Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2018 Feb;286(2):486-498.
doi: 10.1148/radiol.2017170550. Epub 2017 Sep 11.

Linearity, Bias, and Precision of Hepatic Proton Density Fat Fraction Measurements by Using MR Imaging: A Meta-Analysis

Takeshi Yokoo  1 , Suraj D Serai  1 , Ali Pirasteh  1 , Mustafa R Bashir  1 , Gavin Hamilton  1 , Diego Hernando  1 , Houchun H Hu  1 , Holger Hetterich  1 , Jens-Peter Kühn  1 , Guido M Kukuk  1 , Rohit Loomba  1 , Michael S Middleton  1 , Nancy A Obuchowski  1 , Ji Soo Song  1 , An Tang  1 , Xinhuai Wu  1 , Scott B Reeder  1 , Claude B Sirlin  1 ; RSNA-QIBA PDFF Biomarker Committee  1
Affiliations
Review

Linearity, Bias, and Precision of Hepatic Proton Density Fat Fraction Measurements by Using MR Imaging: A Meta-Analysis

Takeshi Yokoo et al. Radiology. 2018 Feb.

Abstract

Purpose To determine the linearity, bias, and precision of hepatic proton density fat fraction (PDFF) measurements by using magnetic resonance (MR) imaging across different field strengths, imager manufacturers, and reconstruction methods. Materials and Methods This meta-analysis was performed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A systematic literature search identified studies that evaluated the linearity and/or bias of hepatic PDFF measurements by using MR imaging (hereafter, MR imaging-PDFF) against PDFF measurements by using colocalized MR spectroscopy (hereafter, MR spectroscopy-PDFF) or the precision of MR imaging-PDFF. The quality of each study was evaluated by using the Quality Assessment of Studies of Diagnostic Accuracy 2 tool. De-identified original data sets from the selected studies were pooled. Linearity was evaluated by using linear regression between MR imaging-PDFF and MR spectroscopy-PDFF measurements. Bias, defined as the mean difference between MR imaging-PDFF and MR spectroscopy-PDFF measurements, was evaluated by using Bland-Altman analysis. Precision, defined as the agreement between repeated MR imaging-PDFF measurements, was evaluated by using a linear mixed-effects model, with field strength, imager manufacturer, reconstruction method, and region of interest as random effects. Results Twenty-three studies (1679 participants) were selected for linearity and bias analyses and 11 studies (425 participants) were selected for precision analyses. MR imaging-PDFF was linear with MR spectroscopy-PDFF (R2 = 0.96). Regression slope (0.97; P < .001) and mean Bland-Altman bias (-0.13%; 95% limits of agreement: -3.95%, 3.40%) indicated minimal underestimation by using MR imaging-PDFF. MR imaging-PDFF was precise at the region-of-interest level, with repeatability and reproducibility coefficients of 2.99% and 4.12%, respectively. Field strength, imager manufacturer, and reconstruction method each had minimal effects on reproducibility. Conclusion MR imaging-PDFF has excellent linearity, bias, and precision across different field strengths, imager manufacturers, and reconstruction methods. © RSNA, 2017 Online supplemental material is available for this article. An earlier incorrect version of this article appeared online. This article was corrected on October 2, 2017.

PubMed Disclaimer

Figures

Figure 1a:
Figure 1a:
Flow diagrams show study selection according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for (a) linearity and bias analysis and (b) precision analysis. * = Based on RSNA-QIBA PDFF Biomarker Committee for PDFF criteria as described in Materials and Methods section and Table E1 (online).
Figure 1b:
Figure 1b:
Flow diagrams show study selection according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for (a) linearity and bias analysis and (b) precision analysis. * = Based on RSNA-QIBA PDFF Biomarker Committee for PDFF criteria as described in Materials and Methods section and Table E1 (online).
Figure 2:
Figure 2:
Bar chart shows estimates of percentage compliance by using Quality Assessment of Studies of Diagnostic Accuracy 2 tool. All 28 studies had moderate to high scores and low risk of bias in all seven categories; all fulfilled five or more of the seven quality categories. Summary of scores for each category were as follows: Category 1, Risk of bias–Patient selection, 57% (16 of 28); Category 2, Risk of bias–Index test, 100% (28 of 28); Category 3, Risk of bias–Reference standard, 100% (28 of 28); Category 4, Risk of bias–Flow and timing, 93% (26 of 28); Category 5, Applicability concerns–Patient selection, 61% (17 of 28); Category 6, Applicability concerns–Index test, 93% (26 of 28); and Category 7, Applicability concerns–Reference standard, 96% (27 of 28).
Figure 3a:
Figure 3a:
Scatterplots show linearity and bias of MR imaging–PDFF. (a) Linearity of MR imaging–PDFF against MR spectroscopy-PDFF, both measured in colocalized ROIs in the liver. Red line represents linear regression fit (after correcting for within-participant correlations between replicated measurements, if any) with coefficient of determination indicating very strong linear fit (R2 = 0.96). Estimated intercept (−0.07%) was not significantly different from zero; estimated slope was close to but less than unity (0.97), indicating very slight underestimation of MR imaging–PDFF values at larger MR spectroscopy-PDFF values. (b) Bland-Altman plot of MR imaging–PDFF relative to MR spectroscopy-PDFF as reference technique demonstrates very small mean bias (−0.13%) and limits of agreement within ± 4%. Estimated size of various bias components are presented in Table 3.
Figure 3b:
Figure 3b:
Scatterplots show linearity and bias of MR imaging–PDFF. (a) Linearity of MR imaging–PDFF against MR spectroscopy-PDFF, both measured in colocalized ROIs in the liver. Red line represents linear regression fit (after correcting for within-participant correlations between replicated measurements, if any) with coefficient of determination indicating very strong linear fit (R2 = 0.96). Estimated intercept (−0.07%) was not significantly different from zero; estimated slope was close to but less than unity (0.97), indicating very slight underestimation of MR imaging–PDFF values at larger MR spectroscopy-PDFF values. (b) Bland-Altman plot of MR imaging–PDFF relative to MR spectroscopy-PDFF as reference technique demonstrates very small mean bias (−0.13%) and limits of agreement within ± 4%. Estimated size of various bias components are presented in Table 3.
Figure 4a:
Figure 4a:
(a, b) Scatterplots illustrate participant-level and ROI-level precision of MR imaging–PDFF measurements, respectively, because of differences in field strength, imager manufacturer, reconstruction method, equipment setup, and random noise (effect sizes are detailed in Table 4). Repeatability refers to precision under an identical experimental condition (ie, scan-rescan repeatability by using fixed hardware and software). Reproducibility refers to precision under variable experimental condition (ie, by using different hardware or software). At ROI level (b), MR imaging–PDFF is highly precise with repeatability coefficients (RCs) and reproducibility coefficients (RDCs) indicating the 95th percentile of precision to be approximately ± 3% and ± 4%, respectively. At participant level (a), repeatability and reproducibility are approximately ± 5% and ± 5.5%. Subject-level precision is lower than ROI-level precision because of differences in ROI placement, that is, heterogeneity of underlying steatosis. In both cases, RCs and RDCs are similar to each other, indicating small impact of technical factors (field strength, imager manufacturer, reconstruction method) on the precision of the measurement.
Figure 4b:
Figure 4b:
(a, b) Scatterplots illustrate participant-level and ROI-level precision of MR imaging–PDFF measurements, respectively, because of differences in field strength, imager manufacturer, reconstruction method, equipment setup, and random noise (effect sizes are detailed in Table 4). Repeatability refers to precision under an identical experimental condition (ie, scan-rescan repeatability by using fixed hardware and software). Reproducibility refers to precision under variable experimental condition (ie, by using different hardware or software). At ROI level (b), MR imaging–PDFF is highly precise with repeatability coefficients (RCs) and reproducibility coefficients (RDCs) indicating the 95th percentile of precision to be approximately ± 3% and ± 4%, respectively. At participant level (a), repeatability and reproducibility are approximately ± 5% and ± 5.5%. Subject-level precision is lower than ROI-level precision because of differences in ROI placement, that is, heterogeneity of underlying steatosis. In both cases, RCs and RDCs are similar to each other, indicating small impact of technical factors (field strength, imager manufacturer, reconstruction method) on the precision of the measurement.
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Loomba R, Sanyal AJ. The global NAFLD epidemic. Nat Rev Gastroenterol Hepatol 2013;10(11):686–690. - PubMed
    1. Wong RJ, Cheung R, Ahmed A. Nonalcoholic steatohepatitis is the most rapidly growing indication for liver transplantation in patients with hepatocellular carcinoma in the U.S. Hepatology 2014;59(6):2188–2195. - PubMed
    1. Matteoni CA, Younossi ZM, Gramlich T, Boparai N, Liu YC, McCullough AJ. Nonalcoholic fatty liver disease: a spectrum of clinical and pathological severity. Gastroenterology 1999;116(6):1413–1419. - PubMed
    1. Ekstedt M, Franzén LE, Mathiesen UL, et al. . Long-term follow-up of patients with NAFLD and elevated liver enzymes. Hepatology 2006;44(4):865–873. - PubMed
    1. Rubinstein E, Lavine JE, Schwimmer JB. Hepatic, cardiovascular, and endocrine outcomes of the histological subphenotypes of nonalcoholic fatty liver disease. Semin Liver Dis 2008;28(4):380–385. - PubMed

MeSH terms

LinkOut - more resources

  NODES
twitter 2