Duplex Ultrasound Surveillance Right Renal Artery Criteria for Thoracoabdominal Aortic Aneurysm Branched Endovascular Repair

Objective The objective of this study is to examine DUS criteria that best differentiate normal from abnormal flow indicative of future risk for hemodynamically significant stenosis and/or occlusion in individuals who are treated with the GORE® EXCLUDER® Thoracoabdominal Branch Endoprosthesis (TAMBE device; W.L. 78 Gore & Associates, Flagstaff, AZ) for endovascular repair of their complex aortic aneurysm. Methods This is a prospective, nonrandomized two-armed multi-center trial utilizing duplex ultrasound imaging (in addition to computed tomography imaging) to follow patients with the TAMBE device over a 24-month follow-up time period. The primary arm is comprised of individuals with extent IV thoracoabdominal aortic aneurysms, individuals with pararenal aortic aneurysms. The secondary arm is comprised of individuals with type I, II and III TAAAs. Logistic regression modelling was used to assess associations between a composite of right renal artery events and nine Doppler criteria (based on PSV and EDV values) considered to be associated with those events. Using the R2 statistic and likelihood ratio test, combinations of 2 or 3 criteria that were most strongly associated with right renal artery events were identified by a forward stepwise process. For each of these combinations, a risk score was calculated. Receiver operating characteristic curve analysis was used to evaluate 9 criteria and their combinations for associated risk for right renal artery event. These models were also applied to left renal artery events. Results Receiver operating characteristic curve analyses demonstrated that the model using the criteria; peak systolic velocity for any segment at index scan >= 190 or <= 40, systolic to diastolic ratio for the index scan at distal native renal artery, SD ratio (PSV/EDV) for the index scan at distal native vessel, and the minimum PSV for the index scan yielded the highest area under the curve for right renal artery events (area under the curve, 0.936). The risk score based on three criteria had a sensitivity of 100% for a score ≥ 0, with a specificity of .648 and sensitivity of 1.0 for the right renal artery. Conclusions In receiver operating characteristic curve analyses, several risk scores based on two or three of our nine criteria were highly associated with the occurrence of right renal artery stenosis and occlusion events in individuals post-procedure. Using any of these risk scores, individuals with a score ≥ 0 may warrant additional imaging and possible reintervention to maintain patency in stented renal arteries. Future work is needed to independently validate these models.