Article Text

Download PDFPDF

Aneurysm surgery after the International Subarachnoid Aneurysm Trial (ISAT)
  1. R S Maurice-Williams
  1. Correspondence to:
 R S Maurice-Williams
 Consultant Neurosurgeon, The Royal Free Hospital and Medical School, London, NW3 2QG, UK;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Every so often in the history of medicine a major technical or pharmaceutical innovation leads to a sudden and fundamental shift in practice. Such events as the introduction of general anaesthesia, the use of antisepsis in surgery, and the introduction of streptomycin immediately spring to mind. Similar events have also occurred in the narrow world of neurology and neurosurgery. The appearance of carbamazepine and l-dopa had an immediate and dramatic effect upon the extent to which surgery was used in trigeminal neuralgia and Parkinson’s disease. The computed tomography (CT) scan quickly led to the disappearance of lumbar air encephalography and ventriculography and the redundancy of the radiological expertise that these older forms of investigation required.

It seems likely that the International Subarachnoid Aneurysm Trial (ISAT) will have a comparable effect on aneurysm surgery. Even before the publication of the results of the interrupted trial in The Lancet in October 20021 there had been a progressive shift from surgery to coiling in the treatment of aneurysms whether these had ruptured or not. Since the publication of ISAT it is clear that many units in the UK have almost abandoned direct surgery. This change may have been more marked in the UK than in other countries for, in this country, few patients with aneurysms receive private treatment—meaning that any change in policy has no economic disadvantage for the surgeon.

The gravity of this development for neurosurgery cannot be overestimated. A ruptured aneurysm is the most common neurosurgical condition which requires the highest degree of surgical skill. The condition affects relatively young patients who are otherwise in good health, and treatment can make the difference between a full cure with a normal life expectancy on the one hand and death or hopeless disability on the other. Aneurysm surgery constitutes, or at any rate constituted, a major proportion of that part of cranial neurosurgery that needed the operating microscope and much of higher neurosurgical training was dependent upon it. For many neurosurgeons aneurysm surgery was their life, and it is perhaps no great surprise that some of them have been reluctant to accept the implications of ISAT and have argued that the need for aneurysm surgery will persist despite it.

The main findings of ISAT may be summarised as follows.1 A total of 2143 patients with ruptured intracranial aneurysms were randomly assigned to clipping (n = 1070) or coiling (n = 1073). At one year the outcome was assessed by a modified Rankin score of 36—that is, patients who were dependent or dead. A significant difference was found between the groups and the trial was abandoned; 22.7% of coiled patients were dependent or dead compared with 30.6% of those subjected to surgery. The annualised risk of rebleeding after coiling was 0.16% (2 cases per 1276 patient years) and zero for clipping.

Detailed objections to the findings of the trial have been made both in the UK and in the USA. In an editorial in the British Journal of Neurosurgery Kirkpatrick and colleagues have made a number of cogent points.2 First, only a minority of the patients admitted to the centres participating in the trial were eventually randomised, and they point out that the results obtained may not be applicable to the non-randomised majority. Secondly, they state quite correctly that clipping provides a permanent cure for the overwhelming majority of treated patients although if neck remnants are left these may regrow into aneurysms and rebleed. In contrast, complete occlusion by coiling is achieved in only about 70% of aneurysms and it is known that even fully coiled aneurysms may re-form and bleed. A third point that they make is that the results of the surgically treated group do not seem to be especially good, and that while a high proportion of the coiling may have been carried out by radiologists who were very experienced in this field the same may not have been true of the surgeons concerned who may have had a greater mix of ability. They argue that much better surgical results might have been achieved by experienced and specialised vascular surgeons, and they also argue for the development of specialised centres for both surgery and endovascular treatment.

At about the same time, a lead editorial in the Journal of Neurosurgery by Dr R C Heros, a leading American vascular neurosurgeon, addressed the data accumulated both by ISAT and in a large series of coiled patients reported by Murayama and colleagues.3,4 Heros points out that even in the best hands only about 60% of aneurysms can be fully coiled at the first attempt—the proportion being somewhat greater in small aneurysms with narrow necks. Furthermore, recanalisation of completely coiled aneurysms occurs in between 2% and 37% of aneurysms and the risk of this is greatest in the largest aneurysm sacs.3

These arguments against the long term usefulness of coiling are well thought out but fail to take account of certain key factors. The results of coiling are not static—they are constantly improving. Not only are interventional radiologists gaining more experience but, more importantly, the microtechnology of endovascular techniques is improving and evolving all the time. This steady improvement is apparent to those of us who have access to interventionalists with experience and ability. In the author’s own unit there has been a clear improvement in the results of coiling during the past two to three years and it now appears that virtually all aneurysms can be successfully coiled with minimal mortality and morbidity. In contrast, the technology of surgery has been static since the introduction of the operating microscope into aneurysm surgery in the late 1970s and early 1980s. Indeed it is difficult to see how aneurysm surgery could be improved further and there are reasons for believing that the results of surgery are worsening as surgery is spread amongst a greater number of surgeons, many of them in training. This process is bound to accelerate as the small number of surgeons with a huge aneurysm experience shrinks as a consequence of retirement and death.

Some further points need to be made. First, as the active treatment of aneurysms has become so universal it has perhaps been forgotten that a very high proportion of patients who survive in good condition after the rupture of an aneurysm do well without any treatment at all. The celebrated study of Alvord et al showed that patients who survived between one and three days in grade I after a ruptured aneurysm had an 80% chance of being alive at two years while for those who survived between seven and 21 days the chance rose to 95%.5 Other studies have shown that the natural history of ruptured aneurysms treated conservatively is relatively benign. Sharr and Kelvin found that the great majority of ruptured vertebrobasilar aneurysms treated conservatively remained well many years later6 and the Cooperative Study of the mid 1960s found that of 34 patients who refused operation for a ruptured aneurysm 82% remained alive and well at follow up of up to six years.7 If so many patients who have no treatment at all do well it seems likely that they would do even better after incomplete coiling which at least provides some protection. In any case, aneurysms which are partially coiled or where the aneurysm re-forms can always be re-treated with coiling in the future when it may well be that new and improved endovascular technology will be available. Even surgery is not completely foolproof. Probably only a minority of surgeons carry out routine check angiography, and without it the surgeon cannot be sure that the aneurysm they have “clipped” is indeed fully occluded.

The second point about surgery is emphasised by the results of a recent series of surgical results published in this journal.8 These were the results achieved by an experienced vascular surgeon in the period immediately prior to the introduction of coiling. At first sight the one year surgical results appear good with an overall mortality of 2.6% for all patients regardless of clinical grade with 83.7% in the best category of the Glasgow Outcome Score. However, these overall results disguise the fact that a high proportion of patients experienced complications and subsidiary procedures that would have been largely avoided with endovascular treatment. Faced with the choice it does not seem likely that many neurosurgeons would choose surgery as opposed to coiling for themselves if they had a ruptured aneurysm any more than they would choose surgery as opposed to stereotactic radiotherapy if they had an acoustic neuroma.

Some vascular neurosurgeons believe that there will still be a role for them in treating aneurysms that are not amenable to endovascular treatment or where the latter has failed. The present author believes that they are deluding themselves. It is becoming apparent that the overwhelming majority of aneurysms, whatever their situation, are now amenable to coiling and the scope of endovascular treatment is bound to enlarge as experience and technology improve. While it is true that there may remain a residue of large and/or complex aneurysms that cannot be so treated it is likely that these will present such formidable surgical problems that it would be difficult to justify a surgical attack even today when we still have very experienced aneurysm surgeons. How much more difficult it will be to justify surgery as opposed to conservative management when there are no longer neurosurgeons with the day to day experience of the technical difficulties of routine aneurysm surgery.

The consequences of ISAT are extremely serious not only for vascular neurosurgeons but also for neurosurgery as a whole. At one blow a major part of neurosurgery is under threat of extinction. For those who have specialised in vascular neurosurgery this situation can be fairly described as catastrophic. A worrying feature is that as neurosurgery breaks down into subspecialties, many neurosurgeons will be exposed to the possibility that technical advances will make their experience and expertise redundant. As far as aneurysm surgery is concerned there is no escaping the fact that the writing is on the wall much as it was for the stagecoach in the late 1820s. The ostlers and postilions and even some of the passengers may have attempted to point out the problems associated with the new steam engines but for the detached observer there could be no doubt about the way that the future was going to develop.