Hepcidin-25 among Patients Undergoing Conventional and Extended-hours Hemodialysis; An Inverse Probability of Treatment Weighting Analysis Using Propensity Scores

In patients on maintenance hemodialysis (HD), elevated interleukin 6 (IL-6) levels promote hepcidin synthesis and cause anemia due to impaired iron utilization. Extended-hours HD has been suggested to reduce erythropoietin dosage by improving iron utilization efficiency; however, the association of extended-hours HD with hepcidin remains unclear. The aim of this study is to investigate the association of extended-hours HD with hepcidin, and to assess its subsequent effects on iron metabolism.

Patients on extended-hours or conventional HD between January and March 2020 were included in the study. Hepcidin-25 and IL-6 levels were measured from pre-dialysis samples. Multivariable linear regression analysis was performed to investigate the impact of extended-hours HD on hepcidin-25 and iron metabolism–serum ferritin levels and transferrin saturation (TSAT). Subsequently, mediation analysis was conducted to evaluate the mediating effects of iron metabolism and hepcidin-25, respectively. To minimize the impact of differences in target values for iron metabolism markers across dialysis modalities, propensity score (PS)–inverse probability weighting (IPW) analysis was conducted as a sensitivity analysis. PS were estimated using age, sex, dialysis vintage, iron supplementation, presence of dyslipidemia, log-transformed (ln-) ferritin and ln-TSAT. Restricted cubic spline (RCS) curves were used to evaluate the non-linear associations between hepcidin-25 and ferritin, TSAT, and IL-6, and interactions with extended-hours HD were tested using Wald’s tests.

A total of 380 participants (183 on extended-hours HD and 197 on conventional HD) were included in the analysis. The median ferritin and TSAT levels were 35.3 ng/dL and 16.9 %, respectively, in extended-hours HD group, and 63.8 ng/dL and 20.5%, respectively, in conventional HD group. Extended-hours HD was associated with lower hepcidin-25 levels compared to conventional HD, despite similar IL-6 levels (Figure 1A, B). Extended-hours HD was associated with lower ln-hepcidin-25; -0.57 (95% confidence interval (CI); -0.85 to -0.28), and 35% of this association was mediated by ln-ferritin (P <0.01). Before adjustment for hepcidin-25, extended-HD was significantly associated with lower ln-ferritin; -0.43 (95% CI; -0.62 to -0.23); however, this association was fully mediated by ln-hepcidin-25 (total mediation proportion was 69%). Extended-hours HD was not significantly associated with lower ln-TSAT (Table 1). For the fit of PS analysis, the C-statistics of logistic regression analysis for PS was 0.80, and absolute standardized difference values were less than 0.1 for all covariates. In the PS-IPW analysis, the association between extended-hours HD and hepcidin-25 remained similar. Using the RCS function, extended-hours HD was associated with lower hepcidin-25 levels compared to conventional HD, independently of iron metabolism. The analysis also suggested that, in extended-hours HD, the increase in hepcidin-25 in response to elevated IL-6 levels was attenuated (Figure 2A, B, C).

Extended-hours HD was associated with lower hepcidin-25 levels independently of iron supplementation, ferritin, and TSAT, even after PS-IPW analysis. These results were suggested that extended-hours HD might improve iron utilization by suppressing the hepcidin-25 response to elevated IL-6 levels.