Persistent Iron Deficiency and Dysregulated Hepcidin Levels After Durable Left Ventricular Assist Device Therapy

Iron deficiency (IronDef) is prevalent in >50% patients with heart failure (HF)1 and is associated with poor exercise capacity, quality of life, and survival.2 Iron metabolism is regulated by hepcidin, a hepatically derived peptide hormone. In iron-replete states, hepcidin inhibits gastrointestinal iron absorption and iron release from reticuloendothelial cells.2 Hepcidin is upregulated by inflammatory cytokines regardless of iron stores, contributing to functional IronDef. HF is associated with increased inflammation and inappropriate hepcidin upregulation.2 High hepcidin levels predict unresponsiveness to oral iron repletion.1 Absolute IronDef occurs with poor iron absorption due to gut edema, or with blood loss. Whether hepcidin dysregulation persists after left ventricular assist device (LVAD) implantation is unknown. We hypothesized that, despite improvement in congestion and predisposition to IronDef from blood loss, LVAD patients have persistent hepcidin dysregulation. We retrospectively reviewed LVAD patients who consented to have plasma samples stored for our institution's Biobank. Hepcidin enzyme-linked immunosorbent assays were performed within 2 weeks before and at 1, 3, and 6 months after LVAD. Patients with plasma samples at ≥3 points were included. We queried medical records for concurrent (±2 months) measurements of ferritin, transferrin saturation (Tsat), and other laboratory values. IronDef was defined as ferritin <100 ng/mL (absolute IronDef), or ferritin = 100–299 ng/mL and Tsat <20% (functional IronDef).1 Values were expressed as median (25th–75th interquartile range). The Mann–Whitney U or Kruskal–Wallis tests were used for comparisons across groups. Pearson's correlation was used to find relationships between variables. The value of p < 0.05 was considered statistically significant. Analysis was performed with STATA 15.1 (TX). The study was approved by our center's Institutional Review Board. Between April 2016 and April 2021, 167 LVADs were implanted, of which 87 patients had plasma samples collected. Baseline characteristics are summarized in Table 1A (Supplemental Digital Content 1, https://links.lww.com/ASAIO/A851). All patients received daily 81–325 mg aspirin and warfarin with target international ratios of 2.0–3.0. At 6 months postoperatively, 75 patients remained on LVAD (12 had heart transplant, LVAD explant, or death). Laboratory values are shown in Table 1B (Supplemental Digital Content 1, https://links.lww.com/ASAIO/A851. After LVAD, N-terminal pro B-type natriuretic peptide (NT-proBNP) decreased (pre-LVAD: 4348 pg/mL, 1 month: 2,406 pg/mL, 3 months: 1,802 pg/mL, 6 months: 1,050 pg/mL, p = 0.0001). Hemoglobin improved, and renal and liver functions remained within normal ranges. A subset of patients received oral or intravenous iron to treat IronDef (Table 1C, Supplemental Digital Content 1, https://links.lww.com/ASAIO/A851). Most patients (68, 78.2%) had packed red blood cell (PRBC) transfusions during index hospitalization (<14 days post-LVAD) (Table 1D, Supplemental Digital Content 1, https://links.lww.com/ASAIO/A851). Many required transfusions at postoperative days 14–30 (19, 21.8%), 30–90 (17, 19.5%), and 90–180 (10, 13.3%), and 42–66% of patients at these different time points met criteria for IronDef (Table 1E, Supplemental Digital Content 1, https://links.lww.com/ASAIO/A851). The mean hepcidin/Tsat ratio in the IRON OUT-HF trial was 50 ng/ml/%, correlating with refractoriness to oral iron repletion in patients with HF.1 Hepcidin levels were normalized to Tsat to take into account inhibition of hepcidin expression by IronDef. In our patients, median hepcidin/Tsat ratios remained >50 ng/ml/% (pre-LVAD: 103 ng/ml/%, 1 month: 94 ng/ml/%, 3 months: 70 ng/ml/%, 6 months: 56 ng/ml/%, p = 0.1) (Figure 1A). Hepcidin/Tsat ratios were higher in those with functional versus absolute IronDef (Table 2, Supplemental Digital Content 1, https://links.lww.com/ASAIO/A851). Changes in hepcidin between pre- and post-LVAD periods correlated with changes in NT-proBNP (r = 0.3, p = 0.0002), but negatively correlated with changes in Tsat (r = –0.2, p = 0.02). Hepcidin/Tsat ratios trended higher in patients with PRBC transfusions versus those without, suggesting that those with lower hepcidin levels may require less transfusions post-LVAD (Table 1F, Supplemental Digital Content 1, https://links.lww.com/ASAIO/A851). During the first 6 months, there were no statistically significant differences in hepcidin/Tsat ratios between patients with LVAD-related and non–device-related infections. There were also no significant differences in hepcidin/Tsat ratios in those with and without bleeding (Figure 1B).Figure 1.: Hepcidin/transferrin saturation (Tsat) over time in patients with and without complications. A: Hepcidin/Tsat ratios before and at 1, 3, and 6 months after LVAD implant. Hepcidin/Tsat ratio >50 ng/ml/% indicates refractoriness to oral iron supplementation. B: Hepcidin/Tsat ratios in patients with and without gastrointestinal or any bleeding and infection within 6 months after LVAD. All comparisons between hepcidin/Tsat ratios in patients with and without events had p > 0.05. LVAD, left ventricular assist device.A subset of patients had right heart catheterization at 3 and 6 months post-LVAD. Right atrial pressure (RAP) and pulmonary artery wedge pressure (PAWP) were low, consistent with improved congestion (3 months [n = 25]: RAP = 9 [6–11] mmHg, PAWP = 12 [9–17] mmHg; 6 months [n = 29]: RAP = 8 [5–11] mmHg, PAWP = 12 [7–16] mmHg). Decreased iron absorption from venous congestion would be less likely. Hepcidin/Tsat ratios remained >50 ng/mL/% in these patients (3 months: 84.5 [27.5–183] ng/mL/%; 6 months: 53.3 [15–157.6] ng/mL/%). LVAD patients are at risk of blood loss from gastrointestinal bleeding, contributing to IronDef. Our study proposed hepcidin dysregulation as an important mechanism of IronDef in LVAD patients. Higher hepcidin levels correlated with lower Tsat and would render patients unresponsive to oral iron. We observed no differences in hepcidin/Tsat ratios among those on or off oral iron. Poor iron absorption due to gut edema is less likely with NT-proBNP decrements, normal renal and hepatic function, and normal intracardiac filling pressures. Despite improved survival and reduced HF severity, LVAD patients have impaired exercise capacity.3,4 Previous studies showed that intravenous iron improved exercise tolerance, quality of life, and decreased hospitalization for patients with HF.5–7 Our study suggests that efforts to improve exercise tolerance for LVAD patients should include screening for IronDef and correction with intravenous iron to circumvent hepcidin dysregulation and poor absorption of oral iron. Because of its retrospective nature, our study has limitations. Frequent data missingness for iron studies and Biobank samples may not be at random. We required ≥3 time points of available biospecimens per individual for analysis to limit the impact of missing data. Among individuals with plasma samples at all time points, median hepcidin levels were similar to those reported for the entire cohort (not shown). Inflammatory markers (C-reactive protein [CRP]) were not consistently measured. We could not correlate hepcidin dysregulation to inflammation after LVAD. Bleeding and PRBC transfusion can increase hepcidin expression. There was a nonsignificant trend of increased hepcidin/Tsat ratios in those with transfusion, but no difference in patients with and without bleeding within 6 months post-LVAD (Table 1F, Supplemental Digital Content 1, https://links.lww.com/ASAIO/A851, Figure 1B). Hepcidin expression could theoretically increase due to infection, yet no difference was seen in hepcidin/Tsat ratios in patients with and without infection, underscoring the possibility that hepcidin dysregulation may be primarily driven by inflammation and exacerbated by blood loss. The study was conducted in a small cohort at a single center. More research is needed before data are generalized, and implications for treatment with intravenous iron should be tested in clinical trials. In conclusion, hepcidin dysregulation is an important mechanism of IronDef in LVAD patients. IronDef should be screened for and when present, consideration should be given to circumventing hepcidin excess via intravenous iron.