Association of gadolinium-enhanced magnetic resonance imaging with hepatic fibrosis and inflammation in primary sclerosing cholangitis.
Standard
Association of gadolinium-enhanced magnetic resonance imaging with hepatic fibrosis and inflammation in primary sclerosing cholangitis. / Keller, Sarah; Aigner, Annette; Zenouzi, Roman; Kim, Anne C; Meijer, Arnoud; Weidemann, Sören A; Krech, Till; Lohse, Ansgar W; Adam, Gerhard; Schramm, Christoph; Yamamura, Jin.
in: PLOS ONE, Jahrgang 13, Nr. 3, 2018, S. e0193929.Publikationen: SCORING: Beitrag in Fachzeitschrift/Zeitung › SCORING: Zeitschriftenaufsatz › Forschung › Begutachtung
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Association of gadolinium-enhanced magnetic resonance imaging with hepatic fibrosis and inflammation in primary sclerosing cholangitis.
AU - Keller, Sarah
AU - Aigner, Annette
AU - Zenouzi, Roman
AU - Kim, Anne C
AU - Meijer, Arnoud
AU - Weidemann, Sören A
AU - Krech, Till
AU - Lohse, Ansgar W
AU - Adam, Gerhard
AU - Schramm, Christoph
AU - Yamamura, Jin
PY - 2018
Y1 - 2018
N2 - OBJECTIVE: To evaluate magnetic resonance imaging (MRI) parameters T2 signal, contrast enhancement (CE), and relative liver enhancement (RLE) of extracellular gadolinium-based contrast agent (GBCA)-enhanced MRI as a marker for hepatic fibrosis and inflammation in patients with primary sclerosing cholangitis (PSC).METHODS: 3.0-Tesla MRI scans and liver biopsies of 40 patients (41.2 ± 17.1 years) were retrospectively reviewed. Biopsies were obtained within a mean time of 54 ± 55 days to MRI scans and specimens were categorized according to Ishak modified hepatic activity index (mHAI) and Scheuer staging of fibrosis. T2 signal (N = 40), CE alterations (N = 29), and RLE (N = 29) were assessed by two raters. Mixed-effects regression models were applied to estimate the association between histopathology and MRI parameters.RESULTS: No significant association was observed between T2 signal or CE alterations with stages of fibrosis or mHAI grading. Regression models revealed significant positive associations of portal venous phase RLE with mHAI grade ≥ 7 points [β = 25.5; 95% CI (2.53; 48.62); p = 0.04] and delayed phase RLE with stages of fibrosis [stage 2: β = 35.13; 95% CI (11.35; 58.87); p = 0.007; stage 3/4: β = 69.24; 95% CI (45.77; 92.75); p < 0.001]. The optimal cut-off value of 66.6% delayed phase RLE distinguished fibrosis stages 0-2 from 3-4 with a sensitivity of 0.833 and specificity of 0.972. Inter-rater reliability (IRR) for quantification of RLE was 'excellent' (r = 0.90-0.98). IRR was 'substantial' for detection of T2 signal in the right liver lobe (RL) (Kappa = 0.77) and 'almost perfect' for T2 signal of the left liver lobe (LL) and CE of both lobes (Kappa = 0.87-1.0).CONCLUSION: The simple and reproducible method of RLE quantification on standard extracellular GBCA-enhanced MRI may provide a correlate measure of advanced stages of hepatic fibrosis and potentially also inflammation in PSC patients, if validated in larger cohorts.
AB - OBJECTIVE: To evaluate magnetic resonance imaging (MRI) parameters T2 signal, contrast enhancement (CE), and relative liver enhancement (RLE) of extracellular gadolinium-based contrast agent (GBCA)-enhanced MRI as a marker for hepatic fibrosis and inflammation in patients with primary sclerosing cholangitis (PSC).METHODS: 3.0-Tesla MRI scans and liver biopsies of 40 patients (41.2 ± 17.1 years) were retrospectively reviewed. Biopsies were obtained within a mean time of 54 ± 55 days to MRI scans and specimens were categorized according to Ishak modified hepatic activity index (mHAI) and Scheuer staging of fibrosis. T2 signal (N = 40), CE alterations (N = 29), and RLE (N = 29) were assessed by two raters. Mixed-effects regression models were applied to estimate the association between histopathology and MRI parameters.RESULTS: No significant association was observed between T2 signal or CE alterations with stages of fibrosis or mHAI grading. Regression models revealed significant positive associations of portal venous phase RLE with mHAI grade ≥ 7 points [β = 25.5; 95% CI (2.53; 48.62); p = 0.04] and delayed phase RLE with stages of fibrosis [stage 2: β = 35.13; 95% CI (11.35; 58.87); p = 0.007; stage 3/4: β = 69.24; 95% CI (45.77; 92.75); p < 0.001]. The optimal cut-off value of 66.6% delayed phase RLE distinguished fibrosis stages 0-2 from 3-4 with a sensitivity of 0.833 and specificity of 0.972. Inter-rater reliability (IRR) for quantification of RLE was 'excellent' (r = 0.90-0.98). IRR was 'substantial' for detection of T2 signal in the right liver lobe (RL) (Kappa = 0.77) and 'almost perfect' for T2 signal of the left liver lobe (LL) and CE of both lobes (Kappa = 0.87-1.0).CONCLUSION: The simple and reproducible method of RLE quantification on standard extracellular GBCA-enhanced MRI may provide a correlate measure of advanced stages of hepatic fibrosis and potentially also inflammation in PSC patients, if validated in larger cohorts.
KW - Journal Article
U2 - 10.1371/journal.pone.0193929
DO - 10.1371/journal.pone.0193929
M3 - SCORING: Journal article
C2 - 29513767
VL - 13
SP - e0193929
JO - PLOS ONE
JF - PLOS ONE
SN - 1932-6203
IS - 3
ER -