Quantitative analysis and modelling of hepatic iron stores using stereology and spatial statistics

Hepatic iron overload is a common clinical problem resulting from hyperabsorption syndromes and from chronic transfusion therapy. Not only does iron loading vary between reticuloendothelial stores and hepatocytes, but iron is heterogeneously distributed within hepatocytes as well. Since the accessibility of iron particles to chelation may depend, in part, on their distribution, we sought to characterize the shape and scale of iron deposition in humans with transfusional iron overload. Toward this end, we performed a histological analysis of iron stores in liver biopsy specimens of 20 patients (1.3–57.8 mg iron/g dry tissue weight) with aid of electron and light microscopy. We estimated distributions related to variability in siderosomal size, proximity of iron centres and inter‐cellular iron loading. These distributions could be well modelled by Gamma distribution functions over most of the pathologic range of iron concentrations. Thus, for a given liver iron burden, a virtual iron‐overloaded liver could be created that served as a model for the true histologic appearance. Such a model may be helpful for understanding the mechanics of iron loading or in predicting response to iron removal therapy.