Association between brain imaging signs, early and late outcomes, and response to intravenous alteplase after acute ischaemic stroke in the third International Stroke Trial (IST-3): secondary analysis of a randomised controlled trial

Summary Background Brain scans are essential to exclude haemorrhage in patients with suspected acute ischaemic stroke before treatment with alteplase. However, patients with early ischaemic signs could be at increased risk of haemorrhage after alteplase treatment, and little information is available about whether pre-existing structural signs, which are common in older patients, affect response to alteplase. We aimed to investigate the association between imaging signs on brain CT and outcomes after alteplase. Methods IST-3 was a multicentre, randomised controlled trial of intravenous alteplase (0·9 mg/kg) versus control within 6 h of acute ischaemic stroke. The primary outcome was independence at 6 months (defined as an Oxford Handicap Scale [OHS] score of 0–2). 3035 patients were enrolled to IST-3 and underwent prerandomisation brain CT. Experts who were unaware of the random allocation assessed scans for early signs of ischaemia (tissue hypoattenuation, infarct extent, swelling, and hyperattenuated artery) and pre-existing signs (old infarct, leukoaraiosis, and atrophy). In this prespecified analysis, we assessed interactions between these imaging signs, symptomatic intracranial haemorrhage (a secondary outcome in IST-3) and independence at 6 months, and alteplase, adjusting for age, National Institutes of Health Stroke Scale (NIHSS) score, and time to randomisation. This trial is registered at ISRCTN.com , number ISRCTN25765518. Findings 3017 patients were assessed in this analysis, of whom 1507 were allocated alteplase and 1510 were assigned control. A reduction in independence was predicted by tissue hypoattenuation (odds ratio 0·66, 95% CI 0·55–0·81), large lesion (0·51, 0·38–0·68), swelling (0·59, 0·46–0·75), hyperattenuated artery (0·59, 0·47–0·75), atrophy (0·74, 0·59–0·94), and leukoaraiosis (0·72, 0·59–0·87). Symptomatic intracranial haemorrhage was predicted by old infarct (odds ratio 1·72, 95% CI 1·18–2·51), tissue hypoattenuation (1·54, 1·04–2·27), and hyperattenuated artery (1·54, 1·03–2·29). Some combinations of signs increased the absolute risk of symptomatic intracranial haemorrhage (eg, both old infarct and hyperattenuated artery, excess with alteplase 13·8%, 95% CI 6·9–20·7; both signs absent, excess 3·2%, 1·4–5·1). However, no imaging findings—individually or combined—modified the effect of alteplase on independence or symptomatic intracranial haemorrhage. Interpretation Some early ischaemic and pre-existing signs were associated with