Pathohistological findings in bilateral removed ovaries of mares with behavioral problems

Unwanted behavior is a common presenting problem in mares which often challenges veterinarians. Behavioral problems associated with cyclicity in absence of neoplastic ovarian pathology are often treated with oestrus suppression (Aurich and Kaps, Reprod.Dom.Anim. 2022;00:1-9). However, most options are cost-intensive, might have severe side effects or reversibility is not clearly determined yet. Therefore, bilateral ovariectomy (bOE) is often chosen due to the benefited surgical procedure. Studies evaluating effects of bOE on unwanted behavior in mares from the owner´s perspective revealed an improvement (Melgaard et al., Animals 2020;10:1210). However, there is a lack of histopathological explanation for this phenomenon. Therefore, the aim of the present study was to evaluate bilateral removed ovaries histopathologically in mares with behavioral problems. We hypothesized, that there are non-neoplastic changes in ovaries histologically detectable, which could be responsible for unwanted behavior in mares. In a retrospective case series, ovaries of 20 mares after bOE (OE) via standing laparoscopy using routine procedures (Röcken et al., Vet. Surge. 2011;40:1009-1014) were compared with granulosa cell tumors (GCT, n=10). Samples from removed ovaries were fixed in formaldehyde for further histological evaluation. One additional representative sample of each ovary was sent to a commercial laboratory for histopathological diagnosis of GCT by a Board-certified pathologist.Immunohistochemical evaluation was carried out using Kiel 67 (ki67), Aromatase (AR) and Anti-Muellerian Hormone (AMH) as diagnostic marker. Intensity of staining for ki67, AR and AMH was classified as -, +, ++, and +++ for no, low (<40%), moderate (40-70%) and high (70-100%) expression, respectively. Immunolabeling for ki67, AR and AMH was most intense in granulosa cells of antral and early atretic follicles in OE. In GCT, immunolabeling for AR and AMH was additionally highly intense in Leydig-like cells. In OE, ovaries of 3 mares initially categorized as GCT could not be confirmed and were therefore regarded as early stages due to upregulated ki67 in presence of a corpus luteum. Measurements of AMH in serum performed before surgery were above the reference value in all mares of GCT and in 6/20 (30%) in OE. In GCT, 8/10 (80%) and in OE, 16/20 (80%) mares showed behavioral abnormalities before surgery, which were not correlated with increased AMH in serum. Serum AMH values were not correlated with high expression of AMH in 9 mares, where granulosa cells of antral follicles could be evaluated. All, but one mare, benefit from bOE regarding behavior as reported from the owner. In the present study, immunostaining with ki67, AR or AMH showed no clear difference between OE and GCT. Further investigations regarding different immunohistochemical staining are warranted to find marker for non- or early neoplastic changes related to improvement of behavior in mares after bOE.