Improved fibrosis staging by elastometry and blood test in chronic hepatitis C. (16th October 2013)

• Record Type: Journal Article
• Authors: Calès, Paul; Boursier, Jérôme; Ducancelle, Alexandra; Oberti, Frédéric; Hubert, Isabelle; Hunault, Gilles; de Lédinghen, Victor; Zarski, Jean‐Pierre; Salmon, Dominique; Lunel, Françoise; the ANRS HC EP 23 Fibrostar Study
• Is Part Of: Liver international. Volume 34:Number 6(2014:Aug.)
• Page Start: 907
• Page End: 917
• Publication Date: 2013-10-16
• Subjects: Liver -- Periodicals; Liver -- Diseases -- Periodicals; 616.362
• DOI: 10.1111/liv.12327
• Languages: English
• ISSNs: 1478-3223

Abstract

<abstract abstract-type="main" id="liv12327-abs-0001"> <title>Abstract</title> <sec id="liv12327-sec-0001" sec-type="section"> <title>Aims</title> <p>Our main objective was to improve non‐invasive fibrosis staging accuracy by resolving the limits of previous methods via new test combinations. Our secondary objectives were to improve staging precision, by developing a detailed fibrosis classification, and reliability (personalized accuracy) determination.</p> </sec> <sec id="liv12327-sec-0002" sec-type="section"> <title>Methods</title> <p>All patients (729) included in the derivation population had chronic hepatitis C, liver biopsy, 6 blood tests and Fibroscan. Validation populations included 1584 patients.</p> </sec> <sec id="liv12327-sec-0003" sec-type="section"> <title>Results</title> <p>The most accurate combination was provided by using most markers of FibroMeter and Fibroscan results targeted for significant fibrosis, i.e. 'E‐FibroMeter'. Its classification accuracy (91.7%) and precision (assessed by F difference with Metavir: 0.62 ± 0.57) were better than those of FibroMeter (84.1%, <italic>P</italic> < 0.001; 0.72 ± 0.57, <italic>P</italic> < 0.001), Fibroscan (88.2%, <italic>P</italic> = 0.011; 0.68 ± 0.57, <italic>P</italic> = 0.020), and a previous CSF‐SF classification of FibroMeter + Fibroscan (86.7%, <italic>P</italic> < 0.001; 0.65 ± 0.57, <italic>P</italic> = 0.044). The accuracy for fibrosis absence (F0) was increased, e.g. from 16.0% with Fibroscan to 75.0% with E‐FibroMeter (<italic>P</italic> < 0.001). Cirrhosis sensitivity was improved, e.g. E‐FibroMeter: 92.7% vs. Fibroscan: 83.3%, <italic>P</italic> = 0.004. The combination improved reliability by deleting unreliable results (accuracy <50%) observed with a single test (1.2% of patients) and increasing optimal reliability (accuracy ≥85%) from 80.4% of patients with Fibroscan (accuracy: 90.9%) to 94.2% of patients with E‐FibroMeter (accuracy: 92.9%), <italic>P</italic> < 0.001. The patient rate with 100% predictive values for cirrhosis by the best combination was twice (36.2%) that of the best single test (FibroMeter: 16.2%, <italic>P</italic> < 0.001).</p> </sec> <sec id="liv12327-sec-0004" sec-type="section"> <title>Conclusion</title> <p>The new test combination increased: accuracy, globally and especially in patients without fibrosis, staging precision, cirrhosis prediction, and even reliability, thus offering improved fibrosis staging.</p> </sec> </abstract>