Theriogenology 83 (2015) 222–227
Contents lists available at ScienceDirect
Theriogenology journal homepage: www.theriojournal.com
Diagnostic double-guarded low-volume uterine lavage in mares M. Christoffersen a, *, L. Brandis a, J. Samuelsson a, A.M. Bojesen b, M.H.T. Troedsson c, M.R. Petersen d a
Veterinary Reproduction and Obstetrics, Department of Large Animal Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark Veterinary Clinical Microbiology, Department of Veterinary Disease Biology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark c Maxwell H. Gluck Equine Research Center, University of Kentucky, Lexington, Kentucky, USA d The Fertility Clinic, Section 4071, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 March 2014 Received in revised form 5 September 2014 Accepted 10 September 2014
Endometritis constitutes a major problem in the management of broodmares; hence, diagnostic tests with a high sensitivity and specificity are highly appreciated. The aim of this study was to compare the results from endometrial, cytologic, and bacteriologic examinations obtained by a newly developed, double-guarded, flushing technique versus standard diagnostic tests, the double-guarded swab and biopsy. The described doubleguarded flush technique requires the use of a disposable uterine flushing tube, a sanitary sleeve, a sterile steel speculum, and a 250 mL fluid bag. Endometrial biopsies, swabs, and low-volume lavage samples were obtained from 34 research mares at six different time points in four estrous cycles and were evaluated cytologically and bacteriologically. Endometrial biopsies from the first cycle (n ¼ 34) were examined for the presence of polymorphonuclear neutrophils (PMNs) in the stratum compactum and stratum spongiosum and used as a gold standard for calculation of diagnostic sensitivity and specificity. In all samples, Escherichia coli was most frequently isolated (lavage, 30%; swab, 21%; and biopsy, 12%) followed by b-hemolytic streptococci (lavage, 11%; swab, 8%; and biopsy, 7%). Positive cytology was less likely to occur when E coli was isolated from the diagnostic tests compared with the growth of b-hemolytic streptococci. Isolation of pathogens from uterine samples was highly associated with the presence of PMNs in the stratum compactum and straum spongiosum on histology. Using the presence of PMNs in the tissue specimens as the gold standard for diagnosing endometritis, the sensitivity of low-volume lavage culture was 0.75 and the specificity was 0.72. In conclusion, the double-guarded, low-volume, lavage technique was a rapid and accurate method for diagnosing mares with endometritis, and the risk of false-positive samples is considered to be minimal compared with other flushing techniques described. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Endometritis Mare Low-volume lavage Diagnostic test Escherichia coli b-hemolytic streptococci
1. Introduction
* Corresponding author. Tel.: þ45 35332983; fax: þ45 35332968 E-mail address: [email protected] (M. Christoffersen). 0093-691X/$ – see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.theriogenology.2014.09.008
Endometritis is one of the main reasons for reduced fertility in the mare [1]. Accurate diagnosis of endometritis and identification of pathogens involved are necessary to initiate correct treatment in time to optimize fertility and reduce the risk of bacterial resistance development.
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227
223
Fig. 1. (A) The steel speculum, sanitary sleeve, disposable uterine lavage tube, and fluid bag for the double-guarded, low-volume, lavage technique. (B) Doubleguarded lavage equipment ready for use. (For interpretation of the references to color in this Figure, the reader is referred to the web version of this article.)
Nielsen [2] was the first to evaluate diagnostic sensitivity and specificity in regards to diagnosis of endometritis in the mare. In this study, endometrial samples were collected from the same mares twice using a double-guarded swab and biopsy, which were used for bacterial culture and cytology. Histology was performed on the recovered endometrial biopsy, and the presence of polymorphonuclear neutrophils (PMNs) was used as “gold standard” when determining whether the diagnostic test result was true or false. The diagnostic sensitivity after culture from the commonly used endometrial swab or biopsy was 0.34 and 0.82, respectively. In other words, the risk of a false-negative diagnosis of endometritis was 66% or 18%, using culture from a swab or biopsy, respectively [2]. LeBlanc and co-workers evaluated the diagnostic sensitivity and specificity of a low-volume uterine lavage technique, initially suggested by Ball et al. [3,4]. In a comprehensive study using the same “gold standard” as Nielsen, 2005, sensitivity and specificity of culture and cytology from a low-volume uterine lavage were found to be 0.71 and 0.86, respectively, and 0.80 and 0.67, respectively [4]. Low-volume uterine lavage was found to be a sensitive diagnostic test, with sensitivity and specificity comparable to reports using an endometrial biopsy. These results suggested a marked reduction in the number of false-negative samples when a low-volume lavage is used compared with the double-guarded swab. Using the procedure described by LeBlanc et al. [4], the lavage tube is passed unguarded through the caudal reproductive tract and cervix, with a high risk of contamination from the commensal flora. To help identify false-positive samples, it was suggested to only categorize samples as positive if growth was obtained together with a cloudy efflux fluid and debris on cytology. Using this approach, the number of false-positive samples was estimated to be 11% [4]. Endometrial cytology is well correlated to bacteriologic finding [2,5,6], although subclinical endometritis caused by E. coli has been found to be associated with negative cytology specimens compared with cytology from gram-positive pathogens [2,4,7]. These findings emphasize the importance of using both cytology and bacteriologic findings when diagnosing endometritis in the mare. This study was conducted to determine if a newly developed, double-guarded flushing technique would improve the sensitivity of identifying endometritis in the mare compared with the double-guarded uterine swab and uterine biopsy techniques.
2. Materials and methods 2.1. Mares and sample collection A total of 34 light horse mares of mixed breeds of age 3 to 25 years with an unknown reproductive history were included in the study. All animal procedures were carried out in accordance to and with the approval of the Institutional Animal Care and Use Committee at the University of Kentucky. No abnormalities were noted on the clinical and gynecological examinations performed in the cycle before the experiment started. The mares were found normal on gynecological examination, and a double-guarded, lowvolume, uterine lavage, endometrial biopsy and a uterine swab were obtained during estrous (>30 mm follicle, endometrial edema, and decreased uterine tone). Uterine samples were obtained from each mare six times across four estrus periods, and all uterine samples were obtained before ovulation. The mares were restrained in an examination stock. After rectal palpation and transrectal ultrasonographic examination, the vulva and perineum were washed with chlorhexidine medical scrub (Dermachlor, Butler Animal Supply, Lexington, KY, USA), rinsed three times with water, and dried with paper towels. All three samples, the double-guarded swab, the low-volume double-guarded lavage, and the endometrial biopsy, were obtained from all mares at each sample time point and in the same order. A double-guarded swab (EQUIVET; Kruuse A/S, Langeskov, Denmark) was used to obtain an endometrial culture. The swab was kept in contact with the endometrial surface for at least 30 seconds. To obtain the double-guarded, low-volume, lavage sample, a sterile steel speculum (EQUIVET) covered with a sterile sanitary sleeve (also termed “chemise” and used to cover an embryo transfer instrument; Kruuse A/S; Fig. 1) was passed manually per vaginam into the cervical canal. When placed in the most cranial part of the cervix, the sanitary sleeve was pulled toward the examiner from the outside and the sterile speculum pushed through the sanitary sleeve, through the cervical canal and into the uterine lumen. A disposable uterine lavage tube (EQUIVET) was attached to a 250 mL fluid bag (Ringers Lactate; Butler Animal Supply) and passed through the speculum and into the uterus, followed by retraction of the speculum toward the examiner and out of the mares’ genital tract. Using this guarded approach, the lavage tube was placed in a sterile manner in the uterine body. The tip of the flushing tube was kept in the caudal aspect of the uterine body by a firm hold around
224
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227
Table 1 Distribution of culture results using low-volume lavage (LVL), swab, and biopsy as diagnostic test at six sampling time points in four estrous cycles. Culture results
LVL, n (%)
Swab, n (%)
Biopsy, n (%)
Culture positive Contamination Sterile E. coli b-strep E. coli þ b-strep Other pathogens Total
122 (61)a 7 (4) 70 (35)a 60 (30)a 21 (11) 39 (20)a 2 (1) 199
71 (36)b 4 (2) 124 (62)b 42 (21)b 16 (8) 8 (4)b 5 (3) 199
51 (26)c 5 (3) 143 (71)b 24 (12)c 15 (7) 5 (3)b 7 (3) 199
Abbreviations: b-strep, beta-hemolytic streptococci; E. Coli, Escherichia coli. a,b,c Different letters indicate significant difference (P < 0.05) between the diagnostic tests.
the vaginal part of the cervix. The fluid was infused into the uterus, and the efflux fluid was recovered by gravity flow from the uterus and into the fluid bag in a closed system. A minimum of 150 mL of the efflux fluid was recovered. In some cases, the lavage tube was rotated, advanced, or retracted a few centimeters to avoid suction to the endometrium and to allow fluid contact. If less than 150 mL of fluid was recovered, the uterus was manipulated by rectal palpation. One hand of the examiner maintained the lavage tube in position per vaginam, whereas the other hand was used to manipulate the fluid toward the lavage tube by rectal palpation. Endometrial biopsies were collected using an alligator jaw biopsy punch introduced into the uterus through a sterile speculum (EQUIVET) as described by Nielsen [2]. 2.2. Microbiology The recovered fluid bag was sealed with a pair of forceps and transported to the laboratory and within 4 hours and hung for sedimentation by gravity for minimum 1 hour. A total of 50 mL was aspirated with a sterile needle from the most ventral part of the bag (including sediment) into a 50mL tube (Falcon; Fisher Scientific, Hampton, NH, USA) and centrifuged at 400g for 10 minutes. Of the 50 mL, 48 mL of the supernatant fluid was poured off. The sediment was suspended in the remaining 2 mL. For bacterial culture, a sterile cotton swab was dipped in the 2-mL sample and smeared on a blood agar (5% horse blood). Endometrial biopsies and uterine swabs were streaked on blood agars. All blood agar plates were cultured aerobically for 24 hours at 37 C. Bacterial growth was identified according to colony morphology, Hancock stain morphology, hemolysis, catalase, and potassium hydroxide (3% KOH) tests. Colonies were counted on blood agar plates and scored: no growth or sterile: 4 or lesser colony forming units (CFU) or positive growth: greater than 4 CFU, and culture results were recorded as E. coli, beta-hemolytic streptococci (b-strep), E. coli þ b-strep, or other pathogens. If more than three different bacteria were present at the blood agar plate, the result was recorded as contamination.
slides, which were dried at room temperature and stained with a quick staining method Diff-Quik (Fisher Scientific) and evaluated by light microscopy (400 magnification). Cytologic classification was based on numbers of PMNs present per 200 cells examined. The sample was considered positive for inflammation when greater than 0.5% PMNs were present as described by Nielsen [2]. 2.4. Histology The endometrial biopsies obtained in the first cycle were used for histopathologic examination. Endometrial biopsies were fixed in 10% formalin, embedded in paraffin and sectioned at 5 mm, and stained with hematoxylin and eosin. Slides were examined for the presence of PMNs within the luminal epithelium, stratum compactum, and stratum spongiosum. Infiltration of more than three PMNs per five fields of high magnification (400) was considered as evidence of acute endometritis [8]. 2.5. Statistical analyses Categorical variables were compared using chi-square test, and significance was set at P < 0.05. Sensitivity and specificity were calculated using the presence of PMNs on endometrial histology from biopsies obtained in the first cycle as the gold standard for predicting endometritis. Sensitivity was calculated as the proportion of mares with PMNs in the endometrium, and a positive culture result from the compared tests collected at the same time. Specificity was calculated as the proportion of mares without PMNs in the endometrial tissue and a negative result from the compared test [2]. To test the effect of the different diagnostic tests and cycle number on a dichotomous outcome and evaluation of the effect, the presence of different pathogens was analyzed by logistic regression. A logit transformation of data was used to describe the relationship between the outcome and the explanatory variable. Generalized score test (the Wald test) was used in the type 3 analysis. Significant differences in the outcome of the explanatory variable were identified by using the syntax for estimates. All analyses were performed using SAS 9.3 (SAS Institute, Cary, NC, USA). The level of significance was set to P < 0.05. Graphs were made using the software GraphPad Prism 5.0 (GraphPad software Inc., La Jolla, CA, USA). 3. Results In total, 199 uterine swabs, 199 uterine lavage samples, and 199 biopsies were obtained from the 34 mares at six different sampling time points in four estrous cycles. Samples obtained from cycle 1 (34 biopsies, 34 low-volume lavages, and 34 swabs) were used for calculating the sensitivity, specificity, and positive and negative predictive values for the three diagnostic tests evaluated using PMNs on the histologic slides as gold standard.
2.3. Cytology
3.1. Microbiology
The biopsies, swabs, and sediment from the low-volume lavage samples from the first cycle were smeared on glass
From all samples, bacteria were isolated from 122 of 199 uterine lavages (61%), 71 of 199 swabs (36%), and
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227 Table 2 Presence of polymorphonuclear neutrophils on cytologic slides related to samples with no growth of pathogens (sterile) and samples with growth of E. coli, b-strep, and other pathogens (OP) in cycle 1. Culture results
Swab, n (%)
LVL, n (%)
Biopsy, n (%)
Cyt
Cytþ
Cyt
Cytþ
Cyt
Cytþ
Sterile E. coli b-strep OP Total
26 (76) 5 (15)a 1 (3)b 0 32 (94)
0 0 2 (6) 0 2 (6)
20 (59) 13 (38)a 0a 0 33 (97)
0 0 1 (3) 0 1 (3)
27 (79) 1 (3)a 1 (3)a 0 29 (85)
1 3 1 0 5
(3) (9) (3) (15)
Abbreviations: b-strep, beta-hemolytic streptococci; Cytþ, cytology positive; Cyt, cytology negative; E. coli, Escherichia coli; LVL, low-volume lavage. a,b Different letters indicate significant difference (P < 0.05) of cytologic response to E. coli and b-strep within each diagnostic test.
225
3.2. Cytology The presence of PMNs on cytologic slides obtained from the three diagnostic tests in cycle 1 was not associated with the positive culture of uterine pathogens (Table 2). In total, PMNs were only observed in 8 of 102 (8%) of the cytologic specimens from the 34 biopsies, 34 lavages, and 34 swabs (cycle 1). One mare had positive cytology (biopsy) with no growth of uterine pathogens on any of the uterine samples obtained. The number of cytology-negative but E. coli positive samples was greater compared with cytology-negative samples culture positive for b-strep (P ¼ 0.02) when using the double-guarded swab. 3.3. Histology
51 of 199 biopsies (26%; Table 1). Seven of 199 (4%) uterine lavage samples, 4 of 199 (2%) uterine swabs, and 5 of 199 (3%) biopsies were registered as contaminated. Escherichia coli was the most frequently isolated pathogen from low-volume lavage (30%), swabs (21%), and biopsy specimens (12%), followed by b-strep, which were cultured from low-volume lavage samples (11%), swabs (8%), and biopsies (7%). E. coli and b-strep were isolated in mixed culture from 20% of the low-volume lavage samples, 4% of the swabs, and 3% of the biopsy specimens. E. coli and b-strep were isolated from 61% of all uterine lavage samples, 33% of all uterine swabs, and 22% of all uterine biopsies obtained from the 34 mares at six sampling time points during the four estrous cycles.
In total, biopsies from 12 of 34 mares (35%) had histologic evidence of acute endometritis by the presence of PMNs in stratum compactum and/or stratum spongiosum. Isolation of pathogens from uterine samples, independent of diagnostic test applied, was associated with the presence of PMNs in the endometrium (P ¼ 0.003). Six mares (18%) with negative cultures had PMNs in the endometrial sample, and two mares (6%) had positive cultures from endometrial samples but no presence of PMNs in the histologic evaluation (Table 3). 3.4. Sensitivity and specificity The diagnostic sensitivity and specificity after the culture of samples recovered using the double-guarded,
Table 3 Culture results of the biopsy, low-volume lavage, and swab in relation to the presence or absence of polymorphonuclear neutrophils on histology (gold standard) on samples obtained in cycle 1. Sampling technique Biopsy Cultþ Cult Sum Lavage Cultþ Cult Sum Swab Cultþ Cult Sum Biopsy Cytþ Cyt Sum Lavage Cytþ Cyt Sum Swab Cytþ Cyt Sum
Histologyþ (%)
Histology (%)
Total (%)
Sensitivity
Specificity
6 (18) 6 (18) 12 (35)
2 (6) 20 (59) 22 (65)
8 (24) 26 (76) 34 (100)
0.50
0.90
9 (26) 3 (9) 12 (35)
6 (18) 16 (47) 22 (64)
15 (44) 19 (56) 34 (100)
0.75
0.72
4 (12) 8 (24) 12 (35)
2 (6) 20 (59) 22 (65)
6 (18) 28 (82) 34 (100)
0.33
0.90
4 (12) 8 (24) 12 (35)
0 (0) 22 (65) 22 (65)
4 (12) 30 (88) 34 (100)
0.33
1.00
1 (3) 11 (32) 12 (35)
1 (3) 21 (62) 22 (65)
2 (6) 32 (94) 34 (100)
0.08
0.95
1 (3) 10 (29) 11 (32)
1 (3) 22 (65) 23 (68)
2 (6) 32 (94) 34 (100)
0.09
0.95
Diagnostic sensitivity and specificity of the low-volume lavage or endometrial swab samples when used for culture or cytology. Abbreviations: Cultþ, culture positive; Cult, culture negative; Cytþ, cytology positive; Cy, cytology negative.
226
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227
low-volume, lavage technique; endometrial swab; and biopsy were determined using the presence of PMNs on histology as gold standard (Table 3) [2]. A higher diagnostic sensitivity of bacterial growth was identified when the uterine lavage (0.75) was performed compared with the swab (0.33) and the biopsy (0.5), whereas the specificity of the uterine lavage was lower (Table 3). The sensitivity of cytology was greater when using the endometrial biopsy (0.33) compared with the uterine swab and low-volume lavage (0.09 and 0.08, respectively). 4. Discussion This is, to our knowledge, the first report on the use of a double-guarded, low-volume, lavage technique for the diagnosis of endometritis in the mare. Compared with the low-volume uterine lavage technique described by LeBlanc et al. [4], this modified guarded technique involves a sterile one-time use lavage tube, sterile sanitary sleeve, and a sterile steel speculum. The use of a disposable lavage tube and a closed fluid-tubing system is advantageous for field work and can be conducted by one person. In the present study, the sensitivity of bacterial growth from the double-guarded lavage technique was comparable to the sensitivity reported by LeBlanc et al. [4] using a nonguarded lavage technique (0.75 vs. 0.71) and was found to be superior compared with the uterine biopsy and doubleguarded swab also used in the present study (sensitivity 0.5 and 0.3, respectively). LeBlanc et al. [4] included an evaluation of the efflux fluid and the amount of debris on cytology to reduce the risk of a false-positive diagnosis when a low-volume lavage test was used. If a double-guarded approach is used, we expect that the risk of false-positive samples would decrease, which is clearly indicated by the high specificity observed in the present study (0.72) which is comparable to the specificity of bacterial growth when the nonguarded flush technique is used (0.71) [4]. The low sensitivity of bacterial growth of the endometrial biopsy in the present study compared with the results reported by Nielsen [2] (0.5 vs. 0.82) may result from very few b-strep isolated from the mares. The high sensitivity of bacterial growth from uterine biopsies in the study by Nielsen was primarily due to isolation of b-strep, which has been previously demonstrated to be able to reside deep within the endometrial tissue and, therefore, difficult to isolate from the endometrial surface [9]. E. coli was the most frequently isolated pathogen, which is in contrast to other studies in which b-strep were more frequently isolated from the uterus from broodmares followed by E. coli and other pathogens [2,4,6,10]. The low-volume lavage culture sensitivity observed in the present study was influenced by the increased detection of E. coli in the low-volume lavage fluid. This may be due to a sampling of a larger part of the endometrium when using a lavage technique compared with the double-guarded swab and the endometrial biopsy, in which only a small part of the endometrium is sampled, which was also described by LeBlanc et al. [4], using the low-volume flushing technique. The isolation of b-strep did not depend on the diagnostic test used; however, more b-strep bacteria were isolated from low-
volume lavage fluid, compared with the swab and the biopsy. These data emphasize, as demonstrated by others [2,4], the reduced sensitivity of endometrial swab culture and suggest the use of a low-volume lavage test for the diagnosis of endometritis [4]. The presence of PMNs in the subepithelial tissue of the endometrium evaluated in hematoxylin and eosin– stained histologic slide is a well-documented tool for diagnosing inflammation (with or without infection) in the endometrium [2,4,10]. Polymorphonuclear neutrophils in the subepithelial tissue were highly associated with isolation of pathogens from the diagnostic tests in the present study, which is in accordance with previous studies suggesting that the detection of PMNs in the stratum compactum and/or stratum spongiosum and endometrial epithelium is the best method of diagnosing uterine inflammation [2,11,12]. Only 7 of 21 positive cultures (33%; Table 2) had PMNs on cytology, resulting in very low sensitivities and specificities on cytology from all the diagnostic tests in the present study. No statistical significance was observed between the different diagnostic tests (swab, biopsy, and double-guarded lavage); however, positive cytology was observed from a greater number of smears from endometrial biopsies when compared with the low-volume lavage fluid and the swab. LeBlanc et al. [4] also reported that the low-volume lavage may not reflect the degree of inflammation, and the low cytology sensitivity on endometrial swabs has also previously been reported by Nielsen [2]. LeBlanc et al. [4] suggest that the centrifugation of the flush may disrupt cell walls leading to increased debris, which may make the cytologic slides from the low-volume lavage flushes more difficult to evaluate. Recent independent studies have demonstrated the presence of established pathogens in the uterus of the mare despite the absence of inflammation (PMNs) [6,10]. We recently demonstrated the presence of PMNs on histology in 50% of cytology-negative mares [12], illustrating the risk of a false-negative diagnosis when cytology is used. It is welldocumented that subclinical endometritis caused by E. coli is associated with negative cytology specimens compared with cytology from gram-positive pathogens [2,4,6]. It might be speculated that E. coli is not as chemotactic to attract PMNs into the uterine lumen as other bacteria; however, further studies are needed to investigate this hypothesis. The high number of E. coli positive samples with negative cytology may have resulted in “false” low sensitivities and specificities in the present study. It is therefore recommended to combine both uterine culture and cytology to increase diagnostic sensitivity and specificity when diagnosing endometritis in the mare. The use of a double-guarded, low-volume, lavage technique will allow the benefits of a sensitive test as demonstrated by LeBlanc et al. [4], and at the same time, it avoids the increased risk of contamination. The risk of contamination of the low-volume lavage flush when using the double-guarded lavage system is considered to be minimal. Because of this, the double-guarded approach is advantageous compared with the simple lavage technique used by LeBlanc et al. [4], who had to include an evaluation of flush efflux and debris on cytologic slides to identify false-positive samples.
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227
227
4.1. Conclusion
References
In summary, we demonstrated that isolation of pathogens using a double-guarded, low-volume, lavage technique is associated with PMNs in endometrial tissue, the accepted gold standard when diagnosing endometritis in the mare. The technique can be used in the field and was found to be an accurate and practical method for diagnosing endometritis in the mare based on sensitivity and specificity. The risk of falsepositive samples is considered to be comparable to other guarded techniques.
[1] Pycock J. Infertility in the mare. In: Noakes DE, Parkinson TJ, England GCW, editors. Arthur’s veterinary reproduction and obstetrics. Philadelphia: Saunders; 2001. p. 577–621. [2] Nielsen JM. Endometritis in the mare: a diagnostic study comparing cultures from swab and biopsy. Theriogenology 2005;64:510–8. [3] Ball BA, Shin SJ, Patten VH, Lein DH, Woods GL. Use of low-volume uterine flush for microbiologic and cytologic examination of the mare’s endometrium. Theriogenology 1988;29:1269–83. [4] LeBlanc MM, Magsig J, Stromberg AJ. Use of a low-volume uterine flush for diagnosing endometritis in chronically infertile mares. Theriogenology 2007;68:403–12. [5] Wingfield Digby N, Ricketts S. Results of concurrent bacteriological and cytological examinations of the endometrium of mares in routine stud farm practice. J Reprod Fertil Suppl 1982;32:181–5. [6] Riddle WT, LeBlanc MM, Stromberg AJ. Relationships between uterine culture, cytology and pregnancy rates in a thoroughbred practice. Theriogenology 2007;68:395–402. [7] Overbeck W, Witte TS, Heuwieser W. Comparison of three diagnostic methods to identify subclinical endometritis in mares. Theriogenology 2011;75:1311–8. [8] Ricketts SW, Alonso S. Assessment of the breeding prognosis of mares using paired endometrial biopsy techniques. Equine Vet J 1991;23:185–8. [9] Petersen MR, Nielsen JM, Lehn-Jensen H, Bojesen AM. Streptococcus equi subspecies zooepidemicus resides deep in the chronically infected endometrium of mares. Clin Theriogenol 2009;1:393–409. [10] Nielsen JM, Troedsson MH, Pedersen MR, Bojesen AM, LehnJensen H, Zent WW. Diagnosis of endometritis in the mare based on bacteriological and cytological examinations of the endometrium: comparison of results obtained by swabs and biopsies. J Vet Sci 2010;30:27–30. [11] Schoon HH, Schoon D, Klug E. Die endometriumbiopsie bei der stute in klinish-gynäkologishen kontext. Pferdeheilkunde 1997;5: 453–64. [12] Nielsen JM, Nielsen FH, Petersen MR. Diagnosis of equine endometritisdmicrobiology, cytology and histology of endometrial biopsies and the correlation to fertility. Pferdeheilkunde 2012;1:14.
Acknowledgments The authors would like to thank Kruuse A/S for donations of uterine lavage tubes. The authors would like to thank laboratory technician Anne Dorte Roed, Department of Large Animal Sciences, University of Copenhagen and Dr. Kirsten Scoggin at the Maxwell H. Gluck Equine Research Center, University of Kentucky for laboratory support. The authors would like to thank Mr. Lynn Ennis, Mr. Kevin Gallagher, and the University of Kentucky’s Maine Chance Farm for the management and care of the horses. Author contributions: M. Christoffersen participated in the design and coordination of the study, carried out the study, performed the statistical analyses, and wrote the article. L. Brandis and J. Samuelsson J. helped carrying out the clinical part of the study. M.R. Petersen, A.M. Bojesen and M.H.T. Troedsson participated in the study design and coordination. M.R. Petersen also helped carrying out parts of the clinical part of the study. All authors read and approved the final article.
Contents lists available at ScienceDirect
Theriogenology journal homepage: www.theriojournal.com
Diagnostic double-guarded low-volume uterine lavage in mares M. Christoffersen a, *, L. Brandis a, J. Samuelsson a, A.M. Bojesen b, M.H.T. Troedsson c, M.R. Petersen d a
Veterinary Reproduction and Obstetrics, Department of Large Animal Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark Veterinary Clinical Microbiology, Department of Veterinary Disease Biology, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark c Maxwell H. Gluck Equine Research Center, University of Kentucky, Lexington, Kentucky, USA d The Fertility Clinic, Section 4071, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 March 2014 Received in revised form 5 September 2014 Accepted 10 September 2014
Endometritis constitutes a major problem in the management of broodmares; hence, diagnostic tests with a high sensitivity and specificity are highly appreciated. The aim of this study was to compare the results from endometrial, cytologic, and bacteriologic examinations obtained by a newly developed, double-guarded, flushing technique versus standard diagnostic tests, the double-guarded swab and biopsy. The described doubleguarded flush technique requires the use of a disposable uterine flushing tube, a sanitary sleeve, a sterile steel speculum, and a 250 mL fluid bag. Endometrial biopsies, swabs, and low-volume lavage samples were obtained from 34 research mares at six different time points in four estrous cycles and were evaluated cytologically and bacteriologically. Endometrial biopsies from the first cycle (n ¼ 34) were examined for the presence of polymorphonuclear neutrophils (PMNs) in the stratum compactum and stratum spongiosum and used as a gold standard for calculation of diagnostic sensitivity and specificity. In all samples, Escherichia coli was most frequently isolated (lavage, 30%; swab, 21%; and biopsy, 12%) followed by b-hemolytic streptococci (lavage, 11%; swab, 8%; and biopsy, 7%). Positive cytology was less likely to occur when E coli was isolated from the diagnostic tests compared with the growth of b-hemolytic streptococci. Isolation of pathogens from uterine samples was highly associated with the presence of PMNs in the stratum compactum and straum spongiosum on histology. Using the presence of PMNs in the tissue specimens as the gold standard for diagnosing endometritis, the sensitivity of low-volume lavage culture was 0.75 and the specificity was 0.72. In conclusion, the double-guarded, low-volume, lavage technique was a rapid and accurate method for diagnosing mares with endometritis, and the risk of false-positive samples is considered to be minimal compared with other flushing techniques described. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Endometritis Mare Low-volume lavage Diagnostic test Escherichia coli b-hemolytic streptococci
1. Introduction
* Corresponding author. Tel.: þ45 35332983; fax: þ45 35332968 E-mail address: [email protected] (M. Christoffersen). 0093-691X/$ – see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.theriogenology.2014.09.008
Endometritis is one of the main reasons for reduced fertility in the mare [1]. Accurate diagnosis of endometritis and identification of pathogens involved are necessary to initiate correct treatment in time to optimize fertility and reduce the risk of bacterial resistance development.
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227
223
Fig. 1. (A) The steel speculum, sanitary sleeve, disposable uterine lavage tube, and fluid bag for the double-guarded, low-volume, lavage technique. (B) Doubleguarded lavage equipment ready for use. (For interpretation of the references to color in this Figure, the reader is referred to the web version of this article.)
Nielsen [2] was the first to evaluate diagnostic sensitivity and specificity in regards to diagnosis of endometritis in the mare. In this study, endometrial samples were collected from the same mares twice using a double-guarded swab and biopsy, which were used for bacterial culture and cytology. Histology was performed on the recovered endometrial biopsy, and the presence of polymorphonuclear neutrophils (PMNs) was used as “gold standard” when determining whether the diagnostic test result was true or false. The diagnostic sensitivity after culture from the commonly used endometrial swab or biopsy was 0.34 and 0.82, respectively. In other words, the risk of a false-negative diagnosis of endometritis was 66% or 18%, using culture from a swab or biopsy, respectively [2]. LeBlanc and co-workers evaluated the diagnostic sensitivity and specificity of a low-volume uterine lavage technique, initially suggested by Ball et al. [3,4]. In a comprehensive study using the same “gold standard” as Nielsen, 2005, sensitivity and specificity of culture and cytology from a low-volume uterine lavage were found to be 0.71 and 0.86, respectively, and 0.80 and 0.67, respectively [4]. Low-volume uterine lavage was found to be a sensitive diagnostic test, with sensitivity and specificity comparable to reports using an endometrial biopsy. These results suggested a marked reduction in the number of false-negative samples when a low-volume lavage is used compared with the double-guarded swab. Using the procedure described by LeBlanc et al. [4], the lavage tube is passed unguarded through the caudal reproductive tract and cervix, with a high risk of contamination from the commensal flora. To help identify false-positive samples, it was suggested to only categorize samples as positive if growth was obtained together with a cloudy efflux fluid and debris on cytology. Using this approach, the number of false-positive samples was estimated to be 11% [4]. Endometrial cytology is well correlated to bacteriologic finding [2,5,6], although subclinical endometritis caused by E. coli has been found to be associated with negative cytology specimens compared with cytology from gram-positive pathogens [2,4,7]. These findings emphasize the importance of using both cytology and bacteriologic findings when diagnosing endometritis in the mare. This study was conducted to determine if a newly developed, double-guarded flushing technique would improve the sensitivity of identifying endometritis in the mare compared with the double-guarded uterine swab and uterine biopsy techniques.
2. Materials and methods 2.1. Mares and sample collection A total of 34 light horse mares of mixed breeds of age 3 to 25 years with an unknown reproductive history were included in the study. All animal procedures were carried out in accordance to and with the approval of the Institutional Animal Care and Use Committee at the University of Kentucky. No abnormalities were noted on the clinical and gynecological examinations performed in the cycle before the experiment started. The mares were found normal on gynecological examination, and a double-guarded, lowvolume, uterine lavage, endometrial biopsy and a uterine swab were obtained during estrous (>30 mm follicle, endometrial edema, and decreased uterine tone). Uterine samples were obtained from each mare six times across four estrus periods, and all uterine samples were obtained before ovulation. The mares were restrained in an examination stock. After rectal palpation and transrectal ultrasonographic examination, the vulva and perineum were washed with chlorhexidine medical scrub (Dermachlor, Butler Animal Supply, Lexington, KY, USA), rinsed three times with water, and dried with paper towels. All three samples, the double-guarded swab, the low-volume double-guarded lavage, and the endometrial biopsy, were obtained from all mares at each sample time point and in the same order. A double-guarded swab (EQUIVET; Kruuse A/S, Langeskov, Denmark) was used to obtain an endometrial culture. The swab was kept in contact with the endometrial surface for at least 30 seconds. To obtain the double-guarded, low-volume, lavage sample, a sterile steel speculum (EQUIVET) covered with a sterile sanitary sleeve (also termed “chemise” and used to cover an embryo transfer instrument; Kruuse A/S; Fig. 1) was passed manually per vaginam into the cervical canal. When placed in the most cranial part of the cervix, the sanitary sleeve was pulled toward the examiner from the outside and the sterile speculum pushed through the sanitary sleeve, through the cervical canal and into the uterine lumen. A disposable uterine lavage tube (EQUIVET) was attached to a 250 mL fluid bag (Ringers Lactate; Butler Animal Supply) and passed through the speculum and into the uterus, followed by retraction of the speculum toward the examiner and out of the mares’ genital tract. Using this guarded approach, the lavage tube was placed in a sterile manner in the uterine body. The tip of the flushing tube was kept in the caudal aspect of the uterine body by a firm hold around
224
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227
Table 1 Distribution of culture results using low-volume lavage (LVL), swab, and biopsy as diagnostic test at six sampling time points in four estrous cycles. Culture results
LVL, n (%)
Swab, n (%)
Biopsy, n (%)
Culture positive Contamination Sterile E. coli b-strep E. coli þ b-strep Other pathogens Total
122 (61)a 7 (4) 70 (35)a 60 (30)a 21 (11) 39 (20)a 2 (1) 199
71 (36)b 4 (2) 124 (62)b 42 (21)b 16 (8) 8 (4)b 5 (3) 199
51 (26)c 5 (3) 143 (71)b 24 (12)c 15 (7) 5 (3)b 7 (3) 199
Abbreviations: b-strep, beta-hemolytic streptococci; E. Coli, Escherichia coli. a,b,c Different letters indicate significant difference (P < 0.05) between the diagnostic tests.
the vaginal part of the cervix. The fluid was infused into the uterus, and the efflux fluid was recovered by gravity flow from the uterus and into the fluid bag in a closed system. A minimum of 150 mL of the efflux fluid was recovered. In some cases, the lavage tube was rotated, advanced, or retracted a few centimeters to avoid suction to the endometrium and to allow fluid contact. If less than 150 mL of fluid was recovered, the uterus was manipulated by rectal palpation. One hand of the examiner maintained the lavage tube in position per vaginam, whereas the other hand was used to manipulate the fluid toward the lavage tube by rectal palpation. Endometrial biopsies were collected using an alligator jaw biopsy punch introduced into the uterus through a sterile speculum (EQUIVET) as described by Nielsen [2]. 2.2. Microbiology The recovered fluid bag was sealed with a pair of forceps and transported to the laboratory and within 4 hours and hung for sedimentation by gravity for minimum 1 hour. A total of 50 mL was aspirated with a sterile needle from the most ventral part of the bag (including sediment) into a 50mL tube (Falcon; Fisher Scientific, Hampton, NH, USA) and centrifuged at 400g for 10 minutes. Of the 50 mL, 48 mL of the supernatant fluid was poured off. The sediment was suspended in the remaining 2 mL. For bacterial culture, a sterile cotton swab was dipped in the 2-mL sample and smeared on a blood agar (5% horse blood). Endometrial biopsies and uterine swabs were streaked on blood agars. All blood agar plates were cultured aerobically for 24 hours at 37 C. Bacterial growth was identified according to colony morphology, Hancock stain morphology, hemolysis, catalase, and potassium hydroxide (3% KOH) tests. Colonies were counted on blood agar plates and scored: no growth or sterile: 4 or lesser colony forming units (CFU) or positive growth: greater than 4 CFU, and culture results were recorded as E. coli, beta-hemolytic streptococci (b-strep), E. coli þ b-strep, or other pathogens. If more than three different bacteria were present at the blood agar plate, the result was recorded as contamination.
slides, which were dried at room temperature and stained with a quick staining method Diff-Quik (Fisher Scientific) and evaluated by light microscopy (400 magnification). Cytologic classification was based on numbers of PMNs present per 200 cells examined. The sample was considered positive for inflammation when greater than 0.5% PMNs were present as described by Nielsen [2]. 2.4. Histology The endometrial biopsies obtained in the first cycle were used for histopathologic examination. Endometrial biopsies were fixed in 10% formalin, embedded in paraffin and sectioned at 5 mm, and stained with hematoxylin and eosin. Slides were examined for the presence of PMNs within the luminal epithelium, stratum compactum, and stratum spongiosum. Infiltration of more than three PMNs per five fields of high magnification (400) was considered as evidence of acute endometritis [8]. 2.5. Statistical analyses Categorical variables were compared using chi-square test, and significance was set at P < 0.05. Sensitivity and specificity were calculated using the presence of PMNs on endometrial histology from biopsies obtained in the first cycle as the gold standard for predicting endometritis. Sensitivity was calculated as the proportion of mares with PMNs in the endometrium, and a positive culture result from the compared tests collected at the same time. Specificity was calculated as the proportion of mares without PMNs in the endometrial tissue and a negative result from the compared test [2]. To test the effect of the different diagnostic tests and cycle number on a dichotomous outcome and evaluation of the effect, the presence of different pathogens was analyzed by logistic regression. A logit transformation of data was used to describe the relationship between the outcome and the explanatory variable. Generalized score test (the Wald test) was used in the type 3 analysis. Significant differences in the outcome of the explanatory variable were identified by using the syntax for estimates. All analyses were performed using SAS 9.3 (SAS Institute, Cary, NC, USA). The level of significance was set to P < 0.05. Graphs were made using the software GraphPad Prism 5.0 (GraphPad software Inc., La Jolla, CA, USA). 3. Results In total, 199 uterine swabs, 199 uterine lavage samples, and 199 biopsies were obtained from the 34 mares at six different sampling time points in four estrous cycles. Samples obtained from cycle 1 (34 biopsies, 34 low-volume lavages, and 34 swabs) were used for calculating the sensitivity, specificity, and positive and negative predictive values for the three diagnostic tests evaluated using PMNs on the histologic slides as gold standard.
2.3. Cytology
3.1. Microbiology
The biopsies, swabs, and sediment from the low-volume lavage samples from the first cycle were smeared on glass
From all samples, bacteria were isolated from 122 of 199 uterine lavages (61%), 71 of 199 swabs (36%), and
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227 Table 2 Presence of polymorphonuclear neutrophils on cytologic slides related to samples with no growth of pathogens (sterile) and samples with growth of E. coli, b-strep, and other pathogens (OP) in cycle 1. Culture results
Swab, n (%)
LVL, n (%)
Biopsy, n (%)
Cyt
Cytþ
Cyt
Cytþ
Cyt
Cytþ
Sterile E. coli b-strep OP Total
26 (76) 5 (15)a 1 (3)b 0 32 (94)
0 0 2 (6) 0 2 (6)
20 (59) 13 (38)a 0a 0 33 (97)
0 0 1 (3) 0 1 (3)
27 (79) 1 (3)a 1 (3)a 0 29 (85)
1 3 1 0 5
(3) (9) (3) (15)
Abbreviations: b-strep, beta-hemolytic streptococci; Cytþ, cytology positive; Cyt, cytology negative; E. coli, Escherichia coli; LVL, low-volume lavage. a,b Different letters indicate significant difference (P < 0.05) of cytologic response to E. coli and b-strep within each diagnostic test.
225
3.2. Cytology The presence of PMNs on cytologic slides obtained from the three diagnostic tests in cycle 1 was not associated with the positive culture of uterine pathogens (Table 2). In total, PMNs were only observed in 8 of 102 (8%) of the cytologic specimens from the 34 biopsies, 34 lavages, and 34 swabs (cycle 1). One mare had positive cytology (biopsy) with no growth of uterine pathogens on any of the uterine samples obtained. The number of cytology-negative but E. coli positive samples was greater compared with cytology-negative samples culture positive for b-strep (P ¼ 0.02) when using the double-guarded swab. 3.3. Histology
51 of 199 biopsies (26%; Table 1). Seven of 199 (4%) uterine lavage samples, 4 of 199 (2%) uterine swabs, and 5 of 199 (3%) biopsies were registered as contaminated. Escherichia coli was the most frequently isolated pathogen from low-volume lavage (30%), swabs (21%), and biopsy specimens (12%), followed by b-strep, which were cultured from low-volume lavage samples (11%), swabs (8%), and biopsies (7%). E. coli and b-strep were isolated in mixed culture from 20% of the low-volume lavage samples, 4% of the swabs, and 3% of the biopsy specimens. E. coli and b-strep were isolated from 61% of all uterine lavage samples, 33% of all uterine swabs, and 22% of all uterine biopsies obtained from the 34 mares at six sampling time points during the four estrous cycles.
In total, biopsies from 12 of 34 mares (35%) had histologic evidence of acute endometritis by the presence of PMNs in stratum compactum and/or stratum spongiosum. Isolation of pathogens from uterine samples, independent of diagnostic test applied, was associated with the presence of PMNs in the endometrium (P ¼ 0.003). Six mares (18%) with negative cultures had PMNs in the endometrial sample, and two mares (6%) had positive cultures from endometrial samples but no presence of PMNs in the histologic evaluation (Table 3). 3.4. Sensitivity and specificity The diagnostic sensitivity and specificity after the culture of samples recovered using the double-guarded,
Table 3 Culture results of the biopsy, low-volume lavage, and swab in relation to the presence or absence of polymorphonuclear neutrophils on histology (gold standard) on samples obtained in cycle 1. Sampling technique Biopsy Cultþ Cult Sum Lavage Cultþ Cult Sum Swab Cultþ Cult Sum Biopsy Cytþ Cyt Sum Lavage Cytþ Cyt Sum Swab Cytþ Cyt Sum
Histologyþ (%)
Histology (%)
Total (%)
Sensitivity
Specificity
6 (18) 6 (18) 12 (35)
2 (6) 20 (59) 22 (65)
8 (24) 26 (76) 34 (100)
0.50
0.90
9 (26) 3 (9) 12 (35)
6 (18) 16 (47) 22 (64)
15 (44) 19 (56) 34 (100)
0.75
0.72
4 (12) 8 (24) 12 (35)
2 (6) 20 (59) 22 (65)
6 (18) 28 (82) 34 (100)
0.33
0.90
4 (12) 8 (24) 12 (35)
0 (0) 22 (65) 22 (65)
4 (12) 30 (88) 34 (100)
0.33
1.00
1 (3) 11 (32) 12 (35)
1 (3) 21 (62) 22 (65)
2 (6) 32 (94) 34 (100)
0.08
0.95
1 (3) 10 (29) 11 (32)
1 (3) 22 (65) 23 (68)
2 (6) 32 (94) 34 (100)
0.09
0.95
Diagnostic sensitivity and specificity of the low-volume lavage or endometrial swab samples when used for culture or cytology. Abbreviations: Cultþ, culture positive; Cult, culture negative; Cytþ, cytology positive; Cy, cytology negative.
226
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227
low-volume, lavage technique; endometrial swab; and biopsy were determined using the presence of PMNs on histology as gold standard (Table 3) [2]. A higher diagnostic sensitivity of bacterial growth was identified when the uterine lavage (0.75) was performed compared with the swab (0.33) and the biopsy (0.5), whereas the specificity of the uterine lavage was lower (Table 3). The sensitivity of cytology was greater when using the endometrial biopsy (0.33) compared with the uterine swab and low-volume lavage (0.09 and 0.08, respectively). 4. Discussion This is, to our knowledge, the first report on the use of a double-guarded, low-volume, lavage technique for the diagnosis of endometritis in the mare. Compared with the low-volume uterine lavage technique described by LeBlanc et al. [4], this modified guarded technique involves a sterile one-time use lavage tube, sterile sanitary sleeve, and a sterile steel speculum. The use of a disposable lavage tube and a closed fluid-tubing system is advantageous for field work and can be conducted by one person. In the present study, the sensitivity of bacterial growth from the double-guarded lavage technique was comparable to the sensitivity reported by LeBlanc et al. [4] using a nonguarded lavage technique (0.75 vs. 0.71) and was found to be superior compared with the uterine biopsy and doubleguarded swab also used in the present study (sensitivity 0.5 and 0.3, respectively). LeBlanc et al. [4] included an evaluation of the efflux fluid and the amount of debris on cytology to reduce the risk of a false-positive diagnosis when a low-volume lavage test was used. If a double-guarded approach is used, we expect that the risk of false-positive samples would decrease, which is clearly indicated by the high specificity observed in the present study (0.72) which is comparable to the specificity of bacterial growth when the nonguarded flush technique is used (0.71) [4]. The low sensitivity of bacterial growth of the endometrial biopsy in the present study compared with the results reported by Nielsen [2] (0.5 vs. 0.82) may result from very few b-strep isolated from the mares. The high sensitivity of bacterial growth from uterine biopsies in the study by Nielsen was primarily due to isolation of b-strep, which has been previously demonstrated to be able to reside deep within the endometrial tissue and, therefore, difficult to isolate from the endometrial surface [9]. E. coli was the most frequently isolated pathogen, which is in contrast to other studies in which b-strep were more frequently isolated from the uterus from broodmares followed by E. coli and other pathogens [2,4,6,10]. The low-volume lavage culture sensitivity observed in the present study was influenced by the increased detection of E. coli in the low-volume lavage fluid. This may be due to a sampling of a larger part of the endometrium when using a lavage technique compared with the double-guarded swab and the endometrial biopsy, in which only a small part of the endometrium is sampled, which was also described by LeBlanc et al. [4], using the low-volume flushing technique. The isolation of b-strep did not depend on the diagnostic test used; however, more b-strep bacteria were isolated from low-
volume lavage fluid, compared with the swab and the biopsy. These data emphasize, as demonstrated by others [2,4], the reduced sensitivity of endometrial swab culture and suggest the use of a low-volume lavage test for the diagnosis of endometritis [4]. The presence of PMNs in the subepithelial tissue of the endometrium evaluated in hematoxylin and eosin– stained histologic slide is a well-documented tool for diagnosing inflammation (with or without infection) in the endometrium [2,4,10]. Polymorphonuclear neutrophils in the subepithelial tissue were highly associated with isolation of pathogens from the diagnostic tests in the present study, which is in accordance with previous studies suggesting that the detection of PMNs in the stratum compactum and/or stratum spongiosum and endometrial epithelium is the best method of diagnosing uterine inflammation [2,11,12]. Only 7 of 21 positive cultures (33%; Table 2) had PMNs on cytology, resulting in very low sensitivities and specificities on cytology from all the diagnostic tests in the present study. No statistical significance was observed between the different diagnostic tests (swab, biopsy, and double-guarded lavage); however, positive cytology was observed from a greater number of smears from endometrial biopsies when compared with the low-volume lavage fluid and the swab. LeBlanc et al. [4] also reported that the low-volume lavage may not reflect the degree of inflammation, and the low cytology sensitivity on endometrial swabs has also previously been reported by Nielsen [2]. LeBlanc et al. [4] suggest that the centrifugation of the flush may disrupt cell walls leading to increased debris, which may make the cytologic slides from the low-volume lavage flushes more difficult to evaluate. Recent independent studies have demonstrated the presence of established pathogens in the uterus of the mare despite the absence of inflammation (PMNs) [6,10]. We recently demonstrated the presence of PMNs on histology in 50% of cytology-negative mares [12], illustrating the risk of a false-negative diagnosis when cytology is used. It is welldocumented that subclinical endometritis caused by E. coli is associated with negative cytology specimens compared with cytology from gram-positive pathogens [2,4,6]. It might be speculated that E. coli is not as chemotactic to attract PMNs into the uterine lumen as other bacteria; however, further studies are needed to investigate this hypothesis. The high number of E. coli positive samples with negative cytology may have resulted in “false” low sensitivities and specificities in the present study. It is therefore recommended to combine both uterine culture and cytology to increase diagnostic sensitivity and specificity when diagnosing endometritis in the mare. The use of a double-guarded, low-volume, lavage technique will allow the benefits of a sensitive test as demonstrated by LeBlanc et al. [4], and at the same time, it avoids the increased risk of contamination. The risk of contamination of the low-volume lavage flush when using the double-guarded lavage system is considered to be minimal. Because of this, the double-guarded approach is advantageous compared with the simple lavage technique used by LeBlanc et al. [4], who had to include an evaluation of flush efflux and debris on cytologic slides to identify false-positive samples.
M. Christoffersen et al. / Theriogenology 83 (2015) 222–227
227
4.1. Conclusion
References
In summary, we demonstrated that isolation of pathogens using a double-guarded, low-volume, lavage technique is associated with PMNs in endometrial tissue, the accepted gold standard when diagnosing endometritis in the mare. The technique can be used in the field and was found to be an accurate and practical method for diagnosing endometritis in the mare based on sensitivity and specificity. The risk of falsepositive samples is considered to be comparable to other guarded techniques.
[1] Pycock J. Infertility in the mare. In: Noakes DE, Parkinson TJ, England GCW, editors. Arthur’s veterinary reproduction and obstetrics. Philadelphia: Saunders; 2001. p. 577–621. [2] Nielsen JM. Endometritis in the mare: a diagnostic study comparing cultures from swab and biopsy. Theriogenology 2005;64:510–8. [3] Ball BA, Shin SJ, Patten VH, Lein DH, Woods GL. Use of low-volume uterine flush for microbiologic and cytologic examination of the mare’s endometrium. Theriogenology 1988;29:1269–83. [4] LeBlanc MM, Magsig J, Stromberg AJ. Use of a low-volume uterine flush for diagnosing endometritis in chronically infertile mares. Theriogenology 2007;68:403–12. [5] Wingfield Digby N, Ricketts S. Results of concurrent bacteriological and cytological examinations of the endometrium of mares in routine stud farm practice. J Reprod Fertil Suppl 1982;32:181–5. [6] Riddle WT, LeBlanc MM, Stromberg AJ. Relationships between uterine culture, cytology and pregnancy rates in a thoroughbred practice. Theriogenology 2007;68:395–402. [7] Overbeck W, Witte TS, Heuwieser W. Comparison of three diagnostic methods to identify subclinical endometritis in mares. Theriogenology 2011;75:1311–8. [8] Ricketts SW, Alonso S. Assessment of the breeding prognosis of mares using paired endometrial biopsy techniques. Equine Vet J 1991;23:185–8. [9] Petersen MR, Nielsen JM, Lehn-Jensen H, Bojesen AM. Streptococcus equi subspecies zooepidemicus resides deep in the chronically infected endometrium of mares. Clin Theriogenol 2009;1:393–409. [10] Nielsen JM, Troedsson MH, Pedersen MR, Bojesen AM, LehnJensen H, Zent WW. Diagnosis of endometritis in the mare based on bacteriological and cytological examinations of the endometrium: comparison of results obtained by swabs and biopsies. J Vet Sci 2010;30:27–30. [11] Schoon HH, Schoon D, Klug E. Die endometriumbiopsie bei der stute in klinish-gynäkologishen kontext. Pferdeheilkunde 1997;5: 453–64. [12] Nielsen JM, Nielsen FH, Petersen MR. Diagnosis of equine endometritisdmicrobiology, cytology and histology of endometrial biopsies and the correlation to fertility. Pferdeheilkunde 2012;1:14.
Acknowledgments The authors would like to thank Kruuse A/S for donations of uterine lavage tubes. The authors would like to thank laboratory technician Anne Dorte Roed, Department of Large Animal Sciences, University of Copenhagen and Dr. Kirsten Scoggin at the Maxwell H. Gluck Equine Research Center, University of Kentucky for laboratory support. The authors would like to thank Mr. Lynn Ennis, Mr. Kevin Gallagher, and the University of Kentucky’s Maine Chance Farm for the management and care of the horses. Author contributions: M. Christoffersen participated in the design and coordination of the study, carried out the study, performed the statistical analyses, and wrote the article. L. Brandis and J. Samuelsson J. helped carrying out the clinical part of the study. M.R. Petersen, A.M. Bojesen and M.H.T. Troedsson participated in the study design and coordination. M.R. Petersen also helped carrying out parts of the clinical part of the study. All authors read and approved the final article.
