Screening for nonalcoholic steatohepatitis by using cytokeratin 18 and transient elastography in HIV mono-infection

Background and aim HIV-infected individuals are at high risk of developing nonalcoholic steatohepatitis (NASH), a leading cause of end-stage liver disease in Western countries. Nonetheless, due to the invasiveness of liver biopsy, NASH remains poorly understood in HIV mono-infection. We aimed to characterize the prevalence and predictors of NASH in unselected HIV mono-infected patients by means of non-invasive diagnostic tools. Methods HIV-infected adults without significant alcohol intake or co-infection with hepatitis B or C underwent a routine screening program employing transient elastography (TE) with controlled attenuation parameter (CAP) and the serum biomarker cytokeratin-18 (CK-18). NASH was diagnosed non-invasively as the coexistence of fatty liver (CAP >= 248 dB/m) and CK-18 >246 U/L. Identified cases of NASH were offered a diagnostic liver biopsy. Predictors of NASH were determined by multivariate logistic regression analysis. Results 202 consecutive HIV mono-infected patients were included. NASH was non-invasively diagnosed in 23 cases (11.4%). Among them, 17 underwent a liver biopsy, and histology confirmed NASH in all cases. The prevalence of NASH was higher in patients with hypertriglyceridemia (17.1%), insulin resistance defined by homeostasis model for assessment of insulin resistance (HOMA-IR) (25%), those with detectable HIV viral load (42.9%) and those with elevated ALT (53.6%). After adjustment, higher HOMA-IR (adjusted odds ratio [aOR] = 1.20, 95% CI 1.01-1.43; p = 0.03) and ALT (aOR = 2.39, 95% CI 1.50-3.79; p<0.001) were independent predictors of NASH. Conclusions NASH, diagnosed by a non-invasive diagnostic approach employing CK-18 and TE with CAP, is common in unselected HIV mono-infected individuals, particularly in the presence of insulin resistance and elevated ALT.