Use of uterine bacteriology with cytology as predictive factors for conception.

Endometritis has been reported to be the third most common medical condition of horses (Card, 2005) and is an important cause of reduced fertility in mares (Hurtgen, 2006) in particular the most common cause of embryonic loss before day 35 in normally cycling mares (Knottenbelt, 2003). Previous reports in the literature on the diagnosis and treatment of endometritis in the mare are contradictory and opinions vary considerably (Liu and Troedsson, 2008). Uterine bacteriology with cytology has long been reported to be a useful diagnostic tool in identifying mares with endometritis (Wingfield Digby and Ricketts, 1982) and at risk of reduced 28 day pregnancy rates (Riddle et al., 2007), however, studies have failed to investigate the use of uterine bacteriology and cytology and their predictive value with regard to live foal rates.

Material and Methods:
Guarded swabs for uterine endometrial cytology and bacteriology were simultaneously collected from 1614 mares on 17 stud farms in the Hunter Valley region of NSW by 5 veterinarians at Scone Equine Hospital during the 2008 breeding season. All mares, including maiden (181), foaling (1059), barren (306) and problem (67) had a swab taken for cytological and bacteriological analysis at the beginning of the season prior to covering. The 1614 mares were further broken down into age groups as follows: Group A (young 3-8 years), group B (middle aged 9-13 years), group C (old mares 14-18 years) and group D (oldest mares >18 years). The swabs were transported to the laboratory at Scone Equine Hospital on the same day as collection. Cytology was classified as non-inflammatory 0-5% neutrophils, mild inflammation 5-15% neutrophils, moderate inflammation 16-30% neutrophils and severe inflammation >30% neutrophils (Card, 2005). Bacterial growth was classified as no growth, mixed flora (2 or more organisms) or positive (for one bacterial isolate only). Swabs that had a cytological presence of >10% neutrophils and no bacterial growth were also put in cooked meat medium and cultured anaerobically for 72 hours in order to identify possible anaerobic pathogens. Pregnancy outcome was classified as missed, slipped (in accordance with Australian stud book definitions) or live foal. Pearson’s χ² analysis (Genstat) was used for statistical analysis throughout, a p value of <0.05 was considered significant.

Results:
Age group was significantly (p < 0.001) associated with pregnancy outcome, live foal rates varying between the young (76%), middle aged (74%), old (65%) and oldest mares (31%). There was a significant (p< 0.001) association between reproductive status and pregnancy outcome, with foaling rates varying between 77% for maiden mares and 61% for problem mares. 1518 (94%) of the 1614 paired cytological and bacteriological samples were normal (< 5% neutrophils and no growth). These control mares had a live foal rate of 74%, 19% missed and 7% slipped. There was a significant association (p< 0.005) between positive bacterial growth and decreased pregnancy rates. Of these positive mares 56% had a live foal, 40% missed and 4% slipped. The major contributors to the positive category were beta haemolytic streptococci (62%) with a live foal rate of 71% and Escherichia coli (12%) with a live foal rate of 17%. There was no significant association between mixed bacterial flora and inflammation or live foal rates though they were lower (71%) compared to controls; 22% mares missed and 7% slipped. There was a significant (p< 0.05) association between inflammation and live foal rate. Mares that fell into the mild inflammatory category had a live foal rate of only 33%, while the overall percentage for mares with mild, moderate or severe inflammation having a live foal was 61%. There was a positive association (p< 0.05) between mares that had a positive swab, inflammatory cytology and pregnancy outcome, with poorer live foal rates for these mares (53%), with 41% of mares classified as missed and 6% as slipped.

Discussion:
Individually both uterine bacteriology and cytology are diagnostic in predicting pregnancy outcome, as mares with either a positive culture or inflammation on cytology had lower live foal rates than the control group. However, they are most prognostic when used together and mares with an inflammatory cytology and positive bacteriology were the least likely to produce a live foal (53%). Our data did not support the finding of Riddle et al., which stated that all mares with two or more organisms (mixed flora) isolated in the uterus had moderate or severe inflammation on cytological analysis. The type of pathogen isolated is important with a live foal rate of only 17% in mares with Escherichia coli isolated. Sampling techniques used in this study were viewed to be successful as only 2.7 % of swabs were hypocellular yielding inadequate epithelial cells to allow analysis (hypocellular results indicate poor/inadequate swabbing technique). Mares with a mild inflammatory response had a disproportionally low number of live foals, it is thought that these mares may not have been viewed as high risk and therefore not treated as aggressively as mare with moderate and severe inflammation or may have developed post breeding endometritis. Analysis of the data is ongoing and further research is being undertaken to assess such things as the predictive value of the intermediary and last mare swab versus first mare swab in repeat breeders and the possibility of creating a predictive equation for live foal rates based on mare age, reproductive status, and uterine bacteriology with cytology.

References:
CARD, C. (2005) Post-breeding inflammation and endometrial cytology in mares. Theriogenology, 64, 580-8.
HURTGEN, J. P. (2006) Pathogenesis and treatment of endometritis in the mare: a review. Theriogenology, 66, 560.
KNOTTENBELT, D. C. (2003) Equine stud farm medicine and surgery, London, W. B. Saunders.
LIU, I. K. & TROEDSSON, M. H. (2008) The diagnosis and treatment of endometritis in the mare: yesterday and today. Theriogenology, 70, 415-20.
RIDDLE, W. T., LEBLANC, M. M., PIERCE, S. W. & STROMBERG, A. J. (2005) Relationships between pregnancy rates, uterine cytology, and culture results in a thoroughbred practice in central kentucky. Proceedings of the Annual Convention of the AAEP, 51.
RIDDLE, W. T., LEBLANC, M. M. & STROMBERG, A. J. (2007) Relationships between uterine culture, cytology and pregnancy rates in a thoroughbred practice. Theriogenology, 68, 395-402.
WINGFIELD DIGBY, N. J. & RICKETTS, S. W. (1982) Results of concurrent bacteriological and cytological examinations of the endometrium of mares in routine stud farm practice 1978-1981. J Reprod Fertil Suppl, 32, 181-5.