Keywords Type A aortic dissection - cannulation strategy - stroke - spinal cord injury
Introduction
Type A aortic dissection (TAAD) is a highly lethal disease that is still challenging for cardiac surgeons worldwide.[1 ] The main goal of surgery is the resection of the intimal tear while restoring adequate flow through the true lumen. As well, preventing proximal extension of the dissection helps to prevent death due to cardiac tamponade, ischemia, or aortic rupture.[1 ] These goals are usually obtained by ascending aorta and proximal arch replacement, with varying degrees of distal extension according to the severity of the disease.[1 ] Despite significant innovations in surgical techniques,[2 ]
[3 ]
[4 ]
[5 ]
[6 ] there are still several unanswered questions regarding the effects of intraoperative decisions and their impact on clinical outcomes.[1 ]
One of the main elements that the cardiac surgeon must consider during a TAAD repair concerns potential strategies oriented toward preventing cerebral and visceral malperfusion. Therefore, deciding on an optimal arterial cannulation for cardiopulmonary bypass (CPB) is a critical step of TAAD surgery, as this will guarantee adequate blood flow to end organs while preventing complications related to sustained malperfusion.[4 ]
[6 ]
[7 ]
[8 ]
The principles that guide optimal arterial cannulation include cannulation of the true lumen, avoiding pressurization of the false lumen, achieving cerebral protection, and guaranteeing adequate end-organ perfusion during the procedure.[1 ]
[8 ]
[9 ]
Also, performing an optimal arterial cannulation is important, because evidence has shown that cannulation-related complications are closely associated with postoperative morbidity and mortality.[9 ]
There are multiple different approaches for arterial cannulation.[1 ]
[9 ] These mainly include femoral, direct or side-graft right axillary (innominate), or direct ascending aortic cannulation, each associated with different advantages and pitfalls.[6 ]
[8 ]
[9 ]
[10 ]
On one hand, femoral cannulation has historically been preferred, as it is usually easier and quicker to perform.[4 ]
[6 ] However, femoral arterial cannulation has been associated with a higher risk of stroke and postoperative neurological dysfunction, mainly related to embolism from dislodged atherosclerotic plaques.[4 ]
[6 ]
[10 ]
[11 ] Therefore, studies have shown that the sites for arterial cannulation have shifted during the last decades, with a growing preference for right-axillary cannulation over a femoral cannulation site.[6 ]
[10 ]
[11 ]
[12 ] Axillary cannulation, on the other hand, has also been associated with postoperative complications, especially dissection of the artery and resultant arm ischemia, which provide a burden of postoperative morbidity. Over the last few years, direct aortic cannulation has become increasingly popular in experienced aortic centers. The evidence, however, has been contradictory, with no clear superiority of one cannulation site over others.[11 ] Current guidelines recommend the use of antegrade perfusion techniques (direct ascending aorta, aortic arch, innominate, axillary), as they provide a more physiological blood flow.[13 ]
[14 ]
Therefore, given the inconsistent evidence regarding the optimal arterial cannulation technique for TAAD surgery, this study aims to analyze outcomes related to different cannulation sites in our 20-year experience in aortic dissection surgery in a middle-income nation setting.
Materials and Methods
This was a retrospective single-center cohort study. We included all TAAD procedures presenting in the acute and hyperacute setting from January 2002 to September 2023 at Fundación Cardioinfatil-LaCardio, a referral center for cardiac and aortic surgery in Bogotá, Colombia.
Study groups were defined according to the site of cannulation (aorta, axillary, femoral, innominate). We collected information on baseline demographic characteristics, signs and symptoms upon admission; intraoperative variables included items like the type of the repair, CPB time, and cerebral perfusion time. Our main outcomes were in-hospital mortality, spinal cord injury (SCI), and stroke rate. As secondary outcomes, we considered other complications during postoperative course such as acute kidney injury rate, reintervention, transient lower extremity paresia/paralysis, and hospital length of stay. Data from pre-, intra-, and postoperative variables were collected by a member of the research team based on the cardiovascular surgery service database, which is compliant with the Socierty of Thoracic Surgeons (STS) standards, and by reviewing electronic health records and national databases.
Quantitative variables were initially evaluated using the Shapiro–Wilk test to determine their distribution. Based on this result, they were analyzed using mean and standard deviation for variables with a normal distribution and median and interquartile range for variables with a nonparametric distribution. Variables were analyzed using relative and absolute frequencies. The chi-squared or Fisher's exact test were used if the marginal absolute frequencies were <5 for categorical variables. The Kruskal–Wallis test for continuous nonparametric variables was used to determine the difference between groups. Differences were considered statistically significant if the p -value was less than 0.05. Post hoc tests for adjusting the p -values for multiple comparisons, including Bonferroni and Tukey, were performed. The data were analyzed on STATA 15 (Stata Corp. 2017. Stata Statistical Software: Release 15. College Station, TX: Stata Corp LLC) for Windows.
The current study was reviewed and approved by the Ethics in Research Committee at the Fundación Cardioinfantil/LaCardio. Given the retrospective nature of the study, it was considered a low-risk/no-risk study, and the patient consent requirement was waived.
The current manuscript is compliant with the STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) guideline recommendations.
Results
Out of a total of 260 patients who received aortic arch interventions, 127 patients who fulfilled all the inclusion criteria were included in the final analysis; 87 patients that received aortic arch interventions for indications other than TAAD were initially excluded. An additional group of 10 patients was excluded, because the aortic disease was identified as a Type B dissection. We then excluded 34 patients who received interventions for TAAD but presented to the emergency department outside the prespecified temporality of acute or hyperacute setting as defined by the STS. The sampling algorithm is shown in [Fig. 1 ]. The most common arterial cannulation site was axillary (n = 83, 65%), independent from the type of procedure that was performed. Femoral cannulation was uncommonly performed in this cohort, as only eight patients (6%) received this approach. Eighteen patients received innominate cannulation and 17 were included in the direct ascending aorta cannulation group.
Fig. 1 Sampling algorithm. *87 Patients who received aortic arch interventions for indications other than aortic dissection were excluded. *10 patients with Type B Aortic Dissection (TBAD) were excluded. *34 patients presenting outside of the prespecified time frame (acute or hyperacute setting) were excluded.
Baseline characteristics are presented in [Table 1 ], demonstrating a similar distribution of known risk factors among the four groups. Patients were most commonly male (69%) in the sixth decade of life. Hypertension was a common risk factor in all the groups (71%). These patients more commonly presented with a decline in functional class to NYHA class II (65%). Most patients in all groups received a surgical intervention in the acute setting (24 hours to 2 weeks).
Table 1
Baseline characteristics
Total
Aortic
Axillary
Femoral
Innominate
N = 127
N = 17
N = 83
N = 8
N = 19
p -Value
Age, y
56.8 (48.2–66.0)
59.7 (43.6–69.7)
57.7 (49.1–66.4)
56.7 (42.6–61.3)
54.3 (46.2–63.8)
0.6195
Male
88 (69.3%)
13 (76.5%)
57 (68.7%)
6 (75.0%)
12 (63.2%)
0.8342
Diabetes
8 (6.3%)
1 (5.9%)
4 (4.8%)
0 (0.0%)
3 (15.8%)
0.2956
Hypertension
90 (70.9%)
10 (58.8%)
58 (69.9%)
5 (62.5%)
17 (89.5%)
0.1995
Previous cardiac surgery
15 (11.8%)
3 (17.6%)
8 (9.6%)
0 (0.0%)
4 (21.1%)
0.395
NYHA
I
16 (12.6%)
1 (5.9%)
12 (14.5%)
0 (0.0%)
3 (15.8%)
0.5235
II
82 (64.6%)
11 (64.7%)
53 (63.9%)
5 (62.5%)
13 (68.4%)
0.9849
III
16 (12.6%)
4 (23.5%)
11 (13.3%)
0 (0.0%)
1 (5.3%)
0.2713
IV
13 (10.2%)
1 (5.9%)
7 (8.4%)
3 (37.5%)
2 (10.5%)
0.0681
LVEF%
51.0 (50.0–57.0)
51.0 (48.0–55.0)
51.0 (50.0–55.0)
50.0 (30.0–58.0)
54.0 (50.0–61.0)
0.3846
Dissection timing
0.8912
Hyperacute (<24 h)
11 (8.7%)
2 (11.8%)
7 (8.4%)
1 (12.5%)
1 (5.3%)
Acute (24 h– < 2 wk)
116 (91.3%)
15 (88.2%)
76 (91.6%)
7 (87.5%)
18 (94.7%)
EuroSCORE II
7.0 (5.0–13.0)
8.4 (4.0–18.0)
7.0 (4.0–12.0)
9.0 (6.0–15.0)
8.0 (6.0–16.7)
Operative priority
0.6161
Urgent
31 (24.4%)
5 (29.4%)
18 (21.7%)
1 (12.5%)
7 (36.8%)
Emergent
93 (73.2%)
12 (70.6%)
63 (75.9%)
7 (87.5%)
11 (57.9%)
Emergent salvage
3 (2.4%)
0 (0.0%)
2 (2.4%)
0 (0.0%)
1 (5.3%)
Abbreviation: LVEF, left ventricular ejection fraction; NYHA, New York Heart Association.
The intraoperative characteristics are shown in [Table 2 ]. Patients across all groups received comparable times of CPB, aortic cross-clamp, and cerebral perfusion (p = 0.95, 0.69, 0.89, respectively). The innominate was the preferred cannulation site for total arch replacement (TAR) procedures (58% of TAR cases), while patients undergoing a hemiarch repair more frequently received axillary (70%) or direct ascending aorta cannulation (14%).
Table 2
Intraoperative characteristics
Total
Aortic
Axillary
Femoral
Innominate
N = 127
N = 17
N = 83
N = 8
N = 19
Aortic procedure
0.894
Hemiarch
85 (66.9%)
12 (70.6%)
60 (72.3%)
5 (62.5%)
8 (42.1%)
Total arch
42 (33.1%)
5 (29.4%)
23 (27.7%)
3 (37.5%)
11 (57.9%)
Cardiopulmonary bypass, median (IQR)
222.0 (186.0–259.0)
223.0 (201.0–272.0)
222.0 (181.0–257.0)
238.5 (222.0–281.5)
209.0 (184.0–289.0)
0.9584
Aortic cross-clamp, median (IQR)
132.0 (100.0–165.0)
131.0 (92.0–160.0)
132.0 (101.0–163.0)
153.5 (109.5–208.5)
118.0 (80.0–178.0)
0.6964
Lowest temperature, °C
24.4 (23.0–26.0)
24.0 (22.0–28.0)
25.0 (23.3–26.6)
22.0 (22.0–24.0)
24.0 (24.0–26.0)
0.8981
Cerebral perfusion, min
28.0 (22.0–38.0)
29.5 (25.0–37.0)
26.0 (20.0–35.5)
31.0 (30.0–33.0)
31.0 (24.0–53.0)
0.111
Abbreviation: IQR, interquartile range.
The main outcomes are summarized in [Table 3 ]. The total mortality rate was 18%. We did not find statistically significant differences between groups in terms of risk of in-hospital mortality. The overall prevalence of SCI > 24 hours was low in our cohort (<2%), with no statistically significant differences between groups. Stroke rates were similar in the aortic, femoral, and innominate groups; however, the axillary cannulation site showed a trend toward fewer stroke events (3.6 vs. 12% for direct aortic cannulation, 12.5% for femoral cannulation, 10,5% innominate cannulation, p = 0.4).
Table 3
Postoperative characteristics
Total
Aortic
Axillary
Femoral
Innominate
N = 127
N = 17
N = 83
N = 8
N = 19
Reoperation for bleeding, n
26 (20.5%)
4 (23.5%)
14 (16.9%)
6 (75.0%)
2 (10.5%)
0.0006
Stroke, n
8 (6.3%)
2 (11.8%)
3 (3.6%)
1 (12.5%)
2 (10.5%)
0.4037
Acute kidney injury, n
25 (19.7%)
7 (41.2%)
14 (16.9%)
1 (12.5%)
3 (15.8%)
0.1217
Spinal cord injury
Transitory lower extremity paresia, n
2 (1.6%)
0 (0.0%)
1 (1.2%)
0 (0.0%)
1 (5.3%)
0.552
Permanent lower extremity paralysis, n
2 (1.6%)
0 (0.0%)
2 (2.4%)
0 (0.0%)
0 (0.0%)
0.7887
In-hospital mortality, n
23 (18.1%)
3 (17.6%)
17 (20.5%)
1 (12.5%)
2 (10.5%)
0.754
Length of stay
Preoperative, d
1.0 (0.0–2.0)
1.0 (0.0–4.0)
0.0 (0.0–1.0)
1.0 (0.5–1.5)
1.0 (0.0–5.0)
0.2122
Total ICU stay, d
6.0 (4.0–10.9)
6.1 (4.0–9.0)
6.0 (3.8–11.0)
15.3 (9.9–17.4)
5.1 (3.7–6.6)
0.0964
Total length of stay, d
13.0 (9.0–20.0)
13.0 (10.0–18.0)
12.0 (9.0–20.0)
25.0 (20.0–30.0)
14.0 (8.0–16.0)
0.1839
Abbreviation: ICU, intensive care unit.
Discussion
TAAD, the most common acute aortic syndrome, occurs when there is an intimal tear that allows blood flow through the false lumen.[15 ] According to the most recent STS/SVS classification, TAAD is defined by an entry tear proximal to the origin of the first branch of the aortic arch.[15 ]
[16 ] TAAD is a life-threatening condition with a mortality of rising 1 to 2% per hour after the onset of symptoms. The estimated incidence of TAAD lies around 5 to 30 cases per million people per year.[15 ]
Despite advances and innovations in devices and surgical techniques, TAAD is still a challenging disease for cardiac surgeons, with a persistently high perioperative mortality rate.[9 ]
[15 ]
[17 ] Surgical repair is directed toward preventing complications such as aortic rupture or extension of the flap, while guaranteeing flow through the true lumen to ensure adequate end-organ perfusion.
Choosing an adequate cannulation site is, therefore, a crucial step during surgical planning and the optimal cannulation strategy remains a matter of debate. Several factors such as hemodynamic status, organ perfusion, cerebral protection, and the condition of the aorta need to be considered. The different advantages and pitfalls of every cannulation site need to be considered as well. Femoral arterial cannulation (retrograde flow through the aorta) has been used since 1950[16 ] and has been popular, because it provides rapid access in hemodynamically unstable patients. Despite this advantage, femoral cannulation has been associated with a higher incidence of stroke due to retrograde flow, with a higher risk of embolization of debris to the brain. Femoral cannulation has also been associated with distal arterial injury, need for reintervention for mediastinal bleeding, and a greater chance of false lumen cannulation.[11 ]
[17 ]
[18 ]
[19 ]
[20 ] Accordingly, and despite the infrequent use of the femoral access, our results show that this site was associated with a higher rate of reoperation for bleeding and a longer length of stay as compared with other cannulation strategies, probably related to an increased preoperative severity of the disease.
Recently, other cannulation sites that provide antegrade perfusion have been preferred.[2 ]
[3 ]
[4 ]
[10 ]
[11 ]
[17 ]
[21 ] Axillary artery cannulation has been one of the most studied sites.[2 ]
[3 ]
[4 ]
[7 ]
[10 ]
[11 ]
[17 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]
[25 ] Since it was first described by Villard in 1976, axillary cannulation attaining a major role for TAAD surgery.[10 ] The most recent guidelines[14 ]
[15 ] recommended axillary as the first choice in aortic dissection surgery. The main advantages are related to antegrade cerebral perfusion (ACP), while allowing less deep hypothermia during the procedures.[2 ]
[3 ]
[11 ]
[17 ]
[24 ] Axillary cannulation has also been shown to lead to a lower incidence of stroke as compared with femoral cannulation. Accordingly, in our study, a great preference for axillary cannulation was evident, with low overall stroke events as compared with the other cannulation sites.
Innominate artery cannulation, originally described by Banbury and Cosgrove in 2000, provides another alternative for cannulation.[1 ]
[9 ]
[26 ] This maintains antegrade perfusion while not requiring another surgical wound. Innominate cannulation can be performed either through direct cannulation (with a low risk of vessel injury) or via a side graft, which carries an additional bleeding risk.[5 ]
[21 ]
[23 ]
[26 ] In our study, we found a preference for innominate artery use for TAR, with similar results as compared with axillary cannulation, but lower bleeding reoperation rates, hospital stay, and stroke.
Direct aortic central cannulation can be achieved using different techniques. An early approach described by Lijoi in 1998 proposed a cannulation strategy guided by TEE for identification of the true lumen, which is then cannulated using the Seldinger technique.[9 ] Another technique described by Jakob in 2007 known as Samurai cannulation, uses the venous drainage and left heart vent for dropping the mean blood pressure < 30 mm Hg to allow better identification of the true lumen after aortic transection before inserting the cannula.[9 ] Central direct true lumen cannulation provides benefits attributed to the ACP such as prevention of cerebral embolization, avoidance of dissection propagation, and avoidance of peripheral arterial injury. Disadvantages of central cannulation include: it is time consuming in emergent situations, as it requires identifying the true lumen; it carries a risk of malperfusion, and it carries a higher risk of aortic rupture. In our study, we identified several cases in which central cannulation was used, with no incidence of SCI; however, a relatively high incidence of acute kidney injury was identified in this group of patients.
This study demonstrates the preference of axillary artery cannulation over other sites, in accordance with worldwide trends, which could be attributed to the advantages described above and to surgeon's preferences—with excellent results and a relatively low number of postoperative complications. In particular, a low stroke rate was evident. The introduction of a direct ascending aorta true lumen cannulation has proven to being a safe alternative, although more experience is needed to provide more robust conclusions. Femoral cannulation use was limited in this experience, with a higher incidence of complications, perhaps related to its use in cases with higher preoperative severity of the disease.
Limitations
This study had limitations inherent to its retrospective observational nature, with a potential for bias in selection of patients. Besides, it is likely that the selection of the cannulation strategy was associated with the surgeon's preference and experience, factors not accounted for in the analysis. Arm ischemia as a complication was not measured in axillary artery cannulation. A small number of patients was included in the femoral cannulation group as compared with other cannulation sites, which may limit the interpretation of the results related to this group. This, however, reflects our real-life choice for cannulation sites for TAAD surgery. Finally, there is a lack of published Colombian and Latin American experiences, preventing comparison of our results with other centers with a similar setting.
Conclusion
Based on our single-center 20-year period experience, axillary artery cannulation proved to be safe, with a low stroke rate in patients with TAAD. Direct aortic cannulation was a good alternative with no incidence of SCI, although its use is limited by surgeon's experience. Femoral cannulation is infrequent in our setting, and it had the highest incidence of reoperation for bleeding and a higher hospital stay as compared with other sites.
