THE EVIDENCE FOR THROMBOLYSIS

In 1995 the National Institute of Neurological Disorders and Stroke (NINDS) rt-PA stroke study demonstrated the efficacy of recombinant tissue type plasminogen activator (rt-PA) for the treatment of acute ischaemic stroke (AIS) when administered within three hours of symptom onset.¹ This randomised controlled trial of 624 patients reported a 13% absolute increase in favourable outcome at three months (defined as a modified Rankin Scale score (mRS) 0–1). The number of patients needed to treat (NNT) to result in one additional favourable outcome over placebo was seven. rt-PA was associated with an increased absolute risk of symptomatic intracerebral haemorrhage (SICH) (6% v 1% for placebo) but there was no significant increase in mortality. Intravenous rt-PA was approved for use in stroke by the Food and Drug Administration in the United States the following year. In 1999 the Therapeutics Product Programme in Canada provisionally approved the use of rt-PA on the condition that a registry of treated cases was maintained.² This was the first licensed treatment for AIS and was embraced with enthusiastic optimism by a number of stroke centres, particularly those who had direct experience with rt-PA in the NINDS trial. In September 2002 the European Agency for the Evaluation of Medicinal Products licensed rt-PA on the proviso that a further randomised trial be conducted for treatment three to four hours post-symptom onset (ECASS III) and a European registry is maintained (Safe Implementation of Thrombolysis in Stroke Monitoring System, SITS-MOST; www.emea.eu.int/pdfs/human/referral/334602en.pdf).

Meta-analysis of the current evidence for thrombolysis supports the findings of the NINDS study,³ however the recanalisation rate after intravenous therapy for proximal middle cerebral artery occlusions is probably only 30%.⁴ In an attempt to improve on this recanalisation rate and extend the time window for therapy, trials of intra-arterial (IA) thrombolysis were undertaken. PROACT II was a randomised controlled trial of IA pro-urokinase in 180 patients with angiographically proven middle cerebral artery occlusion and severe stroke (average National Institute of Stroke Scale score of 17).⁵ In spite of a mean time of 5.3 hours from stroke onset to initiation of IA pro-urokinase, this trial demonstrated a 15% absolute increase in favourable outcomes in the treatment group (mRS 0–2). There was an increased risk of SICH (10% v 2%) but no increase in mortality. A number of case series of IA and combined IV and IA rt-PA followed and recent meta-analysis of IA thrombolysis suggests an 18.5% absolute and 80% relative increase in favorable outcomes, NNT of six and reduced mortality in the IA thrombolysis group with initiation of therapy up to six hours from symptom onset.⁶

To put this in perspective, preventive therapies such as aspirin administered following stroke provide an absolute risk reduction (ARR) of 0.9% for early recurrence or death with NNT of 111.⁷ Warfarin in older patients with uncomplicated atrial fibrillation offers an ARR of 1.5% per year compared with aspirin.⁸ The benefits and risks of rt-PA are very similar to carotid endarterectomy for symptomatic high grade carotid stenosis (> 70%) in which the North American Symptomatic Carotid Endarterectomy Trial (NASCET) demonstrated a 17% ARR of ipsilateral stroke over two years, with a NNT of six despite a 5.6% perioperative risk of stroke or death.⁹ Like surgery and warfarin, rt-PA is potentially dangerous. However, there is compelling evidence that when used in accordance with present guidelines, under the direction of an experienced stroke team, rt-PA is a powerful and relatively safe therapeutic agent whose impact on acute ischaemic stroke can be dramatic. A number of Canadian, American, and European stroke centers are now thrombolysing in excess of 10% of all ischaemic strokes.

WHAT HAVE WE LEARNED?

A number of conclusions can be drawn from the past decade of stroke research. First and foremost, in AIS with demonstrated arterial occlusion the only evidence-based approach to improve outcome acutely is disruption of thrombus and reperfusion of the zone of ischaemia.

The overriding priority remains patient safety, but with AIS the order of the day is urgency in administering the thrombolytic. The goal is not just treatment in under three hours but treatment as soon as possible. There is a significant decline in the probability of favourable outcome for every additional 10–15 minutes delay in the initiation of reperfusion therapy.¹⁰ This urgent time frame for investigation and treatment has had the beneficial spin-off of producing a wealth of new information regarding the evolution of AIS.

Multidisciplinary acute stroke teams can be developed and organised to function quickly and efficiently.¹¹ Fifty percent of the patients in the NINDS trial were assessed clinically, CT scanned, randomised, and had treatment started within 90 minutes of symptom onset. In experienced stroke centres door to needle times of 30 minutes are becoming routine in uncomplicated cases and these results are being achieved in public healthcare systems, such as exists in Canada. This requires recruitment, training, and goal directed restructuring of existing resources and personnel.

One size does not fit all. Optimal stroke therapy must be tailored to specific stroke subtypes. Rapid identification of the mechanism of stroke, the site of vascular involvement, and the assessment of tissue viability requires appropriate resources and organisation to conduct CT and/or MR imaging on an emergency basis. Twenty four hour, seven days a week availability of emergency CT, and expertise in neuroradiology and interventional neuroradiology are essential. Increasingly sophisticated imaging techniques are being developed to help guide decision making on the basis of a “tissue window” and extend treatment options beyond the current “time window”.^12–13

All stroke patients benefit from urgent investigation and initiation of treatment, not just those receiving rt-PA. Evidence indicates that care on a stroke ward, under the direction of a stroke team with early and aggressive rehabilitation, improves outcome of all forms of stroke whether they are rt-PA candidates or not.¹⁴

A CALL FOR CHANGE

“Much has been done but when we look ahead at what remains we see only a beginning has been made” (Sir Wm. Osler, 1907). Thrombolytic therapy has sparked a revolution in acute stroke care but rt-PA is not the final answer. rt-PA will eventually be replaced by combinations of pharmacological and mechanical techniques of thrombolysis with improved efficacy and risk/benefit ratio. The presently limited time window for reperfusion therapy will also be extended by combinations of neuro-protective agents and interventions such as hypothermia, increasing the number of patients that can be thrombolysed. Numerous questions remain but in the clinical arena the principle task is to close the gap between what is already known and what is being translated into clinical practice.

Stroke is the leading cause of long term disability in the adult population in Britain, with stroke patients occupying 20% of hospital beds.¹⁵ It is time to give the same priority to the training of multidisciplinary stroke teams and development of stroke units as was given to coronary care units and the management of myocardial infarction. Governments, health trusts, and funding agencies need to made aware of the critical dependence of modern stroke care on these elements and on emergency access to CT and MR scanning. The potential savings, from reductions in the number of patients requiring lengthy hospitalisations and long term care, quickly offset the capital costs.^16–17

CONCLUSION

The call for a change in British stroke care is growing^18–19 and what is needed now is a National Service Framework for Stroke (see www.doh.gov.uk/nsf). Debate over evidence of the efficacy of rt-PA should not distract us. The critical issue is to create the necessary infrastructure of acute stroke teams, stroke units, and neuroradiology facilities to deliver a new standard of stroke care. Acute stroke treatments will continue to be improved and refined, but without organised acute stroke care, patients in Britain will be denied benefit from these developments.