There’s overall agreement that iron deficiency is prevalent and portends a worse prognosis in patients with heart failure (HF), regardless of ejection fraction or anemia. What remains unclear, however, is which of the many definitions of iron deficiency most closely aligns with adverse outcomes.
Iron deficiency (ID) differs in chronic inflammatory conditions, such as chronic HF, and is defined in international guidelines as a ferritin less than 100 ng/mL or ferritin 100-299 ng/mL with a transferrin saturation (TSAT) less than 20%.
A new study examining four definitions of ID in more than 4,000 patients with HF revealed that TSAT and serum iron – but not guideline criteria – were independently associated with higher 5-year all-cause mortality, regardless of HF phenotype.
“The standard definition, the society guideline definition of iron deficiency, simply isn’t related to outcome at all. The lines for mortality are, more or less, superimposed,” senior author Andrew L. Clark, MD, Hull (England) University Teaching Hospital NHS Trust, told this news organization.
“So we do think, therefore, there’s a need for a rethink as to what constitutes a definition of iron definition in people with heart failure.”
The results were published in the Journal of the American College of Cardiology.
Previous studies have shown that guideline-defined ID is an independent predictor of mortality in chronic HF, but others have questioned its diagnostic and prognostic utility. A 2018 study using bone marrow iron staining as the gold standard showed that a TSAT of 19.8% or less or serum iron of 13 mcmol/L or less, but not ferritin, identified HF patients at the highest risk for death.
A 2016 report from the Hull LifeLab cohort also showed that the highest quintiles of ferritin had the worst all-cause and cardiovascular (CV) mortality.
Commenting on the new results, Maria Rosa Costanzo, MD, Midwest Cardiovascular Institute, Naperville, Ill., said “the first clinical implication is that we should not use these guidelines to define iron deficiency.
“The fundamental problem with the definition is that ferritin is not a good marker of iron deficiency because ferritin is an inflammatory marker,” she said. “So you could have high ferritin and still have iron deficiency because heart failure, like many other diseases, is an inflammatory state.”
In the present analysis of 4,422 patients referred to the Hull LifeLab clinic between 2001 and 2019, iron deficiency was defined using international guideline criteria, ferritin less than 100 ng/mL, TSAT less than 20%, and serum iron 13 mcmol/L or less.
In line with previous studies, the prevalence of ID was high, ranging from 44% to 68%, depending on the definition. ID was more common in women and in those with more severe symptoms, anemia, or preserved ejection fraction.
Overall, 5-year mortality was 34.5% (median follow-up, 49 months). Unadjusted mortality was lowest for patients with a serum ferritin less than 100 ng/mL and a TSAT greater than 20% and was highest for those with serum ferritin above 100 ng/mL with a TSAT less than 20%.
Serum iron levels and TSAT were highly correlated with each other (r = 0.92; P < .001). “Serum iron is almost entirely transferrin bound, and therefore a close association between serious iron and TSAT is not surprising,” noted the authors, led by Gabriele Masini, MD, University of Brescia (Italy).
After multivariate adjustment, TSAT less than 20% (hazard ratio, 1.27; P < .001) and serum iron of 13 mcmol/L or less (HR, 1.37; P < .001) were associated with greater all-cause mortality but not with CV mortality.
Serum ferritin less than 100 ng/mL tended to be associated with lower adjusted all-cause mortality (HR, 0.91; P = .09), whereas ferritin greater than 300 ng/mL was associated with lower all-cause (HR, 0.69, P < .001) and CV mortality (HR, 0.78; P = .048).
No association was found for guideline ID criteria and all-cause or CV mortality. Among patients fulfilling guideline ID criteria with a TSAT less than 20% and a ferritin 100 to 299 ng/mL, the adjusted hazard ratio for 5-year mortality was 1.82.