O Pellerin1, M Sapoval, L Trinquart, A Redheuil, A Azarine, G Chatellier, E Mousseaux. 1. Université Paris Descartes, Sorbonne Paris Cité, Faculté de Médecine, 75270 Paris cedex 06, France; Interventional Radiology Department, Assistance publique-Hôpitaux de Paris, Hôpital Européen Georges Pompidou, 20, rue Leblanc, 75908 Paris cedex 15, France. Electronic address: olivier.pellerin@egp.aphp.fr.
Abstract
PURPOSE: To compare the diagnostic performance of MDCTA versus renal angiography in the detection of>50% renal artery stenosis in patients suspected of reno-vascular hypertension. MATERIALS AND METHODS: Between January 2005 and January 2010, 92 MDCTA and renal arteriographies were retrospectively analysed. Renal angiographies were read by one interventional radiologist. Three blinded independent readers (two senior radiologists and one technician) scored MDCTA images using three different approaches. Reader 1 scored stenosis using only MPR and MIP. Reader 2 (technician) used only proprietary automatic arterial segmentation software. Reader 3 used the cited software, using manual diameter measurements. RESULTS: A total of 92 patients, (235 renal arteries) were assessed in which 48 significant stenosis were found by arteriography. Sensitivity, specificity, of MDCTA compared to renal arteriography were respectively per patient for reader 1: (88%; 80%); for reader 2: (58%; 80%); for reader 3: (96%; 90%) (P<.02). CONCLUSION: When using automated vessel analysis software edited by a radiologist, MDCTA studies had a Sensibility/Specificity of 96%/90% to detect>50% renal artery stenosis.
PURPOSE: To compare the diagnostic performance of MDCTA versus renal angiography in the detection of>50% renal artery stenosis in patients suspected of reno-vascular hypertension. MATERIALS AND METHODS: Between January 2005 and January 2010, 92 MDCTA and renal arteriographies were retrospectively analysed. Renal angiographies were read by one interventional radiologist. Three blinded independent readers (two senior radiologists and one technician) scored MDCTA images using three different approaches. Reader 1 scored stenosis using only MPR and MIP. Reader 2 (technician) used only proprietary automatic arterial segmentation software. Reader 3 used the cited software, using manual diameter measurements. RESULTS: A total of 92 patients, (235 renal arteries) were assessed in which 48 significant stenosis were found by arteriography. Sensitivity, specificity, of MDCTA compared to renal arteriography were respectively per patient for reader 1: (88%; 80%); for reader 2: (58%; 80%); for reader 3: (96%; 90%) (P<.02). CONCLUSION: When using automated vessel analysis software edited by a radiologist, MDCTA studies had a Sensibility/Specificity of 96%/90% to detect>50% renal artery stenosis.