| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Commentary |
Department of Neuroradiology, Radcliffe Infirmary, Woodstock Road, Oxford OX2 6HE, UK
The International Subarachnoid Aneurysm Trail (ISAT) was a multicentre randomized control trial of endovascular embolisation of intracranial aneurysms using coils versus the established optimum neurosurgical operation of craniotomy and clipping. It was stopped in May 2002 by its data monitoring committee when they identified that patients randomized to endovascular treatment (EVT) were enjoying a significant benefit. The British Journal of Radiology is a publication used to reporting novel science, why would it think that its readers would want to read about a clinical trial that reported its outcome in the Lancet in October 2002 [1] and stopped recruiting patients nearly 2 years ago? I would propose two explanations to entice you to read on; first that the trail has provoked a great deal of controversy and comment and second that it has affected the balance between the two subspecialties with consequences that have a more general relevance to interventional radiology and radiology in general.
Coils have been used to embolise intracranial aneurysms since 1987 [2] because it was obvious to some practitioners that the then current endosaccular (i.e. intra-aneurysmal) balloon embolisation was causing delayed aneurysm rupture. Within 5 years, technical improvements provided interventional neuroradiology with a controlled detachment coil system (the Guglielmi Detachable Coil – GDC) and it became evident by 1994 that EVT was a viable alternative to neurosurgery for patients with operable aneurysms, particular recently ruptured aneurysms [3]. ISAT was designed pragmatically to compare the two methods of preventing re-bleeding amongst patients after aneurysm rupture, i.e. subarachnoid haemorrhage (SAH). The trial randomized 2143 patients (1070 neurosurgery and 1073 EVT) in 43 hospitals and compared clinical outcomes 2 months and 1 year after randomization. Outcomes were defined on a modified Rankin scale using data provided by the patient (or their carer) on a standardized questionnaire. The frequency of patients being dependant or dead at 1 year was 30.7% for those assigned neurosurgery and 23.7% for their EVT counterparts. This represented relative and absolute risks reductions for this outcome of 22.6% (95% CI 8.9–4.2) and 6.9% (2.5–11.3), respectively, and a resounding win for the coils.
I cannot go on without declaring my bias. ISAT was piloted and run from the Radcliffe Infirmary in Oxford. I am an interventional neuroradiologist, I performed the first GDC operations in the UK at the Radcliffe Infirmary and throughout the ISAT recruitment period, I worked alongside Dr A J Molyneux who forced us into realising that a randomized controlled trial (RCT) was the only way to legitimize coil embolisation and obtained MRC funding for the trial. Does this disqualify me from commenting; I leave you to judge.
For interventional neuroradiology, the ending of recruitment should have solved all our problems. Problems familiar to almost all radiologists of finding the time and resources to take on a new service and the local politics associated with resource shifts. But before discussing the current fallout, it is worth reviewing the published "Comments on ISAT". I said at the beginning that this trial has been the source of much controversy, despite a very clear statistical result.
The trial was controversial before we even started recruitment. Opposition came, surprisingly, from fellow interventional neuroradiologists. The arguments against a trial can be simplified into those that considered such a trial unnecessary since EVT was so obviously the better treatment and therefore it was unethical to randomize patients on the basis of a clinical equipoise, and those that argued that the technique was too young to undergo such scrutiny. The neurosurgical community was supportive or silent. Since ISAT published in October 2002, things have changed. Within months, eight editorial commentaries were published in six mainstream journals [4–11], the Lancet published four letters [12–14] including comments on commentaries [14, 15]. "Position Statements" were issued by three medical societies (the American Association of Neurological Surgeons (AANS) [16], the American Society of Neuroradiology [17] and the German Society of Neurorsurgery [11]) and an article had appeared in the New York Times [18]. Several descriptive and critical articles were printed in a range of small circulation publications sponsored by commercial companies. Much of this interest came from neurosurgeons in the USA, but before going into details the commercial dimension needs addressing. There is no getting away from the fact that for those manufacturing and selling coils, ISAT was God-sent. It is therefore hardly surprising that the public relations departments of such companies went into overdrive; this may have heightened the sense of threat felt by some neurosurgeons.
It had not previously come to my attention that medical societies might wish to issue "Position Statements". 10 days after the ISAT results were published Drs Harbaugh, Heros and Hadley issued a document on behalf of the American Association of Neurological Surgeons and the Congress of Neurological Surgeons which draws attention to the fact that 20% of patients admitted to participating hospitals were randomized, that only one centre was in the USA and that the treated aneurysms were a highly selected subset [16]. The statement goes on to imply that the standard of neurorsurgery in European centres was less than patients might expect from neurosurgeons in the USA, where the degree of subspecialization is different and to point out, along with most other commentators, that the long-term protection against re-bleeding provided by EVT is unknown and that the results should not be interpreted as applying to all aneurysm patients. In an extraordinary statement they go on "Media reports have attributed a 22.6% risk reduction to endovascular coiling compared to craniotomy for aneurysm clipping. The figure of 22.6%, the overall study relative risk reduction, suggests there was a dramatic reduction in the number of poor outcomes among patients whose aneurysms were treated with coiling as compared to those whose aneurysms were surgically clipped. This is not the case. It is the absolute risk reduction that is of greatest importance to patients. Importantly, the absolute risk reduction of 6.9% reported by the ISAT authors should not be inappropriately generalized". I can only agree with the statement's precision and accept that the statement was addressed to a non-medical audience, but I am not sure that the 53 patients, who, after allocation to neurosurgery, were dependant or dead at 1 year and constitute these statistical differences would see the distinction.
I do not intend to review the detailed criticisms, since replies by Dr Molyneux and Mr R Kerr on behalf of the ISAT collaborators have been published [6, 12, 15]. But what do we do with the results of properly conducted international clinical research? Repeat the study in the USA? Not ethical, I hear the pre-ISAT critics chorus. My view is that the data can be applied logically which is echoed by several neurosurgical commentators. Mr K W Lindsay in a thoughtful article [5] points out that what underlies the AANS response is the pressure from the consumer patient for less invasive surgery and argues that despite the lack of evidence of long-term efficacy that "neuroradiologists should now play an important role in the neurovascular team, not just as a diagnostician, but as an interventionalist, participating in management decisions of all patients with aneurysmal subarachnoid haemorrhage". He is in the vanguard of a new cry for neurovascular teams and is not alone in the UK neurological community in recognising the importance of EVT [19].
Where now for interventional neuroradiology? Will we be forced to abandon our radiology home and become a component of a neurovascular (neurosurgical) team? Medical demarcation is not the ultimate issue because in the end, the important goal is that we provide patients with the best treatment as competently as possible. The UK is relatively well provided at the moment with trained interventional neuroradiologists, but the implications of ISAT are that hospitals accepting patients with acute aneurysmal SAH should be able to provide EVT on an emergency basis. Thus such centres should have 2–3 trained endovascular therapists to provide a service and cover for weekends and out-of-hours treatments. We certainly do not have that volume of interventional manpower yet. The solution may lie in rationalizing centres or training neurosurgeons. The former would risk creating second-class neurosurgical units and the latter, potentially, the end of the interventional neuroradiology.
Will neurosurgeons now train in their legions to provide their patients with the choice of endovascular or extravascular surgery? I think this is unlikely because the required technical skills are different, though there almost certainly will be individuals who want to accept the challenge. Internationally there are a number of "hybrid" surgeons and a strong push is detectable in the USA for neurosurgeons to train. It could happen here but currently the consensus amongst UK interventional neuroradiologists is that the skills needed to perform EVT need regular practice and that to maintain skills in both clipping and coiling would be difficult. I hope I am right. If I could direct how this interesting and exciting new branch of medicine develops, I would like to see an independent subspecialty. It is still young but with ISAT the innocence of youth has been lost and if a subspecialty of individuals, exploiting the endovascular route under X-ray guidance can avoid energy sapping demarcation disputes for a little longer its full potential to treat neurovascular diseases would be better realised. What it is called and where it sits in relation to radiology and neurosurgery is not going to deflect the pressure from patients for minimally invasive surgery and from commercial companies to develop and improve endovascular devices. GDC and coil embolisation of intracranial aneurysms are just the beginning.
Received for publication March 22, 2004. Revision received July 12, 2004. Accepted for publication August 9, 2004.
 |
References |
|---|
 Top References |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. Pelz, T. Andersson, P. Lylyk, M. Negoro, and M. Soderman Stroke Review: Advances in Interventional Neuroradiology 2004 Stroke, February 1, 2005; 36(2): 211 - 214. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
