Thrombolysis Before Thrombectomy for Acute Ischemic Stroke

Aug 26, 2023

Thrombolysis Before Thrombectomy for Acute Ischemic Stroke

Intravenous thrombolysis (IVT) and endovascular thrombectomy (EVT) both improve outcomes for patients with acute ischemic stroke. Jeffrey L. Saver, MD, director of UCLA’s Comprehensive Stroke and Vascular Neurology Program and a JAMA Associate Editor, discusses 2 randomized trials comparing outcomes for stroke patients treated with IVT prior to EVT vs EVT alone.

Read More:
Intravenous Thrombolysis Before Endovascular Thrombectomy for Acute Ischemic Stroke: https://ja.ma/35MH5Bm

Effect of Mechanical Thrombectomy Without vs With Intravenous Thrombolysis in Acute Ischemic Stroke: https://ja.ma/35S4EsO

Effect of Endovascular Treatment With vs Without IV Alteplase on Functional Independence After Acute Ischemic Stroke: https://ja.ma/3bM3KSa

Content

0.373 -> >> Howard Bauchner: Hello, and welcome to Conversations with JAMA.

4.56 -> For all of those, for all of you who are listening or watching today, this is a new experience.

10.24 -> This is the first of what we're hoping will be a continuing series of discussions about important

16.32 -> papers that we're publishing, generally on Tuesday. I'm delighted that my first guest is Jeff

22.64 -> Saver. Jeff is a professor neurology, one of the world's experts in stroke and our associate editor

30.16 -> in neurology. Welcome, Jeff. >> Jeffrey Saver: Thank you, Howard.

33.04 -> It's great to be here. Great to be first. >> Howard Bauchner: So, we'll see how it works.

38.48 -> And I hope all of our listeners can indulge us. And just before we finish, I want to

43.92 -> make sure they know that if they have feedback, they're welcome to provide it to me directly at

48.32 -> howard dot, howard dot bauchner at jama network dot org. So, we're going to discuss two papers.

54.8 -> And Jeff was a handling associate editor for these two papers. The first is entitled the effect of

60.64 -> mechanical thrombectomy without versus intravenous thrombolysis on functional outcome among patients

67.44 -> with acute ischemic stroke, the skip randomized clinical trial by Suzuki et al from Japan. And

75.04 -> the second, effective endovascular treatment alone versus IV alteplase plus endovascular treatment

82.72 -> on functional independence in patients with acute ischemic stroke, the Dev randomized clinical trial

89.04 -> by Z and colleagues from China. These two papers are accompanied by an editorial that Jeff has

94.56 -> written entitled Intravenous Thrombolysis Before an Endovascular Thrombectomy for Acute Ischemic

101.36 -> Stroke. So, Jeff, I've just introduced the two studies. Can you tell me what question were they

107.92 -> trying to answer in these two studies? >> Jeffrey Saver: Both of these trials

112.88 -> were seeking to determine if skipping thrombolysis and going straight to thrombectomy

120.8 -> is as good, non inferior, to using both thrombolysis followed by a thrombectomy.

128.56 -> >> Howard Bauchner: So, Jeff, why is this an important question?

132.48 -> >> Jeffrey Saver: Well, this is a key question for the field, because we have been wondering with the

140.72 -> rise of endovascular thrombectomy if intravenous thrombolysis before a thrombectomy should still

148.48 -> be pursued, or as happened with myocardial infarction 20 years ago, we skipped to go straight

157.6 -> to the mechanical approach. The two treatments in the brain have complementary strengths.

166.72 -> Alteplase, as a clot dissolving medicine, is very good for small and medium clots.

175.2 -> Fairly quick, first of all, had medium clots. Whereas

180.96 -> it does not work very well for the big proximal clots that cause the most severe strokes.

188.32 -> The [inaudible] rate for those was thought to be 10 to 15% with just alteplase. In

195.04 -> contrast, in the vascular thrombectomy using stent retrievers or aspiration devices are very

201.76 -> good for the big proximal clots that can go up and mechanically [inaudible] those very quickly,

207.28 -> but they're not very good for the small distal arteries that can't navigate there. So, because

215.84 -> alteplase is not very good for these big proximal clots, and it has drawbacks of increasing the

223.68 -> bleeding rate, most likely, and causing sometimes clots to fragment and go beyond the reach of

231.44 -> thrombectomy, it might be that it doesn't, it not only doesn't help, it might make patients worse.

237.92 -> On the other hand, it could make patients better by opening the artery before you get to the

243.12 -> cath lab and reducing the total brain ischemia time. Or by making the clot more responsive to

251.2 -> the device retrieval. So, these were physical, physiologically plausible reasons on both sides

258.08 -> of this question, something that could only be settled by randomized trials.

261.76 -> >> Howard Bauchner: Now, Jeff, there's thousands of patients each year with stroke.

267.44 -> Does this study question, is it apropos or does it, is it appropriate for the many

274.4 -> individuals who have stroke? >> Jeffrey Saver: It is. You know, the

278.8 -> thrombectomy pivotal trials all treated patients who were alteplase eligible with alteplase. So,

286.56 -> the current standard of care has been that any patient who is eligible for alteplase within the

292.96 -> first 4 1/2 hours of onset should receive that on the way to the cath lab. And probably about

301.76 -> a third of all ischemic strokes are these big ischemic strokes in the internal carotid artery.

308 -> And then one middle cerebral artery, really M2, where alteplase doesn't work very well,

314 -> recanalization rates, 15, 10 to 15%, up to 30%. So, for that one third of patients, this is a key

331.52 -> decision that every clinician makes. And these are the big bad most severe strokes. Whether

338.56 -> they give alteplase or not, to date, we've been giving alteplase generally, except for some series

344.8 -> that reported in open fashion that you seem, you could skip alteplase and helped to stimulate the

352.56 -> stored of these randomized trials. >> Howard Bauchner: So, let's start

355.76 -> with the study from Japan. It involved 204 patients, 23 hospitals, recruited between

366 -> 2017 and 2019. Follow up through October 2019. This is the SKIP randomized clinical

374.72 -> trial. What were the findings in this trial? >> Jeffrey Saver: Well, first, let me note,

380.24 -> a distinct aspect of the SKIP trial design was the regimen that they used. They used

387.92 -> .6 milligrams of alteplase per kilogram rather than what was common dose in the U.S., which is .9

398 -> milligrams per kilogram. So, they used two thirds of the standard U.S. dose.

402.96 -> That's a common dose used in Asia. There is a grading, a greater leading prone tendency

411.92 -> among Asian patients. And so there's a tradition and good data to suggest that the .6 milligram

422.8 -> dose is about as good as the .9 milligram, and might reduce bleeding rates. But it's a question

430.56 -> whether the results that they see might be reduced to, attributed to this lower dose in part. They

439.2 -> had half their patients receive the alteplase on the way to the cath lab, and half went direct

446.32 -> to the cath lab. The final outcome was functional independence three months after stroke. The degree

454.8 -> of recovery after stroke, a Rankin score of 0 to 2. Now, this was a non inferiority trial, because

464.4 -> if alteplase is not they are than skipping it, there's no reason to give it. It will just

471.52 -> increase patient expenses. So, in a non inferior trial, you have to choose your non inferiority

478.24 -> margin. They chose a margin of 10%. Sorry, a margin roughly equivalent to 10% reduction.

490 -> So, in theory, the treatment could be as much as 10% worse, and still be declared non inferior, 10%

500.96 -> being 1 in 10 fewer patients having functionally independent outcome. I think that most clinicians

509.2 -> would feel that's not the minimally clinically important difference. I think the study suggesting

514.32 -> that for a big bad stroke, it's more like 1 or 2 out of 100 patients, not 10 out of 100 patients.

521.76 -> But for both these studies, when you do a non inferiority study and the treatments

528.16 -> are actually equal, the sample sizes to show indistinguishability are infeasibly large.

536.08 -> So, it has become common to use non inferiority margin that shows if the new treatment retains

546.88 -> substantial proportion of benefit of the old treatment, and then to look at other things

553.6 -> to see if you can really suggest equivalence. They found that the nominal rate of functional

562.64 -> independence of three months was a little higher with skipping tPA than with not skipping tPA.

570.96 -> The difference was not statistically significant to prove non inferiority. So, they failed to prove

577.6 -> non inferiority in this individual trial. But, of course, they didn't exclude non inferiority with

583.36 -> the outcome rate actually being nominally higher for the SKIP procedure. So,

591.2 -> they provided convergent evidence with the dev trial, which we'll be speaking about in a moment,

598.32 -> and with the one prior trial in this area, the DIRECT MT trial, which was published a few months

605.84 -> ago, and had demonstrated non inferiority. >> Howard Bauchner: So, now let's go to the other

610.08 -> trial, which is from China by Z. Now, in this, in this trial, there were 234 patients, interestingly

617.76 -> enough, a very similar number. Also randomized, primary endpoint proportion of patients achieving

623.92 -> functional independence at 90 days. Patients randomized to endovascular thrombectomy alone,

631.2 -> or to combined IV thrombolysis and endovascular thrombectomy. What were

636 -> the findings in this study? >> Jeffrey Saver:

639.2 -> This was a study that used the U.S. standard dose of alteplase .9 milligrams per kilogram. And,

647.76 -> again, the final outcome, primary endpoint was degree of disability on the modified Rankin scale.

654.96 -> And, again, the pure direct thrombectomy group nominally had more good outcomes than the combined

665.76 -> thrombolysis and thrombectomy group. And this time, the difference was enough to reach

670.4 -> statistical significance for non inferiority. So, this was a positive non inferiority

675.52 -> trial. Now, there are two interesting aspects of the implementation of these trials that

684.88 -> allow us to think that they pro slightly different questions. The dev trial used a higher dose of

692.4 -> alteplase. And it also had a longer time period between the start of the alteplase and the

700.56 -> arterial puncture to start the thrombectomy. About 40 minutes. Sometime before that higher dose to

708.08 -> work. But they only saw 7 1/2 or so percent of patients who reopened before they got to

717.2 -> the cath lab. The SKIP tPA trials used the lower dose of alteplase. And a very short time period,

725.68 -> only about 8 minutes on average, [inaudible] was started before the procedure started. So,

733.92 -> the dev trial tested the idea more fully that if you give substantial dose of alteplase

742.72 -> well ahead of time, it will reopen enough arteries to make a big difference. And they did not find

749.12 -> that. And the SKIP trial tested the idea that if you give a safer, lower dose of alteplase,

759.12 -> it not so much before you get to the cath lab, but afterwards to clean up small distal fragments that

767.52 -> the retrievers often leave behind, having the alteplase on board could help with that,

771.68 -> would that improve outcome. And with both of those implementations,

778.32 -> the direct thrombectomy group did better. >> Howard Bauchner: Jeff, you're, you're one of

784.08 -> the world's experts on stroke. You've, you've done clinical trials, published in all of the

789.68 -> major journals. You've written extensively about it. You really champion these two papers. You

797.44 -> thought these two research studies, along with the one that had previously been published,

803.44 -> really could begin to impact the field. How do you think about what these two studies mean,

813.28 -> first for the general physician, and then we'll move onto the specialist like you. What do you

818.56 -> think they mean for the general physician? >> Jeffrey Saver: Well, these studies

825.2 -> create a real challenge for clinicians, both the generalists and the specialist physician.

834.96 -> For the generalist physician, I think the message is, as always, make sure patients with stroke

842.56 -> warning signs get to the hospital as quickly as possible. Getting the artery open effectively and

849.2 -> really is the best guarantee of a good outcome for the patient. And whether we take a two treatment

856.16 -> route, thrombolysis plus thrombectomy, or a single treatment route, thrombectomy allowing, that

862.96 -> doesn't affect the need to get to the hospital right away. For the specialist physician, this is

869.28 -> a really challenging study that put into practice, a set of trials to put into practice, because

878.24 -> you have to be aware that there are some patients who you should still give alteplase to,

885.04 -> even with these results. And those are patients who present to an outside hospital that only does

891.28 -> thrombolysis first and is going to get a drip and ship approach. And any time you get thrombectomy,

900 -> it's going to be very prolonged. So, they should get alteplase at that first hospital.

904.8 -> The second are patients in whom you don't think, you're not sure you're going to be able to reach

910.24 -> the artery in the cath lab. They might have a carotid occlusion in the neck on the way up,

915.44 -> or excessively torturous arterial anatomy in the chest. And if you skip thrombolysis and go

924.88 -> to the cath lab and don't get there, the time window for giving thrombolysis might elapse,

930 -> and they might not get any therapy. So, you have to be confident that you are going to be

935.52 -> able to get to that clot very quickly. In those patients in whom you can get to the clot quickly,

944.16 -> these trials suggest that it is reasonable to skip thrombolytic therapy. And since all of those

952.48 -> trials have nominally better outcomes for patients in the SKIP arm, it's a little bit of suggestion

961.44 -> that it might be a superiority strategy. There are three more trials in the wild about to come in

968.96 -> testing this same question. And there's already a plan for a pooled individual patient data analysis

976.32 -> of the six trials. And they might have enough power to tell us if it is a superior strategy

982.88 -> for those patients in whom you're sure you can treat. But it's going to be a headache

986.88 -> for me and for every clinician having to make this decision very quickly when patients arrive.

994.88 -> Do I give thrombolysis or do I skip? Am I sure this is a treatable patient or not?

1001.84 -> It's a good thing for the field, but a challenge for clinicians.

1005.04 -> >> Howard Bauchner: Now, you deal with these types of patients all the time.

1009.2 -> You are at UCLA. Does giving thrombolysis often slow down the movement of the patient to the cath

1015.68 -> lab for the potential thrombectomy? Is there a way in which it may delay getting the person to the

1022.96 -> individual who can do the thrombectomy? >> Jeffrey Saver: It probably slows it down

1026.56 -> a little bit. You have to consent the patient for the thrombolysis, get the

1034.64 -> drug infusion started. We do parallel processing in stroke codes in the Emergency Department.

1043.04 -> And several things are happening simultaneously. So, it probably doesn't slow it down very much.

1048.24 -> But it probably slows it down by a few minutes, which could make a difference.

1052 -> I will mention that in mobile stroke units, in which there's a CAT scan in the ambulance

1060.08 -> and you're able to give tPA, often in the first 60 minutes of onset, right at the scene,

1066.64 -> that's another setting in which we probably don't want to skip thrombolysis when you can get the

1072.16 -> drug started so early. And there's going to be a long time between the drug and the cath lab.

1078.4 -> >> Howard Bauchner: I was going to ask you about mobile stroke units. We had published a paper,

1084.32 -> again, that you championed from Germany a few years ago. I think that was the first

1089.52 -> major report about putting CAT scans in ambulances. You and I know that our friends

1096.08 -> to the north in Calgary have an enormously well organized mobile stroke unit that covers many,

1103.68 -> many square miles in different areas of Canada. We transform people's care if they are appropriately

1113.36 -> treated. It's extraordinary. I've had friends and colleagues who years ago would have walked out of

1119.68 -> a hospital, may not have walked out of a hospital. Now people do walk out of the hospital. What's the

1125.2 -> next great leap that we have to take both in rural and urban America to make sure that excellent

1132.96 -> appropriate stroke care is available? >> Jeffrey Saver: Sure.

1137.12 -> Well, first, let me say, we still would benefit from advances in thrombolysis.

1143.52 -> >> Howard Bauchner: Okay. >> Jeffrey Saver: If,

1144.48 -> instead of a 30% reperfusion rate, we had a 60, 70% reperfusion rate and less bleeding,

1150.96 -> then it would make sense to give thrombolysis first. And this work being done with newer

1155.76 -> generation thrombolytics connect to plays, or with adding G2P3 agents, or DTIs to thrombolysis, to

1165.84 -> plasminogen activators. That has some promise for improving lytic strategy. That's one way

1171.76 -> to go for it. We also still are working on the tantalizing idea of neural protection, giving

1178.48 -> drugs or treatments that allow the brain to tolerate low blood flow for longer so that

1184.48 -> there's still more salvage of brain when patients arrive at the hospital, treatments that could be,

1189.84 -> come in the ambulance, that would be safe, and hemorrhage and ischemic stroke patients,

1194.48 -> so you wouldn't need imaging beforehand. We already have built out, in terms of care delivery,

1202.72 -> a very good system for reaching rural hospitals with telestroke telemedicine, so that almost every

1213.84 -> hospital in the country can have a neurologist available at least by telemedicine within 15

1219.44 -> minutes of patient arrival and get the lytic started. But we are still working on optimizing

1227.68 -> the patient flow from hospitals where they only offer lytic therapy to the thrombectomy

1237.04 -> stroke centers that can offer catheter thrombectomy. And pre hospital systems are

1245.04 -> moving toward two tier routing, where if you have a very severe stroke, likely proximal clot,

1255.36 -> and the comprehensive stroke center, the thrombectomy stroke center is only a few minutes

1259.2 -> further away than the primary stroke center, then it's worth skipping the primary stroke center and

1265.04 -> going there. And this study, these two studies provide even more fuel to support that idea. But

1272.8 -> getting patients from rural areas by helicopter, by [inaudible] aircraft, requires a lot more

1282.4 -> mobility and organizational arrangements. And those are still getting built out.

1287.44 -> >> Howard Bauchner: Last question, Jeff. What's the magic time? It seems to get a little longer.

1294.72 -> But what's the magic time if a clinician is thinking about a patient? How quickly do they want

1300.48 -> that patient seen? What's the outside limits? >> Jeffrey Saver: The magic time is one minute.

1308.4 -> Every patient, every minute that goes by, more brain will likely be lost.

1313.2 -> >> Howard Bauchner: Okay. >> Jeffrey Saver: So, you never want to wait.

1316.16 -> For what's the current time for which there's some evidence of benefit, it's 24 hours for

1323.04 -> endovascular thrombectomy. For the third to a half of patients who are slow progressors, and we can

1330.72 -> tell that with advanced CT and MR imaging who's a slow progressor, and still has salvageable tissue

1337.92 -> up to 24 hours, and who's a fast progressor and already completed their stroke, the slow

1343.76 -> progressors that can be benefit up to 24 hours from onset probably for some even beyond 24 hours.

1350.96 -> But you can't tell who's a slow progressor or a fast progressor just by looking at them. So, time,

1357.6 -> time is [inaudible]. And patients should be moved through the system as quickly as possible.

1362.4 -> >> Howard Bauchner: This is Howard Bauchner, editor in chief of JAMA. This has been

1367.12 -> Conversations with JAMA. And I've been talking with Jeff Saver. Jeff's a professor in neurology,

1372.72 -> University of California, Los Angeles, a senior associate vice chair of neurology,

1377.36 -> and director of the Clinical Neurotherapeutics Research Center in the Department of Neurology.

1383.36 -> We've been discussing two research papers, the effect of mechanical thrombectomy

1387.44 -> without versus with IV thrombolysis on functional outcome among patients with acute ischemic stroke,

1393.36 -> the SKIP trial, from Japan. And the second paper, the effect of endovascular treatment alone

1399.6 -> versus IV alteplase plus endovascular treatment on functional independence in patients with acute

1405.84 -> ischemic stroke, the dev randomized clinical trial from China. These two papers are accompanied by an

1412.32 -> editorial by Jeff and his colleague entitled IV thrombolysis before endovascular thrombectomy for

1419.12 -> acute ischemic stroke. Thanks so much, Jeff. >> Jeffrey Saver: Thank you, Howard.

Source: https://www.youtube.com/watch?v=759nshxu--U