TY - JOUR
T1 - Significance of intercellular spaces (windows) in effusion fluid cytology
T2 - A study of 46 samples
AU - Murugan, Paari
AU - Siddaraju, Neelaiah
AU - Habeebullah, Syed
AU - Basu, Debdatta
PY - 2008/9
Y1 - 2008/9
N2 - The presence and nature of intercellular windows were studied on 46 body cavity fluid samples chosen on the basis of an unequivocal diagnosis on May-Grünwald-Giemsa (MGG)/Papanicolaou-stained smears and cell blocks. Of these, 24 cases had adenocarcinoma (AC) and seven had reactive mesothelium (RM) with 15 having dis tinct populations of both. Mucicarmine and PAS stains were used wherever indicated. The specificity of windows for predicting reac tive mesothelium was evaluated. Intercellular windows were found in all cases of reactive mesothelium coinciding with the presence of the fuzzy peripheral microvillous borders. Surprisingly, as many as 17139 (44%) of the adenocarcinomas also exhibited this feature, of which 13 had a distinctly visible evenly distributed ciliated cell membrane. In addition, 30/39 (77%) cases of AC exhibited a "win dow-like" appearance caused by cytoplasmic vacuolation. None of the adenocarcinoma clusters with true window formation showed positivity for the mucin stains, whereas all the clusters with pseudo windows caused by vacuolation were stained. Thus the specificity of intercellular windows for RM was merely 56%, though the sensitivity was 100%. On the other hand, the absence of windows was 100% specific for adenocarcinoma. Intercellular windows, though a feature of reactive mesothelial cell populations, can also be found in cases of ciliated adenocarcinomas and may not have a significant predictive value.
AB - The presence and nature of intercellular windows were studied on 46 body cavity fluid samples chosen on the basis of an unequivocal diagnosis on May-Grünwald-Giemsa (MGG)/Papanicolaou-stained smears and cell blocks. Of these, 24 cases had adenocarcinoma (AC) and seven had reactive mesothelium (RM) with 15 having dis tinct populations of both. Mucicarmine and PAS stains were used wherever indicated. The specificity of windows for predicting reac tive mesothelium was evaluated. Intercellular windows were found in all cases of reactive mesothelium coinciding with the presence of the fuzzy peripheral microvillous borders. Surprisingly, as many as 17139 (44%) of the adenocarcinomas also exhibited this feature, of which 13 had a distinctly visible evenly distributed ciliated cell membrane. In addition, 30/39 (77%) cases of AC exhibited a "win dow-like" appearance caused by cytoplasmic vacuolation. None of the adenocarcinoma clusters with true window formation showed positivity for the mucin stains, whereas all the clusters with pseudo windows caused by vacuolation were stained. Thus the specificity of intercellular windows for RM was merely 56%, though the sensitivity was 100%. On the other hand, the absence of windows was 100% specific for adenocarcinoma. Intercellular windows, though a feature of reactive mesothelial cell populations, can also be found in cases of ciliated adenocarcinomas and may not have a significant predictive value.
KW - Adenocarcinoma
KW - Intercellular windows
KW - Pseudowindows
KW - Reactive mesothelium
KW - Serous effusions
UR - http://www.scopus.com/inward/record.url?scp=50849092874&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=50849092874&partnerID=8YFLogxK
U2 - 10.1002/dc.20874
DO - 10.1002/dc.20874
M3 - Article
C2 - 18677752
AN - SCOPUS:50849092874
SN - 8755-1039
VL - 36
SP - 628
EP - 632
JO - Diagnostic Cytopathology
JF - Diagnostic Cytopathology
IS - 9
ER -