Priorities in the management of abdominal aortic aneurysm

If the primary aim in managing abdominal aortic aneurysm (AAA) is to prevent rupture, present methods are ineffective. Nearly 5000 people die from AAA each year in England and Wales alone1. Perhaps this should not surprise us, as enthusiasm has been concentrated mainly on improving surgery, which can benefit neither those with an undetected aneurysm (90 per cent)2 nor those with rupture who do not reach hospital alive (over 50 per cent). Such failure would be forgivable if there were no effective alternative. Ultrasonography can detect almost all AAAs. When it is used as a screening tool and combined with elective surgery, the mortality associated with the disease is halved3. The benefit after a single scan is known to continue, with decreasing cost per life year saved and increasing cost effectiveness, for 7 years4. Ashton and colleagues5 in this issue of BJS (pp 696–701) now show that the benefit remains effective for at least 15 years. These facts lead to the inescapable conclusion that resources would be better spent on screening and elective surgery programmes, rather than the development of increasingly sophisticated operative technologies that are likely to reduce overall death from AAA by only a small margin. Ultrasonography not only detects aneurysms efficiently, it also measures their diameters accurately. Selection for surgery is generally based on this measurement, which is an important factor in terms of cost and benefit. Although an aortic diameter of 5·5 cm is widely accepted as a criterion for operative intervention, even at this size the balance of risk may be in favour of surveillance. Several authors over the past 20 years have observed that the risk of rupture of an AAA between 5·0 and 5·9 cm in diameter is only 1 per cent over a year6. In this issue of BJS (pp 709–716) Brown and colleagues7 show that surgery, even in moderately fit patients treated by endovascular means, carries a greater risk; operation is associated with a 30-day mortality rate of 1·6 per cent. It is when the aneurysm diameter reaches 6 cm and beyond that the risk of non-intervention increases rapidly. And yet operation on an aneurysm smaller than 5·5 cm is occasionally suggested. In this scenario not only is the risk of surgery greater than that of the disease, but two-thirds of those with such small lesions will have an unnecessary operation as they would never have died from aneurysm rupture. That there is no survival benefit for early surgery over surveillance of small aneurysms is reaffirmed by the 12-year results of the UK Small Aneurysm Trial on pages 702–708 of this Journal8. In view of these findings, it is hard to sustain any argument in favour of surgery for aneurysms of less than 5·5 cm diameter. Although a low operative mortality rate for elective AAA repair of around 2 per cent has been reported by a few specialist centres, such results are far from universal. The April issue of BJS contains a meta-analysis of 32, mainly American, papers (and hospital episode statistics from the UK) that shows a much higher mean 30-day operative mortality rate of 9·5 per cent9. The 30-day mortality rate for open repair was 6 per cent in the Multicentre Aneurysm Screening Study (MASS)3, 5·5 per cent for the early surgery group in the UK Small Aneurysm Trial8 and 4·2 per cent in the open repair arm of the UK Endovascular Aneurysm Repair (EVAR) 1 trial7. These latter two papers are reported in this issue of BJS. Although the 30-day mortality rate in the endovascular arm of the EVAR 1 trial was only 1·6 per cent, this must be considered in conjunction with the observation that by 4 years 41 per cent of patients had suffered a complication and 20 per cent had required reintervention, with a persistent annual aneurysm rupture rate of 1 per cent10. Furthermore, patients were randomized only after excluding those anatomically unsuitable for endovascular repair. A 30-day mortality rate of 5 per cent might be considered acceptable in conjunction with a screening programme but, in view of the above findings, a longer view seems more sensible. Perhaps the 1-year mortality rate, plus rates of reintervention and complications, would allow patients a fuller understanding of the relative risks. The authors of the meta-analysis published in the April issue of this Journal concluded that for elective AAA surgery high hospital patient volume leads to low mortality9. If this is correct, excluding low-volume hospitals should improve outcome; the authors suggest a watershed of 43 procedures per annum. However, in the UK at least, this excludes more than half the hospitals currently offering vascular surgery, based on the Vascular Society of Great Britain and Ireland database. Three years ago 180 hospitals in the UK undertook vascular surgery and a reduction to around 50 centres has been proposed, an action with major resource allocation implications. Before going down this route we must be sure that volume equates with ability. It may be that high volume smoothes the figures, making them appear more consistently average and allowing individual low achievers to hide in large departments. Ideally, hospitals with low mortality rates should be selected when AAA screening is introduced, but operative outcomes for individual surgeons or departments are notoriously difficult to assess, especially when small numbers are involved. The range of mortality rates in the meta-analysis emphasizes this, stretching as it does from near zero to 100 per cent in low-volume departments9. The Vascular Society of Great Britain and Ireland recommends the formation of networks and the monitoring of their outcomes, with review of those with results outside the 95th percentile. This concept, which takes into account the limited capacity of existing departments, is a more realistic way of improving outcomes and maintaining standards than a major reallocation of resources to a few high-volume centres. If we really want to improve the outcome for those affected by AAA, screening should be introduced as soon as possible. The National Screening Committee in the UK and the equivalent committee in the USA have recommended screening for men aged 65 years to their respective governments. In the USA this has already been established through Medicare. In the UK we await a final decision, but individual units can start to make preparations based the National Screening Committee's recommendations; these are available at http://www.library.nhs.uk/screening.