Discussion
John Eidt (Little Rock, AR): The authors have presented
admirable results in a consecutive series of 105 patients
scheduled for endovascular repair of abdominal aneurysms
with the AneuRx endograft. They had a 2% 30-day mortal-
ity and an average 2-day hospitalization, and they were
plagued, as all endograft centers have been, with the need
for numerous secondary procedures to achieve moderately
good clinical success. I have to admit that these definitions
of success are a little bit confusing, as what is continuing
success and secondary success. We have just finished the
first decade of the endograft experiment and some authors
have already declared the experiment a failure, citing the
frequent problems such as endoleak and graft limb occlu-
sion, requiring secondary interventions. I would caution that
the initial phases of technological development are often
fraught with unforeseen obstacles and I have no doubt that
endovascular surgical techniques will play an important role
in the management of vascular disease in the future. It is
important to remember that the ultimate goal of aneurysm
treatment is prevention of rupture and endografts have been
generally successful in achieving this goal with reported
long-term rupture rate of less than 3% in most reports. And
as I think I understand your report today, there was one
rupture in today’s report. I think it’s worthwhile to be able
to offer an older, sicker patient, the potential for a very high
assurance of rupture prevention with a less morbid proce-
dure. Results with endografts are device-specific. The
AneuRx graft is a modular graft with a nytenol skeleton
covered by polyester fabric. It does not have fixation hooks
but relies on friction, stent incorporation, and so-called
column strength or longitudinal stiffness for fixation. Fric-
tion generated by a self-expanding nytenol stent is really
quite low. Fibrous incorporation is unpredictable and col-
umn strength works poorly in tortuous vessels. At the sum-
mer vascular meetings last year, a group from Italy reported
migration of greater than 10 mm in over 25% of their cases.
Migration may result in the development of a proximal
attachment site type 1 endoleak. Type 1 endoleaks are the
most dangerous and the most likely to lead to rupture.
You’ve described 9 patients with type 1 endoleak, which is
nearly 10% of the total group. This is a slightly higher
proportion of type 1 endoleaks that most other reports. So
my first question relates to the cause for these type 1
endoleaks. What was the etiology of these type 1 endoleaks?
Was it inappropriate patient and graft selection preopera-
tively or do you think it was due to some type of intraop-
erative technical error. For example, do you think you were
too aggressive in trying to use endografts in every patient
even though some may have had inappropriate anatomy?
Did you undersize the grafts or do you think that these
endograft endoleaks represent technical errors for not plac-
ing the graft close to the renal arteries? So the question is
patient selection or technical error?
Britt Tonnessen: I think the answer to that question is
patient selection. The majority of the first half of these
patients were done during phase 2 clinical trials before the
graft was widely available on the market, and so the group
was definitely more aggressive. Other studies that we’ve
done have shown that angulation is an issue that we didn’t
recognize early on and that migration certainly plays a role
as far as contributing to type 1 endoleaks. In the presenta-
tion, we divided the patients into the first half and the last
half. There were six type 1 endoleaks in the first half and
only three in the latter half. This reflects our learning curve.
John Eidt: You said that these patients were high risk or
higher risk than your open repairs. How does that really
compare with your open repair patients? You said 95% of
your endograft repairs were in either class 3 or 4 ASA class.
How does the open repair group compare to that?
Britt Tonnessen: We haven’t yet broken down the open
repair group into ASA class. What I can say is that a
significant number of the patients who underwent endovas-
cular repair had COPD and wouldn’t have been considered
candidates for open repair.
John Eidt: My last question is really sort of an extension
of that. You had 11 people that died during the first year. Do
you think you should be more selective in refusing en-
dograft repair for some of these very old, very sick patients
with limited life expectancy? I mean, are we really doing
them any favors by repairing a moderate-sized aneurysm
that may not really effect their life expectancy?
Britt Tonnessen: I think that thus far the data’s borne
out that perhaps the very old, very sick patients may be
some of the best candidates for endovascular repair. Of the
10% mortality rate at one year, it was very diffuse as far as
the various causes of death. A couple died from their pre-
existing coronary disease, from CHF, pneumonia, renal
failure, lung cancer, so it was very unpredictable as far as
which patients died from what. There were four patients that
had renal insufficiency, two of those died and both of those
had some procedure-related complications as far as acute
renal failure. So perhaps renal insufficiency is one factor to
seriously look into prior to endovascular repair.
567B.H. Tonnessen et al. / The American Journal of Surgery 184 (2002) 561–567