The results of the UK-based PISTE trial demonstrate that the outcomes are consistent with the published data from similar mechanical thrombectomy acute ischaemic stroke trials and closer to those trials that aimed for a rapid intervention approach.
PISTE (Pragmatic ischaemic stroke thrombectomy evaluation) co-principal investigator, Keith Muir (University of Glasgow, UK), presented the data for the first time at the International Stroke Conference (ISC; 17–19 February, Los Angeles, USA). The trial was funded by the UK Stroke Association, and in part by the National Institute for Health Research Health Technology Assessment programme.
Eleven centres across the UK participated in the PISTE trial which enrolled 65 patients on an intent-to-treat basis. Thirty-two patients were randomised to receive IV t-PA only and 33 were randomised to receive IV t-PA and additional intra-arterial therapy. Patients were selected using simple imaging.
“Where PISTE sits is a subtly different place within the ecosystem of acute thrombectomy trials in that we have had trials based on simple imaging, but these also had a policy of waiting to assess response to IV therapy—both MR CLEAN and REVASCAT either explicitly or implicitly waited for a response. The trials which used complex imaging (CT perfusion, multiphase collaterals or MRI) selected favourable profiles, and these were the trials which additionally had a policy of proceeding as fast as possible to intervention. PISTE occupied a small niche of proceeding as fast as possible to intervention on the basis of simple imaging of CT and CT angiography alone,” Muir explained.
In the PISTE trial IV t-PA had to be started within 4.5 hours after symptom onset, and enrolment, randomisation and procedure commencement (groin puncture) had to be started within 90 minutes of the start of IV t-PA treatment (groin puncture had to be within 5.5 hours after symptom onset). The primary outcome was modified Rankin Scale score of 0–2 at 90 days.
Like many of the other thrombectomy trials, PISTE was discontinued prematurely, which resulted in a small sample size and some baseline imbalances. In the intra-arterial therapy arm there was by chance an older population, and also a higher proportion of patients with diabetes and a higher stroke severity with a median NIHSS score of 18 compared with 14 in the t-PA only group.
In both groups the median ASPECTS score was 9. It was predominantly M1 occlusions and there was a range of collateral scores across the two groups.
Investigators excluded patients who had protocol deviations, which included extensive ischaemia on baseline CT, allocation crossovers on both sides and those who were ineligible on the basis of having an incorrect occlusion site. They therefore had a per protocol population of 28 patients in the t-PA only group and 30 patients in the intra-arterial therapy group.
In terms of timing, Muir reported that in both groups there was a median of 120 minutes from symptom onset to IV t-PA start, and a median of 150 minutes from symptom onset to randomisation. In the intra-arterial group, the time from IV t-PA start to groin puncture was 82 minutes median; randomisation to groin puncture was 58 minutes median; groin puncture to device removal was 49 minutes median and the total time from symptom onset to procedure end was 256 minutes.
“It places us where we wanted to be in terms of early IV start with early reperfusion and no delay, more comparable to the SWIFT PRIME, EXTEND-IA and ESCAPE protocols compared with the other simple imaging-based trials,” Muir explained.
General anaesthesia was used in 31% of patients, the remainder being treated under conscious sedation or local anaesthesia only. Two-thirds of the patients in the intra-arterial therapy arm were treated using stent retrievers, and the rest using aspiration.
In terms of technical success, the PISTE trial had a mTICI 2b-3 rate of 87% (26/30 patients) at the end of the procedure. At 24 hours post-procedure however, CT angiography showed occlusion in six of 27 patients from the intra-arterial therapy arm that had the follow-up scanning. Muir explained that the investigators are looking more closely at this to see if they can speculate as to why there may have been reocclusion in this small number of cases.
In the per protocol population at 90 days post-procedure, there was significant outcome in both the primary and secondary outcome measures. Primary outcome: mRS 0-2: OR 4.92 (1.23, 19.69), p=0.021; and secondary outcome mRS 0-1: OR 14.6 (2.11, 101.5), p=0.005; and mRS distribution: OR 4.47 (1.45, 13.8), p=0.009.
“When we looked at all the primary and secondary clinical outcomes, we saw direction of effect clearly in favour of endovascular therapy across all of the efficacy outcomes. We saw no difference in safety measures, including mortality, symptomatic haemorrhage rates, PH 1/2 intracerebral haemorrhage rates, and favourable effect in terms of days spent in usual residence in favour of intra-arterial therapy,” Muir said.
He added that the outcomes in PISTE are consistent with the published data from similar trials and closer to those trials that aimed for a rapid intervention approach.
“In conclusion, we did achieve the planned timelines for rapid intra-arterial therapy. We had a very high rate of modified TICI2b-3. The primary endpoint was non-significant but there was a consistent odds ratio in the direction of the intra-arterial arm and the secondary endpoint showed significant benefit. All efficacy endpoints were consistent with intra-arterial benefit and we had primary and secondary endpoints significant in the per protocol population,” Muir stated.
NeuroNews spoke to Muir about the uptake of mechanical thrombectomy in the UK since the PISTE trial.
What has been the uptake of thrombectomy procedures at your centre and in the UK generally since the PISTE trial?
Implementation of thrombectomy has been discussed widely but will require a significant investment in training and careful review of stroke service configuration. The UK as a whole awaits the health economic review from NICE which will be important in determining whether or not thrombectomy becomes adopted widely. Our own centre has insufficient neurointerventional staff to offer this intervention at present, and many other stroke centres are in a similar position. A number of PISTE centres are offering treatment for limited hours, and only in one service in London is there an immediate plan to offer thrombectomy 24/7.