Reproducibility and influencing factors of 31P MR spectroscopy in rabbit liver with two-dimensional chemical shift imaging

Objective: To investigate the reproducibility and influencing factors of relative quantification of phosphorus metabolites with two-dimensional chemical shift imaging (2D CSI) in rabbit liver. Methods: Using 2D CSI MRS, 500 ml phosphate (NaH2PO4) solution phantom with 0.05 mol/L concentration and one healthy rabbit were scanned 30 times respectively in one day and rescanned 30 times in the next day, and the stability of MR scanner and reproducibility of within-run and between-days in the same individual were analyzed. Each of thirty rabbits was scanned and rescanned one time respectively in different days, and the reproducibility of between-days in one group was analyzed. The data were statistically analyzed with t tests. Results: (1) Phosphate solution phantom had a good reproducibility of within-run with the coefficient variation (CV) of 4.92% and 5.12% respectively in different two days. No significant change of phosphorus metabolites was detected in between-days, which was 16.68±0.82 and 16.56± 0.85 respectively (t=0.665, P>0.05). (2) The CV of metabolites in one healthy rabbit ranged from 8.04% to 34.13%. Among the metabolites, β-ATP had the best reproducibility with the CV less than 10%. PME was 0.88±0.28 and 0.88±0.30, PDE was 4.35±0.66 and 4.35±0.66, Pi was 0.95±0.30 and 0.97±0.28, α-ATP was 5.58±0.60 and 5.61±0.61, β-ATP was 2.70±0.22 and 2.71± 0.22, γ-ATP was 2.20±0.63 and 2.18±0.44 respectively, no significant changes of metabolites were detected in between-days (P>0.05). (3) The CV of metabolites in 30 healthy rabbits ranged from 8.48% to 36.21%. Among the metabolites, β-ATP had the best reproducibility with CV less than 10%. PME was 0.84±0.30 and 0.79±0.28, PDE was 4.29±0.72 and 3.94±0.84, Pi was 0.91±0.28 and 0.92± 0.31, α-ATP was 5.65±0.66 and 5.36±0.60, β-ATP was 2.71±0.23 and 2.66±0.25, γ-ATP was 2.07±0.29 and 1.99±0.37 respectively, no significant changes of metabolites were detected in between-days (P>0.05). Conclusions: The relative quantification of hepatic β-ATP may be most reliable among the phosphorus metabolites for rabbit liver because of its good reproducibility and small CV. The quantification of phosphorus metabolites by 31P MRS with 2D CSI in rabbit liver is affected by many factors. (authors)