91 - Doppler Ultrasound is Insufficient for Identifying Renal Artery Stenosis in Williams Syndrome

Poster Number: 91
Publication Number: 91.303

Elisa McCarthy, Lucile Packard Children's Hospital Stanford/Stanford University School of Medici - - Palo Alto, CA, Palo Alto, CA, United States; Kirstie Lechich, Lucile Packard Children's Hospital Stanford, Palo Alto, CA, United States; Claudia Algaze, Lucile Packard Children's Hospital Stanford, Palo Alto, CA, United States; Gregory T. Adamson, Stanford University School of Medicine, Palo Alto, CA, United States; Ronnie T. Collins, Stanford University School of Medicine, Palo Alto, CA, United States

Background: Cardiovascular abnormalities occur in 80% of patients with Williams syndrome (WS), the majority comprised of arterial stenoses. Renal artery stenosis (RAS) occurs in up to 60% of patients with WS. Renal ultrasound with Doppler (US) has been recommended for evaluation of RAS in WS.

Objective: We sought to determine the value of US in assessing RAS in WS.

Design/Methods: We identified all patients with genetically-confirmed WS at our center who had undergone both US and abdominal angiography via CT, MRI, or cardiac catheterization. We reviewed all reports and images from available studies to determine the presence of RAS. Stenosis was defined as present on US based on published flow velocity thresholds, as well as arterial dimensions. Stenosis was defined as present on angiographic studies when there was ≥15% luminal reduction relative to a neighboring segment, with specific attention to the renal artery ostium compared to the mid-portion of the artery. We summarized the data using frequencies and percentages, or medians and interquartile ranges (IQR), as appropriate and determined the sensitivity, specificity, and predictive values of US for RAS.

Results: We identified 55 patients with WS who had undergone US. Of those, 26 (47%) had additional imaging and constituted the study cohort. There were 17 males (65%). Median age at time of US was 2.0 years [IQR: 1.3, 10.2]. Additional renal artery imaging included 14 CTs (54%), 10 catheterizations (38%), and 2 MRI (8%). Interval between US and additional imaging was 5.5 months [IQR: 0.6, 12.6]. RAS was not identified on any US, including in the 29 patients who did not have angiography. RAS was present in 58% (15/26) of the angiographic studies: 6/14 CTs (43%), 7/10 catheterizations (70%), and on both MRI studies. The positive predictive value of US for identifying RAS was 0, and the negative predictive value was 42%.Conclusion(s): US is insufficient to identify RAS in WS and is likely to provide false reassurance. Given it can be performed without sedation, CT angiography is a better first-line modality for assessing RAS in WS.