Four intrauterine treatment strategies were evaluated for effectiveness in mares that were confirmed to be susceptible to chronic uterine infection. Pretreatment samples were obtained at detection of estrus, and a genital strain of Streptococcus zooepidemicus was infused into the uterus when a preovulatory (> 35 mm) follicle was detected. At 12 hours after inoculation, mares were assigned to 1 of 4 selected treatment groups: autologous plasma, 100 ml (n = 5); potassium penicillin, 5 million U in 100 ml of phosphate-buffered saline solution (PBSS; n = 5); 10 mg of prostaglandin F2alpha in 100 ml of PBSS (n = 5); and large-volume lavage with normal saline solution (1,000 ml increments). A fifth group, treated with vehicle alone (100 ml of PBSS), served as a negative control (n = 7). All treatments were administered into the uterus. To assess the effectiveness of the treatment, samples for culture and cytologic examination were collected at 96 hours after bacterial inoculation. An effect of treatment was observed on the number of uterine neutrophils (P = 0.02) and growth of S zooepidemicus (P < 0.01). Intrauterine treatment with potassium penicillin, prostaglandin F2alpha, and large-volume uterine lavage significantly reduced the growth of S zooepidemicus (P < 0.01) as well as the number of neutrophils (P < 0.02). Autologous plasma reduced the number of neutrophils (P < 0.05), but not growth of S zooepidemicus. There was significant correlation between the number of uterine neutrophils and growth of S zooepidemicus for each treatment group (r = 0.57; P < 0.05).

Effect of estrogen (E2) and progesterone (P4) on uterine antibacterial activity and immunoglobulin concentrations in mares was studied. In 2 in vitro experiments, 6 mixed-breed mares were ovariectomized, and uterine fluid and blood serum were analyzed. Antibacterial assay methods were used to determine inhibitory effects on Streptococcus zooepidemicus of uterine fluid samples collected on days 3, 5, and 8, and serum obtained on day 8 of treatment. Single radial immunodiffusion methods were used to quantify amounts of IgA and IgG in uterine fluid and serum on days 3, 5, 8, and 14 of treatment. Neither E2 nor P4 increased activity of serum and uterine fluid against S zooepidemicus. Numbers of colony-forming units per milliliter of bacteria were significantly (P < 0.01) lower in control Hanks' balanced salt solution with 1.0% gelatin (HBSSG) than in uterine fluids. Bacterial numbers were significantly (50%) greater in uterine fluids and serum than in HBSSG controls for both treatments. Both fluids, especially serum, supported significantly (P < 0.01) more growth of S zooepidemicus than did HBSSG when incubated for 0, 2, and 4 hours. These findings are in contrast to previous reports of antibacterial activity in the uterus of sexually intact mares undergoing an estrous cycle: great reduction of bacterial count in uterine fluid from mares in diestrus, and significant increases in bacterial numbers in uterine fluid or serum from mares in estrus. Treatment comparisons between serum and uterine fluid IgA and IgG concentrations were not significantly different, although overall IgA concentration in the uterus was higher than concentration in serum. The IgG concentration in uterine fluid was higher in P4- than E2-treated mares. However, IgG concentration was significantly (P < 0.01) higher in uterine fluid on day 8 in P4-treated mares than on day 3 or 5. Results of this study indicate that neither immunoglobulin concentration nor hormone treatment has a direct effec.

The relationship between histologic lesions in endometrial biopsy specimens and susceptibility to chronic uterine infection (CUI) in mares was investigated. Mares were allotted to 4 groups on the basis of degree of endometrial lesions. Mares in group 1 (n = 6) had no pathologic changes, mares in group 2 (n = 5) had only mild pathologic changes, group-3 mares (n = 7) had moderate changes, and group-4 mares (n = 7) had severe inflammatory and fibrotic endometrial changes. Susceptibility to CUI was determined by the inflammatory response to intrauterine inoculation of 5 X 10(6) Streptococcus zooepidemicus. The inoculum was given on the third day of behavioral estrus and in the presence of a follicle > 30 mm. Mares with > 1 neutrophil/5 high-magnification (400 X) microscopic fields and > 20 colonies of S zooepidemicus at 96 hours after inoculation were considered to be susceptible to CUI. There was a significant association between biopsy grade and susceptibility to CUI among the groups. Histologically normal endometrium was associated with resistance to CUI, and severe histopathologic changes in the endometrium were associated with susceptibility to CUI. Mild to moderate endometrial lesions did not correlate consistently with susceptibility or resistance to CUI.