Too good to treat? Ischemic stroke patients with small computed tomography perfusion lesions may not benefit from thrombolysis

Abstract

OBJECTIVE: Although commonly used in clinical practice, there remains much uncertainty about whether perfusion computed tomography (CTP) should be used to select stroke patients for acute reperfusion therapy. In this study, we tested the hypothesis that a small acute perfusion lesion predicts good clinical outcome regardless of thrombolysis administration. METHODS: We used a prospectively collected cohort of acute ischemic stroke patients being assessed for treatment with IV-alteplase, who had CTP before a treatment decision. Volumetric CTP was retrospectively analyded to identify patients with a small perfusion lesion (<15ml in volume). The primary analysis was excellent 3-month outcome in patients with a small perfusion lesion who were treated with alteplase compared to those who were not treated. RESULTS: Of 1526 patients, 366 had a perfusion lesion <15ml and were clinically eligible for alteplase (212 being treated and 154 not treated). Median acute National Institutes of Health Stroke Scale score was 8 in each group. Of the 366 patients with a small perfusion lesion, 227 (62%) were modified Rankin Scale (mRS) 0 to 1 at day 90. Alteplase-treated patients were less likely to achieve 90-day mRS 0 to 1 (57%) than untreated patients (69%; relative risk [RR] = 0.83; 95% confidence interval [CI], 0.71-0.97; p = 0.022) and did not have different rates of mRS 0 to 2 (72% treated patients vs 77% untreated; RR, 0.93; 95% CI, 0.82-1.95; p = 0.23). INTERPRETATION: This large observational cohort suggests that a portion of ischemic stroke patients clinically eligible for alteplase therapy with a small perfusion lesion have a good natural history and may not benefit from treatment.