Counting Small Hypointense Spots Confounds the Quantification of Functional Islet Mass Based on Islet MRI

Iron-containing fragmented islets or free iron released from dying cells could confound the interpretation of MRI of iron nanoparticle-labeled islets. Exclusion of small hypointense spots could be a useful strategy to avoid such artifact. We investigated whether this strategy could improve the estimation of functioning islet mass after islet transplantation. Using a rat syngeneic intraportal islet transplantation model, we quantitatively assessed the relationships between total area, number of hypointense spots on MRI that belong to each size quartile and glycemic control of the recipients. The total area of hypointense spots on MRI was greater in the recipients that achieved diabetes reversal (p = 0.002), whereas the total number of hypointense spots was not different (p = 0.757). Exclusion of small hypointense spots improved the association between the number of hypointense spots and the blood glucose level of the recipients (p < 0.001). Ex-vivo imaging and histologic study confirmed that some small hypointense spots represent the phagocytosed free iron. Exclusion of small hypointense spots improved the quantification of the functional islet mass based on islet MRI. This would be a useful principle in the development of an algorithm to estimate functioning islet mass based on islet MRI.