Results
Deep Hypothermia versus Moderate Hypothermia
Deep hypothermic circulatory arrest (DHCA) is considered by many experts to be the
standard of care for surgical repair of AAAD.[5]
[6] Deep hypothermia decreases brain metabolism by approximately 50% per 6°C drop in
organ temperature and enables full neurologic recovery after the interval of interruption
in brain perfusion.[7]
[8] However, due to prolonged duration of cardiopulmonary bypass (CPB) typically associated
with the profound hypothermia induced by DHCA, increased complication rates including
postoperative bleeding, endothelial dysfunction, neuronal apoptosis, and postoperative
pulmonary complications are reported to exist.[5] In addition, DHCA induces vasoconstriction and decreases regional cerebral blood
flow.[5] The combination of complications and unfavorable physiologic changes associated
with DHCA have led to a gradual shift toward using moderate degrees of hypothermia
(MH).[5]
Recent studies have found MH to be independently associated with lower risk of mortality
and major adverse cardiac and cardiovascular events during AAAD repair.[5]
[9] A retrospective cohort study by Algarni et al[5] of 128 patients compared the two strategies of cooling (DHCA, <20°C; and MH, 22–28°C)
to repair AAAD at a single center, and their results are shown in [Table 1]. Algarni et al[5] reported significantly higher rates of stroke with persistent neurologic deficit
(21 and 13%, p = 0.042) and low cardiac output syndrome (26 and 5%, p < 0.001) in the DHCA group compared with the MH group, respectively. Mortality was
almost two-fold higher in the DHCA group than the MH group (28 and 16%, p = 0.07).[5] However, in addition to these findings, CPB time and blood transfusion were significantly
higher in the DHCA group than the MH group (p = 0.04).
Table 1
Summary of study findings on hypothermia
|
Study
|
Patient
|
Finding
|
|
Algarni et al[5]
|
Risk of stroke, low cardiac output syndrome, and mortality between medium and deep
hypothermia
|
Significantly higher rates of stroke with persistent neurologic deficit (21 and 13%,
p = 0.042) and low cardiac output syndrome (26 and 5%, p < 0.001) with profound hypothermia compared with moderate. Mortality was almost two-fold
higher in the profound group than the moderate group (28 and 16%, p = 0.07)
|
|
Stamou et al[10]
|
Survival rates of 324 patients undergoing AAAD repair with either DHCA, retrograde
or anterograde cerebral perfusion
|
No significance between types of cerebral protection used. Predictors of operative
mortality were hemodynamic instability and CPB time
|
|
Bakhtiary et al[12]
|
Clinical results of 120 patients undergoing AAAD repair with mild systemic hypothermia
|
Permanent neurologic deficits were seen in 4.2% of patients. The 30-day mortality
rate was 5%. Follow-up of 2.8 years showed a survival rate of 87%
|
Abbreviations: AAAD, acute Type- A aortic dissection; CPB, cardiopulmonary bypass;
DHCA, deep hypothermic circulatory arrest.
These findings raise questions as to whether hypothermic temperature may also play
a confounding role, as supported by a recent study by the senior author.[10] We compared survival between 324 patients undergoing AAAD repair with either DHCA,
retrograde, or anterograde cerebral perfusion.[10] Using multivariable logistic regression, we found that independent predictors of
operative mortality were hemodynamic instability and CPB time, not type of cerebral
protection used.[10] The strongest negative effect of DHCA originates from increased CPB times and subsequent
length of operation in comparison with MH.[11]
[12]
[13] Extended CPB times during cardiac surgery are implicated in increased risk of acute
renal insufficiency, stroke, and mortality.[13]
[14]
[15] These effects can be compounded based on the condition of the patient. Diminished
hematocrit and glycemic levels can increase perioperative risk during the use of CPB.[16]
[17] In our study, the median CPB time was 219 minutes for the DHCA group and 173.5 minutes
for the MH group (p < 0.001).[13] Also, the number of patients reaching the extended CPB time of 240 minutes in the
DHCA group tripled that of the MH group (p < 0.015). An increased prevalence of postoperative risk found using DHCA might actually
arise secondary to increase cardiopulmonary bypass times.[13] However, limitations of moderate hypothermia may include higher risk of injury to
the distal organs secondary to warmer temperatures, especially if body arrest time
is prolonged. Further, patients who need more complex repairs, such as total aortic
arch replacement may be better served by DHCA, or moderate hypothermia with dual perfusion
of the brain via axillary artery and the body via the femoral artery.
Although additional studies are needed to investigate this controversy, MH with selective
antegrade cerebral perfusion seems to be a safe strategy that accomplishes excellent
outcomes with relatively low rate of neurologic complications and lower CPB times
compared with DHCA. Both techniques, however, should aim toward limiting cardiopulmonary
bypass times with efficient planning of the operative steps, such as completing the
aortic root repair while cooling the patient to hypothermia.
Axillary versus Femoral Cannulation
Considerable debate remains regarding the optimal cannulation site in patients undergoing
AAAD repair, specifically comparing clinical outcomes of axillary artery cannulation
(AXC) with femoral artery cannulation (FAC).[18]
[19]
[20]
[21]
[22] Concern of flow reversal in the thoracoabdominal aorta with FAC exists and has contributed
to a trend of using the AXC site for CPB.[23] Recent studies focus on early and late outcomes of AXC versus FAC and inform the
year-old controversy.[18]
[20]
A meta-analysis by Ren et al reviewed nine nonrandomized studies comparing outcomes
in patients undergoing AAAD repair with AXC or FAC, with results shown in [Table 2].[18] Fixed-effect modeling showed significantly lower incidence in short-term mortality
(odds ratio [OR] = 0.25; 95% confidence interval [CI]: 0.15–0.42; p < 0.01) and in neurologic dysfunction (OR = 0.46; 95% CI: 0.29, 0.72; p < 0.01) in the AXC group. A meta-analysis by Benedetto et al,[23] composed largely of retrospective studies, showed central cannulation, including
the axillary artery, to be superior to peripheral cannulation of the femoral artery
in the short term, as shown in [Table 2]. These findings are hypothesized to be due to flow reversal in the thoracoabdominal
aorta with femoral artery cannulation, increasing risk of brain or organ malperfusion.[19]
[24] However, a recent study of 215 patients by Klotz et al[25] found no significant differences in postoperative neurologic deficits (p = 0.449) or 30-day mortality (p = 0.699) between patients undergoing central and femoral cannulation. Despite this,
most literature trends in favor of axillary cannulation, and most surgeons have adopted
an antegrade perfusion strategy with axillary artery cannulation.[26]
Table 2
Meta-analysis results of axillary versus femoral cannulation
|
Outcome
|
Odds ratio
|
Relative risk
|
95% confidence interval
|
p-Value
|
|
Ren et al
[18]:
|
|
Short-term mortality
|
0.25
|
–
|
0.15–0.42
|
0.01
|
|
Neurologic dysfunction
|
0.46
|
|
0.29–0.72
|
0.01
|
|
Malperfusion incidence
|
0.84
|
–
|
0.37–1.90
|
0.67
|
|
Benedetto et al
[23]:
|
|
In-hospital mortality
|
–
|
0.59
|
0.48–0.7
|
<0.01
|
|
Permanent neurologic deficit
|
–
|
0.71
|
0.55–0.9
|
0.005
|
Note: odds ratios and relative risk are shown as comparison of axillary/central artery
cannulation versus femoral/peripheral artery cannulation.
With regard to long-term survival, data are limited. However, a retrospective study
of 305 patients showed comparable 5-year survival between the axillary and femoral
cannulation (p = 0.52).[20] Cox's regression analysis demonstrated predictors of long-term mortality to be age
(p < 0.001), stroke (p < 0.001), prolonged CPB time (p = 0.001), hemodynamic instability (p = 0.002), and renal failure (p = 0.001).[20] Additional studies demonstrated similar findings, presenting evidence that repair
with AXC reduces overall mortality and neurologic complications when compared with
FAC.[27]
[28]
[29]
[30]
[31] These findings are summarized in [Table 3].
Table 3
Summary of study findings on cannulation site
|
Study
|
Patient
|
Finding
|
|
Stamou et al[20]
|
5-year survival in patients undergoing AXC vs. FAC
|
No difference in 5-year survival between groups undergoing AXC versus FAC
|
|
Moizumi et al[27]
|
Pre- and postoperative predictors of hospital death in patients with AAAD
|
Vischeral ischemia (OR = 18.4, p = 0.0028) and absence of axillary artery perfusion (OR = 8.2, p = 0.0014) were independent preoperative and operative predictors of hospital death
|
|
Reuthebuch et al[28]
|
Clinical and neurological outcomes of patients undergoing subclavian artery cannulation
versus femoral artery cannulation
|
Significantly improved neurological outcome (p = 0.0057), decreased postoperative bleeding (p < 0.0001), decreased incidence of MI (p < 0.0001), and decreased 30-day mortality (p = 0.0179) in patients undergoing subclavian artery cannulation compared with FAC
|
|
Pasic et al[30]
|
Neurological complications and hospital mortality in patients undergoing AAAD repair
with AXC versus FAC
|
Postoperative complications occurred in both groups, at nonsignificantly higher rates
in FAC compared with AXC
|
|
Etz et al[31]
|
Mortality and stroke in patients undergoing AAAD repair with AXC versus FAC
|
AXC had significantly better outcomes than FAC (p = 0.02)
|
|
Benedetto et al[23]
|
Meta-analysis of 4,476 patients comparing central and peripheral cannulation in patients
undergoing aortic surgery
|
Central cannulation (AXC) showed decreased in-hospital mortality (RR = 0.59, p < 0.001) and permanent neurological dysfunction (RR = 0.71, p = 0.005) when compared with peripheral cannulation (FAC)
|
|
Klotz et al[25]
|
Postoperative cerebral infarction, dialysis, and 30-day mortality in patients undergoing
AAAD repair with either AXC or FAC
|
Comparable postoperative cerebral infarction and 30-day mortality between the groups
(p = 0.699). Nonsignificantly higher rates of need for dialysis in patients undergoing
FAC (p = 0.073)
|
Abbreviations: AAAD, acute Type- A aortic dissection; AXC, axillary artery cannulation;
FAC, femoral artery cannulation; OR, odds ratio; RR, risk ratio.
Debate is maintained through the studies that found no difference in survival or complication
rates between AXC and FAC.[32]
[33]
[34] However, despite this, majority of the evidence demonstrates that perfusion through
the AXC site may be clinically advantageous to FAC. Furthermore, these findings demonstrate
that, regardless of cannulation strategy adopted, it is critical to carefully monitor
procedures and respond adequately to adverse events.[26]
Direct Aortic Cannulation
This technique, which also avoids retrograde flow in the downstream aorta, is an alternative
to time-consuming axillary artery access. The Hannover group reported their experience
of direct aortic cannulation in 122 patients with aortic dissection.[35] Malperfusion occurred in three patients (2.5%). Hospital mortality was 15% for the
entire cohort (18 patients). Permanent neurological dysfunction was detected in 15
patients (12%), whereas temporary neurological dysfunction occurred in 21 (17%). Total
arch replacement was performed in 31 patients (25%).
Technical Aspects of Repair
An additional area of controversy within AAAD repair is the different techniques of
proximal and distal root reconstruction. Choice of reconstruction technique is largely
based on viability and function of affected tissue; however, perioperative outcomes
are poorly studied.[36] Importantly, surgeon preference may play a role in which technique is utilized,
and it is therefore essential to fully understand the risk of each technique.
The most common surgical techniques for proximal root reconstruction include aortic
valve (AV) resuspension for structurally normal valves and sinuses, aortic valve replacement
(AVR) for a structurally abnormal valve but intact sinuses, and root replacement if
both the valve and sinuses are abnormal.[36] A retrospective cohort study by Gunn et al[37] found that the actuarial 10-year survival rates were greatest in AV resuspension,
followed by root replacement and AVR (72, 56, and 36%, respectively), and were significantly
increased in patients who underwent AV resuspension as compared with AVR (p = 0.011). This finding is consistent with the premise that increasingly compromised
tissue predisposes to greater risk. Gunn et al also showed independent predictors
of operative mortality to be hemodynamic instability (OR = 1.9; 95% CI: 0.03, 0.75;
p = 0.021) and CPB time greater than 200 minutes (OR = 1.9; 95% CI: 0.04, 0.54; p = 0.004). Again, consistent with the extent of abnormalities, CPB was significantly
longer in root replacement compared with AV resuspension (p < 0.001) or AVR (p = 0.027).[36]
While most surgical repair focuses on proximal repairs as described previously, the
dissection often propagates beyond the arch to the aortic bifurcation, described as
a DeBakey-I dissection. Among proximal strategy repairs, most patients are left with
a patent “Type-B” dissection, or false lumen, which yields a reoperation rate of more
than 30% to address a dissecting aneurysm. To address this, standard proximal repair
may be supplemented by thoracic stent-grafting through the open arch. A study by Pochettino
et al[38] demonstrated that antegrade stent graft deployment in patients with DeBakey-I dissections
obliterated the false lumen in 80% of patients. Furthermore, short-term results were
comparable between the stented and nonstented groups, despite longer CPB times in
the stented group. In patients with DeBakey-I dissections; therefore, consideration
of antegrade stent grafting should be given to lower morbidity and mortality.
With regard to construction of the distal anastomosis, open distal anastomosis under
circulatory arrest or distal aortic clamping with hypothermic cardiopulmonary bypass
are the primary surgical approaches. Recent studies comparing techniques demonstrate
comparable outcomes and survivals.[39]
[40]
[41]
[42]
[43]
[44] Although outcomes are similar, distal aortic clamping has been reported to distort
the posterior tip of the clamp and does not allow resection of the injured clamping
site, both of which may lead to higher reoperation for bleeding rates.[45] As such, open distal anastomosis under circulatory arrest is preferred technically.[46]
Hemodynamic Instability
Although previous studies related hemodynamic instability to differences in early
and late outcomes following AAAD, no studies have previously quantified late survival
between hemodynamically stable and unstable patients.[47] A recent study by Conway et al[48] was consistent with prior findings, as shown in [Table 4]. This study demonstrated significantly higher rates of postoperative complications
in patients with hemodynamic instability, including cardiac arrest (p < 0.001), operative mortality (p < 0.001), and acute renal failure (p = 0.001). Late survival followed a similar trend, with decreased late survival among
patients presenting with hemodynamic instability. At 1 year, 82% of hemodynamically
stable patients and 57% of hemodynamically unstable patients were alive, and at 10
years, 63 and 44% patients, respectively.[48] From these studies, the authors found that excessive mortality occurs early in the
postoperative course in patients presenting with hemodynamic instability. As such,
treatment of these patients must be individualized.
Table 4
Summary of study findings on surgical era
|
Study
|
Patient
|
Finding
|
|
Fann et al[76]
|
Surgical survival rates of patients with AAAD between 1963 and 1992
|
Earlier operative year, hypertension, cardiac tamponade, renal dysfunction, and older
age were independent determinants of operative death.
|
|
Conway et al[77]
|
Early postoperative outcomes and actuarial-free survival in patients undergoing AAAD
repair between 2000 and 2005 and 2006 and 2010
|
Operative mortality was significantly higher in earlier surgical era (24% in 2000–2005,
12% in 2006–2010; p = 0.013). Earlier date of surgery, hemodynamic instability, and CPB >200 minutes
were independent determinants of operative mortality
|
Abbreviations: AAAD, acute Type- A aortic dissection; CPB, cardiopulmonary bypass.
Malperfusion syndromes include cardiac, cerebral, renal, mesenteric, iliofemoral,
innominate, and spinal and are associated with high hospital mortality and increased
postoperative complications.[49]
[50] A study of 221 AAAD patients by Geirsson et al[51] found malperfusion in 26.7% of patients, with more than 30% of these patients experiencing
two or more malperfusion syndromes. Cardiac (p = 0.02) and cerebral malperfusions (p < 0.001) were significant risk factors for in-hospital mortality, and cerebral malperfusion
was a significant risk factor for decreased long-term survival (p = 0.0002). Recommended treatment of malperfusion syndromes is rapid restoration of
flow into the true lumen and obliteration of the false lumen to restore flow to all
distal aortic branches.
In patients who are high risk for open repair, including those with significant comorbidities
or anatomic challenges, endovascular treatment may provide an alternative. Although
reports of endovascular repair are typically limited by small sample size, studies
have shown promising outcomes for in-hospital and 30-day mortality rates. A study
by Vallabhajosyula et al[52] demonstrated zero in-hospital and 30-day mortality in patients treated endovascularly
who were prohibited from open repair due to hemodynamic instability, extreme frailty,
malignancy, and severe fibrosis or osteomyelitis of the mediastinum. Similar studies
reported comparable outcomes in small series.[53]
[54]
[55]
[56] These studies demonstrate that in patients who are hemodynamically unstable, have
prohibitive comorbidities, or present anatomical challenges, endovascular repair of
the ascending aorta is technically feasible. Although feasible in small series this
technique is challenging and is presently not an acceptable treatment of Type-A aortic
dissection.
Fate of the False Lumen
Persistent patent false lumen in the aorta is common in AAAD and may be associated
with poor long-term prognosis.[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
Analysis of the natural history of the residual aorta after AAAD repair provides insight
into the outcomes for patients with persistent patent false lumen. The presence of
a patent false lumen has been shown to be a significant risk factor for aortic enlargement,
increasing the likelihood for reoperation.[73] These findings are consistent with those of Park et al,[62] who found that the primary indication for reoperation following AAAD repair was
progressive enlargement of the false lumen, affecting 43% of patients. Enlarged aortic
diameter has been shown to be an independent predictor for chronic dissection.[74] In addition to aortic enlargement, patent false lumen can lead to multiple reentries
between the false and true lumen which requires reintervention, as described by Rylski
et al.[75] Endovascular repair may be used to seal the entry between the true and false lumen
to decrease the blood flow through the false lumen and promote stabilization through
thrombosis in the descending aorta. Additionally, in patients with DeBakey-I dissections,
antegrade thoracic stent grafting can be utilized to obliterate the false lumen in
up to 80% of patients, as described by Pochettino et al.[38] Endovascular interventions have been shown to be well tolerated with antegrade stent
graft deployment being a safe method to obliterate the thoracic false lumen. This
type of “elephant trunk” thoracic stent-grafting provides equivocal short-term results
compared with standard, open repair and lowers morbidity and mortality in the long-term.
Given the increased risk of aortic enlargement, reentry, and chronic dissection, it
is indicated to monitor patients with a patent false lumen more closely to assess
lumen status. Furthermore, select patient groups may benefit from endovascular repair
or supplemental stent grafting.
Outcomes in the Current Era
The culmination of previously discussed controversies is an analysis of the clinical
outcomes following repair over time. A previous study described a decline in operative
mortality over the period from 1963 to 1992, with findings summarized in [Table 4].[76] However, with broad understanding of disease and advances in surgical technique,
how has the survival trend changed in the current era?
To address this question, Conway et al[77] compared 111 patients who underwent repair between 2000 and 2005 with 140 patients
who underwent repair between 2006 and 2010. This study demonstrated that operative
mortality was significantly influenced by surgical era, with a 24% operative mortality
rate in patients treated between 2000 and 2005 compared with 12% in patients treated
between 2006 and 2010 (p = 0.013). Independent predictors of operative mortality as described by multivariate
logistic regression included hemodynamic instability (OR = 17.8; 95% CI: 0.05–0.35;
p < 0.001), CPB time > 200 minutes (OR = 9.5, 95% CI: 0.14–0.64; p = 0.002), and earlier date of surgery (OR = 5.8; 95% CI: 1.18, 5.14; p = 0.016). Additionally, actuarial 5-year survival was significantly worse for patients
treated earlier (64% for 2000–2005, 77% for 2006–2010, p < 0.001).[77] These findings demonstrate that the early clinical outcomes of repair of Type A
aortic dissection have improved over time.
