Conference Coverage

Broadly implementing stroke embolectomy faces hurdles




NASHVILLE, TENN. – Results from three randomized controlled trials presented at the International Stroke Conference , plus the outcomes from a fourth trial first reported last fall, immediately established embolectomy as standard-of-care treatment for selected patients with acute ischemic stroke.

Stroke experts interviewed during the conference, however, said that making embolectomy routinely available to most U.S. stroke patients who would be candidates for the intervention will take months, if not years.

They envision challenges involving the availability of trained interventionalists, triage of patients to the right centers, and reimbursement issues as some of the obstacles to be dealt with before endovascular embolectomy aimed at removing intracerebral-artery occlusions in acute ischemic stroke patients becomes uniformly available.

Dr. Pooja Khatri

Dr. Pooja Khatri

Yet another challenge will arise when stroke-treatment groups that did not participate in the trials strive to replicate the success their colleagues reported by implementing the highly streamlined systems that were used in the trials for identifying appropriate stroke patients and for delivering treatment. Those systems were cited as an important reason why those studies succeeded in producing positive outcomes when similar embolectomy trials without the same efficiencies reported just a year or two ago failed to show benefit.

“The evidence makes it standard of care, but the challenge is that our systems are not set up. This is the big thing we will all go home to work on,” said Dr. Pooja Khatri , professor of neurology and director of acute stroke at the University of Cincinnati.

“You talk to everyone at this meeting, and what they want to go home and figure out is how can we deliver this care. It’s really challenging, at a myriad of levels,” said Dr. Colin P. Derdeyn , professor of neurology and director of the Center for Stroke and Cerebrovascular Disease at Washington University in St. Louis.

Growing endovascular availability

Arguably the most critical issue in rolling out endovascular stroke interventions more broadly is scaling up the number of centers that have the staff and systems in place to perform them. Clearly, the scope of providers able to deliver this treatment currently falls substantially short of what will be needed. “It’s kind of daunting to think about the [workforce] needs,” Dr. Khatri said in a talk at the conference, which was sponsored by the American Heart Association.

“In the United States, we’ve been building out a two-tier system, with comprehensive stroke centers capable of delivering this [endovascular embolectomy] treatment” and primary stroke centers capable of administering intravenous treatment with tissue plasminogen activator (TPA), the first treatment that patients eligible for embolectomy should receive, said Dr. Jeffrey L. Saver, professor of neurology and director of the stroke center at the University of California, Los Angeles, and lead investigator for one of the new embolectomy studies.

Dr. Opeolu M. Adeoye

Dr. Opeolu M. Adeoye

“Work groups have suggested about 60,000 U.S. stroke patients could potentially be treated with endovascular therapy, and we’d need about 300 comprehensive stroke centers to do this.” Dr. Saver estimated the current total of U.S. comprehensive stroke centers to be 75, a number that several others at the meeting pegged as more like 80, and they also noted that some centers are endovascular ready but have not received official comprehensive stroke center certification from the Joint Commission.

The extent to which availability of U.S. embolectomy remained limited through most of 2013 was apparent in data reported at the conference by Dr. Opeolu M. Adeoye , an emergency medicine physician and medical director of the telestroke program of the University of Cincinnati. During fiscal year 2013 (Oct. 2012 to Sept. 2013), 386,144 Medicare patients either older than 65 years or totally disabled had a primary hospital discharge diagnosis of stroke; of those, 5.1% had received thrombolytic therapy with intravenous TPA and 0.8% had undergone embolectomy. In a second analysis, he looked at stroke discharges and reperfusion treatments used in the 214 U.S. acute-care hospitals currently enrolled in StrokeNet, a program begun in 2013 by the National Institute of Neurological Disorders and Stroke to organize U.S. centers interested in participating in stroke trials.

During the same period, the 214 StrokeNet hospitals discharged 44,282 Medicare eligible patients who met the same age or disability criteria, with a TPA-treatment rate of 7.9% and an endovascular treatment rate of 2.2%. Although the StrokeNet hospitals treated roughly 11% of U.S. stroke patients in the specified demographic, they administered about 20% of the reperfusion treatment, Dr. Adeoye reported. He also highlighted that the 7.9% rate of TPA treatment among the StrokeNet hospitals correlated well with prior estimates that 6%-11% of stroke patients fulfill existing criteria for TPA treatment

A wide disparity existed in the rate of reperfusion use among StrokeNet hospitals. Sixty-seven hospitals, 31% of the StrokeNet group, treated at least 20 stroke patients with TPA during the study period, while 100 (47%) of the StrokeNet hospitals treated fewer than 10 acute stroke patients. The rate of those doing embolectomies was substantially lower, with 10 hospitals (5%) doing at least 20 endovascular procedures while 90% did fewer than 10.

Although Dr. Adeoye expressed confidence that the number of U.S. centers doing embolectomy cases will “change rapidly” following the new reports documenting the efficacy of the approach, he also acknowledged the challenges of growing the number of high-volume endovascular centers.

Centers that have been equivocal about embolectomy will now start doing it in a more concerted way, he predicted, but if cases get spread out and some sites do only a few patients a year, the quality of the procedures may suffer. “The more cases a site does, the better,” he noted, adding that regions that funnel all their stroke patients to a single endovascular site “do really well.”

“Right now, many hospitals want to do everything to get [fully] reimbursed and not send their patients down the line. There is a financial incentive to build up the stroke service at every hospital,” Dr. Derdeyn noted.

Another aspect to sorting out which centers start offering endovascular treatment will be the need to locate them in a rational way, as happened with trauma centers. Until now, placement of comprehensive stroke centers often depended on hospitals developing the capability as a marketing tool, noted Dr. Larry B. Goldstein , professor of neurology and director of the stroke center at Duke University in Durham, N.C. A hospital might achieve comprehensive stroke center certification, so a second center a few blocks away then follows suit seemingly to keep pace in a public-relations battle for cachet. The result has been an irrational clustering of centers with endovascular capability. He cited the situation in Cleveland, where comprehensive centers run by the Cleveland Clinic and Case Western stand a few dozen feet apart.

Challenges for triage

An analysis published last year by Dr. Adeoye and his colleagues showed that 56% of the U.S. population lived within a 60-minute drive of a hospital able to administer endovascular stroke treatment; by air, 85% had that access ( Stroke 2014;45:3019-24 ). For TPA, 81% of people lived within a 60-minute drive of a center able to administer intravenous lytic treatment and 97% could reach these hospitals within an hour by air. While those numbers sound promising, though, fulfilling that potential depends on getting the right patients to the right hospitals.

Patient triage is perhaps the most vexing issue created by embolectomy’s success. For at least the short term, a limited number of centers will have the staffing and capacity to deliver endovascular embolectomy on a 24/7 basis to acute ischemic stroke patients who have a proximal blockage in a large cerebral artery. The relatively small number of sites able to offer embolectomy, and the much larger number of sites able to administer thrombolytic therapy with TPA, set the stage for some possible tension, or at least confusion, within communities over where an ambulance should bring an acute ischemic stroke patient.

Dr. Colin P. Derdeyn

Dr. Colin P. Derdeyn

“In some places they have trained the EMS [emergency medical services] to recognize severe strokes that are likely to benefit [from embolectomy], and they take those patients to places with endovascular capability. But there are some states with laws against doing this. There are major issues when EMS bypasses hospitals,” Dr. Derdeyn noted. “That’s the million-dollar question: How do you identify the stroke patients [with severe strokes who would benefit from embolectomy] and get them to where they need to go.” Like Dr. Adeoye, Dr. Derdeyn believes that endovascular treatment for stroke needs to be centralized at a relatively small number of high-volume centers.

“You can imagine that the fastest way to get stroke patients treated is to have them all go to one place, but that is much easier said than done,” Dr. Khatri said.

“Stent retrievers get cerebral arteries open, but that is not the biggest challenge. For the short term, the key issue is to get the correct patients to the correct hospitals where they can be treated by the correct team,” said Dr. Mayank Goyal , professor of diagnostic imaging at the University of Calgary (Alta.) and lead investigator for two of the three trials presented at the conference.

“You need a neurologist capable of deciding whether it really is a stroke, and pretty high-level imaging to identify the large-vessel occlusions that will benefit. Acquiring a CT angiography (CTA) image of the brain is a push-button process, but figuring out what the CTA says is not push button, especially the more sophisticated perfusion CT imaging to assess collateral circulation. I don’t see this capability happening in every catheterization laboratory,” Dr. Derdeyn said in an interview.

Another issue is volume. “Telemedicine may allow broader use of [more sophisticated] imaging, but if a place is only doing 20 endovascular procedures a year, they won’t have the best outcomes. Most small hospitals that today give patients TPA see 20 cases or fewer a year, and perhaps 5 patients will have a large-vessel occlusion,” Dr. Derdeyn said.

Before the new reports documenting the safety and efficacy of endovascular treatment, “we did not have the justification to bypass primary stroke centers and take patients directly to comprehensive stroke centers,” Dr. Khatri said. Now, “there is clear evidence that patients with severe strokes should not go to the nearby primary stroke center” but instead head directly for the centers capable of performing embolectomy. But Dr. Khatri also acknowledged that a complex calculation is needed to balance the trade off: Is it better to take a stroke patient more quickly to a nearby hospital that can only start TPA and perhaps later forward the patient to an embolectomy-ready hospital, or to transport the patient somewhat further to a facility able to deliver both TPA and embolectomy?

Dr. Khatri said that, in the Cincinnati area, “we have scheduled a retreat for March to start to plan how this will happen.” Her region includes just one comprehensive stroke center that already performs endovascular treatments for stroke, at the University of Cincinnati, which sits amid 16 other hospitals that perform acute stroke care and can administer TPA. “Ambulance-based triage will be a big issue,” she predicted.

Other aspects of improved triage will be training ambulance personal to better identify the more severe stroke patients who will most likely need endovascular treatment and improving communication between ambulance crews and receiving hospitals to further speed up a process that depends on quick treatment to get the best outcomes.

The ideal is “having paramedics call and tell us what is happening [in their ambulance] and give us as much information as possible so we can start planning for the patient’s arrival. Most hospitals don’t do this now; relatively few have their system well organized,” Dr. Goyal said in an interview. A finely orchestrated emergency transport system and hospital-based stroke team was part of the program developed at the University of Calgary by Dr. Goyal and his associates and which they credited with contributing to the successful embolectomy trial they led, called ESCAPE (Endovascular Treatment for Small Core and Anterior Circulation Proximal Occlusion with Emphasis on Minimizing CT to Recanalization Times)( N. Engl. J. Med. 2015 Feb. 11 [doi:10.1056/NEJMoa1414905] ). Dr. Goyal said that he is now visiting hospitals around the world to assist them in setting up stroke-response systems that mimic what was successful in Calgary and the other centers that participated in ESCAPE.

Improving triage with better screening

A key to improved ambulance triage will be identifying a simple, evidence-based method for assessing stroke severity that ambulance personnel can use to determine what sort of care a patient needs and where the patient needs to go to. Although a couple of U.S. sites have begun pilot studies of field-based CT units that allow stroke patients to undergo imaging-based assessment in the field, clinical evaluation remains the main tool used in the ambulance.

One possible tool is the Los Angeles Motor Scale (LAMS), a stroke-assessment scoring system developed by Dr. Saver and his associates for ambulance use about a decade ago ( Prehosp. Emerg. Care 2004;8:46-50 ). “A LAMS score of 4 or 5 [on a scale of 0-5] is a good starting point, and with time it might improve,” Dr. Goyal said.

The National Institutes of Health Stroke Scale (NIHSS) is a clinical assessment tool not designed for prehospital use, but a new analysis reported at the meeting showed value in using the NIHSS to identify stroke patients who are good candidates for endovascular treatment, further suggesting that a simple screening tool could potentially work in the ambulance to identify patients who probably need embolectomy.

The new analysis combined data from two randomized trials: The IMS (Interventional Management of Stroke) III trial, the results of which, published in early 2013, showed no incremental benefit of endovascular therapy plus TPA over TPA alone for patients with acute ischemic stroke ( N. Engl. J. Med. 2013;368:893-903 ); and the MR CLEAN (Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands) trial, the results of which, published in January, showed a significant incremental benefit from endovascular treatment – it was the first of the four studies recently reported to show this benefit ( N. Engl. J. Med. 2015;372:11-20 ).

The combined data included all patients from both studies with a NIHSS score of at least 20, indicating a very severe stroke. This produced a total of 342 patients, of whom 191 received intravenous TPA plus endovascular treatment and 152 received only TPA. After 90 days, 24% of the patients treated with endovascular treatment and 14% of those treated only with TPA had a modified Rankin Scale score of 0-2, indicating no or limited disability, Dr. Joseph P. Broderick reported at the conference. After adjustments for age and other potential confounders, treatment with endovascular therapy produced a statistically significant 85% improvement in patients achieving an acceptable modified Rankin Scale score at 90 days, said Dr. Broderick, professor of neurology and director of the neuroscience institute at the University of Cincinnati.

“The NIHSS score is a surrogate for clot size. It is an imperfect measure, especially at lower levels, but when the score is 20 or higher it means the patient has a big clot” that will likely not fully respond to TPA but potentially will respond to embolectomy, Dr. Broderick said in an interview. “A patient with a NIHSS score of 20 or higher has about a 95% risk of having an ongoing major artery occlusion despite TPA treatment.”

“The challenge is that we don’t have a fully validated [prehospital] scoring system. Several groups are trying to create one; in the meantime we may come up with certain clinical thresholds” that can reliably guide ambulance crews on the best place to take each stroke patient, Dr. Khatri said.

Dr. Khatri has received research support from Penumbra and Genentech. Dr. Derdeyn has received honoraria from Penumbra and holds equity in Pulse Therapeutics. Dr. Saver has been a consultant to and received research support from Covidien. Dr. Adeoye has been a speaker for Genentech. Dr. Goldstein had no disclosures. Dr. Goyal has been a consultant to and received research support from Covidien and holds a patent on using CT angiography to diagnose stroke. Dr. Broderick has received research support from Genentech.
On Twitter @mitchelzoler

Next Article:

VIDEO: Challenges abound in rolling out stroke embolectomy

Related Articles