Accurate and simple method for quantification of hepatic fat content using magnetic resonance imaging: a prospective study in biopsy-proven nonalcoholic fatty liver disease

• Published in: Journal of gastroenterology Vol. 45; no. 12; pp. 1263 - 1271
• Main Authors: Hatta, Tomoko, Fujinaga, Yasunari, Kadoya, Masumi, Ueda, Hitoshi, Murayama, Hiroaki, Kurozumi, Masahiro, Ueda, Kazuhiko, Komatsu, Michiharu, Nagaya, Tadanobu, Joshita, Satoru, Kodama, Ryo, Tanaka, Eiji, Uehara, Tsuyoshi, Sano, Kenji, Tanaka, Naoki
• Format: Journal Article
• Language: English
• Published: Japan Japan : Springer Japan 01-12-2010; Springer Japan; Springer; Springer Nature B.V
• Subjects: Abdominal Surgery; Adult; Aged; Aged, 80 and over; Biliary Tract; Biopsy; Colorectal Surgery; Fatty liver; Fatty Liver - diagnosis; Female; Gastroenterology; Hepatology; Humans; Liver biopsy; Liver Cirrhosis - complications; Magnetic resonance imaging; Magnetic Resonance Imaging - instrumentation; Magnetic Resonance Imaging - methods; Male; Medicine; Medicine & Public Health; Methods; Microscopy - methods; Middle Aged; Original Article—Liver; Pancreas; Phantoms, Imaging; Prospective Studies; Reproducibility of Results; Severity of Illness Index; Surgical Oncology; Water-oil phantom; Young Adult; Liver biopsy; Water–oil phantom; Fatty liver; Magnetic resonance imaging
• ISSN: 0944-1174; 1435-5922
• DOI: 10.1007/s00535-010-0277-6

Background To assess the degree of hepatic fat content, simple and noninvasive methods with high objectivity and reproducibility are required. Magnetic resonance imaging (MRI) is one such candidate, although its accuracy remains unclear. We aimed to validate an MRI method for quantifying hepatic fat content by calibrating MRI reading with a phantom and comparing MRI measurements in human subjects with estimates of liver fat content in liver biopsy specimens. Methods The MRI method was performed by a combination of MRI calibration using a phantom and double-echo chemical shift gradient-echo sequence (double-echo fast low-angle shot sequence) that has been widely used on a 1.5-T scanner. Liver fat content in patients with nonalcoholic fatty liver disease (NAFLD, n = 26) was derived from a calibration curve generated by scanning the phantom. Liver fat was also estimated by optical image analysis. The correlation between the MRI measurements and liver histology findings was examined prospectively. Results Magnetic resonance imaging measurements showed a strong correlation with liver fat content estimated from the results of light microscopic examination (correlation coefficient 0.91, P < 0.001) regardless of the degree of hepatic steatosis. Moreover, the severity of lobular inflammation or fibrosis did not influence the MRI measurements. Conclusions This MRI method is simple and noninvasive, has excellent ability to quantify hepatic fat content even in NAFLD patients with mild steatosis or advanced fibrosis, and can be performed easily without special devices.