Key Words
Thoracoabdominal aortic aneurysm - Visceral debranching - Hybrid operation - Thoracic
endovascular aortic repair (TEVAR)
Introduction
Surgical repair of thoracoabdoiminal aortic aneurysms (TAAA) is a formidable undertaking
for both patient and surgeon. Clinical outcomes at high-volume centers are excellent,
with cumulative 30-day mortality rates reported as less than 10% in selected series[1]
[2]. Larger, more representative databases show that real-world experience does not
duplicate these results. In a study using the National Inpatient Sample (NIS) from
1988 to 1998 the overall morbidity was shown to be greater than 50% and mortality
was 23%[3]. Totally endovascular approaches are available at select institutions, and are reported
to have better outcomes with low rates of mortality, morbidity, and spinal cord ischemia[4]
[5]
[6]; unfortunately, these are not generalizable due to restrictions on access to devices
and need for custom fabrication.
Hybrid procedures offer an alternative approach to TAAA management. Relying on a combination
of standard open techniques and off-the-shelf endovascular stent grafts, they are
broadly applicable to a wide range of patient anatomy. First described by Quinones-Baldrich
and colleagues at the University of California Los Angeles (UCLA), these approaches
rely on debranching of the visceral aorta followed by standard endovascular aneurysm
repair, covering the entire affected aortic segment[7]. Because hybrid approaches avoid the extensive two cavity exposure, aortic cross-clamping,
and mechanical circulatory support that comprise open TAAA repair, they offer the
theoretical advantage of being less invasive. Despite this, results following hybrid
repair have been mixed, with outcomes varying widely depending on both patient and
surgeon related factors. In reality, hybrid repair is no less of an undertaking for
either patient or surgeon, but rather a different undertaking that can offer specific
advantages in distinct subsets of patients.
Technical Details
Hybrid TAAA repair is a two-staged procedure, consisting of open visceral debranching
followed by aortic endografting. The first stage, visceral debranching, is most commonly
achieved from a transabdominal approach via midline laparotomy. In the absence of
prior laparotomy, this allows ready exposure of all of the visceral vessels as well
as the iliac arteries. Retrograde bypasses are typically constructed from the common
iliac arteries using prosthetic conduit and the target vessels are ligated proximal
to the revascularization to prevent endoleaks after placement of the aortic stent
graft. An iliac conduit is also fashioned to facilitate arterial access for the endovascular
portion of the repair. The second stage of the hybrid repair, consisting of aortic
endografting to exclude the aneurysm, can be performed at the same operative setting
or, as has become our practice, in a delayed fashion. By delaying the aortic endografting
portion of the procedure, the patient is given time to recover and the physiological
insult is distributed over time. The patients typically remain hospitalized between
the first and second stages.
One technical advantage of the hybrid repair is the ability to tailor the visceral
reconstruction to patient specific anatomy. For any individual patient, there may
be multiple potential bypass configurations and there are a number of important principles
to guide in selecting the most appropriate option: these are best illustrated through
the example of a recent patient treated at our institution. A 66 year-old female presented
with a 7cm Type III TAAA ([Figure 1]). She had previously undergone an aortic arch replacement as well as a TEVAR for
a descending aortic aneurysm and the remainder of the aorta was aneurysmal. She had
atrial fibrillation and chronic obstructive pulmonary disease and had previously suffered
from a pulmonary embolus, acute respiratory distress syndrome, and recurrent pneumonias.
She was not felt to be a candidate for traditional open repair and was accordingly
offered a hybrid procedure.
Figure 1. Three-dimensional reconstruction of CTA demonstrating previous aortic arch replacement
and TEVAR, presenting now with a Type III TAAA.
The visceral debranching was performed via a midline laparotomy to allow a direct
anterior approach to the renal and visceral vessels. The distal common iliac arteries
were isolated bilaterally to serve as inflow for the bypass grafts. When selecting
a site on the iliac artery to seat the proximal bypass anastomosis, care must be taken
to insure that there is sufficient artery proximal to the bypass to allow the iliac
limb of the endograft to obtain a distal seal. Although iliac artery based bypasses
are by-and-large the most common configuration, the native infrarenal aorta can be
used as a basis for the grafts, assuming it is free of aneurysmal disease. Alternatively,
if there is extensive aortic or aorto-iliac aneurysmal or occlusive disease, the infrarenal
aorta can be replaced with a tube graft sewn to the aortic bifurcation or with a bifurcated
graft to the common iliac arteries, and the visceral bypasses based on the replaced
segment. In either case, attention, again, must be paid to leave enough normal aorta
or proximal graft to provide an appropriate seal zone for the planned endograft.
After isolating the bypass inflow, the target vessels were isolated. Celiac artery
revascularization was accomplished via a bypass to the common hepatic artery and was
done in a manner that allowed preservation of hepatic and gastric perfusion. From
the anterior approach it is usually most straightforward to select the hepatic artery
as the recipient site for the bypass as this can be isolated in the omental bursa,
or lesser sac. The proximal celiac artery must also be dissected, but only enough
to allow the vessel to be ligated after completion of the bypass. By revascularizing
the hepatic artery but ligating the celiac trunk, the left gastric and splenic arteries
can continued to be perfused by retrograde filling of the proximal hepatic artery.
In this patient, the superior mesenteric artery (SMA) was isolated in the omental
bursa as it emerged from behind the pancreas. Although the SMA can also be easily
identified at the base of the transverse mesocolon, the former location has the advantage
of being well proximal to the origin of the middle colic artery, and allows for the
ligation of the SMA just proximal to the bypass. The left renal artery was isolated
proximally as it branched from the aorta and the right renal artery was identified
as it emerged from behind the vena cava; when preparing the renal arteries, care must
be taken to insure that the bypass is proximal to any early renal branches.
Prosthetic bypass grafts are used for the revascularization, which can be accomplished
with a variety of straight, bifurcated, and custom-branched graft configurations.
Target vessels can be revascularized individually or in series with jump grafts. In
this patient, bilateral bifurcated Dacron grafts were used to revascularize the right
renal and hepatic arteries from the right common iliac artery, and the left renal
and superior mesenteric arteries from the left common iliac artery. We have subsequently
changed our practice to use a jump graft from the SMA to hepatic artery; this allows
us to reserve one of the Dacron limbs for use as a conduit to facilitate subsequent
introduction of the stent-graft. The grafts are tunneled retroperitoneally so that
they are excluded from the peritoneum and separated from the bowel; the iliac conduit
is left buried in the lower abdominal subcutaneous tissue for later exposure.
Aortic endografting for exclusion of the aneurysm can be done concomitantly or in
a delayed fashion, and there is not significant data to unambiguously recommend one
approach over another[8]. Proponents of simultaneous procedures argue that the staged approach leaves the
patient susceptible to rupture in the intervening period, although there is little
data, other than anecdotal experience, to support this approach[9]. Our practice is to perform the stent-graft procedure after the patient has recovered
from the debranching, but on the same admission. The delayed approach offers the advantage
that it allows the patient to recover from the many of the physiological insults of
the debranching procedure before exposing them to the risks associated with the placement
of the stent grafts. This is especially important in terms of renal function, as it
allows the kidneys to recover from the ischemia associated with the bypass before
subjecting them to the nephrotoxic iodinated contrast necessary for the endovascular
portion of the case.
The patient presented here had a relatively uneventful post-operative course and returned
to the operating room on post-operative day 19 for endovascular stent-grafting. A
combination of thoracic endografts and a bifurcated abdominal device were used to
extend from the previous TEVAR into the proximal common iliac arteries ([Figure 2]). The choice of endovascular device should be based on surgeon experience and preference
and patient anatomy; there is no systematic advantage of one device over another for
this procedure. Because of the previous intervention and extent of planned coverage,
a spinal drain and somatosensory evoked potential (SSEP) monitoring were used for
spinal cord protection. The patient tolerated the procedure well and was discharged
home on post-operative day 9 from the stent-graft.
Figure 2. Three-dimensional reconstruction (A) and maximum intensity projection images (B)
of follow-up CTA, demonstrating excluded aneurysm and patent ilio-mesenteric and ilio-renal
bypasses.
Clinical Outcomes
The hybrid approach has many theoretic advantages ([Table 1]). Although certainly not a "minimally invasive" procedure, visceral disbranching
can be accomplished via a standard laparotomy, without entering the chest cavity,
and does not involve aortic cross-clamping or mechanical circulatory support. Because
of this, hybrid repair is thought to have the theoretical advantage of being "less-invasive"
than open TAAA repair and theoretically results in less physiological derangement.
It was initially hoped that hybrid repair would result in less coagulopathy, ischemia–reperfusion
injury, bacterial translocation, sepsis, end-organ damage, and renal failure, culminating
in reduced length of stay and lower morbidity and mortality. In practice, the results
of published series have been mixed ([Table 2]), leaving ambiguity as to the exact role for hybrid repairs[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24].
Table 1.
Theorectical advantages of visceral debranching and hybrid TAAA repair
|
Technical advantage
|
Theoretic benefit to patient
|
|
No thoracotomy
|
Decreased post-operative pain,- decreased pulmonary complications
|
|
No aortic cross-clamp
|
Less end-organ ischemia, minimized ischemia reperfusion injury, decreased renal faiulre,
decreased spinal ischemia
|
|
No need for cardiopulmonary support
|
Less hemodynamic instability, decreased coagulopathy
|
Table 2.
Major published series of hybrid TAAA repair with more than 15 patients
|
Author
|
Year
|
Patients
|
30-day Mortality
|
Permanent paraplegia
|
Endoleaks
|
Graft patency
|
Overall survival
|
Mean Follow-up (mos)
|
|
Zhou ([24])
|
2006
|
31
|
3.2%
|
0%
|
6%
|
95%
|
90%a
|
16
|
|
Black ([10])
|
2006
|
22
|
23%
|
0%
|
42%
|
98%
|
NA
|
9.5
|
|
Lee ([15])
|
2007
|
17
|
24%
|
0%
|
12%
|
96%
|
76%[a]
|
8
|
|
Van de Mortel ([23])
|
2008
|
16
|
31%
|
0%
|
13%
|
95%
|
69%[a]
|
13
|
|
Quinones-
Baldrich ([19])
|
2009
|
20
|
0%
|
6.6%
|
30%
|
100%
|
76%[a]
|
17
|
|
Donas ([11])
|
2009
|
58
|
8.6%
|
3.4%
|
17%
|
97%
|
74%[a]
|
22
|
|
Drinkwater ([9])
|
2009
|
107
|
15%
|
8.4%
|
33%
|
87%[*]
|
NA
|
NA
|
|
Patel ([18])
|
2009
|
23
|
17%
|
4%
|
23%
|
90%
|
68%[b]
|
6
|
|
Kabbani ([13])
|
2010
|
36
|
8.3%
|
0%
|
39%
|
93%
|
80%[a]
|
6
|
|
Patel ([17])
|
2010
|
29
|
3.4%
|
3.4%
|
34%
|
95%
|
NA
|
NA
|
|
Kuratani ([14])
|
2010
|
86
|
2.3%
|
0%
|
17%
|
99%
|
86%[c]
|
88
|
|
Smith ([21])
|
2011
|
24
|
12.5%
|
8.3%
|
12%
|
99%
|
NA
|
12
|
|
Hughes ([12])
|
2012
|
58
|
9%
|
4%
|
NA
|
95%
|
62%[c]
|
26
|
|
Tshomba ([22])
|
2012
|
52
|
14%
|
1.9%
|
7.7%
|
93%
|
77%[d]
|
24[**]
|
|
Rossett ([20])
|
2014
|
76
|
34%
|
11%
|
3%
|
99%
|
NA
|
30
|
|
Massoni ([16])
|
2014
|
45
|
24%
|
13%
|
NA
|
79%
|
45%[e]
|
26
|
* 30 day
** median
a at end of follow-up period
b 1-year Kaplan-Meier survival
c 5-year Kaplan-Meier survival
d freedom from aortic- realted deaths at end of follow-up
e 6-year Kaplan-Meier survival
In 2009 Quinones-Baldrich and colleagues from the University of California Los Angeles
(UCLA) reported the follow up for their first patient and the results of their overall
experience in 20 cases[19]. At 10 years, the index patient for hybrid repair was alive and well without the
need for further intervention and no aneurysm related morbidity. Among a mixed cohort
of high-risk patients who underwent hybrid procedures for aortic arch pathology (three
patients) and thoracoabdominal and juxtarenal aortic aneurysms (17 patients), they
reported nine major complications in six patients (32%), one case of permanent paraplegia
(out of 15 patients at risk, 6.6%), and no perioperative mortality (0%). With a mean
of 16 months of follow- up they identified 3 endoleaks (30%, one type I, three type
II), and no bypass thrombosis or aneurysm enlargement. There were two reinterventions
(10%, one type I and one type II endoleak). The overall survival was 76% at 2 years.
Based on their durable outcomes, low morbidity and mortality, and little need for
reintervention, the UCLA group concluded that hybrid repair offered significant advantages
to traditional open repair, especially for high-risk patients.
Hughes and colleagues from Duke report similar outstanding outcomes[12]. Among 58 patients who underwent hybrid repair, there was an 9% rate of perioperative
mortality and a 4% rate of permanent paraplegia. Interestingly, the paraplegia rate
was nil among the final 25 patients in the series, all of whom had staged repairs.
Those patients who underwent staged repairs also had shorter combined operative times,
decreased intraoperative transfusions, and were more likely to be extubated in less
than 24 hours than those who had single-stage procedures. With a median follow up
of 26 months, there was a 95% graft patency rate, and all thromboses were clinically
silent. There was no intervention for required for any endoleaks.
The excellent early results achieved in these studies are complemented by recent publications
attesting to the durability of the visceral artery bypass grafts and low rates of
late aortic related death. In 46 patients with 164 grafts, Shaherdan and colleagues
report an 86% 5-year primary patency rate[25]. Patency at 5-years for individual bypasses ranged from 69% for the right renal
artery to 100% for the hepatic artery, with the left renal and SMA between 87% and
88%. Among the 32 patients surviving past the perioperative period, there 6 deaths
due to procedure related respiratory failure, and only 2 deaths due to aortic or branch
vessel complications.
The experience reported by Patel and colleagues from Massachusetts General Hospital
(MGH) stands in stark contrast. A recognized high volume center for aortic surgery,
MGH has traditionally reported outstanding results with open TAAA repair[26]. When Patel and colleagues examined the outcomes of hybrid procedures in 23 high-risk
individuals who were not candidates for open TAAA (type I-III) repair, they reported
a 4.3% rate of permanent paraplegia and a 26% in-hospital mortality rate, all of which
were higher than what they observed in their contemporaneous open experience (3.9%
permanent paraplegia and 10% mortality)[18]. With a mean follow-up period of 166 days, there were 7 graft thromboses out of
70 grafts (10%) and five endoleaks (22%, three type I and two type II), three of which
required reintervention. The poor results obtained in the patients who underwent hybrid
repair lead the MGH group to conclude that the morbidity and mortality profile should
limit the use of the hybrid repair, and that many patients unfit for open repair were
simply unfit for surgical intervention of any kind.
This sobering appraisal is further supported by a study from the North American Complex
Abdominal Aortic Debranching Registry that demonstrated a 14% SCI rate among 159 patients
treated at a total of 13 institutions[27]. This rate significantly exceed that which has been reported for open repair in
most specialized aortic centers[1]
[2], and suggests that despite avoiding many of the issues related to spinal cord perfusion
in traditional open repairs, hybrid approaches are not able to improve upon those
outcomes.
The different outcomes obtained in these studies likely reflect differences in the
underlying cohorts between the centers and highlight the importance of patient selection
in hybrid repair. In examining the specifics of the patient cohorts, the MGH cohort
appears to be higher risk with regard to patient comorbidities and physiology, suggesting
that although hybrid repair may offer some advantage in morbidity and mortality, it
is certainly not a low-risk endeavor. Despite the outstanding outcomes of the UCLA
and Duke groups, the MGH experience demonstrates that there are, in fact, limits to
how far the hybrid technique can be pushed, and that in extremely high-risk patient
populations, a non-operative approach may be advisable.
The largest single series in the literature is a multi-intuitional European study
reported by Drinkwater and colleagues[9]. Reporting on 107 consecutive hybrid repairs, they had a 93% technical success rate,
an 8% rate of permanent paraplegia, and a 15% 30-day mortality rate. There was 86%
graft patency at 30-days and an initial endoleak rate of 33%. Although these results
are quite similar to those reported by MGH, Drinkwater et al. offer an alternative
interpretation of their experience. They highlight that the even though their cohort
is high risk, based on the patients' comorbidities and the fact that many were previously
denied open repair, their outcomes compare favorably those reported in the non-selected,
real-world studies of open TAAA repairs, where morbidity ranges from 19% to 23%[3]
[28]. This leads them to conclude that despite the fact that hybrid TAAA carries a significant
associated morbidity and mortality, it offers a viable alternative to traditional
open repair in high risk patients.
A recent meta-analysis of 19 studies encompassing 507 patients demonstrated similar
results[29]. The authors report a pooled rate of 30-day mortality at 12%, a pooled rate of permanent
paraplegia of 4.5%, and a pooled rate of renal insufficiency of 8.8%. During a mean
follow-up period of 34 months, the graft patency was 96% and there was a 23% rate
of endoleak, with 27% of those patients requiring reintervention. Because a pooled
analysis was performed, these results are likely heavily biased by those of Drinkwater
and colleagues, which comprised 20% of the patients[9], but, nonetheless, provide the best estimates of the morbidity and mortality associated
with hybrid TAAA repair. Based on these data, hybrid repairs carry a significant,
but not necessarily prohibitive, associated morbidity and mortality, especially when
considering the overall risk profile of the patients being offered this type of intervention.
Current role in TAAA Management
Based on the available literature and our clinical experience, careful patient selection
is key. Although hybrid repairs may allow high-risk patients to have outcomes equal
to those published from administrative surgical databases, they do not approach those
reported at select centers of excellence with significant expertise in open TAAA repair,
even when the hybrid repairs are performed at the same high volume centers. Although
they may offer a viable alternative for patients at slightly higher than average operative
risk, or for those who have isolated high risk comorbidities (i.e. chronic lung disease),
patients who are deemed unfit for open TAAA repair are likely unfit for hybrid repair
as well.
Not all "high-risk" patients, however, are the same. In relatively healthy individuals
who are considered high risk due to anatomic features, the hybrid approach offers
a reasonable treatment option. In fact, hybrid repair may have the most to offer as
a re-operative approach to recurrent TAAA. By allowing the surgeon to avoid densely
scarred operative fields and obliterated tissue planes, the hybrid approach facilitates
repair for individuals who would otherwise not have a viable open approach to their
pathology. This is best demonstrated by both the Duke and UCLA experiences, in which
they reported outstanding outcomes in a cohort for which the hybrid repair constituted
a repeat aortic operation (55% to 60% of the patients)[19]
[30]. This has been reinforced by our experience, in which the best outcomes are achieved
in patients who have specific anatomic reasons that make them unsuitable for open
TAAA repair.
In the end, it is clear that despite many theoretical advantages, and the impression
that it is less invasive than standard open TAAA repair, hybrid repair is still a
significant undertaking with real risks of associated morbidity and mortality. Although
there is a clear role for hybrid repair in patients who are good physiological candidates
for operative intervention, but have specific anatomical challenges that preclude
traditional open repair, hybrid repair has little to offer for the true physiologically
high-risk patient. It should be remembered that hybrid repair is no less of an operation
than traditional open repair, it is just a different one.
