European Journal of Cardio-Thoracic Surgery Advance Access published December 11, 2014

ORIGINAL ARTICLE

European Journal of Cardio-Thoracic Surgery (2014) 1–6 doi:10.1093/ejcts/ezu454

Impact of surgical experience on outcome in surgery of acute type A aortic dissection† Aristidis Lenos, Petros Bougioukakis, Vadim Irimie, Michael Zacher, Anno Diegeler and Paul P. Urbanski* Cardiovascular Clinic Bad Neustadt, Bad Neustadt, Germany * Corresponding author. Cardiovascular Clinic Bad Neustadt, Herz- und Gefaess-Klinik, Salzburger Leite 1, 97616 Bad Neustadt, Germany. Tel: +49-977-1662416; fax: +49-977-1651219; e-mail: [email protected], [email protected] (P.P. Urbanski). Received 28 August 2014; received in revised form 28 October 2014; accepted 4 November 2014

Abstract OBJECTIVES: The study was conducted to evaluate our results of acute aortic dissection repair taking into account the impact of surgical experience in aortic surgery. METHODS: Between August 2002 and March 2013, 162 consecutive patients (mean age: 63 ± 14 years) underwent surgery for acute type A aortic dissection. All patients were operated on by one of the clinic’s attending surgeons with wide experience in cardiac surgery (at least 2000 procedures performed personally), however about one-half of the patients (75 patients, 46%) were operated by the aortic team (AT) surgeons with profound experience in complex aortic pathologies. All perioperative data were collected prospectively and retrospective statistical analysis was performed using uni- and multivariate analyses to identify predictors for surgical adverse outcome (AO) containing in-hospital and/or 90-day mortality and new permanent neurological and organ dysfunctions. RESULTS: AO was observed in 36 patients (22.2%) including in-hospital mortality in 22 (13.6%). Multivariate logistic regression analysis identified surgery not performed by the AT as the strongest predictor for AO (odds ratio: 14.1; 95% confidence interval: 3.5–55.6; P < 0.0001) followed by any malperfusion, myocardial infarction and creatinine level. Two groups were built according to the surgery performed by the AT (Group AT) or by the surgeons not on the AT (Group No-AT). The comparison of the groups showed no relevant differences regarding the preoperative characteristics, especially compromised consciousness, malperfusion and extent of dissection. Yet, the outcomes in Group AT vs No-AT were significantly different presenting AO: 8.0 vs 34.5% (P < 0.0001), in-hospital mortality: 4.0 vs and 21.8% (P < 0.001), new permanent neurological deficit: 2.7 vs 11.5% (P = 0.03), even if valve-sparing repairs and complete arch replacements were much more frequent in Group AT. The groups also differed considerably in regard to cannulation and perfusion management, which might play a decisive role in surgical outcome. CONCLUSIONS: Aortic repair in acute type A dissection, when performed by highly specialized aortic surgeons, offers not only much better outcomes but also provides significantly higher rate of curative albeit valve-sparing aortic repairs. Patient-centred care in referral aortic centres with surgery performed by specialized teams should be striven for to improve surgical results in acute aortic dissection surgery. Keywords: Aortic dissection • Aortic surgery • Surgical experience • Surgical performance

INTRODUCTION Despite current progress in aortic surgery, the mortality and neurological morbidity in the surgery of acute aortic dissection have not changed considerably for decades and still average
15 —25%, especially in multicentre reports [1–3]. On the other hand, there are also reports demonstrating excellent results with an exceptionally high rate of early survival, regardless of the surgical method used or the extent of surgery [4–7]. Yet, the latter very often has one characteristic in common—they reveal that only one † Presented at the 28th Annual Meeting of the European Association for CardioThoracic Surgery, Milan, Italy, 11–15 October 2014.

or very few surgeons performed all procedures, which indicates that surgical experience and performance could have had a decisive impact on surgical outcome. Taking this aspect into account, we analysed the results of acute aortic dissection surgery performed in our clinic during the past decade, and examined, in addition to other several perioperative variables, the impact of surgical experience on the operative and early clinical outcome.

PATIENTS AND METHODS Between August 2002 and March 2013, 162 consecutive patients (mean age 63 ± 14 years) underwent surgery for acute type A

© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

AORTIC SURGERY

Cite this article as: Lenos A, Bougioukakis P, Irimie V, Zacher M, Diegeler A, Urbanski PP. Impact of surgical experience on outcome in surgery of acute type A aortic dissection. Eur J Cardiothorac Surg 2014; doi:10.1093/ejcts/ezu454.

A. Lenos et al. / European Journal of Cardio-Thoracic Surgery

2

Table 1: Preoperative patient characteristics Characteristics

Age Sex male Hypertension COPD Previous neurological event Previous cardiac surgery CADa Unconscious/intubated Malperfusion Cerebral Myocardial ischaemia Circulatory critical state Creatinine (mg/dl) Pericardial effusion Aortic valve defect Insufficiency Mixed/stenosis Artificial valve Marfan syndrome

n (%) or mean ± SD (range) All; 162 (100)

AT; 75 (46.3)

No-AT; 87 (53.7)

P-value

63 ± 14 (23–88) 108 (66.7) 128 (79) 17 (10.5) 7 (4.3) 14 (8.6) 14 (8.6) 34 (21.0) 28 (17.3) 15 (9.3) 12 (7.4) 23 (14.1) 1.1 ± 0.5 (0.4–3.8) 85 (52.5)

62 ± 15 (23–88) 49 (65.3) 61 (81.4) 11 (14.7) 3 (4.0) 9 (12.0) 9 (12.0) 19 (25.3) 15 (20.0) 9 (12.0) 5 (6.7) 8 (10.6) 1.1 ± 0.4 (0.5–3.0) 43 (57.3)

63 ± 14 (29–87) 59 (67.8) 67 (77.0) 6 (6.9) 4 (4.6) 5 (5.7) 5 (5.7) 15 (17.2) 13 (14.7) 6 (6.9) 7 (8.0) 15 (17.2) 1.1 ± 0.6 (0.4–3.8) 42 (48.3)

NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

102 (63.0) 10 (6.2) 5 (3.1) 7 (4.3)

56 (74.7) 5 (6.6) 3 (4) 5 (6.7)

46 (52.9) 5 (5.8) 2 (2.3) 2 (2.3)

NS

COPD: chronic obstructive pulmonary disease (requiring long-term therapy in anamnesis); CAD: coronary artery disease. With surgical relevant stenosis (among 28 patients in whom coronary artery visualization through heart catheterization or cardiac CT was performed).

a

aortic dissection in our clinic (Table 1). Four of them presented a non-A-non-B type of dissection and were operated because the primary entry was located in the aortic arch (Table 2). During this time, only 1 patient referred to our clinic with acute dissection was refused surgery because of a deep coma caused by a dissection-related occlusion of both common carotid arteries. Twenty-two patients were in a critical circulatory state requiring catecholamines, 2 of whom needed mechanical cardiopulmonary resuscitation. All but one patient underwent preoperative computed tomography (CT) angiography. If this examination had not already been done before referral, or if the information that was essential from a surgical standpoint was missing, it was performed in our clinic immediately after admission and included the middle neck area, for imaging of the aortic arch branches. None of the patients under our care died during preoperative diagnostics. All patients were operated on by one of the clinic’s attending surgeons who had wide experience in cardiac surgery with at least 2000 procedures performed personally. However, not all of them had equivalent experience in aortic surgery, especially with complex aortic pathologies, which are normally treated by the aortic team (AT) managed by the senior author (P.P.U.). The average number of acute aortic dissection surgeries among the seven surgeons not belonging to the AT was 11 (range: 4–22); however, this number does not correspond to their personal experience in repair of the acute dissection because only 2 of them were employed in our centre throughout the entire study duration. All perioperative data, the most important of which are presented in detail in Tables 1–4, were collected prospectively. For the purpose of the study we included, as an additional variable, if the surgery was performed by an AT-surgeon or not. The outcomes contained 30-day mortality and morbidity as well as in-hospital and/or 90-day mortality. A retrospective statistical evaluation using uni- and multivariate analyses with logistic

Table 2:

Aortic pathology

Variables

n (%) All; 162 (100)

Extent of dissection D-a D-ab D-abc D-b D-bc Involvement of supra-aortic arteries Involvement of iliac arteries Primary intimal tear E-a E-ab E-abc E-b E-bc E-c

P-value

AT; 75 (46.3)

No-AT; 87 (53.7)

41 (25.3) 43 (26.5) 74 (45.7) 1 (0.6) 3 (1.9) 61 (37.7)

17 (22.7) 24 (32.0) 33 (44.0) 0 (0) 1 (1.3) 26 (35.0)

24 (27.6) 19 (21.8) 41 (47.1) 1 (1.1) 2 (2.3) 35 (40.0)

NS

48 (29.6)

21 (28.0)

27 (31.0)

NS

128 (79.0) 11 (6.8) 2 (1.2) 17 (10.5) 1 (0.6) 3 (1.9)

56 (74.7) 7 (9.3) 1 (1.3) 10 (13.3) 0 (0) 1 (1.3)

72 (82.8) 4 (4.6) 1 (1.1) 7 (8.0) 1 (1.1) 2 (2.3)

NS

NS

D: extent of dissection; E: entry localization; a: ascending aorta; b: aortic arch; c: descending aorta.

regression models was performed to identify independent predictors for adverse outcome (AO), which was defined as in-hospital or 90-day mortality or, in case of survival, a new permanent neurological and/or organ dysfunction. The study was approved by the Institutional Review Board of Rhön Clinics Campus Bad Neustadt.

A. Lenos et al. / European Journal of Cardio-Thoracic Surgery

A variety of different arterial cannulation sites and protection methods were used according to the decision of each individual surgeon and particular clinical situation (Table 3). In the majority of cases, the generally accepted surgical principles consisting of

Table 3: Surgical strategy Variables

Cannulation site Aorta FA CAA left CAA right IA CAA left + FA CAA right + FA Switch of cannulationa Cerebral protection UCP BCP Straight CA No-CA X-clamp of left SAb Open distal repairc

n (%) All; 162 (100)

AT; 75 (46.3)

No-AT; 87 (53.7)

10 (6.2) 13 (8.0) 45 (27.8) 71 (43.8) 3 (1.9) 14 (8.6) 6 (3.7) 107 (66.0)

3 (4.0) 5 (6.7) 22 (29.3) 27 (36.0) 1 (1.3) 12 (16.0) 5 (6.7) 65 (86.7)

7 (8.0) 8 (9.2) 23 (26.4) 44 (50.6) 2 (2.3) 2 (2.3) 1 (1.1) 42 (48.3)

137 (84.0) 1 (0.6) 12 (7.4) 12 (7.4) 94 (58) 155 (95.7)

68 (90.7) 0 (0) 6 (8.0) 1 (1.3) 56 (74.7) 75 (100)

69 (79.3) 1 (1.1) 6 (6.9) 11 (12.6) 38 (93.7) 80 (92)

P-value

0.003