Fwd: RE: [Histonet] FNA'S

Stephen Peters M.D. petepath <@t> yahoo.com
Thu Mar 2 11:40:37 CST 2006

Hi Rogerson
  I believe there is a great deal of information that can be learned only from the 
  smearing of cells that cannot be seen in paraffin histology and more importantly in the newer liquid medium methods which do not smear cells. As you mentioned, the crushing nuclei of small cell carcinoma and many other small blue cell tumors, also referred to a Azzopardi Phenomenon"  is an important diagnostic clue. How a cell responds to the pressure of smearing tells us a lot about the structural integrity of the cell. Imagine smearing a jellyfish and a turtle under a giant slide. I think cells have similar structural differences which would yield equally different results. Epithelial cells, which function as structural lining and evolved 
  to withstand such stressed tend to be more tightly cohesive and less prone to give up naked nuclei. Cells such as endocrine cells, whose role is less structural and involved in secretory function tend to have more delicate cytoplasm easily disrupt and give off many naked nuclei. Neuroendocrine tumors also fall into this category. Just the breast changes that go on with lactation result in much more delicate cells and numerous naked nuclei. Dense fibrous lesions are often very reluctant to smearing. A fragment of large cell lymphoma can ce confused with poorly differentiated carcinomas on a thin prep but when smeared, the lack of cohesion is 
  a strong clue. When I get a spinal lesion with a differential of menengioma 
  versus schwannoma, I can usually tell simply by how the tissue reacts when 
  making the smears. A schwannoma will role under the slide like a piece of 
  rubber band, and the menengioma often smears with relative ease and 
  surprising discohesion. I feel quite strongly about this matter and believe we 
  are making a mistake by taking Non medical cytology and squirting them in a jar of fixative, to have the cells sucked onto a slide without smearing. I won't mention any names!

More information about the Histonet mailing list