Reperfusion Therapy – Evidence behind the Guidelines

Multiple studies have been carried out on thrombolysis for stroke, each with strengths and flaws. The seminal papers include:

NINDS [23]

This pioneering paper used four primary outcome measures of stroke recovery to assess whether treatment with tPA (tissue plasminogen activator) resulted in clinical benefit at three months in 300 patients. The outcome measures used were scores on the Barthel index, modified Rankin scale, Glasgow outcome scale, and NIHSS. At 3 months there was a 12% absolute difference in the Modified Rankin Score in patients treated with tPA by 3 hours. Symptomatic intracerebral haemorrhage within 36 hours occurred in 6.4% of patients given tPA but only 0.6% of patients given placebo (P <0.001). Mortality at three months was 17% in the tPA group and 21% in the placebo group (P = 0.30).

ECASS-I [24]

Patients were randomised to treatment with tPA or placebo within 6 hours from the onset of symptoms. Primary end points included Barthel Index (BI) and modified Rankin Scale (RS) at 90 days. The trial showed a 6% absolute improvement in modified Rankin Scale, but at the cost of an increase in mortality with tPA (22% vs 15% placebo) and an increase in intracranial haemorrhage.


Patients were again randomised to treatment with tPA or placebo within 6 hours from the onset of symptoms with changes to blood pressure control, CT criteria and randomisation. There was no statistical difference in modified Rankin Score (the primary end point) or mortality, but again, intracranial haemorrhage was increased in the tPA arm (8.8% vs 3.4%).


800 patients were randomised to receive tPA between 3 and 4.5 hours after their stroke, with a primary endpoint of disability at 90 days assessed with modified Rankin scores. More patients had a favourable outcome with alteplase (52.4% vs 45.2%) with no significant mortality difference, but with a higher rate of symptomatic intracranial haemorrhage (2.4% vs. 0.2%, P = 0.008).

IST-3 [26]

A large international multicentre open label randomised control trial, it took 10 years to recruit 3000 patients, many of them outside the existing European licence for alteplase at the time. The study did not reach its primary endpoint: it found a non-significant 2% benefit from tPA (alive and independent, as defined by an Oxford Handicap Score) at 6 months. There was an increase in significant intracranial haemorrhage (7% vs 1%) and early deaths but no difference in overall mortality at 6 months. It suggested that more elderly patients could benefit from thrombolysis than previously thought