Volume-Outcome Relationships in Surgical and Endovascular Repair of Aortic Dissection

Background. As surgical mortality decreases and endovascular utilization increases, it is unknown whether volume-outcome relationships exist in thoracic aortic dissection repair. We characterized volume-outcome relationships for surgical and endovascular management of thoracic aortic dissection. Methods. Patients aged more than 18 years undergoing repair of thoracic aortic dissection in the United States between 2010 and 2014 were identified in seven all-payer state inpatient administrative databases. Patients were divided into groups based on type of repair: surgical repair of type A dissection (TAAD), surgical repair of type B dissection (TBAD), and endovascular repair (TEVAR). Hierarchical logistic regression models evaluated the association between hospital volume and in-hospital mortality. Results. Overall in-hospital mortality rate was 13.4% (890 of 6650), highest after TAAD (463 of 2918, 15.9%), followed by TBAD (270 of 1934, 14.0%) and TEVAR (157 of 1798, 8.7%). Volume-outcome relationships for adjusted in-hospital mortality were demonstrated for TAAD and TBAD (P-trend < .001), but not TEVAR (P-trend = .11). Adjusted in-hospital mortality differed most for TAAD (fewer than 3 cases per year: 21%, 95% confidence interval, 18% to 24%; vs 11 or more cases per year: 12%, 95% confidence interval, 8% to 16%; P < .001) and TBAD (fewer than 2 cases per year: 18%, 95% confidence interval, 15% to 22%; vs 11 or more cases per year: 9%, 95% confidence interval, 5% to 12%; P < .001), whereas TEVAR did not differ between quartiles. Adjusted mortality was lower at centers with 26 or more overall annual thoracic dissection repairs, compared with any of the three lower-volume quartiles (P < .001). Conclusions. This study demonstrated lower mortality at high-volume hospitals for overall repair of aortic dissection, persisting separately for surgical repair of TAAD and TBAD, but not TEVAR. As endovascular technology advances and practice patterns consequently change, analyses should focus on understanding the balance between procedural volume, mortality, and access to care for thoracic aortic dissection. (C) 2019 by The Society of Thoracic Surgeons