Intraarterial therapy background

Intraarterial therapy for acute ischemic stroke can be broadly divided into chemical clot dissolution with locally delivered fibrinolytic agents (ie, intraarterial fibrinolysis) and mechanical thrombectomy with various devices. Compared with intravenous fibrinolysis alone, intraarterial therapies can recanalize large arterial occlusions earlier and more frequently (58; 47; 35). Intraarterial fibrinolysis with streptokinase for large vessel occlusion was initially described in the 1980s prior to the FDA approval of intravenous rtPA (61). The first trial to evaluate the efficacy of intraarterial fibrinolysis was the Prolyse in Acute Cerebral Thromboembolism II trial (PROACT II), published in 1999, which demonstrated better outcomes with intraarterial r-pro-urokinase versus intravenous heparin alone (16). Although the PROACT II results showed clinical benefit of intraarterial thrombolysis, the results may not be currently applicable, as the control group did not receive intravenous fibrinolysis, which has become the standard of care for many acute ischemic stroke patients.

Although initial intraarterial fibrinolysis results were promising, the excitement was dampened by multiple follow-up trials, which failed to demonstrate clinical benefit of intraarterial therapy compared with intravenous fibrinolysis alone. Starting in the mid-1990s, NINDS funded a series of studies examining the efficacy and safety of intraarterial therapy (both chemical clot dissolution and eventually mechanical thrombectomy) for the treatment of severe acute ischemic strokes. These were known as the Emergency Management of Stroke (EMS) (33) and International Management of Stroke trials (IMS, IMS II, and IMS III) (26; 25; 10). Two additional randomized, controlled trials published in 2013, Mechanical Retrieval and Recanalization of Stroke Clots Using Embolectomy (MR RESCUE) (28) and SYNTHESIS (14), also failed to demonstrate statistically-significant clinical benefit of intraarterial therapy compared with standard medical management of acute ischemic stroke. The failure of these trials to demonstrate efficacy of intraarterial therapy was largely attributed to methodologic shortcomings, including long intervals between symptom onset and intraarterial therapy, lack of pretreatment noninvasive vascular imaging to confirm the presence of large vessel occlusions, and limited use of newer generation thrombectomy devices (ie, stent retrievers).

Advances in stroke imaging

Noninvasive cross-sectional imaging plays a crucial role in the workup of acute ischemic stroke. The widespread availability of computed tomography (CT) scanners at most medical centers in developed countries has made CT the modality of choice for initial diagnostic evaluation. CT technology has rapidly improved over the past decade, by decreasing acquisition times and patient radiation exposure while improving scan quality. All acute ischemic stroke patients should undergo noncontrast CT (NCCT) of the head to exclude intracranial hemorrhage, evaluate for early signs of infarction, and exclude other potential causes of neurologic deficits. Excluding intracranial hemorrhage on NCCT is the crucial first step, as the presence of hemorrhage is a contraindication to IV-rtPA (40). Infarction of greater than one third of the middle cerebral artery territory puts the acute ischemic stroke patient at increased risk of hemorrhagic conversion if IV-rtPA is administered or mechanical thrombectomy is performed (40). The Alberta Stroke Program Early CT Score (ASPECTS) (06) is a commonly used 10-point quantitative topographic score used to assess for early ischemic changes on NCCT. ASPECTS is calculated by evaluating standardized regions of the middle cerebral artery territory for early ischemic changes (ie, hypodensity). A point is subtracted for each region that demonstrates ischemic changes. A normal NCCT will have an ASPECTS score of 10, whereas lower scores indicate more extensive ischemic involvement. Higher ASPECTS scores (8 to 10) are associated with greater benefit from IV-rtPA and better functional outcomes, whereas lower ASPECTS scores (≤ 7) are associated with worse functional outcome and increased rates of symptomatic intracranial hemorrhage (sICH). The ASPECTS score can be rapidly calculated in the acute setting and provides a reasonable estimation of core infarct size. ASPECTS limitations include high inter- and intra-observer variability, use limited to the middle cerebral artery territory emphasis, watershed infarcts, and patients with extensive white matter changes or chronic infarcts. Additionally, it should be noted that a normal appearing NCCT (ASPECTS score of 10) does not exclude the presence of a large infarct core, as NCCT is much less sensitive than DWI-MRI (the gold standard for measurement of ischemic core infarct size) in the acute setting (06; 48; 04; 50). Patients with severe neurologic deficits based on the National Institute of Health Stroke Scale (NIHSS) score should also undergo a CT angiogram (CTA) of the head and neck (including aortic arch to head vertex) to evaluate for an intracranial large vessel occlusion, which is usually defined as occlusion of the intracranial internal carotid artery, M1 or proximal M2 segments of the middle cerebral artery, or the basilar artery. In rare circumstances, patients may undergo initial imaging evaluation with magnetic resonance imaging (MRI) of the head as well as magnetic resonance angiography (MRA) of the head and neck. However, CT/CTA is preferred over MRI/MRA, as CT scanners are more widely available and acquisition times are much faster.

More advanced noninvasive imaging techniques such as CT perfusion, multiphase CTA, and MR perfusion are being increasingly utilized during the work-up of acute ischemic stroke patients. These imaging tools can measure potentially salvageable ischemic brain tissue (ie, penumbra) and grade collateral circulation in the ischemic arterial territory. Many centers are using these techniques to select acute ischemic stroke patients who may benefit from mechanical thrombectomy (18). Advanced imaging techniques will likely play an increasingly pivotal role in the work-up and triage of acute ischemic stroke patients in the future.

New advances supporting mechanical thrombectomy

Several major technical advances in the neuroendovascular field occurred during the time that the IMS III, MR RESCUE, and SYNTHESIS trials were still recruiting patients. First, small studies showed that mechanical thrombectomy with newer generation devices, known as stent retrievers, was more effective than intraarterial fibrinolysis or first generation thrombectomy devices like the concentric Merci device (Stryker Inc., Fremont, CA) (38; 45; San Roman et al 2012; 55). These wire-mounted, self-expanding stents proved to be faster and easier to deploy than first-generation thrombectomy devices and had higher recanalization rates.

The second major advance was the use of noninvasive imaging techniques to better select acute ischemic stroke patients with large vessel occlusion and salvageable brain tissue who may benefit from intraarterial thrombectomy. The widespread availability, accuracy, and speed of noninvasive angiographic techniques, such as CTA and to a lesser extent MRA, proved critical to identify large vessel occlusions that could be amenable to endovascular therapy. Additionally, the recognition that noninvasive imaging techniques could define ischemic core infarcts and potentially reversible brain injury (ie, penumbra) also proved crucial to patient selection and overall safety and efficacy of mechanical thrombectomy (30; 36; 43; 34; 56; 17). As poignantly demonstrated in IMS III, the subgroup of patients randomized on the basis of having large vessel occlusion on CTA showed a statistically significant benefit from endovascular revascularization (10).

Mechanical thrombectomy for large vessel occlusions

The shortcomings of previous intraarterial therapy trials were addressed in 5 landmark prospective randomized open-label blind-endpoint trials published in 2015 that demonstrated the superiority of mechanical thrombectomy to best medical therapy in carefully identified patients with large vessel occlusions. Although all 5 study designs differed in certain details, they shared several common core principles: (1) each randomized acute ischemic stroke patients to standard medical therapy (mostly IV-rtPA) or standard medical therapy plus intraarterial thrombectomy; (2) each used noninvasive neuroimaging protocols to carefully select patients with a large vessel occlusion of the anterior circulation, a small core infarction, and the presence of salvageable brain tissue; (3) each required strict time parameters and implemented more efficient workflows to improve reperfusion; and (4) each used newer generation stent retrievers almost exclusively. Listed in chronological order of publication, they are: Multicenter Randomized Clinical Trial of Endovascular Treatment (MR CLEAN) (08), Endovascular Treatment for Small Core and Anterior Circulation Proximal Occlusion with Emphasis on Minimizing CT to Recanalization Times (ESCAPE) (20), Extending the Time for Thrombolysis in Emergency Neurological Deficits-Intra-Arterial (EXTEND-IA) (13), Randomized Trial of Revascularization with Solitaire FR Device versus Best Medical Therapy in the Treatment of Acute Stroke Due to Anterior Circulation Large Vessel Occlusion Presenting within Eight Hours of Symptom Onset (REVASCAT) (27), and Solitaire with the Intention for Thrombectomy as Primary Endovascular Treatment (SWIFT PRIME) (54). A summary of the individual trials, with their clinical applicability and results, are shown in Tables 1 through 3 below.

Table 1. Study Demographics

Study	Control/intervention	Media age (years)	Median baseline NIHSS	Large vessel occlusion distribution (%)	Median ASPECTS	Stent retriever usage (%) – intervention group only	Median time from stroke onset to groin puncture (min) – intervention group only
MR CLEAN	267/233	66/66	18/17	ICA 1.1/0.4 ICA + M1 28.2/25.3 M1 62.0/66.1 M2 7.9/7.7 A1-A2 0.8/0.4	9/9	81.5	260
ESCAPE	150/165	70/71	17/16	ICA + M1 26.5/27.6 M1/all M2 71.4/68.1 M2 2.0/3.7	9/9	86.1	185
EXTEND-IA	35/35	70/69	13/17	ICA 31/31 M1 51/57 M2 17/11	Not recorded	100	210
REVASCAT	103/103	67/66	17/17	ICA 1/0 ICA + M1 27/26 M1 64/65 M2 8/10	8/7	100	269
SWIFT PRIME	98/98	66/65	17/17	ICA 16.0/18.3 M1 77/68 M12 6/14	9/9	100	224
HERMES	653/634	68/68	17/17	ICA 22/21 M1 69/69 M2 7/8 Other 2/2	9/9	89.6	285 (Median stroke onset to reperfusion)

Table 2. Select Inclusion Criteria

Study	Age inclusion criteria (years)	Select imaging inclusion criteria (all studies required large vessel occlusion)	Initial NIHSS	Baseline mRS (pre-stroke)	Stroke onset to groin puncture (hours)
MR CLEAN	≥18 (no upper age limit)	ASPECTS documented, but not used as inclusion criteria	≥2	Not part of inclusion criteria	≤6
ESCAPE	≥18 (no upper age limit)	-CT ASPECTS ≥6 AND -Moderate-good collaterals	≥6	Barthel Index >90 (similar to mRS ≤1)	≤12
EXTEND-IA	≥18 (no upper age limit)	-CT Perfusion target mismatch profile	Documented, but not part of inclusion criteria	≤2	≤6
REVASCAT	18-80 (eventually up to 85)	-CT ASPECTS ≥7, revised to ≥9 OR -MRI ASPECTS ≥6	≥6	≤1	≤8
SWIFT PRIME	18-80	-CT Perfusion target mismatch profile OR -CT/MRI ASPECTS ≥7 OR -Infarct <1/3 of MCA territory on CT or MRI	8-29	≤2	≤6

Table 3. Select Study Outcomes

Study	Median mRS at 90 days (control or intervention)	mRS 0-2 at 90 days (%)	mRS score reduction at 90 days (shift analysis), OR (95% CI)	Death at 90 days (%)	Symptomatic ICH (%)	TICI 2b/3 (%)-Intervention group only
MR CLEAN	4/3	19.1/32.6	1.67 (1.21-2.30)	22/21	6.4/7.7 (at 90 days)	59
ESCAPE	4/2	29.3/53	3.1 (2.0-4.7)	19/10.4	2.7/3.6 (at 90 days)	72.4
EXTEND-IA	3/1	40/71	2 (1.2-3.8)	20/9	6/0 (at 36 hours)	86
REVASCAT	Not recorded	43.7/28.2	1.7 (1.05-2.8)	16/18	2/2 (at 90 days)	66
SWIFT PRIME	3/2	35/60	1.7 (1.23-2.33)	12/9	3/0 (at 27 hours)	88
HERMES	Not recorded	26.5/46	2.49 (1.76-3.53)	18.9/15.3	4.3/4.4	71

Each study demonstrated that mechanical thrombectomy significantly reduced disability rates in carefully selected patients with acute ischemic stroke caused by large vessel occlusion in the proximal anterior circulation. The patients who benefited most had a combination of large vessel occlusion, small core infarction, and large salvageable brain tissue as identified on noninvasive neuroimaging (CTA/MRA) prior to intervention. After publication of these studies, the investigators pooled their patient-level data together to create the HERMES collaboration, a meta-analysis of over 1200 patients and 600 thrombectomy cases (21). Overall, the HERMES data demonstrated that mechanical thrombectomy more than doubled the odds for a better outcome (mRS 0-2) compared to standard medical therapy alone (common OR 2.49, 95% CI 1.76-3.53). The number needed to treat with endovascular thrombectomy to reduce disability was 2.6. TICI 2b/3 reperfusion was achieved in 71% of patients randomized to endovascular thrombectomy in the 2015 trials compared to 25% and 41% achieved in the MR RESCUE and IMS III trials, respectively. Baseline NIHSS was 17 in both groups; the NIHSS at 24 hours was significantly lower in the intervention group (median NIHSS 8 in the intervention group, median NIHSS 15 in the control group), and improvement from baseline at 24 hours was significantly greater in the treatment group (median change in NIHSS -6.4 in the intervention group, median change in NIHSS -2.6 in the control group). The authors concluded that for every 100 patients treated with mechanical thrombectomy, 38 will have less disability at 90 days compared with standard medical management, and an additional 20 more patients will be functionally independent at 90 days (mRS 0-2). There was not a significant difference in serious adverse events (mortality and symptomatic intracranial hemorrhage) between the groups.

The HERMES meta-analysis favored mechanical thrombectomy across multiple subgroups including age, sex, NIHSS, tPA administration, ASPECTS, and time from onset. There was clinical benefit across all age groups, including patients 80 years old or older (80-year-old and older subgroup common OR 3.68, 95% CI 1.95-6.92). Mechanical thrombectomy was favored in both the male and female subgroups analyzed in the HERMES meta-analysis (males common OR 2.54, 95% CI 1.92-3.36 and females common OR 2.38, 95% CI 1.46-3.88). Most of the eligible patients across the 5 trials received intravenous-rtPA (1090/1278, 85%). Clinical outcomes favored mechanical thrombectomy in both subgroups (common OR 2.45, 95% CI 1.68-3.57 in the group receiving intravenous-rtPA and common OR 2.43, 95% CI 1.30-4.55 in the group not receiving intravenous-rtPA). Given this, the authors concluded that endovascular eligibility should be decided irrespective of eligibility for alteplase. HERMES analysis of NIHSS subgroups favored mechanical thrombectomy for higher NIHSS subgroups (11-15, 16-20, 21 or higher), whereas the benefit of mechanical thrombectomy in the lower NIHSS subgroup (10 or lower) was not statistically significant. The lack of conclusive efficacy of mechanical thrombectomy in the lower NIHSS subgroup may be due to insufficient patient enrollment.

The extent of pretreatment core infarction on initial diagnostic imaging has been recognized as a critical determinant of clinical outcome in patients who have had acute ischemic stroke and are receiving reperfusion therapies (either systemic thrombolysis or mechanical thrombectomy). Across all 5 trials, average ASPECTS was 9. Four of the trials required specific criterion to exclude patients with large core infarction on baseline imaging. Three of the trials (ESCAPE, SWIFT PRIME, REVASCAT) excluded patients with low ASPECTS scores (lower than 6) whereas EXTEND-IA selected patients with small core infarcts and salvageable brain tissue. MR CLEAN allowed enrollment across the spectrum of ASPECTS scores, thus, allowing subgroup analysis of patients with low ASPECTS scores (0 to 5). The HERMES analysis concluded that similar favorable outcomes with mechanical thrombectomy were seen in subgroups with high baseline ASPECTS scores (9 to 10) and moderate baseline ASPECTS scores (6 t 10). The subgroup with baseline ASPECTS scores 0 to 5, although underpowered, did not favor mechanical thrombectomy (common OR 1.24, 95% CI 0.62-2.49). However, a retrospective study demonstrated that patients with low ASPECTS may still benefit from thrombectomy (53).

Three of the 2015 studies (MR CLEAN, EXTEND-IA, SWIFT PRIME) required a 6-hour window from stroke onset to randomization. REVASCAT and ESCAPE included patients up to 8 and 12 hours, respectively, from stroke onset to randomization. Median onset to randomization in all 5 studies was 195 minutes (approximately 3 hours) with onset to reperfusion of 285 minutes (approximately 4.5 hours). The HERMES analysis also demonstrated efficiency for thrombectomy for patients randomized beyond 300 minutes after stroke onset (common OR 1.76, 95% CI 1.05-2.97) including up to 420 minutes. However, given the inclusion criteria of each trial, the HERMES analysis could not address if mechanical thrombectomy was efficacious in the delayed window (more than 6 hours from stroke onset).

The cumulative findings of these 5 trials provided American Heart Association (AHA) class 1 level of evidence for acute stroke intervention in carefully selected patients with large vessel occlusions (40). Per the 2018 AHA guidelines, there is Class 1a evidence for mechanical thrombectomy with a stent retriever if patients meet all of the following criteria: (1) prestroke mRS score 0 to 1; (2) causative occlusion of the internal carotid artery or MCA segment 1(M1); (3) age older than 18 years (4) NIHSS score higher than 6; and (6) treatment can be initiated (groin puncture) within 6 hours of symptom onset (49).

Extended window thrombectomy

Two landmark studies published in early 2018 addressed the efficacy of mechanical thrombectomy in the extended window (more than 6-hour) period: the DWI or CTP Assessment with Clinical Mismatch in the Triage of Wake-up and Late Presenting Strokes Undergoing Neurointervention with Trevo trial (DAWN) (44) and Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke trial (DEFUSE 3) (01). Although each had specific inclusion criteria, the overall paradigm was focused on selecting patients beyond 6 hours with large mismatches between the ischemic core and salvageable penumbra.

In DAWN, patients with large vessel occlusion of the anterior circulation who were last known well in the previous 6 to 24 hours were randomly assigned to thrombectomy plus standard medical care or standard care alone. To qualify, patients needed to exhibit a mismatch between clinical deficits and infarct volume as determined by MRI or CT perfusion. The mismatch criteria were defined according to age, with patients older than 80 years needing a NIHSS greater than 10 and a core infarct volume less than 20 ml to qualify. Patients between 18 and 80 years of age needed a NIHSS greater than 10 and a core volume less than 30 ml or a NIHSS greater than 20 and a core volume less than 50 ml. Additional inclusion criteria included an mRS 0 to 1, no evidence of intracranial hemorrhage, and no evidence of infarct core involving more than one third of the territory of the middle cerebral artery on baseline CT or MRI. All thrombectomies were performed with Trevo device, and concomitant stenting of the cervical internal carotid artery was not allowed. Primary endpoints included the rate of functional independence (defined as mRS 0 to 2) at 90 days.

Patients were well-matched in both groups, with median age of 70 years, median NIHSS of 17, and core volume of approximately 8 ml. The occluded vessel was most commonly the first segment of the MCA (80%), followed by the intracranial internal carotid artery (20%). The median time between last known well and study randomization was 12.2 hours in the thrombectomy group and 13.3 hours in the control group. The median time from last known well to recanalization was 13.6 hours. The trial was stopped early at 31 months after interim analysis of the first 200 enrolled patients demonstrated clear superiority of mechanical thrombectomy. Compared to standard medical care, thrombectomy greatly increased rates of functional independence (49% vs. 13%; ARR 35) with a number needed to treat of only 2.8 patients for every 1 patient with preserved functional independence at 90 days. These benefits were maintained across multiple subgroups, including sex, age, baseline NIHSS, and time between onset and randomization (less than 12 hours and 12 to 24 hours). Thrombectomy restored greater than 50% of flow (TICI 2b) or full reperfusion (TICI 3) in 84% of patients, with higher rates of recanalization at 24 hours (77% vs 39%; RR 2). Change in median infarct volume at 24 hours was significantly larger in the control group (+1ml vs +13ml, p< 0.0001). There was no difference in serious adverse events, including symptomatic intracerebral hemorrhage or 90-day mortality, between the two groups.

DEFUSE3, published several months after DAWN, randomized patients presenting with large vessel occlusion of the anterior circulation who were last known well between 6 and 16 hours to thrombectomy plus standard medical therapy versus medical therapy alone. The main inclusion criteria were a core less than 70 ml, a penumbra volume greater than 15 ml, and a penumbra-to-core ratio greater than 1.8. Thrombectomy was preformed with any FDA-approved device, at the discretion of the operator, and carotid stenting was allowed. Primary outcome was ordinal score on the modified Rankin Scale. Like DAWN, both groups in DEFUSE had the median age of 70 years and NIHSS score of 16. In the thrombectomy group, the median time from symptom onset to baseline imaging was 10 hours and 29 minutes, with a median ischemic core and penumbra volumes of 9.4 ml and 114.7 ml respectively. In the control group, median time from symptom onset to baseline imaging was 9 hours and 55 minutes, with core and penumbra volumes of 10.1 ml and 116.1 ml respectively.

The DEFUSE3 trial was also stopped early after interim analysis of first 182 patients demonstrated clear efficacy for mechanical thrombectomy. The thrombectomy group had lower median mRS score at 90 days (3 vs. 4; OR 2.77) and had a higher percentage of patients who were functionally independent (mRS 0- to 2) at 90 days (45% vs. 17%; ARR 28%) with 3.6 patients needed to preserve functional independence in 1 patient. Thrombectomy patients were more likely to have complete recanalization at 24 hours (78% vs. 18%). There was a trend towards lower median infarct growth at 24 hours (23 ml vs. 33 ml; p=0.08). Unlike in the DAWN trial, DEFUSE3 demonstrated a clear all-cause mortality benefit at 90 days (14% vs. 26%; p=0.05). There were no differences in adverse events or symptomatic intracranial hemorrhages between the intervention and control groups.

Overall, both studies demonstrated the superiority of endovascular thrombectomy in the treatment of large vessel occlusions of the anterior circulation in carefully selected patients beyond the traditional intervention period of 6 hours from last known well. Both studies focused on selecting patients with large mismatches between core lesion and salvageable penumbra. Although similar in design, both studies also had some key differences. Foremost, DAWN had a larger time window, including patients up to 24 hours from last known well. DEFUSE3 had broader inclusion criteria, including lower functioning baseline (mRS 0 to 2 vs mRS 0-1), milder clinical deficits (NIHSS greater than 6 vs. 10), and larger maximum core volume (70 mL vs. 50 mL) and did not qualify different mismatch criteria based on age. Additionally, DAWN required thrombectomies be performed with the Trevo device whereas DEFUSE3 allowed any FDA-approved thrombectomy device and allowed for concomitant cervical internal carotid artery stenting.

As a result of these 2 landmark studies, the 2018 AHA/ASA guidelines provide class 1a evidence for mechanical thrombectomy in select patients with acute ischemic strokes within 6 to 16 hours of last known well (49). Given the discrepancy in interventional time frame between the 2 studies, endovascular thrombectomy in patients presenting within 16 to 24 hours since last known well merits class IIa evidence.

Adjuvant management

As the field of endovascular therapy for acute ischemic strokes has matured, several studies have looked at perioperative management and adjuvant therapies to optimize outcomes. These have focused particularly on perioperative sedation and thrombolysis.

The 2 primary modes of perioperative sedation during endovascular therapy are generalized anesthesia via endotracheal intubation and conscious sedation using intravenous benzodiazepines. Generalized anesthesia allows for greater technical success by preventing the patient from moving during the procedure, particularly those with a dominant hemispheric stroke that interferes with verbal comprehension. However, anesthetic induction and positive pressure mechanical ventilation may result in perioperative hypotension, a potentially deleterious effect in patients with acute ischemic stroke.

Since 2016, 3 randomized clinical trials have attempted to address this controversy between generalized anesthesia and conscious sedation: Sedation vs. Intubation for Endovascular Stroke Treatment (SIESTA), Anesthesia During Stroke (ANSTROKE), and General or Local Anesthesia in Intra-Arterial Therapy (GOLIATH). The trials failed to demonstrate significant differences in patient outcomes; however, all were single center with limited sample size. Given these issues, a metaanalysis combining all 3 studies using 365 individual patient data was preformed (51). Overall, general anesthesia was associated with a significant decrease in mean arterial pressure (MAP), including lower periprocedural mean MAP values (96 vs. 100 mm HG, p< 0.01) and episodes of MAP below 70 mm HG (32% vs. 14%, p< 0.001). Sedation type did not affect patient outcome, yet a significant relationship was demonstrated between periprocedural hemodynamics and outcomes. A U-shaped association was demonstrated, with poor outcomes associated with critical MAP threshold of less than 70 mm HG for more than 10 minutes and greater than 90 mm HG for more than 45 minutes. Although it is still unclear if generalized anesthesia is superior to conscious sedation, these studies demonstrated the need to maintain periprocedural MAP between 70 to 90 mm HG. Following thrombectomy with adequate reperfusion (TICI 2b/3), intensive systolic blood pressure goal less than 140 mmHg may be reasonable to prevent reperfusion injury based on the DAWN trial. For patients with TICI 0 to 2a, less data exist to guide management; however, patients may benefit from more permissive blood pressure goals to prevent further ischemia.

Currently, patients who presented with large vessel occlusion within 4.5 hours from onset are eligible for both thrombolysis and endovascular therapy. However, there is uncertainty regarding the role of thrombolysis administered before and during thrombectomy in patients with ischemic stroke. Potential complications of this combination therapy include distal embolization from thrombolysis and increased risk of cerebral hemorrhage. A recent multicenter, randomized noninferiority trial was conducted to compare outcomes from thrombectomy with or without thrombolysis (using alteplase) in 656 eligible patients (60). Overall, endovascular monotherapy was considered noninferior to combination therapy with thrombolysis for modified Rankin scale scores between 0 to 3 and mortality at 90 days. However, the combination therapy resulted in significantly higher rates of successful reperfusion before thrombectomy (7.0% vs. 2.4%) and overall successful reperfusion (84.5% vs. 79.4%).

Future directions

Given the success of these trials in 2015 and 2018, the field of endovascular therapy for acute ischemic strokes is projected to expand even further in the future. Ongoing clinical trials are investigating increasing the intervention window beyond 24 hours in patients with salvageable penumbras. The adage of “time is tissue” may well be replaced with “penumbra is tissue.” Further studies will also need to investigate the efficacy for thrombectomy of more distal vessel occlusions. Of the 5 major 2015 thrombectomy trials, only MR CLEAN and EXTEND-IA included patients with distal middle cerebral artery occlusions (M2 segment). There were inadequate numbers of patients with occlusion of other vessels, including M3, anterior cerebral arteries, and those in the vertebrobasilar circulation, who were enrolled to allow assessment of clinical efficacy in these territories as well. Further randomized trials investigating the potential role of mechanical thrombectomy in occlusions of these vascular territories are needed.

Acute ischemic strokes due to large vessel occlusion involving the vertebrobasilar system, more specifically the basilar trunk, were not evaluated in the 2015 stroke trials. Despite the high rates of morbidity and mortality associated with basilar artery occlusions, there is a lack of evidence from randomized control trials demonstrating the efficacy of intraarterial therapy in this subset of stroke patients (29; 03; 12; 23; 46). Several small, nonrandomized studies have demonstrated good functional outcomes ranging from 30% to 73% in acute ischemic stroke patients with basilar artery occlusions treated with intraarterial therapy (02; 11; 41; 05; 37; 39). A metaanalysis published in 2014 concluded that recanalization of acute basilar artery occlusions leads to a 2-fold reduction in mortality and 1.5-fold reduction in death or disability (31). Although promising, the results of the study should be interpreted with caution, as none of the studies included in the metaanalysis were randomized-control trials. A prospective registry known as the Basilar Artery International Cooperation Study (BASICS) published in 2009 did not demonstrate the superiority of intraarterial therapy of acute basilar artery occlusions over standard medical care (57). Therefore, currently there is no proven superior treatment strategy for the treatment of acute basilar artery occlusions. The BASICS group is currently enrolling patients in a randomized, control trial evaluating the efficacy and safety of intraarterial therapy in acute basilar artery occlusions versus standard medical therapy. The results of this trial will likely influence the treatment paradigm of patients with acute basilar artery occlusions in the near future.

Additionally, future trials will further investigate the effects of thrombolysis prior to endovascular therapy. In particular, the TIMELESS trial will explore if the novel thrombolytic agent tenecteplase can improve recanalization rates and patient outcomes in carefully selected patient undergoing endovascular therapy up to 24 hours after onset of stroke symptoms (NCT03785678).