Multicentre analysis of current strategies and outcomes in open aortic arch surgery: heterogeneity is still an issue.

Objectives: The study was conducted to evaluate, on the basis of a multicentre analysis, current results of elective open aortic arch surgery performed during the last decade.
Methods: Data of 1232 consecutive patients who underwent aortic arch repair with reimplantation of at least one supra-aortic artery between 2004 and 2013 were collected from 11 European cardiovascular centres, and retrospective statistical examination was performed using uni- and multi-variable analyses to identify predictors for 30-day mortality. Acute aortic dissections and arch surgeries not involving the supra-aortic arteries were not included.
Results: Arch repair involving all 3 arch arteries (total), 2 arch arteries (subtotal) or 1 arch artery (partial) was performed in 956 (77.6%), 155 (12.6%) and 121 (9.8%) patients, respectively. The patients' characteristics as well as the surgical techniques, including the method of cannulation, perfusion and protection, varied considerably between the clinics participating in the study. The in-hospital and 30-day mortality rates were 11.4 and 8.8% for the entire cohort, respectively, ranging between 1.7 and 19.0% in the surgical centres. Multivariable logistic regression analysis identified surgical centre, patient's age, number of previous surgeries with sternotomy and concomitant surgeries as independent risk factors of 30-day mortality. The follow-up of the study group was 96.5% complete with an overall follow-up duration of 3.3 ± 2.9 years, resulting in 4020 patient-years. After hospital discharge, 176 (14.3%) patients died, yielding an overall mortality rate of 25.6%. The actuarial survival after 5 and 8 years was 72.0 ± 1.5% and 64.0 ± 2.0, respectively.
Conclusions: The surgical risk in elective aortic arch surgery has remained high during the last decade despite the advance in surgical techniques. However, the patients' characteristics, numbers of surgeries, the techniques and the results varied considerably among the centres. The incompleteness of data gathered retrospectively was not effective enough to determine advantages of particular cannulation, perfusion, protection or surgical techniques; and therefore, we strongly recommend further prospective multicentre studies, preferably registries, in which all relevant data have to be clearly defined and collected.
(© The Author 2016. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.)