ID ensues when iron supply is insufficient to meet the body’s needs or to cover the iron lost physiologically or pathologically
[5]. ID may manifest itself in two distinct forms with intertwined pathophysiology, namely functional and absolute ID. Absolute ID (AID) reflects depleted iron stores, while functional ID (FID) is characterised by reduced availability of iron despite sufficient or overly abundant iron stores due to suboptimal iron trafficking induced by hepcidin. It is crucial to note that studies in HF differ in defining ID
[16][17][18][19]. The most widely used definition of ID, which is also adopted by the European Society of Cardiology (ESC), is a ferritin level <100 μg/L (reflecting AID) or ferritin (100 to 300 μg/L) with a transferrin saturation (TSAT) <20% (reflecting FID)
[20]. This definition of ID has, however, been criticised as it has never been validated against a gold standard and remains a subject of considerable debate, especially in patients with acute HF
[17][18][21][22][23][24].
The ESC definition of ID (also called the FAIR-HF definition
[25]) is limited by relying heavily on ferritin levels, thereby labelling patients with TSAT ≤20%, but a ferritin >300 μg/mL as iron sufficient, while labelling those with isolated hypoferritinaemia (ferritin <100 μg/mL with a TSAT >20%) as iron deficient. This latter category was found to be iron sufficient when compared to bone marrow staining, the golden standard for diagnosing ID
[23]. Although ferritin is one of the most widely used biomarkers to detect iron deficiency
[26], serum ferritin levels can be profoundly influenced by several factors such as inflammation, infection and malignancy, making it falsely elevated in an inflammatory state such as HF and thus does not correlate with iron availability
[27]. Herein showed that, compared to bone marrow iron staining, the ESC definition of ID has a sensitivity of 82.4% and a specificity of 72% for detecting ID in patients with HF
[23]. Serum iron (≤13 μmol/L) and TSAT (≤19.8%) were significantly better cutoffs than the FAIR-HF definition, with areas under the curves (AUC) of 0.922 and 0.932, respectively. Adding ferritin to either definition did not result in a significant increase in the AUC, suggesting that ferritin does not contribute to more accurate identification of truly iron deficient HF patients, and as such, both serum iron ≤13 μmol/L and TSAT ≤19.8% are good indicators of ID as standalone. Prognostically, these two definitions are independently associated with a higher incidence of all-cause mortality, while isolated hypoferritinemia did not
[23]. Several studies corroborated these findings
[16][17][19][23][28][29][30]. More recently, it was found that persistent ID (defined as a serum iron ≤13 µmol/L) was associated with poor prognosis, while its resolution was associated with improved outcomes. Similar trends were found when defining ID as TSAT <20%, but not when defined as per the FAIR-HF criteria
[17]. Remarkably, Cleland et al. found that higher ferritin levels (and not lower levels) were significantly associated with a higher risk of all-cause or cardiovascular mortality, further questioning the correlation between ferritin and iron availability in patients with HF. Additionally, subgroup analysis of individual patient data meta-analysis (
n = 839) pooled from four double-blind, randomised controlled trials (RCTs) showed that although intravenous ferric carboxymaltose (FCM) generally reduces recurrent cardiovascular hospitalisations and cardiovascular mortality, patients with TSAT <20.1% benefit more from FCM iron than those with TSAT >20.1% even if ferritin levels were low
[31]. Similarly, in the IRON-CRT trial, it was found that HF patients with TSAT <20% benefit more from FCM iron than if TSAT was >20% in terms of cardiac contractility and left ventricular ejection fraction (LVEF)
[32]. However, similar interaction was not found in the AFFIRM-AHF trial
[33]. The aforementioned findings confirm the accuracy of TSAT <20% and serum iron ≤13 µmol/L in identifying truly iron deficient HF patients while questioning the diagnostic and prognostic usage of ferritin in detecting ID in HF.
On the other hand, these two definitions of ID might have their own limitations, as serum iron is subjected to circadian variations
[34], and TSAT might be falsely elevated in malnutrition and advanced stages of renal insufficiency
[18][35][36][37]. Recent studies proposed serum soluble transferrin receptor (sTfR) as the most auspicious novel ID-related biomarker since circulating sTfR levels reflect the iron demand of the body in addition to the erythroid proliferation rate quantitatively
[38]. In a similar approach to the bone marrow study, Sierpinski et al. found that ID defined as serum sTfR of ≥1.25 mg/L is more accurate in identifying ID when compared to bone marrow staining in clinically stable patients with HF
[24]. Of note, adding sTfR to multivariable models for predicting 3-year all-cause mortality in patients with HF abolishes the prognostic value of serum ferritin and TSAT after adjusting for all other prognosticators. These findings suggest that elevated serum sTfR is a better surrogate for depleted intracellular iron. In line with these findings, Leszek et al. found that only serum sTfR significantly correlated to myocardial and mitochondrial iron status, but not ferritin, serum iron or TSAT
[39], indicating that sTfR reflects tissue iron demands more accurately. Nevertheless, the lack of assay standardisation restricts its implementation in clinical routines
[37].
To summarise, defining ID in HF using classical biochemical iron parameters appears to be not straightforward. Mounting evidence suggests that ferritin should not be taken into consideration when diagnosing ID in patients with HF but may be used as a safety parameter to avoid the administration of iron to patients with potential iron overload; TSAT or serum iron alone are better indicators of systemic ID, while sTfR might outperform them all. Although the current ESC definition of ID performed thus far reasonably good in general, it is broad and unspecific in identifying those who are truly iron deficient; the high prevalence of ID in HF might have precluded the importance of choosing a more accurate definition to identify those who are truly iron deficient and need IV iron. Identifying truly iron-deficient patients is crucial as inaccurate diagnoses of ID
[17] and unnecessary treatment with FCM might dilute the benefits of IV iron and lead to increased risks such as hypophosphatemia
[40]. Furthermore, in light of existing evidence indicating different mechanisms leading to myocardial and systemic ID
[41] (as discussed below) as well as poor accuracy of systemic biomarkers in detecting myocardial ID, which might be a major driver behind clinical improvements upon iron supplementation, future studies should evaluate other ID-related surrogates in order to identify HF patients that might benefit from iron supplementation on a systemic and cellular level.