Uterine biopsy is a technique that is performed in many species to evaluate the health of the reproductive tract and assess the tissue for causes of subfertility. Although some causes of reproductive failure are genetic, many cases may be attributable to acquired and potentially treatable conditions. Clients may be frustrated when dogs with features that may potentially contribute to the health and betterment of a breed do not successfully reproduce.
Endometrial biopsy is a well-established procedure in human1,2 and equine medicine.3–8 Transcervical methods developed for collection of equine endometrial tissues in the 1960s were the basis for development of standardized histopathologic biopsy sample grading systems that are widely used in equine reproductive medicine. The most commonly used grading or scoring system is the modified Kenney-Doig system published in 1986.5 The nature and severity of lesions are reflected in the category score, and this score has been shown to be associated with reproductive outcomes, thus providing a useful tool in the diagnosis and treatment of subfertility in mares.3,4,6,7
Diagnostic techniques for endometrial lesions in dogs have lagged behind those used for horses. To our knowledge, the various lesions present in the canine endometrium in cases of subfertility have not been classified, which limits the ability to estimate a patient's prognosis on the basis of diagnosis. Relatively few investigations of the use of uterine biopsy in dogs have been performed.9–12 One of the earliest studies10 described a transcervical technique used in 23 Beagles; however, 3 of those dogs had excessive bleeding that necessitated ovariohysterectomy. Given the potential for severe consequences of this complication, collection of incisional biopsy samples of the uterus by means of laparotomy with a caudal abdominal approach became favored. This procedure has been performed to investigate fertility issues in dogs many times; however, to our knowledge, the prevalence of uterine lesions in a large group of dogs has not been reported, and evaluation procedures for uterine tissue in dogs have not been standardized.
The normal dynamic changes of the canine uterus, including normal reproductive cycle changes13 and the postpartum uterine involutionary process,14 have been previously described. Pyometra is a clinically important disease, and its pathogenesis and relationship with CEH have been extensively studied.11,15,16 However, there has been a lack of information regarding the nature and prevalence of lesions that occur in bitches with subfertility. In 1 study,12 the relative prevalence of various uterine lesions from samples obtained in clinical cases was described; however, only 21 subfertile bitches were included.
The purpose of the study reported here was to determine the prevalence of histologic findings in biopsy samples from a large population of subfertile bitches and evaluate potential associations between age and various types of uterine lesions. We hypothesized that lesions that typically worsen with increasing age, such as fibrosis and cystic changes, would be more prevalent in older than in younger subfertile bitches. By identifying the prevalence of lesions in this population, we hoped to provide initial data to support further investigation into the clinical implications of such findings.
Materials and Methods
Case selection—Submission request forms from 1 small animal veterinary practice that provides reproductive medical services to a large number of patients were reviewed to identify case dogs for which uterine biopsy samples were submitted to the Cornell University Animal Health Diagnostic Center Surgical Pathology Service from January 1, 1990, through December 31, 2005. Dogs were excluded from analysis if uterine tissue was collected through other means, such as ovariohysterectomy or necropsy. If a patient had undergone biopsy more than once, only the first tissue submission was evaluated. Although some cases lacked written clinical histories, they were still included if an incisional endometrial biopsy was performed, because the submitting clinic routinely performs this procedure as an investigation of reproductive health and subfertility.
Medical records review—Submission requests and histopathology reports for all eligible cases were evaluated. Age, breed, history, and clinical diagnosis were reviewed by 1 author (ATG) and entered into spreadsheets. Case histories were reviewed, and dogs were assigned to groups on the basis of reason for biopsy (failure to conceive, fetal death, small litter, reproductive cycle abnormalities, and other [including assessment of uterus after treatment for pyometra]). Additional information from the submission forms was also tabulated if available, including the submitting veterinarian, a description of gross reproductive tract findings, and the type of fixative solution used. The morphological diagnosis and histologic description of samples were recorded, and subjective assessment of lesion severity (mild, moderate, or severe) was noted if included in the histopathology report. The pathologist who examined the sample was also identified.
Lesions were grouped according to their tissue depth, chronicity of the infiltrate, and concurrent inflammatory and cystic changes. Cases were carefully tabulated to ensure that dogs with multiple lesions were not counted twice in analysis. Deep lesions (ie, those that developed in the muscle or serosal layers) were evaluated because these lesions are likely to be missed with superficial sampling methods (such as transcervical biopsy).
Tissue collection and processing—Tissue collection methods were similar for all biopsy samples. When possible, biopsies were performed close to the previously anticipated delivery date for nongravid bitches or during cesarean section for pregnant bitches; however, some samples were obtained at other times as requested by an owner or when necessitated by the clinical condition of the patient. Uterine biopsies were routinely performed through a caudal ventral midline laparotomy with exploration of the intra-abdominal portions of the reproductive tract, including visual examination and digital palpation of the tissues. A full-thickness transmural incisional uterine biopsy sample was generally obtained with a scalpel from any grossly abnormal region of the uterus or, when the tract was grossly normal or diffusely affected, from approximately the midpoint of the left uterine horn. When tissue was obtained from a grossly abnormal uterine horn during caesarian section, an incision separate from the hysterotomy incision was used in an effort to obtain a better sample for diagnostic purposes. The size of tissue samples was not recorded on the submission form or the pathology report.
The tissues were placed in Bouin solution or neutral-buffered 10% formalin fixative solution and submitted to the pathology service, where they were embedded in paraffin, sliced at 4 μm, routinely processed, and stained with H&E. Slides were examined by the Anatomic Pathology Service, which includes a team of resident and board-certified veterinary pathologists. All resident work was verified by a board-certified pathologist.
Statistical analysis—Distribution of data was assessed by examination of a histogram and use of a Shapiro-Wilk normality test. Age had a Gaussian distribution and was summarized as mean and SD. The mean ages of dogs in each severity group (mild, moderate, and severe) for the histologic findings of endometrial fibrosis and CEH were compared with a 1-way ANOVA. Adjustment for multiple comparisons (when appropriate; ie, mean age by severity of CEH and fibrosis) was made with the Tukey honestly significant difference test. Discrete data were summarized as percentages, and comparisons of discrete variables among groups were made with a χ2 test of independence. Dogs with missing data for particular variables (eg, age) were omitted from analysis involving those variables. Statistical tests were performed with commercially available software.a Values of P ≤ 0.05 were considered significant.
Results
Of 403 sample submissions reviewed, 4 cases were excluded because the submitted tissue sample was too small to yield a diagnosis. The remaining 399 cases met the study inclusion criteria.
Of the 399 cases, the mean ± SD age of dogs that underwent uterine biopsy was 4.7 ± 1.6 years (range, 0.5 to 9.3 years). Forty-two (10.5%) dogs were < 3 years of age, 77 (19.3%) were 3 to < 4 years of age, 84 (21.1%) were 4 to < 5 years of age, 78 (19.5%) were 5 to < 6 years of age, and 89 (22.3%) were ≥ 6 years of age. Age information was not available for 29 (7.3%) dogs.
A total of 102 breeds were represented in the study. Among the 396 dogs for which breed information was available, the most common breeds were Golden Retriever (n = 25 [6.3%]), Labrador Retriever (25 [6.3%]), Rottweiler (20 [5.1%]), and German Shepherd Dog (18 [4.5%]). Breed information was not available in 3 (0.7%) cases. Given the large number of breeds represented, lesion prevalence was not evaluated for individual breeds.
Although the differences were not evaluated statistically, the most common reason for biopsy related to lack of conception, which was defined by lack of fetuses identified after breeding (274/399 [68.7%]). Less commonly, dogs reportedly delivered smaller than expected litters or lost conceptuses (37 [9.3%]) or had reproductive cycle abnormalities (24 [6.0%]). Cycle abnormalities included abnormally frequent or short estrous cycles, so-called split heats (ie, 2 estrous cycles within a 2- to 4-week period, where the first estrus is short and anovulatory and the second is ovulatory and of normal duration), persistent estrus, anovulatory cycles, or lack of a discernible estrous cycle. Fourteen dogs had biopsies performed for other reasons (eg, owner request [n = 1], previous treatment of pyometra [8], history of subfertility during previous estrus cycle [3], biopsy performed at time of surgical artificial insemination [1], or history of medical estrous cycle management with mibolerone solution [1]), and a clinical history was not recorded when the sample was submitted for 49 (12.3%) of 399 dogs. However, given that endometrial biopsies are performed at the submitting clinic to gain prognostic information regarding future reproductive success for patients, the cases without histories available were retained in the data set. Fetal loss during the current reproductive cycle was noted in 87 (21.8%) cases and was not found or not reported in the remaining cases.
In 29 of 399 (7.3%) cases, biopsy samples were collected at the time of caesarian section. Samples from most (380 [95.2%]) dogs were submitted by 1 veterinarian, with those for remaining dogs (19 [4.8%]) submitted by 6 other associates of the practice. Forms submitted by the veterinarian who performed the surgery indicated the uterine tissue was grossly abnormal in 304 of 399 (76.2%) cases and grossly normal in 20 (5.0%) cases. In 75 (18.8%) cases, submissions did not include a gross description of the reproductive tract. Types of uterine abnormalities included segmental distension of the tissue, uterine asymmetry, presence of mucoid luminal secretions, and abnormal tissue consistency detected by digital palpation of the uterus.
Records indicated that most (271/399 [67.9%]) biopsies were performed during the late portion of luteal phase; however, true assessment of the reproductive cycle stage in the study was difficult because of an absence of data on concurrent vaginal cytology or hormonal testing. Breeding was reportedly timed with progesterone tests in 178 (44.6%) cases, but this was not known or not reported for the remaining cases.
Bouin solution was used for sample fixation in 334 of 399 (83.7%) cases, and in 1 (0.3%) case, 10% formalin solution was used. In 64 (16.0%) cases, fixative type was not recorded.
Tissue samples for most (316/399 [79.2%]) cases were reviewed by 1 author who is board certified in both veterinary pathology and theriogenology (DHS). Evaluations for the remaining 83 (20.8%) cases were performed by other pathologists (15 individuals) in the Cornell University Section of Anatomic Pathology. Although these pathologists generally consult with the theriogenologist (DHS) on cases involving canine uterine biopsy, this information was not recorded.
Uterine tissues frequently had combinations of lesions present, yielding a mean of 2.7 morphological diagnoses/case. Lesion types and prevalence were summarized (Table 1). The most prevalent lesion was endometritis (170/399 [42.6%] cases), followed by cystic changes of the endometrium, including CEH (133 [33.3%]). Cystic endometrial hyperplasia ranged from mild to severe, resulting in pronounced distortion of endometrial architecture in some cases (Figure 1). Edema was identified in 131 (32.8%) cases, and hemosiderophages were noted in 121 (30.3%). In 111 (27.8%) cases, no lesion was detected histologically.
Photomicrographs of uterine tissue in full-thickness biopsy samples collected from a clinically normal bitch in anestrus (A) and 1 of 399 bitches examined because of subfertility (B). In panel A, notice the architecture of the endometrium and endometrial glands (arrows) and layers of myometrium (bracketed region). In panel B, the endometrial architecture is distorted and cystic endometrial glands (†) are present. Stars indicate the uterine lumen.
Citation: Journal of the American Veterinary Medical Association 244, 2; 10.2460/javma.244.2.180
Prevalence (No. [% of group]) of various histopathologic findings in full-thickness uterine biopsy samples obtained from 399 bitches with subfertility, grouped according to age.
| Age (y) | |||||||
|---|---|---|---|---|---|---|---|
| Variable | < 3 (n = 42) | 3 to < 4 (n = 77) | 4 to < 5 (n = 84) | 5 to < 6 (n = 78) | ≥ 6 (n = 89) | Unknown (n = 29) | All (n = 399) |
| Endometritis | 20 (47.6) | 28 (36.4) | 35 (41.7) | 40 (51.3) | 37 (41.6) | 10 (34.5) | 170 (42.6) |
| Chronic | 10 (23.8) | 15 (19.5) | 16 (19.0) | 29 (37.2) | 15 (16.9) | 4 (13.8) | 89 (22.3) |
| Acute | 3 (7.1) | 6 (7.8) | 5 (6.0) | 4 (5.1) | 8 (9.0) | 4 (13.8) | 30 (7.5) |
| Subacute | 7 (16.7) | 7 (9.1) | 14 (16.7) | 6 (7.7) | 14 (15.7) | 3 (10.3) | 51 (12.8) |
| Eosinophilic | 3 (7.1) | 5 (6.5) | 7 (8.3) | 5 (6.4) | 6 (6.7) | 3 (10.3) | 29 (7.3) |
| Cystic changes (including CEH) | 8 (19.0) | 23 (29.9) | 34 (40.5) | 27 (34.6) | 29 (32.6) | 12 (41.4) | 133 (33.3) |
| Edema | 13 (31.0) | 32 (41.6) | 32 (38.1) | 27 (34.6) | 23 (25.8) | 4 (13.8) | 131 (32.8) |
| Hemosiderosis | 9 (21.4) | 21 (27.3) | 28 (33.3) | 31 (39.7) | 20 (22.5) | 12 (41.4) | 121 (30.3) |
| Fibrosis | 12 (28.6) | 14 (18.2) | 24 (28.6) | 26 (33.3) | 16 (18.0) | 9 (31.0) | 101 (25.3) |
| Periglandular | 6 (14.3) | 12 (15.6) | 21 (25.0) | 22 (28.2) | 14 (15.7) | 6 (20.7) | 81 (20.3) |
| Interstitial | 1 (2.4) | 2 (2.6) | 5 (6.0) | 3 (3.8) | 0 (0.0) | 0 (0.0) | 11 (2.8) |
| Myometrial | 5 (11.9) | 0 (0.0) | 1 (1.2) | 0 (0.0) | 5 (5.6) | 3 (10.3) | 14 (3.5) |
| Mucometra | 5 (11.9) | 7 (9.1) | 11 (13.1) | 7 (9.0) | 12 (13.5) | 4 (13.8) | 46 (11.5) |
| Vascular hypertrophy | 5 (11.9) | 11 (14.3) | 7 (8.3) | 9 (11.5) | 10 (11.2) | 4 (13.8) | 46 (11.5) |
| Gland loss and gland atrophy | 5 (11.9) | 8 (10.4) | 3 (3.6) | 7 (9.0) | 8 (9.0) | 2 (6.9) | 33 (8.3) |
| Uterine involution | 5 (11.9) | 7 (9.1) | 4 (4.8) | 5 (6.4) | 5 (5.6) | 2 (6.9) | 28 (7.0) |
| Metritis | 2 (4.8) | 3 (3.9) | 4 (4.8) | 2 (2.6) | 2 (2.2) | 2 (6.9) | 15 (3.8) |
| Gland ectasia | 4 (9.5) | 2 (2.6) | 2 (2.4) | 3 (3.8) | 2 (2.2) | 2 (6.9) | 15 (3.8) |
| Necrosis | 0 (0.0) | 2 (2.6) | 3 (3.6) | 2 (2.6) | 5 (5.6) | 2 (6.9) | 14 (3.5) |
| PEH | 1 (2.4) | 3 (3.9) | 2 (2.4) | 3 (3.8) | 5 (5.6) | 0 (0.0) | 14 (3.5) |
| Reproductive cycle stage mismatch | 0 (0.0) | 3 (3.9) | 4 (4.8) | 4 (5.1) | 1 (1.1) | 0 (0.0) | 12 (3.0) |
| Pyometra | 1 (2.4) | 1 (1.3) | 3 (3.6) | 0 (0.0) | 4 (4.5) | 0 (0.0) | 9 (2.3) |
| Adenomyosis | 0 (0.0) | 1 (1.3) | 1 (1.2) | 1 (1.3) | 1 (1.1) | 2 (6.9) | 6 (1.5) |
| Subinvolution of placental sites | 0 (0.0) | 1 (1.3) | 0 (0.0) | 1 (1.3) | 0 (0.0) | 1 (3.4) | 3 (0.8) |
| Arteritis | 0 (0.0) | 1 (1.3) | 0 (0.0) | 0 (0.0) | 1 (1.1) | 0 (0.0) | 2 (0.5) |
| Mucocoele | 0 (0.0) | 0 (0.0) | 1 (1.2) | 0 (0.0) | 1 (1.1) | 0 (0.0) | 2 (0.5) |
| Polyp | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 1 (1.1) | 0 (0.0) | 1 (0.3) |
| Serosal inclusion cyst | 0 (0.0) | 0 (0.0) | 1 (1.2) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 1 (0.3) |
| No lesions | 12 (28.6) | 25 (32.5) | 22 (26.2) | 15 (19.2) | 27 (30.3) | 10 (34.5) | 111 (27.8) |
Some dogs had > 1 lesion type.
PEH = Physiologic endometrial hyperplasia.
Of the 170 dogs with endometritis, 89 (52.4%) had lymphocytic or lymphoplasmacytic infiltrates detected and were classified as having chronic disease. Fifty-one (30.0%) had subacute disease with mixed inflammation. Acute endometritis with neutrophilic or eosinophilic infiltrates, or both, was detected in 30 (17.6%) of these dogs (Figure 2). Some degree of fibrosis was detected in 101 of 399 (25.3%) dogs, with periglandular fibrosis in 81 (80.2%), interstitial fibrosis in 11 (10.9%), and fibrosis of the myometrium in 14 (13.9%) of these 101 cases of fibrosis.
Photomicrograph of endometrial tissue from a subfertile bitch with severe acute endometritis that had fetal loss reported during the same breeding cycle when the biopsy was obtained. Numerous eosinophils (arrow) have infiltrated the interstitium and are seen exocytosing through the endometrial epithelial lining.
Citation: Journal of the American Veterinary Medical Association 244, 2; 10.2460/javma.244.2.180
When dogs were grouped according to lesion type, mean age was not significantly different among groups. Similarly, when dogs were grouped by age, no significant increase in prevalence of any lesion was found for any group, compared with dogs of other ages. Given the presence of multiple lesions in many dogs, statistical comparisons of the distribution of lesions across age groups were not made.
No significant association was found between the presence of endometritis and CEH (P = 0.77). Of the 133 dogs with CEH, endometritis was concurrently diagnosed in 58 (43.6%). Mean age was not significantly (P = 0.64) different among dogs with mild (4.7 ± 1.4 years; n = 66), moderate (4.6 ± 1.2 years; 17), or severe (4.2 ± 1.9 years; 9) endometrial fibrosis. Among dogs with CEH, the mean age of those with moderate lesions (5.2 ± 1.7 years; n = 34) was significantly (P < 0.05) higher than that of dogs with mild lesions (4.4 ± 1.1 years; 45). Mean age did not differ significantly among dogs with mild versus severe (4.6 ± 0.3 years; n = 23) or moderate versus severe CEH lesions.
Deep lesions were found in 36 of 399 (9.0%) dogs evaluated, including dogs with metritis (n = 15 [3.8%]), myometrial fibrosis (14 [3.5%]), adenomyosis (6 [1.5%]), and serosal inclusion cysts (1 [0.3%]). Chronic lesions (lymphocytic endometritis, lymphoplasmacytic endometritis, periglandular fibrosis, myometrial fibrosis, and interstitial fibrosis) were detected in 195 of 399 (48.9%) cases. Acute lesions (neutrophilic endometritis, eosinophilic endometritis, and edema) were found in 159 (39.8%) cases.
Evaluation of the potential association between eosinophilic endometritis and fetal loss revealed that a significantly (P = 0.008) higher proportion of dogs with eosinophilic endometritis (41.4% [12/29]) had clinical histories suggestive of fetal loss, compared with the 75 of 370 (20.3%) with fetal loss among dogs with other diagnoses.
Discussion
The 2 most common histopathologic findings in the study were endometritis (170/399 [42.6%] dogs) and cystic endometrial changes (including CEH; 133 [33.3%]).
Our hypothesis that age would be associated with severity of cumulative lesions was not supported by the data. Although dogs with moderate CEH were significantly older than those with mild CEH (5.2 vs 4.4 years), mean age of dogs with severe CEH (4.6 years) did not differ significantly from that of dogs with mild or moderate lesions. There were also no significant differences in the mean ages of dogs with mild, moderate, or severe fibrosis. The lack of significance could be attributable to the range of lesions or lack of definitive criteria for categorization as mild, moderate, and severe. More research is needed to further examine the possible relationships between age and severity of cumulative lesions.
Cystic endometrial hyperplasia has been described as the most common uterine disease of bitches17; however, in our study population of 399 subfertile bitches, endometritis was found in 170 (42.6%) patients and CEH was found in 133 (33.3%). Only recently have studies18,19 focused on endometrial inflammation in bitches, which is somewhat surprising given its importance and the attention that endometritis in mares has received.4,8,20
In a study by Fontaine et al,18 endometritis was a common finding (38% prevalence) in a population of 26 infertile bitches assessed through cytologic evaluation and microbial culture of samples collected with a transcervical technique, suggesting that endometritis likely plays a role in canine subfertility, as has been described for other species. Most (7/10) cases of endometritis in that study18 were attributed to bacterial infection, and half (5/10) responded favorably to antimicrobial treatment, with successful breeding reported afterward. The specific causes for endometrial inflammation in our study, if known, were not recorded in the submission histories, and results of microbial culture and other diagnostic tests were not available. The findings of Fontaine et al,18 along with the prevalence of endometritis identified in the present study, suggest that presence and severity of endometritis may be useful criteria in the future development of an endometrial scoring system for use in dogs.
Investigation of the concurrent presence of inflammatory and cystic changes in our study indicated that 58 (43.6%) of 133 dogs with CEH also had endometritis; no significant association was found between these 2 lesion types. Dow15 also evaluated the concurrent presence of endometritis and cystic hypertrophy in the canine endometrium. Results of that study15 indicated that CEH and endometritis frequently occurred together, but in some cases, CEH occurred without inflammatory changes. Other researchers have described variability among cystic lesions, once solely termed CEH,21 and these lesions were shown to have different lectin binding patterns, compared with those found in the endometrium of clinically normal bitches.22 In a more recent investigation,12 new nomenclature for further differentiation of cystic changes was suggested.
Dow15 also found that acute inflammation was more common in the early portion of the luteal phase and chronic inflammation was more common in the late portion of the luteal phase in clinically normal and affected dogs. Chronic endometritis was more commonly detected (89/399 [22.3%]) than acute endometritis (30/399 [7.5%]) in subfertile bitches of the present study, and this may relate to the fact that most (271/399 [67.9%]) dogs had samples collected during the late part of the luteal phase; however, this difference was not tested statistically. Because there were relatively few submissions in our study from patients in the early part of the luteal phase, the data for further assessment were limited.
Uterine fibrosis was detected in 101 of 399 (25.3%) dogs in the present study but was typically mild, even in cases of chronic endometritis. In the scoring system used for mares, the amount of fibrosis is one of the criteria used in assigning a prognostic score, with values assigned on the basis of the number of fibrocyte layers present around glands and vessels.4,8 Given the low overall prevalence of endometrial fibrosis, it is plausible that the lesion is either more subtle in the dog than in the mare or less prevalent overall in the dog.
Edema and hemosiderophages were common findings (detected in samples from 131 [32.8%] and 121 [30.3%] of 399 dogs, respectively) in our study. Edema can often be attributed to an artifact of sample collection and tissue handling. In a previous study,23 hemosiderophages were noted on the cytologic evaluation of samples from clinically normal bitches during early and mid anestrus. In the present study, samples were typically collected in the late portion of the luteal phase. We interpret the presence of hemosiderophages in the mid and late luteal periods as persistence of the hemophagocytosis that normally occurs in response to pro-estrual bleeding in bitches and continues for several weeks. In our experience, hemorrhage is not a common feature of either acute or chronic endometritis, with the exception of very marked lesions. Further investigation is needed to evaluate the presence of hemosiderophages during different stages of the reproductive cycle.
The prevalence of eosinophilic endometritis (29/399 [7.3%] dogs) was a surprising finding. Given the presence of eosinophils noted on cytologic evaluation of a sample from a postpartum bitch in early anestrus in an earlier study,23 cases in our study were reviewed, and a significant (P = 0.008) association was found between eosinophilic endometritis and fetal loss. Further investigation is needed to determine the clinical importance of eosinophilic infiltrates in this tissue. We speculate whether the presence of eosinophils could be a cause of fetal loss or could possibly be associated with placental maturation and late pregnancy or other tissue responses. In horses, clusters of eosinophilic infiltrates have been found in proximity to sloughing endometrial cups.24
In many (111/399 [27.8%]) cases, no clinically relevant lesions were found in biopsy samples that might explain subfertility in dogs of the present study. This may be suggestive of nonuterine causes of reproductive issues, such as breeding management problems, endocrinopathies, oocyte incompetence, ovulatory failure, oviductal defects, and male fertility factors, or it may reflect the subtlety of clinically related changes in the canine endometrium. In the cases where lesions were detected, there was a mean of 2.7 diagnoses/case, exemplifying a continuum of changes present within these tissues. This shows the need for a scoring system that will allow for a more specific, clinically relevant diagnosis on the basis of histologic findings.
In general, the quality of the samples in the present study was considered excellent. Only 4 (1.0%) cases had too little tissue submitted for adequate diagnostic evaluation. This suggests that, overall, surgical incisional biopsy yields a suitable sample. In mares and women, transcervical endometrial biopsies are generally obtained, rather than full-thickness uterine biopsies. Cytologic evaluation of the endometrium in dogs may also be useful; however, 1 group of investigators described difficulty in differentiating leukocytes and pyknotic endometrial epithelial cells in uterine samples collected for this purpose.23 This is less problematic for incisional biopsy samples collected for histologic evaluation because the endometrial cells are evaluated in their normal architecture.
A recent report9 described successful transcervical collection of canine endometrial samples. However, the results suggested possible predisposition to pyometra in dogs that had samples collected in mid to late diestrus or a possible lack of sensitivity for detection of pyometra when transcervical collection methods were used.
The development of safe transcervical tissue collection techniques could potentially expand the use of endometrial biopsy in dogs in clinical and research settings.9,18,19 In addition to amount of tissue, acquisition of representative tissue layers should be considered in comparing grasp-type biopsy to full-thickness incisional biopsy of the endometrium. Deep lesions were found in samples from 36 of 399 (9.0%) dogs in the present study. These lesions could have been missed with more superficial sample collection techniques.10 However, results of 1 study9 performed to compare transcervical and full-thickness incisional biopsy techniques in dogs indicated that many of the samples obtained transcervically included basal glands and myometrium.
The present study had several limitations, such as lack of complete clinical history and other patient information in some cases because of the retrospective study design. The population was relatively small, and sufficient information was not available to assess the prevalence of subfertility among breeds. Also, the number of animals that underwent ultrasonography or other perioperative diagnostic testing could not be determined. Presurgical imaging or other testing results may have directed decisions to pursue ovariohysterectomy rather than uterine biopsy, eliminating some dogs that might have had more severe or grossly apparent lesions from the study population. Patient age also may have influenced this decision. The mean age of dogs that underwent biopsy was 4.7 years, with a range from 0.5 to 9.3 years. Most patients in the study (281/399 [70.4%]) were < 6 years old, and some owners of older dogs may have removed them from breeding programs or elected spaying rather than pursuing diagnostic biopsy when reproductive problems developed. Finally, clinical follow-up was not available for the dogs in this study. Future studies that assess the clinical prognosis for patients treated according to their biopsy results are needed.
ABBREVIATION
| CEH | Cystic endometrial hyperplasia |
Statistix 9, Analytic Software, Tallahassee, Fla.
References
1. Revel A. Multitasking human endometrium: a review of endometrial biopsy as a diagnostic tool, therapeutic applications, and a source of adult stem cells. Obstet Gynecol Surv 2009; 64:249–257.
2. Clark TJ, Mann CH, Shah N, et al. Accuracy of outpatient endometrial biopsy in the diagnosis of endometrial cancer: a systematic quantitative review. BJOG 2002; 109:313–321.
3. de la Concha-Bermejillo A, Kenny PC. Prognostic value of endometrial biopsy in the mare: a retrospective analysis. J Am Vet Med Assoc 1982; 181:680–681.
4. Kenney RM. Cyclic and pathologic changes of the mare endometrium as detected by biopsy, with a note on early embryonic death. J Am Vet Med Assoc 1978; 172:241–262.
5. Kenney RM, Doig PA. Equine endometrial biopsy. In: Morrow DA, ed. Current therapy in theriogenology. 2nd ed. Philadelphia: WB Saunders Co, 1986;723–729.
6. Schlafer DH. Equine endometrial biopsy: enhancement of clinical value by more extensive histopathology and application of new diagnostic techniques? Theriogenology 2007; 68:413–422.
7. Snider TA, Sepoy C, Holyoak GR. Equine endometrial biopsy reviewed: observation, interpretation, and application of histopathologic data. Theriogenology 2011; 75:1567–1581.
8. Van Camp SD. Endometrial biopsy of the mare—a review and update. Vet Clin North Am Equine Pract 1988; 4:229–245.
9. Christensen BW, Schlafer DH, Agnew DW, et al. Diagnostic value of transcervical biopsies in domestic dogs compared to full thickness uterine sections. Reprod Domest Anim 2012; 47:342–346.
10. Günzel-Apel AR, Wilke M, Aupperle H, et al. Development of a technique for transcervical collection of uterine tissue in bitches. J Reprod Fertil Suppl 2001;(57):61–65.
11. Hadley JC. The development of cystic endometrial hyperplasia following serial uterine biopsies. J Small Anim Pract 1975; 16:249–257.
12. Mir F, Fontaine E, Albarix O, et al. Findings in uterine biopsies obtained by laparotomy from bitches with unexplained infertility or pregnancy loss: an observational study. Theriogenology 2013; 79:312–322.
13. Anderson AC, Simpson ME. The ovary and reproductive cycle of the dog (Beagle). Los Altos, Calif: Geron-X Inc, 1973.
14. Al-Bassam MA, Thomson RG, O'Donnell L. Normal postpartum involution of the uterus in the dog. Can J Comp Med 1981; 45:217–232.
15. Dow C. The cystic hyperplasia-pyometra complex in the bitch. Vet Rec 1958; 70:1102–1110.
16. Schlafer DH, Gifford AT. Cystic endometrial hyperplasia, pseudoplacentational endometrial hyperplasia, and other cystic conditions of the canine and feline uterus. Theriogenology 2008; 70:349–358.
17. Verstegen J, Dhaliwal G, Verstegen-Onclin K. Mucometra, cystic endometrial hyperplasia, and pyometra in the bitch: advances in treatment and assessment of future reproductive success. Theriogenology 2008; 70:364–374.
18. Fontaine E, Levy X, Grellet A, et al. Diagnosis of endometritis in the bitch: a new approach. Reprod Domest Anim 2009; 44:196–199.
19. Ribeiro AP, Vicente WR, Apparício M, et al. Uterine leucocyte infiltration after artificial insemination in bitches. Theriogenology 2006; 66:1462–1464.
20. Blanchard TL, Garcia MC, Kintner LD, et al. Investigation of the representativeness of a single endometrial sample and the use of trichrome staining to aid in the detection of endometrial fibrosis in the mare. Theriogenology 1987; 28:445–450.
21. Bartel C, Schokypl S, Walter I. Pseudo-placentational endometrial cysts in a bitch. Anat Histol Embryol 2010; 39:74–80.
22. Leitner M, Aurich JE, Galabova G, et al. Lectin binding patterns in normal canine endometrium and in bitches with pyometra and cystic endometrial hyperplasia. Histol Histopathol 2003; 18:787–795.
23. Watts JR, Wright PJ, Lee CS. Endometrial cytology of the normal bitch throughout the reproductive cycle. J Small Anim Pract 1998; 39:2–9.
24. Grünig G, Triplett L, Canady LK, et al. The maternal leukocyte response to the endometrial cups in horses is correlated with the developmental stages of the invasive trophoblast cells. Placenta 1995; 16:539–559.
