Objective To report our experience with stent supported intracranial recanalization for acute ischemic stroke beyond 8 h of symptoms onset.
Background Acute ischemic stroke (AIS) therapy is often limited to an 8 h window using mechanical means. However, recent reports have shown delayed recanalization beyond 8 h might be a viable option in a subset of patients.
Methods A retrospective review was performed of our AIS database for patients who underwent stent supported intracranial recanalization beyond 8 h of symptom onset. Clinical and angiographic data were reviewed. Outcome was measured using modified Rankin Scale (mRS) scores at 30 and 90 days.
Results 12 patients (11 men and one woman) underwent delayed stenting for AIS. Mean age was 49 years (range 37–73) and mean National Institutes of Health Stroke Scale was 17 (range 8–29, median 15). Mean time from stroke onset to intervention was 66.1 h (range 10–168 h, median 46 h). 10 patients presented with a Thrombolysis in Myocardial Infarction (TIMI) score of 0 and the remaining two had a TIMI of 1. Recanalized vessels included: left middle cerebral artery (n=6), basilar trunk (n=2), vertebrobasilar junction (n=3) and internal carotid artery (ICA)-T (n=1). Four patients had prior attempts of embolectomy/thrombolysis using mechanical and chemical means. Stents used included: six balloon mounted stents, five Wingspan and one Enterprise self-expanding intracranial stent. Recanalization, defined as a TIMI score of 2 or more, was achieved in 11 patients. Two patients (17%) had intracranial hemorrhage. Thirty day mRS of ≤3 was achieved in six patients (50%). Seven patients (58%) had a 90 day mRS of ≤2.
Conclusion Stent supported intracranial recanalization is a safe and feasible approach in a selective group of patients presenting with acute ischemic stroke beyond 8 h of symptom onset.
Statistics from Altmetric.com
Acute ischemic strokes (AIS) compromise up to 88% of strokes, and timely recanalization of these occlusions often leads to improved neurological outcome.1–3 Current recommendations for thrombolytic therapy for stroke are limited to 4.5 h for intravenous therapy4 5 or 8 h for intra-arterial treatment.6–10 Among the various approaches used to achieve intracranial recanalization, stent supported angioplasty has been retrospectively studied by various investigators as an alternative approach to the more established clot retrieval and thrombolytic strategies.11–15 Initial data from these studies show that stenting for AIS can be done with relative safety and efficacy. Further, all studies have been limited to a time window 8 h from stroke onset. However, it is reasonable in selected subgroups of patients to expect benefit from stent supported revascularization beyond the 8 h window. We here present our center experience in delayed recanalization beyond the 8 h window.
We performed a retrospective review of our stroke intervention database (2006–2010) of patients presenting with AIS who underwent endovascular therapy after obtaining the institutional review board approval.
The following variables were reviewed: clinical and imaging details, including demographic details, clinical presentation, National Institutes of Health Stroke Scale (NIHSS) at presentation and imaging findings. A stroke neurologist carefully and independently recorded the time from symptom onset to treatment. Various treatments used for recanalization, including use of intra-arterial tissue plasminogen activator (tPA) or GPIIb/IIIa inhibitors, mechanical devices (Mechanical Embolus Removal in Cerebral Ischemia (MERCI) or Penumbra device) and stents were included. The pre- and postprocedure Thrombolysis in Myocardial Infarction (TIMI) scores were recorded. Primary outcome measures were safety and angiographic evidence of recanalization. Safety was defined as periprocedural complications (stroke, subarachnoid hemorrhage, surgery site bleeding and infections) and/or reperfusion complications (reperfusion injury, hemorrhagic transformation). Secondary outcome measures included clinical outcome measured by the modified Rankin Scale (mRS) score at 30 days and 90 days.
Selection and inclusion criteria
Clinical diffusion mismatch or CT clinical mismatch was used to select patients with acute stroke to determine the presence of a smaller infarct core than expected based on the clinical examination/NIHSS. Patients who had an NIHSS score of >8 at presentation without evidence of large territory infarct on CT scan even after 8 h of symptoms onset were deemed to be good candidates for delayed revascularization. In some cases, perfusion or non-invasive angiographic imaging was used as a selection criterion.
Patients were excluded if any of the following was present: CT head showing completed infarct, NIHSS <8, absence of diffusion–perfusion mismatch or those with greater than two-thirds middle cerebral artery (MCA) territory infarct.
Patients were given a loading dose of 325 mg of aspirin and 450–600 mg of clopidogrel prior to the procedure. Under general anesthesia, groin access was obtained with a 6 French sheath, and a 6 French guiding catheter was placed in the cervical vessel. Activated clotting time was maintained over 250 s after it was decided to deploy a stent. Intracranial arteries were selectively catheterized and angiographic images taken to identify the site of occlusion. A microcatheter was then navigated in the occluded vessel with the support of a 0.014 inch microwire, and was placed distal to the occlusion. Based on the site and type of the thrombus as well as the operator's discretion, chemical (tPA, Abciximab (Reopro; Eli Lily and Company, Indianapolis, Indiana, USA)) or mechanical embolectomy was attempted. Devices used for mechanical angioplasty included: MERCI Retrieval System (Concentric Medical Inc, Mountain View, California, USA), Balloons viz Gateway PTA Balloon Catheter (Boston Scientific Corporation, Natick, Massachusetts, USA) and Maverick PTCA Balloon Catheters (Boston Scientific Corporation). The stent system was prepped and exchanged over the microwire to place in the diseased segment. In the initial part of the study, balloon expandable stents, including Minivision (Abbott Vascular, Abbott Park, Illinois, USA) and Multilink Guidant Vision Stent (Abbott Vascular) were used. Later, self-expanding intracranial stents, including Wingspan (Boston Scientific Corporation) and Enterprise (Cordis Neurovascular Inc, Miami Lakes, Florida, USA) were used. All coronary stent placement and balloon angioplasty were done using the technique of balloon under dilatation (80% of the recommended values for the coronary circulation). Post-stent deployment digital subtraction images were then taken. A postprocedural CT head scan was done to identify any hemorrhagic complication from the procedure. The patients were then transferred to our dedicated neurocritical care unit for further observation and monitoring.
Twelve patients (11 men and one woman) were included in this study. There were seven African–Americans (58.33%) and five Caucasians (41.67%). Mean age was 49 years (37–73). Mean NIHSS was 17 (range 8–29). Mean time from stroke onset to puncture was 66.1 h (range 10–168). Preprocedure CT head imaging showed early evidence of infarction in five patients but all less than one-third of the vessel territory. Preprocedural MRI of the brain was available for seven patients and MR brain perfusion imaging was available for two of them (table 1).
All patients were found to have acute occlusion. Ten patients (83.3%) had a TIMI score of 0 and two (16.7%) had a TIMI of 1. The level of occlusion was left MCA M1 segment (n=6), basilar trunk (n=2), dominant vertebrobasilar junction (n=3) and Carotid T (n=1). Thrombolysis/clot retrieval using mechanical and chemical means was attempted in four patients (three intra-arterial tPA, one intra-arterial/intravenous Abciximab and one MERCI retriever). Intra-arterial tPA was given in extremely small doses (2–4 mg) to make the clot softer. Stenting was done in all patients. Types of stents placed were: six balloon mounted stents (five Minivision and one Vision), five Wingspan stents and one Enterprise self-expanding intracranial stent (table 2). Additional post-stent deployment balloon angioplasty was done in two patients for residual stenosis, while another patient received intra-arterial Abciximab (5 mg).
Recanalization, defined as a TIMI score of 2 or more, was achieved in 11 patients (91.7%). Post- stent deployment, six patients (50%) had a TIMI of 3, five patients (41.67%) TIMI 2 and one patient (8.33%) TIMI 1 (table 2). Recanalization was not achieved in one patient with long segment ICA-T/MCA occlusion despite placement of two overlapping long Enterprise stents. This patient had progression of stroke in the postoperative period and poor neurological outcome (mRS 3) (table 3).
Postoperative outcome and 30 day mRS
Complications included two (16.33%) intracranial hemorrhages, both in the ganglionic region. None of these hemorrhages was associated with worsening of NIHSS. Further, a third patient (8.33%) had procedure related small interpeduncular subarachnoid hemorrhage, probably secondary to small vessel perforation with wire. However, there was no change in his neurological examination. Two patients had progression of their stroke and required intubation, prolonging their ICU and hospital stay. One of these two patients further developed hospital acquired pneumonia. Another case was complicated by hyperperfusion syndrome resulting in bilateral thalamic bleeding with intraventricular extension in the vascular territory of the recanalized basilar artery.
An independent stroke neurologist evaluated 30 day mRS. Six patients (50%) had 30 day mRS scores of ≤3, out of which two (16.67%) had an mRS score of 2. One patient (8.33%) died following progression of stroke and withdrawal of care (table 3).
Follow-up outcome at 90 days
At 90 days, mean follow-up mRS was 2.75. Seven patients (58%) had a 90 day mRS score of ≤2 and were independent for all of their daily activities. mRS scores of the remaining patients were: three had a score of 4 (25%), one a score of 5 (8.33%) and one a score of 6 (8.33%) (table 3).
Follow-up angiogram was available in seven patients. Five did not have any evidence of in-stent stenosis or thrombosis. Two patients had severe, flow limiting (90%) in-stent stenosis, with leptomeningeal collaterals filling the vascular territory. Interestingly, both of these patients had M1 occlusions, and both were doing well clinically (mRS 2) at the 90 day follow-up (figures 1 and 2).
The current pathophysiological model of ischemic stroke postulates the existence of an infracted core which is surrounded by a layer of penumbra further surrounded by a zone of oligemia.16 Positron emission tomography studies have shown that there is potentially salvageable penumbral tissue for 48 h or longer after stroke onset.17 Thus in a selective subset of patients, the penumbral tissue is potentially salvageable even after a long time. Multiple studies have pointed to a relationship between extensive collateralization and early flow augmentation through the recruited collaterals as mechanisms responsible for the sustenance of penumbra.18 19 However, the viability of these collaterals is not known, and there have been reports in the literature about collateral failure and secondary neurological worsening.19 In the face of lack of recanalization, leptomeningeal collateral support may be insufficient to salvage the penumbra.19 20 Thus patients with penumbral tissue sustained on collateral support may be amenable to delayed recanalization.
Various imaging studies have been developed which can identify this ischemic penumbra and subsequently patients amenable to delayed revascularization. Diffusion weighted imaging–perfusion weighted imaging (DWI–PWI) mismatch and DWI reversibility are two MRI correlates that reflect the existence of the penumbra and the potentiality to salvage it through reperfusion, respectively.16 21 CT based protocols have also been described, and along with MR perfusion imaging, have been employed to extend the window of intravenous and intra-arterial therapies for up to 9 h after the ictus.5 22–24 These techniques have also been used for delayed revascularization using endovascular techniques and intra-arterial therapies.19 25 26 It has also been shown that DWI reversibility in these scenarios correlate well with good clinical results.16 21 27 In this study, however, we relied on the clinical diffusion mismatch or CT clinical mismatch as a criterion to select patients for endovascular intervention. A few patients, in whom intervention was done after 24 h of onset, were also treated based on the absence of large infarct (more than one-third of MCA territory) on plain CT scan. More recently, we have been using both MR and CT perfusion to delineate patients who are amenable to delayed revascularization.
Until recently there was only anecdotal evidence of delayed endovascular therapy after the 8 h window.26 Natarajan et al published a series of 30 patients where recanalization was attempted beyond 8 h.28 They used multimodal therapy, including chemical and mechanical, to achieve recanalization in 66.7% of their patients. Further, 30%, or nine out of 30 patients, were treated with intracranial stenting. Symptomatic intracranial hemorrhage was seen in 10% while the mortality rate was 23.3%. Thirty three per cent of patients had a clinical outcome of mRS ≤3. Janjua et al have shown that clinical diffusion–perfusion mismatch can be used to select patients for endovascular therapy beyond 8 h.29 Seventy-two per cent of their patients showed neurological improvement (NIHSS improvement of ≥4) post-recanalization. Currently, the DAWN (DWI/PWI and CTP Assessment in the Triage of Wake-up and Late Presenting Strokes Undergoing Neurointervention) trial is a multicenter study underway to select patients presenting after ≥8 h of stroke onset for endovascular therapy using CT and MRI based protocols.25 To the best of our knowledge, this is the first report in which delayed intracranial stenting was employed in all patients with AIS.
In our study, recanalization was achieved in 91.7% of patients. This compares favorably with data from intravenous thrombolytic studies, Prolyse in Acute Cerebral Thromboembolism (PROACT)-II and multi-MERCI trials where rates of recanalization were 35%, 66% and 70%, respectively.3 8 30 Although the incidence of intracranial hemorrhage was 16.3% in our series, none of these was associated with decline in NIHSS. Rates of symptomatic intracranial hemorrhage were 6.4%, 4.5%, 10% and 9.8% in the National Institute of Neurological Disorders and Stroke (NINDS), Desmoteplase in Acute Ischemic Stroke (DIAS-2), PROACT-II and the multi-MERCI trials, respectively, showing that angioplasty and stenting can be performed safely even in the setting of an organized, longstanding occlusive clot.4 8 9 21 30 Mortality, another indicator for safety, was 8% in our series which is lower compared with the literature (16–45% in the other trials).4 6–9 31 Lastly, 90 day favorable clinical outcome (mRS ≤1–2) was seen in 39%, 40%, 40% and 67% of patients in the NINDS, DIAS-2, PROACT-II and multi-MERCI trials, respectively.4 8 9 21 30
Multiple authors have also published their data on stenting done within 8 h of acute stroke onset.12–15 32 Most of them report median time to intervention/time to needle between 4 and 6 h. However, the median time to intervention in our case was 46 h. Our data were fairly comparable with 83–100% recanalization rate in their studies. The frequency of intracranial hemorrhage varied between 0% and 30% while mortality varied between 25% and 40% in their series. A favorable outcome, defined in all as mRS ≤3 at discharge/90 days, was seen in 22–67% of these patients. In our series, a favorable outcome, defined as mRS ≤2 at 90 days, was seen in 58% of patients (table 4).
Our study has several important limitations. Our series is qualitative, small and not enough to achieve statistical power. Additionally, not all recanalization procedures in our series were done based on uniform imaging protocols, due to the developing nature of these imaging modalities. This automatically introduces a selection bias and makes it difficult to generalize our results to other populations. Nevertheless, this shows that in a subgroup of patients, delayed revascularization can be safely and effectively performed by deployment of intracranial stents without any increased rates of complications, and with good neurological outcome.
Angioplasty and stenting can be safely and effectively done even beyond 8 h of stroke onset. In a subset of the population, delayed recanalization can result in a favorable clinical outcome. Current studies are underway to develop imaging protocols that can better select and define patient subsets that will benefit most by delayed endovascular therapy for AIS.
Competing interests None.
Ethics approval The data collection and analysis in this study was performed with the approval of the Institutional Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.