HIV coinfection shortens the survival of patients with hepatitis C virus–related decompensated cirrhosis


  • Potential conflict of interest: Nothing to report.

HIV Coinfection Shortens the Survival of Patients With Hepatitis C Virus–Related Decompensated Cirrhosis

To the Editor:

We read with interest the article by Pineda et al. in which HIV coinfection as an independent risk factor for decreased survival in patients with hepatitis C virus (HCV)-related cirrhosis was identified.1 The authors estimated that the 1-year and 5-year survival rates were 74% and 44%, respectively, for HCV-monoinfected patients and 54% and 25%, respectively, for HIV/HCV-coinfected patients. Interestingly, liver-related deaths occurred with similar frequency in HCV-monoinfected patients (86%) and in HIV/HCV-coinfected patients (81%). Why HIV coinfection reduces survival in HCV patients who have cirrhosis remains a critical question.

The authors note that 43% of HIV-coinfected patients were on highly active antiretroviral therapy (HAART), but they do not indicate how many of the HIV/HCV-coinfected patients who died were on HAART. HCV coinfection increases the chance of HAART-associated hepatotoxicity by 2- to 10-fold compared with HIV-monoinfected patients.2 It is not fully clear why HCV coinfection contributes to hepatotoxicity in HIV patients on HAART, but the HCV core protein may play a role. Glutathione has mitochondrial and cytoplasmic antioxidant properties that may offer protection against drug-induced liver injury, and the HCV core protein decreases intracellular glutathione stores.3 HIV infection is also associated with depletion of glutathione, and HIV and HCV may synergistically increase susceptibility to drug-induced hepatotoxicity.4 Furthermore, HAART and the HCV core protein have been implicated in causing mitochondrial toxicity, which can lead to dyspnea, abdominal pain, weight loss, vomiting, cardiomyopathy, pancreatitis, and lactic acidosis.5 Such sequelae of HAART may have a more deleterious effect in patients who have advanced liver disease. Identifying how many deceased patients took HAART and whether they took HAART when they had more histologically advanced liver disease could determine the contribution of HAART to the increased risk of death.

The authors also note that hepatic encephalopathy more frequently caused first hepatic decompensation and liver-related death in HIV/HCV-coinfected patients than in HCV-monoinfected patients. Although the authors suggest that HIV patients may have been more likely to take central nervous system depressants through substance abuse, one cannot overlook the effect of HIV infection itself on cognition. HIV neuropathology includes cerebral atrophy and diffuse white matter damage with myelin loss.6 Clinical manifestations include difficulties with concentration and memory loss.7 Furthermore, other neurological sequelae of HIV/AIDS—including primary central nervous system lymphoma, progressive multifocal leukoencepalopathy, and opportunistic infections—remain important causes of HIV-related neurological deficits.6, 8 Pineda et al. defined hepatic encephalopathy as patient confusion or disorientation “in the absence of any other likely nonhepatic causes.”1 Greater detail of the criteria used to distinguish hepatic encephalopathy from HIV-related neuropsychiatric deficits would perhaps clarify if encephalopathy in these patients was due solely to underlying end-stage liver disease.

The study by Pineda et al. represents an important contribution to the body of existing data on HIV/HCV coinfection. Despite its retrospective study design, it raises important questions about why HIV increases mortality in HIV/HCV coinfected patients, thus providing a foundation for much-needed prospective studies on the interplay of HIV and HCV infection in liver disease.

Carla W. Brady M.D.*, Andrew J. Muir M.D., M.H.S.†, * Division of Gastroenterology, Durham VA Medical Center, Durham, NC, † Division of Gastroenterology, Duke Clinical Research Institute, Duke University Medical Center, Durham, NC.