Duplex ultrasound surveillance of renal branch grafts after fenestrated endovascular aneurysm repair

Objective: The use of duplex ultrasound (DUS) examinations for surveillance after fenestrated endovascular aneurysm repair (FEVAR) is not well-studied. Our objective was to further characterize normal and abnormal duplex findings in renal branch grafts after FEVAR. Methods: We retrospectively reviewed a single-center experience involving consecutive patients treated with Cook ZFEN devices between 2012 and 2017. Postoperative imaging consisted of a computed tomography (CT) scan at 1 month, 6 months, 1 year, and annually thereafter. As experienced progressed, DUS examination with or without concurrent CT scans were obtained in a nonstandardized protocol, particularly for patients with decreased renal function. Renal patency loss was defined as occlusion or stenosis of greater than 50% evaluated on 3-day renal artery center-line imaging. Results: A total of 116 patients were treated with FEVAR, of which 60 (51.7%) had concurrent CT and renal DUS images available for review. Six patients (10%) had limited ultrasound studies owing to bowel gas and were excluded. The study cohort therefore included 54 patients receiving of 94 renal fenestrated stents with a mean follow-up of 23 months. Twelve cases of renal patency loss in 10 patients (9 stenoses, 3 occlusions) were found on CT scanning, 11 (91.6%) of which had concurrent abnormalities found on ultrasound examination. Stents with compression at the junction of the main body exhibited significantly elevated mean Peak systolic velocities (PSV) compared with nonstenosed stents (349.2 cm/s vs 115.3 cm/s; P = .003). Stenosis in the most proximal portion of the stent (ie, within the main body) showed no difference in proximal PSV (86.0 cm/s vs 131.9 cm/s; P = .257); however, dampened PSV showed significant differences in the mid (17.5 cm/s vs 109.9 cm/s; P = .027) and distal (19.0 cm/s vs 78.3 cm/s; P = .028) segments compared with nonstenosed stents. All occluded stents demonstrated no flow detection. Proximal PSV served as a strong classifier for junctional stenosis (area under the curve, 0.98). A combined criterion of proximal PSV of greater than 215 cm/s or distal PSV of less than 25 cm/s resulted in a sensitivity of 91.6% and specificity of 85.3% for detecting patency loss. All stents that were compromised underwent successful secondary reintervention and restoration of patency. Conclusions: DUS imaging is a clinically useful modality for surveillance of renal branch grafts after FEVAR. Patterns of segmental velocity elevation (proximal PSV, >215 cm/s) and dampening in the distal renal indicate potential hemodynamic compromise and should prompt more aggressive workup or imaging and likely be considered for secondary intervention.