BACKGROUND

Thrombolytic therapy is effective for acute myocardial infarction, a disease with some similarities to acute ischaemic stroke. Meta-analyses suggest a net benefit in acute ischaemic stroke.

OBJECTIVES

To assess different thrombolytic agents, and different regimens, in acute ischaemic stroke.

SEARCH STRATEGY

We searched the Cochrane Stroke Group trials register (last searched to June 2003), MEDLINE (1966 to July 2003) and EMBASE (1980 to July 2003). We handsearched four Japanese journals, contacted researchers and pharmaceutical companies, and attended relevant conferences.

SELECTION CRITERIA

Randomised and quasi-randomised trials of different doses of a thrombolytic agent, or different agents, or the same agent given by different routes, in people with confirmed acute ischaemic stroke.

DATA COLLECTION AND ANALYSIS

Two reviewers independently assessed trial eligibility and quality, and extracted the data.

MAIN RESULTS

Ten trials involving 1641 patients, 8 conducted in Japan, 1 in China and 1 in the USA, were included. Concealment of allocation was poorly described. Different doses (of tissue plasminogen activator or urokinase) were compared in seven trials (n = 1072 patients). Different agents (tissue plasminogen activator versus urokinase; tissue-cultured urokinase versus conventional urokinase) were compared in three trials (n = 688 patients). One trial compared different routes of administration (intravenous plus intraarterial tissue plasminogen activator versus intraarterial tissue plasminogen activator alone, n = 35 patients). As some trials compared different agents and different doses, some patients contributed to two analyses. A higher dose of thrombolytic therapy was associated with a three-fold increase in fatal intracranial haemorrhages (Odds ratio (OR) 3.25, 95% confidence interval (CI) 1.32 to 7.97) compared with a lower dose of the same agent (based on 16 events among 539 higher-dose patients and 4 events among 533 lower-dose patients in 7 trials). There was no statistically significant difference in early (OR 1.01, 95% CI 0.58 to 1.74) or late (OR 0.94, 95% CI 0.58 to 1.53) deaths between lower and higher doses. Data were inadequate to assess the effect of dose on functional outcome. No statistically significant difference was shown between different thrombolytic agents tested. The data from the pilot trial comparing different routes of administration were inconclusive.

REVIEWERS' CONCLUSIONS: These scant data suggest that higher doses of thrombolytic agents may lead to higher rates of bleeding. However, the evidence is inadequate to conclude whether lower doses of thrombolytic agents are more effective than higher doses, or whether one agent is better than another, or which route of administration is the best, in acute ischaemic stroke.