Quantitative analysis of liver standardized uptake value repeatability in SPECT/CT implications for clinical practice

Purpose. The aim of this study was to assess the repeatability of the SUV normalized by liver volume (SUVL) between two liver receptor SPECT/CT studies performed on different days in patients with ICG-R15 values within normal range.Methods. 935 patients who underwent liver receptor scintigraphy between January 2010 and August 2018 were included. Patients who underwent liver resection, hepatic arterial embolization or had ICG-R15 >10% between scans were excluded, and 38 patients were finally included in the analysis. The repeatability of SUVL between scans was assessed using the intraclass correlation coefficient (ICC) (1.1) between SUVLmax, SUVLpeak and SUVLmean at the first and second scan and the additive and proportional errors from the Bland-Altman analysis.Results. In ICC (1,1), SUVLmax, SUVLpeak and SUVLmean were all greater than 0.8, indicating almost perfect repeatability; neither additive nor proportional errors were observed in the Bland-Altman analysis.Conclusions. In patients with ICG-R15 values within the normal range, the SUV Liver (SUVL) between two liver receptor SPECT/CT studies performed on different days was repeatability over time. It was suggested that the SUVL of liver receptor scintigraphy could be an indicator that could be used for follow-up over time in the assessment of liver fibrosis.