Acute Ischemic Stroke

Amrou Sarraj, MD, on the Optimization of Patient Selection for Endovascular Therapy in AIS

A new study published in the Annals of Neurology showed that patients with acute ischemic stroke with favorable imaging profiles on both noncontrast computed tomography (CT) and CT perfusion (CTP) imaging were more likely to receive endovascular therapy and had higher rates of functional independence.1

The study also indicated that patients with discordant profiles had achieved reasonable functional independence rates with thrombectomy, but those with an unfavorable CTP had higher rates of adverse outcomes associated with endovascular therapy.

These findings were based on the results of a phase 2, multicenter, prospective cohort study of 361 patients with large‐vessel occlusions in the anterior circulation who had presented up to 24 hours after the last known well had been conducted.

In the study, patients underwent a unified prespecified imaging evaluation via CT, CT angiography, and CTP with Rapid Processing of Perfusion and Diffusion software mismatch determination. Patients received either endovascular therapy (n=285 [79%]) or medical management, which was nonrandomized and at the discretion of treating physicians. Favorable profiles based on predefined criteria were adjudicated by an independent, blinded, neuroimaging core laboratory.

Of the 285 patients who underwent endovascular therapy, 87% had favorable CTs, 91% had favorable CTPs, 81% had favorable profiles on both, 16% had discordant profiles, and 3% had unfavorable profiles on both.

Other key findings included:

  • Favorable profiles on CT and CTP correlated similarly with 90‐day functional independence rates (56% vs 57%, respectively) without a difference in the functional outcomes rates in relation to treatment time window (early 0-6 hours vs late >6-24 hours).
  • Having a favorable profile on both modalities significantly increased the odds of receiving thrombectomy, compared with discordant profiles, with an adjusted odds ratio of 3.97.
  • Functional independence was achieved in 58% of patients with favorable profiles on both imaging modalities, compared with 38% with discordant profiles and 0% with unfavorable profiles on both modalities.
  • Among patients with discordant profiles, thrombectomy resulted in higher functional independence rates, compared with medical management alone.
  • Among patients with favorable CT and unfavorable CTP profiles, endovascular therapy was associated with high rates of symptomatic intracranial hemorrhage (sICH, 24%) and mortality (53%).


Neurology Consultant discussed the clinical implications of these findings with lead author Amrou Sarraj, MD, who is an associate professor and director of the Vascular Neurology Fellowship Program at McGovern Medical School at UTHealth in Houston, vice chair of Clinical Quality, and chief of General Neurology Service at Memorial Hermann–Texas Medical Center.

NEURO CON: What prompted you to conduct your study on optimizing patient selection for endovascular treatment in acute ischemic stroke?

Amrou Sarraj: Endovascular thrombectomy has been the cornerstone of the management of stroke patients presenting with large vessel occlusion in the anterior circulation over the last few years, since several randomized controlled trials proved the efficacy and safety of this intervention initially up to 6 hours and up to 24 hours from the last known well. However, for patients presenting up to 6 hours after stroke onset, some of these trials used noncontrast CT to identify patients who may benefit from thrombectomy, whereas the others used advanced perfusion imaging.

The DAWN and DEFUSE 3 trials had assessed the safety and efficacy of thrombectomy in patients presenting beyond 6 hours of stroke onset, with both trials utilizing advanced perfusion imaging modalities to identify the best thrombectomy candidates.2,3 Thus, it is not clear which imaging modality identifies thrombectomy candidates better and whether noncontrast CT can be utilized to select patients for thrombectomy beyond 6 hours after stroke onset.

Moreover, since most trials excluded patients with unfavorable profiles on one or both of the imaging modalities (also known as the large core), outcomes of thrombectomy in these patients is also unknown. In addition, it is unknown whether thrombectomy is beneficial when one imaging modality shows a favorable profile, but the other suggests a poor profile. To answer these important questions, we conducted the SELECT study.

NEURO CON: Were your study findings surprising, or did you anticipate them?

AS: Many of our findings were reported for the first time, since SELECT was the first multicenter cohort study to prospectively compare the clinical utility of noncontrast CT vs CT perfusion and their correlation with clinical outcomes after thrombectomy. All patients received a unified imaging profile with blinded assessment by an independent core laboratory. In the SELECT study, CT and perfusion both performed similarly in terms of identifying candidates who may achieve functional independence for thrombectomy. The high concordance between noncontrast CT and perfusion imaging that reached 81% was probably the most surprising finding, but this is probably why they both performed similarly in identifying candidates who benefit from thrombectomy.

Additionally, patients with a favorable profile on only 1 of the 2 imaging modalities also demonstrated better functional and safety outcomes with thrombectomy compared with medical management only. While such scenarios are frequent in clinical practice, we did not have any documented results of thrombectomy procedures in such patients before SELECT. The SELECT 2 trial, the randomized phase of the study, will assess if patients with unfavorable profiles on either CT or perfusion (or both) would still benefit from thrombectomy compared with medical management only, even with lower likelihood of functional independence and higher risk of complications.

NEURO CON: Your study showed that patients with unfavorable CTP had higher adverse outcomes. What are the implications of this finding for future research and clinical practice?

AS: We demonstrated that patients with a favorable CT, but an unfavorable CTP, had higher rates of adverse outcomes, including mortality, neurological worsening and symptomatic ICH. These outcomes, however, should be measured against those in patients who received medical management only. While there were higher rates of mortality and symptomatic intracerebral hemorrhage, thrombectomy was still associated with higher rates of functional independence compared with patients who received medical management only.  We aim to further assess how the efficacy and safety of EVT changes with unfavorable profiles on noncontrast CT and perfusion imaging in the ongoing SELECT 2 trial.

NEURO CON: What key takeaways do you hope to leave with neurologists on this topic?

AS: Since no randomized data are available on how selection by different imaging modalities affects thrombectomy outcomes, questions that emerged in daily clinical practice such as, “which imaging modality is superior?” or “should influence the treatment decision more than the other?” were previously unanswered.

The SELECT trial showed in a prospective cohort design that noncontrast CT and perfusion images correlated similarly with functional independence after thrombectomy. It also showed that these findings did not differ between patients who presented within 6 hours of stroke onset and those who presented beyond 6 hours. This demonstrated the potential for using noncontrast CT to identify thrombectomy candidates in the later treatment window.

A very important finding demonstrated in the SELECT trial was the potential benefit of thrombectomy in patients with discordant imaging profiles (favorable on one modality, but unfavorable on the other). We will be further assessing these findings in a randomized fashion in the SELECT 2 trial in order to determine the ultimate answer on imaging selection for thrombectomy, as well as the efficacy and safety of the intervention in patients with large-core strokes.

In the meantime, the results from the SELECT trial will help guide physicians on how to utilize CT and CTP for selecting patients for thrombectomy and deciding whether to proceed with thrombectomy in patients with large-core strokes.

NEURO CON: What are some other areas of future research?

AS: A few of the burning questions in current stroke management include:

  1. Whether thrombectomy is safe and efficacious in patients with unfavorable imaging profiles, or so-called large-core strokes.
  2. Whether both imaging modalities are needed to make a decision regarding thrombectomy or using only one is sufficient.
  3. Whether noncontrast CT can be utilized alone to identify thrombectomy candidates presenting beyond 6 hours after stroke onset.


We hope to shed light on these important questions in the ongoing SELECT2 randomized trial.

—Christina Vogt

Reference:

  1. Sarraj A, Hassan AE, Grotta J, et al. Optimizing patient selection for endovascular treatment in acute ischemic stroke (SELECT): a prospective, multicenter cohort study of imaging selection. Ann Neurol. 2020;87(3):419-433.  https://doi.org/10.1002/ana.25669
  2. Nogueira RG, Jadhav AP, Haussen DC, et al; DAWN Trial investigators. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct. N Engl J Med. 2018;378:11-21. doi:10.1056/NEJMoa1706442
  3. Albers GW, Marks MP, Kemps S, et al; DEFUSE 3 investigators. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging. N Engl J Med. 2018;378:708-718. doi:10.1056/NEJMoa1713973