Perfusion imaging now selects patients for thrombolysis out to 24 hours

Design

PROSPERO-registered systematic review and meta-analysis of randomised controlled trials (CRD420261304314)

Population

Six trials, 1,182 patients given intravenous thrombolysis and 1,176 best medical treatment, presenting 4.5-24 hours after acute ischaemic stroke and selected by advanced perfusion imaging

Primary outcome

Excellent and good functional outcome at 3 months, with symptomatic intracranial haemorrhage and 90-day mortality as safety endpoints

Effect

Higher likelihood of excellent and good functional outcome, consistent to 24 hours in exploratory subgroups; increased symptomatic intracranial haemorrhage; no significant difference in 90-day mortality

The 4.5-hour thrombolysis window was set by time because time was the only variable available. Perfusion imaging replaces it with tissue status, and enough randomised trials have now reported to pool them. This meta-analysis included six trials comparing intravenous thrombolysis with best medical treatment in acute ischaemic stroke presenting in the extended window, all selecting patients on advanced perfusion criteria, predominantly CT perfusion: 1,182 treated and 1,176 controls.

Thrombolysis increased both excellent and good functional outcomes at three months. Exploratory subgroup analysis by treatment timing found the benefit consistent out to 24 hours. Symptomatic intracranial haemorrhage was more frequent with thrombolysis. All-cause mortality at 90 days did not differ between groups.

That combination - more independence, more bleeding, no excess deaths - is the familiar thrombolysis trade, now shown to hold when the selection is done by imaging rather than by clock. The reason it holds is that perfusion selection excludes the patients in whom the bleeding risk buys nothing.

The operational consequence lands on radiology rather than on stroke medicine. Extending eligibility to 24 hours means CT perfusion has to be available, processed and interpreted at any hour, and it means late-presenting patients who were previously triaged out now need urgent imaging. In Indian hospitals where perfusion software exists but out-of-hours processing does not, that gap is the thing that decides whether this evidence reaches patients. Note the subgroup analysis by timing was exploratory, so 24 hours is the outer edge of what these trials support rather than a validated threshold.

• Ensure CT perfusion acquisition and post-processing are available around the clock, not only in hours
• Do not exclude a late-presenting stroke patient from imaging on time alone
• Quote the increased symptomatic haemorrhage risk explicitly when consenting
• Treat the 24-hour boundary as exploratory - it comes from subgroup analysis
• Agree a local pathway for who reviews perfusion maps overnight before the pathway is needed

The statistics, in plain English

The pooled result reports direction and significance rather than a single effect size across both functional outcomes, so the honest summary is that thrombolysis improved the odds of independence without quantifying it here. That symptomatic haemorrhage rose while mortality did not is the crucial pairing: a treatment can cause more bleeds and still leave more people independent, because most of the additional bleeds are not fatal. The timing subgroups were exploratory, meaning they were examined after the fact and are hypothesis-generating - the trials were not individually powered to test a 24-hour cut-off.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.