Article Text
Abstract
Introduction The predictive accuracy of ARFI elastography (virtual touch quantification, VTq™) for the non-invasive assessment of liver fibrosis is increasingly recognised, particularly in viral hepatitis. There is less information on its role in other aetiologies of chronic liver disease. We report the results of a preliminary study examining liver stiffness (LS) measurements by ARFI in patients with iron overload.
Method Forty-nine measurements of LS in 43 patients were performed to investigate possible iron overload. LS estimation by ARFI was performed using a standard validated protocol with a Siemens Acuson S2000 + 4C1 probe. Biopsies were performed in 19 (39%) from the same region of liver, immediately after LS measurement. Clinical, biochemical, ultrasonic and histopathological data were collated retrospectively. Predictive accuracy variables were determined for fibrosis stage, using both standard published and local1calibrations.
Results Aetiologies of iron overload in the biopsy cohort were: genetic haemochromatosis (n = 4); hyperferritinaemia with NAFLD or alcohol excess (8); hyperferritinaemia of unknown cause (3); transfusion-dependent thalassaemia (2); C282Y/H63D compound heterozygote (1); and high ferritin from renal cell carcinoma (1). Validation criteria for ARFI technical quality (SD/mean <0.3) were reached in all 19 biopsy patients, but were exceeded in 3/49 (6%) examinations overall; one procedure failed due to poor reproducibility. Metavir stages were F0/1/2/3/4 in 6/3/3/2/5, respectively. Sensitivity, specificity, positive and negative predictive values for: a) the prediction of F4 (cirrhosis) by ARFI were 1.0, 0.71, 0.56 and 1.0, respectively, using published reference standards, and 0.80, 0.79, 0.57 and 0.92, respectively, using local calibration1; b) the exclusion of ≥F2 fibrosis were 0.80, 0.78, 0.80 and 0.78, respectively, with both calibrations. No correlation was seen between LS and either a) necro-inflammation, histological grade of siderosis or scan technical quality; or b) serum ferritin levels.
Conclusion ARFI elastography may prove a useful non-invasive technique for excluding F2 fibrosis and identifying F4 cirrhosis secondary to iron overload. The high incidence of dual pathology, particularly NAFLD and alcohol-related, may have influenced predictive accuracy in this small “real world” cohort. The degree of hepatic iron overload did not correlate with LS. Further biopsy-controlled studies are needed to validate the role of ARFI in this patient group.
Disclosure of interest None Declared.
Reference
Sherman D, et al. J Hepatol. 2014;60(1):S413