Background:  This study examines a population of patients with HIV/hepatitis B virus (HBV) co-infection, where the hepatitis B surface antigen (HBsAg) is positive for 6 months or more and then becomes HBsAg negative (HB seroreversion). This group of patients is compared with HIV/HBV-co-infected patients who have not undergone seroreversion. The main comparisons being investigated are ART, liver function, hepatitis- and HIV-related factors, such as CD4 counts and plasma HIV RNA.

Methods:  A retrospective matched case-control design was used to investigate this issue.  Cases and 2 matched controls were selected from the Victorian HIV Service database at the Alfred Hospital in Melbourne, Australia. All controls were HBsAg⁺ at the time of selection. Matching was based on date of birth and whether the patient was first seen at the Victorian HIV Service pre-HAART or within the era of HAART.

Result:  This investigation has found that 8.5% (20 of 236) of individuals with chronic HBV infection underwent seroreversion in the era of HAART, 95% of whom (n = 19, 95%) were male compared to 92.5% (n = 37) of the controls. Mean age was 40.6 (SD 7.7) vs 35.1 (SD 6.2) for the controls. And, 95% (n = 19) of the cases and 90% (n = 38) of the controls were on HAART. No significant difference was found in CD4 or plasma HIV RNA at nadir/peak, commencement of HAART, or at 3, 6, or 12 months prior to HBV seroreversion; 15% (n = 3) of cases and 45% (n = 18) of controls were HBeAg⁺ during follow-up; 25% (n = 3) and 8% (n = 3) of the case and controls tested positive for HCV antibody. Alanine aminotransferase was significantly lower in cases than in controls at time of seroreversion and 3, 6, and 12 months prior (24 vs 50 cells/µL).

Conclusions:  Our study provides preliminary evidence that potent HIV and HBV antiviral treatment is associated with seroreversion of HBV. Ongoing prospective studies will further elucidate predictive factors and long-term outcome.

*p<0.05; **p<0.01; ^interquartile range