PURPOSE

Liver cirrhosis is an abnormal condition in which liver experiences irreversible scarring. Whole-liver T1rho MR imaging is feasible to detect and assess human liver cirrhosis ^[1]. T1rho values in the liver may be heterogeneous. ROI-based analysis may suffer from sampling errors due to the user-defined placement of the ROI and the choice of slice location. The aim of our study was to find an approach that allow the detection and measurement of the T1rho values of the hepatic segments.

METHODS

MR Experiment Fifteen healthy volunteers (mean age 38 years. range 23–63 years, 7F/8M) without a history of liver diseases were recruited for the study conducted. The MRI experiment was conducted using a Philips 3.0T clinical scanner (Achieva TX, Best, the Netherlands). A 16-channel SENSE XL Torso coil was used for signal reception, using a 3D whole-liver coverage sequence. Pulse Sequence and Imaging Parameters The basic sequence consisted of a self-compensated spin-locking T1rho preparation pulse and was subsequently phase and frequency encoded for image acquisition using a standard alternating phase turbo field echo (TFE) acquisition. The principle of selective excitation technique was used for fat suppression. The other specific parameters for the whole-liver T1rho sequence were as our previous study ^[2]. Image Analysis ROI was positioned in each of the eight functionally independent segments divided according to Couinaud. T1rho of each segment were described as mean ± standard deviation (SD). Statistical Analysis in the reproducibility analysis, for each segment, as well as for the mean value, the variability between two different analyses was evaluated by calculating the inter-class correlation coefficient (ICC). One-way repeated measures analysis of variance (ANOVA) was used to evaluate whether there was a significant difference between T1rho values in different hepatic segments.

RESULTS

An excellent ICC (0.937) was obtained. The mean hepatic T1rho of all the healthy volunteers is 40.46 ± 4.55ms. Each segment T1rho distribution was shown in figure 1. The ANOVA test showed no significant differences in the segmental T1rho value (P>0.05). Table 1 showed the percentage of deviation of the hepatic segments from the mean T1rho ranged from -4.55% (segment VI) to 3.04% (segment IV). Coefficient correlations, the correspondent P values were also shown in Table 1.

DISCUSSION

Hepatic macro-molecules accumulation in the extracellular matrix can be heterogeneous across the liver ^[3]. The bias of T1rho across all hepatic segments should be studied for the further explore the heterogeneity of liver cirrhosis patients. No find of the segmental variability of the T1rho in our healthy volunteers study indicated that it might be potential bio-marker for the evaluation of heterogeneity deposition in liver cirrhosis patients.

CONCLUSION

In conclusion, there is no significant segmental variability of the hepatic T1rho across eight functionally independent segments.