Approval to start
Hospital centres in IST-3 must have the approval of the national co-ordinator before applying for ethics approval. Appropriate local Ethics Committee approval must be sought for each participating hospital. Proof of such approval must be sent to the trial office before recruitment can be started in each centre. The trial must be run according to local procedures and law.
Trial centre requirements
A number of guidelines have stated thrombolysis should only be considered if the patient is admitted to a specialist centre with appropriate experience and expertise[22, 24]. Hospitals participating in IST-3 should have an organised acute stroke service. The components of effective stroke unit care have been identified, so the service should be configured along those lines and also meet local standards and guidelines. In brief, the facilities (details of these requirements are specified in the separate operations manual) should include:
Written protocol for the acute assessment of patients with suspected acute stroke to include interventions to reduce time from onset to treatment.
mmediate access to CT or MR brain scanning (preferably 24 hours a day).
A treatment area where thrombolysis may be administered and the patient monitored according to trial protocol, preferably an acute stroke unit.
Patients with mild, moderate or severe strokes are potentially eligible if the following criteria are met:
Symptoms and signs of clinically definite acute stroke.
Time of stroke onset is known and treatment can be started within six hours of this onset.
CT or MRI brain scanning has reliably excluded both intracranial haemorrhage and structural brain lesions which can mimic stroke (e.g cerebral tumour)
The patient has previously been randomised in IST-3
Major surgery, trauma (e.g. major fall at time of stroke) or gastrointestinal or urinary tract haemorrhage within the previous 21 days. Arterial puncture at a non-compressible site within the previous 7 days.
Any known defect in coagulation (e.g. currently on oral anticoagulants with an INR > 1.3 OR current treatment with heparin [unless APPT within normal laboratory limits] OR treatment with low molecular weight heparin or heparinoid OR treatment with ximelagatran).
Known defect of clotting or platelet function (but patients on antiplatelet agents can be randomised).
The patient is female and of childbearing potential (unless it is certain that pregnancy is not possible) or breast feeding.
Hypo- or hyperglycaemia sufficient to account for the neurological symptoms; the patient should be excluded if their blood glucose is < 3.0 or > 20.0 mmol/L ('stick testing' is a sufficiently accurate test for this purpose).
Symptoms considered likely to resolve completely within the next few hours (i.e. a TIA)
Patient has had a stroke within the previous 14 days or has had treatment for acute ischaemic stroke with thrombolytic therapy within the past 14 days.
Patient was already dependent in activities of daily living before the present acute stroke
Patient has other life threatening illness (e.g. advanced cancer) likely to lead to death within a few months.
Likely to be unavailable for follow-up e.g. no fixed home address.
Patient has Systolic Blood Pressure < 90 mm Hg or > 220 mm Hg or Diastolic Blood Pressure < 40 mm Hg or > 130 mm Hg
High blood pressure (BP) before randomisation
A persistently high blood pressure can be associated with a poor outcome after stroke, though high pre-treatment blood pressure was not an independent predictor of symptomatic intracranial haemorrhage with rt-PA. Some patients with high blood pressure (i.e. systolic BP > 185 mm Hg and/or diastolic > 110 m Hg) can therefore be treated with rt-PA. The randomisation system will only accept patients with systolic BP between 90–220 mm Hg and diastolic BP between 40–130 mm Hg. Although these data provide some guidance, the decision about whether or not to include a patient with persistently high levels of blood pressure in the trial must rest with the physicians' judgement.
Uncertainty principle (absence of proof)
Further inclusion and exclusion criteria are not specified precisely but are guided by the uncertainty principle (or absence of proof for that particular patient). If, for whatever reason, the clinician is convinced that a patient fulfilling the above criteria should be treated, the patient should be treated with rt-PA and NOT randomised. If the clinician is convinced that a patient should not be treated (for whatever reason), the patient should NOT be included in the trial. Patients should only be randomised if they fulfil the eligibility criteria AND the clinician is substantially uncertain about the balance of risks and benefits of rt-PA for that individual.
IST-3 will be run according to the standards laid out in the MRC Guidelines for Good Clinical Practice in Clinical Trials (United Kingdom) and in keeping with the EU directive on Clinical Trials. These guidelines are based on the ICH Harmonised Tripartite Guideline for Good Clinical Practice and the Declaration of Helsinki. Local Ethics Committee (or local equivalent) approval is needed for each participating centre before recruitment can begin. The consent process was developed, in line with recent recommendations, with consumer involvement. Consent is supported by a patient (or carer) information leaflet (Appendices 2 and 3) and is adapted to local ethical requirements and the clinical state of the patient:
If patients can understand and write, signed consent must be obtained.
Patients who can comprehend, but are unable to write, may provide verbal witnessed consent.
The patient's relative or spouse may act as the patient's personal legal representative and provide assent to trial inclusion (consent) if the patient is acutely mentally incompetent as a result of their stroke (e.g. aphasia or decreased conscious level).
Under certain strict criteria, if no relative is available, some local ethics committees may permit a professional legal representative, such as an independent doctor, to enable those patients unable to give consent to be recruited (this is acceptable in certain emergency situations and sometimes previously called 'a waiver of consent').
The requirements of the relevant ethics committee should be adhered to at all times.
All patients MUST have a pre-randomisation brain scan to exclude intracranial haemorrhage. CT scans should cover the entire brain from the foramen magnum to the vertex with 4 – 5 mm thick slices through the posterior fossa and 8 – 10 mm thick for the cerebral hemispheres, with no slice gap. Scans should be windowed on a width of 80 Hounsfield Units (HU) and a centre level of 35 – 40 HU. This is particularly important if scans are to be sent as printed film. All patients (irrespective of treatment allocation) MUST have a follow-up scan at 24–48 hours. In addition a repeat scan is required if the patient deteriorates neurologically or intracranial haemorrhage is suspected for any reason. Although CT scanning is preferred, MR brain imaging is allowed provided there is sufficient radiological support in the hospital to interpret the scans and a gradient echo (T2 *) is included to exclude haemorrhage (haemorrhage can be overlooked on several other types of MR imaging sequence) and Diffusion Weighted Imaging (DWI) is required to identify the recent infarct. All scans performed during the first 7 days following randomisation are to be sent to Edinburgh for coding. The two sets of CT scans per patient (more, if the patient had extra scans due to suspected complications) are to be sent to the Edinburgh trials office, either by post, or (subject to certain conditions) by electronic transfer of DICOM files (details of methods of file transfer and copies of the Scan transfer forms are given in the trial operations manual). If sending a hard copy film, the original is to be sent, as this allows better conversion to an electronic file (a copy should be made and kept at the treating hospital). Hard copy scans will be digitised and converted to DICOM files. All images will be coded with all original identifiers stripped from the record. Each scan can then be assessed, blind to patient details, and to whether the scan is pre- or post treatment. Each scan will be assessed by an international panel of expert radiologists by means of an internet web-based computer system.
Advanced imaging substudies
IST-3 will permit advanced imaging substudies in centres with appropriate facilities and local expertise. Such studies could include CT angiography, MR diffusion and perfusion imaging, carotid duplex and transcranial doppler imaging. Any such proposed sub-studies must be approved by the IST-3 Trial Steering Committee.
The clinician enters patients by telephone call to an automated randomisation system available 24 hours a day. The randomisation system requests a few key items of baseline data, which are then entered with the telephone keypad. A web-based randomisation is being planned for 2006. When the data have been entered and checked, the computer generates the treatment allocation. The system includes a standard minimisation algorithm which ensures that the treatment groups are balanced for key prognostic factors. The algorithm balances allocation on stroke severity (calculated as the patient's predicted probability of a poor outcome, calculated from a validated prognostic model based on key clinical variables measured at baseline). Patients allocated 'immediate rt-PA' should be treated as soon as possible after the randomisation call is completed.
All patients should have intravenous access in place and be administered intravenous fluid therapy according to local acute care protocols. Patients allocated 'immediate rt-PA' should be given recombinant tissue-type plasminogen activator (Alteplase, Boehringer Ingelheim; or Activase, Genentech) in a total dose of 0.9 mg per kg of body weight up to a maximum of 90 mg. Ten per cent of the dose is given as an intravenous bolus delivered over one minute followed by the rest of the infusion over the next 60 minutes. Patients allocated 'control' must avoid treatment with rt-PA and should receive stroke care in the same clinical environment as those allocated 'immediate rt-PA'. Both treatment groups must have their blood pressure monitored closely over the first 24 hours, according to the IST-3 protocol, and this must be documented. Both groups should receive the same general supportive care.
All patients entered in the trial, whether allocated rt-PA or control, must be managed according to local acute stroke care protocols, in the same clinical environment. Such protocols are not specified by the trial, but will generally include the components of effective stroke unit care. Soon after admission, intravenous access, monitoring of physiological variables, correction of any abnormalities, and where clinically appropriate, intravenous fluid therapy should be initiated.
Blood pressure: monitoring and intervention
The NINDS group specified a detailed protocol for the active lowering of blood pressure, though it was unclear whether this policy was beneficial or harmful to patients in the trial. The Blood Pressure in Acute Stroke Collaboration (BASC), have since reviewed all the relevant randomised controlled trials of blood pressure lowering in acute stroke and concluded (as did the International Society for Hypertension [ISH]  that there were no data from reliable randomised controlled trials to guide the management of high blood pressure in patients with acute stroke. Blood pressure tends to fall in the acute phase of stroke and in view of the conclusions of the BASC and ISH, no particular IST-3 protocol for blood pressure management will be specified. To monitor any interaction between blood pressure and response to treatment in IST-3, data on blood pressure levels and the use of blood pressure lowering treatments will be collected. This aspect of the trial will be monitored by the Data Monitoring Committee.
Symptomatic intracranial bleeding
Intracranial haemorrhage should be suspected if any of the following occur during the infusion or within 24 hours of randomisation:
If any of these events occur, any rt-PA infusion should be stopped and the patient examined for possible reasons for the deterioration. Blood should be taken to measure prothrombin time (PT), activated partial thromboplastin time (APPT), fibrinogen, full blood count and group and save serum. CT scanning must be performed immediately, irrespective of the allocated treatment group. If CT scanning confirms intracranial haemorrhage, rt-PA must not be restarted. Management should follow local protocols and will usually require consultation with a haematologist and a neurosurgeon. For patients who have received rt-PA there is no reliable evidence available to recommend any one treatment strategy over another, but fibrinolytic inhibitors such as tranexamic acid may be useful. In the rare instance that fibrinogen levels are low (<1 g/L) after rt-PA therapy, cryoprecipitate (containing fibrinogen and factor VIII) may be required. Fibrinogen assays vary but the Clauss technique is considered the best method if available.
Asymptomatic intracranial bleeding
If asymptomatic intracranial bleeding (haemorrhagic transformation of the infarct or parenchymatous haematoma) is detected on the repeat CT scan performed at about 24 hours after randomisation, no specific action is needed, but it may be necessary to delay the start of long-term antiplatelet or anticoagulant therapy. The degree of delay is a matter for the responsible clinician to determine, but will be influenced by factors such as the degree and extent of haemorrhage, the patient's clinical condition, the nature of the planned treatment and the indication for its use.
If significant extra-cranial bleeding develops, any rt-PA infusion must be stopped immediately. Blood should be taken to assess prothrombin time (PT), activated partial thromboplastin time (APPT), fibrinogen, full blood count and cross match. Appropriate supportive therapy should be given, with monitoring of blood pressure, maintenance of circulating blood volume with intravenous fluids and transfusion of blood as appropriate. The results of the investigations will guide emergency treatment. Management should follow local protocols and will usually require consultation with a haematologist. For patients who have received rt-PA there is no reliable evidence available to recommend any one treatment strategy over another, but fibrinolytic inhibitors such as tranexamic acid may be useful. If fibrinogen levels are low (<1 g/L) cryoprecipitate (containing fibrinogen and factor VIII) may be required. Fibrinogen assays vary but the Clauss technique is considered the best method if available.
Allergic or hypersensitivity reactions
Anaphylactic reactions can occur following rt-PA administration for acute ischaemic stroke, but occur rarely. If there are any signs of anaphylactic response or hypersensitivity (e.g. peri-orbital swelling, tongue swelling, urticarial rash) any rt-PA infusion should be stopped immediately. Patients require urgent medical assessment ('airway, breathing and circulation'). Treatment will depend on the severity of the reaction. Intravenous steroids and antihistamines may be sufficient for mild reactions. Adrenaline (nebulised or intramuscular) and intubation may be required for severe reactions. Local advice from the emergency medicine team should be sought. All such reactions should be recorded on the 7-day hospital form.
Other aspects of treatment
Antithrombotic treatment should not be given within the first 24 hours of the start of rt-PA treatment
There is some evidence to suggest that early aspirin, given with thrombolytic therapy, may increase the risk of death. Antithrombotic treatment (antiplatelet drugs and heparin) should therefore be avoided in the first 24 hours after start of trial treatment for patients who have received rt-PA. Patients treated with rt-PA should first have a repeat CT brain scan, performed at 24–48 hours after treatment, and start long-term antiplatelet therapy with aspirin or other agents, only if the second CT has excluded intracranial haemorrhage. Patients allocated control should start long-term antiplatelet therapy with aspirin (or other effective antiplatelet agent) after randomisation according to usual practice. There are no data to suggest that this delay in initiating antiplatelet drugs materially disadvantages rt-PA allocated patients. The modest benefit of aspirin, given at 24–48 hours after onset of stroke symptoms, was similar to that when given within the first few hours. Conversely, the earlier use of aspirin for patients allocated control is therefore unlikely to introduce a major difference between rt-PA and control groups and will anyway reflect usual clinical practice for control patients. All antithrombotic medication used in the first week following treatment will be recorded on the 7 day trial form to monitor deviation from the protocol and assess risk factors for side effects.
Long-term antiplatelet drugs
Unless there is a clear contraindication, all patients should be considered for long-term antiplatelet therapy with aspirin (or other effective antiplatelet) for routine secondary prevention of vascular events. Treatment should not be started until 24 hours after any rt-PA infusion (see above). At discharge, all patients will be given a card recording their participation in the study and their General Practitioners should be informed by letter.
Other treatments in hospital
All other aspects of treatment are at the discretion of the responsible clinician.
All patients will be followed up, whether they complied with their treatment or not. Follow-up will be at seven days, hospital discharge, transfer to another hospital or death, whichever occurs first. The Hospital Co-ordinator at each collaborating centre must complete the hospital follow-up form for each patient, and send it to the IST-3 Trial Office, enclosing a copy of the pre-and all post-randomisation CT scans.
Six months after randomisation, General Practitioners (or Hospital Co-ordinators) will be contacted by the IST-3 Trial Office staff to check that their patient is alive and that they may be approached for follow-up. If appropriate, the IST-3 Trial Office staff will then mail a postal questionnaire to patients, to record dependency and health related quality of life. The exact procedures for follow-up in each country will be decided by the National Co-ordinator and the IST-3 Management committee. Central follow-up (telephone or postal) has been found to be cost-effective, efficient and also ensures blinding of the assessment. If a patient dies after a hospital follow-up form has been completed (up to 7 days from randomisation), and within 6 months of randomisation, the clinician can conveniently inform the IST-3 Trial Office by completing and returning a simple form to reduce the risk of the co-ordinating centre mailing a questionnaire to a patient who has died. The precise date of death will be very important for survival analyses.
To assess the durability of any treatment benefit beyond 6 months, patients recruited in the UK (and in other countries where appropriate funding has been obtained) will be followed up one year after the six month assessment and annually thereafter (dependent on sufficient funding). These data will also permit more detailed health economic modelling and to test the hypothesis, that the level of disability at six months predicts long-term survival.
Patients may withdraw consent to participate in the trial at any stage. This may involve withdrawal from trial treatment (in which case any rt-PA infusion should be stopped) or withdrawal from trial follow-up. If the latter, it is essential to obtain their consent at the point of withdrawal to obtain follow-up information on their outcome from other sources, e.g. from their hospital records, their general practitioner or central health services data.