Ovarian remnant syndrome in a 2-year-old cat with inconclusive diagnostics

Abstract

Case summary

A 2.5-year-old, spayed female domestic shorthair cat was evaluated for estrus-like behavior (lordosis, vocalization and rolling) occurring on a cyclical basis every 3 weeks. Clinical signs began 6 months after ovariohysterectomy and exploratory abdominal surgery was performed; no ovarian tissue was found. On evaluation at a referral hospital, serum anti-Müllerian hormone (AMH) concentration was 0.14 ng/ml, consistent with a spayed female, and serum progesterone concentration was 1.3 ng/ml, which is not consistent with the presence of active luteal tissue. The cat returned while exhibiting estrus behavior and 90% of the vaginal epithelial cells seen on vaginal cytology were superficial. Serum AMH concentration at that time was again within the interval consistent with a spayed female (0.10 ng/ml). Gonadorelin was administered to induce ovulation; 3 weeks later, serum progesterone concentration was measured at 1.6 ng/ml, not consistent with the presence of active luteal tissue. Exploratory abdominal surgery was performed and a small 2 mm piece of ovarian tissue was removed from near the location of the right ovarian pedicle. No estrus-like behavior was observed in the weeks after the surgery.

Relevance and novel information

This case study demonstrates that a very small amount of ovarian tissue can produce enough estradiol to stimulate estrus behavior in the cat and induce cornification of the vaginal epithelium, yet produce too little AMH to confirm its presence. Similarly, despite the formation and presence of luteal tissue, there was apparently too little progesterone secretion to confirm what is generally seen in an intact queen.

Case description

A 2.5-year-old, spayed female domestic shorthair cat was referred to the Department of Clinical Sciences at the College of Veterinary Medicine at Auburn University for evaluation of estrus-like behavior, including lordosis, vocalization and rolling, which occurred on a cyclical basis every 3 weeks. The clinical signs began approximately 6 months after ovariohysterectomy (OHE), which was performed when the cat was 19 months of age. An exploratory abdominal surgery was performed by the referring veterinarian 11 months after the original OHE and no ovarian tissue was found; the cyclic, estrus-like behavior continued. The patient had no identifiable exposure to exogenous hormones.

On the day of evaluation at the referral hospital, the patient was not exhibiting estrus-like behavior. Blood was drawn and submitted to the Clinical Endocrinology Laboratory, School of Veterinary Medicine, University of California, Davis for determination of serum anti-Müllerian hormone (AMH) concentration to detect remnant ovarian tissue (ovarian remnant syndrome [ORS]). Serum AMH was analyzed using canine-specific assay AMH ELISA (AL-116; Ansh Labs) as previously validated for cats. ¹ The AMH serum concentration was 0.14 ng/ml and within the reference interval (RI) for a spayed female (0.01–0.16 ng/ml¹). Serum progesterone concentration, analyzed using a previously validated ELISA, ² was 1.3 ng/ml, not indicative of active luteal tissue.

The cat returned 3 weeks later for a follow-up examination while she was exhibiting estrus-like behavior, on approximately day 3 or 4 of estrus. At this time, vaginal cytology was performed using a saline-moistened, cotton-tipped swab that was placed in the vestibule and advanced dorso-cranially to swab the cranial vagina as described. ³ A modified Romanowsky stain was used to identify approximately 50% nucleated superficial vaginal epithelial cells and 40% anuclear squamous cells, indicative of the systemic presence of estrogen (Figure 1). Serum was submitted for AMH, and the results were once again within the RI for spayed females (0.10 ng/ml). The patient was administered 1 ml of gonadorelin (43 µg/ml IM, Fertagyl; Merck Animal Health USA) to induce ovulation. The cat then returned for serum progesterone 3 weeks after gonadotropin-releasing hormone (GnRH) administration to determine if ovulation had occurred. The serum progesterone concentration was 1.6 ng/ml; again, not consistent with the presence of luteal tissue.

Figure 1

Superficial epithelial cells of a cranial vaginal cytology while the feline patient was exhibiting estrus-like behavior. Approximately 90% of the epithelial cell population were superficial cells.  Modified Romanowsky stain

Despite the inconclusive diagnostics, but in light of the recurrence of apparent estrous clinical signs (behavior and vaginal cornification), abdominal exploratory surgery was performed 1 month later based on surgeon availability. A small, 2 mm piece of ovarian tissue was confirmed on histopathology at the location of the right ovarian pedicle (Figure 2). Histologically, the examined section of excised tissue consisted of densely vascularized fibrous connective tissue surrounded by mature adipose tissue, multifocally mineralized material within haphazardly arranged smooth muscle fibers and an ovarian remnant measuring approximately 2 × 2 mm (Figures 3 and 4). The ovarian tissue recapitulated all the elements of a normal ovary, including a focal cystic tertiary follicle, secondary follicle, corpus luteum and stromal tissue. The focus of mineralization was interpreted as dystrophic mineralization secondary to previous surgery in the area. After removal of the remnant tissue, the cat no longer exhibited estrus-like behavior.

Figure 2

Tissue surgically removed around the right ovarian pedicle during exploratory abdominal surgery. The tissue contains a very small 2 mm piece of ovary

Figure 3

Sub-gross view of the surgically excised ovarian remnant tissue. The ovarian tissue measures approximately 2 mm in diameter. Hematoxylin and eosin stain. Marker = 1 mm

Figure 4

Higher magnification view of the ovarian remnant tissue, showing the secondary follicle (arrow) and the corpus luteum (triangle) within a normal stromal background. Hematoxylin and eosin stain. Marker = 100 μm

Discussion

ORS refers to the presence of ovarian tissue in the cat, which causes behavioral signs of estrus after an ovariectomy or OHE. ³ The presence of the ovarian tissue is caused by incomplete surgical removal of ovarian tissue. Classic signs of behavioral estrus in the ORS cat are vocalization, lordosis, restlessness, head rubbing, rolling and allowing copulation to occur. These signs typically occur recurrently in a cyclical pattern. It is important to distinguish ORS from exogenous hormonal exposure and adrenal neoplasia, or to determine whether the behavior is not hormonally driven. If ovarian tissue remains in the cat, there is risk for cystic endometrial hyperplasia-pyometra complex of any remaining uterine or cervical tissue, mammary neoplasia and continued undesirable estrus behavior. Estrus behavior may begin shortly after the initial ovariectomy or OHE procedure or the estrus behavior may be delayed for months or years postoperatively. ³

Estrus behavior is caused by the presence of estrogen. The main source of estrogen in the queen is granulosa cells in growing antral follicles of the ovary. Estradiol assays from blood samples can be unreliable in the cat owing to significant time and day variations of systemic estrogen, but a serum estradiol concentration greater than 20 pg/ml while exhibiting estrus-like behavior can be suggestive of ORS as this is the reported RI for intact queens in the follicular phase. ³ Vaginal cytology is a more reliable indicator of the presence of systemic estrogen perhaps because it represents a cumulative exposure. Queens typically have peak cornification of vaginal epithelial cells, or presence of superficial vaginal epithelial cells, when systemic estradiol is highest. As estradiol rises, the percentage of anuclear squamous cells will increase from 5% to 40% of the cell population and intermediate cells will decrease to 10% or less of the population. ⁴ Vaginal cytology with predominantly superficial cells and anuclear squamous cells is indicative of systemic estrogen exposure. The vaginal cytology sample should be taken when the cat is exhibiting estrus-like behavior. Confirming the presence of estrogen exposure is suggestive of ORS but not diagnostic. Exogenous estrogen exposure either from hormonal creams used by owners who are interacting with the cat or from phytoestrogens in the diet can produce similar results in the cat. ⁵

The main source of progesterone in the queen is luteal cells in the corpus luteum of the ovary. Cats are induced ovulators and must ovulate to be in diestrus to diagnose the presence of luteal tissue based upon systemic progesterone concentrations. Spontaneous ovulation occurs in approximately 30% of cats,^6,7 but diestrus occurs more consistently if ovulation is induced either through copulation or through the administration of a GnRH agonist or human chorionic gonadotropin (hCG). To most reliably induce ovulation (with either 25–50 µg IM of a GnRH agonist or 250–500 IU IM of hCG), cats need to be in estrus (or exhibiting estrus-like behavior) and ideally within day 2–4 of estrus to be most responsive to induction protocols. There are several ovulation induction protocols for queens involving one or two injections of either GnRH or hCG.^8,9 Serum progesterone can then be sampled 1–3 weeks after induction. Peak progesterone concentrations are seen 14–22 days after ovulation and will remain elevated until approximately 40 days after ovulation, when it will then begin to drop in the non-pregnant cat. ⁴ A serum progesterone concentration greater than 2 ng/ml indicates the presence of luteal tissue. It should be noted that proper sample handling is important for accurate hormone analysis; blood samples must be allowed to adequately clot, serum separator tubes should be avoided, serum must be pulled off samples quickly, and serum must be kept refrigerated or frozen in a plastic tube until it is to be analyzed. ¹⁰

AMH is only secreted from granulosa cells in the follicles of the ovary and thus can be useful in the diagnosis of ORS. ¹¹ Serum AMH is an especially sensitive diagnostic tool in cats compared with other species because there is little overlap between serum concentrations seen in intact vs spayed populations.^1,12 In addition, serum AMH can be sampled at any point during the estrous cycle, although concentrations tend to be higher in the follicular phase of the cycle. ¹ However, false negatives are always possible, but especially when only very small pieces of ovarian tissue remain. For instance, hemi-ovariectomy in mares halves systemic AMH concentrations. ¹³ Based on this observation, the RI for females, with one ovary would be half that of intact females and presumably even smaller fractions of ovaries generate even lower AMH concentrations in horses and other species. This is presumably because the smaller the ovarian remnant, the fewer remaining follicles there are to secrete AMH. Simply put, the smaller the remnant, the more difficult it is to detect endocrinologically and otherwise. Therefore, it is recommended that AMH concentrations are determined while cats are exhibiting signs of estrus to increase the likelihood of a reliable diagnostic result. ¹

The case report presented here offers an unusual diagnostic dilemma because the patient had low serum AMH concentrations, in a range determined consistent with ovariectomized cats, even while displaying estrus behavior and vaginal cornification, or the presence of superficial epithelial cells, indicative of estrogen exposure. In addition, after ovulation induction, serum progesterone samples did not reach a concentration typically associated with luteal tissue. It is unknown how many corpora lutea are required in the queen to cause serum progesterone concentrations to reach above 2 ng/ml. This case suggests a single corpus luteum may not be enough luteal tissue to elevate serum progesterone concentrations above baseline in the queen.

This case report highlights the limitations of relying on hormone assays alone to diagnose ORS. Although diagnostically useful in the majority of cases, very small pieces of ovarian tissue may not be detectable by serum hormonal concentrations of either AMH or even progesterone when luteal tissue is present. In some cases, advanced diagnostic imaging, such as abdominal ultrasonography or CT, is a valuable tool when other diagnostics are inconclusive. ¹⁴

Conclusions

This case report describes a case of ORS with inconclusive diagnostic test results. It highlights that a very small piece of ovarian tissue can produce enough estrogen to cause estrus behavior in the cat and cornification of the vaginal epithelial cells without significantly elevating serum AMH concentrations. In addition, a single corpus luteum can produce insufficient serum progesterone concentrations to confirm the presence of luteal tissue such as those generally seen in intact queens.