Innovative postmarket device evaluation using a quality registry to monitor thoracic endovascular aortic repair in the treatment of aortic dissection

Objective: United States Food and Drug Administration (FDA)-mandated postapproval studies have long been a mainstay of the continued evaluation of high-riskmedical devices after initial marketing approval; however, these studies often present challenges related to patient/physician recruitment and retention. Retrospective single-center studies also do not fully represent the spectrum of real-world performance nor are they likely to have a sufficiently large enough sample size to detect important signals. In recent years, The FDA Center for Devices and Radiological Health has been promoting the development and use of patient registries to advance infrastructure and methodologies for medical device investigation. The FDA 2012 document, "Strengthening the National System for Medical Device Post-market Surveillance,"highlighted registries as a core foundational infrastructure when linked to other complementary data sources, including embedded unique device identification. The Vascular Quality Initiative (VQI) thoracic endovascular aortic repair for type B aortic dissection project is an innovative method of using quality improvement registries to meet the needs of device evaluation after market approval. Here we report the organization and background of this project and highlight the innovation facilitated by collaboration of physicians, the FDA, and device manufacturers. Methods: This effort used an existing national network of VQI participants to capture patients undergoing thoracic endovascular aortic repair for acute type B aortic dissection within a registry that aligns with standard practice and existing quality efforts. The VQI captures detailed patient, device, and procedural data for consecutive eligible cases under the auspices of a Patient Safety Organization (PSO). Patients were divided into a 5-year follow-up group (200 acute; 200 chronic dissections) and a 1-year follow-up group (100 acute; 100 chronic). The 5-year cohort required additional imaging details, and the 1-year group required standard VQI registry data entry. Results: The sample size of patients in each of the 5-year acute and chronic dissection arms was achieved <= 24 months of project initiation, and data capture for the 1-year follow-up group is also nearly complete. Data completeness and follow-up has been excellent, and the two FDA-approved devices for dissection are equally represented. Conclusions: Although the completeness of long-term follow-up is yet to be determined, the rapidity of data collection supports the use of this construct for device assessment after market approval. The alignment of this effort with routine clinical practice and ongoing quality improvement initiatives is critical and has required minimal additional effort by practitioners, thus facilitating patient inclusion. Importantly, the success and development of this unique project has helped inform FDA strategy for future device evaluation after market approval.