Quantitative analysis of hepatic steatosis in goose using quantitative CT / 中华放射学杂志

ABSTRACT

Objective:To evaluate the precision of quantitative CT (QCT) in measuring fat content of goose liver and to explore the influence of tube voltage on liver fat measurement.Methods:From January to March 2017, 22 Landes geese were selected to establish fatty liver models with overfeeding. The QCT was performed under 2 different tube voltages, 80 and 120 kV. In addition to the voltage, the other scanning parameters were the same. Three ROI were placed in the centre part of the left lobe, upper and lower half of the right lobe on the 3D reconstructed images. The volume percentage of goose liver fat (Fat% QCT) was calculated according to the formula provided by Mindways. Immediately after the QCT examinations, each goose was dissected, and the whole liver was removed. Then samples were taken in the corresponding areas of the ROI by QCT. The fat (triglyceride) content of the samples was measured by the biochemical extraction method. The Spearman correlation tests were used to determine the correlation between the QCT measurements of 80 kV and 120 kV groups and the results of biochemical extraction method. The Wilcoxon signed-rank test was performed to compare Fat% QCT between 80 kV and 120 kV groups. Bland-Altman plots were used to assess agreement between the two groups′ measurements. Results:The Fat% QCT of both 80 kV and 120 kV group positively correlated highly with biochemical extraction ( r s=0.936, P<0.001, r s=0.927, P<0.001) respectively. No statistically significant difference of Fat% QCT was detected between 80 kV and 120 kV group ( Z=-0.141, P=0.888). Bland-Altman analysis showed a small difference of Fat% QCT between the 80 kV and 120 kV group, in which mean values were 0.20%. Conclusions:Using biochemical extracted triglyceride as the reference, under the conditions of low-dose 80 kV and standard 120 kV tube voltages, QCT measurement of fat content is accurate and reliable in goose liver.