Serum ferritin and bone marrow iron stores. I. Correlation with absence of iron in biopsy specimens.

Serum iron, total iron-binding capacity, and percentage saturation of transferrin have classically been used to demonstrate a hypoferremic state; however, these tests may not discriminate between depleted iron stores and conditions associated with defective reticuloendothelial release of iron. Estimation of stainable iron in the bone marrow biopsy specimen is then the most practical way to assess body iron stores. With the availability of a radioimmunoassay procedure for serum ferritin, we undertook a prospective study to determine whether serum ferritin concentrations might replace assessment of the marrow biopsy iron stores as an indicator of hypoferremia. Iron stores were absent from bone marrow biopsy specimens from 104 patients. A good correlation between low serum ferritin levels and absence of iron stores in biopsy specimens was found for 91 patients (87.5%). Thirteen (12.5%) had normal serum ferritin concentrations with absence of biopsy iron. These individuals had hematopoietic malignancies or active hepatic disease, or were receiving iron therapy. In this group, a bone marrow biopsy would still be necessary for evaluation of a hypoferremic state, even though the serum ferritin concentration might be normal.