Local Hospitals Still Have a Role in Treating Severe Stroke

November 09, 2020

For patients suspected of having a large-vessel occlusion stroke, direct transfer to a comprehensive stroke center capable of endovascular therapy that is farther away than a local primary stroke center does not necessarily produce better functional stroke outcomes, a new study has shown.

In the RACECAT trial, functional outcomes were similar for patients suspected of having a large-vessel occlusion stroke who were located in areas not currently served by a comprehensive stroke center, whether they were first taken to a local primary stroke center or whether they were transported over a longer distance to a comprehensive center.

Dr Marc Ribo

"Under the particular conditions in our study where we had a very well-organized system, a 'mothership' transfer protocol for patients with suspected large-vessel occlusion has not proven superior over the 'drip and ship' protocol, but the opposite is also true," lead investigator Marc Ribo, MD, concluded.

Ribo, who is assistant professor of neurology at Hospital Vall d’Hebron, Barcelona, Spain, presented the RACECAT results at the European Stroke Organisation-World Stroke Organisation (ESO-WSO) Conference 2020 on November 7.

Ribo told Medscape Medical News that there is a feeling among the stroke community that patients with a suspected large-vessel occlusion should be transferred directly to a comprehensive stroke center capable of performing endovascular thrombectomy, even if there is a nearer, smaller primary stroke center where patients are usually taken first for thrombolysis.

"Many stroke neurologists believe we are losing time by sending patients with severe stroke to a local hospital and that we should skip this step, but this is controversial area," he commented. "Our findings suggest that we should not automatically bypass local stroke centers."

Ribo pointed out that the local centers performed very well in the study, with very fast "in/out" times for patients who were subsequently transferred for thrombectomy.

"On the basis of our results, we recommend that if a local stroke center can perform well like ours did ― if they are within the time indicators for treating and transferring patients ― then they should keep receiving these patients. But if they are not performing well in this regard, then they should probably be bypassed," he commented.

The RACECAT trial was well received by stroke experts at an ESO-WSO 2020 press conference at which it was discussed.

Stefan Kiechl, MD, Medical University Innsbruck, Innsbruck, Austria, described the trial as "outstanding," adding: "It has addressed a very important question. It is a big achievement in stroke medicine."

Patrik Michel, MD, Lausanne University Hospital, Lausanne, Switzerland, commented: "This is a very important and highly sophisticated trial in terms of design and execution. The message is that it doesn't matter which pathway is used, but it is important to have a well-organized network with highly trained paramedics."


The RACECAT trial was conducted in the Catalonia region of Spain. Twenty-seven hospitals participated, including seven comprehensive stroke centers and 20 local stroke centers.

The trial included stroke patients with suspected large-vessel occlusion stroke, as determined on the basis of evaluation by paramedics using the criteria of a RACE scale score >4 and on the basis of a call to a vascular neurologist. For inclusion in the study, patients had to be in a geographical area not served by a comprehensive stroke center and to have an estimated arrival time to a comprehensive center of less than 7 hours from symptom onset in order that thrombectomy would be possible.

Of 7475 stroke code patients evaluated, 1401 met the inclusion criteria and were randomly assigned to be transferred to a local hospital or to a comprehensive stroke center farther away.

Baseline characteristics were similar between the two groups. The patients had severe strokes with an average National Institutes of Health Stroke Scale score of 17. It was later confirmed that 46% of the patients enrolled in the study had a large-vessel occlusion stroke.

Results showed that time from symptom onset to hospital arrival was 142 minutes for those taken to a local center and 216 minutes for those taken to a comprehensive stroke center. Of those taken to a local hospital, 86% arrived within 4 hours of symptom onset and so were potential candidates for thrombolysis, compared with 76% of those taken to a comprehensive center.

Of the patients taken to a local hospital, 60% were given thrombolysis, vs 43% of those taken immediately to a comprehensive center. On the other hand, 50% of patients who were taken directly to a comprehensive center underwent thrombectomy, compared to 40% who were first taken to a local center.

For patients who received thrombolysis, time to tissue plasminogen activator administration was 120 minutes for those treated at a local hospital, vs 155 minutes for those taken directly to a comprehensive center.

For patients who received thrombectomy, time from symptom onset to groin puncture was 270 minutes if they were first taken to a local hospital and were then transferred, vs 214 minutes for those taken directly to the comprehensive center.

The primary efficacy endpoint was functional outcome using Modified Rankin Scale (mRS) shift analysis at 90 days for ischemic stroke patients. This showed a "completely flat" result, Ribo reported, with an adjusted hazard ratio of 1.029 for patients taken to a comprehensive center in comparison with those taken to a local center.

"There was absolutely no trend towards benefit in one group over the other," he said.

What About Hemorrhagic Stroke?

The study also evaluated functional outcomes for the whole population enrolled. "If we make the decision just based on thrombectomy-eligible patients, we may harm the rest of the patients, so we did this study to look at the whole population of severe stroke patients," Ribo said.

Of the study population, 25% of patients were found to have had a hemorrhagic stroke.

"The problem is, at the prehospital level, it is impossible to know if a patient is having a large-vessel occlusion ischemic stroke or a hemorrhagic stroke," Ribo explained. "We have to make a decision for the whole population, and while a longer transport time to get to a comprehensive stroke center might help a patient with a large-vessel occlusion ischemic stroke, it might not be so appropriate for patients with a hemorrhagic stroke who need to have their blood pressure stabilized as soon as possible."

For the whole population, the mRS shift analysis at 90 days was also neutral, with an adjusted hazard ratio of 0.965.

When considering only patients with hemorrhagic stroke, the adjusted hazard ratio for the mRS shift analysis at 90 days was 1.216, which was still nonsignificant (95% CI, 0.864 – 1.709). This included a nonsignificant increase in mortality among those taken directly to a comprehensive center.

"If we had better tools for a certain diagnosis in the field, then we could consider taking large-vessel occlusion ischemic stroke patients to a comprehensive center and hemorrhagic stroke patients to the local stroke center, but so far, we don't have this option apart from a few places using mobile stroke units with CT scanners," Ribo noted.

Transfer times to comprehensive centers in the study ranged from 30 minutes to 2.5 hours. "There might well be a difference in outcomes for short and long transfers, and we may be able to offer different transfer protocols in these different situations, and we are looking at that, but the study was only stopped in June, and we haven’t had a chance to analyze those results yet," Ribo added.

Complications during transport occurred in 0.5% of those taken to a local hospital and in 1% of those taken directly to a comprehensive center. "We were concerned about complications with longer transfers, but these numbers are quite low. Intubations were very low ― just one patient taken to a local center, vs three or four in the longer transfer group," he added.

For both local and comprehensive centers, treatment times were impressive in the study. For local hospitals, the average in/out time was just 60 minutes for patients who went to a comprehensive center; for patients receiving thrombolysis, the average door to needle time was around 30 minutes.

Time to thrombectomy in the comprehensive center for patients transferred from a local hospital was also very fast, with an average door to groin puncture time of less than 40 minutes. "This shows we have a very well-oiled system," Ribo said.

"There is always going to be a balance between a quicker time to thrombolysis by taking a patient to the closest hospital but a quicker time to thrombectomy if patients are taken straight to the comprehensive center," he concluded. "But in our system, where we are achieving fast treatment and transfer times, our results show that patients had timely access to reperfusion therapies regardless of transfer protocol, and under these circumstances, it is fine for the emergency services to take stroke patients to the closest stroke center."

Results Applicable Elsewhere?

During the discussion at an ESO-WSO 2020 press conference, other experts pointed out that the Catalonia group is a leader in this field, being the pioneers of the RACE score used in this study for paramedics to identify suspected large-vessel occlusions. This led to questions about the applicability of the results.

"The performance by paramedics was very good using the RACE scale, and the performance times were very impressive. Are these results applicable elsewhere?" Kiechl asked.

Ribo said the combination of the RACE score and a call with a vascular neurologist was of "great value" in identifying appropriate patients. Half of the patients selected in this way for the trial were confirmed to have a large-vessel occlusion. "That is a good result," he added.

He noted that the performance of the local hospitals improved dramatically during the study. "They had an incentive to work on their times. They could have lost most of their stroke patients if their results came out worse. We told them they had an opportunity to show that they have a role in treating these patients, and they took that opportunity."

Ribo said there were lessons here for those involved in acute stroke care. "When creating stroke transfer policies in local networks, the performances of individual centers need to be taken into account. If primary stroke centers are motivated and can work in a well-coordinated way and perform to within the recommended times, then they can keep receiving stroke code patients. This should be possible in most developed countries."

Noting that the in/out time of 60 minutes at local hospitals was "very impressive," Kiechl asked how such fast times were achieved.

Ribo responded that to a great extent, this was due to ambulance staff. "We have trained the paramedics to anticipate a second transfer after delivering the patient to the local hospital so they can prepare for this rather than waiting for a second call."

Ribo pointed out that there were other advantages in taking patients to local centers first. "For those that do not need to be transferred on, they will be closer to relatives. It is very difficult for the family if the patient is hundreds of miles away. And there may be a cost advantage. We did look at costs, but haven't got that data yet."

He said, "If local stroke centers do not treat so many stroke code patients, they will lose their expertise, and that will be detrimental to the remaining patients who are taken there. We want to try to maintain a good standard of stroke care across a decent spread of hospitals ― not just a couple of major comprehensive centers," he added.

Commenting on the study for Medscape Medical News, Jesse Dawson, MD, University of Glasgow, United Kingdom, who was chair of the plenary session at which the study was presented, said: "RACECAT is very interesting but needs a lot of thought to dissect.

"My take-away is that we know that time to reperfusion is key, and we need to get these times as low as possible, but we don't need to chase a particular care pathway," Dawson said. "Thus, if your country/geography suits 'drip and ship' better, this is acceptable. If direct to endovascular is possible or you are close to such a center, then this is ideal. But within those paradigms, be as fast as possible."

He added that results of the subgroups with regard to transfer time will be helpful.

The RACECAT study was funded by Fundacio Ictus Malaltia Vascular.

European Stroke Organisation-World Stroke Organisation (ESO-WSO) Conference 2020: Presented November 7, 2020.

For more Medscape Neurology news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.