A 36-year-old African woman presented in November 2002 with advanced HIV disease. Her baseline CD4+ T-cell count was 20 cells/mm3 (reference range 312-2064) and HIV-1 RNA 530,000 copies/mL (Roche Amplicor HIV-1 Monitor; dynamic range 400-750,000). The patient had no other risk factors for liver disease, serologic tests for hepatitis B and C were negative, and baseline liver function tests were within normal limits. Combination antiretroviral therapy was initiated with lamivudine, tenofovir, and nevirapine.
She was monitored at 3-monthly intervals uneventfully until May 2004. At that time, the CD4 count was 192 cell/mm3 and HIV-1 RNA <50 copies/mL. Elevation of liver enzymes was noted with AST 101 U/L (normal 15-46) and ALT 110 U/L (normal 11-66) and, during the following weeks, rose to a peak AST of 192 and ALT 226. Previously normal, albumin fell to 2.7 g/dL (normal 3.6-5.0) and prothrombin time was 19.5 seconds (normal 12.0-14.6) with an INR of 1.6. The subsequent evaluation included repeat serologic and virologic tests for hepatitis A, B, and C, which were all negative. An antinuclear antibody screen was negative, smooth muscle antibody IgG titer was 1:20 (negative <1:20), and liver kidney microsome antibodies were negative at <1:20. The liver was normal in size and echo texture on ultrasound. During this evaluation, antiretroviral therapy was interrupted because of concern for potential drug-induced hepatotoxicity.
A liver biopsy was performed which revealed extensive portal and periportal fibrosis with evidence of bridging (Fig. 1). The portal regions contained a mixed inflammatory infiltrate composed of lymphocytes and plasma cells with evidence of interface activity and necrosis. No significant cholestasis or steatosis was found.
A provisional diagnosis of autoimmune hepatitis was made and treatment started with prednisone. The liver function tests quickly normalized and after 6 weeks of prednisone, antiretroviral therapy was reintroduced with lamivudine and tenofovir as before, but fosamprenavir with ritonavir rather than nevirapine. Although drug-induced hepatic toxicity appeared unlikely, the treating clinicians felt is was prudent to exclude nevirapine from subsequent regimens. Prednisone was tapered over 16 weeks without apparent consequence until the liver enzymes rose again 2 weeks after discontinuation, which further supports the diagnosis of autoimmune hepatitis. Prednisone was restarted with a slower taper planned after her liver enzymes have normalized.
Evaluation of elevated liver enzymes in HIV-infected patients is often complex. Causes include primary hepatic disease, such as viral hepatitis, and drug toxicity, particularly from antiretroviral agents. In some instances, a combination of these elements is present. Moreover, the presence of opportunistic infections, alcohol and other drug abuse, metabolic complications of HIV disease and its therapies, and the use of complementary therapies may complicate the picture.1 In this case, a careful evaluation, including liver biopsy, revealed a treatable and unrelated cause of liver disease.
Although most cases involve asymptomatic elevations of liver enzymes, more severe and even fatal hepatotoxicity has been reported with nevirapine.2,3 This nonnucleoside reverse transcriptase inhibitor is used in many first-line regimens, particularly in resource-limited settings, and for the prevention of perinatal transmission. However, the acute hepatic toxicity associated with nevirapine has led to FDA public health advisories and changes to the product labeling in the United States and in Europe.4 These include recommendations against use of the drug in women with CD4 cell counts greater than 250 cells/mm3 who seem to be especially at risk. Hepatic events typically occur within the first 6 weeks of therapy, although late-onset hepatitis has been reported.5 Overall, postmarketing surveillance data do not support greater incidence of hepatic events after 18 weeks of therapy than with comparator drugs.
This case illustrates some of the difficulties encountered when evaluating elevated liver enzymes in a patient with HIV disease receiving nevirapine therapy. Data to guide clinical decision making are lacking with regard to antiretroviral therapy, especially with late-onset hepatitis. Most clinicians feel it is prudent to withdraw potentially hepatotoxic medications whenever possible. This case also illustrates the benefits of a careful evaluation, preferably with hepatology expertise available, and examination of liver histology.
1. Dieterich D. Managing antiretroviral-associated liver disease. J Acquir Immune Defic Syndr
. 2003;34(suppl 1):S34-S39.
2. Bartlett J. Severe liver toxicity in patients receiving two nucleoside analogues and a non-nucleoside reverse transcriptase inhibitor. Presented at the 8th Conference on Retroviruses and Opportunistic Infections, Chicago, IL, USA, February 4-8, 2001 [abstract 19].
3. Sanne I, Mommeja-Marin H, Hinkle J, et al. Severe hepatoxicity associated with nevirapine use in HIV-infected subjects. J Infect Dis
4. Boehringer Ingelheim Pharmaceuticals. Viramune (nevirapine) package insert. January 2004.
5. Clarke S, Harrington P, Condon C, et al. Late onset hepatitis and prolonged deterioration in hepatic function associated with nevirapine therapy. Int J STD AIDS