European Journal of Cardio-Thoracic Surgery Advance Access published May 14, 2014
Total aortic arch replacement with frozen elephant trunk in acute type A aortic dissections: are we pushing the limits too far?† Malakh Shrestha*‡, Felix Fleissner‡, Fabio Ius, Nurbol Koigeldiyev, Tim Kaufeld, Erik Beckmann, Andreas Martens and Axel Haverich Department of Cardiothoracic, 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, 30625 Hannover, Germany. Tel: +49-511-5326238; fax: +49- 511-5328156; e-mail: [email protected] (M. Shrestha). Received 13 November 2013; received in revised form 12 March 2014; accepted 28 March 2014
Abstract OBJECTIVES: Acute type A aortic dissection (AADA) is a surgical emergency. In patients with aortic arch and descending aorta (DeBakey type I) involvement, performing a total aortic arch replacement with frozen elephant trunk (FET) for supposedly better long-term results is controversial. We hereby present our results. METHODS: From February 2004 to August 2013, 52 patients with acute aortic dissection DeBakey type I received a FET procedure at our centre (43 males, age 59.21 ± 11.67 years). All patients had an intimal tear in the aortic arch and/or proximal descending aorta. Concomitant procedures were Bentall (n = 15) and aortic valve repair (n = 30). RESULTS: Cardiopulmonary bypass (CPB), X-clamp and cardiac arrest times were 262 ± 64, 159 ± 45 and 55 ± 24 min, respectively. The 30day mortality rate was 13% (n = 7). Stroke and re-thoracotomy for bleeding were 12% (n = 6) and 23% (n = 12), respectively. Postoperative recurrent nerve palsy and spinal cord injury rates were 10% (5 of 52) and 4% (2 of 52), respectively. Follow-up was 40 ± 24 months. During follow-up, no patient died and no patient required a reoperation for the aortic arch. CONCLUSIONS: Our results with FET in AADA show acceptable results. Total aortic arch replacement with an FET in AADA patients does demand high technical skills. In spite of this, we believe FET improves long-term outcomes in cases of AADA with intima tear or re-entry in the aortic arch or the descending aorta (DeBakey type I). Modern grafts with four side branches as well as sewing collars for the distal anastomosis have helped to further ‘simplify’ the FET implantation. However, such a strategy is not appropriate in all AADA cases; it should be implemented only in experienced centres and only if absolutely necessary. Keywords: Frozen elephant trunk • Aortic surgery • Aneurysm • Acute dissection
INTRODUCTION Acute type A aortic dissection is an absolute emergency with extremely dismal outcomes without surgical therapy. There is still no consensus regarding the ‘ideal’ surgical therapy for the treatment of acute type A aortic dissection involving the aortic arch and the descending aorta (DeBakey type I). To simplify the two-stage repair of the combined disease of the aortic arch and the proximal descending aorta, ‘Elephant trunk (ET) technique’ was introduced by Borst et al. at our centre in March 1982 [1]. Over the years, the indication for this technique has been expanded to include aortic dissections, both acute and chronic [2, 3]. Endovascular stent-graft technology was introduced by Dake et al. to treat aortic pathology in 1998 [4]. However, a totally endovascular treatment of the aortic arch pathology is still difficult because of the supra-aortic vessels.
A combination of the above two techniques resulted in the so-called frozen elephant technique [4, 5]. It therefore represents an evolution of the classical elephant procedure. Initially, the ‘frozen elephant trunk’ (FET) technique was used for the treatment of aortic aneurysms. Later, the indications for FET were expanded to patients with dissection involving the arch and the descending aorta (DeBakey type I) due to false lumen thrombosis around the stent graft, since a patent false lumen had been identified as a risk factor for aortic dilatation and late operation. The purpose of this study was to assess the results of the frozen ET technique for the treatment of acute aortic dissection (DeBakey type I).
MATERIALS AND METHODS Patients
†
Presented at the 27th Annual Meeting of the European Association for CardioThoracic Surgery, Vienna, Austria, 5–9 October 2013. ‡ Both authors contributed equally to this work.
At our centre, the first FET implantation was done in August 2001 for thoracic aortic aneurysm. With more experience, we expanded
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
AORTIC SURGERY
ORIGINAL ARTICLE
European Journal of Cardio-Thoracic Surgery (2014) 1–6 doi:10.1093/ejcts/ezu185
M. Shrestha et al. / European Journal of Cardio-Thoracic Surgery
2
the indication for FET to chronic type A aortic dissection, acute type B aortic dissection and finally to acute type A aortic dissection (AADA, February 2004). The ethical committee of our institution gave approval for this retrospective study. Between February 2004 and August 2013, 52 consecutive patients underwent complete aortic arch replacement and stenting of the proximal descending aorta with the FET technique in a setting of acute DeBakey type I dissection. Patient records were retrospectively reviewed. Clinical follow-up ended in August 2013 and was 100% complete. This was done by visiting the patients at our outpatient clinic or contacting them at home or their physician by phone. Preoperative demographic patient characteristics are reported in Table 1. All patients had an intimal tear in the distal aortic arch and/or proximal descending aorta. Overall, this study population showed a high operative risk (all acute aortic dissection, n = 52; all emergent procedures, n = 52). The patients were predominantly male (43 of 52, 83%) and middle aged (age average 59.21 (±11.67, range from 41 to 79 years). Eleven patients (21%) showed acute neurological deficits, 4 (8%) displayed a cardiac tamponade and 13 (25%) showed signs of acute visceral or peripheral malperfusion. Because of the long period of the study, the patient cohort was divided into three groups for the sake of objective analysis. Especially for the last group, we modified our techniques in accordance with lessons learnt from the recently published 10-year experience (2001–11) with FET.
Group I (2004–10). During this period, the FET technique was performed using the Custom-made ‘Chavan–Haverich’ hybrid graft as well as the Jotec E-vita. A total of 25 patients (male: n = 22, age: 62.79 ± 9.66 years) were operated on. Twelve patients received Chavan–Haverich and 13 Jotec E-Vita, respectively. Forty percent (10 of 25) of these patients underwent concomitant aortic valve sparing root procedures.
Group II (2010–11). During this period, the FET technique was performed using the Thoraflex hybrid graft as part of the CE certification trial. During the first phase of the trial, only the hybrid graft without the delivery system was available. A total of 16 patients (male: n = 12, age: 56.81 ± 13.96 years) received this Thoraflex hybrid graft. Sixty-nine percent (11 of 16) of these patients underwent concomitant aortic valve sparing root procedures.
Group III (Since 2013). During this period, the FET technique was performed using the Thoraflex hybrid graft as part of the post CE certification registry. We also modified our surgical technique for this trial. A total of 11 patients (male: n = 9, age: 54.55 ± 10.08 years) were operated on [6]. Almost 82% (9 of 11) of these patients underwent concomitant aortic valve sparing root procedures.
Surgical technique Because of the long period, obviously, surgical techniques have evolved through the years.
Early years (2004–10). Custom-made ‘Haverich–Chavan’ (Curative GmbH, Dresden, Germany) hybrid grafts were implanted in 12 patients. This early custom-made hybrid graft was very stiff and could not be completely collapsed. The hybrid graft was delivered into the descending aorta with the help of a special forceps. This sometimes resulted in technical difficulty in placing the stented portion into the descending aorta, especially in smaller sized dissected descending aorta. We changed to the Jotec E-vita hybrid graft after its availability. Thirteen acute dissection patients received E-vita grafts. Although the E-vita graft was a considerable advance over previous custom-made grafts, in our experience, there were still some significant technical shortcomings. Therefore, together with Vascutek Terumo, we developed a novel FET hybrid graft that would be technically simple to implant and also incorporate innovative features: (i) The Gel-coat on the graft for classical haemostatic properties. (ii) Different sizes of the un-stented and stented parts to ensure that a second graft for the aortic arch replacement would not be necessary in patients with large descending aorta. (iii) A sewing collar to simplify the distal anastomosis. (iv) Branched grafts for the aortic arch to facilitate active haemostasis. (v) Novel delivery system for technically simple and rapid deployment.
Description of the Thoraflex hybrid graft. The Thoraflex hybrid graft (Vascutek, Terumo) consists of a 4-branched arch graft with a stent graft at the distal end. The proximal part is a conventional gel-coated woven polyester graft. The stented section of the graft is a self-expanding endo-prosthesis constructed of
Table 1: Preoperative characteristics Variables
Total (n = 52)
Group 1 (n = 25)
Group 2 (n = 16)
Group 3 (n = 11)
P-value
Total Male Age (years) >60 years of age Marfan Neurological deficits Shock Cardiac tamponade Malperfusion
52 (100%) 43 (83%) 59 ± 12 24 (46%) 1 (2%) 7 (13%) 1 (2%) 4 (8%) 13 (25%)
25 (100%) 22 (88%) 62.79 ± 9.66 14 (56%) 1 (4%) 5 (20%) 1 (4%) 2 (8%) 8 (32%)
16 (100%) 12 (74%) 56.81 ± 13.96 7 (44%) 0 4 (5%) 0 1 (6%) 4 (5%)
11 (100%) 9 (82%) 54.55 ± 10.08 3 (27%) 0 3 (27%) 0 1 (9%) 1 (9%)
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
Continuous data summarized as mean ± SD; categorical and binary data summarized as frequency (percentage). SD: standard deviation.
thin-wall polyester and nitinol ring stents, which are attached to fabric with braided polyester sutures. The grafts are available in different sizes (26–40 mm in diameter for the stented portion). The length of the stented part is either 100 or 150 mm. Another unique feature of this graft is that the proximal un-stented and distal stented parts are available in different sizes (e.g. un-stented part 32 mm in diameter and stented part with 40 mm diameter).
Lessons learned. We recently published our 30-year experience (1982–2012) with conventional elephant trunk technique as well as our 10-year experience (2001–11) with FET. We identified several parameters that still needed to be improved and initiated several significant changes in perioperative management. (i) Reduction in perioperative stroke rate: Use of carbon dioxide (CO2) sufflation in the operative field to prevent air embolism. (ii) Myocardial protection (‘Beating Heart’ aortic arch replacement): To minimize the myocardial ischaemia time, we initiated the ‘beating heart’ aortic replacement technique. As soon as the aortic root or other cardiac procedure was performed, the heart was perfused antegradely and the aortic arch placed on ‘beating heart’. (iii) Spinal cord protection: To minimize spinal cord ischaemia, cerebrospinal fluid (CSF) drainage was performed so that spinal pressure remained below 12 mmHg. The CSF drainage catheter was implanted just before the operation or immediately postoperatively in the intensive care unit (ICU). (iv) Shorten the circulatory arrest and cardiopulmonary bypass (CPB) times: To minimize coagulopathy, perfusion of the body was restarted immediately after the distal anastomosis and ‘re-warming’ begun. (v) Haemostasis: Time was taken to perform each anastomosis carefully to have ‘perfect’ haemostatic anastomoses.
Surgical technique since 2013. We have standardized the surgical technique as far as possible. After a standard median sternotomy, extracorporeal circulation is initiated with cannulation of the aorta and the right atrium. This technique of cannulation of the ascending aorta is performed even in AADA, as has been published by our group [7]. The left side of the heart is vented through the right superior pulmonary vein. CO2 is blown in to the surgical field. During this period, the patient is cooled to a nasopharyngeal temperature of 26°C, the aortic root/ascending aortic procedure or other concomitant procedures (e.g. CABG) are performed if necessary. Blood cardioplegia has been our preferred method of myocardial protection. Cardioplegia is repeated approximately every 30 min. After the aortic root or other cardiac concomitant procedure, de-airing of the myocardium is performed. Antegrade perfusion of the myocardium is initiated through the ascending aorta/ aortic prosthesis at the rate of 100–200 ml/min (via Cardioplegia cannula). Care is taken that the pressure in the aortic root does not exceed a mean pressure of 80–90 mmHg. Cerebral protection by moderate hypothermic circulatory arrest with selective antegrade cerebral perfusion (SACP) has been our preferred technique. After reaching the desired temperature (26°C), the systemic circulation is arrested and the aorta opened. With the patient in the Trendelenburg position, catheters (Medtronic DPL, USA) are introduced into the left carotid artery and the innominate artery for SACP. The left sub-clavian artery is clamped or occluded with a Fogarty catheter (Baxter, Deerfield, IL, USA), thus avoiding a steal
3
phenomenon as well as to prevent blood flowing into the operative field. Cerebral perfusion is initiated at a rate of 10 ml/kg/min. The blood temperature of SACP is 24°C. The aorta is transected either between the left common carotid artery and the left sub-clavian artery or distal to the left subclavian artery. The self-expanding FET part of the Thoraflex hybrid prosthesis is placed through the opened aortic arch and deployed in the descending aorta. A sewing collar between the graft segments simplifies the ‘distal anastomosis’. In patients where the distal aortic arch is not aneurysmatic, the distal anastomosis is done between the left carotid and the left sub-clavian arteries. After the distal anastomosis is completed, the left sub-clavian artery is anastomosed to the third branch of the arch graft. The perfusion to the lower part of the body and the sub-clavian artery is then restarted via the fourth branch of the graft that is connected to the inner curvature of the prosthesis vis-à-vis to the supra-aortic vessels. Rewarming of the patient is started leaving the temperature of the SACP at 28–30°C. The proximal end of the graft is anastomosed, either to the native ascending aorta or the ascending aortic graft. The previously placed myocardial perfusion cannula is removed and the myocardial perfusion is performed directly from the CPB cannula. In this way, the rewarming time and consequently the total operation time is reduced to a minimum. The first and the second branches of the dacron graft are then anastomosed to the innominate and the left carotid artery, respectively. Once the CPB is discontinued, the fourth branch used for antegrade perfusion is ligated and resected.
Statistical analysis Data analyses. Data were collected and analysed retrospectively. SPSS 13.0 (SPSS, Inc., Chicago, IL, USA) was used to perform data analysis. The primary end-points were in-hospital mortality and distal aortic operation. Categorical and continuous variables were summarized as percentages and means ± standard deviation (SD), respectively. The independent samples Student’s t-test or, where appropriate, the non-parametric Mann–Whitney U-test and the one-way ANOVA (analysis of variance) test and Fisher’s exact test were used for comparisons of variables among each group. Two-tailed P-values of ≤0.05 were considered significant.
RESULTS The detailed intra- and perioperative data are given in Tables 2 and 3.
Group 1 CPB and X-clamp times were 255 ± 74 and 156 ± 51 min, respectively. The intraoperative mortality and 30-day mortality were 4% (n = 1) and 20% (n = 5), respectively. There was one intraoperative death. This patient could not be weaned from the cardiopulmonary support and died on venoarterial extracorporeal membrane oxygenation support after receiving venous bypass as ultima ratio despite all efforts.
AORTIC SURGERY
M. Shrestha et al. / European Journal of Cardio-Thoracic Surgery
M. Shrestha et al. / European Journal of Cardio-Thoracic Surgery
4
Table 2: Operative data Variables
Total (n = 52)
Group 1 (n = 25)
Group 2 (n = 16)
Group 3 (n = 11)
P-value
CPB time (min) X-clamp time (min) Circulatory arrest times (min) SACP SACP time (min) Concomittant procedures CABG Bentall Aortic valve reconstruction (total) David procedure Yacoub procedure Root stabilization Blood products RBC (n) Platelets (n) Fresh frozen plasma (n)
262 ± 64 159 ± 45 55 ± 24 n = 52 (100%) 87 ± 40
255 ± 74 156 ± 51 64 ± 25 n = 25 (100%) 66 ± 28
260 ± 62 168 ± 49 50 ± 28 n = 16 (100%) 92 ± 42
274 ± 48 147 ± 37 58 ± 16 n = 11 (100%) 116 ± 34
n.s. n.s. n.s. n.s.
Total aortic arch replacement with frozen elephant trunk in acute type A aortic dissections: are we pushing the limits too far?† Malakh Shrestha*‡, Felix Fleissner‡, Fabio Ius, Nurbol Koigeldiyev, Tim Kaufeld, Erik Beckmann, Andreas Martens and Axel Haverich Department of Cardiothoracic, 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, 30625 Hannover, Germany. Tel: +49-511-5326238; fax: +49- 511-5328156; e-mail: [email protected] (M. Shrestha). Received 13 November 2013; received in revised form 12 March 2014; accepted 28 March 2014
Abstract OBJECTIVES: Acute type A aortic dissection (AADA) is a surgical emergency. In patients with aortic arch and descending aorta (DeBakey type I) involvement, performing a total aortic arch replacement with frozen elephant trunk (FET) for supposedly better long-term results is controversial. We hereby present our results. METHODS: From February 2004 to August 2013, 52 patients with acute aortic dissection DeBakey type I received a FET procedure at our centre (43 males, age 59.21 ± 11.67 years). All patients had an intimal tear in the aortic arch and/or proximal descending aorta. Concomitant procedures were Bentall (n = 15) and aortic valve repair (n = 30). RESULTS: Cardiopulmonary bypass (CPB), X-clamp and cardiac arrest times were 262 ± 64, 159 ± 45 and 55 ± 24 min, respectively. The 30day mortality rate was 13% (n = 7). Stroke and re-thoracotomy for bleeding were 12% (n = 6) and 23% (n = 12), respectively. Postoperative recurrent nerve palsy and spinal cord injury rates were 10% (5 of 52) and 4% (2 of 52), respectively. Follow-up was 40 ± 24 months. During follow-up, no patient died and no patient required a reoperation for the aortic arch. CONCLUSIONS: Our results with FET in AADA show acceptable results. Total aortic arch replacement with an FET in AADA patients does demand high technical skills. In spite of this, we believe FET improves long-term outcomes in cases of AADA with intima tear or re-entry in the aortic arch or the descending aorta (DeBakey type I). Modern grafts with four side branches as well as sewing collars for the distal anastomosis have helped to further ‘simplify’ the FET implantation. However, such a strategy is not appropriate in all AADA cases; it should be implemented only in experienced centres and only if absolutely necessary. Keywords: Frozen elephant trunk • Aortic surgery • Aneurysm • Acute dissection
INTRODUCTION Acute type A aortic dissection is an absolute emergency with extremely dismal outcomes without surgical therapy. There is still no consensus regarding the ‘ideal’ surgical therapy for the treatment of acute type A aortic dissection involving the aortic arch and the descending aorta (DeBakey type I). To simplify the two-stage repair of the combined disease of the aortic arch and the proximal descending aorta, ‘Elephant trunk (ET) technique’ was introduced by Borst et al. at our centre in March 1982 [1]. Over the years, the indication for this technique has been expanded to include aortic dissections, both acute and chronic [2, 3]. Endovascular stent-graft technology was introduced by Dake et al. to treat aortic pathology in 1998 [4]. However, a totally endovascular treatment of the aortic arch pathology is still difficult because of the supra-aortic vessels.
A combination of the above two techniques resulted in the so-called frozen elephant technique [4, 5]. It therefore represents an evolution of the classical elephant procedure. Initially, the ‘frozen elephant trunk’ (FET) technique was used for the treatment of aortic aneurysms. Later, the indications for FET were expanded to patients with dissection involving the arch and the descending aorta (DeBakey type I) due to false lumen thrombosis around the stent graft, since a patent false lumen had been identified as a risk factor for aortic dilatation and late operation. The purpose of this study was to assess the results of the frozen ET technique for the treatment of acute aortic dissection (DeBakey type I).
MATERIALS AND METHODS Patients
†
Presented at the 27th Annual Meeting of the European Association for CardioThoracic Surgery, Vienna, Austria, 5–9 October 2013. ‡ Both authors contributed equally to this work.
At our centre, the first FET implantation was done in August 2001 for thoracic aortic aneurysm. With more experience, we expanded
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
AORTIC SURGERY
ORIGINAL ARTICLE
European Journal of Cardio-Thoracic Surgery (2014) 1–6 doi:10.1093/ejcts/ezu185
M. Shrestha et al. / European Journal of Cardio-Thoracic Surgery
2
the indication for FET to chronic type A aortic dissection, acute type B aortic dissection and finally to acute type A aortic dissection (AADA, February 2004). The ethical committee of our institution gave approval for this retrospective study. Between February 2004 and August 2013, 52 consecutive patients underwent complete aortic arch replacement and stenting of the proximal descending aorta with the FET technique in a setting of acute DeBakey type I dissection. Patient records were retrospectively reviewed. Clinical follow-up ended in August 2013 and was 100% complete. This was done by visiting the patients at our outpatient clinic or contacting them at home or their physician by phone. Preoperative demographic patient characteristics are reported in Table 1. All patients had an intimal tear in the distal aortic arch and/or proximal descending aorta. Overall, this study population showed a high operative risk (all acute aortic dissection, n = 52; all emergent procedures, n = 52). The patients were predominantly male (43 of 52, 83%) and middle aged (age average 59.21 (±11.67, range from 41 to 79 years). Eleven patients (21%) showed acute neurological deficits, 4 (8%) displayed a cardiac tamponade and 13 (25%) showed signs of acute visceral or peripheral malperfusion. Because of the long period of the study, the patient cohort was divided into three groups for the sake of objective analysis. Especially for the last group, we modified our techniques in accordance with lessons learnt from the recently published 10-year experience (2001–11) with FET.
Group I (2004–10). During this period, the FET technique was performed using the Custom-made ‘Chavan–Haverich’ hybrid graft as well as the Jotec E-vita. A total of 25 patients (male: n = 22, age: 62.79 ± 9.66 years) were operated on. Twelve patients received Chavan–Haverich and 13 Jotec E-Vita, respectively. Forty percent (10 of 25) of these patients underwent concomitant aortic valve sparing root procedures.
Group II (2010–11). During this period, the FET technique was performed using the Thoraflex hybrid graft as part of the CE certification trial. During the first phase of the trial, only the hybrid graft without the delivery system was available. A total of 16 patients (male: n = 12, age: 56.81 ± 13.96 years) received this Thoraflex hybrid graft. Sixty-nine percent (11 of 16) of these patients underwent concomitant aortic valve sparing root procedures.
Group III (Since 2013). During this period, the FET technique was performed using the Thoraflex hybrid graft as part of the post CE certification registry. We also modified our surgical technique for this trial. A total of 11 patients (male: n = 9, age: 54.55 ± 10.08 years) were operated on [6]. Almost 82% (9 of 11) of these patients underwent concomitant aortic valve sparing root procedures.
Surgical technique Because of the long period, obviously, surgical techniques have evolved through the years.
Early years (2004–10). Custom-made ‘Haverich–Chavan’ (Curative GmbH, Dresden, Germany) hybrid grafts were implanted in 12 patients. This early custom-made hybrid graft was very stiff and could not be completely collapsed. The hybrid graft was delivered into the descending aorta with the help of a special forceps. This sometimes resulted in technical difficulty in placing the stented portion into the descending aorta, especially in smaller sized dissected descending aorta. We changed to the Jotec E-vita hybrid graft after its availability. Thirteen acute dissection patients received E-vita grafts. Although the E-vita graft was a considerable advance over previous custom-made grafts, in our experience, there were still some significant technical shortcomings. Therefore, together with Vascutek Terumo, we developed a novel FET hybrid graft that would be technically simple to implant and also incorporate innovative features: (i) The Gel-coat on the graft for classical haemostatic properties. (ii) Different sizes of the un-stented and stented parts to ensure that a second graft for the aortic arch replacement would not be necessary in patients with large descending aorta. (iii) A sewing collar to simplify the distal anastomosis. (iv) Branched grafts for the aortic arch to facilitate active haemostasis. (v) Novel delivery system for technically simple and rapid deployment.
Description of the Thoraflex hybrid graft. The Thoraflex hybrid graft (Vascutek, Terumo) consists of a 4-branched arch graft with a stent graft at the distal end. The proximal part is a conventional gel-coated woven polyester graft. The stented section of the graft is a self-expanding endo-prosthesis constructed of
Table 1: Preoperative characteristics Variables
Total (n = 52)
Group 1 (n = 25)
Group 2 (n = 16)
Group 3 (n = 11)
P-value
Total Male Age (years) >60 years of age Marfan Neurological deficits Shock Cardiac tamponade Malperfusion
52 (100%) 43 (83%) 59 ± 12 24 (46%) 1 (2%) 7 (13%) 1 (2%) 4 (8%) 13 (25%)
25 (100%) 22 (88%) 62.79 ± 9.66 14 (56%) 1 (4%) 5 (20%) 1 (4%) 2 (8%) 8 (32%)
16 (100%) 12 (74%) 56.81 ± 13.96 7 (44%) 0 4 (5%) 0 1 (6%) 4 (5%)
11 (100%) 9 (82%) 54.55 ± 10.08 3 (27%) 0 3 (27%) 0 1 (9%) 1 (9%)
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
Continuous data summarized as mean ± SD; categorical and binary data summarized as frequency (percentage). SD: standard deviation.
thin-wall polyester and nitinol ring stents, which are attached to fabric with braided polyester sutures. The grafts are available in different sizes (26–40 mm in diameter for the stented portion). The length of the stented part is either 100 or 150 mm. Another unique feature of this graft is that the proximal un-stented and distal stented parts are available in different sizes (e.g. un-stented part 32 mm in diameter and stented part with 40 mm diameter).
Lessons learned. We recently published our 30-year experience (1982–2012) with conventional elephant trunk technique as well as our 10-year experience (2001–11) with FET. We identified several parameters that still needed to be improved and initiated several significant changes in perioperative management. (i) Reduction in perioperative stroke rate: Use of carbon dioxide (CO2) sufflation in the operative field to prevent air embolism. (ii) Myocardial protection (‘Beating Heart’ aortic arch replacement): To minimize the myocardial ischaemia time, we initiated the ‘beating heart’ aortic replacement technique. As soon as the aortic root or other cardiac procedure was performed, the heart was perfused antegradely and the aortic arch placed on ‘beating heart’. (iii) Spinal cord protection: To minimize spinal cord ischaemia, cerebrospinal fluid (CSF) drainage was performed so that spinal pressure remained below 12 mmHg. The CSF drainage catheter was implanted just before the operation or immediately postoperatively in the intensive care unit (ICU). (iv) Shorten the circulatory arrest and cardiopulmonary bypass (CPB) times: To minimize coagulopathy, perfusion of the body was restarted immediately after the distal anastomosis and ‘re-warming’ begun. (v) Haemostasis: Time was taken to perform each anastomosis carefully to have ‘perfect’ haemostatic anastomoses.
Surgical technique since 2013. We have standardized the surgical technique as far as possible. After a standard median sternotomy, extracorporeal circulation is initiated with cannulation of the aorta and the right atrium. This technique of cannulation of the ascending aorta is performed even in AADA, as has been published by our group [7]. The left side of the heart is vented through the right superior pulmonary vein. CO2 is blown in to the surgical field. During this period, the patient is cooled to a nasopharyngeal temperature of 26°C, the aortic root/ascending aortic procedure or other concomitant procedures (e.g. CABG) are performed if necessary. Blood cardioplegia has been our preferred method of myocardial protection. Cardioplegia is repeated approximately every 30 min. After the aortic root or other cardiac concomitant procedure, de-airing of the myocardium is performed. Antegrade perfusion of the myocardium is initiated through the ascending aorta/ aortic prosthesis at the rate of 100–200 ml/min (via Cardioplegia cannula). Care is taken that the pressure in the aortic root does not exceed a mean pressure of 80–90 mmHg. Cerebral protection by moderate hypothermic circulatory arrest with selective antegrade cerebral perfusion (SACP) has been our preferred technique. After reaching the desired temperature (26°C), the systemic circulation is arrested and the aorta opened. With the patient in the Trendelenburg position, catheters (Medtronic DPL, USA) are introduced into the left carotid artery and the innominate artery for SACP. The left sub-clavian artery is clamped or occluded with a Fogarty catheter (Baxter, Deerfield, IL, USA), thus avoiding a steal
3
phenomenon as well as to prevent blood flowing into the operative field. Cerebral perfusion is initiated at a rate of 10 ml/kg/min. The blood temperature of SACP is 24°C. The aorta is transected either between the left common carotid artery and the left sub-clavian artery or distal to the left subclavian artery. The self-expanding FET part of the Thoraflex hybrid prosthesis is placed through the opened aortic arch and deployed in the descending aorta. A sewing collar between the graft segments simplifies the ‘distal anastomosis’. In patients where the distal aortic arch is not aneurysmatic, the distal anastomosis is done between the left carotid and the left sub-clavian arteries. After the distal anastomosis is completed, the left sub-clavian artery is anastomosed to the third branch of the arch graft. The perfusion to the lower part of the body and the sub-clavian artery is then restarted via the fourth branch of the graft that is connected to the inner curvature of the prosthesis vis-à-vis to the supra-aortic vessels. Rewarming of the patient is started leaving the temperature of the SACP at 28–30°C. The proximal end of the graft is anastomosed, either to the native ascending aorta or the ascending aortic graft. The previously placed myocardial perfusion cannula is removed and the myocardial perfusion is performed directly from the CPB cannula. In this way, the rewarming time and consequently the total operation time is reduced to a minimum. The first and the second branches of the dacron graft are then anastomosed to the innominate and the left carotid artery, respectively. Once the CPB is discontinued, the fourth branch used for antegrade perfusion is ligated and resected.
Statistical analysis Data analyses. Data were collected and analysed retrospectively. SPSS 13.0 (SPSS, Inc., Chicago, IL, USA) was used to perform data analysis. The primary end-points were in-hospital mortality and distal aortic operation. Categorical and continuous variables were summarized as percentages and means ± standard deviation (SD), respectively. The independent samples Student’s t-test or, where appropriate, the non-parametric Mann–Whitney U-test and the one-way ANOVA (analysis of variance) test and Fisher’s exact test were used for comparisons of variables among each group. Two-tailed P-values of ≤0.05 were considered significant.
RESULTS The detailed intra- and perioperative data are given in Tables 2 and 3.
Group 1 CPB and X-clamp times were 255 ± 74 and 156 ± 51 min, respectively. The intraoperative mortality and 30-day mortality were 4% (n = 1) and 20% (n = 5), respectively. There was one intraoperative death. This patient could not be weaned from the cardiopulmonary support and died on venoarterial extracorporeal membrane oxygenation support after receiving venous bypass as ultima ratio despite all efforts.
AORTIC SURGERY
M. Shrestha et al. / European Journal of Cardio-Thoracic Surgery
M. Shrestha et al. / European Journal of Cardio-Thoracic Surgery
4
Table 2: Operative data Variables
Total (n = 52)
Group 1 (n = 25)
Group 2 (n = 16)
Group 3 (n = 11)
P-value
CPB time (min) X-clamp time (min) Circulatory arrest times (min) SACP SACP time (min) Concomittant procedures CABG Bentall Aortic valve reconstruction (total) David procedure Yacoub procedure Root stabilization Blood products RBC (n) Platelets (n) Fresh frozen plasma (n)
262 ± 64 159 ± 45 55 ± 24 n = 52 (100%) 87 ± 40
255 ± 74 156 ± 51 64 ± 25 n = 25 (100%) 66 ± 28
260 ± 62 168 ± 49 50 ± 28 n = 16 (100%) 92 ± 42
274 ± 48 147 ± 37 58 ± 16 n = 11 (100%) 116 ± 34
n.s. n.s. n.s. n.s.
