Neurointerventional therapy for large vessel occlusion stroke : the new standard of care

Corresponding author: Doungporn Ruthirago Contact Information: Doungporn.ruthirago @ttuhsc.edu DOI: 10.12746/swrccc2016.0416.216 For the past two decades, intravenous tissue plasminogen activator (IV tPA) has been the gold standard treatment of acute ischemic stroke (AIS) for patients presenting to the hospital in the first 4.5 hours after symptom onset. However, in patients with AIS due to intracranial large vessel occlusion (LVO), IV tPA has very poor recanalization rates. This group of patients has significantly worse outcomes than those without LVO. Endovascular therapy has evolved significantly since the first trial in 1998. With the publication of recent trials using modern stent-retriever devices and selection of patients with LVO, endovascular therapy has become the standard of care for patients with the most severe ischemic strokes. In this article we outline the two decade evolution of this therapy.

For the past two decades, intravenous tissue plasminogen activator (IV tPA) has been the gold standard treatment of acute ischemic stroke (AIS) for patients presenting to the hospital in the first 4.5 hours after symptom onset.However, in patients with AIS due to intracranial large vessel occlusion (LVO), IV tPA has very poor recanalization rates.This group of patients has significantly worse outcomes than those without LVO.Endovascular therapy has evolved significantly since the first trial in 1998.With the publication of recent trials using modern stent-retriever devices and selection of patients with LVO, endovascular therapy has become the standard of care for patients with the most severe ischemic strokes.In this article we outline the two decade evolution of this therapy.

First generation endovascular trials
In an attempt to increase recanalization rates in AIS patients, the first randomized controlled trial (RCT) using intra-arterial thrombolysis (pro-urokinase), Prolyse in Acute Cerebral Thromboembolism (PROACT), was completed in 1998. 1 This study showed superiority in recanalization in acute LVO stroke compared with placebo but unfortunately also an increased risk of symptomatic hemorrhage.In spite of this initial result, pro-urokinase was taken off of the market.
With the hope for better treatment options, additional trials of endovascular intervention were completed.The interventions included intra-arterial thrombolysis using alteplase tPA and the Merci thrombectomy device.In 2013, three studies were published, "IMS III, MR RESCUE, and SYNTHESIS Expansion". 2,3,4 Tse three multicenter, prospective RCTs showed no benefit in the intervention arm but also showed no additional risk of symptomatic bleeding after the intervention.Several concerns were raised regarding some aspects of these trials, including non-universal determination of LVO, use of first generation lower-efficacy devices such as the Merci device, the use of intra-arterial tPA without device in some trials (e.g., 66% of endovascular patients in the SYNTHESIS trial were treated with intra-arterial tPA alone), high utilization of intra-arterial tPA and heparin, and slow randomization to arterial puncture time (Table 1 and Table 2).
While these three trials were being conducted, two additional RCTs, SWIFT and TREVO, examined the efficacy and safety of second generation devices (stent retrievers) compared to first generation devices (Merci). 5,6 he studies demonstrated significantly better recanalization rates with the new devices without any excess in hemorrhagic complications (Table 3).

Pavis L. et al.
Neurointerventional therapy Having learned the limitations in the initial studies and with cautious enthusiasm from the device vs. device trials, new endovascular stroke trials were designed and conducted.These second generation RCTs had universal determination of the presence of LVO prior to randomization.Several of the studies initiated optimized workflow protocols to achieve faster randomization to arterial puncture times, and all of the studies used second generation stent-retriever mechanical thrombectomy devices.The first of these, MR CLEAN, was published in late 2014. 7his study compared endovascular therapy to best medical treatment in stroke patients with LVO who presented within hours of symptom onset.Following the presentation of this study at the 2014 World Stroke Congress, multiple endovascular trials were stopped by the respective data safety and monitoring boards due to efficacy of endovascular therapy on interim analysis.The four RCTs (ESCAPE, REVASCAT, EXTEND IA, and SWIFT PRIME) showed benefit of endovascular stroke therapy and were presented and published in 2015 (Table 4). 8,9,10,11 Thse five studies benefited from the first generation trials by optimizing study design and conduct.However, there were still unanswered questions and limitations to these studies (Table 5).

guidelines, data synthesis and meta-analyses
In the summer of 2015 the American Hearth Association/American Stroke Association guidelines were updated to reflect changes in the evidence base.The recommendations based on Class I evidence included administration of tPA in eligible patients even if endovascular therapy was considered and administration of endovascular therapy to adult patients with appropriate LVO ischemic stroke presenting within six hours of symptom onset who meet additional clinical and radiological criteria (stroke severity, CT ASPECT score ≥6).The recommendations emphasized the preference for second generation devices (stent-retrievers) over intra-arterial tPA.Also strongly recommended was the need for rapid evaluation of stroke patients with non-invasive imaging to determine presence of LVO stroke.After the publication of the guidelines several attempts at data synthesis using traditional meta-analyses and individual patient data (IPD) meta-analysis were conducted.The traditional meta-analysis combined the results from first and second generation trials and demonstrated that the overall result still demonstrated superiority of endovascular therapy compared to best medical therapy in acute ischemic stroke. 12Subsequent to this the HERMES clinical trial collaboration pooled the results of the trials in an individual patient data (IPD) meta-analysis to address some of the subgroups who were under-represented in the individual trials (Table 5). 13The results confirmed the overall superiority of endovascular therapy.Also, the IPD meta-analysis demonstrated substantial efficacy of endovascular therapy within the late >5 hour window without an increase in symptomatic intracranial hemorrhage; most of the patients treated late were treated between 5-8 hours.Furthermore, patients who were ineligible for tPA also benefited from endovascular therapy compared with conservative management.Interestingly, the IPD meta-analysis confirmed initial observations from subgroup analysis of the individual trials in that the treatment effect was modified by age.Although the therapy was positive in all age groups, patients who were older than 80 years had an even higher benefit from endovascular therapy.
Last, it is important to note that the overall ef-fect size of endovascular therapy is large.The number needed to treat for benefit (NNTB) for functional independence with endovascular therapy in AIS ranges from 4 in the ESCAPE and SWIFT PRIME trials to 7 in MR CLEAN and REVASCAT.When meaningful improvement in disability (modified Rankin scale shift) was used, the NNTB was 2.6 based on the HERMES meta-analysis.This contrasts with the previous standard of care, intravenous tPA vs. placebo, which had an NNTB to achieve normal functional status of 8-15 depending on the time to treatment delay.A recent study analyzing cost-effectiveness among second generation endovascular RCTs found that adding endovascular treatment to standard stroke therapy such as IV tPA is not only cost-effective but also cost saving. 14ile stent retriever thrombectomy was the first endovascular technique to show efficacy in pivotal randomized controlled trials of large vessel occlusion stroke, other modern techniques have also been developed.Aspiration thrombectomy using large bore intracranial catheters either alone or in combination with stent retrievers and augmentation of stent retrievers with aspiration via balloon guided catheters have shown promising results in observational studies. 15,16he first randomized trial comparing some of these techniques with a novel stent retriever (Penumbra 3D Revascularization Device) has been completed with encouraging preliminary results, suggesting compa-

conclusions
Several clinical trials, registries, and thousands of treated patients were required to refine the patient selection, device development, stroke response workflow, and clinical trials design.All of this was necessary to establish a firm evidence base for this "new" therapy.However, there are many unanswered questions left: What to do with patients who wake up with stroke symptoms and patients with unknown time of onset?What is the upper limit of the time window?Do we need additional imaging or clinical criteria to select patients at delayed time windows?Should pediatric stroke patients also be included?Is there a role for augmenting the effect of mechanical thrombolysis using medications, such as antithrombotic medications, neuroprotective agents or cell-based therapies?The next generation of endovascular stroke RCTs is already underway to address some of these questions.
Until the results of these next studies are available, current healthcare providers, health system directors, and policy makers have the task of implementing endovascular therapy for all eligible patients to decrease the disability from this disease.This will require restructuring of programs, additional personnel, and system wide coordination.The time is now to deliver the new standard of care.

Table 1
Strengths and weaknesses of first generation endovascular stroke RCTs

Table 3
First generation devices versus second generation devices RCTs National institute of health stroke scale; LVO-large vessel occlusion; TIMI-thrombolysis in myocardial infarction; TICI-thrombolysis in cerebral infarction

Table 4
Second generation endovascular RCTs minogen activator; mRS-modified ranking scale; sICH-symptomatic intracerebral hemorrhage; NNTB-number needed to treat for benefit; NR-not reported

Table 5
Strengths and weaknesses of second generation endovascular stroke RCTs