Detectability of Renal Arteries in Living Renal Donors Using Multi-Detector CT: Comparison among Axial and Three-Dimensional Reconstruction Images

摘要

Purpose: The purpose was to compare the effectiveness of MDCT reconstruction methods that improve detectability for detecting the renal arteries in patients undergoing laparoscopic nephrectomy irrespective of the ability of the radiologist. Methods: 128 patients with left nephrectomy and either one or multiple renal arteries were enrolled. Radiologists with varying levels of expertise (5 board-certified and 3 non-certified) interpreted all MDCT images, including axial, VR and MIP images, independently and determined the numbers of renal arteries. The interpretation times, sensitivity, specificity and diagnostic accuracy rates were analyzed. A receiver operating characteristic analysis was used to compare the results from the various reconstruction methods. Results: At surgery, there were 156 arteries: 101 donors had one renal artery and 27 had multiple arteries. The interpretation time for board-certified radiologists was shorter than that for non-certified radiologists, and the interpretation time using MIP images was significantly shorter than that of using the other reconstruction methods. The sensitivity, specificity and diagnostic accuracy on axial images were significantly higher than those on other images. The average Az value for detection of the numbers of renal arteries using axial images was higher than that of using other images. Conclusion: MDCT is helpful in reviewing the numbers of donated renal arteries for radiologists with different levels of expertise, and different types of reconstructions.