Conference Coverage

Collateral flow flags stroke patients for late thrombectomy


Patients with acute ischemic stroke presenting late at the hospital can be selected for endovascular thrombectomy by the presence of collateral flow on CT angiography (CTA), a new study shows.

The MR CLEAN-LATE trial found that patients selected for thrombectomy in this way had a greater chance of a better functional outcome than patients who did not receive endovascular therapy.

The study was presented at the 14th World Stroke Congress in Singapore by study investigator Susanne Olthuis, MD, of Maastricht (the Netherlands) University Medical Center.

Patients in the intervention group were more likely to show a benefit on the primary endpoint of modified Rankin Scale (mRS) score at 90 days with a significant common odds ratio of 1.68, a finding that received applause from attendees of the plenary WSC session at which the study was presented.

“This means that patients treated with endovascular therapy in this trial had about a 1.7 times higher chance of achieving a better functional outcome at 90 days,” Dr. Olthuis said.

“Selection based on collateral flow identifies an additional group of patients eligible for late-window endovascular therapy in addition to those eligible based on perfusion and clinical criteria,” Dr. Olthuis concluded.

“We recommend implementation of collateral selection in routine clinical practice as it is time efficient. The CTA is already available, and it involves a low-complexity assessment. The only distinction that needs to be made is whether or not there are any collaterals visible on CTA. If collaterals are absent or there is any doubt, then CT perfusion [CTP] imaging can still be used,” she added.

Co–principal investigator Wim H. van Zwam, MD, interventional radiologist at Maastricht, said in a comment:“My take-home message is that now in the late window we can select patients based on the presence of collaterals on CT angiography, which makes selection easier and faster and more widely available.

“If any collaterals are seen – and that is easily done just by looking at the CTA scan – then the patient can be selected for endovascular treatment,” Dr. van Zwam added. “We don’t need to wait for calculations of core and penumbra volumes from the CTP scan. There will also be additional patients who can benefit from endovascular therapy who do not fulfill the CTP criteria but do have visible collaterals.”

Explaining the background to the study, Dr. Olthuis noted that endovascular thrombectomy for large vessel occlusion stroke is safe and effective if performed within 6 hours and the effect then diminishes over time. In the original trial of endovascular treatment, MR CLEAN, patients with higher collateral grades had more treatment benefit, leading to the hypothesis that the assessment of collateral blood flow could help identify patients who would still benefit in the late time window.

The current MR CLEAN-LATE trial therefore set out to compare safety and efficacy of endovascular therapy in patients with acute ischemic stroke in the anterior circulation presenting within 6-24 hours from symptom onset with patients selected based on the presence of collateral flow on CTA.

At the time the trial was starting, the DAWN and DEFUSE 3 trials reported showing benefit of endovascular therapy in patients presenting in the late window who had been selected for endovascular treatment based on a combination of perfusion imaging and clinical criteria, so patients who fitted these criteria were also excluded from MR CLEAN-LATE as they would now be eligible for endovascular therapy under the latest clinical guidelines.

But the study continued, as “we believed collateral selection may still be able to identify an additional group of patients that may benefit from endovascular therapy in the late window,” Dr. Olthuis said.

The trial randomly assigned 502 such patients with a National Institutes of Health Stroke Scale (NIHSS) score of at least 2 and with collateral flow grades of 1-3 to receive endovascular therapy (intervention) or control.

Safety data showed a slightly but nonsignificantly higher mortality rate at 90 days in the control group (30%) versus 24% in the intervention group.

The rate of symptomatic intracranial hemorrhage was higher in the intervention group (6.7%) versus 1.6% in the control group, but Dr. Olthuis pointed out that the rate of sICH in the intervention group was similar to that in the endovascular groups of the DAWN and DEFUSE 3 trials.

The primary endpoint – mRS score at 90 days – showed a shift toward better outcome in the intervention group, with an adjusted common OR of 1.68 (95% confidence interval, 1.21-2.33).

The median mRS score in the intervention group was 3 (95% CI, 2-5) versus 4 (95% CI, 2-6) in the control group.

Secondary outcomes also showed benefits for the intervention group for the endpoints of mRS score 0-1 versus 2-6 (OR, 1.63); mRS 0-2 versus 3-6 (OR 1.54); and mRS 0-3 versus 4-6 (OR, 1.74).

In addition, NIHSS score was reduced by 17% at 24 hours and by 27% by 5-7 days or discharge in the intervention group. Recanalization at 24 hours was also improved in the intervention group (81% vs. 52%) and infarct size was reduced by 32%.

Dr. Olthuis explained that collateral grade was defined as the amount of collateral flow in the affected hemisphere as a percentage of the contralateral site, with grade 0 correlating to an absence of collaterals (and these were the only patients excluded).

Grade 1 included patients with 50% or less collaterals, grade 2 more than 50%, and grade 3 excellent collaterals – around 100%. “We included grade 1, 2 and 3, and subgroup analysis suggested no treatment interactions between different collateral grades in the patients included,” she said.

Dr. van Zwam noted that there has been evidence from other studies suggesting that the presence of collateral flow could be used to select patients for late thrombectomy, but MR CLEAN-LATE is the first randomized trial to show this and provides confirmation that this strategy is valid.

“Our results show that patients can be selected with just standard CT angiography imaging and that CT perfusion is not necessary. This will make it easier and faster to select patients especially for centers in low-resource areas who do not yet have CT perfusion imaging,” he commented.

“But even in centers where CT perfusion imaging is performed, these results should mean that we do not have to wait to analyze these results before going ahead with thrombectomy. It will also give us an additional tool, as some patients do not meet the criteria on perfusion imaging but still have identifiable collaterals and thus would now qualify for endovascular thrombectomy,” he added.


Recommended Reading

Air pollution tied to stroke risk, subsequent CV events, and death
MDedge Cardiology
Apixaban outmatches rivaroxaban in patients with AFib and valvular heart disease
MDedge Cardiology
Diabetes becoming less potent risk factor for CVD events
MDedge Cardiology
Risk score refines TIA management for PCPs, emergency docs
MDedge Cardiology
Stroke management: There’s an app for that
MDedge Cardiology