Utility of Ultrasound-Guided Attenuation Parameter for the Quantification of Steatosis with Reference to MRI-Based Proton Density Fat Fraction: A Multicenter, Prospective Study

Background: Ultrasound-guided attenuation parameter (UGAP) is a tool recently developed for noninvasive evaluation of hepatic steatosis. However, there are still few reports on its usefulness in clinical practice. This prospective multicenter study analyzed the diagnostic accuracy of grading hepatic steatosis with reference to magnetic resonance imaging-based proton density fat fraction (MRI-PDFF) in a large cohort of patients with chronic liver disease. Methods: A total of 1010 patients with chronic liver disease were prospectively enrolled from six Japanese liver centers. MRI-PDFF and UGAP values were analyzed to examine whether the distribution was normal using a normal probability plot. Linearity was evaluated using intra-class correlation coefficients (ICCs) between MRI-PDFF and UGAP values. Bias, defined as the mean difference between MRI-PDFF and UGAP values, was assessed by Bland–Altman analysis. UGAP cutoffs for pairwise MRI-PDFF-based hepatic steatosis grade were determined using area under the receiver operating characteristics curve (AUROC) analyses. Findings: UGAP values were shown to be normally distributed. In contrast, since PDFF values were not normally distributed, they were log-transformed (MRI-logPDFF). UGAP values were significantly correlated with MRI-logPDFF (ICC=0·768). Additionally, Bland–Altman analysis showed good agreement between MRI-logPDFF and UGAP with a mean bias of 0·0002% and a narrow range of agreement (95% CI: -0·015–0·015). The AUROCs for distinguishing ≥S1 (MRI-PDFF>5·2%), ≥S2 (MRI-PDFF>11·3%), and S3 (MRI-PDFF>17·1%) were 0·910 (95% CI: 0·891–0·928), 0·912 (95% CI: 0·894–0·929), and 0·894 (95% CI: 0·873–0·916), respectively. Clinical Trial: This study was registered in the UMIN Clinical Trials Registry (UMIN000041196). Interpretation: UGAP has an excellent diagnostic accuracy for grading hepatic steatosis with reference to MRI-PDFF. Additionally, UGAP has good linearity and negligible bias, suggesting that UGAP has excellent technical performance characteristics that can be widely used in clinical trials and patient care. Funding: GE Healthcare and JSPS KAKENHI [grant number JP 21K07987]. Declaration of Interest: Kento Imajo, Hidenori Toyoda, Satoshi Yasuda, Yasuaki Suzuki, Katsutoshi Sugimoto, Hidekatsu Kuroda, Tomoyuki Akita, Junko Tanaka, Yutaka Yasui, Masayuki Kurosaki, Namiki Izumi, Atsushi Nakajima, and Takashi Kumada have no competing interests to disclose. Ethical Approval: This study was approved by the Ogaki Municipal Hospital and by the local research ethics committee at each center.