The performance of duplex ultrasonography for the assessment of renal artery stenosis in Takayasu’s arteritis patients

Abstract

Abstract Background: This study intends to analyze the renal artery hemodynamic parameters in Takayasu’s arteritis (TAK) patients, so as to explore the diagnostic value of duplex ultrasonography in the involved renal artery of TAK patients. Methods: 115 TAK patients with 314 renal arteries were retrospectively analyzed, who were admitted to Peking Union Medical College Hospital from 2017 to 2022 and underwent both renal artery ultrasonography and angiography within 4 weeks. Seven ultrasonic parameters were compared among groups with different degrees (noninvolvement, <50% stenosis, 50%-69% stenosis, 70%-99% stenosis) of renal artery stenosis (RAS). Receiver operating characteristic curves (ROCs) were used to obtain the optimal threshold of renal artery PSV (RPSV), the renal-aortic PSV ratio (RAR), and renal-interlobar PSV ratio (RIR) for diagnosing different degrees of RAS in TAK patients. Results: There were statistically significant differences in RAR and RIR among the four groups (all P < 0.05), while there were no statistically significant differences in RPSV and AT between the moderate stenosis group (50%-69% stenosis) and the severe stenosis group (70%-99% stenosis). The discrimination of interlobar PSV(IPSV) and interlobar RI (IRI) was not significant, while IEDV was not statistically significant among the four groups. For TAK patients, the optimal thresholds of RPSV for diagnosing renal artery involvement (RAI), ≥ 50 % RAS and ≥ 70 % RAS were 143cm/s, 152cm/s and 183cm/s, respectively, with the sensitivities, specificities and accuracies all higher than 80%. The optimal thresholds of RIR for detecting RAI, ≥50% RAS and ≥70% RAS are 4.6, 5.6 and 6.4, respectively with satisfied diagnostic efficiencies. The areas under the curve (AUCs) of RPSV and RIR are 0.908 and 0.910 for diagnosing ≥ 50% RAS, and 0.876, 0.882 for diagnosing ≥ 70% RAS. When aortic PSV ≥ 140cm/s, RAR has an unsatisfied diagnostic efficiency and when aortic PSV < 140cm/s, RAR ≥ 2.2 can be used as the diagnostic threshold for ≥ 70 % RAS with a sensitivity, specificity, and overall accuracy of 84.00%, 89.93% and 89.08%, respectively. Conclusion: RPSV and RIR are still most valuable ultrasonic diagnostic parameters for RAS in TAK patients, but the optimal thresholds of them are different from the diagnostic criteria established based on atherosclerotic patients. For RAR, its diagnostic value should be judged according to the degree of aortic stenosis.