Using MRI to non-invasively and accurately quantify preoperative hepatic steatosis. Issue 8 (August 2017)

• Record Type: Journal Article
• Authors: Clarke, Callisia N.; Choi, Haesun; Hou, Ping; Davis, Catherine H.; Ma, Jingfei; Rashid, Asif; Vauthey, Jean-Nicolas; Aloia, Thomas A.
• Is Part Of: HPB. Volume 19:Issue 8(2017)
• Page Start: 706
• Page End: 712
• Publication Date: 2017-08
• Subjects: Liver -- Diseases -- Periodicals; Biliary tract -- Diseases -- Periodicals; Pancreas -- Diseases -- Periodicals; 616.362005
• DOI: 10.1016/j.hpb.2017.04.009
• Languages: English
• ISSNs: 1365-182X

Abstract

Abstract: Background: The obesity epidemic has significantly increased the incidence and severity of hepatic steatosis in liver surgery patients and liver donors, potentially impacting postoperative liver regeneration and function. Development of a non-invasive means to quantify hepatic steatosis would facilitate selection of candidates for liver resection and transplant donation. Methods: An IRB-approved protocol prospectively enrolled 28 patients with liver tumors requiring hepatic resection. In all patients, fast dual-echo gradient-echo MR images were acquired using 2-Point Dixon technique in 2D and 3D. The degree of steatosis was quantified by percent fat fraction (%FF) from in- and out-of-phase, and water-only and fat-only images. The technique-specific %FFs were compared to intraoperative and histopathological findings. Results: For patients with >30% steatosis by histology, the mean %FF was 22% (SD ± 5.2%) compared to a mean %FF of 5.0% (SD ± 2.1%, p = 0.0001) in patients with <30% steatosis. Using scaled values for the MR-calculated %FF, all patients with >30% pathologic steatosis could be identified preoperatively. Conclusions: Quantitative MRI identified patients with clinically-relevant steatosis with 100% accuracy. These findings could have significant impact on the management of liver resection patients and transplant donors.