BACKGROUND/AIMS: The aim of this study was to compare the performance of 11 biochemical scores to estimate liver fibrosis in HIV/HBV co-infection. METHODS: Performance was evaluated using the Receiver Operating Characteristics (ROC) curve method. The Kappa index was used to study overall agreement with liver biopsy results. Interpretative algorithms were established by optimizing sensitivity and specificity and the percentage of correctly classified patients. RESULTS: One hundred and eight patients (F0-F1, n = 47; F2, n = 28; F3, n = 17; F4, n = 16) were considered for the evaluation of serum biomarker performance. The AUROCs of the Fibrotest, Hepascore, Fibrometer, and Zeng's scores ranged from 0.74 to 0.77 for significant fibrosis (> or = F2), from 0.79 to 0.84 for advanced fibrosis (> or = F3) and from 0.87 to 0.92 for cirrhosis (F4). Thresholds defined for each stage of fibrosis were close to those previously published for the Fibrotest and Hepascore. Strict concordance with biopsies correctly classified 50% of the patients. CONCLUSIONS: Fibrotest, Fibrometer, Hepascore, and Zeng's score were the most accurate non-invasive biochemical scores for liver fibrosis assessment in HIV/HBV co-infection. Global performance of biomarkers was not significantly improved by a decision tree combining the results of two biochemical scores.
BACKGROUND/AIMS: The aim of this study was to compare the performance of 11 biochemical scores to estimate liver fibrosis in HIV/HBV co-infection. METHODS: Performance was evaluated using the Receiver Operating Characteristics (ROC) curve method. The Kappa index was used to study overall agreement with liver biopsy results. Interpretative algorithms were established by optimizing sensitivity and specificity and the percentage of correctly classified patients. RESULTS: One hundred and eight patients (F0-F1, n = 47; F2, n = 28; F3, n = 17; F4, n = 16) were considered for the evaluation of serum biomarker performance. The AUROCs of the Fibrotest, Hepascore, Fibrometer, and Zeng's scores ranged from 0.74 to 0.77 for significant fibrosis (> or = F2), from 0.79 to 0.84 for advanced fibrosis (> or = F3) and from 0.87 to 0.92 for cirrhosis (F4). Thresholds defined for each stage of fibrosis were close to those previously published for the Fibrotest and Hepascore. Strict concordance with biopsies correctly classified 50% of the patients. CONCLUSIONS: Fibrotest, Fibrometer, Hepascore, and Zeng's score were the most accurate non-invasive biochemical scores for liver fibrosis assessment in HIV/HBV co-infection. Global performance of biomarkers was not significantly improved by a decision tree combining the results of two biochemical scores.
Authors: Richard K Sterling; Wendy C King; Abdus S Wahed; David E Kleiner; Mandana Khalili; Mark Sulkowski; Raymond T Chung; Mamta K Jain; Mauricio Lisker-Melman; David K Wong; Marc G Ghany Journal: Hepatology Date: 2019-08-19 Impact factor: 17.425
Authors: S C Raftopoulos; J George; M Bourliere; E Rossi; W B de Boer; G P Jeffrey; M Bulsara; D J Speers; G MacQuillan; H L I Ching; N Kontorinis; W Cheng; J Flexman; S Fermoyle; P Rigby; L Walsh; D McLeod; L A Adams Journal: Hepatol Int Date: 2011-07-12 Impact factor: 6.047
Authors: Ho Soo Chun; Jae Seung Lee; Hye Won Lee; Beom Kyung Kim; Jun Yong Park; Do Young Kim; Sang Hoon Ahn; Seung Up Kim Journal: Dig Dis Sci Date: 2021-09-02 Impact factor: 3.487
Authors: Lars Peters; Jacqueline Neuhaus; Amanda Mocroft; Vincent Soriano; Jürgen Rockstroh; Gregory Dore; Massimo Puoti; Ellen Tedaldi; Bonaventura Clotet; Bernd Kupfer; Jens D Lundgren; Marina B Klein Journal: Antivir Ther Date: 2011