A Single Test Combining Blood Markers and Elastography is More Accurate Than Other Fibrosis Tests in the Main Causes of Chronic Liver Diseases

TitreA Single Test Combining Blood Markers and Elastography is More Accurate Than Other Fibrosis Tests in the Main Causes of Chronic Liver Diseases
Type de publicationArticle de revue
AuteurDucancelle, Alexandra , Leroy, Vincent, Vergniol, Julien, Sturm, Nathalie, Le Bail, Brigitte, Zarski, Jean-Pierre, Nguyen Khac, Eric, Salmon-Céron, Dominique, de Ledinghen, Victor, Calès, Paul
EditeurLippincott, Williams & Wilkins
TypeArticle scientifique dans une revue à comité de lecture
Année2017
LangueAnglais
DateAoût 2017
Numéro7
Pagination639-649
Volume51
Titre de la revueJournal of Clinical Gastroenterology
ISSN1539-2031
Mots-clésAdult, Aspartate Aminotransferases, Biomarkers, Elasticity Imaging Techniques, End Stage Liver Disease, Female, Humans, Liver, Liver Cirrhosis, Liver Diseases, Male, Middle Aged, Platelet Count, Prospective Studies, Retrospective Studies, ROC Curve, Severity of Illness Index
Résumé en anglais

BACKGROUND AND GOAL: International guidelines suggest combining a blood test and liver stiffness measurement (LSM) to stage liver fibrosis in chronic hepatitis C (CHC) and non-alcoholic fatty liver disease (NAFLD). Therefore, we compared the accuracies of these tests between the main etiologies of chronic liver diseases.

STUDY: Overall, 1968 patients were included in 5 etiologies: CHC: 698, chronic hepatitis B: 152, human immunodeficiency virus/CHC: 628, NAFLD: 225, and alcoholic liver disease (ALD): 265. Sixteen tests [13 blood tests, LSM (Fibroscan), 2 combined: FibroMeters] were evaluated. References were Metavir staging and CHC etiology. Accuracy was evaluated mainly with the Obuchowski index (OI) and accessorily with area under the receiver operating characteristics (F≥2, F≥3, cirrhosis).

RESULTS: OIs in CHC were: FibroMeters: 0.812, FibroMeters: 0.785 to 0.797, Fibrotest: 0.762, CirrhoMeters: 0.756 to 0.771, LSM: 0.754, Hepascore: 0.752, FibroMeter: 0.750, aspartate aminotransferase platelet ratio index: 0.742, Fib-4: 0.741. In other etiologies, most tests had nonsignificant changes in OIs. In NAFLD, CHC-specific tests were more accurate than NAFLD-specific tests. The combined FibroMeters had significantly higher accuracy than their 2 constitutive tests (FibroMeters and LSM) in at least 1 diagnostic target in all etiologies, except in ALD where LSM had the highest OI, and in 3 diagnostic targets (OIs and 2 area under the receiver operating characteristics) in CHC and NAFLD.

CONCLUSIONS: Some tests developed in CHC outperformed other tests in their specific etiologies. Tests combining blood markers and LSM outperformed single tests, validating recent guidelines and extending them to main etiologies. Noninvasive fibrosis evaluation can thus be simplified in the main etiologies by using a unique test: either LSM alone, especially in ALD, or preferably combined to blood markers.

URL de la noticehttp://okina.univ-angers.fr/publications/ua18788
DOI10.1097/MCG.0000000000000788
Lien vers le document

https://insights.ovid.com/crossref?an=00004836-201708000-00013

Titre abrégéJ. Clin. Gastroenterol.
Identifiant (ID) PubMed28692443