Effect of alteplase on the CT hyperdense artery sign and outcome after ischemic stroke

Abstract

Objective: To investigate whether the location and extent of the CT hyperdense artery sign (HAS) at presentation affects response to IV alteplase in the randomized controlled Third International Stroke Trial (IST-3).

Methods: All prerandomization and follow-up (24-48 hours) CT brain scans in IST-3 were assessed for HAS presence, location, and extent by masked raters. We assessed whether HAS grew, persisted, shrank, or disappeared at follow-up, the association with 6-month functional outcome, and effect of alteplase. IST-3 is registered (ISRCTN25765518).

Results: HAS presence (vs absence) independently predicted poor 6-month outcome (increased Oxford Handicap Scale [OHS]) on adjusted ordinal regression analysis (odds ratio [OR] 0.66, p < 0.001). Outcome was worse in patients with more (vs less) extensive HAS (OR 0.61, p = 0.027) but not in proximal (vs distal) HAS (p = 0.420). Increasing age was associated with more HAS growth at follow-up (OR 1.01, p = 0.013). Treatment with alteplase increased HAS shrinkage/disappearance at follow-up (OR 0.77, p = 0.006). There was no significant difference in HAS shrinkage with alteplase in proximal (vs distal) or more (vs less) extensive HAS (p = 0.516 and p = 0.580, respectively). There was no interaction between presence vs absence of HAS and benefit of alteplase on 6-month OHS (p = 0.167).

Conclusions: IV alteplase promotes measurable reduction in HAS regardless of HAS location or extent. Alteplase increased independence at 6 months in patients with and without HAS.

Classification of evidence: This study provides Class I evidence that for patients within 6 hours of ischemic stroke with a CT hyperdense artery sign, IV alteplase reduced intra-arterial hyperdense thrombus.

Figure 1. Separate ordinal regression analyses show odds ratios for the effect of alteplase treatment on 6-month functional outcome in the full group (n = 2,961) and in the subgroups with and without a hyperdense artery sign (HAS) on prerandomization CT

Odds ratio >1 (right of line) indicates better outcome (lower 6-month Oxford Handicap Scale). Results for the HAS and no HAS groups are adjusted for the effect of age, time from stroke onset to scan (hours), and NIH Stroke Scale (NIHSS) score. Results for the full group are adjusted for the effect of age, time from stroke onset to scan (hours), NIHSS, and presence/absence of HAS.

Figure 2. Ordinal regression analyses for the effect of treatment on change in HAS extent from prerandomization to follow-up scan

Ordinal regression analyses with change in HAS segment number (shrinkage, no change, growth) as the dependent variable assessing the importance of prerandomization HAS location (proximal = internal carotid artery, middle cerebral artery mainstem, vertebral or basilar arteries; distal = anterior or posterior cerebral arteries or sylvian branches of the middle cerebral artery) and HAS extent on the effect of alteplase. *Location analysis does not include patients with HAS in both proximal and distal arteries. Odds ratio <1 (left of line) indicates HAS shrinkage, >1 (right of line) indicates HAS growth. Results are adjusted for the effect of age, time from stroke onset to scan (hours), and NIH Stroke Scale score.