Endometrial cytology and culture specimens (n=2123) were collected concurrently with a guarded uterine culture instrument from 970 mares (738 barren, 1230 foaling and 155 maiden mares) during three breeding seasons (2001-2004). Results were compared to the 28-d pregnancy rate for the cycle from which the samples were taken. Cytological smears were evaluated for inflammation at x100 and graded as: not inflammatory (0-2 neutrophils/field), moderate inflammation (2-5 neutrophils/field), severe inflammation (>5 neutrophils/field), or hypocellular (scant epithelial cells and no neutrophils). Uterine culture swabs were plated within 6h, incubated for 72 h and results determined at 24, 48, and 72 h. Approximately, 20% (n=423) cytology samples were positive for inflammation (>2 neutrophils), whereas approximately 11% (n=231) of cultures had microorganisms recovered. A majority (64%) of the positive cultures (147/231) had inflammation on cytology smears. Streptococcus equi subsp. zooepidemicus was associated with more positive cytology results than coliforms (P<0.01). Mares with positive cytology or culture had lower pregnancy rates than mares with normal findings (P<0.01). Lowest pregnancy rates were recorded for mares with severe endometrial inflammation (21%, versus moderate inflammation 48%). Isolation of a microorganism from mares with endometrial inflammation was not associated with a further reduction in pregnancy rates. In barren, foaling and maiden mares, cytology was positive in 28, 17, and 5%, respectively, and culture was positive in 12.2, 11.1, and 3.2%. Foaling and maiden mares had higher pregnancy rates than barren mares (62, 69, and 44%, respectively, P<0.001). In conclusion, a positive cytology was twice as common as a positive culture, and isolation of microorganisms was associated with reduced pregnancy rates, even in the apparent absence of inflammation.