MP12-17 VARIABILITY OF PIXEL BRIGHTNESS INTENSITY CAN BE USED TO FURTHER CHARACTERIZE A STONE’S COMPOSITION

Abstract

You have accessJournal of UrologyImaging/Radiology III1 Apr 2014MP12-17 VARIABILITY OF PIXEL BRIGHTNESS INTENSITY CAN BE USED TO FURTHER CHARACTERIZE A STONE’S COMPOSITION Yaniv Larish, Leonard Glickman, and Jay A. Motola Yaniv LarishYaniv Larish More articles by this author , Leonard GlickmanLeonard Glickman More articles by this author , and Jay A. MotolaJay A. Motola More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2014.02.452AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Non-contrast CT (NCCT) is the gold standard of imaging in the diagnosis of urinary calculi. Composition of stones can be determined based on a stone’s pixel brightness intensity as measured by mean Hounsfield units (mHU) with varying degrees of certitude. In this study, we sought to evaluate the variability of pixel intensity by measuring the standard deviation of HU values (HUsd) within specific stone types to improve the accuracy of determining a stone’s composition. METHODS A single-institution, retrospective, chart review was performed from June 2009 to July 2013 identifying patients with GU calculi. All patients had a NCCT scan of the abdomen and pelvis which confirmed the presence of a GU stone. All calculi were treated with endoscopic management during which stone specimen was retrieved for analysis. Stone composition was determined by infrared spectroscopy. CT scans were reviewed by two urologists who were blinded to the stone’s chemical composition. The radiographic characteristics of the stone, including mHU and HUsd were measured. RESULTS 155 patients with 191 distinct stone episodes were identified. Patients with multiple stones in a single renal unit were excluded from analysis. The mean time between CT scan and surgery was 27 days. 72%, 17%, 6%, and 5% of the stones were composed of calcium oxalate (CaOx), uric acid (UrA), calcium phosphate (CP), and cystine (CY), respectively. Inter-rater reliability demonstrated good agreement between the two reviewers (r=0.89 and 0.74 for mHU and HUsd respectively). The mHU values were 697.78, 425.83, 706.86, and 634.67 for CaOx, UrA, CP, and CY, respectively. There was a significant difference when comparing UrA to each of the other three stone types: CaOx, CP, and CY (p<.0001, 0.0028, 0.0482, respectively). The HUsd values were 126.22, 61.56, 128.22, and 60.16 for CaOx, UrA, CP, and CY, respectively. There was a significant difference comparing UrA to CaOx and to CP (p<.0001, 0.0008, respectively) and a significant difference comparing CY to CaOx and to CP (p= 0.0001 and 0.0034 respectively). CONCLUSIONS The HU information obtained from CT scans has been limited to the mHU number. The HUsd is a previously ignored and readily obtainable variable that should be used to increase the clinician’s confidence in determining a stone’s composition and guiding treatment. © 2014FiguresReferencesRelatedDetails Volume 191Issue 4SApril 2014Page: e112 Advertisement Copyright & Permissions© 2014MetricsAuthor Information Yaniv Larish More articles by this author Leonard Glickman More articles by this author Jay A. Motola More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...