Comprehensive Learning Curve Analysis of a Long-Term Experience With Thoracic Endovascular Aneurysm Repair.

Purpose: To analyze the learning curve for thoracic endovascular aortic repair (TEVAR) in a single center over a period of 25 years.

Materials and methods: In total, 390 consecutive standard TEVAR procedures undertaken between 1996 and 2021 were included in a retrospective, observational, single-center study. Cumulative sum charts were elaborated for the entire center experience (primary outcome) as well as for the first and second implanting physicians. Data on procedural variables (contrast volume, operative and fluoroscopy time), 30-day major adverse events (MAEs) and clinical success, and endoleak and reintervention rates were secondary outcomes and subdivided into 4 quartiles of experience (Q1-Q4) or presented as first 2 versus latest 2 quartiles (Q1-Q2 vs Q3-Q4).

Results: The mean follow-up was 4.3±4.0 years. The center's learning curve was achieved after 75 procedures, and it was similar for the first implanting physician. The surgeon coming thereafter had a significantly shorter curve (10 TEVARs). Comparing Q1-Q2 with Q3-Q4, 30-day MAEs (16.1 vs 11.3%, p=0.164), 30-day mortality (11.4% vs 3.6%, p=0.003), and intraoperative additional maneuvers (21.5% vs 13.3%, p=0.033) were reduced along with an improvement in clinical success (85.9% vs 90.3%, p=0.190). From Q1 to Q4, operative time (139.8±65.5 to 76.7±43.7 min, p=0.001), fluoroscopy time (15.1±8.8 to 7.1±5.1 min, p<0.001), and contrast volume (244.0±112.1 to 104.3±46.1 mL, p<0.001) showed a considerable reduction. Late endoleak and aortic-related mortality declined significantly from Q1-Q2 to Q3-Q4 (24.1% to 15.5%, p=0.033 and 18.6% vs 8.2%, p=0.006, respectively). Operative time (p=0.021), contrast volume (p=0.016), and fluoroscopy time (p=0.004) were independent risk factors for endoleak, causing a 1.3-fold risk increase for both each 60 minutes of additional operative time (p=0.021) and every 100 mL of additional contrast medium (p=0.016). Each 10-minute increase in fluoroscopy time determined a 1.4-fold risk increment (p=0.004).

Conclusion: The learning curve shortened significantly over time with non-negligible clinical outcome improvements, suggesting that specific endovascular training is mandatory to become an effective TEVAR performer.

Clinical impact: For the first time in literature, the standard TEVAR's learning curve has been evaluated at a single vascular surgery center over a period of 25 years. The learning curve for the center and the first physician historically undertaking TEVAR was achieved at the 75th treated patient. The learning curve of the surgeons coming thereafter was significantly shorter (10 cases). This quarter-century demonstrated that intraoperative learning-related variables were associated with long-term clinical outcomes and all have improved over time. Centers approaching TEVAR for the first time and training program providers could use these data to aim to offer better clinical outcomes.