Cytology of 100 samples of abdominal fluid from 100 horses with abdominal disease

A total of 100 samples of abdominal fluid (AF) from 100 horses with abdominal disease were evaluated by cytology. Cytology results were subsequently correlated with the final outcome of the disease. The horses were classified into 4 groups: Group I, horses that were treated with conventional (nonsurgical) therapy and recovered; Group II, horses that had surgery and survived; Group III, horses that had surgery but died; and Group IV, horses that were subjected to euthanasia prior to surgery. Statistical analysis showed that both nucleated cell count and total neutrophils were significantly higher in Group III than in Group I; and that the total mesothelial cells were significantly higher in Group III than in Groups I and II. No significant differences were found for erythrocyte counts and fluid total protein levels among the 4 groups. The findings suggest that classifying AF as transudate, modified transudate and exudate, as well as grading of inflammation as mild, moderate and severe on the basis of nucleated cell count (NCC) and fluid total protein (AFTP) can be greatly misleading. Differential identification of the nucleated cells was found to be far more reliable than the NCC alone, with or without the AFTP, and rendered valuable information, which overruled many times a diagnosis of transudate or modified transudate. Bands, metamyelocytes, toxic changes, plasma cells, and neutrophils penetrating rafts or fronds of mesothelial cells, supported a diagnosis of inflammation, even when the NCC, and the AFTP (interpreted according to currently accepted values) suggested otherwise. Several morphological features were found, including some cell types for which little or no mention was found at all in 22 major studies of this fluid in horses. Among these were 'reddish neutrophils', large granular lymphocytes (LGL), plasma cells, Mott cells, blasts, and a unique hitherto undescribed granular mesothelial cell.