Influence of Hepatitis C Virus Co-Infection and Hepatitis C Virus Treatment on Risk of Chronic Kidney Disease in HIV-Positive Persons

Amanda Mocroft; Lene Ryom; Cristiana Oprea; Qiuju Li; Andri Rauch; Christoph Boesecke; Vilma Uzdaviniene; Dalibor Sedlacek; Josep M. Llibre; Karine Lacombe; Lars N. Nielsen; Eric Florence; Inka Aho; Nikoloz Chkhartishvili; János Szlavik; Gordana Dragovic; Clifford Leen; Helen Sambatakou; Therese Staub; Montse Laguno; Hila Elinav; Janez Tomažič; Lars Peters


AIDS. 2020;34(10):1485-1495. 

In This Article


The EuroSIDA Study

Persons were included from the EuroSIDA study, a large prospective observational cohort of almost 23 000 HIV-1-positive patients followed in 100 hospitals in 35 European countries with Israel and Argentina. Individuals were enrolled into 10 cohorts from 1994 onward. In cohort 10, all HIV-positive patients were also required to be positive for anti-HCV antibodies (HCV-RNA positive, negative or unknown status). At recruitment, in addition to demographic and clinical data, a complete ART history was obtained together with the most recent CD4+ cell counts and HIV-RNA measurements, as well as all HCV tests, HCV-RNA, HCV genotype, hepatitis B surface antigen (HBsAg), and HBV-DNA. Data is collected prospectively at clinical sites and sent to the coordinating centre at yearly intervals. At each follow-up visit, all CD4+ cell counts, HIV-RNA, HCV tests, HCV-RNA, genotype, and HBsAg results measured since last follow-up are collected, together with start and stop dates for antiretroviral drugs and HCV and HBV drugs. Detailed information about data collected in EuroSIDA can be found at

Methods and Definitions

CKD was defined as a confirmed (>3 months apart) eGFR less than 60 ml/min per 1.73 m2 for those with first eGFR greater than 60 ml/min per 1.73 m2 and a confirmed (>3 months apart) 25% decline in eGFR for those with baseline eGFR 60 ml/min per 1.73 m2 or less. eGFRs were calculated using the CKD-EPI formula.[17] All persons with known HCV serostatus and prospective follow-up after 1 January 2004 (start of standardized collection of serum creatinine) were eligible for inclusion. Persons with less than three eGFRs during prospective follow-up were excluded, as were persons with less than 3 months follow-up. Baseline was defined as the first prospective visit in EuroSIDA after 1 February 2004 at which both eGFR and HCV serostatus were measured, and where HCV-RNA was known for those anti-HCV-positive. Persons aged less than 16 at baseline or without a CD4+ cell count and HIV viral load in the 12 months before or 1 month after baseline were excluded.

On the basis of time-updated HCV antibody tests, HCV-RNA and HCV treatment, we defined five HCV groups

  1. Anti-HCV-negative

  2. HCV antibody-positive, HCV-RNA-negative, untreated (spontaneous clearers)

  3. HCV antibody-positive, HCV-RNA-positive, untreated (chronic infections)

  4. HCV antibody-positive, HCV-RNA-negative, treated (successfully treated with any HCV therapy; cured)

  5. HCV antibody-positive, HCV-RNA-positive, treated (treated, HCV-RNA positive)

All groups anti-HCV positive were defined on the basis of a single HCV-RNA measurement; for example, persons were classified as spontaneous clearers based on the latest value of HCV-RNA. Those HCV-RNA positive after treatment included persons who did not achieve SVR, persons without an end of treatment response, persons who were HCV-RNA-positive having started treatment more recently and those reinfected with HCV. Persons were followed until their last visit (median June 2018), date of death, or CKD, whichever occurred first. Person-years of follow-up (PYFU) and CKD events accrued according to current HCV strata using the last observation carried forward and persons could contribute PYFU to multiple groups.

In those that developed CKD, we performed an exploratory analysis looking at reversal of CKD. This was defined as a confirmed (> 3 months apart) increase in eGFR to greater than 60 ml/min per 1.73 m2 among persons with at least two further eGFRs and 3 months follow-up after CKD. Baseline for this analysis was date of developing CKD, and individuals were followed to the first of reversal of CKD or last eGFR.

Statistical Analysis

Characteristics of individuals were compared across strata using chi-squared statistics for categorical variables and the Kruskall--Wallis test for continuous variables. Incidence rates of CKD per 1000 PYFU were calculated within HCV groups, and Poisson regression was used to compare these rates with those cured as the reference group. Different models were investigated; the first adjusted only for the Data Collection on Adverse events of Anti-HIV Drugs (D:A:D) study CKD risk score,[18] without including the component because of HCV coinfection. Liver fibrosis stage (as previously described;[19] this was included as a baseline measurement as it may lie on the causal pathway between HCV status and CKD) and the HCV strata defined above were also included in this model. As the D:A:D CKD risk score does not include all the variables, which differed between the HCV strata, we also investigated a more extensive model adjusting for many more potential confounding variables. This second model adjusted for a greater number of potential confounding factors, all fixed at baseline (sex, HIV exposure group, region of Europe (North, Central West, South, Central East, East and Argentina[20]), eGFR, HIV viral load, prior AIDS, cardiovascular disease, non-AIDS defining malignancies (NADM), end-stage liver disease (ESLD; ascites, hepatorenal syndrome, grade III/IV hepatic encephalopathy, unspecified liver decompensation, oesophageal variceal bleeding, spontaneous bacterial peritonitis, liver transplantation and hepatocellular carcinoma). Further information about these events is available at We also adjusted for smoking status (never smoked, current smoker, past smoker, unknown smoking status), hypertension, BMI, use of nephrotoxic antiretrovirals [antiretrovirals: tenofovir, atazanavir (unboosted and/or ritonavir-boosted), indinavir, and lopinavir], use of nephrotoxic drugs (foscarnet, acyclovir, pentamidine, cidofovir, amphotericin B), CD4+ cell count, nadir CD4+, age, liver fibrosis, and baseline date. A third model adjusted for baseline liver fibrosis and the components of the D:A:D CKD risk score (including use of nephrotoxic antiretrovirals and HCV status as defined in this study) at baseline as separate variables rather than a composite score. The model was additionally adjusted for starting integrase inhibitors, shown to increase serum creatinine levels,[21] as a time updated variable. As results were consistent across models, our results focus on model 3, which had the lowest Akaike Information Criterion.

We performed a wide range of sensitivity analyses to investigate the robustness of our results to different assumptions. We performed a sensitivity analysis where the last HCV-RNA measurement was carried forward for a maximum of 12 months. This reduces the bias from HCV-RNA measurements measured many years previously being used to stratify persons into HCV strata. We also excluded persons with stage F3/F4 liver fibrosis at baseline, as well as PYFU and CKD events occurring after the development of F3/F4 liver fibrosis in the subgroup of persons at high risk for CKD using the D:A:D CKD risk score,[18] and an analysis limited to after 2014, when DAAs became more widely available for persons included in the EuroSIDA study.[22] We also explored a more rigorous definition of CKD as a confirmed 25% decline to less than 60 ml/min per 1.73 m2.[1] We repeated our analyses separately among those treated and cured or HCV-RNA-positive after treatment in those not exposed, or only exposed, to DAA-based regimens.

All analyses were performed in SAS version 9.4 (Statistical Analysis Software, Cary, North Carolina, USA).