Soluble transferrin receptor as a marker of tissue iron deficiency: impact on cardiac, systemic, hematinic and neurohormonal biomarkers in patients with heart failure without systemic iron deficiency

Abstract

Abstract Background Soluble receptor of transferrin (sTfR) is a marker of tissue iron status and may help to inform on subtle iron depletion and increased iron demand at tissular level even in the absence of overt systemic iron deficiency or anaemia. In this regard, the impact of raised sTfR levels as a marker of subtle tissue iron deficiency (ID) on serum biomarkers in non-anaemic HF patients with otherwise normal systemic iron status has not been evaluated. Purpose The aim of our study was to describe the association between sTfR as a marker of increased iron demand and tissue iron deficiency on cardiac, systemic, neurohormonal and hematinic biomarkers in non-anaemic patients with HF and normal systemic iron status. Methods We conducted an observational, prospective, cohort study of 1120 consecutive patients with chronic HF regardless the level of LVEF (DAMOCLES study). Patients were included if they had normal haemoglobin levels (≥12 g/dL), normal systemic iron status (serum iron >33 µg/dL, ferritin >100ng/mL and % transferrin saturation >20%) and available biomarker data. sTfR was measured with the Beckman Coulter enzyme immunoassay (higher levels indicate increased iron demand). The primary objective was to explore the association between sTfR and NT-proBNP levels. Secondary objectives were to explore the association between sTfR, and a wide array of serum biomarkers. Pearson’s correlation coefficient (r) was used to estimate correlations in bivariate analyses. The sex-and-age adjusted associations between sTfR, and the biomarkers were explored using General Additive Models (GAM) and linear regression models. Results The final study cohort consisted in 215 patients. Mean age was 70±12 years, mean LVEF was 43±15% and 62 (29%) were women. Mean sTfR values were 1.42±0.66 mg/L. In the whole cohort, median [Q1-Q3] NT-proBNP (pg/mL) was 1,125[587-2,668]. Log-transformed sTfR (log[sTfR]) showed significant correlations with log[NT-proBNP] (r=0.230; p-value=0.001), log[cTnT] (r=0.197; p-value=0.028), log[albumin](r=-0.221; p-value=0.001), C-reactive protein (r=0.215; p-value=0.002), serum erythropoietin levels, MCH (r=-0.247; p-value<0.001) and MCHC (r=-0.209; p-value<0.001). As shown in Table 1, these findings were confirmed in multivariate age-and-sex adjusted linear regression models. In age-and-sex adjusted Generalized Additive Models (GAM) (Figure 1), we confirmed that higher levels of sTfR were associated with increased levels of biomarkers indicating cardiac damage (cTnT, NT-proBNP), active inflammatory status (C-reactive protein) and increased cellular stimulation in response to tissue hypoxia (endogenous erythropoietin). Conclusions In a cohort of HF patients without iron deficiency or anaemia, higher levels of sTfR indicating increased iron demand and tissue ID were associated with a worse biomarker profile indicating early subclinical cardiac and/or systemic damage even in the absence of overt systemic iron deficiency or anaemia.