Transferrin saturation is a superior prognostic biomarker for iron deficiency after acute decompensated heart failure: a retrospective clinical real-world cohort study

Abstract Background Iron deficiency (ID) is the most common extracardiac comorbidity in heart failure (HF) and associated with worse outcomes. While a uniform definition of ID in HF remains lacking, several biomarkers for ID, including ferritin and transferrin saturation (TSAT), are used in both clinical trials and clinical practice. However, which biomarker is a superior prognostic predictor for ID in HF remains unsettled. In addition, real-world clinical data on ID screening and intravenous (IV) iron treatment is largely missing. Purpose To determine which biomarker is a superior prognostic predictor for the composite endpoint of HF rehospitalisation (HHF) and all-cause mortality in a real-world cohort of patients with HF and coexisting ID, and to investigate frequency and predictors of ID screening and IV iron treatment. Methods In this retrospective cohort study, all patients aged 18-85 years hospitalised at a tertiary university hospital with new onset HF with reduced ejection fraction (HFrEF) in 2016-2020 were consecutively included from medical records by ICD-10 code I50.0-I50.9 and followed until 31 December 2021. Patient characteristics, comorbidities, medications, laboratory results and data on follow-up, including if ID was tested for, IV iron administration, HHF and all-cause mortality was collected. First available iron status after admission for index hospitalisation was used. Predictors for ID screening and IV iron use were identified with univariate logistic regression. The association between ID markers and the composite endpoint was analysed with Cox regression adjusted for age, sex, baseline anemia, estimated glomerular filtration rate (eGFR) <30 mL/min/1.73 m2 and N-terminal pro-B type natriuretic peptide (NT-proBNP) below median value with median ferritin and TSAT values set as references respectively. Results In all, 753 patients were included (66.5 ± 12.7 years, 508 (67.5%) men, ejection fraction 29.5 ± 6.9% and mean NT-proBNP 6241 ± 7183 ng/L). Of these, 484 (64.3%) were screened for ID and 235 (48.6%) fulfilled criteria for ID. Of these, 94 (40.0%) received IV iron. Of screened patients, 28.6% had ID with anemia and 20.8% ID without anemia (Table 1). Factors associated with screening for ID were follow-up at hospital, anemia, eGFR <30 mL/min/1.73 m2 and ischemic heart disease, and for IV iron additionally chronic kidney disease and peripheral artery disease. Median ferritin was 140 µg/L and TSAT 0.2. Both ferritin and TSAT displayed an inversely linear association with outcomes, while TSAT <0.2 showed a stronger association with increased risk of HHF and all-cause mortality compared to median values (Figure 1-2). Conclusions In a real-world clinical setting, TSAT was a superior prognostic biomarker for worse outcomes compared to ferritin in patients hospitalised for acute HF. While both ID screening and IV iron use remained greatly underutilised in clinical practice, identification of a reliable biomarker for ID is critical.