Article Text
Abstract
Introduction Treatment of acute large vessel occlusion (LVO) stroke secondary to intracranial atherosclerotic disease (ICAD) is more nuanced with disparate and infrequently reported outcomes. The deployment of balloon-mounted stents presents an expedient approach with renewed feasibility in the modern era of supple intermediate catheters.
Methods A prospectively maintained endovascular stroke database was searched for patients undergoing intracranial stenting with balloon-mounted stents for acute LVO. Demographic, angiographic, and clinical data were extracted to determine procedural technique and success, measured both angiographically and clinically.
Results Sixty patients underwent stenting for acute LVO secondary to ICAD. Mean presenting National Institutes of Health Stroke Scale (NIHSS) score was 18 and 62% of treated sites were in the posterior circulation. Cases were performed under IV conscious sedation unless the patient was already intubated. Successful recanalization was achieved in 93% of cases (Thrombolysis in Cerebral Infarction (TICI) grade 2b in 48% and TICI grade 3 in 45%). Mean improvement in NIHSS score on post-procedure day 1 was 3.4. Good outcome (modified Rankin Scale score 0–2) at 3 months was 34% and the mortality rate was 34%. The rate of peri-procedural symptomatic hemorrhage was 8% and the rate of acute/subacute stent thrombosis was 7%. In this small cohort, patient age, sex, presenting NIHSS, comorbidities, smoking, tissue plasminogen activator administration, and stent location were not significant predictors of recanalization or good outcome.
Conclusion Treatment of acute LVO stroke with balloon-mounted stents for ICAD has reperfusion rates and clinical outcomes comparable to thrombectomy, with higher rates of hemorrhage and mortality. Because stent placement was performed after attempted thrombectomy, a trial of primary stenting versus thrombectomy should be considered in patients suspected of underlying ICAD.
- atherosclerosis
- balloon
- catheter
- stent
- stroke
Statistics from Altmetric.com
Introduction
While acute large vessel occlusion (LVO) in the USA is most commonly thromboembolic and thus responsive to thrombectomy (stentriever, aspiration or both),1–4 the management of acute LVO secondary to intracranial atherosclerotic disease (ICAD), especially when the lesion is a high-grade stenosis with superimposed thrombus, is technically more challenging and of potentially greater risk. The coronary experience has revealed that a durable effective result often requires deployment of a stent.5–7 While the SAMMPRIS trial failed to show a benefit of stenting with the self-expandable Wingspan stent in the elective setting,8 this technique remains necessary as a reperfusion strategy in the management of an important subset of acute stroke cases. Traditionally, self-expanding stents have been used for this purpose; however, as more centers perform thrombectomy under IV conscious sedation and, given the importance of expedient revascularization, the technique of intracranial angioplasty followed by stenting presents considerable challenges as it is necessary to perform multiple tedious exchanges.
In the era of modern flexible intracranial catheters that can support stiffer balloon-mounted stent constructs, the advantage of a swift single pass with balloon-mounted stents, avoiding a wire exchange, via small delivery catheters (≤0.056 inch) is an important and potentially easier consideration. In this study we review our experience with the use of these stents in acute ischemic stroke in cases of presumed underlying ICAD, highlighting the technique and clinical outcomes.
Methods
We performed a retrospective review of a prospectively maintained IRB-approved endovascular thrombectomy database from January 2004 to June 2018 for patients undergoing thrombectomy and subsequent placement of an intracranial balloon-mounted stent for ICAD, deemed to be the lesion responsible for the acute LVO. Patients treated with dedicated intracranial self-expanding low radial force stents were excluded (n=5), and those treated on a semi-elective basis subacutely for recurrent symptomatic known stenotic lesions in the setting of aggressive medical therapy but stable examinations (‘SAMMPRIS failures’) were also excluded (n=25).
From the database and medical record, patient age, sex, medical comorbidities, occlusion location, and presenting National Institutes of Health Stroke Scale (NIHSS) score were noted. Procedural details including anesthesia (general vs conscious sedation), catheters used, stent used, angiographic and clinical results were obtained along with peri-procedural complications. ICAD was defined radiographically (often with a deployed stentriever demonstrating the lesion), or alternatively inferred in cases of repeated reocclusion/inability to recanalize a vessel without an obvious embolic source or vessel dissection. Symptomatic hemorrhage referred to radiographically demonstrable extravascular intracranial hemorrhage judged to be the etiologic cause of concomitant clinical deterioration. Angiographic and clinical follow-up results were also extracted.
Procedural technique
Over the studied time period the procedural technique evolved considerably. Procedures were performed under IV conscious sedation unless the patient already presented to the angiography suite intubated (n=19). Initial support systems in the early period entailed a 6F Envoy catheter with or without the support of a Shuttle sheath. This evolved to the use of a Shuttle sheath with an 0.057 inch distal access catheter. Since 2013, support systems have typically included the use of a 6F long sheath supporting a 115 cm Navien 0.058 inch or CAT5 intermediate catheter. After initial attempted thrombectomy suggests occlusion secondary to intrinsic disease (often manifesting as repeated reocclusion after recanalization), the intermediate catheter is advanced over a microcatheter/exchange length microwire combination to the stenotic site. The microcatheter is advanced across the stenosis over the microwire and a gentle run is performed via the microcatheter to ensure an intraluminal location. The microcatheter is then exchanged for a balloon-inflated coronary stent, which is deployed (Figure 1). Minivision and Integrity stents were used throughout the treatment period; Driver and Taxus stents were also used from 2004 to 2010. Intraprocedurally, the patient is loaded with eptifibatide. After a post-deployment CT or MRI confirms no hemorrhage, the patient is then loaded with aspirin and clopidogrel. In the event of survivable post-procedural hemorrhage, antiplatelet therapy is held until radiographic and clinical stability is achieved.(figure 1).
Statistical analysis
Statistical analysis was performed using IBM SPSS Statistics 23 (Armonk, New York, USA). Univariate analyses were performed using analysis of variance for continuous variables and Pearson χ2 tests for categorical variables. Multivariate logistic regression analysis was subsequently performed to identify independent predictors of successful recanalization (Thrombolysis in Cerebral Infarction (TICI) 2b or 3) and 90-day modified Rankin Scale (mRS) score 0–2, evaluating patient age, sex, presenting NIHSS score, medical comorbidities (hypertension, hyperlipidemia, coronary artery disease, diabetes, atrial fibrillation), smoking status, tissue plasminogen activator (tPA) administration, and stent location and type for p values ≤0.2 on univariate analysis.
Results
Sixty patients underwent stenting for acute LVO secondary to ICAD (table 1). Mean patient age was 66 (SD 15, range 34–91). Thirty-three patients were male (55%). Medical comorbidities included hypertension in 77%, dyslipidemia in 58%, diabetes in 22%, and known coronary artery disease in 20%. Twenty patients were smokers (33%). Mean presenting NIHSS score was 18 (median 17, SD 7.4, range 5–37). Five patients received IV tPA (8%). Occlusion sites included the basilar artery (n=26, 41%), middle cerebral artery (n=14, 22%), distal vertebral artery (V4, n=13, 21%; occluded, atretic or hypoplastic contralateral V4), supraclinoid internal carotid artery (n=4, 7%), and cavernous internal carotid artery (n=3, 5%).
Successful recanalization was achieved in 93% of cases (TICI 2b in 48%, TICI 3 in 45%). In a multivariate analysis, patient age, sex, presenting NIHSS, medical comorbidities, smoking status, tPA administration, and stent location and type were not significant predictors of recanalization. In seven cases, multiple/telescoping stents were used (12%). There were 11 peri-procedural complications (18%), including symptomatic intracranial hemorrhage (n=5, 8%), peri-procedural compounded ischemic stroke secondary to stent thrombosis within 48 hours (n=4, 7%), retroperitoneal hematoma (n=1, 2%), and carotid-cavernous fistula (n=1, 2%).
After treatment, the mean change in the NIHSS score on post-procedure day 1 was 3.4 (SD 8, median 2). At discharge, nine patients had an mRS score of 0–2 (15%); 16 died during their hospitalization (27%). Carrying forward patients who died prior to discharge, 56 patients had available 3-month clinical follow-up: 19 were mRS 0–2 (34%), 27 were mRS 0–3 (48%), and 19 had died (34%). Nineteen patients with anterior circulation lesions had 3-month clinical follow-up: 7 were mRS 0–2 (37%), 9 were mRS 0–3 (47%), and 5 had died (26%). Thirty-seven patients with posterior circulation lesions had 3-month clinical follow-up: 12 were mRS 0–2 (32%), 18 were mRS 0–3 (49%), and 14 had died (38%). In a multivariate analysis, patient age, sex, presenting NIHSS, medical comorbidities, smoking status, tPA administration, and stent location (LVO site) and type were not significant predictors of a good outcome (mRS 0–2).
Among 42 patients who survived the hospitalization period with available follow-up (mean 2.3 years, median 0.9 years) there were 12 recurrent ischemic events. Seven occurred after the initial hospitalization and five were acute/subacute in-stent thrombotic events during the initial hospitalization incorporated above as peri-procedural complications. Twelve patients underwent delayed (6–12 month post-stent) digital subtraction angiography; two patients had asymptomatic moderate in-stent stenosis.
Discussion
In an era of increasing rates of revascularization for acute LVO, endovascular practitioners will face with greater frequency a proportion of cases with occlusion secondary to ICAD. In light of the results from the SAMMPRIS trial8 that may hamper experience with intracranial stenting, an easier, expedient but safe technique is necessary to treat these lesions in a patient population that is often awake and potentially non-cooperative.
Balloon-mounted stents may be limited in their applicability to intracranial lesions owing to a relatively stiff delivery platform that can be challenging to navigate around the carotid siphon. However, the relevance of this limitation is now waning as a host of modern supple intermediate catheters are available that can be easily navigated across the siphon. For balloon-mounted stents up to 4 mm in diameter, such as Integrity or Vision, a minimal catheter diameter of only 0.056 inches is needed. As such, we typically employ 115 cm 0.058 inch Navien or CAT5 catheters to deliver these stents without difficulty. This series comprises 18 cases prior to our use of the 0.058 inch Navien in 2013 (2 cases of stent implantation per year) compared with 42 cases since then (8 cases of stent implantation annualized), reinforcing the impact of this introduction on procedural feasibility. We did not, however, retrospectively note a difference in revascularization or clinical outcomes between these two periods, potentially in light of the small numbers in the early cohort (table 1).
Another potentially perceived disadvantage of these stents is the shorter delivery platform which may not ‘reach’ the intended target through a 115 cm intermediate catheter. An important approach to mitigate this limitation entails advancing a Stiff Amplatz wire through the long 6F sheath and intermediate catheter to straighten and thus shorten more proximal tortuosity. With the wire in place through these catheters, the long 6F sheath (Infinity or Neuron Max) can then be advanced as distal as feasible (high cervical or petrous internal carotid artery).
Balloon-mounted stents are cost-efficient alternatives to dedicated intracranial stents which have significant advantages, particularly in the context of procedures performed in awake patients—namely, a rapid exchange delivery platform and ‘single-pass’ deployment without the need for lengthy exchanges for angioplasty and then stenting. The use of bare metal compared with drug-eluting stents in the intracranial circulation will ultimately require greater comparative scrutiny as there are few data regarding their relative efficacy in the intracranial circulation.9
Prior to the introduction of stentrievers, primary intracranial stenting with the Wingspan system for acute LVO was performed in a Food and Drug Administration-approved study of 20 patients.10 It is interesting to note from this 2009 study that there were two cases (10%) where the Wingspan system could not be delivered to the lesion, requiring bailout using an Enterprise stent. The authors dedicated two additional papers to reinforcing the importance of deliverability in this era preceding the introduction of more supple intermediate catheters, advocating the deployment of the Enterprise stent for acute LVO.11 12 While subsequently demonstrated as feasible in presumably thromboembolic occlusions,13 for patients with ICAD, lower radial force stents designed for stent-assisted coiling may need to be crossed again after deployment and treated with angioplasty with a non-compliant balloon. This contrasts with a swift single pass with more cost-efficient balloon-mounted coronary stents.
Alternative approaches to acute intracranial stenting include the administration of glycoprotein IIb/IIIa inhibitors14 after thrombus removal without treatment of the underlying plaque or intracranial angioplasty without stenting.15 In our experience with the former approach, freshly recanalized intrinsic disease carries a prohibitively high risk of acute reocclusion; frequently, in preparation for acute intracranial stenting, we have seen recanalized severe intrinsic disease progress to interval reocclusion on interval runs. Should the vessel remain patent, one can extract from SAMMPRIS8 a recurrent risk of transient ischemic attack or stroke of at least 6% at 30 days. Intracranial angioplasty carries a risk of intracranial dissection and generally necessitates the use of dual antiplatelet therapy after treatment anyhow to facilitate a durable result.
Our study underscores a generally more guarded prognosis in patients with acute LVO secondary to ICAD: the rate of good outcome (mRS 0–2) was 34% at 3-month follow-up despite successful recanalization achieved in 93% of cases. However, most cases were performed for posterior circulation lesions where outcomes are known to be worse than the anterior circulation.12 Of potential meaningful comparison to the BEST trial,16 it is noteworthy that our rate of mRS 0–3 outcome for posterior circulation lesions was 49%, potentially a more comparable result for this location.
Lesion location did not significantly impact the angiographic or clinical results in our study; the rate of good outcome for anterior circulation lesions was 37%. This contrasts with the HERMES Collaboration which reported an overall 46% rate of mRS 0–2 outcome at 3 months in pooled data from five randomized trials for thrombectomy.17 Similarly, our rate of symptomatic hemorrhage (8%) is greater than that reported in HERMES (4%). These differences may be explained by a more arduous procedure that typically entails both thrombectomy and stent placement as well as requiring administration of a glycoprotein IIb/IIIa inhibitor intraprocedurally followed by loading with aspirin and clopidogrel postprocedurally. Importantly, the first-pass effect is in fact lost in our study as stenting was not performed on the first pass in any case.
Outcomes and symptomatic hemorrhage rates in this study are similar to those reported by Al Kasab et al for another Western population treated for acute LVO secondary to ICAD.18 In contrast, a study in a Korean population with ICAD by Kang et al reported a lower rate of symptomatic hemorrhage post-procedure (1%) and a 57% good outcome rate (mRS 0–2), although it is unclear whether these results would have good external validity to Western centers.14 This study included patients with a lower presenting NIHSS score (median 11 vs 17 in our study) and had a far greater proportion of anterior circulation occlusions (85% vs 38% in our study). Another report of 56 Korean patients with acute LVO secondary to ICAD reported successful recanalization in 80% of cases (lower than this study) but comparable favorable outcomes to patients with LVO without ICAD (46.4% vs 46.9%).19 In a study of 140 Chinese patients with acute large artery occlusion, Jia et al found a prevalence of ICAD of 34% and reported comparable radiographic and clinical outcomes between patients with and without ICAD.20 Along with the report by Kang et al, comparing these studies with our own and that of Al Kasab et al underscores a potential discrepancy in outcomes between Western and Eastern populations with acute LVO secondary to ICAD.
Conclusion
With modern supple intermediate catheters, successful recanalization of acute LVO secondary to ICAD can often be achieved with traditional balloon-mounted stents (93% of cases in this series). Nevertheless, follow-up rates of clinical good outcome in our Western population remain lower than those seen for patients undergoing thrombectomy for LVO secondary to embolic disease. This is at least in part due to a lost ‘first-pass effect’ in this study population, which suggests that, to accumulate further evidence, a clinical trial comparing primary first-pass stenting with thrombectomy and subsequent stenting in patients with acute LVO secondary to presumed ICAD should be considered.
References
Footnotes
Contributors Conception and design: BAG, AJ, TGJ. Drafting the article: BAG, AJ, TGJ. Acquisition of data/data analysis: BAG, SMD, GW. Reviewed and revised article prior to submission: all authors. Study supervision: TGJ.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests BTJ: Consultant, Medtronic. BAG: Consultant, Microvention. TGJ: Consultant, Stryker Neurovascular (PI DAWN-unpaid); Ownership interest: Anaconda; Advisory Board/Investor: FreeOx Biotech, Advisory Board/Investor; Route92, Advisory Board/Investor; Blockade Medical, Consultant; Honoraria: Cerenovus.
Provenance and peer review Not commissioned; externally peer reviewed.
Patient consent for publication Not required.