Author + information
- Received September 25, 2018
- Revision received January 4, 2019
- Accepted January 7, 2019
- Published online March 25, 2019.
- aDepartment of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota
- bDepartment of Neurosurgery, Jacobs School of Medicine and Biomedical Sciences, Canon Stroke and Vascular Research Center, University at Buffalo, Buffalo, New York
- cDepartment of Neurosurgery, Gates Vascular Institute at Kaleida Health, Buffalo, New York
- dJacobs Institute, Buffalo, New York
- ↵∗Address for correspondence:
Dr. L. Nelson Hopkins, Department of Neurosurgery, University at Buffalo, 100 High Street, Suite B4, Buffalo, New York 14203.
• Ischemic stroke is now treatable with catheter intervention. Revascularization must be accomplished rapidly for optimal results, but there are not enough comprehensive stroke centers or neurointerventionists to achieve this.
• Cardiologists have excellent catheter skills and extensive experience with emergency intervention, and catheterization labs are widespread, suggesting a possible solution for needed immediate intervention worldwide.
• Stroke intervention must be rapidly expanded geographically and with a dramatic increase in the number of willing and trained interventionists to meet the enormous public health need.
Stroke is a catastrophic event for patients and their families. Given the frequency of approximately 750,000 events annually with their associated morbidity and mortality, stroke has assumed increasing importance. Scientific study has identified several diseases categorized under the broad term of “stroke” that form the rationale for current treatment strategies. This paper reviews new information, especially on ischemic stroke (particularly large-vessel occlusions), which identifies the potential for new approaches that can dramatically improve outcome but will require the need to enhance and embrace the care team required to deliver optimal care and address current unmet clinical needs.
Stroke is a major concern to patients, their families, and society as a whole, given the associated mortality and morbidity (1). Occurring in approximately 750,000 U.S. patients and 15,000,000 patients worldwide, stroke requires optimization of therapeutic strategies. Treatments for acute ischemic stroke (AIS) continue to evolve with the accumulation of new scientific data from randomized controlled trials and registries. Accordingly, the guidelines have been updated in the 2018 American Heart Association/American Stroke Association stroke early management document (2). Key principles remain, namely, that the brain is exquisitely sensitive to ischemia and that time to treatment is of paramount importance, not only in improving patient outcomes, but also in providing a metric for quality of care. The central role of early thrombolysis in eligible patients continues to be emphasized. Overlying these principles of care, clinical trial results have now accumulated regarding more optimal patient selection criteria, changes in imaging modalities and parameters, as well as new strategies of care. Of particular importance has been documentation of the role of mechanical thrombectomy and expansion of the time window for endovascular treatment from <6 h for thrombolysis to 24 h for select patients on the basis of perfusion imaging findings. This evolution in strategies of care is following a trajectory similar to what was seen in the field of care for ST-segment elevation acute myocardial infarction (STEMI). This paper explores issues in implementation of the most recent strategies of care for AIS, assimilating the lessons learned in the treatment of STEMI. Included is a critical discussion of training for interventional cardiologists and other qualified interventionists as part of stroke teams to deal with unmet needs of patients with these disabling clinical events, especially in areas devoid of comprehensive stroke centers (CSCs).
Background: Treatment of Acute Arterial Occlusion
Earlier efforts on treatment of acute arterial occlusion focused on STEMI were based on the pathophysiology associated with acute thrombotic occlusion of a major coronary vessel.
Accordingly, emphasis was placed on the use of thrombolytic therapy with the development of strategies to efficiently administer it and to then gauge its effectiveness. As the field evolved, interventional strategies with initially percutaneous transluminal coronary angioplasty and then stent implantation were studied in multiple registries and randomized controlled trials. There was initial substantial debate regarding thrombolysis versus mechanical percutaneous coronary intervention (PCI) (3). Concerns were related to the magnitude of improvement in outcome, the risk–benefit ratio, and delivery-of-care models, including patient transport to centers capable of delivering care. Those issues were subsequently resolved by the data such that guidelines indicate that optimal care of STEMI is primary PCI if it is available in experienced centers within 4 to 6 h of symptom onset. Subsequent metrics of quality include a door-to-balloon time of 90 min (4). In addition, strategies were developed for patients seen at centers in which PCI was not immediately available and included the concept of initial thrombolysis followed by urgent transfer to a PCI center (5).
The evolution of treatment strategies for acute arterial occlusion in the brain with acute stroke is following the same trajectory (2). The issues are similar in terms of patient selection, risk–benefit ratio, and magnitude of improvement in the delivery of care at central and peripheral referral hospitals, including concepts of spoke and hub institutions. The issues are also similar with the evolution of stroke intervention, beginning with thrombolytics and only recently evolving to Level of Evidence: I of substantial benefit for intervention to reopen the artery. Given the magnitude of the clinical problem, resolution of these issues is of great importance.
Although the basic concept of reopening an acutely occluded major artery in the heart or brain has some similarities, several factors clearly differentiate AIS from STEMI. Acute myocardial infarction usually results from rupture of a pre-existing plaque in a coronary artery; by contrast, cardioembolic strokes comprise approximately 20% of AIS with an additional 10% to 15% of emboli originating in carotid artery plaques (6). In these patients, there is a high rate of recurrence. Thus, the major goal in AIS is most commonly removal of the offending embolus, rather than stenting the plaque, to restore flow (Central Illustration).
There are also important differences in the specific target organ bed. Both the heart and the brain have ischemic thresholds beyond which success rates are low if reperfusion therapy is initiated. The effects of organ dysfunction are variable in the different beds. Myocardial infarction with suboptimal reperfusion may lead to varying degrees of left ventricular dysfunction for which medical therapy may result in marked improvement and even render the patient asymptomatic at normal levels of activity. By contrast, failure of reperfusion for stroke leads to significant, often disabling neurological dysfunction or, at a minimum, residual cognitive defects that may be very important to the patient and family. In addition, recent stroke studies confirm that brain tissue is more sensitive to ischemia than myocardium, so the time to revascularization is more critical with AIS (7,8).
Another factor pertinent to this discussion is the relative fragility of the intracranial arteries as compared with coronary arteries. Structurally, major intracranial arteries are approximately one-third to one-half the thickness of coronary arteries and are predominately composed of media (9,10). In addition, very small and fragile perforating branches from major intracranial arteries can inadvertently be instrumented with guidewires during endovascular intervention, resulting in vessel laceration and intracranial hemorrhage. Such iatrogenic hemorrhage from intracranial arteries may be fatal or result in worsening of the stroke and reducing the chances for recovery. Finally, to reach the area of occlusion, the operator must be able to cannulate the internal carotid artery and navigate the devices required for thrombectomy distal to the intracranial carotid siphon.
The discussion in the preceding text highlights the need for stroke interventionists to acquire an understanding of basic craniovascular pathophysiology and anatomy, and receive training in safely accessing the fragile cerebrovasculature. A mastery of safe and effective use of the tools developed specifically for intracranial procedures is essential. Also important is an understanding of the clinical evaluation of the patient who may (or may not) have an evolving stroke. Finally, the appropriate choice of imaging modalities is fundamental to accurately diagnose AIS with emergent large-vessel occlusion (ELVO) to help determine whether intervention may be feasible or helpful.
Clinical Trials Data
Although early randomized controlled trials comparing intravenous tissue plasminogen activator to catheter-based intervention for AIS showed no significant benefit for intervention, they demonstrated its safety. The IMS (Interventional Management of Stroke) III, MR RESCUE (Mechanical Retrieval and Recanalization of Stroke Clots Using Embolectomy), and SYNTHESIS EXPANSION (Synthesis Expansion; A Randomized Controlled Trial on Intra-Arterial Versus Intravenous Thrombolysis in Acute Ischemic Stroke) trials compared endovascular stroke therapies to systemic thrombolysis (11–13). These trials had significant limitations. Large-vessel occlusion was not confirmed in many cases, delays in initiating the therapy were common, and clot-removal devices were early generation and less effective than current tools (14,15).
The next generation of stroke trials reported in 2015 compared modern mechanical stroke thrombectomy stent retrievers plus intravenous thrombolytics with intravenous thrombolytics alone (Table 1) (16). These trials showed substantial benefit for endovascular intervention compared with intravenous thrombolysis alone up to 6 h from stroke symptom onset or time “last known to be well.”
Data in support of thrombectomy during the 6-h to 12-h window were not as robust due to the limited number of patients treated outside the 6-h window and were derived from post hoc analysis of the original HERMES (Highly Effective Reperfusion Evaluated in Multiple Endovascular Stroke Trials) collaboration dataset (17). Subsequent studies, namely, the DAWN (Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention With Trevo) (18) and DEFUSE 3 (Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3) (19) trials, focused on the treatment of strokes >6 h on the basis of perfusion imaging. These findings now form the scientific rationale of guideline-based care.
Goyal et al. (20) performed a meta-analysis of individual patient data from the 2015 randomized controlled trials. The collective enrollment was 1,287 subjects; 634 were assigned to endovascular thrombectomy and 653 to a control group that included best medical care. The primary outcome was the degree of disability on the modified Rankin scale (mRs) at 90 days. Preceding intravenous alteplase was used in 83% of the endovascular group and 87% of the control group, and endovascular treatment was administered within 180 min of presentation in 70% of cases. The group receiving endovascular thrombectomy had significantly less disability at 90 days compared with the control group (adjusted common odds ratio: 2.49; 95% confidence interval: 1.76 to 3.53; p < 0.001). The number needed to treat to reduce disability by 1 or more points on the mRs was 2.6.
The thrombolytic agent used in combination with thrombectomy in these earlier trials most commonly was alteplase. The use of alteplase has been studied further in a head-to-head comparison with the more fibrin-specific tenecteplase (21). In this study, the primary outcome was reperfusion of >50% of the ischemic territory or absence of a retrievable thrombus at the time of intervention. The median time from stroke symptom onset to initiation of intravenous thrombolysis was 125 min for tenecteplase and 134 min for alteplase. The study investigators found that tenecteplase before thrombectomy was associated with better functional outcome and higher incidence of reperfusion. Accordingly, practice may continue to evolve with more selective thrombolytic drugs to further optimize care and outcome.
Data from these trials showed that optimal results were time-dependent. Meta-analysis of individual patient data from the original HERMES collaboration showed that each 1-h delay to reperfusion was associated with a less favorable degree of disability (common odds ratio: 0.84; 95% confidence interval: 0.76 to 0.93) (20). The next set of trials focused on the potential to expand the time to treatment and therefore expand the number of patients who might benefit. The results have important implications because symptom recognition as well as transport times may involve significant delays before the patient reaches medical care. In addition, many smaller, more remote medical centers may not have the capability to manage acute stroke as recommended by treatment guidelines. The importance of imaging triage in patients arriving outside the 6-h window was demonstrated in the DEFUSE 3 and DAWN trials (18,19). These trials demonstrated marked improvement for patients last known well up to 16 to 24 h before randomization (in the DAWN trial) or initiation of thrombectomy (in the DEFUSE 3 trial), including patients who awaken having had a stroke during the night (“wake-up stroke”). In these patients, there was often severe clinical impairment at baseline presentation. However, when perfusion imaging was applied, there were some patients with large areas of ischemia, but considerably smaller areas of core infarction, that were amenable to treatment.
The importance of specific perfusion imaging data for optimal patient selection in this “mismatch” group was studied in the DEFUSE 3 trial, a multicenter, randomized, open-label trial that evaluated 182 patients 6 to 16 h after they were last known to be well (19). Patient selection criteria included a proximal middle cerebral artery or internal carotid artery occlusion with an initial infarct size of <70 ml, but a ratio of the volume of ischemic tissue on perfusion imaging to infarct volume of ≥1.8, indicating that the region of tissue ischemia was greater than the area of infarction. The primary endpoint was the ordinal score on the mRs. A 90-day mRs score of 0 to 2 (indicating less disability) was documented in 45% of the thrombectomy group compared with 17% in the control group (p < 0.001). There was also a reduction in 90-day mortality in the thrombectomy group versus the control group (p = 0.05). These and other data indicate that the time window for dramatic improvement in outcome with invasive intervention, based on perfusion imaging, may be significantly longer in selected patients than what had been initially thought, broadening strategies for care of patients with AIS.
As stroke therapy awareness grew and hospital reimbursement for treating AIS with thrombolytics increased, the concept of stroke centers was developed and defined. Individual states developed their own definitions of stroke centers on the basis of their demographics and politics. Currently, there are 4 basic types of stroke centers (namely, stroke-ready, primary, comprehensive, and thrombectomy-capable) defined and accredited at a national level. The agencies that provided accreditation for all 4 types are The Joint Commission (22) and the Healthcare Facilities Accreditation Program (23). Det Norse Veritas (24) and the Center for Improvement in Healthcare Quality (25) provide accreditation for all but thrombectomy-capable centers. Comprehensive and thrombectomy-capable centers have stroke-intervention capability, but only CSCs must consistently have 24/7/365 coverage.
Primary stroke centers must have the capability to administer intravenous tissue plasminogen activator and have specialty physician backup available, but not necessarily immediately. CSCs must have a well-organized stroke team capable of rapid triage, imaging, and stroke intervention available 24/7/365. As of May 2018, there are approximately 212 of these centers in the United States. The number of neurointerventionists varies, from estimates of 800 to 1,100 (Table 2). Whether all these centers have the capability of 24/7/365 coverage is unclear from personal observation. In addition, time from stroke onset to emergency room arrival is variable, depending upon whether the patient is seen at a local hospital and then transferred to the CSC or presents initially to the CSC. Time to intervention is variable, depending on the stroke pathways at each center.
The Joint Commission has published guidance for the certification of neurointerventionists who routinely take calls for endovascular treatment of stroke (Table 3) (26). These requirements are detailed and include training and experience. They cover neurologists, radiologists–neuroradiologists, and neuroendovascular trainees with intracranial and extracranial training and expertise. Specific training mandates have recently been removed, but controversy exists as to final requirements. The guidelines do not cover interventional cardiologists, interventional radiologists, and interventional vascular surgeons (i.e., extracranial interventionists) (Table 2) (26). However, each of these latter groups has training and established expertise in evaluation and endovascular treatment of the extracranial carotid artery.
Central questions relate to whether these extracranial interventionists should or might be active members of the stroke team and how they might best be integrated into it. The answer appears to be local. In large centers with multiple active neurointerventionists who are able to staff the laboratories 24/7/365 and provide very rapid procedural coverage, extracranial interventionists may not be needed for catheter-based stroke intervention. Conversely, as stroke intervention becomes more mainstream, especially when regulators begin to impose the door-to-needle mandates that are now common for STEMI, having additional trained physicians on the interventional stroke team including appropriately trained extracranial interventionists could be a welcome addition.
For this to happen, these interventionists must receive effective training. Experience with carotid artery stenting (CAS) for access is invaluable for stroke intervention. Vetting of the interventionists in the CREST trial (Carotid Revascularization Endarterectomy versus Stenting Trial) required appropriate clinical evaluation of stroke symptoms (and confounding symptoms) and expertise in cervical angiography, embolic protection, and successful stent placement (27). The results of the CREST trial, compared with those of other carotid stenting trials, showed excellent results for CAS among the multiple interventional specialties discussed here. No safety disadvantages were detected for those interventional specialists who did not receive intracranial training. The next steps for these carotid-skilled interventionists are for clinical neurology training and intracranial interventional training. The degree of neurology training ideally should be determined locally. For example, for stroke teams for which a vascular neurologist is always present, complete clinical neurology expertise may not be required of the interventionist, regardless of specialty.
Many of the 5,534 hospitals in the United States have abundant neurological expertise, but do not have an adequate number of neurointerventionists (if any) for 24/7/365 coverage sufficient to avoid physician burnout. Even at the approximately 212 CSCs, such around-the-clock coverage can be a burden because some of these centers have only 1 or 2 stroke-trained neurointerventionists on staff. There are many more cardiology and other interventionists than neurointerventionists, and they are more widely distributed (Table 2) (28). Extracranial interventionists could be trained and could become experienced in intracranial thrombectomy. Training a skilled interventional cardiologist, radiologist, or vascular surgeon in many instances will be different from training a neurointerventionist. And it is important to note that the level of training required varies greatly and depends on the individual’s personal experience and skill set. Interventional cardiologists, interventional radiologists, and interventional vascular surgeons must learn the basics of anatomy, pathophysiology, diagnosis of ELVO, neurotechnology, and methodology if they have interest in joining a stroke intervention team. An advantage of including interventional cardiologists is that they have worked under the time constraints imposed as metrics of care for 24/7/365 delivery of treatment of acute myocardial infarction. In addition, they have extensive STEMI experience opening occluded arteries on awake patients with a rapidly moving target. Interventional cardiologists skilled in STEMI and trained in stroke intervention in areas of the world with no neurointerventionists could have a significant positive impact on AIS.
In rural areas and in small- to medium-sized communities without CSCs or “stroke-ready” teams, skilled extracranial interventionists can play a critically important role in stroke intervention. Today in the United States, patients with AIS and ELVO are often initially evaluated in local hospitals but must be transferred to a CSC for intervention. Unfortunately, the time it takes for initial evaluation in the outside hospital, arranging and transferring the patient to the CSC, re-evaluation, imaging, and preparing for and performing the intervention often result in unacceptable delays to reperfusion.
The interventional workforce needs to be expanded for stroke teams of the future (Central Illustration). The 800 to 1,100 neurointerventionists need help if the availability of urgent mechanical thrombectomy for eligible stroke patients is to be optimized at the 5,000+ hospitals across the United States. Neurointerventional training programs in neuroradiology, neurology, and neurosurgery will not be sufficient to provide the necessary people power. Interventional cardiology, radiology, and vascular surgery can add value for the stroke teams of the future. Combined, these interventional groups number up to nearly 10,000 (Table 2). Many strokes originate in the heart, and cardiologists can add value to the stroke team. Brain imaging continues to advance rapidly, and its role in patient selection continues to expand; interventional radiologists can add skill and experience to image interpretation. Because of STEMI, in-hospital vascular emergencies, ruptured aortic and other vascular aneurysms, interventional cardiologists, radiologists, and vascular surgeons have developed the mindset needed for urgent 24/7/365 intervention. Recognition of the importance of intervention for STEMI has resulted in widespread distribution of interventional cardiologists and catheterization laboratories. The existing paradigm of evaluation in local hospitals, then transferring patients with ELVO to major CSCs in larger cities routinely puts patients outside the ideal time window for optimal revascularization. Emergent, mechanical stroke intervention locally by a stroke team and then transferring the patient with a reperfused brain (if necessary for complex cases) to a brain rehabilitation center could preserve the ideal time window, resulting in better outcomes. Uncomplicated interventions with good results would not require transfer of all patients, given that evaluation and preventive strategies could well be handled by the vascular neurologist with help from the cardiologist and other interventionists.
What is required is a willingness on the part of the neurointerventional community to train interested interventional cardiologists, radiologists, and vascular surgeons in stroke intervention, incorporate these interventionists into stroke teams, and make interdisciplinary collaboration the norm for this compelling public health issue. Another less obvious benefit will be the synergies that spring from interaction and collaboration among different vascular disciplines.
Given the mandate for rapid intervention, new technology may offer help. There is currently significant interest in robotics for catheter-based intervention that hopefully will lead to rapid remote robotic stroke intervention in areas underserved by stroke interventionists. Multidisciplinary collaboration will be essential for this to occur.
The sight of a paralyzed, aphasic, cognitively devastated patient destroyed by AIS returning to a functional human being within minutes after brain reperfusion is achieved is perhaps the most dramatic event any physician will ever see. With everyone working together, the future for stroke patients is indeed bright.
Dr. Hopkins has received grant/research support from Canon Medical Systems Corporation; has financial interests/stock in Boston Scientific, Cerebrotech, Endostream, Endomation, Silk Road, Ostial Corporation, Imperative Care, StimSox, Photolitec, ValenTx, Ellipse, Axtria, NextPlain, and Ocular; and has a board/trustee/officer position in Imperative Care Inc. Dr. Holmes has reported that he has no relationships relevant to the contents of this paper to disclose.
Listen to this manuscript's audio summary by Editor-in-Chief Dr. Valentin Fuster on JACC.org.
- Abbreviations and Acronyms
- acute ischemic stroke
- carotid artery stenting
- comprehensive stroke center
- emergent large-vessel occlusion
- modified Rankin scale
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- Received September 25, 2018.
- Revision received January 4, 2019.
- Accepted January 7, 2019.
- 2019 American College of Cardiology Foundation
- Benjamin E.J.,
- Virani S.S.,
- Callaway C.W.,
- et al.
- Powers W.J.,
- Rabinstein A.A.,
- Ackerson T.,
- et al.
- Ibanez B.,
- James S.,
- Agewall S.,
- et al.
- Madan M.,
- Halvorsen S.,
- DiMario C.,
- et al.
- Wessler B.S.,
- Kent D.M.
- Hopkins L.N.,
- Holmes D.R. Jr..
- Fayad Z.A.,
- Fuster V.,
- Fallon J.T.,
- et al.
- Harteveld A.A.,
- Denswil N.P.,
- Van Hecke W.,
- et al.
- Ciccone A.,
- Valvassori L.,
- Nichelatti M.,
- SYNTHESIS Expansion Investigators
- Qureshi A.I.,
- Abd-Allah F.,
- Aleu A.,
- et al.
- Mokin M.,
- Siddiqui A.H.,
- Levy E.I.,
- Hopkins L.N.
- Nogueira R.G.,
- Jadhav A.P.,
- Haussen D.C.,
- et al.,
- DAWN Trial Investigators
- Campbell B.C.V.,
- Mitchell P.J.,
- Churilov L.,
- et al.,
- EXTEND-IA TNK Investigators
- The Joint Commission
- ↵Healthcare Facilities Accreditation Program (HFAP). Stroke Certification. Available at:. https://hfap.org/CertificationPrograms/StrokeCertification.aspx. 29112018.
- ↵DNV-GL. Stroke Care Certification Programs: Roadmap to Stroke Care Excellence. Available at:. https://www.dnvgl.us/assurance/healthcare/stroke-certs.html. 29112018.
- ↵Center for Improvement in Healthcare Quality. Disease Specific Certification. Available at:. https://www.cihq.org/disease_specific.asp. 29112018.
- The Joint Commission
- Narang A.,
- Sinha S.S.,
- Rajagopalan B.,
- et al.