Congenital Anomalies of the Aortic Arch in Acute Type-A Aortic Dissection: Implications for Monitoring, Perfusion Strategy, and Surgical Repair Bryan G. Maxwell, MD, MPH,* Katherine B. Harrington, MD,† Ramin E. Beygui, MD,† and Daryl A. Oakes, MD‡ Objective: To assess whether management of acute Stanford type-A aortic dissection differs in patients with congenital anomalies of the aortic arch compared with standard institutional practice. Design: Retrospective analysis of all consecutive patients from 2001 through 2011. Setting: Quaternary referral center for surgical management of thoracic aortic disease. Participants: All patients with arch anomalies who underwent surgery for acute Stanford type-A aortic dissection during the study period (n ¼ 43). Interventions: Surgical management, anesthetic monitoring, and perfusion strategy were analyzed in a retrospective fashion. No new interventions were undertaken as part of this study. Measurements and Main Results: Management differed most in patients with an aberrant right subclavian artery (n ¼ 5), because the institutional standard of right axillary artery cannulation with left upper extremity arterial pressure monitoring was not possible. In patients with one of two
E
MERGENT SURGICAL MANAGEMENT of acute Stanford type-A aortic dissection requires careful attention to the anatomy of the aortic arch in order to optimize surgical and neurologic outcome. Numerous anatomic variations and abnormalities of the arch exist in the population, which may have important implications for perioperative management.1 Although case reports have documented successful treatment of type A aortic dissection in the setting of anatomic anomalies of the aortic arch, these reports have been sporadic and incomplete in their analysis of perioperative monitoring, perfusion strategies, and surgical technique.2–4 Key aspects of usual institutional management for type-A dissection include the following: Preoperative cannulation of the left radial or brachial artery for invasive blood pressure monitoring, central venous access with an introducer sheath in the right internal jugular vein, noninvasive cerebral oximetry monitoring (Fore-Sight™, CAS Medical Systems, Branford, CT) with evaluation for absolute decreases or relative asymmetry during deep hypothermic circulatory arrest (DHCA) or selective antegrade cerebral perfusion (SACP), and right axillary artery cannulation with a sidearm Dacron graft via a separate infraclavicular incision. If arch or hemiarch repair is required, SACP is achieved via the right axillary artery graft with clamping of the innominate artery and back-clamping of the left carotid artery. The arterial line is placed preferentially on the left to allow for monitoring of systemic perfusion pressure while the right axillary or innominate artery is clamped. The author sought to systematically review contemporary institutional experience with the perioperative management of acute type-A aortic dissection in patients with arch anomalies, with a focus on assessing the hypothesis that the management of these patients differs in important ways from standard institutional practice in patients with normal arch anatomy.
"bovine" arch patterns (n ¼ 32), management differed in the conduct of selective antegrade cerebral perfusion, which could include clamping above or below the takeoff of the left common carotid artery (and, therefore, produced unilateral or bilateral antegrade cerebral perfusion). All patients with a connective tissue disorder exhibited a bovine arch pattern. Management of patients with a right arch (n ¼ 3) reflected the opposite of management for normal anatomy (for patients with traditional mirror-image branching) or opposite that of the aberrant right subclavian group (for patients who had a corresponding aberrant left subclavian artery). Conclusions: Rational management reflected the anatomic variations observed. These results support the importance of interdisciplinary planning, especially in an emergency, to optimize outcome. & 2013 Elsevier Inc. All rights reserved. KEY WORDS: type-A aortic dissection, acute aortic dissection, aortic arch anomalies, abnormal aortic arch, perfusion strategy, cannulation strategy
METHODS The authors performed a retrospective cohort analysis of all identifiable instances of operative intervention for acute Stanford type-A aortic dissection in patients with anatomic anomalies of the aortic arch occurring during the interval from January 1, 2001 to December 31, 2011. This research received Institutional Review Board approval. Potential cases initially were identified through a combination of International Statistical Classifications of Diseases (ICD-9) diagnostic codes for type-A dissection and free-text searches within the institutional clinical database for permutations on the terms “aortic arch anomaly” and “type-A dissection” (see Appendix). The authors identified 6 major patterns of abnormal arch anatomy for the search (Fig 1). Two patterns reflect what commonly is referred to as a "bovine" arch: An arch with a common origin of the innominate artery and left common carotid (Fig 1C) and the origin of the left common carotid off of the innominate artery (Fig 1D). Several prior reports have pointed out that in addition to being ambiguous, “bovine arch” is a misnomer, because neither of these patterns reflects the arch branching pattern found in cattle (a single common brachiocephalic trunk that gives rise to all head and extremity vessels; Fig 1B).5 The other 4 branching patterns identified were a 4-vessel arch with an aberrant right subclavian artery originating from the distal arch or
From the *Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD; and the †Departments of Cardiothoracic Surgery and ‡Anesthesiology, Perioperative and Pain Medicine, Stanford University School of Medicine, Stanford, CA. Address correspondence and reprint requests to Bryan G. Maxwell, MD, MPH; Johns Hopkins University School of Medicine, Department of Anesthesiology and Critical Care Medicine; 1800 Orleans Street Zayed 6208P; Baltimore, MD, 21287. Tel.: þ(410) 955 9147; Fax: þ(410) 955 0994 E-mail: [email protected] © 2013 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2013.12.001
Journal of Cardiothoracic and Vascular Anesthesia, Vol ], No ] (Month), 2013: pp ]]]–]]]
1
2
MAXWELL ET AL
Fig 1. Taxonomy of arch branching patterns. (A) Normal aortic arch. (B) True bovine pattern (ie, single common brachiocephalic trunk). (C) Common origin of the innominate artery and left common carotid (1 of 2 “bovine arch” patterns in humans). (D) Origin of the left common carotid off of the innominate artery (second of 2 “bovine arch” patterns in humans). (E) Four-vessel arch with an aberrant right subclavian artery (ARSCA) originating from the distal arch or proximal descending aorta. (F) Four-vessel arch with the left vertebral artery coming directly off the aorta. (G) Right aortic arch. (H) Double aortic arch.
proximal descending aorta (Fig 1E), a 4-vessel arch with the left vertebral artery coming directly off the aorta (Fig 1F), a right aortic arch (Fig 1G), and a double aortic arch (Fig 1H). The embryologic etiologies of these branching patterns are thought to involve a complex interplay of cell-signaling pathways and hemodynamic forces and remain under active investigation.6 Potential cases were reviewed individually to exclude cases that did not meet inclusion criteria and to avoid double-counting of cases. The clinical database techniques used in this project were supported by the NIH/NCRR CTSA award number UL1 RR025744. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.7 Criteria for inclusion in the retrospective cohort were adult patients (age greater than or equal to 18 years) who underwent operative intervention for aortic dissection within 48 hours of admission to the institution. Individual case review was performed using details from the operative report, imaging reports, and primary source images from computed tomography (CT) scans and intraoperative transesophageal echocardiography (TEE) to confirm the following information: The dissection was acute in nature, the dissection involved the ascending aorta between the aortic root and the takeoff of the first arch branch vessel, the aortic arch anatomy represented one of several patterns of abnormal arch branching, and the patient underwent operative repair of the dissection. In addition to the operative report and imaging studies, anesthesia records, perfusion records, progress notes, and discharge summaries were used to determine operative and postoperative variables including cannulation strategy, method of surgical repair, site of primary intimal tear (PIT), bypass times and nadir temperature, use of DHCA or SACP, and incidence of in-hospital neurologic complications and mortality.
These perioperative variables were used to compare the management of patients with arch anomalies to the standard institutional practice for management of acute type-A dissection in patients with normal arch anatomy. RESULTS
A total of 161 potential cases were identified from the database search criteria. After individual case review, 43 met criteria for inclusions. Of these, 32 were 1 of the 2 “bovine arch” patterns: 23 had a common origin of the innominate artery and left common carotid artery (Fig 1C) and 9 had a left common carotid off the innominate (Fig 1D). No cases of true bovine anatomy (single brachiocephalic trunk; Fig 1B) were identified. There were 5 patients with an aberrant right subclavian artery (ARSCA; Fig 1E), 3 patients with a 4-vessel arch with separate takeoff of the left vertebral artery (Fig 1F), 3 patients with a right aortic arch (Fig 1G), and no patients with a double aortic arch (Fig 1H). Table 1 contains demographic variables and Table 2 contains surgical variables for each group. All patients undergoing composite valve graft or ascending aorta replacement alone were repaired on cardiopulmonary bypass (CPB) without circulatory arrest. In all patients undergoing hemiarch replacement, deep hypothermic circulatory arrest with or without selective antegrade cerebral perfusion was employed. The PIT was in the root (21%) or ascending aorta (77%) in all but 1 case; 1 patient (ARSCA group) had a PIT in the arch with retrograde type-A extension.
3
ARCH ANOMALIES AND TYPE-A DISSECTION
Table 1. Characteristics of Patients by Arch Pattern Group
Age (mean ⫾ SD) Male White Connective Tissue Disorder Marfan Other Congenital Anomalies Bicuspid Aortic Valve Anomalous Coronary Artery Repaired TOF Repaired PAPVR
Bovine Arch
ARSCA
Vertebral
Right Arch
All
n ¼ 32
n¼5
n¼3
n¼3
n ¼ 43
47 28 27 6 5
⫾ 15 (88) (84) (19) (16)
57 2 4 0
⫾9 (40) (80)
44 1 1 0
3 2 0 0
(9) (6)
1 0 0 0
(20)
0 0 0 0
⫾9 (33) (33)
⫾ 27 (33) (67)
33 1 2 0
0 0 1 1
(33) (33)
47 32 34 6
⫾ 15 (74) (79) (14)
4 2 1 1
(9) (5) (2) (2)
NOTE: Values are number (percentage) unless otherwise specified as mean ⫾ standard deviation. Abbreviations: ARSCA, aberrant right subclavian artery; PAPVR, partial anomalous pulmonary venous return; SD, standard deviation; TOF, tetralogy of Fallot.
the takeoff of the left common carotid); this was more common in patients with the origin of the left common carotid off the innominate (Fig 1D; 8/8 patients ¼ 100%) than in patients with a common origin of the innominate and left common carotid (Fig 1C; 4/8 patients ¼ 50%). Unilateral cerebral perfusion resulted from clamping above the takeoff of the left common carotid in 4 of 15 patients (27%); all these cases occurred in patients with a common origin of the innominate and left
Among patients with either of the 2 bovine arch patterns (n ¼ 32), arterial cannulation was via the right axillary artery in 30 (94%) patients and via the right femoral artery in 2 (6%). In all patients undergoing hemiarch replacement (with or without aortic valve replacement; n ¼ 15), axillary artery cannulation was performed and selective antegrade cerebral perfusion via the right axillary artery was used. In 11 of 15 patients (73%), bilateral cerebral perfusion resulted (clamping occurred below
Table 2. Surgical Characteristics Group
Arterial Line Site Left upper extremity Right upper extremity Femoral Arterial Bypass Cannulation Right axillary* Left axillary* Femoral Ascending aorta Surgical Repair Ascending only CVG Ascending þ hemiarch CVG þ hemiarch Cases without DHCA/SACP Nadir temp (1C) CPB time (min) Cross-clamp time (min) Cases with DHCA/SACP Nadir temp (1C) CPB time (min) Cross-clamp time (min) SACP DHCA DHCA/SACP time (min)
Bovine Arch
ARSCA
Vertebral
Right Arch
All
n ¼ 32
n¼5
n¼3
n¼3
n ¼ 43
31 0 1
(97)
30 0 2 0
(94)
12 5 5 10
(38) (16) (16) (31)
(3)
(6)
29 161 114
⫾1 ⫾1 ⫾4
22 260 189 15 0 37
⫾3 ⫾ 63 ⫾ 56 (100) ⫾ 16
0 5 0 0 0 5 0 2 0 3 0
(100)
(100)
(40) (60)
29 97 61
⫾1 ⫾ 17 ⫾ 11
19 207 111 0 3 35
⫾2 ⫾ 86 ⫾ 37 (100) ⫾ 14
3 0 0
(100)
2 1 0
3 0 0 0
(100)
0 1 2 0
0 0 3 0
19 201 97 3 0 21
(100)
⫾1 ⫾ 18 ⫾ 17 (100) ⫾1
(67) (33)
(33) (67)
3 0 0 0
(100)
28 105 60
⫾1 ⫾ 57 ⫾ 32
36 6 1
(84) (14) (2)
33 1 9 0
(77) (2) (21)
17 5 11 10 22 29 132 86 21 20 237 154 18 3 34
(40) (12) (26) (23) (51) ⫾1 ⫾ 34 ⫾ 27 (49) ⫾2 ⫾ 66 ⫾ 62 (90) (10) ⫾ 14
NOTE. Values are number (percentage) or mean ⫾ standard deviation, as appropriate. Abbreviations: ARSCA, aberrant right subclavian artery; CPB, cardiopulmonary bypass; CVG, composite valve graft; DHCA, deep hypothermic circulatory arrest (without SACP); SACP, selective antegrade cerebral perfusion. *Includes axillary, subclavian, or innominate.
4
common carotid (Fig 1C; 4/8 ¼ 50% of patients in this group). No cerebral oximetry abnormalities were recorded in this group. Six patients (19%) had an identified connective tissue disorder (5 of these had Marfan syndrome). Other associated anomalies in this group included 3 patients with bicuspid aortic valve and 2 patients with an anomalous left main coronary off a single coronary ostium in the right coronary sinus. Of patients with ARSCA (n ¼ 5), 1 patient had an unusual combination of a bovine arch (common origin of right and left common carotid arteries) and aberrant retroesophageal right subclavian artery in the setting of a large arch aneurysm; the remaining 4 patients had a more typical 4-vessel arch and a retroesophageal right subclavian artery as the most distal branch vessel. In all patients, arterial line placement occurred in the right radial artery and arterial cannulation for bypass occurred via the right femoral artery with retrograde perfusion. The patients in this group who required hemiarch replacement (n ¼ 3) did not undergo SACP; instead, DHCA was used for the arch repair component (the only use of pure DHCA among all study patients). Cerebral protection was afforded by hypothermia only without antegrade or retrograde cerebral perfusion during circulatory arrest. The only associated connective tissue disorder or congenital abnormality in this group was 1 patient with a bicuspid aortic valve. All patients in the 4-vessel arch with vertebral artery group (n ¼ 3) had arterial cannulation via the right axillary artery and arterial line placement in the left radial artery. All surgical repairs consisted of replacement of the ascending aorta and hemiarch with SACP. There were no connective tissue disorders or associated congenital abnormalities identified in this group. Of patients with a right-sided aortic arch (n ¼ 3), 1 patient had a mirror-image 3-vessel aortic arch with left-sided innominate artery, right common carotid, and right subclavian arterial branching. For this patient, perioperative management involved the mirror-image equivalent of standard management for a patient with a left arch and normal branching pattern: The arterial line was placed in the right radial artery, arterial cannulation occurred at the left axillary artery via a separate infraclavicular incision, and surgical repair was an isolated repair of the ascending aorta. The remaining 2 patients had a right-sided aortic arch with an aberrant retroesophageal left subclavian artery—a mirror-image equivalent of the 4-vessel aberrant right subclavian artery group. These 2 patients both had arterial lines placed in the left radial artery, femoral arterial cannulation for bypass, and surgical repair involving isolated repair of the ascending aorta. There were no associated connective tissue disorders in the right arch group. Associated congenital abnormalities in this group included 1 patient with repaired tetralogy of Fallot and 1 patient with repaired partial anomalous pulmonary venous return. Outcomes included one mortality that occurred, in an 88year-old woman (common origin of the left carotid and innominate arteries) after a postoperative course complicated by multisystem organ failure. One nonfatal neurologic complication occurred, in a 41-year-old man (common origin of the left carotid and innominate arteries) who underwent successful
MAXWELL ET AL
repair that involved ascending aorta and hemiarch replacement (nadir temperature 171C, bypass time 227 minutes, aortic crossclamp time 142 minutes, unilateral SACP for 56 minutes). Postoperative imaging revealed a significant right middle cerebral artery stroke from which the patient recovered partially (he was discharged to home with some residual weakness but could perform all activities of daily living independently). His first comprehensive neurologic examination occurred postoperatively but, of note, the presenting symptoms of his dissection (which extended into the bilateral carotid arteries on his first CT scan) involved right eye visual changes and left lower extremity paresis. He had no neurologic pathology identified in the left hemisphere (after right unilateral ACP). DISCUSSION
The identified cases represent a frequency of arch anomalies that is consistent with the best prior estimate of the frequency of arch anomalies in the general population: An analysis of an unselected population of consecutive patients (n ¼ 861) undergoing CT scans in Leicester, United Kingdom, identified a bovine pattern in 20% of patients, a 4-vessel arch with the left vertebral artery off the arch in 6%, a 4-vessel arch with an aberrant right subclavian artery in 0.5%, and a right-sided arch in 0.2%.1 These estimates are consistent with the prior literature, including 1 angiography series8 and several smaller autopsy studies.9–11 Although population-based methodologies would be required to adequately assess whether individuals with arch anomalies have a higher incidence of aortic dissection compared with the general population, the results do not provide any grounds to suspect a significant departure from the baseline risk. Other authors have speculated that abnormal arch anatomy may be a risk factor for aortic dissection: For instance, that the acute angle of take-off in an aberrant right subclavian artery constitutes a locus of weakness in the aortic wall that may predispose this as the site of dissection.3 However, in the absence of retrograde dissection involving the ascending aorta, this particular example would represent a type-B dissection and as such is beyond the scope of this study. No evidence from the current study supported the association between arch anomalies and dissection insofar as there was only one case with a distal PIT that became a type-A dissection because of retrograde extension. The group that demonstrated the most significant departures from standard management was the aberrant right subclavian group. Site of cannulation in acute type-A dissection remains a controversial topic. Ascending aortic cannulation has been suggested as a safe option12,13 to avoid the risks of false lumen perfusion and atheroemboli from the descending aorta that are inherent in femoral cannulation with retrograde perfusion,14 though a recent, large (n ¼ 473) retrospective comparison of femoral to central cannulation (ascending aorta, axillary, or innominate artery) found no difference in mortality or permanent neurologic disability in the femoral cannulation group.15 For similar reasons, other centers prefer left axillary cannulation for initiation of bypass, but this technique requires the additional step of direct cannulation of the innominate
ARCH ANOMALIES AND TYPE-A DISSECTION
artery and/or left common carotid artery through the arch if antegrade cerebral perfusion is desired.16 Cannulations of the right axillary artery or innominate artery17 have been advocated as alternatives. Both avoid the need to manipulate the dissected ascending aorta (which also may have significant atheroma present, because atheromatous disease shares risk factors with aortic dissection18) and allow for subsequent SACP without additional cannulation sites. Innominate cannulation has the advantage of simplicity and avoiding a need for a separate incision.19 Axillary artery cannulation via a separate incision (standard practice) has the advantage of allowing the initiation of CPB and cooling before opening the sternum and mediastinum to maximize safety by minimizing the risk and consequences of iatrogenic rupture. However, the abnormal anatomy in the aberrant right subclavian group required a departure from this practice. These patients all had contralateral arterial line placement, femoral arterial cannulation instead of a separate infraclavicular dissection for axillary cannulation, and hemiarch replacement under pure hypothermic circulatory arrest instead of SACP. In this situation, if ACP is desired, it would have to be accomplished through direct carotid artery cannulation. The same is true if right axillary artery cannulation is performed in patients with ARSCA: Separate direct carotid artery cannulation is required to accomplish ACP because the right axillary artery does share an origin with the right carotid artery.20 The management implications of the 2 bovine arch branching patterns center around the use of antegrade cerebral perfusion. If right axillary cannulation is used for bypass, 2 options exist for SACP: A clamp can be applied to the innominate artery caudad to the takeoff of the left common carotid artery (resulting in antegrade perfusion of both carotid arteries—ie, bilateral ACP) or cephalad to the takeoff of the left common carotid artery (resulting in traditional perfusion of only the right carotid—ie, unilateral ACP). Bilateral ACP may be easier (and in this series was more common) in patients in whom the left common carotid artery originates from the innominate artery (see Fig 1D) than in those in whom the left common carotid and innominate arteries have a common origin at the aorta (see Fig 1D). When bilateral ACP is used, a patent circle of Willis is not necessary for bilateral cerebral perfusion, and cerebral oximetry should be interpreted with this in mind. For instance, if asymmetric cerebral oximetry values are observed, other causes should be considered, such as impaired venous drainage (from large-caliber jugular venous cannulae or head position with jugular vein kinking). If unilateral ACP is used, contralateral cerebral perfusion remains dependent on a patent circle of Willis and adequate pressure to reach the contralateral hemisphere. Concern for relative asymmetry of cerebral oximetry as opposed to absolute (presumably symmetric) values should be pursued with an understanding of this anatomic variant’s implications for SACP. As with any SACP case, management options would include performing separate cannulation of the left common carotid artery for bilateral ACP or raising the SACP flow or perfusion pressure. In addition, with a bovine arch branching pattern (see Fig 1C or 1D), an additional option exists: Moving the proximal innominate artery clamp to below the takeoff of the left common carotid artery to convert to bilateral ACP.
5
An alternative approach for conduct of CPB and SACP is the use of the innominate artery for arterial perfusion. Subsequent to the period of this study, the authors infrequently have used this approach when the innominate artery is found to be free of dissection. The innominate artery should be assessed carefully by preoperative imaging and direct inspection to rule out extension of the dissection. In this scenario placement of a 6-mm-to-8-mm conduit to the proximal innominate artery (a “chimney”) through the median sternotomy provides access for the conduct of CPB as well as SACP. Using the undissected innominate artery obviates the need for a second incision for exposure of the right axillary artery for placement of the chimney. Although this study was too small to provide insights into the effect of unilateral versus bilateral ACP on outcomes, the authors noted that one surgeon included language in an operative note that suggested a belief that the safe duration of SACP might be longer in a patient with a bovine arch in whom antegrade perfusion was provided to both carotid arteries by clamping of the bovine trunk below the left carotid takeoff. Further studies would be required to assess whether this belief is held widely (and, therefore, if it affects surgeon behavior) or if any outcome data support it. The large German registry of type-A dissection repair found no evidence of a neurologic outcome or mortality difference between patients receiving unilateral versus bilateral antegrade cerebral perfusion, though duration was not specifically examined between these groups.21 The authors did not observe any episodes of significant asymmetry of cerebral oximetry values during unilateral ACP; however, it was suspected that the retrospective methodology was less sensitive in detecting such a phenomenon than would be a prospective investigation. The authors speculated that cerebral oximetry offers value in detecting inadequate cerebral perfusion in the setting of abnormal branching patterns (with or without incomplete intracranial vascular connections at the circle of Willis), but this study was not powered to demonstrate this value. The group with a 4-vessel arch, including a direct takeoff of the left vertebral artery, did not differ significantly from standard management. However, it should be noted that this anatomic pattern likely would have implications in planning for distal repair. Antegrade stent graft placement at the time of open repair (often termed a “frozen elephant trunk”) or subsequent retrograde thoracic endovascular stent graft deployment would risk compromise of both the left subclavian and left vertebral arteries. Because vertebral artery flow often is implicated in maintaining collateral perfusion to the left upper extremity when the subclavian artery is sacrificed,22 4-vessel arch anatomy would require modification of standard technique (eg, carotid-subclavian transposition or bypass) to reduce the risk of arm ischemia. Patients with a right aortic arch fell into 2 groups: 1 patient had a mirror-image equivalent of normal anatomy, with contralateral equivalents for the placement of invasive monitoring and site of cannulation for bypass. Two patients had a mirror-image equivalent of the aberrant right subclavian group. It is noteworthy that 2 of 3 cases of right arch were associated with other congenital lesions that had required operative repair (410 years prior in both cases). It is possible that the dissection observed in these patients could be related to
6
MAXWELL ET AL
previous cannulation of the ascending aorta, though given the long time interval between events, a causal relationship might be less plausible. The authors observed that within this survey of arch anomalies, all the patients with an identified connective tissue disorder (Marfan syndrome and/or bicuspid aortic valve) were in 1 of the 2 bovine arch groups. Larger studies would be needed to further evaluate this observation, but this center is among several international referral centers for connective tissue disorders, and, as such, it is interesting that none of the other abnormal patterns was observed in this subpopulation. The principal limitations of this study were those of a single-institution, retrospective design, which has limited outcome information and does not allow for a robust estimation of whether individuals with variant aortic arch anatomy are at increased or decreased risk of aortic dissection relative to the population average. However, the patterns observed even in this small series showed consistency, and the departures from usual management—most notably in the ARSCA group—were physiologically justifiable. These results supported the conclusion that careful attention to arch anatomy should be a part of the planning of acute type-A aortic dissection on the part of anesthesiologists, surgeons, and
perfusionists. If high-quality CT images are not available or do not clearly confirm the arch anatomy, TEE has a potential role in identifying some variant branching patterns (eg, right arch, aberrant right subclavian) but is likely to lack sensitivity in detecting others (bovine arch, 4-vessel arch with left vertebral artery). This role would need to be evaluated further in subsequent studies to determine accuracy. Although these operations usually occur under emergent conditions with limited opportunity for leisurely preoperative workup and discussion, interdisciplinary communication is key in ensuring that all parties involved understand the implications of variant arch anatomy for modifications in monitoring, perfusion, and surgical repair strategies. ACKNOWLEDGEMENTS The authors gratefully acknowledge the clinical informatics assistance and support of Gomathi Krishnan.
APPENDIX A. SUPPORTING INFORMATION
Supplementary material cited in this article is available online at http://dx.doi.org/10.1053/j.jvca.2013.12.001.
REFERENCES 1. Jakanani GC, Adair W: Frequency of variations in aortic arch anatomy depicted on multidetector CT. Clin Radiol 65:481-487, 2010 2. Li Q-L, Zhang X-M, Zhang X-M: Aortic dissection originating from an aberrant right subclavian artery. J Vasc Surg 46: 1270-1273, 2007 3. Kikuchi K, Makuuchi H, Oono M, et al: Surgery for aortic dissection involving an aberrant right subclavian artery. Jpn J Thorac Cardiovasc Surg 53:632-634, 2005 4. Rasmussen DK, Dougherty J: Aortic dissection with vascular abnormalities. J Am Osteopath Assoc 111:407-409, 2011 5. Layton KF, Kallmes DF, Cloft HJ, et al: Bovine aortic arch variant in humans: clarification of a common misnomer. AJNR Am J Neuroradiol 27:1541-1542, 2006 6. Wang Y, Dur O, Patrick MJ, et al: Aortic arch morphogenesis and flow modeling in the chick embryo. Ann Biomed Eng 37: 1069-1081, 2009 7. Lowe HJ, Ferris TA, Hernandez PM, et al: STRIDE—an integrated standards-based translational research informatics platform. AMIA Annu Symp Proc 2009:391-395, 2009 8. Natsis KI, Tsitouridis IA, Didagelos MV, et al: Anatomical variations in the branches of the human aortic arch in 633 angiographies: clinical significance and literature review. Surg Radiol Anat 31:319-323, 2009 9. Edwards JE: Anomalies of the derivatives of the aortic arch system. Med Clin North Am 32:925-949, 1948 10. Gupta M, Sodhi L: Variations in branching pattern, shape, size and relative distances of arteries arising from arch of aorta. Nepal Med Coll J 7:13-17, 2005 11. Bhatia K, Ghabriel MN, Henneberg M: Anatomical variations in the branches of the human aortic arch: a recent study of a South Australian population. Folia Morphol (Warsz) 64:217-223, 2005 12. Reece TB, Tribble CG, Smith RL, et al: Central cannulation is safe in acute aortic dissection repair. J Thorac Cardiovasc Surg 133: 428-434, 2007
13. Kamiya H, Kallenbach K, Halmer D, et al: Comparison of ascending aorta versus femoral artery cannulation for acute aortic dissection type A. Circulation 120:S282-286, 2009 14. Tiwari KK, Murzi M, Bevilacqua S, et al: Which cannulation (ascending aortic cannulation or peripheral arterial cannulation) is better for acute type A aortic dissection surgery? Interact Cardiovasc Thorac Surg 10:797-802, 2010 15. Di Eusanio M, Pantaleo A, Petridis FD, et al: Impact of different cannulation strategies on in-hospital outcomes of aortic arch surgery: a propensity-score analysis. Ann Thorac Surg 96:16561663, 2013 16. Kano M, Chikugo F, Shimahara Y, et al: Left axillary artery perfusion in surgery of type A aortic dissection. Ann Thorac Cardiovasc Surg 14:22-24, 2008 17. Huang FJ, Wu Q, Ren CW, et al: Cannulation of the innominate artery with a side graft in arch surgery. Ann Thorac Surg 89: 800-803, 2010 18. Augoustides JGT, Harris H, Pochettino A: Direct innominate artery cannulation in acute type a dissection and severe thoracic aortic atheroma. J Cardiothorac Vasc Anesth 21:727-729, 2007 19. Di Eusanio M, Ciano M, Labriola G, et al: Cannulation of the innominate artery during surgery of the thoracic aorta: our experience in 55 patients. Eur J Cardiothorac Surg 32:270-273, 2007 20. Battaloglu B, Secici S, Colak C, et al: Aberrant right subclavian artery and axillary artery cannulation in type a aortic dissection repair. Ann Thorac Surg. e1-e2 , 2013 21. Krüger T, Weigang E, Hoffmann I, et al: Cerebral protection during surgery for acute aortic dissection type A: results of the German Registry for Acute Aortic Dissection Type A (GERAADA). Circulation 124:434-443, 2011 22. Woo EY, Bavaria JE, Pochettino A, et al: Techniques for preserving vertebral artery perfusion during thoracic aortic stent grafting requiring aortic arch landing. Vasc Endovascular Surg 40: 367-373, 2006
E
MERGENT SURGICAL MANAGEMENT of acute Stanford type-A aortic dissection requires careful attention to the anatomy of the aortic arch in order to optimize surgical and neurologic outcome. Numerous anatomic variations and abnormalities of the arch exist in the population, which may have important implications for perioperative management.1 Although case reports have documented successful treatment of type A aortic dissection in the setting of anatomic anomalies of the aortic arch, these reports have been sporadic and incomplete in their analysis of perioperative monitoring, perfusion strategies, and surgical technique.2–4 Key aspects of usual institutional management for type-A dissection include the following: Preoperative cannulation of the left radial or brachial artery for invasive blood pressure monitoring, central venous access with an introducer sheath in the right internal jugular vein, noninvasive cerebral oximetry monitoring (Fore-Sight™, CAS Medical Systems, Branford, CT) with evaluation for absolute decreases or relative asymmetry during deep hypothermic circulatory arrest (DHCA) or selective antegrade cerebral perfusion (SACP), and right axillary artery cannulation with a sidearm Dacron graft via a separate infraclavicular incision. If arch or hemiarch repair is required, SACP is achieved via the right axillary artery graft with clamping of the innominate artery and back-clamping of the left carotid artery. The arterial line is placed preferentially on the left to allow for monitoring of systemic perfusion pressure while the right axillary or innominate artery is clamped. The author sought to systematically review contemporary institutional experience with the perioperative management of acute type-A aortic dissection in patients with arch anomalies, with a focus on assessing the hypothesis that the management of these patients differs in important ways from standard institutional practice in patients with normal arch anatomy.
"bovine" arch patterns (n ¼ 32), management differed in the conduct of selective antegrade cerebral perfusion, which could include clamping above or below the takeoff of the left common carotid artery (and, therefore, produced unilateral or bilateral antegrade cerebral perfusion). All patients with a connective tissue disorder exhibited a bovine arch pattern. Management of patients with a right arch (n ¼ 3) reflected the opposite of management for normal anatomy (for patients with traditional mirror-image branching) or opposite that of the aberrant right subclavian group (for patients who had a corresponding aberrant left subclavian artery). Conclusions: Rational management reflected the anatomic variations observed. These results support the importance of interdisciplinary planning, especially in an emergency, to optimize outcome. & 2013 Elsevier Inc. All rights reserved. KEY WORDS: type-A aortic dissection, acute aortic dissection, aortic arch anomalies, abnormal aortic arch, perfusion strategy, cannulation strategy
METHODS The authors performed a retrospective cohort analysis of all identifiable instances of operative intervention for acute Stanford type-A aortic dissection in patients with anatomic anomalies of the aortic arch occurring during the interval from January 1, 2001 to December 31, 2011. This research received Institutional Review Board approval. Potential cases initially were identified through a combination of International Statistical Classifications of Diseases (ICD-9) diagnostic codes for type-A dissection and free-text searches within the institutional clinical database for permutations on the terms “aortic arch anomaly” and “type-A dissection” (see Appendix). The authors identified 6 major patterns of abnormal arch anatomy for the search (Fig 1). Two patterns reflect what commonly is referred to as a "bovine" arch: An arch with a common origin of the innominate artery and left common carotid (Fig 1C) and the origin of the left common carotid off of the innominate artery (Fig 1D). Several prior reports have pointed out that in addition to being ambiguous, “bovine arch” is a misnomer, because neither of these patterns reflects the arch branching pattern found in cattle (a single common brachiocephalic trunk that gives rise to all head and extremity vessels; Fig 1B).5 The other 4 branching patterns identified were a 4-vessel arch with an aberrant right subclavian artery originating from the distal arch or
From the *Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD; and the †Departments of Cardiothoracic Surgery and ‡Anesthesiology, Perioperative and Pain Medicine, Stanford University School of Medicine, Stanford, CA. Address correspondence and reprint requests to Bryan G. Maxwell, MD, MPH; Johns Hopkins University School of Medicine, Department of Anesthesiology and Critical Care Medicine; 1800 Orleans Street Zayed 6208P; Baltimore, MD, 21287. Tel.: þ(410) 955 9147; Fax: þ(410) 955 0994 E-mail: [email protected] © 2013 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2013.12.001
Journal of Cardiothoracic and Vascular Anesthesia, Vol ], No ] (Month), 2013: pp ]]]–]]]
1
2
MAXWELL ET AL
Fig 1. Taxonomy of arch branching patterns. (A) Normal aortic arch. (B) True bovine pattern (ie, single common brachiocephalic trunk). (C) Common origin of the innominate artery and left common carotid (1 of 2 “bovine arch” patterns in humans). (D) Origin of the left common carotid off of the innominate artery (second of 2 “bovine arch” patterns in humans). (E) Four-vessel arch with an aberrant right subclavian artery (ARSCA) originating from the distal arch or proximal descending aorta. (F) Four-vessel arch with the left vertebral artery coming directly off the aorta. (G) Right aortic arch. (H) Double aortic arch.
proximal descending aorta (Fig 1E), a 4-vessel arch with the left vertebral artery coming directly off the aorta (Fig 1F), a right aortic arch (Fig 1G), and a double aortic arch (Fig 1H). The embryologic etiologies of these branching patterns are thought to involve a complex interplay of cell-signaling pathways and hemodynamic forces and remain under active investigation.6 Potential cases were reviewed individually to exclude cases that did not meet inclusion criteria and to avoid double-counting of cases. The clinical database techniques used in this project were supported by the NIH/NCRR CTSA award number UL1 RR025744. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Research Resources or the National Institutes of Health.7 Criteria for inclusion in the retrospective cohort were adult patients (age greater than or equal to 18 years) who underwent operative intervention for aortic dissection within 48 hours of admission to the institution. Individual case review was performed using details from the operative report, imaging reports, and primary source images from computed tomography (CT) scans and intraoperative transesophageal echocardiography (TEE) to confirm the following information: The dissection was acute in nature, the dissection involved the ascending aorta between the aortic root and the takeoff of the first arch branch vessel, the aortic arch anatomy represented one of several patterns of abnormal arch branching, and the patient underwent operative repair of the dissection. In addition to the operative report and imaging studies, anesthesia records, perfusion records, progress notes, and discharge summaries were used to determine operative and postoperative variables including cannulation strategy, method of surgical repair, site of primary intimal tear (PIT), bypass times and nadir temperature, use of DHCA or SACP, and incidence of in-hospital neurologic complications and mortality.
These perioperative variables were used to compare the management of patients with arch anomalies to the standard institutional practice for management of acute type-A dissection in patients with normal arch anatomy. RESULTS
A total of 161 potential cases were identified from the database search criteria. After individual case review, 43 met criteria for inclusions. Of these, 32 were 1 of the 2 “bovine arch” patterns: 23 had a common origin of the innominate artery and left common carotid artery (Fig 1C) and 9 had a left common carotid off the innominate (Fig 1D). No cases of true bovine anatomy (single brachiocephalic trunk; Fig 1B) were identified. There were 5 patients with an aberrant right subclavian artery (ARSCA; Fig 1E), 3 patients with a 4-vessel arch with separate takeoff of the left vertebral artery (Fig 1F), 3 patients with a right aortic arch (Fig 1G), and no patients with a double aortic arch (Fig 1H). Table 1 contains demographic variables and Table 2 contains surgical variables for each group. All patients undergoing composite valve graft or ascending aorta replacement alone were repaired on cardiopulmonary bypass (CPB) without circulatory arrest. In all patients undergoing hemiarch replacement, deep hypothermic circulatory arrest with or without selective antegrade cerebral perfusion was employed. The PIT was in the root (21%) or ascending aorta (77%) in all but 1 case; 1 patient (ARSCA group) had a PIT in the arch with retrograde type-A extension.
3
ARCH ANOMALIES AND TYPE-A DISSECTION
Table 1. Characteristics of Patients by Arch Pattern Group
Age (mean ⫾ SD) Male White Connective Tissue Disorder Marfan Other Congenital Anomalies Bicuspid Aortic Valve Anomalous Coronary Artery Repaired TOF Repaired PAPVR
Bovine Arch
ARSCA
Vertebral
Right Arch
All
n ¼ 32
n¼5
n¼3
n¼3
n ¼ 43
47 28 27 6 5
⫾ 15 (88) (84) (19) (16)
57 2 4 0
⫾9 (40) (80)
44 1 1 0
3 2 0 0
(9) (6)
1 0 0 0
(20)
0 0 0 0
⫾9 (33) (33)
⫾ 27 (33) (67)
33 1 2 0
0 0 1 1
(33) (33)
47 32 34 6
⫾ 15 (74) (79) (14)
4 2 1 1
(9) (5) (2) (2)
NOTE: Values are number (percentage) unless otherwise specified as mean ⫾ standard deviation. Abbreviations: ARSCA, aberrant right subclavian artery; PAPVR, partial anomalous pulmonary venous return; SD, standard deviation; TOF, tetralogy of Fallot.
the takeoff of the left common carotid); this was more common in patients with the origin of the left common carotid off the innominate (Fig 1D; 8/8 patients ¼ 100%) than in patients with a common origin of the innominate and left common carotid (Fig 1C; 4/8 patients ¼ 50%). Unilateral cerebral perfusion resulted from clamping above the takeoff of the left common carotid in 4 of 15 patients (27%); all these cases occurred in patients with a common origin of the innominate and left
Among patients with either of the 2 bovine arch patterns (n ¼ 32), arterial cannulation was via the right axillary artery in 30 (94%) patients and via the right femoral artery in 2 (6%). In all patients undergoing hemiarch replacement (with or without aortic valve replacement; n ¼ 15), axillary artery cannulation was performed and selective antegrade cerebral perfusion via the right axillary artery was used. In 11 of 15 patients (73%), bilateral cerebral perfusion resulted (clamping occurred below
Table 2. Surgical Characteristics Group
Arterial Line Site Left upper extremity Right upper extremity Femoral Arterial Bypass Cannulation Right axillary* Left axillary* Femoral Ascending aorta Surgical Repair Ascending only CVG Ascending þ hemiarch CVG þ hemiarch Cases without DHCA/SACP Nadir temp (1C) CPB time (min) Cross-clamp time (min) Cases with DHCA/SACP Nadir temp (1C) CPB time (min) Cross-clamp time (min) SACP DHCA DHCA/SACP time (min)
Bovine Arch
ARSCA
Vertebral
Right Arch
All
n ¼ 32
n¼5
n¼3
n¼3
n ¼ 43
31 0 1
(97)
30 0 2 0
(94)
12 5 5 10
(38) (16) (16) (31)
(3)
(6)
29 161 114
⫾1 ⫾1 ⫾4
22 260 189 15 0 37
⫾3 ⫾ 63 ⫾ 56 (100) ⫾ 16
0 5 0 0 0 5 0 2 0 3 0
(100)
(100)
(40) (60)
29 97 61
⫾1 ⫾ 17 ⫾ 11
19 207 111 0 3 35
⫾2 ⫾ 86 ⫾ 37 (100) ⫾ 14
3 0 0
(100)
2 1 0
3 0 0 0
(100)
0 1 2 0
0 0 3 0
19 201 97 3 0 21
(100)
⫾1 ⫾ 18 ⫾ 17 (100) ⫾1
(67) (33)
(33) (67)
3 0 0 0
(100)
28 105 60
⫾1 ⫾ 57 ⫾ 32
36 6 1
(84) (14) (2)
33 1 9 0
(77) (2) (21)
17 5 11 10 22 29 132 86 21 20 237 154 18 3 34
(40) (12) (26) (23) (51) ⫾1 ⫾ 34 ⫾ 27 (49) ⫾2 ⫾ 66 ⫾ 62 (90) (10) ⫾ 14
NOTE. Values are number (percentage) or mean ⫾ standard deviation, as appropriate. Abbreviations: ARSCA, aberrant right subclavian artery; CPB, cardiopulmonary bypass; CVG, composite valve graft; DHCA, deep hypothermic circulatory arrest (without SACP); SACP, selective antegrade cerebral perfusion. *Includes axillary, subclavian, or innominate.
4
common carotid (Fig 1C; 4/8 ¼ 50% of patients in this group). No cerebral oximetry abnormalities were recorded in this group. Six patients (19%) had an identified connective tissue disorder (5 of these had Marfan syndrome). Other associated anomalies in this group included 3 patients with bicuspid aortic valve and 2 patients with an anomalous left main coronary off a single coronary ostium in the right coronary sinus. Of patients with ARSCA (n ¼ 5), 1 patient had an unusual combination of a bovine arch (common origin of right and left common carotid arteries) and aberrant retroesophageal right subclavian artery in the setting of a large arch aneurysm; the remaining 4 patients had a more typical 4-vessel arch and a retroesophageal right subclavian artery as the most distal branch vessel. In all patients, arterial line placement occurred in the right radial artery and arterial cannulation for bypass occurred via the right femoral artery with retrograde perfusion. The patients in this group who required hemiarch replacement (n ¼ 3) did not undergo SACP; instead, DHCA was used for the arch repair component (the only use of pure DHCA among all study patients). Cerebral protection was afforded by hypothermia only without antegrade or retrograde cerebral perfusion during circulatory arrest. The only associated connective tissue disorder or congenital abnormality in this group was 1 patient with a bicuspid aortic valve. All patients in the 4-vessel arch with vertebral artery group (n ¼ 3) had arterial cannulation via the right axillary artery and arterial line placement in the left radial artery. All surgical repairs consisted of replacement of the ascending aorta and hemiarch with SACP. There were no connective tissue disorders or associated congenital abnormalities identified in this group. Of patients with a right-sided aortic arch (n ¼ 3), 1 patient had a mirror-image 3-vessel aortic arch with left-sided innominate artery, right common carotid, and right subclavian arterial branching. For this patient, perioperative management involved the mirror-image equivalent of standard management for a patient with a left arch and normal branching pattern: The arterial line was placed in the right radial artery, arterial cannulation occurred at the left axillary artery via a separate infraclavicular incision, and surgical repair was an isolated repair of the ascending aorta. The remaining 2 patients had a right-sided aortic arch with an aberrant retroesophageal left subclavian artery—a mirror-image equivalent of the 4-vessel aberrant right subclavian artery group. These 2 patients both had arterial lines placed in the left radial artery, femoral arterial cannulation for bypass, and surgical repair involving isolated repair of the ascending aorta. There were no associated connective tissue disorders in the right arch group. Associated congenital abnormalities in this group included 1 patient with repaired tetralogy of Fallot and 1 patient with repaired partial anomalous pulmonary venous return. Outcomes included one mortality that occurred, in an 88year-old woman (common origin of the left carotid and innominate arteries) after a postoperative course complicated by multisystem organ failure. One nonfatal neurologic complication occurred, in a 41-year-old man (common origin of the left carotid and innominate arteries) who underwent successful
MAXWELL ET AL
repair that involved ascending aorta and hemiarch replacement (nadir temperature 171C, bypass time 227 minutes, aortic crossclamp time 142 minutes, unilateral SACP for 56 minutes). Postoperative imaging revealed a significant right middle cerebral artery stroke from which the patient recovered partially (he was discharged to home with some residual weakness but could perform all activities of daily living independently). His first comprehensive neurologic examination occurred postoperatively but, of note, the presenting symptoms of his dissection (which extended into the bilateral carotid arteries on his first CT scan) involved right eye visual changes and left lower extremity paresis. He had no neurologic pathology identified in the left hemisphere (after right unilateral ACP). DISCUSSION
The identified cases represent a frequency of arch anomalies that is consistent with the best prior estimate of the frequency of arch anomalies in the general population: An analysis of an unselected population of consecutive patients (n ¼ 861) undergoing CT scans in Leicester, United Kingdom, identified a bovine pattern in 20% of patients, a 4-vessel arch with the left vertebral artery off the arch in 6%, a 4-vessel arch with an aberrant right subclavian artery in 0.5%, and a right-sided arch in 0.2%.1 These estimates are consistent with the prior literature, including 1 angiography series8 and several smaller autopsy studies.9–11 Although population-based methodologies would be required to adequately assess whether individuals with arch anomalies have a higher incidence of aortic dissection compared with the general population, the results do not provide any grounds to suspect a significant departure from the baseline risk. Other authors have speculated that abnormal arch anatomy may be a risk factor for aortic dissection: For instance, that the acute angle of take-off in an aberrant right subclavian artery constitutes a locus of weakness in the aortic wall that may predispose this as the site of dissection.3 However, in the absence of retrograde dissection involving the ascending aorta, this particular example would represent a type-B dissection and as such is beyond the scope of this study. No evidence from the current study supported the association between arch anomalies and dissection insofar as there was only one case with a distal PIT that became a type-A dissection because of retrograde extension. The group that demonstrated the most significant departures from standard management was the aberrant right subclavian group. Site of cannulation in acute type-A dissection remains a controversial topic. Ascending aortic cannulation has been suggested as a safe option12,13 to avoid the risks of false lumen perfusion and atheroemboli from the descending aorta that are inherent in femoral cannulation with retrograde perfusion,14 though a recent, large (n ¼ 473) retrospective comparison of femoral to central cannulation (ascending aorta, axillary, or innominate artery) found no difference in mortality or permanent neurologic disability in the femoral cannulation group.15 For similar reasons, other centers prefer left axillary cannulation for initiation of bypass, but this technique requires the additional step of direct cannulation of the innominate
ARCH ANOMALIES AND TYPE-A DISSECTION
artery and/or left common carotid artery through the arch if antegrade cerebral perfusion is desired.16 Cannulations of the right axillary artery or innominate artery17 have been advocated as alternatives. Both avoid the need to manipulate the dissected ascending aorta (which also may have significant atheroma present, because atheromatous disease shares risk factors with aortic dissection18) and allow for subsequent SACP without additional cannulation sites. Innominate cannulation has the advantage of simplicity and avoiding a need for a separate incision.19 Axillary artery cannulation via a separate incision (standard practice) has the advantage of allowing the initiation of CPB and cooling before opening the sternum and mediastinum to maximize safety by minimizing the risk and consequences of iatrogenic rupture. However, the abnormal anatomy in the aberrant right subclavian group required a departure from this practice. These patients all had contralateral arterial line placement, femoral arterial cannulation instead of a separate infraclavicular dissection for axillary cannulation, and hemiarch replacement under pure hypothermic circulatory arrest instead of SACP. In this situation, if ACP is desired, it would have to be accomplished through direct carotid artery cannulation. The same is true if right axillary artery cannulation is performed in patients with ARSCA: Separate direct carotid artery cannulation is required to accomplish ACP because the right axillary artery does share an origin with the right carotid artery.20 The management implications of the 2 bovine arch branching patterns center around the use of antegrade cerebral perfusion. If right axillary cannulation is used for bypass, 2 options exist for SACP: A clamp can be applied to the innominate artery caudad to the takeoff of the left common carotid artery (resulting in antegrade perfusion of both carotid arteries—ie, bilateral ACP) or cephalad to the takeoff of the left common carotid artery (resulting in traditional perfusion of only the right carotid—ie, unilateral ACP). Bilateral ACP may be easier (and in this series was more common) in patients in whom the left common carotid artery originates from the innominate artery (see Fig 1D) than in those in whom the left common carotid and innominate arteries have a common origin at the aorta (see Fig 1D). When bilateral ACP is used, a patent circle of Willis is not necessary for bilateral cerebral perfusion, and cerebral oximetry should be interpreted with this in mind. For instance, if asymmetric cerebral oximetry values are observed, other causes should be considered, such as impaired venous drainage (from large-caliber jugular venous cannulae or head position with jugular vein kinking). If unilateral ACP is used, contralateral cerebral perfusion remains dependent on a patent circle of Willis and adequate pressure to reach the contralateral hemisphere. Concern for relative asymmetry of cerebral oximetry as opposed to absolute (presumably symmetric) values should be pursued with an understanding of this anatomic variant’s implications for SACP. As with any SACP case, management options would include performing separate cannulation of the left common carotid artery for bilateral ACP or raising the SACP flow or perfusion pressure. In addition, with a bovine arch branching pattern (see Fig 1C or 1D), an additional option exists: Moving the proximal innominate artery clamp to below the takeoff of the left common carotid artery to convert to bilateral ACP.
5
An alternative approach for conduct of CPB and SACP is the use of the innominate artery for arterial perfusion. Subsequent to the period of this study, the authors infrequently have used this approach when the innominate artery is found to be free of dissection. The innominate artery should be assessed carefully by preoperative imaging and direct inspection to rule out extension of the dissection. In this scenario placement of a 6-mm-to-8-mm conduit to the proximal innominate artery (a “chimney”) through the median sternotomy provides access for the conduct of CPB as well as SACP. Using the undissected innominate artery obviates the need for a second incision for exposure of the right axillary artery for placement of the chimney. Although this study was too small to provide insights into the effect of unilateral versus bilateral ACP on outcomes, the authors noted that one surgeon included language in an operative note that suggested a belief that the safe duration of SACP might be longer in a patient with a bovine arch in whom antegrade perfusion was provided to both carotid arteries by clamping of the bovine trunk below the left carotid takeoff. Further studies would be required to assess whether this belief is held widely (and, therefore, if it affects surgeon behavior) or if any outcome data support it. The large German registry of type-A dissection repair found no evidence of a neurologic outcome or mortality difference between patients receiving unilateral versus bilateral antegrade cerebral perfusion, though duration was not specifically examined between these groups.21 The authors did not observe any episodes of significant asymmetry of cerebral oximetry values during unilateral ACP; however, it was suspected that the retrospective methodology was less sensitive in detecting such a phenomenon than would be a prospective investigation. The authors speculated that cerebral oximetry offers value in detecting inadequate cerebral perfusion in the setting of abnormal branching patterns (with or without incomplete intracranial vascular connections at the circle of Willis), but this study was not powered to demonstrate this value. The group with a 4-vessel arch, including a direct takeoff of the left vertebral artery, did not differ significantly from standard management. However, it should be noted that this anatomic pattern likely would have implications in planning for distal repair. Antegrade stent graft placement at the time of open repair (often termed a “frozen elephant trunk”) or subsequent retrograde thoracic endovascular stent graft deployment would risk compromise of both the left subclavian and left vertebral arteries. Because vertebral artery flow often is implicated in maintaining collateral perfusion to the left upper extremity when the subclavian artery is sacrificed,22 4-vessel arch anatomy would require modification of standard technique (eg, carotid-subclavian transposition or bypass) to reduce the risk of arm ischemia. Patients with a right aortic arch fell into 2 groups: 1 patient had a mirror-image equivalent of normal anatomy, with contralateral equivalents for the placement of invasive monitoring and site of cannulation for bypass. Two patients had a mirror-image equivalent of the aberrant right subclavian group. It is noteworthy that 2 of 3 cases of right arch were associated with other congenital lesions that had required operative repair (410 years prior in both cases). It is possible that the dissection observed in these patients could be related to
6
MAXWELL ET AL
previous cannulation of the ascending aorta, though given the long time interval between events, a causal relationship might be less plausible. The authors observed that within this survey of arch anomalies, all the patients with an identified connective tissue disorder (Marfan syndrome and/or bicuspid aortic valve) were in 1 of the 2 bovine arch groups. Larger studies would be needed to further evaluate this observation, but this center is among several international referral centers for connective tissue disorders, and, as such, it is interesting that none of the other abnormal patterns was observed in this subpopulation. The principal limitations of this study were those of a single-institution, retrospective design, which has limited outcome information and does not allow for a robust estimation of whether individuals with variant aortic arch anatomy are at increased or decreased risk of aortic dissection relative to the population average. However, the patterns observed even in this small series showed consistency, and the departures from usual management—most notably in the ARSCA group—were physiologically justifiable. These results supported the conclusion that careful attention to arch anatomy should be a part of the planning of acute type-A aortic dissection on the part of anesthesiologists, surgeons, and
perfusionists. If high-quality CT images are not available or do not clearly confirm the arch anatomy, TEE has a potential role in identifying some variant branching patterns (eg, right arch, aberrant right subclavian) but is likely to lack sensitivity in detecting others (bovine arch, 4-vessel arch with left vertebral artery). This role would need to be evaluated further in subsequent studies to determine accuracy. Although these operations usually occur under emergent conditions with limited opportunity for leisurely preoperative workup and discussion, interdisciplinary communication is key in ensuring that all parties involved understand the implications of variant arch anatomy for modifications in monitoring, perfusion, and surgical repair strategies. ACKNOWLEDGEMENTS The authors gratefully acknowledge the clinical informatics assistance and support of Gomathi Krishnan.
APPENDIX A. SUPPORTING INFORMATION
Supplementary material cited in this article is available online at http://dx.doi.org/10.1053/j.jvca.2013.12.001.
REFERENCES 1. Jakanani GC, Adair W: Frequency of variations in aortic arch anatomy depicted on multidetector CT. Clin Radiol 65:481-487, 2010 2. Li Q-L, Zhang X-M, Zhang X-M: Aortic dissection originating from an aberrant right subclavian artery. J Vasc Surg 46: 1270-1273, 2007 3. Kikuchi K, Makuuchi H, Oono M, et al: Surgery for aortic dissection involving an aberrant right subclavian artery. Jpn J Thorac Cardiovasc Surg 53:632-634, 2005 4. Rasmussen DK, Dougherty J: Aortic dissection with vascular abnormalities. J Am Osteopath Assoc 111:407-409, 2011 5. Layton KF, Kallmes DF, Cloft HJ, et al: Bovine aortic arch variant in humans: clarification of a common misnomer. AJNR Am J Neuroradiol 27:1541-1542, 2006 6. Wang Y, Dur O, Patrick MJ, et al: Aortic arch morphogenesis and flow modeling in the chick embryo. Ann Biomed Eng 37: 1069-1081, 2009 7. Lowe HJ, Ferris TA, Hernandez PM, et al: STRIDE—an integrated standards-based translational research informatics platform. AMIA Annu Symp Proc 2009:391-395, 2009 8. Natsis KI, Tsitouridis IA, Didagelos MV, et al: Anatomical variations in the branches of the human aortic arch in 633 angiographies: clinical significance and literature review. Surg Radiol Anat 31:319-323, 2009 9. Edwards JE: Anomalies of the derivatives of the aortic arch system. Med Clin North Am 32:925-949, 1948 10. Gupta M, Sodhi L: Variations in branching pattern, shape, size and relative distances of arteries arising from arch of aorta. Nepal Med Coll J 7:13-17, 2005 11. Bhatia K, Ghabriel MN, Henneberg M: Anatomical variations in the branches of the human aortic arch: a recent study of a South Australian population. Folia Morphol (Warsz) 64:217-223, 2005 12. Reece TB, Tribble CG, Smith RL, et al: Central cannulation is safe in acute aortic dissection repair. J Thorac Cardiovasc Surg 133: 428-434, 2007
13. Kamiya H, Kallenbach K, Halmer D, et al: Comparison of ascending aorta versus femoral artery cannulation for acute aortic dissection type A. Circulation 120:S282-286, 2009 14. Tiwari KK, Murzi M, Bevilacqua S, et al: Which cannulation (ascending aortic cannulation or peripheral arterial cannulation) is better for acute type A aortic dissection surgery? Interact Cardiovasc Thorac Surg 10:797-802, 2010 15. Di Eusanio M, Pantaleo A, Petridis FD, et al: Impact of different cannulation strategies on in-hospital outcomes of aortic arch surgery: a propensity-score analysis. Ann Thorac Surg 96:16561663, 2013 16. Kano M, Chikugo F, Shimahara Y, et al: Left axillary artery perfusion in surgery of type A aortic dissection. Ann Thorac Cardiovasc Surg 14:22-24, 2008 17. Huang FJ, Wu Q, Ren CW, et al: Cannulation of the innominate artery with a side graft in arch surgery. Ann Thorac Surg 89: 800-803, 2010 18. Augoustides JGT, Harris H, Pochettino A: Direct innominate artery cannulation in acute type a dissection and severe thoracic aortic atheroma. J Cardiothorac Vasc Anesth 21:727-729, 2007 19. Di Eusanio M, Ciano M, Labriola G, et al: Cannulation of the innominate artery during surgery of the thoracic aorta: our experience in 55 patients. Eur J Cardiothorac Surg 32:270-273, 2007 20. Battaloglu B, Secici S, Colak C, et al: Aberrant right subclavian artery and axillary artery cannulation in type a aortic dissection repair. Ann Thorac Surg. e1-e2 , 2013 21. Krüger T, Weigang E, Hoffmann I, et al: Cerebral protection during surgery for acute aortic dissection type A: results of the German Registry for Acute Aortic Dissection Type A (GERAADA). Circulation 124:434-443, 2011 22. Woo EY, Bavaria JE, Pochettino A, et al: Techniques for preserving vertebral artery perfusion during thoracic aortic stent grafting requiring aortic arch landing. Vasc Endovascular Surg 40: 367-373, 2006
