Severe liver inflammation and liver failure in people taking HAART is due to the effects of infection with hepatitis C or B, rather than the toxicities of anti-HIV drugs, according to a study conducted at the University of Brescia and published in the March 2003 edition of the Journal of Acquired Immune Deficiency Syndromes.
Between March 1997 and June 1998, 755 HIV-positive patients who were prescribed anti-HIV therapy at the HIV clinic at the University of Brescia were included in a study to determine the rate of severe hepatotoxicity and outcome of severe hepatotoxicity in patients on HAART.
At baseline, all patients were ask to complete a questionnaire regarding alcohol consumption, under went a physical examination and routine liver tests (ALT, AST, bilirubin, albumin and prothrombin time), had HIV viral load and CD4 cell measured and were checked for markers of infection with hepatitis viruses. HAART was independently prescribed by the patient’s physician and follow-up was provided within a month of starting HAART and then every four months when all patients had liver function tests.
The results of liver function tests were categorised using a scale determined by the US AIDS Clinical Trials Group (ACTG): relevant hepatotoxicity; severe hepatotoxicity; and, liver failure.
If severe hepatotoxicity was diagnosed, patients were asked about drug and alcohol consumption, had a liver ultrasound and body scan, and tests to determine blood lactic levels and for hepatitis B, C and D. Patients were also investigated for opportunistic infections which can affect the liver.
Liver biopsies were performed within three months on 16 patients with evidence of severe hepatotoxicity. HAART was stopped in patients with severe hepatotoxicity, but restarted in all patients after a four month “wash-out” if they showed normalisation of liver function.
In total, 105 cases of relevant hepatotoxicity were observed (17 cases per 100 person years). The only variable found to be significantly associated with relevant hepatotoxicity were anti-hepatitis C reactivity and elevated ALT levels.
Twenty-six of the 105 cases of relevant hepatotoxicity met the definition of severe hepatotoxicity (a rate of 4.2 per 100 person years). Seven of these patients went on to develop liver failure (rate 1.1 per 100 person years).
Treatment regimen did not significantly affect the incidence of severe hepatotoxicity (2 NRTIs and a PI: 4 per 100 person years; 6 per 100 person in people treated with two NRTIs; and, no cases in people treated with an NNRTI and two NRTIs). However, as the investogators themselves acknowledge, there were only 62 treatment episodes of NNRTI therapy, amounting to 41 person years of follow-up, compared to 716 PI treatment episodes amounting to 478 person years.
Patients with severe hepatotoxicity were more likely to be male; to have acquired HIV through injecting drug use; be younger; be coinfected with hepatitis B, C or D; and have abnormal liver function markers at baseline.
Evidence of hepatitis C virus was present in 25 of the 26 patients with severe hepatotoxicity and one person also hepatitis B antigens. In all 26 patients, a direct relationship between an increase in ALT levels and increase in CD4 cell count from baseline was observed.
Liver biopsies were performed on 16 patients, with the results suggesting damage caused by immune reaction. In two patients, liver biopsies had been performed before anti-HIV therapy was initiated, with the post-HAART biopsies showing a worsening of inflammation.
The only patients with severe hepatotoxicity to die were the seven with liver failure. In six of these patients the liver failure was irreversible and they were not subsequently treated with HAART, dying of end stage liver failure within three months to one year. In one patient, liver failure was associated with lactic acidosis and pancreatitis and the patient died within three days. However, one patient did show improvement in liver function after HAART was withdrawn, but as his CD4 cell count was below 50 cells/mm3 he developed several opportunistic infections and it was necessary to recommence anti-HIV therapy, the patient dying of liver failure within a further three months.
Information was available on 17 patients with severe hepatotoxicity but not liver failure. HAART was stopped for a four-month “wash-out” period and then restarted. Seven patients (41%) relapsed to severe liver toxicity within two to four weeks of treatments restarting. However, in the remaining patients, no evidence of relapse was observed in 20 – 32 months of treatment.
In the seven patients who did relapse, five, all of whom had HCV, were treated with alpha interferon. One patient showed clearance of HCV and normalised ALTs. The remaining four were treated with HAART during alpha interferon therapy and did not show a relapse to severe hepatotoxicity within 30 months.
The investigators note that the 17% incidence of liver toxicity recorded in their study was similar to that seen in earlier investigations. However, they were concerned primarily with the 4.2% rate of severe liver toxicity and 1.1% rate of liver failure.
They suggest that immune reconstitution reaction against hepatitis virus antigens in the liver played a major role in the occurrence of severe hepatotoxicity. Indeed, severe liver damage could have been related to the “shock caused by immune reconstitution” against hepatitis viruses in people with severely damaged immune systems prior to starting HAART.
The investigators did not, therefore believe that anti-HIV drugs themselves were responsible for the severe liver toxicities seen in the study, noting that 59% of patients who had stopped HAART due to liver problems were able to restart therapy with the same regimens after a “wash out” period.
”Severe hepatotoxicity was related to preexisting chronic viral hepatitis” conclude the investigators, recommending that people starting HAART who are coinfected with hepatitis and have low CD4 cell counts receive careful monitoring. However they add, that according to their data, the risk of liver toxicity resulting in liver failure was low for coinfected patients with a CD4 count between 350 and 200 cells/mm 3. As regards the patients who developed liver failure, they suggest that liver failure would have been best prevented by the early treatment of chronic hepatitis C whilst the immune system was still relatively intact.
Further information on this website
Hepatitis B - Overview
Hepatitis C - Overview
Hepatitis B - Factsheet
Hepatitis C - Factsheet
Draft BHIVA guidelines for treatment of HIV/HBV coinfection
Draft BHIVA guidelines for treatment of HIV/HCV coinfection
HIV and hepatitis - Booklet in the information for HIV-positive people series
Puoti M et al. Severe hepatotoxicity during combination antiretroviral treatment: incidence, liver histology, and outcome. Journal of Acquired Immune Deficiency Syndromes, 32: 259 – 267, 2003.