Innominate artery cannulation for proximal aortic ...

ORIGINAL ARTICLE

European Journal of Cardio-Thoracic Surgery 48 (2015) 937–942 doi:10.1093/ejcts/ezu534 Advance Access publication 1 February 2015

Cite this article as: Preventza O, Garcia A, Tuluca A, Henry M, Cooley DA, Simpson K et al. Innominate artery cannulation for proximal aortic surgery: outcomes and neurological events in 263 patients. Eur J Cardiothorac Surg 2015;48:937–42.

Innominate artery cannulation for proximal aortic surgery: outcomes and neurological events in 263 patients† Ourania Preventzaa,b,*, Andrea Garciab, Alexandra Tulucab, Matthew Henryb, Denton A. Cooleya, Kiki Simpsonc, Faisal G. Bakaeena,b,c, Lorraine D. Cornwellb,c, Shuab Omerb,c and Joseph S. Cosellia,b a b c

Department of Cardiovascular Surgery, The Texas Heart Institute, Houston, TX, USA Division of Cardiothoracic Surgery, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, TX, USA The Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX, USA

* Corresponding author. BCM 390 One Baylor Plaza, Houston, TX 77030, USA. Tel: +1-832-3559910; fax: +1-832-3559920; e-mail: [email protected] (O. Preventza). Received 12 September 2014; received in revised form 1 December 2014; accepted 4 December 2014

Abstract OBJECTIVES: To determine whether innominate artery cannulation is the ideal perfusion strategy for delivering antegrade cerebral perfusion (ACP) during surgery on the proximal ascending aorta and transverse aortic arch. METHODS: A total of 263 patients underwent innominate artery cannulation with a side graft for surgery on the proximal aorta. Operations performed were ascending and proximal arch replacement (n = 213, 81.0%), ascending and total arch replacement (n = 33, 12.6%) and ascending aortic replacement (n = 12, 4.6%). Concomitant or other procedures included aortic root replacement and repair (n = 113, 43.0%), aortic valve replacement or repair (n = 118, 44.9%), coronary artery bypass (n = 40, 15.2%), antegrade stent graft delivery to the proximal descending thoracic aorta for aneurysm or dissection (n = 28, 10.7%), mitral valve repair (n = 11, 4.2%), patent foramen ovale repair (n = 3, 1.1%) and tricuspid valve repair (n = 2, 0.8%). Twenty-seven patients (10.3%) presented with acute or subacute Type I aortic dissection, and 45 (17.1%) had a previous sternotomy. Median cardiopulmonary bypass (CPB), cardiac ischaemia and ACP times were 126.0 [95–163 interquartile range (IQR)], 91.0 (73–121 IQR) and 21.0 (16–32 IQR) min. Bilateral ACP was delivered in 235 patients (90.7%). RESULTS: The operative mortality rate was 4.9% (n = 13). Nine patients (3.4%) had postoperative stroke, which was permanent in 5 (1.9%) of them. Multivariate analysis associated risk of stroke or temporary neurological deficit with acute or subacute Type I aortic dissection (P = 0.028) and age (P = 0.015). Renal disease (P = 0.036) and CPB time (P = 0.011) were independent risk factors for operative mortality. Circulatory arrest time was identified as a risk factor for mortality (P = 0.038). CONCLUSIONS: Innominate artery cannulation can be performed safely and poses a low risk of neurological events in procedures requiring hypothermic circulatory arrest. The technique for cannulating this artery should be part of the routine armamentarium of cardiac and aortic surgeons, and the innominate artery is among the preferred perfusion sites for delivering ACP.

INTRODUCTION The causes of neurological events in patients undergoing proximal aortic surgery are multifactorial. Underlying pathology and cerebral protection strategies play a significant role in neurological outcome in such patients [1]. Antegrade cerebral perfusion (ACP) during a period of modified circulatory arrest has been shown to improve neurological outcomes. Several sites and techniques have been evaluated for the purpose of establishing antegrade flow [2]. Both central and peripheral artery cannulation during proximal aortic surgery have been advocated by well-respected groups, so † Presented at the 28th Annual Meeting of the European Association for CardioThoracic Surgery, Milan, Italy, 11–15 October 2014.

debate still persists [3–8]. We reviewed our experience with innominate artery cannulation in patients undergoing proximal arch and total aortic arch surgery and the overall outcomes and neurological events in these patients.

MATERIALS AND METHODS From a prospectively maintained clinical database, we obtained data that had been collected over a recent 2.5-year period from 263 consecutive patients {median age, 60.0 years [interquartile range (IQR), 51–70]; 191 [72.6%] male}. All procedures had been done via median sternotomy with innominate artery cannulation through a side graft for arterial inflow during cardiopulmonary

© The Author 2015. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

Downloaded from https://academic.oup.com/ejcts/article-abstract/48/6/937/2464876 by guest on 21 April 2018

AORTIC SURGERY

Keywords: Innominate artery cannulation • Aortic surgery • Axillary artery cannulation

938

O. Preventza et al. / European Journal of Cardio-Thoracic Surgery

bypass (CPB) (Table 1). The Institutional Review Board of Baylor College of Medicine approved the study. The procedures for which innominate artery cannulation was used, as well as the concomitant operations and intraoperative times, are listed in Table 2. Follow-up data were obtained from hospital records, clinic visits and the Social Security death index. The innominate artery was not cannulated if any of the following conditions existed: atherosclerosis at the base of the innominate, aneurysmal dilatation of the innominate, redo sternotomy with the aorta in close proximity to the sternum, the necessity to initiate CPB before opening the sternum and Type I aortic dissection extending into the innominate artery. Open distal anastomosis was performed in all except 17 cases. In these 17 cases, the embolic risk in case of cannulation of the distal arch, as well as the ability to gain more length of extra aorta during performance of the distal anastomosis, was the reason for innominate artery cannulation.

Definitions Pulmonary disease was defined as a history of restrictive or obstructive lung disease. Preoperative renal disease was defined as haemodialysis dependence or a creatinine level ≥1.5 mg/dl. We have previously described the outcome definitions used in the present study [5]. Briefly, operative death was defined as death within 30 days or before hospital discharge, stroke as a new brain injury evident clinically or radiographically after the procedure, and a temporary neurological deficit (TND) as postoperative agitation or confusion with no injury detected by magnetic resonance

imaging or computed tomography. Postoperative renal injury was defined as the need to start dialysis or a doubling of the patient’s serum creatinine level. ‘Aortic risk’ was used to indicate high-risk procedures: total arch operations, Stage I elephant trunk procedures, a diagnosis of acute or subacute aortic dissection, redo sternotomy and emergency operations. Intraoperative times were defined as follows: the ACP time was defined as the period of circulatory arrest during which the patient was receiving ACP. The circulatory arrest time was the total time of circulatory arrest with and without ACP. The cardiac ischaemia time was the interval from the initiation of circulatory arrest or cross-clamping until clamp removal. Finally, the CPB time was the period of CPB, not including the ACP time or the circulatory arrest time.

Intraoperative procedure We have previously reported our surgical technique of innominate artery cannulation using an 8-mm Terumo Vascutek Gelweave graft (Vascutek Terumo, Renfrewshire, Scotland) [5]. Five thousand units of heparin is given by the anaesthesia team before a partial occluding clamp is placed on the innominate artery. Near-infrared spectroscopic monitoring is used for cerebral oximetry during the partial clamping of the innominate. An arteriotomy is performed, and the graft is sewn to the artery in an end-to-side fashion with a continuous 6–0 Prolene suture. The mean arterial blood pressure is maintained at 70–90 mmHg. The innominate artery graft is used to deliver ACP via the right common carotid artery during open distal proximal aortic repair. Our target flow is 10–15 ml/kg/min, which is adjusted according to the near-infrared spectroscopy signal [5]. Left-sided cerebral perfusion into the left common carotid artery is achieved via a 9-F Pruitt perfusion balloon-tip catheter (LeMaitre Vascular, Inc., Burlington, MA, USA) connected

Table 1: Preoperative characteristics (n = 263) Age (years) Male gender Diagnosis of acute or subacute Type I aortic dissection Diagnosis of chronic dissection Aortic riska Hypertension Chronic obstructive pulmonary disease Prior coronary artery disease or myocardial infarction Emergency/urgent need for operation Previous sternotomy Renal disease Diabetes Prior proximal aortic surgeryb Congestive heart failure Cerebrovascular event (stroke with or without deficit, TIA) Prior percutaneous coronary intervention Prior thoraco-abdominal surgery Prior coronary artery bypass grafting More than one previous sternotomy

60.0 (51–70) 191 (72.6) 27 (10.3) 18 (6.8) 96 (36.5) 221 (84.0) 93 (35.4) 68 (25.9) 51 (19.4) 45 (17.1) 33 (12.6) 31 (11.8) 28 (10.7) 25 (9.5) 20 (7.6) 14 (5.3) 13 (4.9) 7 (2.7) 4 (1.5)

Data are reported as median [interquartile range (25–75%)] for continuous variables and number (percentage) for categorical variables. TIA: transient ischaemic attack. a Aortic risk refers to total arch operations, Stage I elephant trunk procedures, redo sternotomy, acute or subacute Type I aortic dissection and emergency procedures. b Prior proximal aortic surgery includes previous aortic root, ascending aorta, hemiarch and arch operations.


Table 2:

Indications for surgery (n = 263)

Ascending and hemiarch Ascending and total arch replacement with or without ET Ascending replacement only Aortic valve replacement only Concomitant procedures Aortic root repair or replacement Aortic valve replacement Aortic valve resuspension and commissuroplasty Coronary artery bypass grafting Antegrade delivery of a covered thoracic stent graft Maze Mitral valve repair Patent foramen ovale repair Tricuspid valve repair Intraoperative times (min) Cardiopulmonary bypass time (n = 263) Cardiac ischaemia time (n = 263) Antegrade cerebral perfusion time (n = 246) Circulatory arrest time (n = 246) Pure circulatory arrest time (n = 21) Bilateral antegrade cerebral perfusion

213 (81.0) 33 (12.6) 12 (4.6) 5 (1.9) 113 (43.0) 64 (24.3) 54 (20.5) 40 (15.2) 28 (10.7) 19 (7.2) 11 (4.2) 3 (1.1) 2 (0.8) 126.0 (95–163) 91.0 (73–121) 21.0 (16–32) 21.0 (16–32) 6.0 (4–11) 235 (90.7)

Data are reported as median [interquartile range (25–75%)] for continuous variables and number (percentage) for categorical variables. ET: elephant trunk.


Statistical analysis For univariate analysis, to test for significant differences between groups (i.e. patients who died versus patients who did not, and patients who had neurological events versus patients who did not), we used the non-parametric Mann–Whitney U-test for continuous variables and the χ 2 test or the Fisher’s exact test, when necessary, for categorical variables. Significance was defined as P < 0.05. For the multivariate analysis, exact logistic regression was performed. Initially, a nominal logistic regression was performed with 17 preoperative and intraoperative variables that were significant univariate predictors of the outcome being modelled: age, ascending aortic or arch disease, prior thoraco-abdominal surgery, redo sternotomy, prior proximal aortic surgery, congestive heart failure (CHF), renal disease, chronic obstructive pulmonary disease, cardiac disease [i.e. existing coronary artery disease, a previous myocardial infarction, or previous coronary artery bypass grafting (CABG) or percutaneous transluminal coronary angioplasty], previous stroke (with and without deficits), acute or subacute Type I aortic dissection, concomitant aortic valve replacement, concomitant CABG, CPB time, circulatory arrest time, ACP time and cardiac ischaemia time. The variables identified in this step with a P-value cut-off of 0.1 became candidates for the exact logistic regression models, which was necessary because of the rarity of the outcome events being modelled. All statistical analyses were conducted with SAS version 9.1 (SAS Institute, Inc., Cary, NC, USA).

RESULTS Neurological events Nine patients (3.4%) suffered postoperative strokes, 2 of which were fatal (Table 3). Four patients had complete recovery, and 1 had partial recovery. Table 4 gives the significant univariate predictors of stroke, new neurological events (stroke or TND) and mortality. Multivariate analysis of data from the complete cohort of patients (including 17 patients who did not require open distal

Table 3: Postoperative outcomes (n = 263) In-hospital mortality Stroke Permanent Temporary neurological deficit Paralysis Paraparesis Ventilation for more than 48 h Renal injury Myocardial infarct Tracheostomy Pericardial window Reoperation for bleeding Wound dehiscence

13 (4.9) 9 (3.4) 5 (1.9) 10 (3.8) 1 (0.4) 2 (0.8) 65 (24.7) 24 (9.1) 2 (0.8) 26 (9.9) 8 (3.0) 15 (5.7) 8 (3.0)

anastomosis with circulatory arrest and ACP) identified age as an independent risk factor for stroke alone (P = 0.032). For stroke or any new TND, age and the diagnosis of acute or subacute Type I aortic dissection were identified as independent risk factors (P = 0.015 and P = 0.028, respectively) (Table 5). None of the intraoperative times (CPB, ACP and circulatory arrest time), when inserted into the multivariate model, were risk factors for a new neurological event (Table 6). (The 17 patients who did not require open distal anastomosis were excluded from this analysis because their operations did not involve CPB, ACP or circulatory arrest.) In this series, 3 patients developed spinal cord ischaemia (1 had paralysis, and 2 had transient paraparesis). All 3 patients had undergone a concomitant descending thoracic aortic intervention. The patient who had paralysis underwent a frozen elephant trunk procedure for arch and descending aortic aneurysm, and the 2 patients with transient paraparesis had been treated for Type I aortic dissection with the antegrade delivery of a covered endovascular stent.

Mortality The overall operative mortality rate was 4.9% (n = 13). Eight patients died within 30 days, 2 patients died at 1 month and 1 patient each died at 2, 3 and 4 months, respectively. Two patients had a fatal stroke, 1 after total arch replacement for mega-aorta and 1 after a proximal arch operation for acute Type I aortic dissection with malperfusion. Four patients (2 of whom had redo operations) required prolonged extracorporeal membrane oxygenation (ECMO) support; none of them could be separated from ECMO, and all 4 died. One patient, who was morbidly obese and had dilated cardiomyopathy and an ejection fraction of 20%, had a postoperative myocardial infarction and was taken immediately to the catheterization laboratory, but the intervention was unsuccessful. Another patient, while recovering on the floor after a proximal arch operation, developed asystole and died. One patient had a Stage I elephant trunk procedure from which he was recovering well. Because of the large size of the thoraco-abdominal aorta (8 cm) and the new onset of severe back pain, the patient underwent emergency completion Stage II elephant trunk open repair but died 12 days after the second operation. Finally, 4 patients died from multiorgan failure. In univariate analysis, prior thoraco-abdominal aortic surgery (P = 0.020), CHF (P = 0.0040) and preoperative renal disease (P = 0.014) were risk factors for mortality, as were intraoperative ACP time (P = 0.026), CPB time (P = 0.041) and circulatory arrest time (P = 0.023) (Table 4). The multivariate logistic analysis model that included all patients (including the 17 who did not require open distal anastomosis with circulatory arrest and ACP) showed that preoperative renal disease and CPB time were independent predictors of operative mortality (P = 0.036 and P = 0.011, respectively) (Table 5). When the multivariate analysis was performed only in patients with open distal repair (i.e. excluding the 17 patients mentioned previously), prior thoraco-abdominal aortic surgery, preoperative CHF and circulatory arrest time were found to be independent predictors of operative mortality (Table 6).

Other outcomes

Data are reported as number (percentage).

Table 3 outlines the other adverse outcomes.


AORTIC SURGERY

to the arterial line by a Y-connector. The temperature used during the arch repair was 24–28°C.

939


940

Table 4: Univariate analysis of mortality, stroke, temporary neurological deficit (TND) and combined stroke/TND

Ascending and hemiarch Ascending and arch Prior thoraco-abdominal aortic surgery Preoperative congestive heart failure Preoperative renal disease Antegrade cerebral perfusion time (min) Cardiopulmonary bypass time (min) Circulatory arrest time (min) Cardiac ischaemia time (min) Aortic risk

Concomitant aortic root operation Acute or subacute Type I aortic dissection Aortic risk Age (years)

Concomitant aortic valve replacement Prior proximal aortic surgery Circulatory arrest time (min)

Concomitant aortic root repair or replacement Acute or subacute Type I aortic dissection Preoperative renal disease Age (years) Antegrade cerebral perfusion time (min) Circulatory arrest time (min)

Mortality = No (n = 250)

Mortality = Yes (n = 13)

P-value

205 (82.0) 30 (12.0) 10 (4.0) 20 (8.0) 28 (11.2) 20.0 (16–30) 125.0 (95–161) 20.0 (16–31) 91 (73–119) 88 (35.2)

8 (61.5) 3 (23.1) 3 (23.1) 5 (38.5) 5 (38.5) 43.0 (19–59) 153.0 (133–238) 43.0 (19–78) 111.0 (80–162) 8 (61.5)

0.078 0.21 0.020 0.0040 0.014 0.026 0.041 0.023 0.056 0.075

Stroke = No (n = 254)

Stroke = Yes (n = 9)

P-value

112 (44.1) 24 (9.5) 90 (35.4) 60.0 (51–70)

1 (11.1) 3 (33.3) 6 (66.7) 67.0 (63–73)

0.083 0.053 0.078 0.051

TND = No (n = 253)

TND = Yes (n = 10)

P-value

59 (23.3) 25 (9.9) 20.0 (16–32)

5 (50.0) 3 (30.0) 29.5 (26–69)

0.067 0.078 0.050

Stroke or TND = No (n = 244)

Stroke or TND = Yes (n = 19)

P-value

110 (45.1) 22 (9.0) 28 (11.5) 60.0 (51–69) 20.0 (16–30) 20.0 (16–32)

3 (15.8) 5 (26.3) 5 (26.3) 65.0 (55–74) 30.0 (19–69) 30.0 (18–69)

0.015 0.033 0.073 0.060 0.019 0.040

Data are reported as median [interquartile range (25–75%)] for continuous variables and number (percentage) for categorical variables.

Table 5: Multivariate exact logistic regression analysis for all patientsa Risk factor

P-value

In-hospital mortality risk factors Preoperative renal disease 0.036 Cardiopulmonary bypass time 0.011 Postoperative stroke risk factors Age 0.032 Postoperative stroke or TND risk factors Age 0.015 Acute or subacute Type I 0.028 aortic dissection

Odds ratio

95% CI for odds ratio

Table 6: Multivariate exact logistic regression analysis for patients who required open distal anastomosis (17 patients who did not require antegrade cerebral perfusion or circulatory arrest time were excluded) Risk factor

5.61 1.01

1.11–27.72 1.00–1.02

1.07

1.01–1.14

1.05 4.86

1.01–1.09 1.18–17.69

CI: confidence interval; TND: temporary neurological deficit; ACP: antegrade cerebral perfusion. a This model does not include ACP or circulatory arrest time as independent variables because not every case involved the use of circulatory arrest or ACP.

DISCUSSION Cannulation of the innominate artery for CPB was initially described in 2000 by Banbury and Cosgrove [9]. Several modifications, including direct cannulation and the use of side grafts, have been since reported [10, 11].


P-value

In-hospital mortality risk factors Preoperative CHF 0.0019 Circulatory arrest timea 0.038 Prior thoraco-abdominal 0.032 aortic surgery Postoperative stroke risk factors Ageb 0.048 Postoperative stroke or TND risk factors b Age 0.020 Acute or subacute Type I 0.034 aortic dissection

Odds ratio

95% CI for odds ratio

14.74 1.43 0.079

2.58–86.83 1.02–1.97 0.008–0.82

1.80

1.01–3.60

1.57 4.63

1.07–2.43 1.12–16.92

CHF: congestive heart failure; CI: confidence interval; TND: temporary neurological deficit. a Circulatory arrest time is categorized in 15-min intervals. b Age is categorized as follows: ≤30, 31–40, 41–50, 51–60, 61–70, 71–80 and ≥80 years.

Brain protection is the cornerstone of proximal aortic surgery. Evolving data show that ACP helps protect the brain during complex aortic reconstruction and produces better neurological outcomes



short period of time (2.5 years). Our postoperative stroke and TND rates were 3.4 and 3.8%, respectively. Di Eusanio et al. [10], who used the same side graft technique in 55 patients, reported no strokes and a 1.8% rate of TND. The difference between their results and ours may be due to the limited extent of the proximal aortic operations in their series, as well as their reporting of permanent neurological events only. Their series included 10.9% hemiarch operations and 16.4% total arch operations, compared with 81% hemiarch and 12.6% total arch operations in our series. In addition, our outcomes included all strokes, not just permanent strokes. Of our 9 patients with stroke, 4 had complete recovery at discharge, making the permanent stroke rate 1.9% (5/263). Other groups have associated direct innominate artery cannulation with stroke rates of 0–4.2% [7, 23]. To prevent dissection at the cannulation site and to avoid the effect of turbulent flow close to a heavily atherosclerotic or dissected arch, we prefer to attach a side graft to the innominate artery instead of cannulating it directly. Type I aortic dissection and age were independent predictors of the risk of stroke or TND in our series (as shown by the multivariate analysis models both with and without the patients who required open distal anastomosis), as others have also reported [7, 24]. In our earlier series of 68 patients [5], age and ACP time also predicted neurological events. However, in the current study, ACP time was not found to be an independent predictor of stroke. With regard to mortality, in the full patient cohort, preoperative renal disease was an independent predictor of mortality, in accordance with other studies [7]. In addition, CPB time was an independent predictor of operative mortality (P = 0.011). Of interest, CPB was not a predictor of either stroke or mortality for patients requiring open distal anastomosis. In the multivariate logistic regression analysis that included only patients with open distal anastomosis (i.e. the analysis from which 17 patients were excluded), circulatory arrest time was found to be an independent predictor of operative mortality, as others have also found [7].

Limitations This was a non-randomized, observational study whose cohort consisted solely of patients who underwent innominate artery cannulation. No effort was made to compare alternative cannulation sites. Surgical bias with regard to operative strategy in the care of these complex pathologies is inherent.

CONCLUSION For aortic procedures requiring open distal anastomosis and hypothermic circulatory arrest, innominate artery cannulation poses little neurological risk, and the ability to cannulate this artery should be part of the routine armamentarium of cardiac and aortic surgeons, and the innominate should be considered one of the preferred perfusion sites for delivering ACP. Age and Type I aortic dissection are independent predictors of stroke in this patient population. To our knowledge, this is the largest reported series to date of innominate artery cannulations in patients undergoing proximal aortic surgery.

ACKNOWLEDGEMENTS Stephen N. Palmer, assisted with the editing of the manuscript.

AORTIC SURGERY

than cannulating the femoral artery with only deep hypothermic circulatory arrest [7, 12]. Hypothermia at various levels is a component of most strategies for cerebral protection. Profound hypothermia alone has been shown to be effective by investigators at Yale [13]. Among the commonly used cannulation sites, the innominate artery is less often chosen than others, and the optimal cannulation site is still a subject of debate [5–7, 14, 15]. The main advantage of axillary cannulation when compared with femoral cannulation is the ability to deliver ACP through the initial cannulation site in technically challenging aortic repairs. Although axillary cannulation is well tolerated, it is not without potential adverse consequences, including the need to make an additional incision, brachial plexus injury, arm ischaemia, localized dissection and malperfusion [16–18]. Compared with femoral and axillary cannulation, innominate cannulation provides the following benefits: there is no risk of brachial plexus injury and arm ischaemia, and there is no need for an additional incision (the innominate cannulation site is easily accessible via median sternotomy). Cerebral atheroembolism is minimized because, unlike femoral arterial cannulation, innominate cannulation does not result in retrograde flow through a potentially atherosclerotic thoraco-abdominal aorta. Also, in cases of acute or chronic aortic dissection, the malperfusion that occasionally follows femoral arterial cannulation is avoided. We began using innominate artery cannulation in proximal aortic surgery on a routine basis in June 2011. We have reported our initial experience with this technique in 68 patients [5]. It is our practice to use unilateral or bilateral ACP in aortic reconstructions in which open distal anastomosis is required. To facilitate the use of this strategy, we have incorporated innominate and right axillary cannulation [5, 18]. Currently, in selected cases requiring redo sternotomy in which the aortic pathology is not close to the posterior aspect of the sternum, the innominate artery is our preferred site; otherwise, right axillary cannulation is our first choice. In cases of severe haemodynamic instability, in which there is a need to establish immediate CPB, we occasionally access the femoral artery percutaneously or by cut-down. Femoral artery cannulation used to be the main form of cannulation used for patients with aortic dissections and aneurysms of the transverse arch, and it has been used extensively by certain aortic surgical groups. Ayyash et al. [19] reported 767 femoral artery cannulations for thoracic aortic surgery; 572 of them were for ascending aortic and arch operations. The majority were elective cases, and 12.2% were emergency cases. The operative survival rate was 95%, the stroke rate was 1.6% and the rate of local wound complications was 3.2%. The use of axillary artery cannulation was shown nearly 20 years ago to be simple and safe for CPB and cerebral perfusion [20], although it is not without pitfalls [17, 21]. In a cohort of 869 patients at Mount Sinai hospital who underwent replacement of the ascending aorta or repair or replacement of the aortic root, using axillary cannulation rather than femoral or direct aortic cannulation reduced mortality and stroke risk [12]. Svensson et al. [7] at the Cleveland Clinic reported that the stroke rates associated with axillary, direct aortic and direct innominate cannulation were 4, 7.8 and 4.2%, respectively. In a recent survey of European cardiac surgery centres, the axillary artery was the most commonly used site of cannulation for both acute (54%) and chronic (48%) aortic pathology. For cases of acute aortic pathology, the axillary artery was the second most preferred site (40%), followed by the femoral (33%) and innominate arteries (11%) [22]. Our study is not designed to compare innominate artery cannulation with cannulation at other sites. Rather, it is an observational study of a large number of consecutive patients treated over a

941

942


Conflict of interest: Joseph S. Coselli serves as a consultant to and receives royalties from Vascutek Ltd, a subsidiary of Terumo Corporation.

REFERENCES [1] Hagl C, Ergin MA, Galla JD, Lansman SL, McCullough JN, Spielvogel D et al. Neurologic outcome after ascending aorta-aortic arch operations: effect of brain protection technique in high-risk patients. J Thorac Cardiovasc Surg 2001;121:1107–21. [2] Khaladj N, Shrestha M, Meck S, Peterss S, Kamiya H, Kallenbach K et al. Hypothermic circulatory arrest with selective antegrade cerebral perfusion in ascending aortic and aortic arch surgery: a risk factor analysis for adverse outcome in 501 patients. J Thorac Cardiovasc Surg 2008;135: 908–14. [3] Fusco DS, Shaw RK, Tranquilli M, Kopf GS, Elefteriades JA. Femoral cannulation is safe for type A dissection repair. Ann Thorac Surg 2004;78: 1285–9; discussion 9. [4] Khaladj N, Shrestha M, Peterss S, Strueber M, Karck M, Pichlmaier M et al. Ascending aortic cannulation in acute aortic dissection type A: the Hannover experience. Eur J Cardiothorac Surg 2008;34:792–6; discussion 6. [5] Preventza O, Bakaeen FG, Stephens EH, Trocciola SM, de la Cruz KI, Coselli JS. Innominate artery cannulation: an alternative to femoral or axillary cannulation for arterial inflow in proximal aortic surgery. J Thorac Cardiovasc Surg 2013;145:S191–6. [6] Strauch JT, Spielvogel D, Lauten A, Lansman SL, McMurtry K, Bodian CA et al. Axillary artery cannulation: routine use in ascending aorta and aortic arch replacement. Ann Thorac Surg 2004;78:103–8; discussion 8. [7] Svensson LG, Blackstone EH, Rajeswaran J, Sabik JF III, Lytle BW, Gonzalez-Stawinski G et al. Does the arterial cannulation site for circulatory arrest influence stroke risk? Ann Thorac Surg 2004;78:1274–84. [8] Tiwari KK, Murzi M, Bevilacqua S, Glauber M. Which cannulation (ascending aortic cannulation or peripheral arterial cannulation) is better for acute type A aortic dissection surgery? Interact CardioVasc Thorac Surg 2010;10:797–802. [9] Banbury MK, Cosgrove DM III. Arterial cannulation of the innominate artery. Ann Thorac Surg 2000;69:957. [10] Di Eusanio M, Ciano M, Labriola G, Lionetti G, Di Eusanio G. Cannulation of the innominate artery during surgery of the thoracic aorta: our experience in 55 patients. Eur J Cardiothorac Surg 2007;32:270–3. [11] Huang FJ, Wu Q, Ren CW, Lai YQ, Yang S, Rui QJ et al. Cannulation of the innominate artery with a side graft in arch surgery. Ann Thorac Surg 2010; 89:800–3. [12] Etz CD, Plestis KA, Kari FA, Silovitz D, Bodian CA, Spielvogel D et al. Axillary cannulation significantly improves survival and neurologic outcome after atherosclerotic aneurysm repair of the aortic root and ascending aorta. Ann Thorac Surg 2008;86:441–6; discussion 6–7. [13] Ziganshin BA, Rajbanshi BG, Tranquilli M, Fang H, Rizzo JA, Elefteriades JA. Straight deep hypothermic circulatory arrest for cerebral protection during aortic arch surgery: safe and effective. J Thorac Cardiovasc Surg 2014;148:888–900. [14] LeMaire SA, Price MD, Parenti JL, Johnson ML, Lay AD, Preventza O et al. Early outcomes after aortic arch replacement by using the Y-graft technique. Ann Thorac Surg 2011;91:700–8. [15] Sabik JF, Nemeh H, Lytle BW, Blackstone EH, Gillinov AM, Rajeswaran J et al. Cannulation of the axillary artery with a side graft reduces morbidity. Ann Thorac Surg 2004;77:1315–20.


[16] Rescigno G, Aratari C, Matteucci ML. Axillary artery cannulation pitfalls. J Thorac Cardiovasc Surg 2009;138:251; author reply 2. [17] Schachner T, Nagiller J, Zimmer A, Laufer G, Bonatti J. Technical problems and complications of axillary artery cannulation. Eur J Cardiothorac Surg 2005;27:634–7. [18] Wong DR, Coselli JS, Palmero L, Bozinovski J, Carter SA, Murariu D et al. Axillary artery cannulation in surgery for acute or subacute ascending aortic dissections. Ann Thorac Surg 2010;90:731–7. [19] Ayyash B, Tranquilli M, Elefteriades JA. Femoral artery cannulation for thoracic aortic surgery: safe under transesophageal echocardiographic control. J Thorac Cardiovasc Surg 2011;142:1478–81. [20] Sabik JF, Lytle BW, McCarthy PM, Cosgrove DM. Axillary artery: an alternative site of arterial cannulation for patients with extensive aortic and peripheral vascular disease. J Thorac Cardiovasc Surg 1995;109:885–90; discussion 90–1. [21] Sinclair MC, Singer RL, Manley NJ, Montesano RM. Cannulation of the axillary artery for cardiopulmonary bypass: safeguards and pitfalls. Ann Thorac Surg 2003;75:931–4. [22] De Paulis R, Czerny M, Weltert L, Bavaria J, Borger MA, Carrel TP et al. Current trends in cannulation and neuroprotection during surgery of the aortic arch in Europe. Eur J Cardiothorac Surg 2015;47:917–23. [23] Ji S, Yang J, Ye X, Wang X. Brain protection by using innominate artery cannulation during aortic arch surgery. Ann Thorac Surg 2008;86:1030–2. [24] Misfeld M, Leontyev S, Borger MA, Gindensperger O, Lehmann S, Legare JF et al. What is the best strategy for brain protection in patients undergoing aortic arch surgery? A single center experience of 636 patients. Ann Thorac Surg 2012;93:1502–8.

APPENDIX. CONFERENCE DISCUSSION Scan to your mobile or go to http://www.oxfordjournals.org/page/6153/1 to search for the presentation on the EACTS library

Dr P. Bougioukakis (Bad Neustadt, Germany): Of course, we have some questions. It seems to us that the rate of the innominate cannulation in your cohort is high. How often did you use a different approach during the study duration, like the axillary artery? Second, did you retrospectively analyze the patients with neurological events regarding the images on account of the quality of the innominate artery? And if you did so, did you find any indications that the more peripheral approach for cannulation will be a better method for these patients? I keep it short and simple because we are running out of time. Thank you. Dr Preventza: With regards to your first question, during this study period, which was 2.5 years, these 263 patients represent approximately two-thirds of the cases. There is another one-third that represents the axillary cannulation. With regards to your second question, no, we did not go back to review the patients who suffered neurological outcome. But as part of our approach, we evaluate the pathology of the innominate artery. So if there is an atheroma at the base of the innominate artery, or if there is dissection of the innominate due to type I aortic dissection, or in case of redo sternotomy with an aorta that is in close proximity to the sternum, then we use alternative sites of cannulation and most preferably the right axillary cannulation. So, all these issues with innominate artery cannulation are eliminated as a consequence of surgical bias and judgment.