Increased liver stiffness is associated with mortality in HIV/HCV coinfected subjects: The French nationwide ANRS CO13 HEPAVIH cohort study.

• Main Authors: Sarah Shili-Masmoudi, Philippe Sogni, Victor de Ledinghen, Laure Esterle, Marc-Antoine Valantin, Isabelle Poizot-Martin, Anne Simon, Eric Rosenthal, Karine Lacombe, Gilles Pialoux, Olivier Bouchaud, Anne Gervais-Hasenknoff, Cécile Goujard, Lionel Piroth, David Zucman, Stéphanie Dominguez, François Raffi, Laurent Alric, Firouzé Bani-Sadr, Caroline Lascoux-Combe, Daniel Garipuy, Patrick Miailhes, Daniel Vittecoq, Claudine Duvivier, Hugues Aumaître, Didier Neau, Philippe Morlat, François Dabis, Dominique Salmon, Linda Wittkop, ANRS CO13 HEPAVIH study group
• Format: Article
• Language: English
• Published: Public Library of Science (PLoS) 2019-01-01
• Series: PLoS ONE
• Online Access: https://doi.org/10.1371/journal.pone.0211286

BACKGROUND:The association between liver stiffness measurements (LSM) and mortality has not been fully described. In particular the effect of LSM on all-cause mortality taking sustained virological response (SVR) into account needs further study. METHODS:HIV/HCV participants in the French nation-wide, prospective, multicenter ANRS CO13 HEPAVIH cohort, with ≥1 LSM by FibroScan (FS) and a detectable HCV RNA when the first valid FS was performed were included. Cox proportional hazards models with delayed entry were performed to determine factors associated with all-cause mortality. LSM and SVR were considered as time dependent covariates. RESULTS:1,062 patients were included from 2005 to 2015 (69.8% men, median age 45.7 years (IQR 42.4-49.1)). 21.7% had baseline LSM >12.5 kPa. Median follow-up was 4.9 years (IQR 3.2-6.1). 727 (68.5%) were ever treated for HCV: 189 of them (26.0%) achieved SVR. 76 deaths were observed (26 liver-related, 10 HIV-related, 29 non-liver-non-HIV-related, 11 of unknown cause). At the age of 50, the mortality rate was 4.5% for patients with LSM ≤12.5 kPa and 10.8% for patients with LSM >12.5 kPa. LSM >12.5 kPa (adjusted Hazard Ratio [aHR] = 3.35 [2.06; 5.45], p<0.0001), history of HCV treatment (aHR = 0.53 [0.32; 0.90], p = 0.01) and smoking (past (aHR = 5.69 [1.56; 20.78]) and current (3.22 [0.93; 11.09]) versus never, p = 0.01) were associated with all-cause mortality independently of SVR, age, sex, alcohol use and metabolic disorders. CONCLUSION:Any LSM >12.5 kPa was strongly associated with all-cause mortality independently of SVR and other important covariates. Our results suggest that close follow-up of these patients should remain a priority even after achieving SVR.