ORIGINAL ARTICLE
European Journal of Cardio-Thoracic Surgery 48 (2015) 258–263 doi:10.1093/ejcts/ezu488 Advance Access publication 18 December 2014
Cite this article as: Rylski B, Beyersdorf F, Desai ND, Euringer W, Siepe M, Kari FA et al. Distal aortic reintervention after surgery for acute DeBakey type I or II aortic dissection: open versus endovascular repair. Eur J Cardiothorac Surg 2015;48:258–63.
Distal aortic reintervention after surgery for acute DeBakey type I or II aortic dissection: open versus endovascular repair† Bartosz Rylskia,b,*, Friedhelm Beyersdorfa, Nimesh D. Desaib, Wulf Euringera, Matthias Siepea, Fabian A. Karia, Prashanth Vallabhajosyulab, Wilson Y. Szetob, Rita K. Milewskib and Joseph E. Bavariab a b
Heart Centre Freiburg University, Freiburg, Germany Hospital of the University of Pennsylvania, Philadelphia, PA, USA
* Corresponding author. Heart Centre Freiburg University, Hugstetter Str. 55, 79106 Freiburg, Germany. Tel: +49-761-27028180; fax: +49-761-27028670; e-mail: [email protected] (B. Rylski). Received 8 July 2014; received in revised form 21 October 2014; accepted 30 October 2014
Abstract OBJECTIVES: Aortic dissection DeBakey type I and II may require distal reinterventions after initial proximal repair. We evaluated outcomes following open versus endovascular approaches to distal aortic pathologies after surgery for acute dissection. METHODS: One hundred and forty-one consecutive patients underwent 152 distal reinterventions after previous type I or II dissection repair [63 (first quartile, 55; third quartile, 72) years old; initially 86% DeBakey type I; 54% hemiarch, 39% isolated ascending, 7% total arch replacement] at two tertiary centres in the USA and Europe over the last 14 years. Among them, 56 and 85 required reintervention for the aortic arch and the descending aorta, respectively. The median follow-up was 2.1 (first quartile, 0.8; third quartile, 5.8) years (439 patientyears). RESULTS: The median time between acute aortic dissection repair and descending aortic reintervention was longer in the open group (2.7 (first quartile, 0.8; third quartile, 6.7) vs 0.6 (first quartile, 0.1; third quartile, 3.5) years, P < 0.01). There was one irreversible spinal ischaemia in the open and one stroke in the endovascular group. Two patients in the open and none in the endovascular group required re-exploration for bleeding. Two open and 4 endovascular patients required more than 1 distal reintervention (6 vs 8%, P = 1). Descending aortic open-repair patients experienced higher in-hospital mortality (23 vs 0%, P < 0.01) and lower survival at 1 and 5 years (74 ± 8% vs 96 ± 3%, 65 ± 9% vs 92 ± 5%, P < 0.01, respectively). CONCLUSIONS: Endovascular intervention for descending aortic pathologies after DeBakey type I or II dissection surgical repair is associated with lower in-hospital mortality and better survival, and does not raise the likelihood of later reinterventions at the mid-term follow-up. Keywords: Aortic dissection • Aortic surgery • Aortic aneurysm • Endovascular reintervention
INTRODUCTION In most acute ascending aortic dissection patients, the dissection process extends beyond the left subclavian artery and emergency proximal aortic repair does not eliminate all dissected aortic segments [1, 2]. The residually dissected distal aorta grows at a rate between 1 and 7 mm/year [3, 4]. Therefore some, especially younger, survivors develop later dissected aortic aneurysm requiring secondary intervention. Furthermore, initial replacement of all dissected aortic tissue in DeBakey type II dissections does not eliminate the risk of later adverse aortic events like new-onset acute DeBakey type III dissection, penetrating atherosclerotic † Presented at the 28th Annual Meeting of the European Association for CardioThoracic Surgery, Milan, Italy, 11–15 October 2014.
ulcers or suture aneurysm [1, 5]. The overall risk of late distal aortic interventions following proximal repair of acute ascending aortic dissection ranges between 6 and 15% [1, 4, 5]. Within recent decades, endovascular techniques have been widely adopted to treat even complicated thoracic and abdominal aortic disease [6]. However, since currently available stent grafts were designed for aortic aneurysm and not aortic dissection repair, much controversy exists over the role of endovascular treatment among patients presenting a chronic dissection process such as residual dissection after proximal aortic repair for type I or II dissection. Our pooled databases covering 14 years of aortic interventions from two tertiary aortic centres in the USA and Europe provide a unique opportunity to analyse clinical data on individuals with a history of acute DeBakey type I or II dissection and later distal
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
B. Rylski et al. / European Journal of Cardio-Thoracic Surgery
MATERIALS AND METHODS Patient population The institutional review committees at both participating centres approved this study and the need for informed consent was waived. The Hospital of the University of Pennsylvania in Philadelphia and Heart Center Freiburg University in Freiburg and Bad Krozingen reviewed their aortic databases covering patients operated on between 2000 and 2013. Patients with Marfan, Loeys-Dietz and Ehlers-Danlos syndromes were excluded from the analysis. A total of 141 patients (University of Pennsylvania, 100; Heart Center Freiburg, 41) with a history of proximal aortic repair for acute type I or II aortic dissection (87 initially operated on at the study centres) underwent aortic reintervention for pathologies of the aortic arch or descending aorta and comprise the study population.
Reinterventions The decision regarding open versus endovascular aortic repair was based upon patients’ aortic anatomy and comorbidities, and was made by each surgeon individually. Indications for reintervention were a maximum aortic diameter ≥5.5 cm, rapid aneurysmal degeneration with growth rate >1 cm/year, suture line aneurysm, malperfusion syndrome due to compression of the true aortic lumen, penetrating atherosclerotic ulcers and large re-entries between false and true lumen in the descending aorta. The first endovascular reintervention took place in the fifth year of the study. Figure 1 illustrates the rate of endovascular distal aortic reinterventions per year throughout the study period. The first endovascular reinterventions were performed for true aortic lumen compression or Crawford I aneurysm with greatest diameter in the proximal descending thoracic aorta. Abdominal and aortic arch aneurysms were endovascularly treated in the later study period. Replacement of all aneurysmal aortic segments was the aim in patients in the open group. In the endovascular group, our goal was to remodel the aorta by covering the large proximal re-entries to induce thrombosis of the false lumen. Stent grafts were
oversized by 5–10% in relation to the true lumen diameter of the proximal landing zone. Post-deployment balloon dilatation of the stent grafts was avoided except at overlapping stent graft segments. Distal extension was usually carried distally to the diaphragm or coeliac axis to promote remodelling of the entire thoracic aorta.
Patient follow-up Surveillance follow-up data were obtained by contacting the patients’ general practitioners, the patients themselves and their family members via aortic clinic office visits or by the Social Security Death Index. The complete follow-up was available in 130 of 132 hospital survivors (99%). Patients were followed up a total of 439 patient-years, with a median follow-up among survivors of 2.1 (first quartile, 0.8; third quartile, 5.8) years. According to current guidelines [7], the follow-up protocol included postoperative computed tomography (CT) angiography before discharge, clinical examination and CT angiography 6 and 12 months postoperatively, and annually thereafter at our institutional aortic outpatient clinics or at their regional hospitals for patients living in remote areas.
Statistical analysis Continuous data are presented as median (first quartile; third quartile); categorical variables are given as counts and percentages. For comparison of continuous variables the Mann–Whitney rank sum test was employed. Categorical variables were compared using Fisher’s exact test. Survival was analysed using the Kaplan– Meier method and log-rank calculations. All statistical calculations were performed using SigmaPlot, version 12 (Systat Software, San Jose, CA, USA).
RESULTS Demographics Patient demographics are summarized in Table 1. Their average age was 63 years and over 70% were males in both groups. Most of the open and endovascular patients had a history of DeBakey type I dissection (83 vs 91%, P = 0.47). Distal aortic repair for acute aortic dissection included isolated ascending replacement (no arch replacement) in 49% and hemiarch replacement in 48% open patients, whereas the majority (63%) of endovascular patients underwent initially hemiarch replacement. Two percent of open and 22% of endovascular patients (P < 0.01) underwent concomitant antegrade thoracic endovascular aortic repair (TEVAR) during the surgery for acute aortic dissection. There were no group differences with regard to hypertension, diabetes mellitus, renal failure, coronary artery disease and history of stroke (Table 1). Reintervention was undertaken as an emergency procedure in 6 open and 5 endovascular patients.
Distal aortic reinterventions: aortic arch Figure 1: The rate of endovascular distal aortic reinterventions per year throughout the study period.
The majority of aortic arch reinterventions took place in an open fashion (20 hemiarch and 32 total arch redo replacements). Four patients underwent endovascular (hybrid) total arch repair with
AORTIC SURGERY
reintervention. The aim of this study was to examine the outcomes of open versus endovascular reinterventions on distal aortic pathologies after surgery for acute ascending aortic dissection.
259
260
B. Rylski et al. / European Journal of Cardio-Thoracic Surgery
Table 1: Demographics, history of acute ascending aortic dissection and current clinical presentation
Age (years) Over 80 years old Male gender History of aortic dissection DeBakey type I DeBakey type II Proximal repair Aortic valve repair Aortic root replacement Wheat procedure Distal repair No arch replacement Hemiarch replacement Total arch replacement Antegrade TEVAR Current clinical presentation Hypertension Diabetes mellitus Hyperlipidaemia Renal failure COPD Current smoker Coronary artery disease BAV History of stroke Current distal aortic pathology Aortic arch aneurysm Thoracoabdominal aneurysm Crawford I Crawford II Crawford III Crawford IV Abdominal aortic aneurysm True lumen compression ‘De novo’ type III dissection PAU
P-value
All (n = 141)
Open group (n = 87)
Endovascular group (n = 54)
63 (55; 72) 8 (6) 100 (71)
63 (55; 72) 5 (6) 61 (71)
63 (56; 71) 3 (6) 39 (72)
0.89 1.00 0.41
123 (87) 18 (13)
74 (85) 13 (15)
49 (91) 5 (9)
0.44 0.44
110 (78) 26 (18) 6 (4)
66 (76) 16 (18) 6 (7)
44 (81) 10 (19) 0
0.53 0.84 0.08
55 (39) 76 (54) 10 (7) 14 (10)
43 (49) 42 (48) 2 (2) 2 (2)
12 (22) 34 (63) 8 (15) 12 (22)
European Journal of Cardio-Thoracic Surgery 48 (2015) 258–263 doi:10.1093/ejcts/ezu488 Advance Access publication 18 December 2014
Cite this article as: Rylski B, Beyersdorf F, Desai ND, Euringer W, Siepe M, Kari FA et al. Distal aortic reintervention after surgery for acute DeBakey type I or II aortic dissection: open versus endovascular repair. Eur J Cardiothorac Surg 2015;48:258–63.
Distal aortic reintervention after surgery for acute DeBakey type I or II aortic dissection: open versus endovascular repair† Bartosz Rylskia,b,*, Friedhelm Beyersdorfa, Nimesh D. Desaib, Wulf Euringera, Matthias Siepea, Fabian A. Karia, Prashanth Vallabhajosyulab, Wilson Y. Szetob, Rita K. Milewskib and Joseph E. Bavariab a b
Heart Centre Freiburg University, Freiburg, Germany Hospital of the University of Pennsylvania, Philadelphia, PA, USA
* Corresponding author. Heart Centre Freiburg University, Hugstetter Str. 55, 79106 Freiburg, Germany. Tel: +49-761-27028180; fax: +49-761-27028670; e-mail: [email protected] (B. Rylski). Received 8 July 2014; received in revised form 21 October 2014; accepted 30 October 2014
Abstract OBJECTIVES: Aortic dissection DeBakey type I and II may require distal reinterventions after initial proximal repair. We evaluated outcomes following open versus endovascular approaches to distal aortic pathologies after surgery for acute dissection. METHODS: One hundred and forty-one consecutive patients underwent 152 distal reinterventions after previous type I or II dissection repair [63 (first quartile, 55; third quartile, 72) years old; initially 86% DeBakey type I; 54% hemiarch, 39% isolated ascending, 7% total arch replacement] at two tertiary centres in the USA and Europe over the last 14 years. Among them, 56 and 85 required reintervention for the aortic arch and the descending aorta, respectively. The median follow-up was 2.1 (first quartile, 0.8; third quartile, 5.8) years (439 patientyears). RESULTS: The median time between acute aortic dissection repair and descending aortic reintervention was longer in the open group (2.7 (first quartile, 0.8; third quartile, 6.7) vs 0.6 (first quartile, 0.1; third quartile, 3.5) years, P < 0.01). There was one irreversible spinal ischaemia in the open and one stroke in the endovascular group. Two patients in the open and none in the endovascular group required re-exploration for bleeding. Two open and 4 endovascular patients required more than 1 distal reintervention (6 vs 8%, P = 1). Descending aortic open-repair patients experienced higher in-hospital mortality (23 vs 0%, P < 0.01) and lower survival at 1 and 5 years (74 ± 8% vs 96 ± 3%, 65 ± 9% vs 92 ± 5%, P < 0.01, respectively). CONCLUSIONS: Endovascular intervention for descending aortic pathologies after DeBakey type I or II dissection surgical repair is associated with lower in-hospital mortality and better survival, and does not raise the likelihood of later reinterventions at the mid-term follow-up. Keywords: Aortic dissection • Aortic surgery • Aortic aneurysm • Endovascular reintervention
INTRODUCTION In most acute ascending aortic dissection patients, the dissection process extends beyond the left subclavian artery and emergency proximal aortic repair does not eliminate all dissected aortic segments [1, 2]. The residually dissected distal aorta grows at a rate between 1 and 7 mm/year [3, 4]. Therefore some, especially younger, survivors develop later dissected aortic aneurysm requiring secondary intervention. Furthermore, initial replacement of all dissected aortic tissue in DeBakey type II dissections does not eliminate the risk of later adverse aortic events like new-onset acute DeBakey type III dissection, penetrating atherosclerotic † Presented at the 28th Annual Meeting of the European Association for CardioThoracic Surgery, Milan, Italy, 11–15 October 2014.
ulcers or suture aneurysm [1, 5]. The overall risk of late distal aortic interventions following proximal repair of acute ascending aortic dissection ranges between 6 and 15% [1, 4, 5]. Within recent decades, endovascular techniques have been widely adopted to treat even complicated thoracic and abdominal aortic disease [6]. However, since currently available stent grafts were designed for aortic aneurysm and not aortic dissection repair, much controversy exists over the role of endovascular treatment among patients presenting a chronic dissection process such as residual dissection after proximal aortic repair for type I or II dissection. Our pooled databases covering 14 years of aortic interventions from two tertiary aortic centres in the USA and Europe provide a unique opportunity to analyse clinical data on individuals with a history of acute DeBakey type I or II dissection and later distal
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
B. Rylski et al. / European Journal of Cardio-Thoracic Surgery
MATERIALS AND METHODS Patient population The institutional review committees at both participating centres approved this study and the need for informed consent was waived. The Hospital of the University of Pennsylvania in Philadelphia and Heart Center Freiburg University in Freiburg and Bad Krozingen reviewed their aortic databases covering patients operated on between 2000 and 2013. Patients with Marfan, Loeys-Dietz and Ehlers-Danlos syndromes were excluded from the analysis. A total of 141 patients (University of Pennsylvania, 100; Heart Center Freiburg, 41) with a history of proximal aortic repair for acute type I or II aortic dissection (87 initially operated on at the study centres) underwent aortic reintervention for pathologies of the aortic arch or descending aorta and comprise the study population.
Reinterventions The decision regarding open versus endovascular aortic repair was based upon patients’ aortic anatomy and comorbidities, and was made by each surgeon individually. Indications for reintervention were a maximum aortic diameter ≥5.5 cm, rapid aneurysmal degeneration with growth rate >1 cm/year, suture line aneurysm, malperfusion syndrome due to compression of the true aortic lumen, penetrating atherosclerotic ulcers and large re-entries between false and true lumen in the descending aorta. The first endovascular reintervention took place in the fifth year of the study. Figure 1 illustrates the rate of endovascular distal aortic reinterventions per year throughout the study period. The first endovascular reinterventions were performed for true aortic lumen compression or Crawford I aneurysm with greatest diameter in the proximal descending thoracic aorta. Abdominal and aortic arch aneurysms were endovascularly treated in the later study period. Replacement of all aneurysmal aortic segments was the aim in patients in the open group. In the endovascular group, our goal was to remodel the aorta by covering the large proximal re-entries to induce thrombosis of the false lumen. Stent grafts were
oversized by 5–10% in relation to the true lumen diameter of the proximal landing zone. Post-deployment balloon dilatation of the stent grafts was avoided except at overlapping stent graft segments. Distal extension was usually carried distally to the diaphragm or coeliac axis to promote remodelling of the entire thoracic aorta.
Patient follow-up Surveillance follow-up data were obtained by contacting the patients’ general practitioners, the patients themselves and their family members via aortic clinic office visits or by the Social Security Death Index. The complete follow-up was available in 130 of 132 hospital survivors (99%). Patients were followed up a total of 439 patient-years, with a median follow-up among survivors of 2.1 (first quartile, 0.8; third quartile, 5.8) years. According to current guidelines [7], the follow-up protocol included postoperative computed tomography (CT) angiography before discharge, clinical examination and CT angiography 6 and 12 months postoperatively, and annually thereafter at our institutional aortic outpatient clinics or at their regional hospitals for patients living in remote areas.
Statistical analysis Continuous data are presented as median (first quartile; third quartile); categorical variables are given as counts and percentages. For comparison of continuous variables the Mann–Whitney rank sum test was employed. Categorical variables were compared using Fisher’s exact test. Survival was analysed using the Kaplan– Meier method and log-rank calculations. All statistical calculations were performed using SigmaPlot, version 12 (Systat Software, San Jose, CA, USA).
RESULTS Demographics Patient demographics are summarized in Table 1. Their average age was 63 years and over 70% were males in both groups. Most of the open and endovascular patients had a history of DeBakey type I dissection (83 vs 91%, P = 0.47). Distal aortic repair for acute aortic dissection included isolated ascending replacement (no arch replacement) in 49% and hemiarch replacement in 48% open patients, whereas the majority (63%) of endovascular patients underwent initially hemiarch replacement. Two percent of open and 22% of endovascular patients (P < 0.01) underwent concomitant antegrade thoracic endovascular aortic repair (TEVAR) during the surgery for acute aortic dissection. There were no group differences with regard to hypertension, diabetes mellitus, renal failure, coronary artery disease and history of stroke (Table 1). Reintervention was undertaken as an emergency procedure in 6 open and 5 endovascular patients.
Distal aortic reinterventions: aortic arch Figure 1: The rate of endovascular distal aortic reinterventions per year throughout the study period.
The majority of aortic arch reinterventions took place in an open fashion (20 hemiarch and 32 total arch redo replacements). Four patients underwent endovascular (hybrid) total arch repair with
AORTIC SURGERY
reintervention. The aim of this study was to examine the outcomes of open versus endovascular reinterventions on distal aortic pathologies after surgery for acute ascending aortic dissection.
259
260
B. Rylski et al. / European Journal of Cardio-Thoracic Surgery
Table 1: Demographics, history of acute ascending aortic dissection and current clinical presentation
Age (years) Over 80 years old Male gender History of aortic dissection DeBakey type I DeBakey type II Proximal repair Aortic valve repair Aortic root replacement Wheat procedure Distal repair No arch replacement Hemiarch replacement Total arch replacement Antegrade TEVAR Current clinical presentation Hypertension Diabetes mellitus Hyperlipidaemia Renal failure COPD Current smoker Coronary artery disease BAV History of stroke Current distal aortic pathology Aortic arch aneurysm Thoracoabdominal aneurysm Crawford I Crawford II Crawford III Crawford IV Abdominal aortic aneurysm True lumen compression ‘De novo’ type III dissection PAU
P-value
All (n = 141)
Open group (n = 87)
Endovascular group (n = 54)
63 (55; 72) 8 (6) 100 (71)
63 (55; 72) 5 (6) 61 (71)
63 (56; 71) 3 (6) 39 (72)
0.89 1.00 0.41
123 (87) 18 (13)
74 (85) 13 (15)
49 (91) 5 (9)
0.44 0.44
110 (78) 26 (18) 6 (4)
66 (76) 16 (18) 6 (7)
44 (81) 10 (19) 0
0.53 0.84 0.08
55 (39) 76 (54) 10 (7) 14 (10)
43 (49) 42 (48) 2 (2) 2 (2)
12 (22) 34 (63) 8 (15) 12 (22)
