Background: The most commonly used definition of iron deficiency (ID; ferritin <100 ng/mL or ferritin 100-300 ng/mL and transferrin saturation [TSAT] <20%) has not been validated in patients with heart failure (HF). We aimed to define and validate the biomarker-based definition of ID in HF, using bone marrow iron staining as the gold standard. Second, we aimed to assess the prognostic value of the optimized definition.
Methods and results: Bone marrow aspiration with iron staining was performed in 42 patients with HF and a reduced left ventricular ejection fraction (≤45%) undergoing median sternotomy for coronary artery bypass grafting. Patients were mostly male (76%) with mild-to-moderate HF and a mean age of 68±10 years. Bone marrow ID was found in 17 (40%) of the HF patients. The most commonly used definition of ID had a sensitivity of 82% and a specificity of 72%. A definition solely based on TSAT ≤19.8% or serum iron ≤13 µmol/L had a sensitivity of 94% and specificity of 84% and 88%, respectively (P<0.05 compared with the former definition). Subsequently, we assessed the incidence of all-cause mortality in 387 consecutive outpatient HF patients (left ventricular ejection fraction ≤45%). In these patients, TSAT ≤19.8% and serum iron ≤13 µmol/L, and not ferritin, were independently associated with mortality.
Conclusions: A TSAT ≤19.8% or a serum iron ≤13 µmol/L shows the best performance in selecting patients with ID and identifies HF patients at the highest risk of death. Our findings validate the currently used TSAT cutoff of <20% for the identification of ID in HF patients, but question the diagnostic value of ferritin.
Keywords: bone marrow; ferric carboxymaltose; ferritin; heart failure; prognosis; transferrin saturation.
© 2018 American Heart Association, Inc.