Best Post of Sept '08: Extramedullary hematopoesis is not an uncommon finding in subdural hematomas

The next in my "Best of the Month" series, in which I feature my favorite post from each month, is from September 26, 2008:

I received a case in consultation this week from a general pathologist who was concerned about clusters of unusual-looking cells within a chronic subdural hematoma (SDH), a specimen the surgeons call a "subdural membrane" (look at the picture above of a resected chronic subdural hematoma from Greenfield's textbook to see why surgeons use that term). I felt that the cluster of cells represented foci of extramedullary hematopoesis (EMH). Another general pathology colleague was doubtful that such a phenomenon was likely in an patient with no other hematological disorders. I countered that perhaps 25% of SDH cases in hematologically intact patients show EMH. This prompted an email to the illustrious Hopkins neuropathologist Peter Burger (pictured), who provided me with a reference to an 2007 article by Juan Rosai and colleagues. In their article, Rosai and friends cited a larger study by Muller et al. (reference 1) in which "a microscopic study of 130 chronic subdural hematomas... found nucleated red blood cells in 41 cases (32%). In a subsequent study, the same authors found erythroblasts in 33% of 38 cases, and this percentage increased to 57% after serial sections were taken." The Rosai article itself (reference 2) describes in detail two cases with the aim of documenting the fact that "such lesions can fool the 'general' surgical pathologists, including some who can hardly be regarded as 'junior'."

References:

1. Muller W, Zimmermann E, Firsching R. Erythropoiesis in chronic subdural hacmatomas. Acta Neurochir (Wien). 1988;93:137-139.

2. Kuhn E, Dorji T, Rodriguez J, Rosai J. Extramedullary Erythropoiesis in Chronic Subdural Hematoma Simulating Metastatic Small Round Cell Tumor. International Journal of Surgical Pathology 15:3 [July 2007] p.288-291.