Quantitative assessment of renal and perirenal adipose tissue distribution at 5 T: a feasibility study

TL;DR

This study shows that a new MRI technique called FACT can accurately measure fat around the kidneys at 5 T, offering a non-invasive alternative to biopsies.

Contribution

The study demonstrates the feasibility and accuracy of the FACT technique for quantifying renal and perirenal fat at ultra-high field strength (5 T).

Findings

FACT-PDFF showed excellent agreement with MRS-PDFF in phantom and clinical studies.

FACT images achieved high quality and excellent repeatability in volunteers.

Renal fat distribution varied significantly between kidneys, genders, and anatomical regions.

Abstract

To evaluate the feasibility and accuracy of Fat Analysis Calculation Technique (FACT), a multi-echo Dixon-like sequence, for quantifying renal and perirenal adipose distribution at 5 T.

Accuracy of FACT-based Proton density fat fraction (FACT-PDFF) was assessed by comparing with magnetic resonance spectroscopy-based PDFF (MRS-PDFF) in phantom study. In vivo FACT images from 24 volunteers (13 males and 11 females) without kidney-related diseases were acquired at 5 T and evaluated independently by two readers. Repeatability of FACT-PDFF was assessed through three consecutive scans. Spearman correlation examined associations between averaged FACT-PDFF and clinical characteristics. Linear regression, intraclass correlation coefficients (ICCs), and Bland-Altman plots assessed consistency and deviations between fat quantification methods and field strengths. The Wilcoxon signed-rank test compared image quality scores between radiologists. The paired t test compared FACT-PDFF differences across all regions of interest between bilateral kidneys and between renal cortex and medulla. Analysis of covariance compared gender-related renal fat differences.

In phantom study, FACT-PDFF showed excellent agreement with MRS-PDFF at both fields (ICCs ≥ 0.995). Linear regression revealed strong correlations (R² ≥ 0.998), and Bland-Altman plots indicated minimal bias. In clinical study, FACT images achieved high quality. Repeatability was excellent (ICCs: 0.837–0.991; CVs: 0.78–4.49%). Significant PDFF differences existed between bilateral kidneys (cortex/medulla: P < 0.001; sinus fat: P = 0.003), cortex vs. medulla (P < 0.001), and genders (right cortex, left medulla, and left perirenal fat: P ≤ 0.044). PDFF was correlated positively with age, weight, body mass index, waist/hip circumference, and waist-to-height ratio (r = 0.412–0.797, P ≤ 0.046).

FACT at 5 T reliably quantifies renal and perirenal adipose distribution. It can offer a non-invasive alternative to biopsy, and may facilitate the understanding of renal adipose distribution and its clinical associations.

The online version contains supplementary material available at 10.1186/s12880-025-02136-8.

Figures