Candice A Bookwalter1, Sudhakar K Venkatesh2, John E Eaton3, Thomas D Smyrk4, Richard L Ehman1. 1. Department of Radiology, Mayo Clinic College of Medicine, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA. 2. Department of Radiology, Mayo Clinic College of Medicine, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA. venkatesh.sudhakar@mayo.edu. 3. Department of Medicine, Mayo Clinic College of Medicine, Mayo Clinic, Rochester, MN, USA. 4. Department of Pathology, Mayo Clinic College of Medicine, Mayo Clinic, Rochester, MN, USA.
Abstract
AIM: To determine correlation of liver stiffness measured by MR Elastography (MRE) with biliary abnormalities on MR Cholangiopancreatography (MRCP) and MRI parenchymal features in patients with primary sclerosing cholangitis (PSC). METHODS: Fifty-five patients with PSC who underwent MRI of the liver with MRCP and MRE were retrospectively evaluated. Two board-certified abdominal radiologists in agreement reviewed the MRI, MRCP, and MRE images. The biliary tree was evaluated for stricture, dilatation, wall enhancement, and thickening at segmental duct, right main duct, left main duct, and common bile duct levels. Liver parenchyma features including signal intensity on T2W and DWI, and hyperenhancement in arterial, portal venous, and delayed phase were evaluated in nine Couinaud liver segments. Atrophy or hypertrophy of segments, cirrhotic morphology, varices, and splenomegaly were scored as present or absent. Regions of interest were placed in each of the nine segments on stiffness maps wherever available and liver stiffness (LS) was recorded. Mean segmental LS, right lobar (V-VIII), left lobar (I-III, and IVA, IVB), and global LS (average of all segments) were calculated. Spearman rank correlation analysis was performed for significant correlation. Features with significant correlation were then analyzed for significant differences in mean LS. Multiple regression analysis of MRI and MRCP features was performed for significant correlation with elevated LS. RESULTS: A total of 439/495 segments were evaluated and 56 segments not included in MRE slices were excluded for correlation analysis. Mean segmental LS correlated with the presence of strictures (r = 0.18, p < 0.001), T2W hyperintensity (r = 0.38, p < 0.001), DWI hyperintensity (r = 0.30, p < 0.001), and hyperenhancement of segment in all three phases. Mean LS of atrophic and hypertrophic segments were significantly higher than normal segments (7.07 ± 3.6 and 6.67 ± 3.26 vs. 5.1 ± 3.6 kPa, p < 0.001). In multiple regression analysis, only the presence of segmental strictures (p < 0.001), T2W hyperintensity (p = 0.01), and segmental hypertrophy (p < 0.001) were significantly associated with elevated segmental LS. Only left ductal stricture correlated with left lobe LS (r = 0.41, p = 0.018). Global LS correlated significantly with CBD stricture (r = 0.31, p = 0.02), number of segmental strictures (r = 0.28, p = 0.04), splenomegaly (r = 0.56, p < 0.001), and varices (r = 0.58, p < 0.001). CONCLUSION: In PSC, there is low but positive correlation between segmental LS and segmental duct strictures. Segments with increased LS show T2 hyperintensity, DWI hyperintensity, and post-contrast hyperenhancement. Global liver stiffness shows a moderate correlation with number of segmental strictures and significantly correlates with spleen stiffness, splenomegaly, and varices.
AIM: To determine correlation of liver stiffness measured by MR Elastography (MRE) with biliary abnormalities on MR Cholangiopancreatography (MRCP) and MRI parenchymal features in patients with primary sclerosing cholangitis (PSC). METHODS: Fifty-five patients with PSC who underwent MRI of the liver with MRCP and MRE were retrospectively evaluated. Two board-certified abdominal radiologists in agreement reviewed the MRI, MRCP, and MRE images. The biliary tree was evaluated for stricture, dilatation, wall enhancement, and thickening at segmental duct, right main duct, left main duct, and common bile duct levels. Liver parenchyma features including signal intensity on T2W and DWI, and hyperenhancement in arterial, portal venous, and delayed phase were evaluated in nine Couinaud liver segments. Atrophy or hypertrophy of segments, cirrhotic morphology, varices, and splenomegaly were scored as present or absent. Regions of interest were placed in each of the nine segments on stiffness maps wherever available and liver stiffness (LS) was recorded. Mean segmental LS, right lobar (V-VIII), left lobar (I-III, and IVA, IVB), and global LS (average of all segments) were calculated. Spearman rank correlation analysis was performed for significant correlation. Features with significant correlation were then analyzed for significant differences in mean LS. Multiple regression analysis of MRI and MRCP features was performed for significant correlation with elevated LS. RESULTS: A total of 439/495 segments were evaluated and 56 segments not included in MRE slices were excluded for correlation analysis. Mean segmental LS correlated with the presence of strictures (r = 0.18, p < 0.001), T2W hyperintensity (r = 0.38, p < 0.001), DWI hyperintensity (r = 0.30, p < 0.001), and hyperenhancement of segment in all three phases. Mean LS of atrophic and hypertrophic segments were significantly higher than normal segments (7.07 ± 3.6 and 6.67 ± 3.26 vs. 5.1 ± 3.6 kPa, p < 0.001). In multiple regression analysis, only the presence of segmental strictures (p < 0.001), T2W hyperintensity (p = 0.01), and segmental hypertrophy (p < 0.001) were significantly associated with elevated segmental LS. Only left ductal stricture correlated with left lobe LS (r = 0.41, p = 0.018). Global LS correlated significantly with CBD stricture (r = 0.31, p = 0.02), number of segmental strictures (r = 0.28, p = 0.04), splenomegaly (r = 0.56, p < 0.001), and varices (r = 0.58, p < 0.001). CONCLUSION: In PSC, there is low but positive correlation between segmental LS and segmental duct strictures. Segments with increased LS show T2 hyperintensity, DWI hyperintensity, and post-contrast hyperenhancement. Global liver stiffness shows a moderate correlation with number of segmental strictures and significantly correlates with spleen stiffness, splenomegaly, and varices.
Entities:
Keywords:
Atrophy; Biliary stricture; Hyperenhancement; Hypertrophy; Segmental stiffness; Signal intensity
Authors: E V Loftus; G C Harewood; C G Loftus; W J Tremaine; W S Harmsen; A R Zinsmeister; D A Jewell; W J Sandborn Journal: Gut Date: 2005-01 Impact factor: 23.059
Authors: W R Kim; T M Therneau; R H Wiesner; J J Poterucha; J T Benson; M Malinchoc; N F LaRusso; K D Lindor; E R Dickson Journal: Mayo Clin Proc Date: 2000-07 Impact factor: 7.616
Authors: R Olsson; A Danielsson; G Järnerot; E Lindström; L Lööf; P Rolny; B O Rydén; C Tysk; S Wallerstedt Journal: Gastroenterology Date: 1991-05 Impact factor: 22.682
Authors: Bojan D Petrovic; Paul Nikolaidis; Nancy A Hammond; John A Martin; Polina V Petrovic; Pankaja M Desai; Frank H Miller Journal: Dig Dis Sci Date: 2007-04-05 Impact factor: 3.199
Authors: John E Eaton; Aditi Sen; Safa Hoodeshenas; Cathy D Schleck; William S Harmsen; Gregory J Gores; Nicholas F LaRusso; Andrea A Gossard; Konstantinos N Lazaridis; Sudhakar K Venkatesh Journal: Clin Gastroenterol Hepatol Date: 2019-11-01 Impact factor: 11.382
Authors: Matthew A Morgan; Rachita Khot; Karthik M Sundaram; Daniel R Ludwig; Rashmi T Nair; Pardeep K Mittal; Dhakshina M Ganeshan; Sudhakar K Venkatesh Journal: Abdom Radiol (NY) Date: 2022-09-05
Authors: Erum A Hartung; Juan S Calle-Toro; Carolina Maya Lopera; Jessica Wen; Robert H Carson; Mohini Dutt; Kathryn Howarth; Susan L Furth; Kassa Darge; Suraj D Serai Journal: Abdom Radiol (NY) Date: 2020-08-05