1. Trial Identification

To prepare, maintain and disseminate large numbers of high quality systematic reviews it is necessary to identify and classify large numbers of potentially relevant trials in all languages. We are, therefore, most grateful to everyone who helps us with translation. For inclusion in our Specialised Register of Trials we do not need a full translation of non-English language reports, but only essential information to be extracted.

Randomised Controlled Trials (RCTs) and Controlled Clinical Trials (CCTs)

A relevant trial is any study performed in humans, comparing two or more interventions concurrently, where the study groups or interventions are definitely, probably or possibly assigned prospectively using a random or quasi-random method.

Eligible RCTs or CCTs must meet the following criteria:

1. A study in one or more humans
2. Comparison of two or more interventions. An intervention may be a drug, surgical or other procedure, physiotherapy, preventive intervention, medical education etc and control measures include placebo, active medication, different drug doses or no treatment.
3. Allocation to intervention is prospective using
   1. Random method (RCT) by random number table, computer generated list, toss of coin etc
   2. Quasi-random method (CCT) by alternate allocation, order of admission, date of birth, medical record number, day of week, ward of admission etc
   3. Possibly random method (CCT). In some research reports (especially abstracts) the authors may not describe in detail the method of allocation although a control group is included and a random or quasi-random method of allocation is implied. These trials (possible CCTs) should be included for further examination and possible clarification with authors.

Words and phrases that may suggest RCTs or CCTs

Special Notes

1. Patients may act as their own controls (ie crossover trials) where the ORDER of the interventions is randomly assigned or alternated.
2. Units of randomisation may be individuals, groups (eg communities or hospitals), organs (eg eyes) or other parts of the body (eg limbs).
3. Any report of a RCT or CCT should be included even when no results are given (eg the method or patient selection procedures may be published independently). A follow up study may also be made of the original study population and long term results reported.
4. "Random selection" usually indicates that some method of random sampling was used to select subjects for the study. It does not imply random allocation of the participants to the treatment or comparison groups.
5. Some articles which begin as a study on animals may subsequently include a study in humans. Please read to the end of the paper.

The following list gives examples of research which is not eligible for inclusion.

1. All uncontrolled research eg where all patients receive the same treatment (unless they are acting as their own controls)
2. Open clinical trial eg where the study physician or the patient decides on the treatment to be administered.
3. Historical controls eg where two treatment groups from different time periods are compared (eg a group from 1990 compared to a group from 1980).
4. Case/control studies. eg studies which compare a group of people with the disease under investigation to a control group without the disease. Controls may be healthy volunteers, patients with other illnesses etc.
5. Trials where the study groups are allocated according to patient characteristics. eg sex, age, severity of disorder, different aetiology (eg different types of epilepsy), geographic location, ability to conform with treatment or intervention under study, etc.
6. Case note reviews and retrospective studies (unless they include further results from a previous trial and comparative results are reported).
7. Studies exclusively in animals
8. Autopsy, tissue and cell studies

In general, if there is doubt about the relevancy of a particular trial please give as much information as you can.

We will also collect systematic reviews of interventions in stroke. These are relatively rare in the literature but describe reviews of clearly formulated questions using systematic and explicit methods to identify, select and critically appraise relevant research, and to collect and analyse data from the studies that are included in the review. Eligible systematic reviews should include a methods section detailing inclusion criteria and a search strategy for identification of trials. Statistical methods (meta-analysis) may be used to analyse and summarise the results of the included studies but this is not essential.

2. Scope of the Stroke Review Group

To be eligible for inclusion in our specialised register of stroke trials, articles should describe interventions in the primary or secondary prevention of stroke, or the treatment or rehabilitation of stroke patients, including those with subarachnoid haemorrhage. Primary prevention trials are restricted to those in which stroke is the major outcome of interest, eg:

1. Carotid surgery for asymptomatic stenosis.
2. Interventions for prevention of stroke in patients with atrial fibrillation
3. Interventions affecting the development of plaque in the carotid arteries. Outcome measures usually include ultrasound measurement of the carotid intima-media thickness.

The primary prevention of generalised vascular disease is beyond the scope of the Stroke Group, as is the primary prevention of stroke in patients with cardiac diseases, eg stroke following myocardial infarction/cardiac surgery.

The following terms are relevant to stroke:

• Stroke/cerebrovascular accident (CVA)
  
  • Ischaemic stroke
    
    • (Cerebral or cerebellar or brainstem or vertebrobasilar) infarction
    • (Cerebral or cerebellar or brainstem or vertebrobasilar) ischaemia
    • (Cerebral or vertebrobasilar) thrombosis
    • (Cerebral or vertebrobasilar) embolism
    • (Cerebral or vertebrobasilar) thromboembolism
  • Haemorrhagic stroke
    
    • (Cerebral or brainstem or cerebellar) haemorrhage or bleeding or haematoma
    • Intracerebral haemorrhage or bleeding or haematoma
    • (Supratentorial or infratentorial) haemorrhage or haematoma
• Transient ischaemic attacks (TIAs)
• Reversible ischaemic neurological deficit (RIND)
• Subarachnoid haemorrhage
• (Intracranial or intracerebral or cerebral) aneurysm
• (Intracranial or intracerebral or cerebral) arteriovenous or venous malformations (AVMs)
• Cerebral sinus venous thrombosis
• Intracranial venous thrombosis
• Carotid endarterectomy
  
  • Carotid artery surgery
  • Carotid artery occlusion
• Carotid artery dissection
  
  • Dissecting carotid artery aneurysm
• Hemiplegia (unless it is clear that some other pathology caused the hemiplegia, eg multiple sclerosis, head injury)
• Aphasia, dysphasia, dysphagia and hemiparesis due to stroke

The following terms are NOT relevant to stroke:

• Vascular dementia
• Multi-infarct dementia
• Chronic cerebrovascular insufficiency (unless it is specifically stated that the patients had suffered a stroke, TIA etc)