Cerebrovascular Disease: Basic Designs, Sample Sizes and Pitfalls

Abstract

Cerebrovascular disease is a major public health problem; over the next 10 years approximately 15 million people will suffer an acute stroke in Europe and the United States [1]. There is a clear need for more effective treatments both to prevent stroke and to treat the acute event. There has, in response to this, been a substantial increase in interest in the design, methodology and analysis of randomised controlled trials in cerebrovascular disease in recent years [2,3]. Indeed, trials have changed clinical practice in the primary prevention of stroke, e.g. treatment of hypertension [4]; the secondary prevention of stroke, e.g. antiplatelet treatment [5], anticoagulation [6], and carotid endarterectomy [7,8]; rehabilitation following stroke, e.g. stroke units [8]; and in the treatment of acute stroke, e.g. Nimodipine in subarachnoid haemorrhage [9] and more recently aspirin [10,11] and thrombolytic therapy [12] in cerebral infarction. There are trials in progress of many other potentially promising treatments such as angioplasty for symptomatic carotid stenosis [13] and neuroprotective treatments for acute ischaemic stroke [3].