu Y, Pan ZL, Chen KM. High-definition computed tomography for coronary artery stents imaging compared with standard-definition 64-row multidetector computed tomography. J Comput Assist Tomogr 2012; 36
Andreini D, Pontone G, Mushtaq S, Bartorelli A, Bertella E, Trabattoni D, et al
., Coronary in-stent restenosis: assessment with CT coronary angiography. Radiology 2012; 265
Oda S, Utsunomiya D, Funama Y, Takaoka H, Katahira K, Honda K. Improved coronary in-stent visualization using a combined high-resolution kernel and a hybrid iterative reconstruction technique at 256-slice cardiac CT – Pilot study. Eur J Radiol. 2013; 82
[Table 1], [Table 2], [Table 3]