Abstract
PURPOSE: Calculation of extracellular volume fraction (ECV) currently receives increasing interest as a potential biomarker for non-invasive assessment of liver fibrosis. ECV calculation requires hematocrit (Hct) sampling, which might be difficult to obtain in a high-throughput radiology department. The aim of this study was to generate synthetic ECV for hepatic applications without the need for Hct sampling. METHODS: In this prospective study participants underwent liver MRI. T1 mapping was performed before and after contrast administration. Blood Hct was obtained prior to MRI. We hypothesized that the relationship between Hct and longitudinal relaxation rate of blood (R1 = 1/T1(blood)) could be calibrated and used to generate the equation for synthetic Htc and ECV calculation. Conventional and synthetic ECV were calculated. Pearson correlation, linear regression and Bland-Altman method were used for statistical analysis. RESULTS: 180 consecutive patients were divided into derivation (n = 90) and validation (n = 90) cohorts. In the derivation cohort, native R1(blood) and Hct showed a linear relationship (Hct(MOLLI) = 98.04 × (1/T1(blood)) - 33.17, R(2) = 0.75, P < 0.001), which was used to calculate synthetic ECV in the validation and whole study cohorts. Synthetic and conventional ECV showed significant correlations in the derivation, validation and in the whole study cohorts (r = 0.99, 0.97 and 0.99, respectively, P < 0.001, respectively) with minimal bias according to the Bland-Altman analysis. CONCLUSION: Synthetic ECV seems to offer an alternative method for non-invasive quantification of the hepatic ECV. It may potentially overcome an important barrier to clinical implementation of ECV and thus, enable broader use of hepatic ECV in routine clinical practice.