By Joseph Broderick
The publication and presentation of four endovascular trials for acute ischaemic stroke has evoked a number of metaphors: “sea change”, “new day”, and “the polar vortex has lifted”. But there is an overwhelming consensus that we have taken a big step forward in acute ischaemic stroke therapy. The trial which opened the floodgates was the MR CLEAN trial (Multicentre randomized clinical trial of endovascular treatment for acute ischaemic stroke in the Netherlands) which presented its results in October at the World Stroke Congress. These clearly positive data prompted the data and safety monitoring boards of other ongoing endovascular trials to review their own data. All of these trials were then stopped early because of clear efficacy and crossing of stopping boundaries within the respective trials. And this is the reason why so many positive trials were presented simultaneously at the International Stroke Conference in Nashville, USA this February – something we may never see again at a single international stroke meeting.
Why were past trials neutral and the recent trials positive? Three primary reasons likely explain the difference. First, all subjects in the latest stent retriever trials had documentation of a large artery occlusion prior to randomisation. When the earlier trials, such as IMS III (Endovascular therapy after intravenous t-PA versus t-PA alone for stroke) began, use of computed tomography angiography (CTA) was quite limited. Use of CTA changed dramatically from 2006 to present. Support for the importance of large artery occlusions, comes from one of the earlier larger trials, IMS III, where a post-hoc analysis of patients with documentation of a large artery occlusion on baseline CTA showed benefit for endovascular therapy. In addition, endovascular therapy benefited those subjects in IMS III with very severe strokes (NIHSS≥20) at 12 months, all of whom had large artery occlusions. This benefit was seen even with use of older clot-retrieval technology.
Secondly, the new stent retriever technology is a game changer with fast and better reperfusion rates compared to prior technology. The higher rates of reperfusion in the stent-retriever trials correspond with better outcomes as compared to prior trials which had very limited use of the newer technology. Third, several of the investigators in IMS III, in which efforts to decrease the time from hospital arrival to deployment of endovascular therapy was a major focus and progress had been made by end of the trial, implemented the discipline of time to treatment even further with much more rapid times from hospital arrival to endovascular therapy in the stent-retriever trials. Time to reperfusion has been clearly demonstrated to be associated with outcome.
Except for MR CLEAN, each of the trials used brain and vascular imaging to explicitly select patients who may most likely to have salvageable brain if reperfusion could be obtained: ASPECTS and CTA collateral flow (ESCAPE), CT perfusion/core (EXTEND IA), and CT perfusion/ASPECTS (SWIFT PRIME). The imaging selection process is one reason why the patient outcomes in EXTEND-IA, SWIFT PRIME, and ESCAPE are all better than in MR CLEAN – for both the endovascular groups as well as the standard group (which included mostly participants treated with IV t-PA). However, MR CLEAN, using baseline CT to rule out larger areas of ischaemic brain and CTA to document large artery occlusion, still found a clear difference in benefit for endovascular therapy within six hours of onset. At present, a good clinical evaluation, a baseline CT, and CTA are probably sufficient to select patients for endovascular therapy within the current treatment window of IV t-PA of four and a half hours. This may evolve over time.
What are the questions that remain unanswered and what work needs to be done? First, we still have limited trial data beyond six hours from stroke onset so we do not know whether patients selected by the various imaging techniques at later time windows will also benefit from endovascular therapy. We need to examine the trials for data regarding the use of general anaesthesia and conscious sedation to see if current local practices should change. We have to find better ways to prevent embolisation in patients with endovascular therapy which is a clear documented risk. Finally, given the demonstration once again of the importance of early and better reperfusion in improving outcome, we need to find methods of medical reperfusion (combination therapies for example) that are better than IV t-PA alone, since many patients will always be distant from comprehensive stroke centres.
The benefit of endovascular therapy in certain populations of stroke patients has major implications for regional stroke systems of care and prehospital triage of stroke patients in communities across the USA and around the world. In some cities in countries outside of the USA, acute stroke triage is centralised and streamlined. No such system currently exists in US communities. It will be critical for all involved with care of stroke patients to be active at the local hospital, regional, state, and national levels to work out the best systems of acute stroke care for our individual communities over the coming years. What is best for patients, not for a given hospital system or systems, should be the focus of these discussions and future changes.
Joseph Broderick is professor of Neurology and director of the Neuroscience Institute at the University of Cincinnati, Cincinnati, USA. He was the principal investigator of the IMS III trial.