Endovascular Treatment of Acute Ischemic Stroke May Be Safely Performed With No Time Window Limit in Appropriately Selected Patients
Background and Purpose—The traditional time window for acute ischemic stroke intra-arterial therapy (IAT) is <6 hours, which is based on pharmacological thrombolysis without penumbral imaging. This study was conducted to determine the safety of patient selection for IAT based on perfusion mismatch rather than time.
Methods—A cohort of consecutive patients treated with IAT was identified by database review. Patients were selected for IAT based on the presence of perfusion mismatch using CT perfusion or MRI regardless of stroke duration. Thrombolytics were minimized after 6 hours in favor of mechanical embolectomy or angioplasty±stenting. Outcomes (National Institutes of Health Stroke Scale, modified Rankin Scale) were assessed by independent examiners. A multivariate analysis was performed to compare those treated <6 hours (early) with those treated >6 hours (late).
Results—Fifty-five patients (mean National Institutes of Health Stroke Scale=19.7±5.7) were treated, 34 early and 21 late, with mean time-to-intervention of 3.4±1.6 hours and 18.6±16.0 hours, respectively. Thrombolysis In Myocardial Ischemia 2 or 3 recanalization was achieved in 82.8% early and 85.7% late patients (P=1.0). Intracerebral hemorrhage occurred in 25.5% overall, but symptomatic intracerebral hemorrhage occurred in 8.8% of the early and 9.5% of the late patients (P=1.0). Thirty-day mortality was similar (29.4% versus 23.8%, P=0.650). At 3 months, 41.2% and 42.9%, respectively, achieved a modified Rankin Scale ≤2 (P=0.902). Only presenting National Institutes of Health Stroke Scale was a predictor of modified Rankin Scale ≤2 (OR 0.794[95% CI 0.68 to 0.92], P=0.009) and death (adjusted OR 1.29[95% CI 1.04 to 1.59], P=0.019).
Conclusions—In appropriately selected patients, IAT for acute ischemic stroke can be performed safely regardless of stroke duration. The concept of an acute ischemic stroke treatment window for IAT should be re-evaluated with a clinical trial selecting patients with perfusion mismatch.
Traditionally, patients with acute ischemic stroke (AIS) have not been treated with intra-arterial revascularization therapy (IAT) beyond 6 hours due to the perceived lack of benefit and increased risk of intracerebral hemorrhage (ICH). This concept is based on anecdotal experience with pharmacological thrombolysis and few randomized clinical trial data.1 With modern imaging studies that assess the ischemic penumbra and the availability of mechanical devices for clot removal, it may be possible to treat patients beyond the 6-hour treatment window without increasing the risk of ICH.2 The theoretical justification for this approach is that by limiting recanalization to those patients with minimal or small completed infarcts and large areas of perfusion mismatch, neuronal function may be spared without increasing the ICH risk, whereas revascularization of mostly infarcted tissue does not result in significant neuronal recovery and increases the risk of ICH because of associated endothelial, vascular and blood–brain barrier injury. Furthermore, pharmacological agents, particularly fibrinolytics, may have an increased propensity to cause ICH.1,3–6 A few case series have reported treatment beyond the 6-hour window suggesting it may be feasible to treat patients presenting late.7,8
Due to the significant number of patients with AIS presenting outside the 6-hour time window at the author’s institution, an IAT protocol was developed using primarily CT, CT angiography, and CT perfusion imaging to select patients regardless of the time window. This retrospective study was conducted to assess the safety of this approach.
Materials and Methods
A prospectively collected and Institutional Review Board-approved database of all patients with AIS was retrospectively reviewed to identify a cohort of consecutive patients treated between March 2007 and April 2009 with IAT. All IAT was carried out by the author. The decision to initiate IAT was based on the clinical diagnosis of AIS with National Institutes of Health Stroke Scale (NIHSS) ≥10 or severe aphasia, the absence of acute hypodensity on CT involving more than one third of the middle cerebral artery (MCA) territory or more than half of the brain stem, the absence of any ICH on CT, the presence of a perfusion mismatch, and in patients presenting <3 hours, a contraindication to intravenous tissue plasminogen activator. Perfusion mismatch was defined by the presence of a low cerebral blood flow region that measured ≥20% of the low cerebral blood volume region. IAT was not performed if the core infarct based on cerebral blood volume maps was more than one third of the MCA territory. Patients with brain stem ischemia >6 hours in duration were defined as having perfusion mismatch if the clinical deficit was disproportionately more severe than the diffusion-weighted imaging lesions. Time to treatment was defined as the time from stroke onset to initiation of IAT. Age >80 years, hypertension >220/120 mm Hg, hyperglycemia >300 mg/dL, and underlying dementia were considered as relative contraindications to pharmacological IAT.
All patients had CT, CT angiography, and CT perfusion within 15 minutes of arrival regardless of renal function. The CT angiography images were obtained using helical 0.75-mm thick scans at 0.5-mm intervals after injection of 100 mL of contrast at 4 mL/s. The CT perfusion images were acquired after the injection of 40 mL of contrast at 8 mL/s with 40 sequential acquisitions at 1-second intervals, each consisting of 2 adjacent 10-mm thick slices.
Six-Fr femoral access was obtained in all patients followed by administration of 2000 to 3000 U of heparin. The decision to proceed with IAT was made after angiography of the affected vessel. If the use of the Merci(Concentric Medical Inc, Mountain View, Calif) system was likely, then the 6-Fr sheath was exchanged for an 8-Fr sheath. For intra-arterial thrombolysis as first treatment, tissue plasminogen activator was given in multiple boluses over 15 to 30 minutes directly into the thrombus for a maximum of 20 mg. Additionally, abciximab, 2 to 10 mg, was given directly into the thrombus if atherothrombosis was the likely etiology. If no recanalization occurred in 30 minutes, then mechanical recanalization was attempted.
Mechanical embolectomy was performed with either the Merci Retriever with proximal occlusion through the Merci 8-Fr guide catheter or the Penumbra System (Penumbra Medical Inc, Alameda, Calif). If complete recanalization was not achieved within the first 30 to 60 minutes, then either thrombolytics or abciximab, if they had not already been given, were given up to the previously mentioned maximum doses. After 3 unsuccessful Merci passes or the use of >2 Merci devices, the Penumbra system was used and vice versa for the Penumbra system.
Angioplasty and stenting were planned as first-line treatment if an underlying atherosclerotic lesion was highly likely or if the other approaches failed. An undersized angioplasty balloon was used at nominal pressures. If an underlying stenosis was confirmed by the presence of a “waist” on the balloon, lesion irregularity characteristic of atherosclerosis, or marked lesion recoil was present, then a Vision (Abbott Vascular, Abbott Park, Ill) stent was deployed at nominal atmospheres for intracranial lesions or an approved carotid stent was deployed for proximal internal carotid artery lesions. If stent deployment was expected before the procedure, 600 mg clopidogrel and 325 mg aspirin were given through a nasogastric tube; otherwise, an abciximab half-weight-based bolus (approximately 8 to 12 mg) was given intra-arterially before stent placement (except if given earlier) followed by postprocedural 600 mg clopidogrel and 325 mg aspirin. Stenting was avoided if there was a large acute infarct on baseline CT or if thrombolytics were used.
Cases were performed under local anesthesia with judicious (25 to 100 μg) fentanyl and (1 to 4 mg) midazolam for severe agitation or discomfort. Comatose patients unable to maintain their airway underwent endotracheal intubation with propofol sedation and paralytics. Systolic blood pressures were maintained between 160 and 200 mm Hg except in those who received intravenous tissue plasminogen activator in whom it was kept ≤185/110 mm Hg. Fluid boluses, phenylephrine, β-blockers, and nicardipine were used as needed. After restoration of perfusion or for those suspected of having intracerebral hemorrhage (ICH), systolic blood pressure was immediately lowered to 90 to 120 mm Hg. In those cases in which no or incomplete recanalization was achieved, the systolic blood pressure was kept at 150 to 180 mm Hg.
Two hours after the start of the intervention or if patients had clinical deterioration, an intraprocedural CT (DynaCT; Siemens Medical Systems, Inc) was performed. If the CT was negative for ICH, then the interventional procedure was continued when appropriate but if positive, the procedure was terminated and corrective measures taken (ie, rapid lowering of systolic blood pressure <120 mm Hg, infusion of 20 to 30 mg protamine sulfate, platelet transfusion to reverse antiplatelets, and fresh–frozen plasma if thrombolytics were given).
Endovascular outcome was assessed using the Thrombolysis In Myocardial Ischemia scale with a score of 2 or 3 considered as recanalization success.9 ICH was defined using the European Cooperative Acute Stroke Study classification.10 All patients were examined by an independent vascular neurologist or stroke nurse practitioner in-hospital and at follow-up, who measured outcomes with the NIHSS and modified Rankin Scale. Patients unable to return to the clinic were contacted by telephone (as part of routine care) to assess their neurological function.
SAS Version 9.2 (SAS Institute, Cary, NC) was used for all data analysis. In the univariate analysis, all categorical variables were analyzed by Pearson χ2 or Fisher exact tests. When comparing late versus early groups, the distribution of continuous variables was examined and independent sample t tests or Mann-Whitney U tests were performed as appropriate. Multiple logistic regression using crude ORs, adjusted ORs, and 95% CIs was computed to assess time-to-treatment association with the various outcomes after controlling for pertinent clinical covariates. Outcome variables consisted of modeling recanalization success, death, dichotomized modified Rankin Scale (≤2 versus ≥3), and symptomatic and asymptomatic ICH. Time to treatment was dichotomized into early (0 to 6 ours) or late (≥6 hours) and for a subanalysis, early was redefined as 0 to 8 hours and late as >8 hours. The covariates included recanalization method (pharmacological, embolectomy, angioplasty/stenting, and combination therapy), site of occlusion (internal carotid artery, MCA, tandem internal carotid artery and MCA, or vertebrobasilar), dichotomized recanalization (unsuccessful [Thrombolysis In Myocardial Ischemia 0 to 1] versus successful [Thrombolysis In Myocardial Ischemia 2 to 3]), and presenting NIHSS.
A total of 55 patients were treated, 34 early and 21 late. The mean presenting NIHSS was 19.7±5.7 (range, 7 to 36). The mean time to treatment was 9.2±12.3 hours (range, 1 to 68 hours), but 3.4±1.6 hours and 18.6±16.0 hours in the early and late groups, respectively; median time to treatment was 5 hours overall and 3.25 hours and 12 hours, respectively. Cardioembolism was the leading stroke cause in the early group (54.8% versus 29.2%), whereas atherothrombosis was more common in the late group (29% versus 45.8%, respectively), P=0.199. The proportion treated with thrombolysis was significantly higher in the early group, 58.7% versus 23.8% (P=0.035), whereas angioplasty/stenting was more common in the late group, 26.5% versus 57.1% (P=0.023). Table 1 summarizes the patient characteristics and procedural details.
Successful recanalization was achieved in 84.0% of all patients (82.8% early and 85.7% late, P=1.0). ICH occurred in 25.5% and symptomatic ICH was seen in 9.1% (8.8% early versus 9.5% late, P=1.0). At 30 days, mortality was similar between the 2 groups (29.4% early versus 23.8% late, P=0.65). Overall, 41.8% achieved an modified Rankin Scale ≤2 at 3 months, 41.2% in the early group and 42.9% in the late group (P=0.902). Table 2 lists the outcomes and complications.
In univariate analysis and multivariate modeling, only presenting NIHSS, not time to treatment or recanalization method, was a predictor of modified Rankin Scale ≤2 (adjusted OR 0.794 [95% CI 0.68 to 0.92], P=0.009) and death (adjusted OR 1.29 [95% CI 1.04 to 1.59], P=0.019), although successful recanalization might also have had an influence (adjusted OR 5.28 [95% CI 0.56 to 49.65], P=0.074). The results were unchanged when only anterior circulation strokes or, in separate analysis, isolated MCA occlusions were considered and they were similarly unchanged when the early group was redefined as 0 to 8 hours.
Early recanalization is the most effective treatment for AIS, yet poorly timed recanalization is a major predictor of ICH, which has been the major obstacle limiting widespread use of recanalization therapy. For patients treated with intravenous tissue plasminogen activator, the effective time window is 3 to 4.5 hours, although the major clinical benefit is in those receiving therapy within 90 minutes.11,12 The time window is longer, up to 6 hours, in patients treated with IAT, offering a robust clinical benefit but with an increased risk of ICH.1 This apparent time dependence has been a fundamental principal of stroke neurology. Yet, the author and his colleagues have anecdotal cases treated beyond 6 hours who had excellent clinical outcomes.7,8,13,14 This current study suggests that in select patients with AIS with potentially salvageable brain tissue defined by CT perfusion (anterior circulation) or clinical–MRI mismatch (posterior circulation), IAT beyond 6 hours is feasible and safe with no increase in the risk of ICH or death. Moreover, this approach may be as effective as <6-hour IAT.
The safety of IAT in the late group in this series may have been due to several factors. First, and likely most important, is the selection of patients with perfusion imaging. Numerous preclinical studies and clinical trials have shown that reperfusion in the presence of a large perfusion mismatch and small necrotic core results in brain tissue salvage,15 whereas reperfusion of necrotic core in the absence of perfusion mismatch is ineffective and may increase the risk of ICH.16 Second, in this series, pharmacological thrombolysis was kept to a minimum, especially in those treated in the late group. There are data to support this approach, not the least of which are the intravenous thrombolysis trials that have clearly shown a relatively short time window for intravenous thrombolysis safety.11,12,17 Animal models have also shown that fibrinolytics, especially tissue plasminogen activator, are associated with a higher propensity to cause ICH independent of their recanalization efficacy.5 Other factors include the individualization of the therapy (ie, thrombolysis, embolectomy, or stenting) to each patient taking into account the known factors that predispose to ICH (ie, hyperglycemia, hypertension, age, size of necrotic core, etc) as well as probable stroke etiology.13,17,18 In general, randomized trials have not taken all of these factors into account when determining type and dosage of therapy, but rather such characteristics were exclusion criteria or were not considered in the treatment algorithm. The author has previously reported on this multimodal approach and subsequently larger series have also shown that a multimodal approach may be more efficacious than fibrinolysis alone.13,17,19
The proportion of patients treated with stents was higher in the late group. This was expected because stenting was reserved for those with an underlying atherosclerotic stenosis and there was an overrepresentation of atherosclerotic occlusions in the late group. The proportion of atherosclerotic strokes was also higher than would have been expected based on the prevalence of large vessel atherosclerosis in AIS in the stroke belt.20 In this series, the late group fared better than would be expected based on duration of ischemia, which in the multivariate analysis was not related to stenting, although the number treated with stenting was small, which likely affected the ability to detect a benefit. The benefit could have been due to, as discussed previously in this article, less neurotoxicity from fibrinolytics. The underlying pathogenesis of the occlusion could also be the explanation because patients with underlying atherosclerotic stenoses likely had developed some degree of chronic pial or other collaterals and therefore were more likely to progress slowly and present later than those with embolic occlusion and they may also have had ischemic preconditioning increasing tolerance to ischemia21,22; the latter could explain in part the lower NIHSS scores in the late group. This finding is of importance because approximately 20% of ischemic strokes are due to large vessel atherosclerosis with 8% to 10% due to intracranial atherosclerosis.23 Furthermore, it is reasonable to postulate that mechanical embolectomy devices (eg, Merci and Penumbra) may not be as safe or effective in patients with an underlying stenosis, which may prevent clot retrieval, increase the risk of vessel injury, or predispose to reocclusion and clinical deterioration. It therefore may be important to attempt to determine the cause of the vascular occlusion before deciding on which endovascular approach to attempt and to consider stenting in those who present late. This concept that therapy be based on known pathophysiological mechanisms has not been systematically studied in the setting of AIS and in the ultra-acute setting determining the etiology of a large vessel occlusion may not be possible; therefore, this approach requires validation. This is also relevant because if stents are placed, patients will need dual antiplatelet therapy acutely as well as chronically and this can increase the risk of ICH.24 In this series, such an association was not found, but the number of patients treated was small.
The limitations of this study are the retrospective nature of the database review and small sample size. The latter limits the ability to derive statistically significant results from the data and the ability to correct for differences between the 2 groups. However, all outcomes were measured by independent clinicians who documented the data prospectively as part of the standard of care for all patients with stroke at our institution. Because all patients treated were included in this series, the bias should have been minimized. Another limitation is the relatively higher number of vertebrobasilar occlusions in the late group, although the majority of patients treated had anterior circulation AIS. This group of patients has been reported to have potentially longer treatment windows, possibly related to more robust collaterals or relative resistance of brain stem neurons to ischemia as well as a higher likelihood of atherosclerotic occlusions, which may allow for greater collaterals. Also, CT-based penumbral imaging is not adequate to assess the brain stem and the use of diffusion-weighted imaging–clinical mismatch has not been validated. Therefore, a study of just patients with anterior circulation strokes may have been more appropriate, although after excluding the vertebrobasilar group, the results of this study did not change. Another limitation of this study is that there is currently no consensus on what constitutes a penumbra and which imaging modality best defines it; an analysis that included outcomes data on patients not treated due to lack of perfusion mismatch would have strengthened the conclusions, but these data are not available. Lastly, based on the published mechanical embolectomy registries, some may argue that an 8-hour time window is a more appropriate definition of early IAT.2,6 A 6-hour cutoff was chosen for this study because the clinical outcomes of the patients studied in those registries have generally not been good and therefore it could not be assumed that IAT up to 8 hours was as safe as <6-hour IAT. The latter is also supported by the results of the only randomized trial, Prolyse in Acute Cerebral Thromboembolism (PROACT) II, which had a 6-hour window and which is arguably the cleanest published data set of IAT.1 Nevertheless, when reanalyzed using an 8-hour cutoff, the results of the study were unchanged.
This study showed that multimodal IAT may be safely used to treat patients who present beyond the traditional 6-hour time window if they are selected based on the presence of perfusion mismatch. However, more data are needed to better define which imaging criteria best define mismatch in clinical practice followed by validation of this approach through larger prospective trials.
I acknowledge the invaluable assistance and expertise of the Stroke Team: Kerri Remmel, MD, Vincent Truong, MD, Rori Spray, ARNP, CNRN, Betsy Wise, ARNP, and Craig Zeigler, PhD.
A.A.-C. is on the speaker’s bureau for BMS/Sanofi Partnership and on the advisory board for Arterain Medical Inc.
- Received January 14, 2010.
- Revision received March 5, 2010.
- Accepted March 25, 2010.
Furlan A, Higashida R, Wechsler L, Gent M, Rowley H, Kase C, Pessin M, Ahuja A, Callahan F, Clark WM, Silver F, Rivera F. Intra-arterial prourokinase for acute ischemic stroke. The PROACT II study: a randomized controlled trial. Prolyse in Acute Cerebral Thromboembolism. JAMA. 1999; 282: 2003–2011.
Smith WS, Sung G, Saver J, Budzik R, Duckwiler G, Liebeskind DS, Lutsep HL, Rymer MM, Higashida RT, Starkman S, Gobin YP, Frei D, Grobelny T, Hellinger F, Huddle D, Kidwell C, Koroshetz W, Marks M, Nesbit G, Silverman IE. Mechanical thrombectomy for acute ischemic stroke: final results of the Multi MERCI trial. Stroke. 2008; 39: 1205–1212.
Nakano S, Iseda T, Yoneyama T, Kawano H, Wakisaka S. Direct percutaneous transluminal angioplasty for acute middle cerebral artery trunk occlusion: an alternative option to intra-arterial thrombolysis. Stroke. 2002; 33: 2872–2876.
Dijkhuizen RM, Asahi M, Wu O, Rosen BR, Lo EH. Rapid breakdown of microvascular barriers and subsequent hemorrhagic transformation after delayed recombinant tissue plasminogen activator treatment in a rat embolic stroke model. Stroke. 2002; 33: 2100–2104.
Ning M, Furie KL, Koroshetz WJ, Lee H, Barron M, Lederer M, Wang X, Zhu M, Sorensen AG, Lo EH, Kelly PJ. Association between tPA therapy and raised early matrix metalloproteinase-9 in acute stroke. Neurology. 2006; 66: 1550–1555.
The Penumbra Pivotal Stroke Trial: safety and effectiveness of a new generation of mechanical devices for clot removal in intracranial large vessel occlusive disease. Stroke. 2009; 40: 2761–2768.
Barnwell SL, Clark WM, Nguyen TT, O'Neill OR, Wynn ML, Coull BM. Safety and efficacy of delayed intraarterial urokinase therapy with mechanical clot disruption for thromboembolic stroke. AJNR Am J Neuroradiol. 1994; 15: 1817–1822.
Larrue V, von Kummer RR, Muller A, Bluhmki E. Risk factors for severe hemorrhagic transformation in ischemic stroke patients treated with recombinant tissue plasminogen activator: a secondary analysis of the European-Australasian Acute Stroke Study (ECASS II). Stroke. 2001; 32: 438–441.
Abou-Chebl A, Bajzer CT, Krieger DW, Furlan AJ, Yadav JS. Multimodal therapy for the treatment of severe ischemic stroke combining GPIIb/IIIa antagonists and angioplasty after failure of thrombolysis. Stroke. 2005; 36: 2286–2288.
Wintermark M, Flanders AE, Velthuis B, Meuli R, van Leeuwen M, Goldsher D, Pineda C, Serena J, van der Schaaf I, Waaijer A, Anderson J, Nesbit G, Gabriely I, Medina V, Quiles A, Pohlman S, Quist M, Schnyder P, Bogousslavsky J, Dillon WP, Pedraza S. Perfusion-CT assessment of infarct core and penumbra: receiver operating characteristic curve analysis in 130 patients suspected of acute hemispheric stroke. Stroke. 2006; 37: 979–985.
Gupta R, Yonas H, Gebel J, Goldstein S, Horowitz M, Grahovac SZ, Wechsler LR, Hammer MD, Uchino K, Jovin TG. Reduced pretreatment ipsilateral middle cerebral artery cerebral blood flow is predictive of symptomatic hemorrhage post-intra-arterial thrombolysis in patients with middle cerebral artery occlusion. Stroke. 2006; 37: 2526–2530.
Gupta R, Vora NA, Horowitz MB, Tayal AH, Hammer MD, Uchino K, Levy EI, Wechsler LR, Jovin TG. Multimodal reperfusion therapy for acute ischemic stroke: factors predicting vessel recanalization. Stroke. 2006; 37: 986–990.
Vora NA, Gupta R, Thomas AJ, Horowitz MB, Tayal AH, Hammer MD, Uchino K, Wechsler LR, Jovin TG. Factors predicting hemorrhagic complications after multimodal reperfusion therapy for acute ischemic stroke. AJNR Am J Neuroradiol. 2007; 28: 1391–1394.
Lin R, Vora N, Zaidi S, Aleu A, Jankowitz B, Thomas A, Gupta R, Horowitz M, Kim S, Reddy V, Hammer M, Uchino K, Wechsler LR, Jovin T. Mechanical approaches combined with intra-arterial pharmacological therapy are associated with higher recanalization rates than either intervention alone in revascularization of acute carotid terminus occlusion. Stroke. 2009; 40: 2092–2097.
Lackland DT, Bachman DL, Carter TD, Barker DL, Timms S, Kohli H. The geographic variation in stroke incidence in two areas of the southeastern stroke belt: the Anderson and Pee Dee Stroke Study. Stroke. 1998; 29: 2061–2068.
Brandt T, von Kummer R, Muller-Kuppers M, Hacke W. Thrombolytic therapy of acute basilar artery occlusion. Variables affecting recanalization and outcome. Stroke. 1996; 27: 875–881.
Zhang J, Yang ZJ, Klaus JA, Koehler RC, Huang J. Delayed tolerance with repetitive transient focal ischemic preconditioning in the mouse. Stroke. 2008; 39: 967–974.
Sacco RL, Kargman DE, Gu Q, Zamanillo MC. Race–ethnicity and determinants of intracranial atherosclerotic cerebral infarction. The Northern Manhattan Stroke Study. Stroke. 1995; 26: 14–20.
Diener HC, Bogousslavsky J, Brass LM, Cimminiello C, Csiba L, Kaste M, Leys D, Matias-Guiu J, Rupprecht HJ. Aspirin and clopidogrel compared with clopidogrel alone after recent ischaemic stroke or transient ischaemic attack in high-risk patients (MATCH): randomised, double-blind, placebo-controlled trial. Lancet. 2004; 364: 331–337.