Endovascular repair is worse than open repair of abdominal aortic aneurysms
BMJ 2005; 331 doi: https://doi.org/10.1136/bmj.331.7518.0-f (Published 22 September 2005) Cite this as: BMJ 2005;331:0-fAll rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
The published results of the UK EVAR 1 trial (1) do not justify the
headlined conclusion of the ‘POEM’ article printed in the BMJ on 24
September. Moreover, the article contains factual errors, which serve
further to mislead.
The adjusted 30-day mortality rate after endovascular repair (EVAR)
was 1.7 percent compared to 4.7 percent after conventional open repair
(OR). The BMJ erroneously reported these figures to be 0.2 percent and 0.5
percent. The two thirds reduction in 30-day mortality was statistically
significant (Odds Ratio 0.37, 95%CI 0.37, p<0.02) and, in practical
terms, translates into a 1 in 60 risk of death within 30-days after EVAR
compared to 1 in 20 after OR. Comparison of total in-hospital mortality
rates showed an even greater advantage for EVAR (1.9 percent vs 6.4
percent; Odds Ration 0.30, 95%CI 0.14-0.62, p<0.001).
Follow up of the patients in the EVAR 1 trial showed that the early
benefit associated with EVAR, in terms of aneurysm-related mortality, was
sustained in the medium term. After a mean of 4 years, 19 patients in the
EVAR group compared to 34 in the OR group had died from causes related to
their aneurysm (4 percent vs 7 percent; Odds Ratio 0.51, 95%CI 0.29-0.92,
p=0.02).
Arguments to the effect that the early survival benefit associated
with EVAR is clinically irrelevant because the trial failed to demonstrate
any significant difference in all-cause mortality in the medium term might
resonate with some health-care bureaucrats. However, they are likely to
cut little or no ice with patients or their relatives. A man in his 70’s
or 80’s with a large aortic aneurysm is far more interested in his chances
of surviving the operation than what may or may not happen to him in 4 or
5 years’ time. Given that the trial demonstrated equivalence in terms of
later survival, and certainly no detriment associated with EVAR, we would
argue that this order of priority is entirely valid.
The BMJ states correctly that, in the EVAR 1 trial, EVAR was
associated with a higher rate of post-operative complications and
secondary interventions than OR. But, the article failed to mention that
fact that neither of these was responsible for any additional deaths.
Secondary interventions did account for a marginally higher cost of EVAR
over 4 years and may well have contributed the early advantage for EVAR in
terms of Health-Related Quality of Life (HRQL) becoming evened out during
the first year. However, EVAR technology is still evolving and there is
convincing evidence from other sources to show that complication and
secondary intervention rates are reducing progressively (2). To consign
EVAR to the history books on the basis of the higher incidence of late
adverse events reported in the EVAR trial is therefore premature at the
very least.
To elderly patients major surgery, with the associated need for post-
operative intensive care and prolonged hospitalisation, represents a
massive ordeal. Understandably many are attracted by the proven benefits
of minimally invasive treatment. EVAR can be carried out under local
anaesthesia. It involves a much shorter stay in hospital without need for
intensive care and recovery is more rapid than after OR. The results of
the EVAR 1 trial do not constitute acceptable grounds for denying these
advantages to properly informed patients. They should be allowed the
freedom to choose. This is possible on the NHS only if the necessary
funding is provided.
In the absence of guidance from NICE, local PCTs must make up their
own minds whether or not to fund EVAR. Negative reports in leading
journals, such as the BMJ are sure to influence them against doing so. We
now have a situation in which EVAR is banned in some areas while it
continues to be available in others - yet another example of post-code
lottery with its associated inequities and injustices.
A reasonable interpretation of the EVAR 1 trial results is that they
showed equivalence of outcome for EVAR and OR in terms of overall
survival, which was the principle end point and that the secondary
endpoints were reasonably well balanced also: although we would argue that
a sustained lower rate of aneurysm-related death after EVAR outweighs the
negative impact of the higher rate of complications and secondary
interventions. The data on HRQL are neutral. To interpret the results as
showing that EVAR is worse than OR is simply not justified.
The editorial article “A POEM a week for the BMJ” (BMJ 2 November
2002) promises articles that will provide doctors with information that is
truly important in order to help them deal with the “information paradox”
– too much information and an inability to find what is needed. Factual
errors and unbalanced analysis represent serious failings in respect of
this intended objective with potential to cause widespread and long-
lasting damage to the welfare of patients.
References
1.The EVAR Trial Participants. Comparison of endovascular aneurysm
repair with open repair in patients with abdominal aortic aneurysm (EVAR 1
Trial), long-term survival, graft durability, quality of life and costs:
Randomised controlled trial. Lancet 2005; 352: 2187-2192
2.Torella F for the EUROSTAR participants. Effect of improvedendograft
designon outcomeof endovascular aneurysm repair. J Vasc Surg 2004; 40: 216
-221
Peter Harris,
Consultant Vascular Surgeon
Royal Liverpool University Hospital
Richard McWilliams,
Consultant Vascular Radiologist,
Royal Liverpool University Hospital
Competing interests:
None declared
Competing interests: No competing interests
The title and interpretation of the EVAR trial results in the filler,
POEM Endovascular repair is worse than open repair of abdominal aortic
aneurysms September 21 2005,1 are both misleading and incorrect.
Mortality figures were quoted both as aneurysm related and all cause
deaths. The 30-day mortality was 1.7% in the EVAR group, which is
significantly lower than 4.7% in the open repair group. Surprisingly, POEM
somehow interpreted 30 day mortality to be less than 1% for both groups
which is clearly an error. The confidence intervals (odds ratio 0.35, 95%
CI 0.16-0.77, p=0.009) make type I error unlikely. If pre intervention
deaths are added to the figures 30-day mortality was calculated at 3.5%
for EVAR and 7.1% for open surgery.2
The advantage of EVAR is sustained at 4 years, with 4% aneurysm
related deaths in the EVAR group compared with 7% in the open repair
group. This confers a 3% absolute reduction in aneurysm related mortality.
The all cause mortality rate was comparable in both groups at 4 years.
EVAR offers a 30 day lower risk treatment compared with open surgery
in patients who are medically fit with an abdominal aortic aneurysm.
Compared with open repair EVAR offers no advantage with respect to all
cause mortality and health related quality of life; and is more expensive,
however, it confers a 3% improved aneurysm related survival which persists
at 4 years.
The POEM conclusions that endovascular repair is worse than open
repair is inaccurate and not evidence based.
Jeremy Crane Specialist registrar
Meryl Davies Specialist registrar
Nick Cheshire Professor of Vascular Surgery
Regional Vascular Unit, St Mary’s Hospital, Paddington, London.
References
1 Endovascular repair is worse than open repair of abdominal aortic
aneurysms
BMJ 2005; 331: 0-f
2 EVAR Trial Participants. Endovascular aneurysm repair versus open
repair in patients with abdominal aortic aneurysm (EVAR trial 1):
randomised controlled trial. Lancet 2005;365: 2179-86
Competing interests:
None declared
Competing interests: No competing interests
We strongly disagree with the analysis and conclusions of the POEM
group regarding the EVAR trial. These results were presented to the
Endovascular Forum of the Vascular Society of Great Britain and Ireland
and the British Society of Interventional Radiologists in June 2005. The
mortality of endovascular repair at 30 days was 1.7% and of open repair
4.7%. We do not understand how the POEM group came up with their mortality
figures of less than 1% for both groups. This 3% advantage in mortality
was also seen in the mid-term results as a 3% decrease in aneurysm related
death in favour of endovascular repair. Most of the complications of
endovascular repair were type 2 endoleaks which we now recognise rarely
require treatment. The benefits of endovascular repair include shorter
hospital stay, and faster return to work and the activities of daily
living, both of which are very important to patients. The consensus at the
Endovascular Forum was that endovascular repair was advantageous for fit
patients (EVAR-1 trial) but patients not fit for open repair were unlikely
to benefit from endovascular repair (EVAR-2 trial).
We believe that most patients, given the option, would choose a
procedure with lower mortality, lower morbidity and equivalent efficacy.
Certainly, our waiting list for endovascular procedures since the
publication of the EVAR results has increased dramatically driven by
patient demand. To conclude that EVAR has no benefits is hugely misleading
and does the POEM group no favours.
Competing interests:
None declared
Competing interests: No competing interests
Royal United Hospital,
Combe Park,
Bath.
BA1 3NG
Dear Editor,
As a surgical Senior House Officer I was surprised to note the
incongruity between the title and text of the Patient-Oriented Evidence
that Matters filler, “Endovascular repair is worse than open repair of
abdominal aortic aneurysms” (BMJ 2005;331 (24 September). Whilst I feel
your synopsis of the results of the EVAR trial 1 (Lancet 2005;365: 2179-
86) are correct, I do not see any evidence here that suggests endovascular
aneuysm repair is worse than open repair as suggested in the title. In
fact, although the improved 30 day survival and absolute reduction in
aneurysm related mortality are small, they do show at least no difference
and probably infer advantage to the endoscopic approach. Their inability
to reach significance represents the lack of power of the study, something
that may resolve with the availability of longer term results. Similarly,
cited disadvantages such as cost will clearly be reduced if endoscopic
workload increases. My experience of health related quality of life in
terms of patient satisfaction is that EVAR patients are at a significant
advantage.
It is important that in an article concerning “evidence that
matters”, one that particularly sets out to influence opinion and working
practices of clinicians, the title must avoid misleading statements and
the “Bottom line” gives a balanced opinion. In this case I believe it has
failed us on both counts.
Yours sincerely,
William Thomas
(williamthomas@doctors.org.uk)
Competing interests:
None declared
Competing interests: No competing interests
Reply
Several readers have provided thoughtful comments and raised
questions about a recent POEM on endovascular aneurysm repair (EVAR).
First, there is a miscalculation of aneurysm-associated mortality: a
decimal place was dropped and in rounding (0.17% rounded to 0.2% and 0.47%
rounded to 0.5%) resulted in the contentious rates. The correct rates
should be 1.7% and 4.7%. We reported the absolute 3% reduction in aneurysm
mortality correctly.
We agree the title can be misleading but only if you focus on the
single secondary outcome of aneurysm-specific mortality. In this unblinded
study, the outcomes least likely to be influenced by the lack of blinding
(all-cause mortality, complications, costs, and need for additional
interventions) give a fairly negative overall impression of EVAR.
Focusing only on cause-specific mortality evades the actual evidence
in this study: all-cause mortality was unchanged. Nonetheless, if one
chose to ignore this, compared with open surgery, one would need to treat
33 patients with EVAR to prevent one aneurysm-associated death. However,
based on the rate of complications and subsequent interventions, one would
only need to treat 3 patients with EVAR for one to be harmed. One of the
authors mentions that the most prevalent complication, a type 2 leak, is
no longer treated. We can only report what actually occurred in this
study.
To summarize this was a well-done study that showed EVAR had no
effect on total mortality, a modest reduction in aneurysm-associated
mortality at the expense of higher complication rates, higher re-
intervention rates, higher costs and no improvement in quality of life.
Based on this study, the evidence looks like on balance, EVAR is more
harmful.
Henry C. Barry, MD, MS
Senior Editor, InfoPOEMs
Competing interests:
None declared
Competing interests: No competing interests