At least 795,000 people suffer a new or recurrent stroke each year in the US.1 Every 40 seconds, someone in the US has a stroke, and stroke is the third leading cause of death in the US.1 These numbers are likely to increase as the US population ages and as stroke awareness increases. In the past, ischaemic stroke has been managed with palliation and rehabilitation and with prevention of the causative entity as the focus. However improvements in physiological imaging (computed tomography [CT] angiography and CT perfusion) and the evolution of endovascular technology have led to a movement towards acute stroke intervention to treat stroke as an emergency, with the goal of revascularisation to save the brain’s ischaemic penumbra.
Much has been learned from cardiology in that the sooner blood flow can be restored to the brain (as in the heart), the better the chance of a clinical recovery. Many centres have the capability to infuse intravenous (IV) thrombolytics, such as tissue plasminogen activator (t-PA), as this treatment is often provided by neurologists, emergency room physicians or internists. Data suggest that the technical success rate of IV t-PA in achieving revascularisation ranges from 35 to 52%, depending on the degree and location of the initial arterial occlusion; however, this therapy can be associated with a reocclusion rate of up to 17%.2–5
Data also suggest that endovascular stroke intervention may be even more effective than IV thrombolytic therapy. The advantages of the intra-arterial (IA) approach for thrombolysis are multiple. First, angiography allows for localisation of the arterial occlusion and allows visualisation of possible recanalisation. In addition, direct delivery of the thrombolytic agent (i.e. t-PA or reteplase) to the occlusion is accomplished with visualisation of clot dissolution. The Prolyse in acute cerebral thromboembolism (PROACT) studies evaluated IA thrombolysis with recombinant prourokinase (rpro-UK) in patients within six hours of a middle cerebral artery (M1 or M2 segment) occlusion stroke.6,7 The PROACT-II study (a phase III prospective, randomised, placebo-controlled study) enrolled 180 patients with a median National Institutes of Health Stroke Scale score of 17 (range four to 30), comparing IA rpro-UK and low-dose heparin with low-dose heparin alone.7 A favourable outcome (modified Rankin scale score of 0–2 at 90 days) was achieved in 40% of patients treated with IA rpro-UK versus only 25% of control subjects, and the recanalisation rate was significantly higher in the rpro-UK group (66%) than in the control group (18%; p<0.001). Although the rate of symptomatic intracerebral haemorrhage was higher in rpro-UK patients (10%) than in control patients (2%; p=0.06), no difference in mortality rate was observed.
Finally, data for mechanical thrombectomy look very promising for the treatment and reversal of stroke caused by large arterial occlusions. Two devices approved by the US Food and Drug Administration are available specifically for mechanical thrombectomy: the Merci retriever (Concentric Medical, Mountain View, CA, US) and the Penumbra aspiration device (Penumbra, Inc., Alameda, CA, US). Both devices are designed for thrombectomy in carefully selected acute stroke patients with large-vessel intracranial occlusions. The Mechanical embolus removal in cerebral ischemia (MERCI) and Multi-MERCI trials evaluated safety and efficacy in the setting of acute stroke within eight hours of onset.8–10 The control arm in these trials was the spontaneous recanalisation rate of 18% from PROACT-II.7 Treatment with the retriever alone resulted in revascularisation in 60 of 111 (54%) vessels and in 77 of 111 (69%) vessels after additional therapy (IA t-PA, mechanical clot disruption). Among patients in whom revascularisation was achieved, there was a two-fold survival advantage and a significantly higher proportion of patients lived without significant disability.
The Penumbra pivotal stroke trial showed that the Penumbra device was safe and effective for revascularisation in patients with acute intracranial large-vessel occlusion.11 This prospective, multicentre, single-arm study included 125 patients with a National Institutes of Health Stroke Scale score of at least 8, presenting within eight hours of symptom onset, and ineligible for or with an occlusion refractory to IV t-PA. Thrombolysis in myocardial infarction grade 2 or 3 revascularisation was obtained in 81.6% of the patients in whom the Penumbra device was used.
In general, mechanical thrombectomy has been proved to be safe and effective for the treatment of acute stroke with large-vessel occlusion in selected patients. Mechanical thrombectomy requires the knowledge and skills of trained interventionists who can choose the patients best suited for this treatment and who have experience in the use of these devices.
Unfortunately, however, not all strokes occur in the vicinity of large comprehensive stroke centres or even smaller centres with on-call neurointerventionists, making endovascular options for revascularisation unavailable at many centres. In addition, the volume of patients who experience acute stroke is too great to be handled by the several hundred neurointerventionists in practice.
This lack of adequately trained and widely geographically spread neurointerventionists causes a relative deficiency in the US. In addition, patients who undergo stroke intervention are at some increased risk of direct vascular injury and/or intracranial haemorrhage. Recognition of stroke symptoms and rapid triage to a centre where the patient can receive treatment safely and physicians can handle not only stroke treatment but also the possible complications of stroke intervention is of primary concern.
How do we resolve a cut-and-dry case of supply and demand? In a perfect world, dedicated neurointerventionists with vast experience and knowledge of neuroendovascular procedures, neuroanatomy and stroke physiology would be able to fill this void. Regrettably, this is not possible, as the number of elective neurointerventional cases is relatively small and cannot support the number of neurointerventionists required to provide around-the-clock coverage for acute stroke patients across the US. Several thousand more committed neurointerventionists would have to be trained to provide 24-hour stroke coverage for all major medical centres.
In 2004, an estimated 1.3 million coronary angioplasty procedures were performed in the US.12 Approximately 2,200 coronary catheterisation laboratories exist in the US, although not all of them have the capability to perform angioplasty. Nonetheless, the manpower of the cardiologists is greater than that of the neurointerventionists. Most acute care hospitals with coronary catheterisation suites also have an emergency protocol for acute myocardial infarction that would allow the incorporation of an emergency stroke system without much additional infrastructure. In addition, most modern cardiac catheterisation laboratories are equipped with fluoroscopy devices that allow for the digital subtraction and road-mapping capabilities that are necessary to perform intracranial work. The techniques involved in acute stroke intervention, including thrombectomy, angioplasty and stenting, are familiar to the interventional cardiologist, although there is not a direct correlation between the characteristics of coronary and intracranial arteries. For an interventional cardiologist with catheter skills, less time may be involved in learning the technical aspects of neurointervention.
The use of interventional cardiologists to aid in covering stroke call in acute stroke programmes is wrought with potential educational, political and economic hurdles. If cardiologists are to perform stroke intervention, their training and education must go beyond ‘weekend warrior’ courses in which they receive, at best, a rudimentary understanding of neurointervention and neurosurgical disease states. Although the skill set for cardiac intervention is similar to that required for neurointervention, the correlation is not exact. For example, arteries of the intracranial circulation are more tortuous and delicate and less forgiving than those of the coronary circulation. While technical nuances could be easily learned by skilled interventional cardiologists, training in neuroanatomy and neurophysiology, interpreting CT angiograms and CT perfusion scans, performing the neurological examination, managing complications and decision-making are the most important aspects and also the most difficult to teach practitioners without a neurointerventional background. Because of these issues, cardiologists should ideally acquire neurointerventional skills and knowledge via additional fellowship training or an apprenticeship in stroke. During that time, the cardiologist could work with a stroke neurologist and stroke interventionists and focus only on stroke treatments and intervention.
Even if cardiologists are properly trained, they face political obstructions/territorial turf issues from other specialties, including interventional radiology, neurology and neurosurgery, even though few institutions have the manpower to offer uninterrupted stroke call coverage. Stroke programmes based on cooperative efforts between neurointerventionists and cardiologists are one way to possibly solve the manpower issues related to acute stroke care and management.
Finally, the national effort to increase public awareness of stroke must continue in order to emphasise the importance of early presentation. Increased stroke awareness also applies to smaller community hospitals that can facilitate faster transfer times to comprehensive stroke centres.
Endovascular therapies for acute stroke intervention have gained momentum as the treatment of choice for strokes caused by large arterial occlusion in certain patients. There is a paucity of neurointerventional practitioners who can offer continuous call coverage for stroke at large, comprehensive stroke centres. Interventional cardiologists may be able to help fill this void.
To perform stroke intervention, cardiologists must master a significant amount of neuroanatomy and neurology. The ability to interpret CT angiography and CT perfusion imaging and perform a rapid neurological examination that localises the ischaemic insult is crucial in choosing an appropriate treatment plan. Understanding symptom progression and causes (oedema, hypoperfusion, haemorrhage or infarct extension) must also be a priority of the training programme.13 Structured training programmes, organised and supervised by neurointerventionists, might allow cardiologists to focus on the early intervention of stroke and the necessary neurological and anatomical knowledge to effectively treat stroke.