Abstract

Background: Serum ferritin is the best single laboratory test to diagnose iron deficiency anemia (IDA). Ferritin levels <20 μg/L are highly specific for IDA, and ferritin levels >100 μg/L usually exclude IDA. However, ferritin concentrations between 20 and 100 μg/L are often inconclusive. The objective of this study was to improve the diagnosis of IDA when ferritin levels are inconclusive.

Methods: We evaluated the predictive performance of classic (ferritin, mean corpuscular volume, transferrin and serum iron) and modern [reticulocyte hemoglobin content, serum transferrin receptor and soluble transferrin receptor (sTfR)/log(ferr)] iron status parameters to diagnose IDA in 2084 anemic, non-hospitalized patients. The results were validated in an independent cohort of 274 anemic patients.

Results: In our study population, 29% (595 patients) of the patients had a ferritin level between 20 and 100 μg/L, hampering diagnosis of IDA. None of the classic or modern parameters was capable of completely separating the IDA population from the non-IDA population. However, using a new parameter, the transferrin/log(ferritin) ratio, the IDA and non-IDA populations can be completely separated. At a cut-off value of 1.70, the transferrin/log(ferritin) ratio indicates IDA in 29% of the patients with inconclusive ferritin levels.

Conclusions: The transferrin/log(ferritin) ratio is a practical new tool that improves diagnosis of iron deficiency when ferritin levels are inconclusive.