ORIGINAL ARTICLE. Abstract INTRODUCTION

Transcription

1 European Journal of Cardio-Thoracic Surgery 49 (2016) doi: /ejcts/ezv150 Advance Access publication 18 April 2015 ORIGINAL ARTICLE Cite this article as: Leontyev S, Tsagakis K, Pacini D, Di Bartolomeo R, Mohr FW, Weiss G et al. Impact of clinical factors and surgical techniques on early outcome of patients treated with frozen elephant trunk technique by using EVITA open stent-graft: results of a multicentre study. Eur J Cardiothorac Surg 2016;49: Impact of clinical factors and surgical techniques on early outcome of patients treated with frozen elephant trunk technique by using EVITA open stent-graft: results of a multicentre study Sergey Leontyev a, *, Konstantinos Tsagakis b,, Davide Pacini c, Roberto Di Bartolomeo c, Friedrich W. Mohr a, Gabriel Weiss d, Martin Grabenwoeger d, Jorge G. Mascaro e, Mauro Iafrancesco e, Ulrich F. Franke f, Nora Göbel f, Thanos Sioris g, Kazimierz Widenka h, Carlos A. Mestres i and Heinz Jakob b a Department of Cardiac Surgery, Leipzig Heart Center, University of Leipzig, Leipzig, Germany b Department of Thoracic and Cardiovascular Surgery, West German Heart Centre Essen, University Hospital Essen, Essen, Germany c Department of Cardiac Surgery, Sant Orsola-Malpighi Hospital, Bologna, Italy d Department of Cardiovascular Surgery, Hospital Hietzing, Vienna, Austria e Queen Elizabeth Hospital Birmingham, Birmingham, UK f Department of Cardiac and Vascular Surgery, Robert Bosch Hospital, Stuttgart, Germany g Tampere University Hospital Heart Center, Tampere, Finland h Szpital Wojewódzki N2, Oddział Kardiochirurgii, Rzeszów, Poland i Department of Cardio Vascular Surgery, Hospital Clinico, University of Barcelona, Barcelona, Spain * Corresponding author. Universität Leipzig, Herzzentrum, Klinik für Herzchirurgie, Strümpellstr. 39, Leipzig, Germany. Tel: ; fax: ; sergey.leontyev@medizin.uni-leipzig.de (S. Leontyev). Received 9 October 2014; received in revised form 22 February 2015; accepted 25 February 2015 Abstract OBJECTIVES: The treatment of patients with extensive thoracic aortic disease involving the arch and descending aorta is often performed, using the frozen elephant trunk (FET) technique. We retrospectively analysed early outcomes with this technique, using a prospective database. METHODS: A total of 509 patients (mean age: 61 ± 11 years) were registered between January 2005 and January 2014 in a multicentre database after FET surgery. Acute or chronic aortic dissection (AD) was the indication for surgery in 350 (68.8%) patients and degenerative or atherosclerotic aneurysm (DA) accounted for 159 (31.2%) patients. A logistic regression model was created to identify independent predictors of inhospital mortality and neurological complications. RESULTS: The average in-hospital mortality was 15.9% (n = 81) with 17.1% for AD patients and 13.2% for DA patients (P =0.2).Independentpredictors of in-hospital mortality were haemodynamic instability [odds ratio (OR): 2.7, P = 0.005], peripheral vascular disease (OR: 2.6, P =0.002), diabetes (OR: 2.1, P = 0.05) and selective cerebral perfusion time >60 min (OR: 2.2, P = 0.005). Patients under 60 years of age and the use of guide wire during FET implantation were protective for early survival. Stroke occurred in 7.7% (n = 39) of patients. Paraplegia or paraparesis occurred in 7.5% (n = 38) of patients. A distal landing zone lower than T10 was an independent predictor for spinal cord injury (OR: 2.3, P = 0.03). CONCLUSIONS: Techniques for faster arch replacement and controlled FET placement should be considered in order to reduce the early mortality and neurological complications after FET surgery. For distal aortic lesions, a two-staged approach is suggested, rather than the FET landing lower than T10. Keywords: Aortic arch surgery Frozen elephant trunk Risk factors INTRODUCTION The treatment of patients with extensive pathology of the thoracic aorta, involving the arch and the descending/thoracoabdominal aorta, is still associated with a significant operative and medium-term Presented at the 28th Annual Meeting of the European Association for Cardio- Thoracic Surgery, Milan, Italy, October The first two authors contributed equally to this study. mortality and a high incidence of neurological complications [1, 2]. During the last 10 years, new technical solutions like frozen elephant trunk (FET) technique have been developed and introduced into clinical practice [3 5]. The FET technique involves the use for extensive aortic arch pathology hybrid stent-graft prosthesis with integrate non-stented vascular prosthesis. We show here the early results of a large multicentre study, using the FET technique for patients with extensive thoracic aorta pathology. The Author Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

2 S. Leontyev et al. / European Journal of Cardio-Thoracic Surgery 661 The objectives of this study were (i) To evaluate the clinical characteristics and early results of aortic arch surgery using the FET technique. (ii) To evaluate the impact of intraoperative details and surgical technique on early outcome. (iii) To determine the predictors of early mortality and neurological (cerebral and spinal) outcomes after aortic arch surgery using the FET technique. PATIENTS AND METHODS A total of 509 consecutive patients underwent aortic arch/descending aorta replacement with FET procedure between January 2005 and January 2014 and were enrolled in the International E-vita Open Registry (IEOR) (Table 1). The selection of these patients was approved by the Institutional Review Board of the University of Essen, Germany, and individual consent for the study was waived. The patients received surgery in nine cardiac surgery centres (Leipzig, Germany; Essen, Germany; Bologna, Italy; Barcelona, Spain; Birmingham, UK; Tampere, Finland; Vienna, Austria; Stuttgart, Germany; Rzeszów, Poland). Table 1: Patients characteristics 509 (100) Age 64.1 ± 11.8 Female 152 (29.9) Arterial hypertension 420 (82.5) CAD 92 (18.1) Hyperlipidaemia 217 (42.6) Diabetes mellitus 45 (8.8) COPD 95 (18.7) Creatinine level >2 mg/dl 53 (10.4) PVD 80 (15.7) LV EF <40% 19 (3.7) LV EF >60% 262 (51.5) LV EF 40 60% 228 (44.6) Previous cerebrovascular accident 29 (5.7) Previous cardiovascular surgery 144 (28.3) Previous EVAR Descending aorta 17 (3.3) Abdominal aorta 6 (1.2) Aortic arch 6 (1.2) Descending and abdominal aorta 3 (0.6) Degenerative aortic aneurysm 159 (31.2) Thoracic aortic aneurysm (TAA) 105 (20.6) Thoraco-abdominal aortic aneurysm (TAAA) 54 (10.6) Aortic dissection 350 (68.8) Acute Type A aortic dissection (AAAD) 170 (33.4) Acute Type B aortic dissection (ABAD) 20 (3.9) Chronic Type A aortic dissection (CAAD) 110 (21.6) Chronic Type B aortic dissection (CBAD) 50 (9.8) Marfan syndrome 19 (3.7) Cystic media necrosis 42 (8.9) Descending aorta diameter (mm) 50 ± 18 Maximal aneurysm diameter (mm) 64 ± 15 Continuous variables are reported as mean and standard deviation; categorical variables are reported as percentages. Data are presented as number/% unless otherwise indicated. CAD: coronary artery disease; COPD: chronic obstructive pulmonary disease; PVD: peripheral vascular disease; LF EF: left ventricle ejection fraction; EVAR: endovascular aortic repair. Operative technique The E-vita Open stent-graft ( Jotec, Hechingen, Germany) was used in all patients. The surgical technique and indication were as previously described [4 6]. No standard surgical protocol for aortic arch surgery or stent-graft placement and implantation was applied. All operations were performed through medial sternotomy. Arterial cannulation was performed via the right subclavian artery (60.7%), the ascending aorta (20.4%) and the femoral artery (9.6%). Antegrade selective cerebral protection (ASCP) (10 15 ml/kg/min, C) was instituted in 99.4% of patients. Spinal fluid drainage was used intra- and postoperatively in 21.6% of patients. The guide was applied for stent graft placement in 69% of patients. The reimplantation of supra-aortic vessels was performed as island (60.9%) or separate (38.3%). The left subclavian artery was sacrificed in 6.5% of patients. All intraoperative data are presented in Table 2. Definitions The early mortality was defined as all-cause in-hospital mortality. Operations were considered emergent if performed within 24 h of hospital admission for cardiovascular instability, and urgent if performed during the same hospital admission. All patients with suspected neurological complications on physical examination underwent computerized tomography or magnetic resonance imaging. Spinal cord injury was defined as new-onset transient or permanent paraparesis or paraplegia. Neurological complications were defined as permanent neurological deficit (PND) for patients with stroke [1]. Statistical analysis All statistical analyses were performed using the SPSS 17.0 (Chicago, IL, USA). Normally distributed continuous variables were reported as mean and standard deviation and were compared by a two-tailed t-test. Continuous variables that were not normally distributed were reported as median and interquartile range and were compared using the Wilcoxon rank-sum test. Categorical variables were reported as frequencies and percentages and were analysed by χ 2 or Fisher s exact test as appropriate. Independent continuous variables were compared by unpaired Student s t-test for comparison of normally distributed data between two groups or Kruskal Wallis test for the comparison of more than two groups as appropriate. Continuous variables are expressed as mean ± SD (standard deviation), and categorical data as proportions throughout the manuscript. Candidate variables for multivariate analysis were selected on the basis of clinical relevance or significance of univariate association with P < 0.05 (Supplementary material). Multicollinearity was assessed using linear regression analysis, where a variance inflation factor >4.0 indicated potential intercorrelation among variables. If multicollinearity existed, correlated variables were either combined into a single variable, or 1 variable from a set of correlated variables was selected for the multivariate analysis. To predict in-hospital mortality or neurological complication, a parsimonious logistic regression model was developed by backward selection, and variables were retained if the final P-value was <0.05. The area under the receiver operating characteristic (ROC) curve was used to assess model (only for in-hospital mortality) discrimination. All statistical analyses were performed using the SPSS 17.0 (Chicago, IL, USA). AORTIC SURGERY

3 662 S. Leontyev et al. / European Journal of Cardio-Thoracic Surgery Table 2: Intraoperative data and operative procedures n = 509 Operation CPB time (min) 240 ± 73 ASCP time (min) 72 ± 31 Cross-clamp time (min) 140 ± 57 Visceral ischaemia (min) 67 ± 28 Temperature of HCA <20 ( C) 46 (9.0) ( C) 270 (53) ( C) 193 (37.9) Arterial cannulation Axillary artery 309 (60.7) Ascending aorta 104 (20.4) Femoral artery 49 (9.6) Other 47 (9.3) ASCP 3 (0.6) None 3 (0.6) Unilateral 13 (2.6) Bilateral 493 (96.9) CSF drainage 110 (21.6) E-vita Open diameter 30 ± 9 Proximal stent level Zone 0 4 (0.8) Zone 1 8 (1.6) Zone 2 70 (13.8) Zone (75.4) Zone 4 43 (8.4) Distal lading zone Unknown 12 (2.4) Th5 8 (1.6) Th6 Th7 125 (24.6) Th8 Th9 283 (55.6) Th10 81 (15.9) True lumen oversize in patients with AD 45 (12.8) Aortic oversize in patients with DA 23 (14.4%) Use of guide wire 351 (69.0) Subtotal aortic arch replacement 37 (7.3) Total aortic arch replacement 469 (92.1) Sacrifice of left subclavian artery 33 (6.5) Supra-aortic vessels reimplantation Island 310 (60.9) Separate 195 (38.3) Aortic valve intervention Aortic valve repair 41 (8.1) AVR biological 45 (8.8) AVR mechanical 26 (5.1) Aortic valve resuspension 41 (8.1) Aortic root intervention ARR biological 39 (7.7) ARR mechanical 32 (6.3) Aortic valve reconstruction (David) 20 (3.9) Aortic valve reconstruction (Yacoub) 28 (5.5) Continuous variables are reported as mean and standard deviation; categorical variables are reported as percentages. Data are presented as number/percentage unless otherwise indicated. CPB: cardiopulmonary bypass; ASCP: antegrade cerebral perfusion; HCA: hypothermic circulatory arrest; CSF: cerebrospinal fluid; AD: aortic dissection; DA: degenerative aneurysm; AVR: aortic valve replacement; ARR: aortic root replacement. RESULTS A total of 509 consecutive patients underwent total replacement of the aortic arch, using the FET technique during the study period. The mean age of patients was 64.1 ± 11.8 years and the Figure 1: Postoperative in-hospital mortality. AAAD: acute Type A aortic dissection; ABAD: acute Type B aortic dissection; CAAD: chronic Type A aortic dissection; CBAD: chronic Type B aortic dissection; TAA: thoracic aortic aneurysm; TAAA: thoraco-abdominal aortic aneurysm. majority (70.1%) were male. The maximum aneurysm diameter was 64 ± 15 mm. Acute or chronic aortic dissection (AD) was the indication for surgery in 350 (68.8%) of patients, while degenerative or atherosclerotic aneurysm (DA) accounted for 159 (31.2%) of patients (Table 1). Preoperative malperfusion syndrome and haemodynamic instability were observed in 51.2% (n = 87) and 28.2% (n = 48) of patients with acute Type A aortic dissection (AAAD). Operative details and postoperative complications All operative details and early postoperative results have been presented in Table 2. The majority of patients (53%, n = 270) were operated with a core body temperature between 20 and 25 C. A higher core body temperature of C was used in 37.9% (n = 193) of patients. Bilateral ASCP was applied for 96.6% of patients with a mean ASCP time of 72 ± 31 min. During the implantation of FET, the guide wire was used for 69% (n = 351) of patients. The mean prosthesis size of the E-vita hybrid open stent-graft was 30 ± 9 mm (range: mm). The distal prosthesis landing zone was T7 and lower in 26.1% (n = 133), between T8 and T9 in 55.6% (n = 283) and more than T10 in 15.9% (n = 81) of patients. In 75.4% (n = 384) of our patients, the proximal stent level was in Zone 3. Postoperative outcome The average in-hospital mortality for the entire cohort of patients was 15.9% (n = 81), and was not statistically significant between patients with AD and DA; 17.1% (n = 60) vs 13.2% (n = 21), respectively (P = 0.2). In-hospital mortality for patients with AAAD was 17.2% (n = 29) (Fig. 1). The most common complications were renal and pulmonary insufficiency, 26.1% (n = 133) and 26.5% (n = 135), respectively. Re-exploration for bleeding was performed in 17.7% of patients (Table 3). Multivariate predictive model of in-hospital mortality Preoperative haemodynamic instability [odds ratio (OR): 2.6, P = 0.005, 95% confidence interval (CI): ), peripheral vascular disease (OR: 2.6, P = 0.002, 95% CI: ) and diabetes (OR: 2.1, P = 0.048, 95% CI: ) were found to be independent predictors of in-hospital mortality. In addition, the early

4 S. Leontyev et al. / European Journal of Cardio-Thoracic Surgery 663 Table 3: Postoperative outcome n = 509 Re-exploration 90 (17.7) Pulmonary insufficiency 135 (26.5) Dialysis 133 (26.1) Dialysis permanent 21 (4.1) Dialysis temporary 112 (22) Hepatic failure 18 (3.5) Bowel ischaemia 26 (5.1) Peripheral ischaemia 25 (4.9) Sepsis 43 (8.4) Stroke 41 (7.7) Spinal cord injury 38 (7.5) Paraparesis permanent 5 (1.0) Paraparesis regressive 14 (2.8) Paraplegia permanent 15 (2.9) Paraplegia regressive 4 (0.8) In-hospital mortality 81 (15.9) Figure 2: Postoperative incidence of stroke. AAAD: acute Type A aortic dissection; ABAD: acute Type B aortic dissection; CAAD: chronic Type A aortic dissection; CBAD: chronic Type B aortic dissection; TAA: thoracic aortic aneurysm; TAAA: thoraco-abdominal aortic aneurysm. Continuous variables are reported as mean and standard deviation; categorical variables are reported as percentages. Data are presented as number/percentage unless otherwise indicated. mortality was also significantly influenced by the duration of ASCP >60 min (OR: 2.2, P = 0.005, 95% CI: ). Conversely, age under 60 years (OR: 0.5, P = 0.016, 95% CI: ) and the use of guide wire (OR: 0.5, P = 0.04, 95% CI: ) during FET implantation were protective for early survival. Model discrimination was measured as ROC 70 (95% CI: 63 76), demonstrating acceptable predictive accuracy. The Hosmer Lemeshow goodness of fit test for the logistic regression model was not significant (χ 2 = 5.0; df = 8; P = 0.7). Multivariate predictive model of stroke and spinal cord injury The incidence of PND was 7.7% (Fig. 2). The independent predictors for PND were previous cerebrovascular accident (OR: 3.7, P = 0.01, 95% CI: ) and arterial cannulation site at the ascending aorta (OR: 2.0, P = 0.048, 95% CI: ). The AAAD was associated with a significantly higher rate of postoperative neurological deficit compared with all other patients, 11.2% (n = 19) vs 5.9% (n = 20) (P = 0.035), respectively. But in AAAD patients without preoperative cerebral malperfusion syndrome, the incidence of stroke was comparable with all other patients; 7.2 vs 5.9%, respectively (P = 0.6). The overall rate of spinal cord injury was 7.5% (n = 38) (Fig. 3), comprising 19 patients with paraparesis (3.75%) and 19 patients with paraplegia (3.75%). Paraparesis symptoms were reversed in a majority of patients (n = 14, 2.8%). Conversely, symptoms of paraplegia were permanent in a majority of patients (n = 15, 2.9%). Multivariate analysis identified a distal landing zone lower than T10 as an independent predictor of permanent spinal cord injury (OR: 2.3, 95% CI: , P = 0.03). DISCUSSION In the current endovascular era, new hybrid solutions have been developed to treat patients with complex thoracic aorta Figure 3: Postoperative incidence of spinal cord injury. AAAD: acute Type A aortic dissection; ABAD: acute Type B aortic dissection; CAAD: Chronic Type A aortic dissection; CBAD: Chronic Type B aortic dissection; TAA: Thoracic aortic aneurysm; TAAA: thoraco-abdominal aortic aneurysm. pathology. The use of hybrid stent graft or frozen elephant trunk offers the opportunity to exclude the descending aortic pathology [7] or enables the preparation of an easier access for a second stage surgical or endovascular thoracoabdominal procedure [8]. The present study, based on IEOR, summarizes the 9-year multicentre international experience, using FET in high-risk patients with extensive aortic pathology, showed the early outcome in these patients. The previous publications from IEOR have reported that the in-hospital mortality for patients undergoing FET surgery was 7 15%, depending on the aortic pathology, [3 7, 9]. Single-centre studies published in the last 5 years, have shown in-hospital mortalities ranging from 7.8 to 17.2% [10 13]. The in-hospital mortality observed in the present study was 15.9% considering all patients, which is comparable with other case series reported in literature. The observed trend in higher mortality must be interpreted with caution, given the fact that 69% of patients suffered from any form of AD and the mortality rate of these patients, although obviously higher, was not significant when compared with patients with DA (Fig. 1). The major advantage of the present study is that it gives an overview of the outcome of FET in patients with the most common and important aortic disease. The in-hospital mortality was comparable in patients with different aortic diseases, and logistic regression analysis was unable to differentiate independent predictors AORTIC SURGERY

5 664 S. Leontyev et al. / European Journal of Cardio-Thoracic Surgery for in-hospital mortality among these (Fig. 1). Despite this, early mortality was influenced by the complicated form of AAAD. Multivariate analysis identified preoperative haemodynamic instability as an independent predictor for early death and >80% of haemodynamically unstable patients suffered from AAAD. Studies have shown that in-hospital mortality for patients with AAAD treated by FET ranged between 13 and 18% [3, 4, 12]. Duration of ASCP more than 60 min was also identified as an independent predictor for death, which showed an indirect influence of complex aortic pathology on early outcome. Another study from Hannover found mega-aortic syndrome to be an independent predictor for in-hospital mortality [14]. The advantage of the current study is its multicentre design, which enabled the comparison of different technical details in order to find the best possible surgical solutions. For example, one of these solutions was the use of a guide wire during the implantation of the frozen elephant stent graft. Pacini et al. [5] showed a negative influence of stent graft placement without guide wire for patients with AAAD using univariate analysis. We are able to show a significant survival benefit, using a guide wire during stent graft implantation for all patients. In our opinion, the use of a guide wire enabled exact positioning and safe placement of the stent graft. In addition, using a guide wire could prevent aortic injury due to the tip of the delivery system, especially in patients with AAAD, and would also help in additional endovascular procedures. One of the disadvantages of using a guide wire is wire-induced aortic injury, which can be prevented using a pig tail catheter in patients with very fragile aortas. In our opinion, the use of a guide wire is part of a modern hybrid aortic surgery concept, which also includes intraoperative angioscopy and/or angiography. The intraoperative angioscopy enabled the visualization of downstream aortic disease, identification of the distal landing zone and also control of stent-graft deployment followed by balloon dilatation, if required [15]. The intraoperative angiogram is more meaningful in acutely ill patients, such as in patients with AAAD, for the diagnosis of coronary status and downstream malperfusion [16]. Neurological complications: permanent neurological deficit and spinal cord injury The early postoperative outcome after extend aortic arch surgery is always significantly influenced by the presence of postoperative neurological complications. Generally, the prevalence of PND or stroke after conventional aortic arch surgery is % [2, 17, 18], while it is 5 13% in patients with FET [4, 6, 12 14]. Postoperative stroke is related with prolonged hospital and intensive care unit stays, increased ventilation time, as well as reduced long-term survival and quality of life [1, 17]. The factors that were shown to significantly influence neurological outcome were age, history of previous central neurological event, renal insufficiency [2], urgent timing of surgery [2], AAAD, prolonged operation and circulatory arrest time [2]. The incidence of neurological complications in the current study was 7.7%. The postoperative PND rate was nearly twice as high among patients who had presented with AAAD (11.2 vs 5.9%, P = 0.035), but AAAD was not identified as an independent predictor for permanent neurological deficit. Multivariate analysis showed that neurological outcome was significantly influenced by previous history of neurological accident and cannulation of the ascending aorta. Direct aortic cannulation through the distal ascending aorta, the brachiocephalic trunk or curvature of the transverse arch compared with vascular, axillar or femoral access is still a subject of debate [17]. Femoral arterial cannulation had a higher in-hospital mortality and worse neurological outcome [2], possibly because of retrograde cerebral embolization. The benefit of axillary cannulation for improvement of neurological outcome, in patients with extend aortic arch surgery has been previously demonstrated [13, 17]. The advantages of axillary cannulation when compared with all other techniques are secure and safe antegrade cerebral perfusion and optimal deairing of the aortic arch branches during reinstitution of cardiopulmonary bypass flow. The ascending aortic cannulation in our patient cohort was associated with a 2-fold higher incidence of stroke (15.1 vs 7.1%, P = 0.009) in all patients and was independent of the type of aortic disease. The 2-fold increase in stroke rate was observed in patients with and without AAAD and when ascending arterial cannulation was performed. The possible explanation of this could be antegrade cerebral embolization during the direct cannulation of supra-aortic vessel and insufficient deairing of the supra-aortic vessel. Neurological complications: spinal cord injury Spinal cord injury is a relatively new complication in open aortic arch surgery. The reported incidence of this catastrophic complication in patients who underwent conventional elephant trunk implantation ranged between 0.4 and 2.8% [19, 20]. Conversely, previously reported data from IEOR revealed a higher incidence of spinal cord injury between 4 and 9% [3, 5, 6], depending on the different types of aortic pathology. The Hannover group has reported the incidence of spinal cord injury to be between 1 and 4% [12, 14]. The groups from Bologna and Essen described the rate of spinal cord injury as 9 and 5%, respectively [11, 21]. On the other hand, some single-centre studies have reported occurrences of spinal cord injury between 21 and 24% after FET implantation [13, 22 24]. In the current study, permanent or regressive paraparesis and paraplegia occurred in 7.5% of patients. In a majority of cases, the paraparesis symptoms were regressive as opposed to paraplegia symptoms. The potential reasons of spinal cord damage during the operation are duration of circulatory arrest, core body temperature during the circulatory arrest, coverage of segmental arteries by stent graft implantation, embolism and postoperative haemodynamic management. Duration of circulatory arrest >40 min at a core body temperature higher than 28 C was previously described as a factor that significantly increased the occurrence of spinal cord injury [13]. On the other hand, Pacini et al. [5] failed to find a significant correlation between spinal cord injury and time of antegrade cerebral perfusion, despite distal arrest times of 75 ± 22 min. Likewise, in the current study, neither body core temperature nor circulatory arrest time was found to influence neurological outcome, especially on spinal cord injury. The coverage of intercostal arteries is an anticipated consequence of FET and the number of occluded segmental arteries is directly related to the distal landing zone of the stent graft. Previous publications considering small series of patients have reported a significantly higher incidence of spinal cord injury in patients with the distal landing zone between T7 and T8 [23, 24]. Flores et al. have observed the combination of a distal landing zone of T7 or lower and a history of previous abdominal aortic aneurysm repair as a strong predictor for spinal cord injury [24].

6 S. Leontyev et al. / European Journal of Cardio-Thoracic Surgery 665 Conversely in large patient series, the distal landing zone was not found to influence the prevalence of spinal cord injury [5, 13]. In the present study, we were able to identify the distal landing zone of T10 and more as an independent predictor for the occurrence of spinal cord injury. An understanding of the spinal cord blood flow could help to determine an optimal solution for spinal cord protection. Experimental studies have shown that blood flows into the spinal cord arteries from segmental/intersegmental arteries, subclavian and iliac arteries [25]. Hybrid stent graft implantation influenced all these inflow pathways: segmental artery ( permanent) by stent graft implantation and subclavian and iliac arteries (temporary) by administration of circulatory arrest. This mechanism could be a reason for increased incidence of spinal cord damage in patients with FET when compared with patients with conventional elephant trunk implantation. The protection of spinal cord, in order to prevent such complications, could be enabled by the use of deeper hypothermia during circulatory arrest, perfusion of left subclavian artery, drainage of spinal fluid and the use of stent-graft matching the patients anatomy. In spite of a large number of patients, we failed to find any advantage of using left subclavian artery perfusion or drainage of spinal fluid on early neurological outcome. Spinal cord injury was observed in only one of 38 patients (3%) with sacrificed left subclavian artery. The intercostal arteries could be compromised by FET implantation owing to the presence of a dissection membrane in patients with AD. Pacini et al. [5] identified the diameter of the false lumen in patients with chronic AD as an independent predictor for spinal cord injury. In the current study, we failed to find a significant influence of the type of aortic pathology (acute/chronic AD or aortic aneurysm) on the occurrence of postoperative spinal cord injury. In our opinion, the occurrence of spinal cord injury is multifactorial and is mostly influenced by a combination of acute ischaemic injury during distal circulatory arrest in mild-to-moderate hypothermia, and postoperative haemodynamic fluctuations after extensive segmental artery occlusion. STUDY LIMITATIONS The current study has several limitations. First, this is a retrospective and non-randomized study based on IEOR. Secondly, because of the multicentre study design, different surgical protocols were followed, which were not standardized across the different centres. In conclusion, the FET operation can be performed in patients with extensive aortic arch pathology with acceptable mortality and neurological complications. ASCP times of >60 min significantly influenced the in-hospital mortality. The standardized use of guide wire during stent graft implantation is recommended. The distal landing zone at the level of T10 has been associated with a significant increase in spinal cord injury after FET implantation. SUPPLEMENTARY MATERIAL Supplementary material is available at EJCTS online. Conflict of interest: Heinz Jacob is consultant to JOTEC GmbH. REFERENCES [1] 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: [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: [3] Jakob H, Tsagakis K, Pacini D, Di Bartolomeo R, Mestres C, Mohr F et al. The International E-vita Open Registry: data sets of 274 patients. J Cardiovasc Surg 2011;52: [4] Tsagakis K, Pacini D, Di Bartolomeo R, Gorlitzer M, Weiss G, Grabenwoger M et al. Multicenter early experience with extended aortic repair in acute aortic dissection: is simultaneous descending stent grafting justified? J Thorac Cardiovasc Surg 2010;140:S [5] Pacini D, Tsagakis K, Jakob H, Mestres CA, Armaro A, Weiss G et al. The frozen elephant trunk for the treatment of chronic dissection of the thoracic aorta: a multicenter experience. Ann Thorac Surg 2011;92: [6] Mestres CA, Tsagakis K, Pacini D, Di Bartolomeo R, Grabenwoger M, Borger M et al. One-stage repair in complex multisegmental thoracic aneurysmal disease: results of a multicentre study. Eur J Cardiothorac Surg 2013;44:e [7] Tsagakis K, Kamler M, Kuehl H, Kowalczyk W, Tossios P, Thielmann M et al. Avoidance of proximal endoleak using a hybrid stent graft in arch replacement and descending aorta stenting. Ann Thorac Surg 2009;88: [8] Folkmann S, Weiss G, Pisarik H, Czerny M, Grabenwoger M. Thoracoabdominal aortic aneurysm repair after frozen elephant trunk procedure. Eur J Cardiothorac Surg 2014;47: [9] Weiss G, Tsagakis K, Jakob H, Di Bartolomeo R, Pacini D, Barberio G et al. The frozen elephant trunk technique for the treatment of complicated type B aortic dissection with involvement of the aortic arch: multicentre early experience. Eur J Cardiothorac Surg 2014;47: [10] Ma WG, Zheng J, Dong SB, Lu W, Sun K, Qi RD et al. Sun s procedure of total arch replacement using a tetrafurcated graft with stented elephant trunk implantation: analysis of early outcome in 398 patients with acute type A aortic dissection. Ann Cardiothorac Surg 2013;2: [11] Di Eusanio M, Pantaleo A, Murana G, Pellicciari G, Castrovinci S, Berretta P et al. Frozen elephant trunk surgery-the Bologna s experience. Ann Cardiothorac Surg 2013;2: [12] Shrestha M, Fleissner F, Ius F, Koigeldiyev N, Kaufeld T, Beckmann E et al. Total aortic arch replacement with frozen elephant trunk in acute type A aortic dissections: are we pushing the limits too far?. Eur J Cardiothorac Surg 2014;47: [13] Leontyev S, Borger MA, Etz CD, Moz M, Seeburger J, Bakhtiary F et al. Experience with the conventional and frozen elephant trunk techniques: a single-centre study. Eur J Cardiothorac Surg 2013;44: [14] Ius F, Fleissner F, Pichlmaier M, Karck M, Martens A, Haverich A et al. Total aortic arch replacement with the frozen elephant trunk technique: 10-year follow-up single-centre experience. Eur J Cardiothorac Surg 2013; 44: [15] Tsagakis K. Angioscopy as a supplement to frozen elephant trunk treatment. Ann Cardiothorac Surg 2013;2: [16] Tsagakis K, Konorza T, Dohle DS, Kottenberg E, Buck T, Thielmann M et al. Hybrid operating room concept for combined diagnostics, intervention and surgery in acute type A dissection. Eur J Cardiothorac Surg 2013;43: [17] Immer FF, Moser B, Krahenbuhl ES, Englberger L, Stalder M, Eckstein FS et al. Arterial access through the right subclavian artery in surgery of the aortic arch improves neurologic outcome and mid-term quality of life. Ann Thorac Surg 2008;85: [18] LeMaire SA, Carter SA, Coselli JS. The elephant trunk technique for staged repair of complex aneurysms of the entire thoracic aorta. Ann Thorac Surg 2006;81: [19] Safi HJ, Miller CC III, Estrera AL, Villa MA, Goodrick JS, Porat E et al. Optimization of aortic arch replacement: two-stage approach. Ann Thorac Surg 2007;83:S [20] Etz CD, Plestis KA, Kari FA, Luehr M, Bodian CA, Spielvogel D et al. Staged repair of thoracic and thoracoabdominal aortic aneurysms using the elephant trunk technique: a consecutive series of 215 first stage and 120 complete repairs. Eur J Cardiothorac Surg 2008;34: AORTIC SURGERY

7 666 S. Leontyev et al. / European Journal of Cardio-Thoracic Surgery [21] Tsagakis K, Dohle D, Benedik J, Lieder H, Jakob H. Overall Essen s experience with the E-vita open hybrid stent graft system and evolution of the surgical technique. Ann Cardiothorac Surg 2013;2: [22] Miyairi T, Kotsuka Y, Ezure M, Ono M, Morota T, Kubota H et al. Open stent-grafting for aortic arch aneurysm is associated with increased risk of paraplegia. Ann Thorac Surg 2002;74:83 9. [23] Mizuno T, Toyama M, Tabuchi N, Wu H, Sunamori M. Stented elephant trunk procedure combined with ascending aorta and arch replacement for acute type A aortic dissection. Eur J Cardiothorac Surg 2002;22: [24] Flores J, Kunihara T, Shiiya N, Yoshimoto K, Matsuzaki K, Yasuda K. Extensive deployment of the stented elephant trunk is associated with an increased risk of spinal cord injury. J Thorac Cardiovasc Surg 2006;131: [25] Etz CD, Kari FA, Mueller CS, Silovitz D, Brenner RM, Lin HM et al. The collateral network concept: a reassessment of the anatomy of spinal cord perfusion. J Thoracic Cardiovasc Surg 2011;141: European Journal of Cardio-Thoracic Surgery 49 (2016) doi: /ejcts/ezv248 Advance Access publication 18 September 2015 EDITORIAL COMMENT Cite this article as: Shrestha M. Re: Impact of clinical factors and surgical techniques on early outcome of patients treated with frozen elephant trunk technique by using EVITA open stent-graft: results of a multicentre study. Eur J Cardiothorac Surg 2016;49: Re: Impact of clinical factors and surgical techniques on early outcome of patients treated with frozen elephant trunk technique by using EVITA open stent-graft: results of a multicentre study Malakh Shrestha* Department of Cardio-thoracic, Transplantation and Vascular Surgery, Hannover Medical School, Hannover, Germany * Corresponding author. Department of Cardiothoracic, Transplantation and Vascular Surgery, Hannover Medical School, Carl-Neuberg Strasse 1, Hannover, Germany. Tel: ; fax: ; shrestha.malakh.lal@mh-hannover.de (M. Shrestha). Keywords: Complex aortic disease Aortic arch surgery Frozen elephant trunk Leontyev et al. have retrospectively evaluated the early outcomes following total aortic arch replacement with frozen elephant trunk (FET) in complex thoracic aortic disease using the EVITA prosthesis [1]. Five hundred and nine patients underwent complete aortic arch replacements for complex pathologies [degenerative or atherosclerotic aortic aneurysms (DAs) or aortic dissections (ADs)] of the aortic arch and descending aorta in several European centres. The in-hospital mortality rate was 15.9% with 17.1% for AD and 13.2% for DA patients. Peri-operative stroke rate was 7.7%. Paraplegia or paraparesis occurred in 7.5% of patients. The authors made the following conclusions: (i) Techniques for faster arch replacement and controlled FET placement should be considered. (ii) A two-stage approach is suggested rather than an FET with a landing zone lower than T10. The authors have correctly stated that treatment of complex aortic arch disease is still associated with significant peri-operative risks. Over the years, various techniques have been proposed to treat complex aortic disease. Deep hypothermic circulatory arrest along with various cerebral perfusion techniques have been proposed for cerebral protection. Borst et al. introduced the Elephant trunk technique (ET) to simplify the second stage of the two-stage procedure to treat disease of the aortic arch extending into the descending aorta [2]. Dake et al. introduced endovascular stent grafts to treat aortic pathology [3]. A combination of the above two techniques (endovascular stent graft and classical ET) resulted in the frozen elephant trunk technique (FET). Initially, various home-made FET prostheses combining conventionally available Dacron grafts and thoracic endovascular aortic repair (TEVAR) grafts were used until a prefabricated hybrid prosthesis became available [4, 5]. However, we need to understand that the FET technique is just one of the tools available to the surgeon to treat complex aortic arch disease. The peri-operative risks after surgery of the aortic arch are mainly procedure dependent and with or without the use of the FET technique. Although the present study tries to explain to some extent the peri-operative risks after the FET technique, we need further information to define optimal indications and reduce risks after complex aortic arch surgery. In the present study, every participating centre had its own surgical protocol (e.g. arterial cannulation, stent graft placement, cerebral protection technique etc.). This means that the reason for peri-operative stroke cannot be evaluated objectively. Ideally, we need a multicentre study with a uniform surgical protocol to better evaluate the results. The FET technique has been used in different aortic pathologies: acute and chronic ADs (Stanford Type A and B, DeBakey I) and atherosclerotic arch and descending aneurysms. It is of utmost importance to understand that these are different disease entities and thus should not be grouped together when analysing the data. SurgeryforacuteADisdoneinemergencysettingsandassuch adds an extra risk to the already high-risk setting of the aortic arch