Learning curve of fenestrated and branched endovascular aortic repair for pararenal and thoracoabdominal aneurysms

Objective: The objective of this study was to review the learning curve for fenestrated-branched endovascular aortic repair (F-BEVAR) of pararenal and thoracoabdominal aortic aneurysms (TAAAs). Methods: We reviewed the clinical data of 334 consecutive patients (255 males, mean age 75 ± 7 years) who underwent F-BEVAR between 2007 and 2016 in a single institution. Outcomes were analyzed in four quartiles of experience (Q1-Q4). Study outcomes included trends in patient characteristics, device design, procedural variables, 30-day mortality, and major adverse events (MAEs). Results: There were 178 patients (53%) treated for pararenal aneurysms and 156 (47%) for TAAAs. During the study period, there was a statistically significant increase in the proportion of TAAAs and in the number of vessels incorporated. Despite this, there was a steady decrease in 30-day mortality (6% in Q1 to 0% in Q4; P <.04) and in the rate of MAEs (60% in Q1 to 29% in Q4; P<.001). By linear regression analysis, there was significant decline in estimated blood loss (1358 ± 1517 mL in Q1 to 486 ± 520 mL in Q4; P <.001), total operating time (325 ± 116 minutes in Q1 to 248 ± 92 minutes in Q4; P <.001), total fluoroscopy time (121 ± 59 minutes in Q1 to 85 ± 39 minutes in Q4; P <.001), contrast volume (201 ± 92 mL in Q1 to 160 ± 61 mL in Q4; P =.002), and radiation dose (4141 ± 2570 mGy in Q2 to 2543 ± 1895 mGy in Q4; P <.001). Independent predictors of MAEs were total operating time (odds ratio [OR], 1.6; 95% confidence interval [CI], 1.3-1.8; P <.001), Society for Vascular Surgery total score (OR, 1.1; 95% CI, 1.02-1.2; P =.009), and quartile 1 (OR, 3.0; 95% CI, 1.7-5.2; P <.001). Conclusions: This study demonstrates significant improvement in perioperative mortality, MAEs, procedural variables, and secondary interventions in patients treated by F-BEVAR, despite the increase in complexity of aneurysm pathology during the study period. Also, better patient selection contributed to improve outcomes.