T1015 Rate and Predictors of Erythropoietic Growth Factors Use in Chronic Hepatitis C Patients with Treatment Induced Anemia

Abstract

analyses to clarify HCV's role in DM risk. Methods: Two reviewers performed searches in multiple databases and abstracted data. Eligibility criteria included: 1) minimum of 200 adults-100 cases, 100 controls; 2) published in English as peer-reviewed manuscript. Studies with only post-transplant data were excluded. We performed random effects meta-analyses that combined studies with similar risk measures and similar control groups. Egger's test was used to assess potential small study bias. Results: We identified 31 unique studies. Many were performed in U.S. (n=10) or Italy (n=6). Control groups included: non-HCV/ HBV infected in 16 studies, and HBV-infected, with other liver diseases (OLD), and HIV mono-infected in n=8, n=3 and n=4 studies respectively. Among studies comparing DM risk in HCV cases to that in non-infected controls, 13 reported odds ratios (OR) and 3 longitudinal studies reported hazard ratios (HR). Pooled estimators for studies with ORs indicated excess risk (ORcrude=2.11, 95% CI 1.48-2.74, ORadjusted=1.76, 95% CI 1.08-2.44). 5 adjusted estimates included BMI, with 4 finding additional excess DM risk. Although Egger's test was significant, it was explained by 1 study's effect. Excess DM risk was also observed by pooling HRs (HRadjusted=1.67, 95% CI, 1.28-2.06). All adjusted HRs included BMI with no evidence of small study bias. Among studies with HBV+ controls, both pooled ORs suggested excess, though the adjusted effect was no longer significant ORcrude=1.69, 95% CI 1.15-2.22, ORadjusted=1.99, 95% CI 0.92-3.06). Although lack of comparability in effect measures (HIV) or in controls (OLD) prevented meta-analysis, most individual studies reported excess DM risk with HCV that was particularly suggestive for HIV. Conclusions: In this meta-analysis, pooled estimators indicated HCV infection increases DM risk, with suggestive evidence that the excess risk persists even after controlling for other risk factors like BMI. The meta-analysis of prospective studies reporting HRs in particular bolsters argument that HCV conveys excess risk. Additional research will be needed to identify other factors that mediate DM risk in HCV-infected cases in comparison to non-infected or HBVinfected controls.