Quantification of hepatic macrosteatosis in living, related liver donors using T1‐independent, T2*‐corrected chemical shift MRI

Abstract

IDEAL is a useful tool for the preoperative diagnosis of HS in potential living liver donors; it can also help to avoid unnecessary biopsies in these patients.

Diagnostic Performance of a Rapid Magnetic Resonance Imaging Method of Measuring Hepatic Steatosis

“…Our results compare favourably to the small number of studies that have used a pathologist’s visual estimate of fat content as the reference standard to define thresholds and assess the diagnostic performance of comparable MRI approaches [7], [9], [15], [22], [23], [26], [27] (see Table S2 in File S1). The MRI technique presented here has higher areas under ROC curves for all liver fat thresholds described in the literature (Table S2 in File S1), except for two studies [7], [15] investigating liver fat in non-diseased cohorts (potential living donor candidates). At 5% or more fat, our AUC was marginally lower (0.962 versus 0.987) than the result from Joe et al [7] and, at 30% or more fat, Lee et al [15] reported an AUC of 0.995 compared to our result of 0.992.…”

“…While several reports have compared histological liver fat estimates with quantitative MRI approaches [7]–[25], there are fewer studies [7], [9], [15], [22], [23], [26], [27] which have assessed the diagnostic performance of MRI using a pathologist’s visual estimate of fat content as the reference standard to identify the clinically relevant thresholds as defined by the Nonalcoholic Steatosis Clinical Research Network (NASH CRN) [2], [28]. The largest of these diagnostic studies [15] investigated a cohort of potential living donors without any documented liver disease.…”

Quantification of the Fat Fraction in the Liver Using Dual-Energy Computed Tomography and Multimaterial Decomposition

“…Previous studies demonstrated that measurement of PDFFs using IDEAL IQ sequence improved the reliability of fat quantification by correcting for confounders such as T2* decay, T1 bias, and noise bias and for the multispectral complexity of fat. 19,40 Image Analysis of CT and MRI for Fat Quantification…”

“…A bandage for pressing on the abdomen of the animals was used in order to reduce respiratory motion artifacts. The iterative decomposition of water and fat using echo-asymmetry and the least-squares estimation quantitative sequence (IDEAL IQ; GE Healthcare) 19,40 were obtained to estimate the proton density fat fraction (PDFF) map in the axial plane using the following imaging parameters: repetition time, 16.2 milliseconds; first time to echo, 1.8 milliseconds; echo spacing 2.1 milliseconds; echo train length, 6; flip angle, 8 degrees; number of excitations, 2; slice thickness, 2 mm; matrix, 256 Â 160; and field of view, 230 mm. IDEAL IQ produced T2*-corrected water, T2*-corrected fat, R2* maps, and PDFF maps.…”

“…13 Therefore, an alternative, noninvasive means of diagnosing hepatic steatosis such as ultrasonography, magnetic resonance imaging (MRI), and computed tomography (CT) have been used for diagnosing fatty liver. [14][15][16][17][18][19] Recent studies have demonstrated that the hepatic fat fraction (HFF) calculated from proton densities determined on magnetic resonance (MR) spectroscopy is equivalent to the tissue triglyceride concentration. 13,20 In addition, an approach of T1-independent, T2*-corrected, chemical shift-based fat-water separation with multipeak fat spectral modeling is a highly accurate method for evaluation of fatty liver disease.…”