Now that the safety and efficacy of mechanical thrombectomy for the treatment of acute ischaemic stroke has been proven in at least five randomised controlled trials, there is some worry within the community that the pressure to deliver this therapy to all appropriate patients will result in inexperienced operators attempting to do the procedures. But who are the ideal operators?
At the Annual Meeting of the European Society of Minimally Invasive Neurological Therapy (ESMINT; 10–12 September, Nice, France) the ideal operator was one of the most heavily discussed topics, with Tommy Andersson (AZ Groeninge Kortrijk, Belgium and Karolinska University Hospital, Stockholm, Sweden) maintaining that the three main criteria for an operator performing mechanical thrombectomy effectively and safely are knowledge, experience and training.
Andersson explained that a requirement for a mechanical thrombectomy operator should be sufficient knowledge. “Knowledge about the brain as an organ, about the intracranial circulation and about the specifics of the intracranial arteries, because they differ a lot from systemic and cardiac arteries,” he said.
Further to knowledge, Andersson said that potential operators need to meet experience criteria. He explained that in some countries there are guidelines outlining the requirements that equal adequate experience. “For instance, in Germany, to perform intracranial interventions independently you should have performed a specific number of procedures. Mechanical thrombectomy is a surgical procedure, and I think people forget that, they compare it with IV tPA, but it is different. With IV tPA, you have a standardised pharmacological procedure, where it is done exactly the same way all over the world. Mechanical thrombectomy however, is not standardised. The time is different, there are more criteria, other patient-related criteria, and it is operator-related if you want to be fast, efficient and avoid complications,” he maintained.
Training, Andersson pointed out, is the most difficult to fulfil given the fact that patients requiring mechanical thrombectomy for the treatment of a stroke are acute patients and are, therefore, not readily available for practice. He suggested that operators should thus also be doing other neurointerventional procedures, “without which they will be unable to have the proper training and experience”.
He reiterated that with the ongoing mechanical thrombectomy “boom” and the subsequent risk that the pressure will prompt inexperienced operators to attempt doing the procedures, the solution is to have “a few good centres and developed infrastructure rather than to have many bad ones”. “I prefer at the moment specialised centres with experience in selection, in technique and in post-treatment care, but there is a lot of pressure now from society, hospital administrators, and mainly from neurologists. Everyone wants to have this now. In our practice, the same stroke clinics that previously hesitated to send patients because of lack of evidence now hesitate because they want to do it themselves.
“In this post-study era, in order to avoid complications and failed or futile revascularisations, I think the infrastructure needs to be organised in accordance with national and regional circumstances, the workflow of selection and technique need to be improved, and the operators need to be trained to get proper knowledge and finally the experience,” Andersson stated.
As for whether specialities other than neurointerventionalists should be attempting mechanical thrombectomy in acute stroke patients, Andersson said that it does not matter the specialty, as long as the individual possesses the adequate training.
“It is not the background, but the training that counts. Why should we accept that untrained people do mechanical thrombectomy when we would not accept it for anything else?” he questioned.
Later at the ESMINT Annual Meeting, Andersson’s sentiments were echoed by Michael Söderman (Karolinska University Hospital, Stockholm, Sweden) who maintained that mechanical thrombectomy in acute stroke should be performed by a trained neurointerventionalist.
As for what a trained neurointerventionalist in this era should be defined as, Söderman explained, “A trained neurointerventionalist does not necessarily mean someone who has gone through the proper training path; it includes specialists with different backgrounds, but who has been trained properly. It can be a neuroradiologist, or a neurosurgeon, or a neurologist, but proper training after being a specialist is probably the most important thing.”
Making his case for why endovascular treatment of stroke should be done by a trained neurointerventionalist, Söderman explained that that person is trained to access the cerebral circulation, has more experience, is faster and there is less risk for complications. That person, he said, also works in a neurovascular team and does triage by neuroradiology, has slimmed patient flow, rapid access to a stroke neurologist and can address related medical issues. Lastly, the trained neurointerventionalist understands neurovascular diseases and that all strokes are not “simple” M1 embolic occlusions.
In terms of experience, Söderman reported that high volume centres do better than low volume operators. He presented data from a nationwide inpatient sample hospital discharge database from 2008 (Mechanical Thrombectomy in Acute Stroke: Utilization Variances and Impact of Procedural Volume on Inpatient Mortality; Peter Adamczyk, MD1: J Stroke Cerebrovasc Dis. 2013) based on 2,749 patients, that showed that high volume centres were independently associated with lower mortality for stroke treatment. That study proposed that centralisation of stroke care should be to regional stroke centres with enough patient load.
Further, he explained that “practise makes the expert”. He used the example of the EVA-3S (Endarterectomy versus angioplasty in patients with symptomatic severe carotid stenosis) study where trial investigators could participate with as little experience as five carotid artery stenting procedures or with any prior carotid artery stenting experience at all when supervision by an experienced tutor was guaranteed. In that study, the 30-day mortality and morbidity incidence was 9.6%; compared with the CREST (Stenting versus endarterectomy for treatment of carotid-artery stenosis) study, with strict credentialing rules for participants, where the mortality and morbidity incidence was 5.6%. In terms of outcome based on operator specialty, the CREST data show that interventional neuroradiologists had the lowest event rate, followed by cardiologists, interventional radiologists and finally, vascular surgeons, Söderman explained.
“The only question in my mind is availability and workload—that is the only problem. We have to deliver stroke service to huge populations, so we have to be available and each centre has to have a reasonable workload. This in my mind is a matter of organisation and statistics and there is no excuse to offer service by insufficiently trained physicians to the population,” Söderman said.
With regards to his thoughts for the way forward, he suggested that a comprehensive stroke centre with 24/7 interventional neuroradiology service is the way it should go. That would in turn provide a centralised model of acute stroke care, in which hyperacute care is provided to all patients with stroke across an entire metropolitan area, and can reduce mortality and length of hospital stay.
Solidifying his stance, Söderman added, “It is probably better for a patient to wait an hour for transportation than to be admitted to a centre where they do not know what they are doing. These cases are not always simple or straightforward. You need experience, you need to review the cases, and you need to understand the physiology before you start to make treatment decisions.”