Liver disease is frequently seen in HIV+ patients as a result of coinfection with hepatitis B (HBV) or C (HCV) viruses, alcohol abuse and/or exposure to hepatotoxic drugs. The aim of this study was to assess the prevalence of liver cirrhosis, its main causes and clinical presentation in HIV+ patients. Observational, cross-sectional, retrospective study of all HIV+ individuals followed at one reference HIV outpatient clinic in Madrid. Liver fibrosis was measured in all cases using transient elastometry (FibroScan). All 2168 HIV+ patients on regular follow-up (76% males, 46% injecting drug users) were successfully examined by FibroScan) between October 2004 and August 2006. Liver cirrhosis was recognized in 181 (overall prevalence, 8.3%), and the main aetiologies were HCV, 82.3%; HBV, 1.6%; dual HBV/HCV, 2.8%; and triple HBV/HCV/ hepatitis delta virus (HDV) infection, 6.6%. The prevalence of cirrhosis differed among patients with distinct chronic viral hepatitis: HCV, 19.2%; HBV, 6.1%; HBV/HCV, 41.7%; and HBV/HCV/HDV, 66.7%. In 12 patients with cirrhosis (6.7%), no definite aetiology was recognized. Overall, cirrhotics had lower mean CD4 counts than noncirrhotics (408 vs 528 cells/microL respectively; P = 0.02), despite similar proportion of subjects with undetectable viraemia on highly active antiretroviral therapy. Clinical manifestations of liver cirrhosis were: splenomegaly, 61.5%; oesophageal varices, 59.8%; ascites, 22.6%; encephalopathy, 12.1%; and variceal bleeding, 6.1%. Liver cirrhosis and hepatic decompensation events are relatively frequent in HIV+ individuals. Chronic HCV and alcohol abuse, but not chronic HBV, play a major role. Transient elastometry may allow the identification of a significant number of HIV+ individuals with asymptomatic liver cirrhosis.

Sustained HCV clearance after IFN-based therapy reduces the risk of liver toxicity during antiretroviral therapy, which should further encourage the treatment of chronic hepatitis C in HIV-coinfected patients. In this population, prescription of PIs or efavirenz decreases and use of dydeoxynucleoside analogues increases the risk of hepatotoxicity.

The recent availability of non-invasive tools to measure liver fibrosis has allowed examination of its extent and determination of predictors in all patients with chronic hepatitis C virus (HCV) infection. On the other hand, most information on hepatic fibrosis in HCV/human immunodeficiency virus (HIV)-coinfected patients has been derived from liver biopsies taken before highly active antiretroviral therapy (HAART) was widely available. All consecutive HCV patients with elevated aminotransferases seen during the last 3 years were evaluated and liver fibrosis measured using transient elastography (FibroScan) and biochemical indexes. Patients were split according to their HIV serostatus. A total of 656 (69.6%) HCV-monoinfected and 287 (30.4%) HIV/HCV-coinfected patients were assessed. Mean CD4 count of coinfected patients was 493 cells/muL and 88% were under HAART (mean time, 4.2 +/- 2.4 years). Advanced liver fibrosis or cirrhosis was recognized in 39% of the coinfected and 18% of the monoinfected patients (P < 0.005). A good correlation was found between FibroScan) and biochemical indexes [AST to platelet ratio index (r = 0.405, P < 0.0001), FIB-4 (r = 0.393, P < 0.0001) and Forns (r = 0.407, P < 0.0001)], regardless of the HIV status. In the multivariate analysis, age >45 years, body mass index (BMI) >25 kg/m(2), and HIV infection were independently associated with advanced liver fibrosis or cirrhosis. HIV/HCV-coinfected patients have more advanced liver fibrosis than HCV-monoinfected patients despite the immunologic benefit of HAART.

Background Cases of severe unexplained liver disease in HIV-infected individuals have recently been reported and are often associated with exposure to didanosine (ddI) and nodular regenerative hyperplasia. Herein, we examine the clinical outcome following ddI removal. Methods From 3,300 HIV-infected patients attending three clinics since 2004, all who exhibited persistently elevated aminotransferases and/or significant liver fibrosis in the absence of any known cause of liver damage were identified. Results Thirty-two individuals (prevalence ∼1%) met the inclusion criteria - all were on antiretroviral therapy. Of these, 84% were male and 68% had acquired HIV through homosexual contact. Liver biopsy was performed in 12, of whom three showed nonspecific advanced liver fibrosis, two nodular regenerative hyperplasia and three showed only periportal fibrosis. On follow up, nine patients developed episodes of hepatic decompensation, mainly as a consequence of portal hypertension; in eight cases (25%) portal thrombosis was diagnosed. No association was found with plasma HIV RNA or CD4+ T-cell count. All patients but three had been exposed to ddI for a median of 44 months; removal of ddI in 27 was followed 12 months later by improvement in clinical and laboratory parameters in 13 (48%) patients. Finally, a trend towards liver fibrosis improvement was recognised using FibroScan®. Conclusions Idiopathic persistent liver enzyme elevations in HIV-infected individuals are often associated with cirrhotic and non-cirrhotic portal hypertension. Although this is a relatively rare condition, prolonged exposure to ddI seems to play a pathogenic role and removal of the drug is associated with clinical and laboratory improvements.

No abstract