CCTA can assess implanted coronary stents for in-stent restenosis (ISR), patency, and occlusion. Performance is acceptable for rule-out in favorable anatomy but is strongly influenced by stent diameter, strut thickness, heart rate, and location. Approximately 8–12% of stents cannot be imaged with diagnostic quality, almost always because of metal-related artifact.
1 Appropriate use & clinical role
Historically, Appropriate Use Criteria considered CCTA generally inappropriate for evaluating ISR regardless of symptoms, because metallic struts degrade lumen visualization. More recent evidence has softened this position:
- The SCCT 2021 Expert Consensus considers CCTA appropriate for symptomatic patients with stent diameter ≥ 3.0 mm, and potentially appropriate for smaller stents with thin struts (< 100 μm) in proximal, non-bifurcation locations.
- The most widely accepted scenario remains the left main (LM) stent ≥ 3 mm — the LM is large-caliber and relatively motion-resistant, so CCTA reliably excludes ISR here.
For small (< 3 mm), bifurcation, or otherwise unfavorable stents, invasive coronary angiography (ICA) remains the reference standard. Combined CCTA + CT perfusion (CTP) is a useful alternative when anatomic assessment alone is insufficient.
2 In-stent restenosis (ISR)
ISR is a recurrent narrowing of > 50% of the lumen diameter within the stent or at its immediate edges (the adjacent 5 mm). It results from neointimal hyperplasia and macrophage accumulation following vascular injury at implantation.
Risk factors: small-diameter stents, long lesions, ostial lesions, bifurcation lesions, and diabetes mellitus.
Four morphologic patterns are described and can be approximated on CCTA:
| Pattern | Description |
|---|---|
| I | Focal lesion (< 10 mm) within the stent |
| II | Diffuse lesion (> 10 mm) within the stent |
| III | Stenosis (> 10 mm) extending outside the stent |
| IV | Totally occluded stent |
A simplified focal-versus-diffuse scheme is often sufficient for reporting.
3 Diagnostic accuracy
Meta-analytic data across 35 studies (4,131 stents) using ≥ 64-slice CT demonstrate:
- Per-stent sensitivity 90%, specificity 89–94%
- Negative likelihood ratio 0.10 — strong rule-out capability for ≥ 50% ISR
- ~92% of stents fully interpretable on CCTA
Adding CT perfusion (CTP) to CCTA improves per-patient diagnostic accuracy from 71% to 87% and eliminates non-diagnostic examinations — making the combined anatomic–functional study valuable when CCTA alone is degraded by a small-diameter stent.
4 Factors that reduce accuracy
Four factors significantly impair stent evaluation:
- Stent diameter < 3.0 mm — blooming artifact can obscure up to 55% of the in-stent lumen.
- Strut thickness ≥ 100 μm — greater metallic artifact and partial-volume averaging.
- Heart rate ≥ 65 bpm — motion artifact compounds other artifact types.
- Bifurcation stents — complex geometry limits lumen visualization.
5 Imaging challenges & artifacts
CCTA is highly prone to false-positive interpretation of stents. Three artifact types dominate:
The dense metallic struts appear artificially thick, visually shrinking the inner lumen and overestimating stenosis. This is the principal source of false positives.
The high density of the metal preferentially absorbs lower-energy photons, producing dark, hypoattenuated regions inside the lumen that mimic true stenosis.
Even slight cardiac motion blurs the struts into the lumen, falsely suggesting ISR. This is the most common cause of non-assessable segments.
6 Techniques for image optimization
The SCCT recommends the following to improve stent imaging:
- Heart rate control — goal < 60 bpm.
- Tube potential ≥ 100 kVp (140 kVp further reduces beam hardening).
- Sharp (high-resolution) reconstruction kernel.
- Model-based / iterative reconstruction.
- Mono-energetic reconstructions when spectral or dual-energy CT is available — high-energy virtual monoenergetic images reduce beam hardening.
- Thin slice reconstructions.
At the workstation, increasing window level and widening window width reduces the apparent brightness of the metal and improves lumen assessment.
Emerging bioresorbable scaffolds are expected to resolve many of these limitations — the struts dissolve over 2–3 years and do not generate metallic artifact.
7 Transluminal Attenuation Gradient (TAG)
Because visual assessment of the in-stent lumen is difficult, the TAG compares contrast attenuation (HU) inside the stent with the vessel segment just proximal to it.
| Ratio (in-stent HU / proximal HU) | Interpretation |
|---|---|
| ≥ 0.8 | Normal — no flow limitation |
| < 0.8 | Loss of contrast opacification — suggests hemodynamically significant ISR |
Caveat: if the stent shows no contrast at all (total occlusion), CCTA cannot reliably distinguish severe ISR from acute stent thrombosis — invasive evaluation is required.
8 Emerging technology
Photon-counting detector CT (PCD-CT). Ultra-high-resolution PCD-CT is a substantial advance for stent imaging. In a prospective comparison with invasive coronary angiography, it demonstrated 100% sensitivity and 100% negative predictive value for ISR, with markedly improved in-stent lumen visualization.
Subtraction CCTA. Subtraction techniques have improved diagnosable rates to 92% versus 51% for conventional CCTA in 3.0–3.5 mm cobalt-chromium stents.
9 Comparison with other modalities
| Modality | Role in stent evaluation |
|---|---|
| Invasive coronary angiography (ICA) | Clinical gold standard for detecting ISR and grading stenosis severity |
| IVUS | Useful invasive adjunct for lumen and stent assessment |
| OCT | Optimal for stent expansion, malapposition, and strut fracture assessment |
| SPECT / PET stress perfusion | Detects mismatched perfusion in hemodynamically significant ISR |
| MRA | Cannot directly visualize in-stent lumen — metallic stents cause susceptibility artifact and signal void; stress MR perfusion can detect hemodynamically relevant ISR functionally |
10 Standardized reporting
- CAD-RADS modifiers. The presence of a stent anywhere in the coronary tree requires the
Smodifier (e.g.,CAD-RADS 3/S). If a stent is entirely non-evaluable due to blooming or motion, use theNmodifier (e.g.,CAD-RADS N/S). - CT-FFR limitation. CT-FFR is generally contraindicated in patients with existing stents — the computational fluid dynamics model does not handle stent geometry reliably.
References
- Narula J, Chandrashekhar Y, Ahmadi A, et al. SCCT 2021 Expert Consensus Document on Coronary Computed Tomographic Angiography. J Cardiovasc Comput Tomogr. 2021;15(3):192–217. PubMed
- Dahdal J, Jukema RA, Remmelzwaal S, et al. Diagnostic Performance of CCTA and CTP Imaging for Clinically Suspected In-Stent Restenosis: A Meta-Analysis. J Cardiovasc Comput Tomogr. 2025;19(2):183–190. PubMed
- Dai T, Wang JR, Hu PF. Diagnostic Performance of CT Angiography in the Detection of Coronary Artery In-Stent Restenosis: Evidence From an Updated Meta-Analysis. Eur Radiol. 2018;28(4):1373–1382. PubMed
- Rief M, Zimmermann E, Stenzel F, et al. CT Angiography and Myocardial CT Perfusion in Patients With Coronary Stents. J Am Coll Cardiol. 2013;62(16):1476–85. PubMed
- American College of Cardiology Foundation Task Force. ACCF/ACR/AHA/NASCI/SAIP/SCAI/SCCT 2010 Expert Consensus Document on Coronary CT Angiography. J Am Coll Cardiol. 2010;55(23):2663–99. PubMed
- Hagar MT, Soschynski M, Saffar R, et al. Ultra-High-Resolution Photon-Counting Detector CT in Evaluating Coronary Stent Patency. Eur Radiol. 2024;34(7):4273–4283. PubMed
- Li J, Guo MT, Yang X, et al. Usefulness of Subtraction CCTA for In-Stent Restenosis Assessment Using 320-Row Area Detector CT. Medicine. 2021;100(51):e28345. PubMed