Effect of patient size on mean sterile water attenuation during multiphase CT examinations.

Abstract

The purpose of this study was to determine the variability of attenuation measurements in a water phantom included in the FOV during multiphase 64-MDCT. Ninety-seven consecutively registered patients undergoing multiphase kidney and liver protocol CT of the abdomen on the same 64-MDCT scanner had a sealed water bottle placed on their anterior abdomen during the examination. Region of interest (ROI) measurements of the mean attenuation (in HU) of the water bottle were made during the unenhanced, dynamic, and delayed phases of contrast enhancement. Generalized estimating equations were used to model mean attenuation in the ROI as a function of cross-sectional patient area, contrast phase, and protocol. Day of month and time of day were covariates. A phantom was created to model the patient study. The mean attenuation values in the water bottle ROIs were -13.1 HU for the kidney protocol and -9.1 HU for the liver protocol in the unenhanced phase, -11.7 HU for the kidney and -9.5 HU for the liver protocol in the dynamic phase, and -11.9 HU for the kidney and 11.0 HU for the liver protocol in the delayed phase. Kidney protocol water bottle ROI attenuation values were lower than the liver values (p = 0.04). In all phases with both protocols, the values differed from 0 HU (all p < 0.0001). Water bottle ROI attenuation decreased as patient cross-sectional area increased (-0.01 HU/cm(2), p < 0.0001). Three patients had absolute water bottle attenuation changes greater than 20 HU between phases. Day of the month (p = 0.08) and time of day (p = 0.93) were not significant factors. Phantom data supported the study findings. The mean attenuation of water decreased as patient diameter increased. Both artifactual enhancement and decrease in enhancement greater than 20 HU were found in three larger patients.