MASS trial results

Alan Scott presented the results of the Multi-centre Aneurysm Screening Study (MASS) at the 37th Annual Meeting of the Vascular Surgical Society of Great Britain and Ireland in Belfast.

According to Scott, there are over 6,000 deaths from ruptured abdominal aortic aneurysms each year. “Half of all cases don’t even make it to hospital and for those who do there is waiting for them the high mortality of emergency surgery (about 80–90%). However, for elective surgery, in most district hospital there is a mortality of around 2–8%.” It was this, said Scott, that led to the MASS study, which aimed to see if it was possible to identify aneurysms and treat them electively, thereby reducing mortality.

The MASS study started in 1997, with recruitment over two years. 67,800 men aged between 65 and 74 were recruited by four screening centres (Portsmouth, Southampton, Winchester and Oxford). The men were randomised into a control group and a group invited to be screened. 33,839 were invited to screening, with 80% attending. Some other losses left 27,147 men who were screened, with 1,333 aortic aneurysms detected.

Maximum anterior-posterior and transverse aortic diameters were measured using ultrasound. The largest one of these two measurements was taken as the indicator for changing clinical management.

Scott said that the overall prevalence of abdominal aortic aneurysms in men between 65 and 74 was found to be 4.9%, slightly less than the investigators expected. The distribution of AAA was as expected two-thirds under 4.5cm, and one third equal to or above 4.5cm.

“The 30-day operative mortality at four years for the 414 elected operations was 6% and for the emergency operations 37%, this was in both the control and invited group together,” said Scott.

“We weren’t expecting to demonstrate significant abnormality or change in all cause mortality as abdominal aortic aneurysms only represent 2% of all the causes of death.”

“At four years there were 113 aneurysm related deaths in the control group, compared with 65 in the invited group. A 42% reduction in AAA related mortality. But as far as the patient is concerned, they are more interested in what is going to happen if they turn up for their screening and in those who attended screening there was a 53% reduction in AAA-related mortality.” There was, however, no significant change in quality of life shown.

In conclusion Scott stated, “We have demonstrated a highly significant reduction in mortality from AAA as a result of screening and that this did not cause any significant quality of life change. We feel that the trial provides strong support for starting a national screening programme, but in order to say this with conviction, we need to know about cost-effectiveness and the costs of screening.”

Martin Buxton of the Health Economics Group Brunel University then presented an evaluation of cost effectiveness to four years of screening AAA. The cost of an elective surgery was calculated as £6,909, compared with emergency surgery at £11,176. This difference being largely attributable to differential use of ITU. At four years, the mean AAA related cost per patient in the invited group of the MASS study was calculated as £98.42 (screening and surgery), while for the control group it was £35.03 per patient (just surgery). The mean survival time in days up to four years for the two groups was very close, reflecting that the AAA-related mortality is relatively low across the groups as a whole. In fact, it leads to only a small difference in survival time of 0.82 days (at the cost of £63, the difference in costs per patient). This translated to around £28,000 per life year gained up to four years, and when QALY adjusted this becomes £36,000 (because value of year of life in this age group is only 0.8). According to Buxton, this is around the benchmark that NICE considers reasonable and if extended to 10 years the cost-effectiveness would be well below this level.

Buxton stated: “On the observed four year data, the cost-effectiveness ratio is already at the margins of what might be called acceptable. So even if there were no further gain, we would still be in with a possibility. We are convinced from the data from this trial that it will fall considerably, even if you look at only 10 years out. It will continue probably beyond that. And we argue again that the clinical and economical analysis of MASS provides clear evidence to support the true cost-effectiveness of screening men aged 65–74 and a strong basis on which to recommend this as a national screening programme.

It is worth noting that the following day saw the presentation of data from the Western Australia randomised controlled trial of screening for AAAs. In this study, screening was found to reduce AAA-related mortality by 28%. This was less than expected, and partly attributed to a very low level of aneurysms in the non-screened group.