Diagnostic value of endoscopically guided transcervical catheterization for cytological and bacteriological sampling in canine endometritis and normal uterus

The aims of this study were: 1) To evaluate the diagnostic value of endometrial cytological samples obtained by endoscopically guided transcervical catheterization (TCC) in canine endometritis and normal uterus; 2) To determine the sensitivity and specificity of TCC for the identification of uterine bacterial growth; 3) To compare vaginal and uterine bacteriological findings. Before elected ovariohysterectomy, vaginal and TCC bacteriological and cytological samples were obtained in ten post-pubertal bitches. After surgery, direct uterine bacteriological sampling and full-thickness biopsy were also carried out. The agreement between endometrial cytology and full-thickness biopsy and between bacteriological results were assessed using Cohen’s kappa (κ) coefficient. Sensitivity and specificity were calculated for cytology and bacteriology cultures. The agreement between the TCC cytology and histopathology was k = 0.62, P < 0.05. Sensitivity of TCC cytology was 66 % and specificity was 100 %, P < 0.01. The agreement between TCC and direct uterine bacteriological sampling was k = 0.60, P = 0.05. Sensitivity of TCC bacteriological samples was 100 % and specificity was 75 %, P < 0.05. Conversely, the agreement between vaginal and TCC or direct uterine bacterial analyses were k = 0.05, P > 0.1 and k = 0.1, P > 0.1, respectively. This study highlights the diagnostic value of endoscopically guided TCC for cytological and bacteriological sampling in normal bitches and those suffering from endometritis. The poor agreement between vaginal and uterine bacteriological results confirms the presence of different microbiota between these organs.