Two magnetic resonance imaging (MRI) techniques topped transient elastography (TE) for diagnosing hepatic fibrosis and steatosis in patients with nonalcoholic fatty liver disease (NAFLD), according to a first-in-kind study.
Magnetic resonance elastography surpassed all other methods for staging fibrosis, while MRI-based measurement of proton density fat fraction (PDFF) was superior for grading steatosis, with liver biopsy used as the comparative gold standard, said Dr. Kento Imajo at Yokohama (Japan) City University Graduate School of Medicine and his associates. “Magnetic resonance imaging–based noninvasive assessment of liver fibrosis and steatosis is a potential alternative to liver biopsy in clinical practice,” the investigators wrote in the March issue of Gastroenterology.
Courtesy of Wikimedia / "Nephron" / Creative Commons License
Assessing liver fibrosis and steatosis is important for staging NAFLD. Although “useful” overall, transient elastography can be unreliable in morbidly obese NAFLD patients or those with ascites because of low-frequency vibrations created by the probe, the researchers noted. To compare TE with MRI-based magnetic resonance elastography and PDFF, they evaluated 142 patients with biopsy-confirmed NAFLD and 10 controls, all of whom they also assessed with five clinical scoring systems for fibrosis – the FIB4 index, the NAFLD fibrosis score, the aspartate aminotransferase (AST) to platelet ratio, the AST-to-alanine transaminase (ALT) ratio, and the BARD score (Gastroenterology. 2015 Dec 8. doi: 10.1053/j.gastro.2015.11.048).
Magnetic resonance elastography detected stage 2 or higher hepatic fibrosis with an area under the receiver operating characteristic (AUROC) curve value of 0.91 (95% confidence interval, 0.86-0.96), compared with 0.82 (0.74-0.89) for transient elastography (P = .001), the investigators reported. The AUROC for MRE also significantly exceeded the AUROCs for all five clinical indexes of fibrosis severity. Furthermore, MRI-based measurement of PDFF identified hepatic steatosis of grade 2 or higher with an AUROC curve value of 0.90 (95% CI, 0.82-0.97), which was significantly greater than the AUROC obtained by using TE to measure the controlled attenuation parameter (0.73; 95% CI, 0.64-0.81; P less than .001).
Adding a measure for serum keratin 18 fragments or ALT did not significantly improve the detection of nonalcoholic steatohepatitis or macrovesicular steatosis affecting at least 5% of hepatocytes by either MRI or TE, the researchers noted. While liver biopsy remains the gold standard for assessing NAFLD, it is associated with sampling errors and intra- and interobserver variability, and these errors could have affected their study results, they acknowledged. The study also did not account for hepatic perfusion, which can elevate liver stiffness measurement independently from liver disease.
Both the magnetic resonance elastography and PDFF techniques require specialized hardware and software that are available from several commercial suppliers, the researchers also noted.
The study was partially supported by the Japanese Ministry of Health, Labor, and Welfare, the Japanese Science and Technology Agency, and Kiban-B, Shingakujuturyouiki. The investigators had no disclosures.