Abstract
The detection and enumeration of sideroblasts depend critically on the method used for iron staining of bone marrow smears. Several methods proposed for semiquantitative evaluation of bone marrow hemosiderin (iron stores) were compared with respect to their suitability for detection of normal and abnormal sideroblasts. Instead of the customary percentage of sideroblasts, the introduction of a sideroblast score is proposed and its diagnostic relevance was prospectively studied. Low sideroblast scores are associated with iron deficiency and hypoproliferative anemia. A normal sideroblast score, despite the absence of stainable hemosiderin, excludes the diagnosis of severe iron depletion. Elevated sideroblast scores may be correlated either with iron overload and/or sideroblastic (sideroachrestic) anemias.