Those who have strokes with favorable imaging profiles have a higher likelihood of receiving endovascular thrombectomy as well as achieving high functional independence rates after 90 days.
Amrou Sarraj, MD
Patients who experience stroke who have favorable imaging profiles on non‐contrast computed tomography (CT) and CT perfusion have a higher likelihood of both receiving endovascular thrombectomy (EVT)
and high functional independence rates, according to new phase 2 study results (NCT02446587).
Additionally, the study, conducted by Amrou Sarraj, MD, associate professor of neurology, UT McGovern Medical School, and colleagues, showed that those who have discordant profiles—unfavorable imaging profiles or unfavorable CT perfusion—were still able to achieve reasonable functional independence rates.
However, those with only poor CT perfusion but favorable CT had more adverse outcomes, with rates of symptomatic intracranial hemorrhage and mortality rates of 24% (P
= .26) and 53% (P
= .72), respectively, though they were not significantly higher.
“The primary imaging modalities used to select patients for EVT are non‐contrast CT and CT‐perfusion,” Sarraj and colleagues wrote. “However, their relative utility is uncertain. We prospectively assessed CT and CTP concordance/discordance and correlated the imaging profiles on both with EVT treatment decisions and clinical outcomes.”
All told, Sarrai and colleagues’ efforts included 361 patients, of whom 79% (n = 285) received EVT. Of those, 87% had favorable CTs, while 91% had favorable CT perfusions, and 81% had favorable profiles on both. All told, 16% of the group had discordant results and 3% had unfavorable results for both CT and CT perfusion.
Ultimately, the adjusted odds ratio (OR) of receiving EVT with a favorable CT and CT perfusion compared to favorable CT with unfavorable CT perfusion was 5.99 (95% CI, 2.25–15.97; P
<.001), and was 3.82 (95% CI, 1.69–8.63; P
= .001) compared to unfavorable CT with favorable CT perfusion.
When patients had favorable imaging profiles on both modalities, the odds of receiving EVT compared to discordant profiles rose significantly (adjusted OR, 3.97; 95% CI, 1.97–8.01; P
<.001). All told, 58% of those with favorable profiles on both achieved functional independence compared to 38% of those with discordant profiles and 0% with both being unfavorable (P
<.001 for trend).
In all, 67 patients had discordant imaging profiles, of which 67% (n = 45) received EVT while 33% (n = 22) received only medical management. Functional independence was more likely to be high with EVT, occurring so in 38% (n = 17) of patients, compared to 18% (n = 4), though it was not statistically significant (P
Successful reperfusion was achieved in 80% (n = 36) of the discordant profile patients (favorable CT/unfavorable CT perfusion: 82%, 14 of 17; unfavorable CT/favorable CT perfusion: 79%, 22 of 28). Patients with successful reperfusion also achieved higher functional independence rates (42%) compared to those without successful reperfusion (22%; P
There was a correlation between 90-functional independence rates, as measured by the modified Rankin Scale (mRS) score, and favorable profiles on both CT and CT perfusion modalities. In total, the probability of good functional independence was 56% with a favorable CT and 57% with favorable CT perfusion (adjusted OR, 1.91; 95% CI, 0.40–9.0; P
“We examined if these findings would differ between those who were treated early versus late,” Sarrai and colleagues. “Similar outcomes were seen between favorable CTs and favorable CT perfusions when the results were dichotomized by the time window. The high overlap between the favorable profiles on CT and CT perfusion probably explains the similar outcomes when the 2 modalities are compared. Since both imaging studies were available to the physicians, it’s possible that findings on both modalities may have affected treatment decisions.”
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. Annals Neurol. Published online January 9, 2020. doi: 10.1002/ana.25669.