Thrombolysis for stroke

Cerebrovascular disease is the third leading cause of disease burden in developed nations, and is predicted to be the fourth ranked disease burden worldwide by 2030 after unipolar depressive disorders, ischaemic heart disease and trauma.1 All of these conditions are characterised by sudden and unpredictable demands requiring an immediately accessible, systemised and multidisciplinary approach to care. The complexities of acute ischaemic stroke in Australia have been addressed by detailed clinical guidelines.2 An emergency care bundle for stroke and transient ischaemic attack has recently been offered by the National Institute of Clinical Studies of the National Health and Medical Research Council.3

When administered to appropriate patients within 3 hours of stroke symptom onset, the benefits of recombinant tissue plasminogen activator (rt-PA) are significant, with treated patients 30% more likely to be in the excellent outcome grade — an absolute increase of 13%4 — and improvements in modified Rankin scores for some other patients with higher modified Rankin scores. The number of patients needed to treat for benefit may be as low as three.5 The associated risk of an intracerebral haematoma causing deterioration is about one in 30.5 The third European Cooperative Acute Stroke Study (ECASS3), a randomised trial of intravenous rt-PA in the 3–4.5 hour window, demonstrated a smaller but statistically significant benefit with no increase in haematoma rate.6 A recent Cochrane review of 26 thrombolysis trials of rt-PA, streptokinase, desmoteplase, urokinase and pro-urokinase, which included 7125 patients, found a significant net benefit in terms of death and dependency.7

The clinical applicability of new therapies may be exaggerated by the Hawthorne effect of clinical trials. The European Safe Implementation of Thrombolysis in Stroke Monitoring Study (SITS-MOST) registry — a mandated requirement of European drug licensing authorities — was established to monitor thrombolysis in day-to-day clinical practice.8 It included centres not experienced with thrombolysis and demonstrated the feasibility and safety of thrombolytic therapy across Europe. The cover of the issue of The Lancet in which the registry outcomes were published declared that rt-PA is “safe and effective in routine clinical use”. The rate of symptomatic intracerebral haemorrhage (ICH), as defined by the National Institute of Neurological Disorders and Stroke (NINDS), was 7.3% in SITS-MOST for both experienced and new thrombolysis centres, with a calculated mortality rate from ICH at 3 months of 1.9%.8 Early deterioration due to ICH in the SITS-MOST registry occurred in 1.7% of cases.

In this issue of the Journal, Simpson and colleagues report the Australian contribution to the Safe Implementation of Thrombolysis in Stroke International Stroke Thrombolysis Register (SITS-ISTR),9 reflecting the local experience of treating acute stroke 15 years on from the NINDS trial.4 Participation in the Australian component of the SITS-ISTR was voluntary. Many centres undertaking thrombolysis did not participate, and this may weaken the generalisability of this new data. Nevertheless, over 500 patients were enrolled and outcomes did not differ from those of the larger international database. The important safety data were reassuring, with a symptomatic ICH rate of 8.3% by the definition used in the NINDS randomised controlled trial, and 1.3% by the SITS-MOST definition. The 3-month ICH mortality rate was 2.2%. A British subset of SITS has been reported recently, with outcomes also comparable to those for the rest of Europe.10

In the SITS-MOST registry, new and experienced centres did not differ in terms of their haemorrhagic complication rates.8 While the Australian report by Simpson et al does not detail information regarding the types of centres involved, implementation of thrombolysis in Australia beyond the centres that participated in the thrombolysis trials is already well established. Audits by the National Stroke Foundation have found that 33 hospitals were regularly treating with rt-PA in 2007,11 and that this increased to 50 by 2009.12 New metropolitan and rural centres can adopt thrombolysis with executive support and leadership from medical and nursing “stroke champions”. This nearly always results in the establishment of a stroke unit and the adoption of a local thrombolysis protocol with coordination of the prehospital emergency services. Mentorships with established metropolitan centres are worthwhile in the early stages. Some direct links using telemedicine for the treatment of the first cases have been employed in Victoria (Associate Professor Bernard Yan, Neurologist and Neurointerventionist, Royal Melbourne Hospital, personal communication).

Early recognition and intervention for stroke requires a tightly coordinated interdisciplinary approach and rates of intravenous thrombolysis administration can be used as a clinical quality indicator for stroke care.13 Ongoing participation in the SITS-ISTR and the Australian Stroke Clinical Registry is crucial for monitoring the progress of this important therapy.

A coordinated system of care for stroke and transient ischaemic attack in Australia has been stalled by the lack of a concerted effort to adopt thrombolysis. Australian registry data provide reassurance that Australian stroke physicians, emergency physicians and systems that support thrombolysis can achieve similar results to those recorded in Europe. We can now move beyond discussing the efficacy and feasibility of implementing this therapy and work toward a more coordinated system of applying the evidence.