1/1页9314: 跳转到查看:13152
发新话题 回复该主题
键盘左右键可以进行前后翻页操作
【有 0 人在聊天】 帮助

Arch Aneurysm Repair With Long Elephant Trunk

Arch Aneurysm Repair With Long Elephant Trunk

Arch Aneurysm Repair With Long Elephant Trunk:
A 10-Year Experience in 111 Patients
Koichi Toda, MD, Kazuhiro Taniguchi, MD, Takafumi Masai, MD,
Toshiki Takahashi, MD, Satoru Kuki, MD, Yoshiki Sawa, MD; Osaka Cardiac Surgery
Research (OSCAR) Group
Department of Cardiovascular Surgery, Japan Labor Health and Welfare Organization Osaka Rosai Hospital, Sakai, Department of
Cardiovascular Surgery, Rinku General Medical Center, Izumisano, Department of Cardiovascular Surgery, National Hospital
Organization, Osaka National Hospital, Osaka, Department of Cardiovascular Surgery, Takarazuka City Hospital, Takarazuka,
Department of Cardiovascular Surgery, Osaka University Graduate School of Medicine, Suita, Japan
Background. We studied the long-term outcome of arch
aneurysm repair with a long elephant trunk (LET) anastomosed
at the base of brachiocephalic artery.
Methods. Between 1998 and 2008, 111 patients underwent
arch aneurysm repair with LET. A 4-branched graft
was sutured to the sinotubular junction, the distal ascending
aorta transected, and a LET inserted into the
aortic arch while selective cerebral perfusion was maintained.
The graft distal end was anastomosed to the LET,
incorporating the distal ascending aorta, and arch vessels
were anastomosed to graft branches.
Results. Concomitantly, 33 patients (30%) underwent
other cardiac procedures, including 11 aortic root replacements.
Two patients died (1.8%) within 30 days and 7
died (6.3%) after 30 days. Perioperative morbidity included
2 (1.8%) with stroke, 3 (2.7%) with paraplegia, and
1 (0.9%) with paraparesis. Postoperative computed tomography
scans revealed complete aneurysmal thrombosis
around the LET in 88 patients (79%), who were
monitored without a second-stage procedure. Among 23
patients with incomplete thrombosis, 19 underwent a
second-stage procedure to complete distal fixation of the
LET. Overall survival was 88%, 83%, and 75%, at 1, 3, and
5 years after aneurysm repair with the LET. No aneurysm
rupture or reexpansion occurred in patients with complete
thrombosis. Four patients with incomplete thrombosis
died of rupture before the second-stage procedure.
Conclusions. Our results demonstrated safety and good
durability of the LET technique and suggest that this
technique is a simple and safe procedure that is applicable
to a variety of arch aneurysms.
(Ann Thorac Surg 2009;88:16 –22)
© 2009 by The Society of Thoracic Surgeons
An extensive aortic arch pathology involving the descending
aorta remains a surgical challenge and an
optimal technique remains controversial. A single-stage
procedure with a bilateral anterior thoracotomy carries
substantial morbidity due to respiratory complications,
although it is an excellent procedure permitting access to
the entire thoracic aorta [1]. A two-stage approach with
an elephant trunk (ET) anastomosed in the distal aortic
arch was introduced by Borst [2] as a less invasive
procedure; however, the cumulative risk of two major
procedures and the additional risk of rupture between
the two procedures have been demonstrated [3].
In 1998 we introduced a total arch replacement (TAR)
with a long elephant trunk (LET) anastomosed at the
base of the brachiocephalic artery (BCA) [4]. This technique
yields a secure and rapid anastomosis in the distal
ascending aorta, where anastomosis and hemostasis can
be performed without difficulties. Previous findings presented
by our group suggest that the technique does not
require subsequent distal anastomosis and could transform
a two-stage ET procedure into a single-stage repair
when complete aneurysmal thrombosis around the LET
is achieved [5].
We have applied this technique for a wide variety of
aortic pathologies, including acute type A [6] and
chronic type B dissection [7], and have also combined
this technique with aortic root replacement [8]. We
review our experience with 111 patients during a
10-year period and analyze early outcome as well as the
long-term safety and effectiveness of our arch aneurysm
repair technique that uses an LET anastomosed at the
base of BCA.
Patients and Methods
This retrospective review of the clinical records was
approved by the appropriate Institutional Review Boards
of the participating institutions, and informed consent
was provided by each patient.
Patients
Between October 1998 and September 2008, 111 patients
with an arch aneurysm underwent TAR with an LET in
Osaka Rosai Hospital and affiliated institutions (Rinku
General Medical Center, National Hospital Organization
Osaka National Hospital, and Takarazuka City Hospital).
The baseline patient clinical profiles including associated
comorbidity are reported in Table 1. The mean patient
age was 69 years, and 11 patients (10%) were older than
80 years at the time of the operation. Among 25 patients
with significant coronary artery disease (CAD), 6 patients
had a history of percutaneous coronary intervention and
9 received concomitant coronary artery bypass grafts
(CABG). Three patients required hemodialysis because
of chronic renal failure, and 3 patients had a poor left
ventricular ejection fraction of less than 0.35.
Surgical Procedures
We have previously reported details regarding the
present surgical and perfusion technique of TAR with an
LET [4, 9]. Briefly, the ascending aorta and arch vessels
are minimally dissected using a median sternotomy;
then, 8-mm Dacron grafts (Gelweave, Vascutek, Germany)
are anastomosed to right and left axillary arteries
below the clavicle. The patient is started on cardiopulmonary
bypass (CPB), with the arterial return to an 8-mm
Dacron graft anastomosed to the right axillary artery
and venous drainage from the bicaval cannula, then
cooled to 25°C.
While the patient is being cooled, the heart is arrested
by antegrade and retrograde cold blood cardioplegia,
after which a 4-branched arch graft (Hemashield Platinum,
MAQUET Cardiovascular LLC, Wayne, NJ) is sutured
into the sinotubular junction (Fig 1A). The size of
the LET is determined by the size of the descending aorta
in preoperative computed tomography (CT) scans. We
measure the inner diameter of the nonaneurysmal descending
aorta and choose a graft one size smaller for
patients with a nondissecting aneurysm. In acute aortic
dissection cases, we choose a LET that is two sizes
smaller to reduce the friction between the graft and aortic
wall.
When the patient has been cooled to 25°C, systemic
perfusion is stopped and selective cerebral perfusion
(SCP) is maintained at a flow rate of 10 mL/kg/min using
the 8-mm Dacron graft anastomosed to the right and left
axillary arteries, with an additional 3-mm small cannula
in the left common carotid artery. The ascending aorta is
then transected at the base of the brachiocephalic artery
(BCA), and the LET is placed into the aortic arch and
descending aorta (or the true lumen of the dissecting
aorta) by pulling the LET with a catching catheter (Amplatz
Goose-Neck Snare, Microvena, MN) introduced
from the femoral artery (Fig 1B). For each patient, we
determine the length of the LET using a CT angiogram
obtained before the operation.
A distal anastomosis is then performed between the
4-branched arch graft and LET tube graft incorporating
the distal ascending aorta (Fig 1C). Before completing the
distal anastomosis, we wash debris from the LET by
perfusing the descending aorta through a small arterial
cannula placed in the femoral artery. Antegrade distal
perfusion is resumed from the side branch of the graft
and the patient is rewarmed. During rewarming, the BCA
and left common carotid artery are divided and anastomosed
to the branches of the graft, and the most distal
branch is anastomosed to the left axillary artery graft,
which is introduced into the pericardial space through a
retroclavicular tunnel. The proximal left subclavian artery
is then closed with clipping or sutures (Fig 1D).
Before closing the chest, the branched graft is entirely
covered with an expanded polytetrafluoroethylene sheet.
Follow-Up
Patients who demonstrated complete thrombosis of the
arch aneurysm in postoperative CT scan images were
discharged and monitored as outpatients. Patients who
did not demonstrate complete thrombosis required a
second-stage procedure, which we have previously described
[4]. Briefly, the descending aorta is exposed
through an anterolateral left thoracotomy. The descending
aorta is clamped under a normothermic femorofemoral
venoarterial bypass, and the LET is pulled out and the
distal end is anastomosed to the descending aorta. In
those patients, preoperative variables were investigated
by review of clinical records, and postoperative variables
were investigated by a review of outpatient records. A
cross-sectional follow-up examination was conducted in
December 2008 by telephone interviews.
Data Analysis
Data were analyzed using StatView 5.0 software (SAS
Institute, Cary, NC). Values are expressed as the mean 
standard deviation. Univariate analyses were conducted
for patient demographics and considered perioperative
risk factors, as summarized in Tables 2 and 3. The Fisher
exact test was used for categoric variables and a Mann-
Whitney U test was used to compare continuous variables.
Survival curves were generated using Kaplan-
Meier methods and compared using the log-rank test. A
value of p  0.05 was considered significant.
Results
Our technique was uniformly applicable for a wide variety
of aortic arch aneurysms, including degenerative in 81
(73%), chronic dissecting in 22 (20%), and acute dissecting
in 8 (7%). The aneurysm in 39 patients (35%) involved the
ascending and arch aorta, 45 patients (41%) had an arch
to distal arch aneurysm, and the aneurysm in the remaining
27 patients extended beyond the level of the carina
(Table 1).
The LET was a mean length of 15.9  3.1 cm (range, 10
to 22 cm) and had a diameter of 24  2 mm (range, 18 to
30 mm). The distal end of the LET, which was marked
with a metal clip, was confirmed in CT scan as located
between the Th5 and Th10 levels (mean, 7.5  1.2).
Thirty-three patients (30%) underwent 35 concomitant
procedures, including aortic root replacements in 11,
aortic valve replacements in 8, CABG in 9, reconstruction
of the left vertebral artery in 5, aortic and mitral valve
replacement in 1, and patch closure of an aortopulmonary
artery fistula in 1. The mean duration values in
minutes for operative details were aortic cross clamp
time, 100  45; CPB, 210  58; hypothermic circulatory
arrest of the lower body for open distal anastomosis, 27 
10; and antegrade SCP time 96  29.
Two patients (1.8%) died in-hospital, at 4 and 7 days
postoperatively, of a thoracoabdominal aortic aneurysm
(TAAA). There were 7 (6.3%) postoperative hospital deaths,
including 3 patients who died of pneumonia at 3 to 5
months, 2 who died of mediastinitis at 2 and 3 months, 1
who died of a TAAA rupture at 45 days, and 1 who died
of sepsis due to cholecystitis at 2 months.
No phrenic nerve palsy or new recurrent nerve palsy
developed. Extubation occurred within 24 hours after the
procedure in 69% of the patients. Postoperative morbidity
included 2 strokes (1.8%), and permanent spinal cord
ischemia (SCI) was documented in 4 patients, including 3
with paraplegia and 1 with paraparesis. SCI was transient
in 5 patients, and resolved within 24 hours after cerebrospinal
fluid drainage and administration of naloxone
(Table 4). Univariate analyses of the influence of periop-
erative factors on postoperative SCI identified a significant
association between the level of the distal end of the
LET and SCI (Table 2).
Postoperative CT scans revealed complete aneurysmal
thrombosis around the LET in 88 patients (79%), who were
followed up without a second-stage procedure. Among 23
patients with incomplete thrombosis, 4 died of a ruptured
aneurysm before the second-stage procedure, and 19 underwent
a second-stage procedure to complete distal fixation
of the LET. The second-stage procedure in 13 of these
patients was during the same admission as the first-stage
procedure. The mean duration between the first and second
procedures for the 13 patients was 11  9 days, and all
received an anastomosis between the distal end of the LET
and descending aorta through a left thoracotomy. In those
13 patients, the operation and the CPB times were 157  55
and 43  17 minutes, respectively.
The remaining 6 patients with incomplete thrombosis
each underwent a second-stage procedure after follow-up
CT scans revealed expansion of a descending or TAAA. The
mean duration between the first and second procedures
was 511  336 days. Three received an anastomosis between
the distal end of the LET and descending aorta
through a left thoracotomy, and the other 3 underwent
TAAA repair using the branched graft.
The average follow-up was 3326 months (range, 4 days
to 94 months), and overall survival was 88%, 83%, and 75%
at 1, 3, and 5 years after aneurysm repair with the LET.
Rupture or reexpansion of the aneurysm was not encountered
in those with complete thrombosis, but 4 of the 23
patients with incomplete thrombosis died of a ruptured
aneurysm before undergoing a second-stage procedure
(Table 5). One patient who refused the operation and died
after discharge was considered as a late death. Aside from
this patient, no other late death resulted from rupture of a
treated aortic arch aneurysm (Table 6).
Table 2. Univariate Analysis of Association Between Patient
Risk Factors and Spinal Cord Ischemia
Variables
No SCI
(n  102)
Mean  SD
or No. (%)
SCI
(n  9)
Mean  SD
or No. (%) p Value
Age, y 69  10 71  10 0.43
CAD 23 (22.5) 2 (22.2) 0.99
Diabetes mellitus 8 (7.8) 1 (11.1) 0.55
Hypertension 55 (53.9) 5 (55.6) 0.99
Cerebrovascular disease 19 (18.6) 0 (0) 0.35
COPD 27 (26.4) 1 (11.1) 0.44
Renal failure 9 (8.8) 1 (11.1) 0.59
Previous AAA repair 19 (18.6) 2 (22.2) 0.68
Aneurysm diameter, mm 58  12 55  10 0.58
Dissecting aneurysm 28 (27.5) 2 (22.2) 0.99
Extension beyond carina 25 (24.5) 2 (22.2) 0.99
CPB time, min 210  59 210  49 0.81
Open distal time, mina 27  10 23  11 0.16
SCP time, min 96  29 93  22 0.86
LET length, cm 15.8  3.2 16.8  2.1 0.47
LET diameter, mm 25 2 24 2 0.91
End of LET (Th) 7.5  1.1 8.4  1.2 0.03
a Hypothermic circulatory arrest time of the lower body for open distal
anastomosis.
AAA  abdominal aortic aneurysm; CAD  coronary artery disease;
COPD  chronic obstructive pulmonary disease; CPB  cardiopulmonary
bypass; LET  long elephant trunk; SCP  selective
cerebral perfusion; SD  standard deviation.
Table 3. Univariate Analysis of Association Between Patient
Perioperative Factors and Complete Aneurysmal Thrombosis
Around Long Elephant Trunk
Variables
Thrombosed
(n  88)
Mean  SD;
No. (%)
Non-thrombosed
(n  23)
Mean  SD;
No. (%) p Value
Age, y 69 9 71  12 0.099
Aneurysm diameter,
mm
56  11 66  13 0.0016
Diameter  7 cm 5 (6) 7 (30) 0.0027
Extension beyond
carina
13 (15) 14 (61) 0.0001
Dissecting aneurysm 25 (28) 5 (22) 0.607
Degenerative
aneurysm
63 (72) 18 (78) 0.607
End of LET (Th) 7.6  1.2 7.5  1.2 0.9772
LET  long elephant trunk; SD  standard deviation.
Table 4. Postoperative Mortality and Morbidity
Variable No. (%)
Early death (30 d) 2 (1.8)
Hospital death 7 (6.3)
Cerebral infarction 2 (1.8)
Spinal cord ischemia
Transient (resolved  24 h) 5 (4.5)
Permanent 4 (3.6)
Mechanical ventilation
24 h 77 (69)
24 to 48 h 13 (12)
48 21 (19)
Table 5. Features of Aneurysmal Ruptures Before
Second-Stage Procedure
Extension of
Aneurysm
Aneurysm
Size, mm
Thrombosis
Around
LET
Rupture
After
LET, d
1. Arch to distal
descending
67 Incomplete 4
2. Ascending to distal
descending
70 Incomplete 7
3. Ascending to distal
descending
88 Incomplete 45
4. Ascending to distal
descending
80 Incomplete 2340
LET  long elephant trunk.
Ann Thorac Surg TODA ET AL 19
2009;88:16–22 ARCH REPAIR WITH LONG ELEPHANT TRUNK
ADULT CARDIAC

TOP

 

A Kaplan-Meier curve demonstrated that there was no
significant difference in overall survival between patients
with complete aneurysmal thrombosis without a secondstage
procedure and those without complete thrombosis
who underwent a second-stage procedure (Fig 2). However,
4 patients without complete aneurysmal thrombosis
who could not accomplish a second-stage procedure had
significantly poor survival.
The geometry and pathology of the thoracic aorta was
compared between patients who demonstrated complete
aneurysmal thrombosis and those who did not (Table 3).
The maximum diameter of the aneurysm was significantly
smaller in patients with complete aneurysmal
thrombosis, and aneurysmal extension beyond the
level of the carina was significantly less frequent in
those patients with complete thrombosis. Univariate
analysis found no significant association between the
type of aneurysm and complete aneurysmal thrombosis
around the LET.
Comment
A direct surgical approach to treat pathologic conditions of
the aortic arch has become an established procedure, with
satisfactory results and durability [10]. However, performing
anastomosis and hemostasis in the deep mediastinum
beyond the left subclavian artery is not always straightforward
because the tissues are fragile and visualization is
poor. The depth of a distal anastomosis is a risk factor for
prolonged distal anastomosis [11]. To overcome these difficulties,
we modified Borst’s ET technique [2], and introduced
TAR with an LET anastomosed at the base of the
BCA as a less invasive procedure. In the present retrospective
study of 111 patients, we found that our LET technique
could be applied for a variety of aortic arch pathologies with
low rates of operative mortality (1.8%) and stroke (1.8%)
compared with recent reports from other large series
[10–13].
Neurologic injury is the principal cause of serious morbidity
and a significant risk factor for in-hospital death after
conventional total arch replacement [14]. We used antegrade
SCP in which several different techniques, including
direct balloon catheter perfusion of the BCA and the left
carotid artery [14], as well as right axillary cannulation with
balloon catheters in the left carotid and left subclavian
arteries [13] have been proposed. We avoid insertion of a
balloon catheter into the BCA, because malposition of a
balloon catheter in that artery is not rare [15]. Instead we
use bilateral axillary artery grafts and a small cannula in the
left carotid artery for SCP.
We choose direct cannulation into the left carotid artery,
except in patients with acute aortic dissection. With our
technique, the cannula is placed in the left carotid artery at
5 cm distal from the aortic arch to avoid manipulation of
atheromatous plaque near the orifice of arch vessels. To
avoid a cerebral embolism associated with cannulation in
the left carotid artery, cannulation is performed after establishing
SCP from right and left axillary arteries, so that
plaque possibly dislodged by cannulation can be flushed to
the aortic arch by collateral back flow.
For additional safety in regard to cerebral circulation,
we use near infrared spectroscopy to monitor cerebral
oxygen saturation. When a difference in cerebral oxygen
saturation is found between the left and right hemispheres,
we check flow in the left and right carotid
arteries by a transesophageal echocardiography (TEE), as
well as left and right radial artery pressure, to rule out
problems with the SCP circuit and cannulation.

TOP

 

In our technique, the branch grafts to the arch vessels are
longer than 8 cm because the aortic arch is left in place and
grafts branch off the main tube graft, which is anastomosed
at sinotubular junction. We anastomose the 4-branched
graft to the sinotubular junction in such a manner that 3 of
the grafts branch off from the great curvature of the main
tube graft, by which they lay on the ventral side of the SVC
and innominate vein. No kinking or occlusion of the grafts
to arch vessels ocurred, except for a patient in which
occlusion of the graft to the left subclavian artery was found
in a CT angiogram 3 years after the operation. Spielvogel
and colleagues [12] reported use of a trifurcated graft for
total arch replacement. Their graft was as long as ours, and
they found no kinking or thrombosis of the graft in a series
of 109 patients.
Upon awaking from anesthesia, 9 (8.1%) patients had
paraplegia or paraparesis and were treated by increasing
mean arterial pressure up to 100 mm Hg, spinal drainage,
and naloxone administration. Five recovered completely
within 24 hours, but paraparesis developed in 1 patient and
paraplegia in 3. No delayed paraplegia developed. A relatively
high incidence of paraplegia has been reported to be
associated with a frozen ET procedure [16], although the
mechanism of SCI appears to be multifactorial and remains
poorly defined. Our analysis of the associations between
patient risk factors and SCI revealed that the distal end of
the LET was significantly deeper in patients with SCI (Table
2). The concept of collateral network circulation in the
spinal cord has been proposed, and the importance of
inputs into this network, including the subclavian arteries,
hypogastric arteries, and segmental vessels, has been recognized
[17].
To maintain collateral network circulation, we perfused
the left subclavian artery when lower body circulation was
stopped for open distal anastomosis. In addition to the
importance of maintaining stable hemodynamics during
and after LET placement to maintain sufficient collateral
network circulation, extensive deployment of an
LET should be avoided. Recently, in addition to using CT
angiogram findings to determine the precise length of
the LET in each patient, we use intraoperative TEE to
confirm that the end of the LET is not placed beyond the
level of Th 8. We also place the suture onto the lesser
curvature of the LET beforehand to fix its length and
prevent over-stretching when the graft is pulled down to
the aortic arch. By following these technical refinements,
we have not had a patient with postoperative SCI.
Overall survival was 88%, 83%, and 75% at 1, 3, and 5
years after aneurysm repair with the LET, which is comparable
with other large series of total arch replacement [1, 3,
10, 13]. We found no significant difference in survival
between patients with complete aneurysmal thrombosis
without a second-stage procedure and those without complete
thrombosis who underwent a second-stage procedure.
Furthermore, we did not encounter rupture or reexpansion
of the aneurysm in patients with complete
aneurysmal thrombosis. Although the mean follow-up period
may not be adequate for firm conclusions, this result
suggests that our technique does not require a second-stage
procedure when the LET results in thrombus exclusion of
the arch aneurysm.
On the other hand, 4 patients without complete aneurysmal
thrombosis who could not accomplish a second procedure
had poor survival, and 3 died of ruptured aneurysm
early after the first-stage procedure (Table 5). We speculate
that early rupture might be related to localized aortic wall
weakening due to hypoxia, which could be induced by
incomplete thrombosis [18]. We consider that incomplete
aneurysmal thrombosis is mainly attributable to incomplete
distal sealing of the trunk, because complete thrombosis
around the LET was seen in patients who underwent
distal fixation of the LET without ligating the intercostal
arteries during the second-stage procedure. Univariate
analysis (Table 3) demonstrates that complete thromboexclusion
of the arch aneurysm is less likely with an LET alone
when the maximum diameter of the aneurysm exceeds 7
cm or the aneurysm extends beyond the level of the carina.
Therefore it is necessary to confirm the thromboexclusion
of the aneurysm early after the LET procedure in the
patients with a large and extensive aneurysm.
When thromboexclusion is not anticipated, a rapid twostage
procedure [4] or replacement of the entire thoracic
aorta or entire aorta using both mediastinal and thoracoabdominal
incisions [19] may be necessary. In this series, 7
patients underwent distal fixation of the LET at the descending
aorta through a left thoracotomy as a rapid twostage
procedure within 1 to 7 days after the first procedure.
Distal fixation of the LET could be accomplished by transcatheter
placement of an endovascular stent graft [20], as
the LET could serve as a useful proximal landing zone for
stent graft placement. Considering the progress in endovascular
technology, this combined procedure may play an
important role in treatment of extensive thoracic aneurysm
with a suitable anatomy.
In conclusion, our results demonstrated safety and
good durability of LET technique and suggest that this
technique is a simple and safe procedure that is
uniformly applicable for qualified patients with a wide
variety of aortic arch aneurysm.

TOP

 
1/1页9314: 跳转到
发表新主题 回复该主题
首页   |    联系我们   |   加入我们    |    关于我们   |   友情链接    |   网站地图   |   企业邮箱