Thoracic aortic surgery enters the era of big data

From the earliest days of cardiovascular surgery, the medical community and the general public were aware of both the significant risks posed by our surgical procedures and the variability in their outcome [1]. Cardiovascular surgeons responded to these concerns by developing clear data definitions, risk models, outcomes assessment systems and eventually open public reporting of key index operations to diminish outcome variation and maintain high-quality standards. This process, led in the USA by the Society of Thoracic Surgeons (STS), initially started with the development of a robust model for predicting coronary artery bypass surgery mortality [2, 3]. Over many iterations, the STS database has been expanded to include a wide variety of cardiovascular, thoracic and congenital heart surgery procedures with custom-designed risk modelling strategies for various procedures of interest. Quality performance metrics for various procedures have also been developed. Although voluntary, the STS Adult Cardiac Surgery Database (ACSD) captures more than 90% of all cardiovascular surgery in the USA and, since 2011, has had a growing tally of international sites [4]. As part of the mission of the STS-ACSD to be a source of high-quality clinical data, a robust auditing function ensures sites are compliant with reporting standards [5]. Aortic arch operations are uniquely risky procedures as they frequently involve diseased aortic arch tissue, require significant manipulation of the cerebral vessels and interruption or alteration of the cerebral circulation. Thoracic aortic operations have rates of death, stroke and renal failure that are orders of magnitude higher than standard operations like coronary artery bypass grafting and aortic valve replacement. Despite this, there is a paucity of large database information on outcomes of aortic operations or even how and why they are being performed. There are several volunteer registries including the International Registry of Aortic Dissection and the German Registry for Acute Aortic Dissection Type A, which focus on aortic dissection, and some limited aortic data are also included in the German Society for Thoracic and Cardiovascular Surgery Registry, Japan Adult Cardiovascular Surgery Database and other national registries. The STS recognized the importance of thoracic aortic surgery and has made iterative improvements in the detail of aortic operations in the STS-ACSD. This process started with the v2.73 modification of the STS-ACSD in July 2011, with further additions performed in July 2014 with v2.81 [6]. With improvements in technology for endovascular repair into the aortic arch and even the ascending aorta, fully understanding the landscape of aortic procedures and their outcomes has become a significant priority. To this end, as of 1 July 2017, the STS-ACSD has undergone a massive upgrade of the aortic surgery component to fully reflect the complexity of open and endovascular thoracic aortic surgery. STS-ACSD v2.9 will capture comprehensive data on every aspect of aortic surgery from the root to the iliac with full phenotypic, anatomic and interventional details. The database can accommodate the full variety of open, endovascular and concomitant or staged hybrid procedures. It is uniquely developed to be able to compare open and endovascular techniques for treating similar disease states to ensure that new technology can be adopted safely. Although we await high-quality data from the new database, we can still learn important insights regarding aortic surgery from the older versions. The current study by Englum et al. [7] is an excellent example of how large volumes of data in aortic surgery can help guide clinical practice. This article examines the role of different cerebral perfusion strategies and temperatures on the outcomes of aortic arch surgery. Although the data are already 2 iterations past in the STS-ACSD, the authors present several important findings. First, the volume of aortic arch surgery is rapidly increasing in the USA and is performed in a wide variety of practice settings with a median volume of only 2 cases per 6-month interval. The overall rates of mortality (12%), stroke (8%) and renal failure (9%) are high and are reflective of disease burden of these patients and the technical complexity of these operations. The high rates of these most severe complications provide further evidence that finding methods to improve the outcomes of aortic surgery deserves further attention and resources. The most important finding of this study was that performing aortic arch surgery without the use of adjunctive cerebral protection methods was associated with significantly worse rates of the composite end-point of mortality or poor neurologic outcomes, even when performed with deep/profound hypothermia. This finding is in the context of a ‘straight hypothermic circulatory arrest’ strategy being used in 25% of all patients and was the single most commonly used strategy. Higher volume centres appear to be more likely to use some form of adjunctive perfusion, which is typically the standard practice in Europe and Asia [8, 9]. While the intricacies of which type of cerebral perfusion is superior remains undefined, the cardiovascular surgery community in the USA should embrace these findings and progress away from straight circulatory arrest as a primary strategy in aortic arch cases. Further details regarding the anatomic complexity of arch procedures, as well as the safety of comparative perfusion strategies, and the use of hybrid procedures will be forthcoming as STS-ACSD v2.9 rolls out and collects even more detailed data. Although arch surgery was first performed with cardiopulmonary bypass and hypothermic circulatory arrest over 50 years ago [10], until recently, our understanding of this procedure was limited to case reports and predominantly single-centre case series. With the evolution of high-quality clinical databases to study procedures such as coronary artery bypass grafting, aortic valve replacement, transcatheter aortic valve replacement and percutaneous coronary intervention, thoracic aortic surgery is just now entering the modern era of high accountability through the big data revolution. With the rapid emergence of endovascular approaches to the arch and ascending aorta, data from the STS-ACSD will continue to guide clinical aortic practice for future generations.