Antithrombotic drugs and ischaemic stroke.

About 80% of strokes are ischaemic. Approximately 12% of patients die within 3 months following stroke, and another 20% are institutionalised or become highly dependent. In early 2013, what is the harm-benefit balance of antithrombotic treatments used in the acute phase of ischaemic stroke? To answer this question, we reviewed the available data using the standard Prescrire methodology. Clinical trials of aspirin in the acute phase of ischaemic stroke consist mainly of two randomised trials including a total of 40 541 patients. After 1 to 6 months, 13 deaths or sequelae resulting in dependence are prevented when 1000 patients are treated with aspirin during the acute phase. Aspirin increases the risk of symptomatic intracranial haemorrhage when it is introduced less than 24 hours after treatment with alteplase. Abciximab, an injectable antiplatelet agent, showed no tangible clinical benefit in 5 randomised placebo-controlled trials in a total of 1275 patients. Clopidogrel and prasugrel, two other antiplatelet agents, have not been evaluated in this setting. However, in case of allergy to aspirin, clopidogrel is a useful alternative in other situations associated with a risk of arterial ischaemia. In a randomised trial including 458 patients, cilostazol and aspirin were similarly effective after 3 months of follow-up, but cilostazol caused cardiac arrhythmias. Ticlopidine has too many adverse effects to consider it a useful drug. Anticoagulant therapy during the acute phase of stroke has an unfavourable harm-benefit balance, including in patients with stroke secondary to cardiac embolism. Low-molecular-weight heparin reduces the risk of pulmonary embolism but has no impact on overall mortality. The aim of thrombolysis is to unclog the affected artery. Intravenous alteplase administration is the best-assessed thrombolytic method. Twelve randomised trials have compared intravenous thrombolysis with alteplase versus no thrombolytic therapy in 7012 patients. Among patients treated within 3 hours after stroke onset, 41% survived without sequelae after alteplase administration, versus 32% in the absence of thrombolysis; alteplase had no statistically significant impact on mortality at the end of follow-up. Efficacy appeared to be similar in patients over 80 years old.The harm-benefit balance may also be favourable when thrombolysis is started more than 3 hours after stroke onset, but when it is initiated more than 4.5 hours after stroke onset, it increases mortality. Four randomised trials showed that intra-arterial thrombolysis with urokinase or pro-urokinase had a beneficial effect, versus no thrombolysis, in a total of 356 patients. In a randomised trial including 362 patients, intra-arterial thrombolysis did not have a better harm-benefit balance than intravenous thrombolysis. Among the various physical and mechanical methods used to remove or dissolve clots, the best-evaluated is ultrasound plus intravenous alteplase thrombolysis; initial results with this procedure warrant further clinical trials. Rapid intervention and patient selection are both crucial in optimising the harm-benefit balance of intravenous alteplase thrombolysis. This treatment should only be used when management begins within the first hours following stroke symptom onset, and when there are no risk factors for bleeding, especially intracranial bleeding. For other patients, aspirin is the only antithrombotic drug known to reduce, albeit only slightly, the risk of death and sequelae following ischaemic stroke.