Evaluation of cytospin precision in low cellularity canine cerebrospinal fluid

Abstract

The cell count and differential of cerebrospinal fluid (CSF) cytologic examination classify CSF as inflammatory or not. The cytospin cell yield is related to cell count, but to our knowledge a relationship has not been characterized and cytospin precision is undocumented in any species. The objective of our study was to calculate intra-assay precision of cellular yield and differential on cytocentrifuged canine CSF, determine the factors that may affect precision, and predict the number of cytospins necessary to confirm mild neutrophilic pleocytosis. Ten concurrent replicate cytospins were created from nonhemorrhagic CSF, obtained from 60 dogs in other terminal studies, with either a manual or calibrated pipetting technique. Up to 500 cells per cytospin were counted and classified on each slide. Coefficient of variation (CV), multiple regression, and probabilities were calculated for relationships between cell yield and independent factors including technique, total nucleated cell count, cell differential, and total protein. Manual and calibrated pipetting had similar CVs (average 31%) for total cell yield, but the calibrated technique had fewer foamy macrophages. CV for neutrophil percentage among low cellularity samples with any neutrophils was 146%. Probability based on linear regression showed that 1 cytospin is sufficient to identify samples with >3% neutrophils. Occasional neutrophils, eosinophils, mitotic figures, phagocytic cells, and ependymal cells were seen in many low cellularity canine CSF samples. Canine CSF cytospin cell yield and differential evaluations are imprecise. Calibrated rather than manual pipetting is recommended.