Article Text
PDF
Abstract
Objective and importance To describe the use of DynaCT angiographic imaging for the evaluation of Wingspan in-stent restenosis (ISR).
Methods Two patients were treated with Wingspan stenting and percutaneous transluminal angioplasty (Patient 1 had treatment of a severe stenosis of the right middle cerebral artery and patient 2 had severe stenosis of the left intracranial internal carotid artery. Both patients developed ISR and were evaluated with high resolution DynaCT angiographic imaging.
Results DynaCT demonstrated circumferential soft tissue density material distributed within the stent as the cause of the stenosis visualized with conventional angiography.
Conclusions These findings support the hypothesis that ISR is caused by neointimal proliferation, rather than vascular re-coil with stent collapse.
Statistics from Altmetric.com
In-stent re-stenosis (ISR) occurs in approximately 30% of patients undergoing intracranial percutaneous transluminal angioplasty and stenting (PTAS) with the Wingspan (Boston Scientific Corp, Fremont, California, USA) system.1 2 Conventional angiography is the gold standard for the diagnosis of ISR. However, while technologies such as intravascular ultrasound are readily available to directly interrogate the tissue responsible for ISR within the coronary vasculature, direct imaging of the tissue in-growth (or other etiology) responsible for intracranial ISR has been limited.
Recent advances in flat panel detector imaging have led to the ability to perform high resolution computerized tomographic angiography after the intra-arterial injection of dilute contrast (IA-CTA). This technique yields very high contrast and high resolution imaging of the intracranial vasculature and implants.3 4 We report the use of DynaCT (Siemens Medical Solutions, Erlangen, Germany) to study the appearance of Wingspan ISR before and after angioplasty (PTA) in two patients.
Methods
Technique for DynaCT angiography
A biplane flat panel detector angiography suite (AXIOM Artis FD Biplane Angiosuite with DynaCT) was used to acquire the images. The technique previously described by Benndorf et al5 for the DynaCT was modified to include the injection of 44 ml of Ultravist 240 (Schering, Berlin, Germany) diluted to 20% strength. A continuous contrast injection is performed during the entire 20-s image acquisition and a 2-s x-ray delay at a rate of 2 cc/s. The DynaCT was performed with 0.4° increments, 512 matrix in projections, 220° total angle, 20°/s, and approximately 15–30 frames/s, for a total of 538 projections. Image post-processing was then performed to correct for scatter, beam hardening, and ring artifacts using a commercially available workstation (Leonardo; Siemens Medical Solutions).
Case reports
Case 1
A 39-year-old man presented with a right M1 stenosis symptomatic with stroke (figure 1A). He was successfully treated with Wingspan (figure 1B).
A 39-year-old man was admitted with complaints of episodic left hemiparesis. MR showed foci of restricted diffusion within the right hemisphere. These symptoms continued to fluctuate while on intravenous heparin therapy. Catheter angiography the following day confirmed a 70%, 6 mm long stenosis of the M1 middle cerebral artery segment (A). The patient was loaded with acetylsalicylic acid (650 mg) and clopidogrel (600 mg) (Plavix; Bristol-Myers Squibb/Sanofi Pharmaceuticals, New York, New York, USA) and subsequently underwent angioplasty and stenting with the Wingspan system (3.0×9 mm Wingspan stent). At the completion of the procedure, the stenosis measured 12% (B). His neurological symptoms resolved over several days, and he was discharged from the hospital on a daily regimen of ASA 325 mg and clopidogrel 75 mg. He experienced symptomatic in stent re-stenosis (ISR) 3.5 months after the original procedure and he was re-treated with angioplasty alone. His symptoms resolved, but recurred 2 months later. Follow-up angiography at this time demonstrated recurrent severe ISR (C). This was successfully treated with percutaneous transluminal angioplasty (PTA) (D) with a marked improvement in the luminal diameter. Curved reformations (E) derived from DynaCT source data following PTA demonstrated no evidence of dissection or an intimal flap within the stented segment. Within the lumen of the stent, soft tissue density material, presumably hyperplastic intimal tissue, was circumferentially distributed about the augmented central channel of contrast defining the lumen of the vessel — correlating well with the conventional angiogram.
Four months after the initial procedure, the patient experienced recurrent right hemispheric transient ischemic attacks (TIAs) while on aspirin and clopidogrel. Follow-up angiography demonstrated a 7-mm long segment of 90% ISR which was uneventfully treated with repeat balloon angioplasty. Re-treatment resulted in the resolution of his neurological symptoms.
At 6 months, the patient again developed right hemispheric TIAs. Cerebral angiography demonstrated long segment (11.5 mm), pre-occlusive (>90%) ISR (figure 1C) which was treated with repeat angioplasty (figure 1D). A DynaCT was performed to assess stent apposition and exclude an in-stent dissection after PTA. The DynaCT demonstrated a circumferential distribution of soft tissue density material within the Wingspan stent surrounding a central channel of intraluminal contrast (figure 1E).
His neurological symptoms again resolved following the angioplasty procedure.
Case 2
A 69-year-old woman on aspirin and clopidogrel presented with a symptomatic left supraclinoid ICA stenosis (figure 2A). She underwent uneventful PTAS with the Wingspan system (figure 2B). At 4 months, she underwent a follow-up angiogram for recurrent TIA that demonstrated a pre-occlusive (>90%) ISR of the supraclinoid ICA. She subsequently underwent uneventful repeat PTA of the supraclinoid ICA with resolution of her symptoms. The patient again developed episodic aphasia at 7 months while still on acetylsalicylic acid (ASA) and clopidogrel. Repeat angiography demonstrated a long segment (17.8 mm), flow-limiting (95%) ISR of the supraclinoid ICA (figure 2C). For this, she underwent another re-treatment with PTA of the supraclinoid ICA (figure 2D). A DynaCT was performed before angioplasty (figure 2E) demonstrating a thick layer of soft tissue density material within the stent marginating a narrow, serpiginous contrast-containing central lumen. Following the angioplasty, DynaCT demonstrated no evidence of in-stent dissection and showed the in-stent tissue to be circumferentially displaced or compressed about an augmented central lumen (figure 2F).
A 69-year-old woman presented with a 15-min episode of aphasia while on acetylsalicylic acid and clopidogrel. Cerebral angiography showed a high grade long segment stenosis of the supraclinoid segment of the carotid artery extending into the proximal M1 segment of the middle cerebral artery (A). She underwent successful, uneventful balloon angioplasty and stenting with the Gateway–Wingspan system (4.0×15 mm stent (B). She was discharged on aspirin and clopidogrel. She returned at 4 and 7 months after the procedure, both times with symptomatic high grade in stent restenosis (ISR) requiring re-treatment with percutaneous transluminal angioplasty (PTA). At 7 months pretreatment angiography (C) demonstrated long segment, pre-occlusive ISR. Following re-PTA, the luminal diameter was markedly improved (D). Curved reformations derived from the pretreatment DynaCT source images (E) demonstrate the stent to be fully and uniformly expanded. Within the stent, there was a thick layer of soft tissue density material, presumably representing hyperplastic intimal tissue marginating a highly stenotic (but still patent) contrast-filled lumen — correlating well with the pretreatment angiogram (C). Curved reformats after PTA (F) demonstrate no change in the position or diameter of the stent; however, the soft tissue density material has been peripherally compressed or displaced with concomitant augmentation of the contrast-containing lumen — again correlating well with the post-PTA angiogram (D).
Discussion
ISR occurs in approximately 30% of patients undergoing intracranial PTAS with the Wingspan system.1 Some patient subgroups have been determined to be particularly susceptible to ISR after Wingspan.2 The etiology, natural history, appropriate imaging surveillance and optimal management of intracranial ISR remain largely unknown. The success of intracranial stenting depends, in part, upon our ability to understand and manage intracranial ISR.
The etiology of Wingspan ISR remains in question. While balloon expandable coronary stents typically become re-stenotic as a result of the in-growth of hyperplastic intimal tissue, some have hypothesized that stenosis within the self-expanding Wingspan stents may be the result of vascular re-coil and stent collapse due to the inadequate radial resistive force of the stent.
The diagnosis of Wingpsan ISR is accurately made by conventional angiography, which demonstrates narrowing of the contrast column through the stented segment. Direct imaging of the tissue in-growth (or other etiology) creating intracranial ISR is limited. CTA can sometimes show soft tissue density material within the confines of the stent; however, the resolution is limited and the appearance has been difficult to differentiate from focal thrombus formation.2 Magnetic resonance angiography (MRA) is limited by non-laminar flow within stenotic or tortuous vascular segments as well as by artifacts from the stent itself and its end-markers. Intraluminal imaging modalities such as intravascular ultrasound and optical coherence tomography have not yet been applied for the clinical imaging of intracranial blood vessels in humans.
DynaCT uses flat-panel detector technology to generate high resolution cone beam CTA images. The level of resolution achieved with DynaCTA provides clear imaging of the commonly used self-expanding intracranial nitinol stents as well as the endoluminal contrast column.5 DynaCT performed prior to angioplasty demonstrated the Wingspan stent to be fully and uniformly expanded within the parent vessel. This appearance of the stent precludes vascular re-coil and stent collapse as a mechanism for the restenotic appearance depicted on angiography.
Prior to treatment, DynaCT showed a thick layer of soft tissue density material distributed within the stent circumferentially around a central serpiginous channel of contrast material defining the remaining, stenotic lumen. This peripheral in-stent material likely represents hyperplastic neointimal tissue—the causative factor for the narrowing observed on angiography. Following the angioplasty, the stent itself did not appear to have been displaced or further expanded. The tissue within the stent demonstrated relatively uniform compaction about an augmented central contrast channel after the angioplasty (correlating with the improved luminal diameter on conventional angiography (figures 1E and 2F). These findings would indicate that the in-stent material has a relatively loose supra-structure that may be peripherally compressed by the angioplasty balloon without any lasting distortion of the actual stent structure.
Conclusions
DynaCT is useful in the evaluation of ISR following Wingspan stenting. In the two patients imaged in this study, a thick layer of soft tissue density, compressible material was found within the stent causing the stenosis. There was no noted collapse of the stent observed.
Footnotes
Funding DF has NIH funding for the SAMMPRIS trial. Other funders: NIH.
Competing interests DF (Boston Scientific, Research Support for US Wingspan Registry); the other author have none.
Ethics approval The images were obtained within the setting of routine patient care. No patient identifiers were included. While the available images of Wingpsan ISR are novel/unique, the disease process is not.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.