Study Overview
Objective. To compare the incidence of hepatic decompensation in patients who are co-infected with HIV and hepatitis C (HCV) and who underwent antiretroviral treatment and patients who are HCV-monoinfected.
Design. Retrospective cohort study.
Participants and setting . This study used the Veterans Aging Cohort Study Virtual Cohort (VACS-VC), which includes electronic medical record data from patients who are HIV-infected and are receiving care at Veterans Affairs (VA) medical facilities in the United States. Inclusion criteria for patients who were co-infected were: detectable HCV RNA, recently initiated antiretroviral therapy (ART), defined as use of ≥ 3 antiretroviral drugs from 2 classes or ≥ 3 nucleoside analogues within the VA system, HIV RNA level > 500 copies/mL within 180 days before starting ART, and were seen in the VACS-VC for at least 12 months after initiating ART. Inclusion criteria for patients who were monoinfected with HCV were detectable HCV RNA, no HIV diagnosis or antiretroviral prescriptions, and seen in the VACS-VC for at least 12 months prior to inclusion into the study. Exclusion criteria were hepatic decompensation, hepatocellular carcinoma, and liver transplant during the 12-month baseline period or receipt of interferon-based HCV therapy.
Main outcome measure. The primary outcome was incident hepatic decompensation, defined as diagnosis of ascites, spontaneous bacterial peritonitis, or esophageal variceal hemorrhage at hospital discharge or 2 such outpatient diagnoses.
Main results. A total of 10,359 patients met inclusion criteria and were enrolled between 1997 and 2010. Of these, 4280 were patients co-infected with HIV and HCV and treated with antiretroviral agents and 6079 were patients who were HCV-monoinfected. Age, race/ethnicity, and history of diabetes, alcohol dependence or abuse, and injection or non-injection drug were similar between the 2 groups. The majority of participants were men. HCV genotype 1 was most prevalent in both groups. There were more patients who had HCV RNA levels ≥ 400,000 IU/mL and/or ≥ 1x10 6 copies/mL in the co-infected group versus the monoinfected group.
Hepatic decompensation occurred more frequently among those who were co-infected and receiving ART (271 [6.3%]) than among those who were monoinfected (305 [5.0%], P = 0.004). The incidence rate was 9.5 events per 1000 person-years (95% CI, 7.6–11.9) among patients co-infected with HIV and HCV and treated with ART and 5.7 events per 1000 person-years (95% CI, 4.4–7.4) among patients who were monoinfected. Variceal hemorrhage was less common among patients who were co-infected as compared to those who were monoinfected (71 [26.2%] vs. 168 [55.1%], P < 0.001). The proportion of patients with ascites (226 [83.4%] in the co-infected group vs. 236 [77.4%] in the monoinfected, P = 0.070) and spontaneous bacterial peritonitis (48 [17.7%] in the co-infected group vs. 68 [22.3%] in the monoinfected, P = 0.171) were similar. After adjustment for age, race/ethnicity, diabetes, BMI, history of alcohol abuse, injection or non-injection drug use, and VA center patient volume, patients who were co-infected and receiving ART had a higher rate of hepatic decompensation than monoinfected patients (hazard ratio, 1.83 [95% CI, 1.54–2.18]).
In subgroup analysis, rates of decompensation remained higher even among co-infected patients who maintained HIV RNA levels < 1000 copies/mL (hazard ratio 1.65 [95% CI 1.20–2.27])
Conclusion. Patients who were co-infected with HIV and HCV and treated with ART had higher rates of hepatic decompensation compared with patients monoinfected with HCV. Good control of HIV viral loads in co-infected patients may not be sufficient to improve health outcomes.
Commentary
Currently, it is estimated that there are 3.5 to 5.5 million people in the United States infected with HCV, accounting for about 1.5% of the population. Approximately 20% to 40% of those infected will develop chronic infection and 10% to 25% of these patients will progress to experience severe liver disease [1]. Yet of the 3.5 million people who are thought be chronically infected with HCV, only 50% are diagnosed and are aware of the infection and a mere 16% are treated for HCV [2].
Estimates suggest that about 10% of those with HCV are also infected with HIV. In the era prior to ART for HIV infections, patients with HIV and HCV most commonly died of HIV-related causes. In the post-ART era, patients are surviving longer and are now experiencing HCV-related comorbidities [3].
This study compares the incidence of hepatic decompensation in patients with HIV and HCV co-infection who are undergoing treatment with ART and those with HCV monoinfection. The results show that patients who were co-infected and treated with ART had higher incidence of hepatic decompensation as compared with those who were monoinfected. This study’s strengths are the large enrollment numbers (> 10,000 patients) and the long follow-up periods (6.8 and 9.9 years for the co-infected and monoinfected cohorts, respectively). As the authors indicate, the weakness of this study is the exclusion of the diagnosis of hepatic encephalopathy and jaundice from their definition of hepatic decompensation. Their reasoning for doing so is that these frequently occur due to unrelated causes, such as narcotic overdose and biliary obstruction. It is possible that this resulted in an underestimation of hepatic decompensation. Finally, 98.8% of the enrolled patients were male. The study results cannot be generalized to women.
Since 2011, the availability of direct-acting antivirals for the treatment of HCV has rapidly increased. These new agents have improved treatment outcomes with better sustained virological response, shorter treatment duration, and better adverse event rates [4]. Telaprevir and boceprevir were first-generation protease inhibitors, and these were followed by simeprevir in 2013. Sofosbuvir also became available in 2013 as the first polymerase inhibitor. These agents were and continue to be evaluated for use in HIV/HCV co-infected patients both in treatment-naive and previously treated patients with good outcomes. A fifth agent, faldaprevir, another protease inhibitor, is expected to become available this year and others are in clinical trials [5]. Sustained virologic response rates of 67% to 88% depending on genotype with regimens using sofosbuvir in co-infected patients for example, have been achieved, which are similar to rates in monoinfected patients [6].
Applications for Clinical Practice
The authors found that management of HIV viral loads to less than 1000 copies/mL reduced the risk for hepatic decompensation. However, the difference in incidence rates between those whose HIV load was < 1000 copies/mL and those whose viral load was ≥ 1000 copies/mL was small (9.4 [95% CI, 5.4–16.2] vs. 9.6 [95% CI, 7.5–12.2]). The findings suggest that control of HIV viral loads in co-infected patients is not sufficient to reduce the rate of liver complications. The authors propose that earlier consideration be given to treatment of HCV infection in co-infected patients to improve health outcomes. The American Association for the Study of Liver Diseases and the Infectious Diseases Society of America have published guidelines for the diagnosis and management of HCV [7]. The difference in hepatic decompensation rates between mono- and co-infected patients should become less relevant as use of direct-acting antivirals expands.
—Mayu O. Frank, MS, ANP-BC and Allison Squires, PhD, RN, New York University College of Nursing