Methodological factors affecting canine vaginal cytology : the influence of sampling site, swab condition, and fixation method

Abstract

Background: Vaginal cytology is widely used for cycle staging in bitches because of its practicality and low cost; however, its diagnostic reliability is strongly infuenced by methodological variability. The present study aimed to evaluate the impact of sampling site, swab type, and fxation method on the interpretation of exfoliative vaginal cytology. Materials and Methods: Exfoliative vaginal cytology samples were obtained from 332 clinically healthy bitches representing all stages of the estrous cycle. Samples were collected from diferent locations (vestibulum vs. vagina) using either dry or saline- moistened swabs and prepared as air-dried or spray-fxed smears, followed by eosin–thiazine staining. Clinical examination and serum progesterone concentrations served as the reference method for cycle staging. Results: The proportions of diferent cell types were signifcantly infuenced by the sampling location. Vestibular smears contained higher proportions of keratinized and intermediate cells (p < 0.0001), whereas basal cells (p = 0.0078) and anuclear superfcial cells (p < 0.0001) were more frequently observed in vaginal smears. Overall, sampling location signifcantly afected the distinction between proestrus and estrus. Swab condition signifcantly infuenced smear quality and cellular presentation but had minimal impact on cycle stage classifcation. Smears obtained using saline-moistened swabs showed signifcantly higher cellular yield per microscopic feld (p < 0.0001). Bacteria, cellular debris, and secretory material were less frequently observed in moistened smears (p < 0.0001), whereas free nuclei were detected more frequently (p < 0.0001). Cotton fbers were more common in dry swab preparations (p < 0.0001). Cellular arrangement also difered between sampling methods (p = 0.003). The fxation method infuenced certain morphological features but did not afect cytological stage classifcation. No signifcant diferences in total cellular yield were observed between air-dried and spray-fxed smears. Conclusion: Among the factors evaluated, only the sampling location exerted a clinically relevant efect on cycle stage determination, particularly for distinguishing proestrus from estrus.