Great talk. thanks. I may have missed it, but was the MR rescue trial discussed? http://www.nejm.org/doi/full/10.1056/NEJMoa1212793 This was published in the same issue of NEJM and at the same stroke conference which specifically addressed the issue of not only interventional but also perfusion scanning/CT angio which did not show benefit of performing either.
Where the rubber meets the road (in conjunction with your talk on EMS later) is when should the patient coming via EMS not even go to a hospital doesn't have interventional? 0-4.5hrs, no study has show benefit of inteventional over tPA, so seems like those folks should go to any hospital that can give tPA regardless of whether they can do interventional. And even the 4.5-8hr or further, the MR rescue trial also showed no benefit. But should we do it anyway b/c we're not going to tPA them?
I agree that we need better trials, but are we going to get one? What do you think of the editorial (http://www.nejm.org/doi/full/10.1056/NEJMe1215730 hopefully that's the right one) where the article points out that though there's no randomized trial showing benefit, medicare has already decided to reimburse this in the thousands, therefore, it's going to be difficult to see that trial due to equipoise?
As always I Iove EMRAP and really like MIke Weinstock but I did not like this section. When pro-intervention/pro-tPA MDs talk about stroke studies they have a bizarre passion that clouds their thinking.
When discussing a study that failed to show what they hope it will show, they point out all the problems with the study, talk about nonsignificant relative benefits in subgroup analysis and talk about future studies.
However, when discussing the ONLY two DBRCT studies that have EVER shown benefit (NINDS, ECASS III) they suddenly are not critical in any way. 2 studies out of over a dozen well done DBRCT have shown benefit and their assessment is (paraphrased) : "we have to keep selecting patients carefully to get it right" meaning "manipulate study design to stack the deck in our favour." There is no discussion of the serious flaws with the only two positive studies and absolutely no mention of the preponderance of negative literature. There is also the prerequisite discussion of registries to support their theories, a trick used anytime you can't find actual evidence to support your argument.
One of the discussants incorrectly stated that NINDS showed more benefit in 0-90 minutes (true) and that ECASS 3 also showed that time mattered. That is in fact false as there was no difference between 3 and 4.5 hours in benefit in ECASS 3.
The only thing I heard during that whole segment that I agreed with was the notion that patients should be enrolled in studies of interventional therapy/ tPA if the issue is to be settled. Until then, there is no good evidence that interventional therapies help stroke patients and on balance decent evidence of harm.
WEINSTOCK: Thanks for these insightful comments - these are good points - studies are only good when they give you the answer you are looking for... of course! I tried to hold Budzik's feet to the fire with his "we need to select patients better" but I will say that he has a decent point that there are new devices they are using that were not used during this study. In a age of rapidly changing technologies, we need to find a way to do studies more rapidly or this will always be a criticism. I am going to forward these comments to Budzik and Eubank for their responses!
Listening to this discussion raised some concerns. Dr. Budzik questions his own study; the primary end point, timing, and patient selection, then offers the "I pulled a clot out and the patient is now better so I don't a trial to tell me it works" rationalization as a substitute for science. These arguments might have convinced some people prior to SYNTHESIS, MR RESCUE and IMS III, but how many negative RCTs need to be published in the NEJM in order to put this controversy to rest? MR RESCUE even addresses the issue of patient selection, demonstrating that patients with a favorable penumbra pattern on neuroimaging still fail to benefit from endovascular therapy.
At a minimum moving forward, if new technologies are in development, they need to be subjected to rigorous study through RCTs. This would avoid the "Mikey likes it" approach that led to the widespread use of retrieval devices without proof of benefit. Current devices are now shown in IMS III and MR RESCUE to be equivalent to IV thrombolytics at best, with SYNTHESIS demonstrating a trend towards worse outcomes in the endovascular therapy group.
I have a few issues with this. First I am tired of believers treating ED physicians as though we are somehow mentally defective. I know how to perform a neurologic exam. I know how to calculate an NIH stoke number. I am very familar with giving TPA. Many of us grew up in the TPA for STEMI era. ED physicians have given this drug many many times. Strangely without resistance. WHY? It was CLEAR!!! that TPA was effective in STEMI. The utility of TPA for stroke is "less clear?" Calling TPA "the GOLD STANDARD" in stroke treatment is just WRONG! did I miss a study? does TPA for Stroke reduce mortality? Last I read TPA in stroke improved function as defined by a scale that was never independantly verified....at 90 days. At one point he compared interventional stroke therapy to coronary stenting. As I recall every percutaneous intervention study showed improved survival as compared to Thrombolytic therapy. I do agree that patient selection continues to be a problem. i agree that there is a patient population that might benefit. However, they have yet to be identified.
RESPONSE FROM GEOFF EUBANK (neurologist on the program):
The purpose of this discussion was not to revisit IV tPA for ischemic stroke(debated in a prior issue of EMRAP), though I/we recognize that it is still controversial for some in the ED world. Recent policy statements from the American College of Emergency Physicians (with the Academy of Neurology) do further support the use of IV tPA for use in acute ischemic stroke. That being said, let me address a few of the comments made. TPA has not been shown to reduce stroke mortality. Agreed. We wish it did improve mortality, but frankly our disease happens to be (apparently) much trickier to treat and a fair bit more resistant and risky than treatment for STEMI's. The evaluation of ischemic stroke also remains often a bit more nuanced than STEMI, it is for less certain variables at times, making treatment decisions challenging. Stroke is, however, THE leading cause of disability in adults and the NINDS was specifically intended and powered to look at that. It did, in fact, show that disability, as measured by the modified Rankin Scale, was reduced in the tPA treated group. Not by data mining or twisting of the data, but by preplanned objectives and analysis. I also agree that numerous studies prior to the NINDS and ECASS trials failed to show benefit.The lessons learned from those trials were applied to subsequent trials. Shorter time window. Avoidance of anti coagulated patients. Tight control of BP. Exclusion of visible infarcts on CT. Timing is critical as evidenced from the trials and also by its criticisms. Much of the benefit was derived by the 0-90 minute patient group. The 90-180 patient group showed less benefit. My comments about ECASS 3 was that the longer it takes for patients to receive tPA, the less benefit we see, though this is an inferred conclusion from the trials, admittedly. Which brings us to the interventional aspect of acute stroke treatment. The rationale for this therapy derives from the fact that patients with large vessel occlusion have a natural history of high disability and mortality, even with those that might receive IV tPA. This known problem leads to hypothesis that a more direct recanalization would provide more or (for the cynical) some benefit. Early intraarterial thrombolysis (PROACT) did show benefit, though with enough safety concerns to preclude FDA approval. Interventional devices were subsequently designed to hopefully provide a safer alternative than chemical thrombolysis. I have been witness to numerous cases where patients have had truly miraculous recoveries, very shortly after a clot is removed. It is very hard to not offer that after you have seen it numerous times. Yet, individual cases cannot provide proof of benefit, just the impetus to propel forward. This is where I may deviate from my interventional colleagues. They are likely correct in their assessment of benefit. But this is precisely where randomized clinical trials need to answer the question of actual benefit. The challenges with new technology does somewhat necessarily lead to technical feasibility issues first (experience, device mechanics, etc.) before they are ready for RCT's. It's hard to withhold a therapy you wholeheartedly believe works. Ideally, physicians would not have to make this decision prior to proof of benefit. Once a product "matures" in centers with the dedication and expertise to study, it should have the spotlight of RCT's shown on it to see if it truly benefit patients. We haven't demonstrated that yet. It should be pointed out that the therapy was not medically harmful either. Learning from failed trials and redesigning trials them to learn is not stacking the deck, its trying to separate the wheat from the chaff. Some patients with certain anatomic or medical characteristics may be excluded, leaving a group that could, in fact, benefit. Moreover, the devices used in the IMS III trial and others have been shown to be far inferior to the latest generation devices (in fact, many patients simply had intraaterial thrombolysis). Additionally, sites that routinely used general anesthesia, may have skewed results in a negative fashion (hemodynamic issues with general anesthesia could have adversely impacted the interventional group). Time to intervention is another area that is improving and, similar to IV tPA, seems to influence outcomes. I truly believe these factors are the difference between a positive and a negative trial. I also believe that if this is true, it will withstand the scrutiny of a RCT. We should insist that they do. Both as an adjunct to IV tPA in the <4.5 hour group and also as a stand alone therapy in the >4.5 hour group. What we have learned from these trials should and has given us (the medical community) the appropriate pause to truly assess whether these therapies are beneficial. I expect our ED colleagues will support these necessary research endeavors, whilst holding our feet to the fire until if/when they do.
Geoffrey Eubank, MD Co-director Ohio Health Riverside Methodist Hospital Stroke Program
Thanks for the informative and insightful discussions that you have provided for the E M R A P listeners. I, in fact, felt that your approach was fairly neutral and as an Emergency Doctor feel that the stroke literature has been misinterpretted to a degree by messrs Hoffman et al.
While Thrombolysis is associated with a variable increase risk of post therapy bleeding there is fairly good evidence for benefit in terms of Morbidity. This has been established in multiple studies, although granted the variable ways of measuring morbidity (Rankin Scoring) and changing of protocols during studies (ECASS III) as well as a lack of homogenity leaves some of the studies open to critique.
After hearing this segment I was more confused about imaging so I took the liberty of reviewing this here: http://emergencypedia.com/2013/05/22/imaging-in-acute-stroke-and-tia/ I would welcome your comments on this...
Thank you for your comments. Thus far, however, I haven't read anyone who has really given a reply regarding the stroke triage question for EMS, should we at this point be diverting patients to stroke centers that have interventional if they can make it in 8hrs? If they are within the 4.5hr window, should they then be transferred to facilities that can provide interventional? My understanding is that medicare is reimbursing for this e.g. Merci reimburses 23k according to: http://www.bonsecours.com/assets/CME/Jensen2.pdf vs $4,900-6,400 for tPA. This makes it difficult to know what's going on.
I'm confused. A few times in the discussion I heard the phrase "so I've given the tPA, they're not getting better, what do I do?"
Am I missing something? I thought all the benefit shown in NINDS was 6 months out after the event, so I wouldn't expect to see a benefit (unless I'm treating a TIA). I am one of the skeptics, but still offer it to patients. I really did like the "script" regarding informed consent.
I don't like the tone in the written summary that IV TPA is accepted as beneficial: "If you compare the chances of the patient improving compared to the chances of complications such as bleeding or making things worse, the evidence favors treatment for the stroke."
Echoing above statements, there have been 12 RCTs looking at IV TPA. 2 showed benefit, 6 showed no benefit, and 4 showed harm. Some argue that it's the time window, i.e. it works if given early. However, the biggest trial, IST-3, showed benefit 0-3h, harm 3-4.5hr, and benefit 4.5-6hrs. This is inconsistent with our theory of why TPA works-biological implausibility.
Another argument, mentioned above, is that the harm trials were the early ones, that we changed protocols, and things work now. But the biggest most recent trial, IST3, is one that showed no benefit.
An excellent, and brief, review from David Newman and his crew is available on the NNT: http://www.thennt.com/nnt/thrombolytics-for-stroke/
By the way, excellent point Jim M! Some people see a patient getting better in front of them while TPA is being administered and think it's the TPA; it's not. Like Jim said, the reported benefit is change in Rankin scale months after the stroke.
To join the conversation, you need to subscribe.
Sign up today for full access to all episodes and to join the conversation.
Episode 140Full episode audio for MD edition238:22 min - 100 MB - M4AResumen Mayo 2013 en españolEspañol80:43 min - 28 MB - MP3EM:RAP 2013 May MP381 MB - ZIPEM:RAP May 2013 Written Summmary857 KB - PDFEM:RAP May 2013 Board Review Questions635 KB - PDFEM:RAP May 2013 Board Review Questions:Answer Sheet661 KB - PDF
EMFreakz - May 2, 2013 5:23 AM
Great talk. thanks. I may have missed it, but was the MR rescue trial discussed? http://www.nejm.org/doi/full/10.1056/NEJMoa1212793 This was published in the same issue of NEJM and at the same stroke conference which specifically addressed the issue of not only interventional but also perfusion scanning/CT angio which did not show benefit of performing either.
Where the rubber meets the road (in conjunction with your talk on EMS later) is when should the patient coming via EMS not even go to a hospital doesn't have interventional? 0-4.5hrs, no study has show benefit of inteventional over tPA, so seems like those folks should go to any hospital that can give tPA regardless of whether they can do interventional. And even the 4.5-8hr or further, the MR rescue trial also showed no benefit. But should we do it anyway b/c we're not going to tPA them?
I agree that we need better trials, but are we going to get one? What do you think of the editorial (http://www.nejm.org/doi/full/10.1056/NEJMe1215730 hopefully that's the right one) where the article points out that though there's no randomized trial showing benefit, medicare has already decided to reimburse this in the thousands, therefore, it's going to be difficult to see that trial due to equipoise?
Pierre M. - May 2, 2013 6:06 AM
As always I Iove EMRAP and really like MIke Weinstock but I did not like this section. When pro-intervention/pro-tPA MDs talk about stroke studies they have a bizarre passion that clouds their thinking.
When discussing a study that failed to show what they hope it will show, they point out all the problems with the study, talk about nonsignificant relative benefits in subgroup analysis and talk about future studies.
However, when discussing the ONLY two DBRCT studies that have EVER shown benefit (NINDS, ECASS III) they suddenly are not critical in any way. 2 studies out of over a dozen well done DBRCT have shown benefit and their assessment is (paraphrased) : "we have to keep selecting patients carefully to get it right" meaning "manipulate study design to stack the deck in our favour." There is no discussion of the serious flaws with the only two positive studies and absolutely no mention of the preponderance of negative literature. There is also the prerequisite discussion of registries to support their theories, a trick used anytime you can't find actual evidence to support your argument.
One of the discussants incorrectly stated that NINDS showed more benefit in 0-90 minutes (true) and that ECASS 3 also showed that time mattered. That is in fact false as there was no difference between 3 and 4.5 hours in benefit in ECASS 3.
The only thing I heard during that whole segment that I agreed with was the notion that patients should be enrolled in studies of interventional therapy/ tPA if the issue is to be settled. Until then, there is no good evidence that interventional therapies help stroke patients and on balance decent evidence of harm.
Mike W. - May 2, 2013 7:16 AM
WEINSTOCK: Thanks for these insightful comments - these are good points - studies are only good when they give you the answer you are looking for... of course! I tried to hold Budzik's feet to the fire with his "we need to select patients better" but I will say that he has a decent point that there are new devices they are using that were not used during this study. In a age of rapidly changing technologies, we need to find a way to do studies more rapidly or this will always be a criticism. I am going to forward these comments to Budzik and Eubank for their responses!
Elise O. L., M.D. - May 12, 2013 5:50 PM
Listening to this discussion raised some concerns. Dr. Budzik questions his own study; the primary end point, timing, and patient selection, then offers the "I pulled a clot out and the patient is now better so I don't a trial to tell me it works" rationalization as a substitute for science. These arguments might have convinced some people prior to SYNTHESIS, MR RESCUE and IMS III, but how many negative RCTs need to be published in the NEJM in order to put this controversy to rest? MR RESCUE even addresses the issue of patient selection, demonstrating that patients with a favorable penumbra pattern on neuroimaging still fail to benefit from endovascular therapy.
At a minimum moving forward, if new technologies are in development, they need to be subjected to rigorous study through RCTs. This would avoid the "Mikey likes it" approach that led to the widespread use of retrieval devices without proof of benefit. Current devices are now shown in IMS III and MR RESCUE to be equivalent to IV thrombolytics at best, with SYNTHESIS demonstrating a trend towards worse outcomes in the endovascular therapy group.
Preston W. - May 12, 2013 6:44 PM
I have a few issues with this. First I am tired of believers treating ED physicians as though we are somehow mentally defective. I know how to perform a neurologic exam. I know how to calculate an NIH stoke number. I am very familar with giving TPA. Many of us grew up in the TPA for STEMI era. ED physicians have given this drug many many times. Strangely without resistance. WHY? It was CLEAR!!! that TPA was effective in STEMI. The utility of TPA for stroke is "less clear?" Calling TPA "the GOLD STANDARD" in stroke treatment is just WRONG! did I miss a study? does TPA for Stroke reduce mortality? Last I read TPA in stroke improved function as defined by a scale that was never independantly verified....at 90 days. At one point he compared interventional stroke therapy to coronary stenting. As I recall every percutaneous intervention study showed improved survival as compared to Thrombolytic therapy. I do agree that patient selection continues to be a problem. i agree that there is a patient population that might benefit. However, they have yet to be identified.
Mike W. - May 21, 2013 9:41 PM
RESPONSE FROM GEOFF EUBANK (neurologist on the program):
The purpose of this discussion was not to revisit IV tPA for ischemic stroke(debated in a prior issue of EMRAP), though I/we recognize that it is still controversial for some in the ED world. Recent policy statements from the American College of Emergency Physicians (with the Academy of Neurology) do further support the use of IV tPA for use in acute ischemic stroke. That being said, let me address a few of the comments made. TPA has not been shown to reduce stroke mortality. Agreed. We wish it did improve mortality, but frankly our disease happens to be (apparently) much trickier to treat and a fair bit more resistant and risky than treatment for STEMI's. The evaluation of ischemic stroke also remains often a bit more nuanced than STEMI, it is for less certain variables at times, making treatment decisions challenging. Stroke is, however, THE leading cause of disability in adults and the NINDS was specifically intended and powered to look at that. It did, in fact, show that disability, as measured by the modified Rankin Scale, was reduced in the tPA treated group. Not by data mining or twisting of the data, but by preplanned objectives and analysis. I also agree that numerous studies prior to the NINDS and ECASS trials failed to show benefit.The lessons learned from those trials were applied to subsequent trials. Shorter time window. Avoidance of anti coagulated patients. Tight control of BP. Exclusion of visible infarcts on CT.
Timing is critical as evidenced from the trials and also by its criticisms. Much of the benefit was derived by the 0-90 minute patient group. The 90-180 patient group showed less benefit. My comments about ECASS 3 was that the longer it takes for patients to receive tPA, the less benefit we see, though this is an inferred conclusion from the trials, admittedly.
Which brings us to the interventional aspect of acute stroke treatment. The rationale for this therapy derives from the fact that patients with large vessel occlusion have a natural history of high disability and mortality, even with those that might receive IV tPA. This known problem leads to hypothesis that a more direct recanalization would provide more or (for the cynical) some benefit. Early intraarterial thrombolysis (PROACT) did show benefit, though with enough safety concerns to preclude FDA approval. Interventional devices were subsequently designed to hopefully provide a safer alternative than chemical thrombolysis. I have been witness to numerous cases where patients have had truly miraculous recoveries, very shortly after a clot is removed. It is very hard to not offer that after you have seen it numerous times. Yet, individual cases cannot provide proof of benefit, just the impetus to propel forward.
This is where I may deviate from my interventional colleagues. They are likely correct in their assessment of benefit. But this is precisely where randomized clinical trials need to answer the question of actual benefit. The challenges with new technology does somewhat necessarily lead to technical feasibility issues first (experience, device mechanics, etc.) before they are ready for RCT's. It's hard to withhold a therapy you wholeheartedly believe works. Ideally, physicians would not have to make this decision prior to proof of benefit. Once a product "matures" in centers with the dedication and expertise to study, it should have the spotlight of RCT's shown on it to see if it truly benefit patients. We haven't demonstrated that yet. It should be pointed out that the therapy was not medically harmful either. Learning from failed trials and redesigning trials them to learn is not stacking the deck, its trying to separate the wheat from the chaff. Some patients with certain anatomic or medical characteristics may be excluded, leaving a group that could, in fact, benefit. Moreover, the devices used in the IMS III trial and others have been shown to be far inferior to the latest generation devices (in fact, many patients simply had intraaterial thrombolysis). Additionally, sites that routinely used general anesthesia, may have skewed results in a negative fashion (hemodynamic issues with general anesthesia could have adversely impacted the interventional group). Time to intervention is another area that is improving and, similar to IV tPA, seems to influence outcomes. I truly believe these factors are the difference between a positive and a negative trial. I also believe that if this is true, it will withstand the scrutiny of a RCT. We should insist that they do. Both as an adjunct to IV tPA in the <4.5 hour group and also as a stand alone therapy in the >4.5 hour group.
What we have learned from these trials should and has given us (the medical community) the appropriate pause to truly assess whether these therapies are beneficial. I expect our ED colleagues will support these necessary research endeavors, whilst holding our feet to the fire until if/when they do.
Geoffrey Eubank, MD
Co-director
Ohio Health Riverside Methodist Hospital Stroke Program
Andrew C. - May 21, 2013 11:31 PM
Dear Mike and Geoff
Thanks for the informative and insightful discussions that you have provided for the E M R A P listeners. I, in fact, felt that your approach was fairly neutral and as an Emergency Doctor feel that the stroke literature has been misinterpretted to a degree by messrs Hoffman et al.
While Thrombolysis is associated with a variable increase risk of post therapy bleeding there is fairly good evidence for benefit in terms of Morbidity. This has been established in multiple studies, although granted the variable ways of measuring morbidity (Rankin Scoring) and changing of protocols during studies (ECASS III) as well as a lack of homogenity leaves some of the studies open to critique.
After hearing this segment I was more confused about imaging so I took the liberty of reviewing this here: http://emergencypedia.com/2013/05/22/imaging-in-acute-stroke-and-tia/
I would welcome your comments on this...
Kind Regards
Andrew Coggins
MBChB MRCP(UK)
EMFreakz - May 26, 2013 2:48 PM
Thank you for your comments. Thus far, however, I haven't read anyone who has really given a reply regarding the stroke triage question for EMS, should we at this point be diverting patients to stroke centers that have interventional if they can make it in 8hrs? If they are within the 4.5hr window, should they then be transferred to facilities that can provide interventional? My understanding is that medicare is reimbursing for this e.g. Merci reimburses 23k according to: http://www.bonsecours.com/assets/CME/Jensen2.pdf vs $4,900-6,400 for tPA. This makes it difficult to know what's going on.
Sean G., M.D. - May 29, 2013 3:46 AM
My take on the lecture was no, the literature says we should be lysing these folks then transferring them.
Jim M., DO - June 20, 2013 1:04 PM
I'm confused. A few times in the discussion I heard the phrase "so I've given the tPA, they're not getting better, what do I do?"
Am I missing something? I thought all the benefit shown in NINDS was 6 months out after the event, so I wouldn't expect to see a benefit (unless I'm treating a TIA). I am one of the skeptics, but still offer it to patients. I really did like the "script" regarding informed consent.
Richard C. - June 23, 2013 11:51 PM
I don't like the tone in the written summary that IV TPA is accepted as beneficial: "If you compare the chances of the patient improving compared to the chances of complications such as bleeding or making things worse, the evidence favors treatment for the stroke."
Echoing above statements, there have been 12 RCTs looking at IV TPA. 2 showed benefit, 6 showed no benefit, and 4 showed harm. Some argue that it's the time window, i.e. it works if given early. However, the biggest trial, IST-3, showed benefit 0-3h, harm 3-4.5hr, and benefit 4.5-6hrs. This is inconsistent with our theory of why TPA works-biological implausibility.
Another argument, mentioned above, is that the harm trials were the early ones, that we changed protocols, and things work now. But the biggest most recent trial, IST3, is one that showed no benefit.
An excellent, and brief, review from David Newman and his crew is available on the NNT: http://www.thennt.com/nnt/thrombolytics-for-stroke/
By the way, excellent point Jim M! Some people see a patient getting better in front of them while TPA is being administered and think it's the TPA; it's not. Like Jim said, the reported benefit is change in Rankin scale months after the stroke.