A novel cytologic sampling technique to diagnose subclinical endometritis and comparison of staining methods for endometrial cytology samples in dairy cows

1. Introduction.

Subclinical endometritis (SCE) is one of the most important reproductive impairments in dairy cows studied in the last decade. It can be defined as the superficial inflammation of the endometrium (no deeper than the stratum spongiosum) ^[1], without visible clinical signs, but significantly affecting reproductive performances^[2,3]. Cytology is considered the best technique to diagnose SCE due to its feasibility and fair reliability ^[2,3]. Consequently, a practical definition for SCE was established by consensus in 2006, stating that SCE is diagnosed when in endometrial cytology, samples taken between 21 and 33 days postpartum (DPP) greater than 18% of harvested cells are identified as polymorphonuclear cells (PMNs), or greater than 10% of the cells are PMNs when samples are taken at 34 to 47 DPP, in the absence of clinical endometritis^[4]. Although endometrial cytology is actually considered the best technique to diagnose SCE^[5], the main problem lies in the vast variety of cutoff values to differentiate affected versus unaffected cows taking into account the days after calving samples are taken ^[6,7].

Cytology samples to diagnose SCE are mainly obtained either by cytobrush (CB) ^[2] or low-volume uterine lavage (LVL) ^[3], either of them providing similar results ^[8,9]. In equine medicine, sampling by cotton swab (CS) is also considered a valid technique to diagnose endometritis ^[10]. However, there seems not to be an ideal cytology technique; each method has specific advantages and disadvantages ^[11]. Yet, CB is recommended as the technique of choice because of its feasibility, safety, and reasonably high-quality samples^[8,9]. The ideal cytology technique should represent an equilibrium between practicability and the possibility to yield highly accurate results. A harmless technique yielding a high number of well-preserved cells is indispensable for reliable cytologic results^[12].

Subclinical endometritis commonly is diagnosed during the voluntary waiting period before insemination, usually from 21 to 64 DPP^[2,3,8,13]. However, diagnosing SCE at that time has two major disadvantages: (1) often, sampling interferes with routine management at the dairy farm (provoking extra handling of animals and hence extra labor); and (2) the percentage of PMNs present in the uterine lumen is a dynamic phenomenon ^[14], which makes it hard to predict the amount of PMNs that will be actually present at the moment the animal is inseminated, creating parameters for SCE diagnosis relatively erratic. Consequently, the prevalence of SCE fluctuates widely among different studies, mainly because of the large variation in both the DPP when samples are taken and the cutoff value applied to define an animal as SCE positive ^[7,15,16]. Therefore, it is imperious to take cytology samples at a standard moment that is both more convenient in relation to the general herd management and allows the use of a universal PMNs threshold. Sampling during artificial insemination (AI) may be an ideal proposal in this perspective because it does not require extra manipulation of the animals and offers opportunities to determine the uterine health status at the moment the animal is fertilized.

After cytologic sampling, smears need to be air-dried, stained, and microscopically evaluated. The staining method should be fast, easy to perform under practical circumstances, and yield high-quality samples allowing accurate interpretation ^[17,18]. According to the literature, in more than 90% of the studies, a modifiedWright-Giemsa staining, such as Diff-Quik, Tincion 15, Haema Quick, Hemacolor, is used ^[3,19–21]. Because these stainings are fast and easy to perform, they are well accepted and widely used, although they have not been compared with a gold standard to evaluate more objectively the PMNsto- epithelial cells ratio. Naphthol-AS-D-chloroacetateesterase (CIAE) is an enzyme histochemical method in which PMNs appear bright red after staining and is therefore regarded as the preferable staining to identify and count PMNs^[22,23]. Because a relatively small number of PMNs is needed to consider a cow positive for SCE, it is essential to have a good quality staining not to miss any PMN and hence sub-evaluate the number of cows suffering from SCE.

Hence, the following were the aims of the present study:

To compare and validate an innovative technique to take endometrial cytology samples using the generally accepted CB as the gold standard;
To compare the PMNs percentage in twin endometrial cytology samples stained with the widely applied Diff-Quik versus the CIAE, a histochemical preparation considered as the gold standard to stain PMNs.

2. Materials and methods.

All experiments mentioned in the present study were carried out with permission of the Ethical Committee of the Faculty of Veterinary Medicine of Ghent University (EC 2013/174).

2.1. Study design.

The study population consisted of 204 Holstein-Friesian cows, 140 from one commercial dairy herd, and 64 that were selected at the slaughterhouse before slaughtering. Cows chosen at the slaughterhouse were of unknown origin and health status and thus no data concerning health and previous reproductive status were available. Cows from the commercial farm were between 31 and 37 DPP at the moment of sampling (Fig.1).

2.2. Examinations and sampling.

Only cows with a body condition score greater than 2 (1–5) ^[24] were enrolled in the study. Vaginal examination was performed by the gloved hand method^[25]. The presence of purulent vaginal discharge (PVD) was not used as an exclusion criterion. Vaginal discharge (VD) was classified as: clear mucus (VD-0), mucus with pus flecks (VD-1), mucopurulent discharge (VD-2), and purulent (or fetid) discharge (VD-3) ^[26]. A transrectal reproductive ultrasound (Tringa, Esaote-Pie Medical, Maastricht, the Netherlands) examination (uterus and ovaries) was performed before sampling. Reproductive ultrasound findings were classified as presence (>0.5 cm) or absence of fluid in the uterine lumen^[2] and presence (Corpus luteum (CL), >2 cm) or absence of corpus luteum ^[27] at the moment of examination.

A special device was developed to take two endometrial cytology samples at the same time. A human use Cytobrush Plus GT (CooperSurgical, Berlin, Germany) was adapted to a stainless steel stylet of an universal insemination gun (Agtech, Manhattan, KS, USA), by heating the top of the stylet with a lighter and fitting it to the base of the handle of the CB. Then, the stylet-CB was introduced in a 2200-long equine infusion pipette individually wrapped (Agtech, Manhattan, KS, USA). Next, a 1.5-cm clean piece of paper tape (Tesa 4322; Hamburg, Germany) was rolled on the top of the equine infusion pipette. To protect the equine infusion pipette (tape and CB) from contact with both the vaginal and cervical wall during sampling, the pipette was covered with a 1200-long Sani-Shield Rod (Agtech, Manhattan, KS, USA).

Once the sampling pipette was armed, the perineal region of the cow was thoroughly cleaned with fresh wáter and dried with paper towel. Next, under rectal guidance, the pipette was introduced into the vagina and manipulated through the cervix. Once in the lumen of the uterine body, the top of the pipette (with the paper tape on it) was released from the Sani-Shield Rod. Then, it was rolled on the dorsalwall of the uterine body with a gentle pressure of the index finger through the rectum. After the pipettepaper tape (cytotape or CT) was rotated twice, the CB was released from the pipette (Fig. 2) and also rolled twice in the dorsal part of the uterine body (just next to the place where CT had been rolled). Once the CB was rolled twice, it was retracted into the pipette, and then the pipette was covered again with the Sani-Shield Rod to prevent contamination with cervical and vaginal cells. Finally, the device was carefully removed from the reproductive tract.

2.3. Preparation and staining of the slides.

Slides for cytologic examination were prepared at the farm and slaughterhouse immediately after sampling. First, the CT samples were gently rolled onto a clean microscope glass slide (Marienfeld, Lauda-Königshofen, Germany) to spread the collected cellular material. After that, two microscope slides were prepared by rolling half of the CB circumference on one slide (first duplicate) and the other half on another slide (second duplicate), in this way, obtaining a suitable and equally distributed quantity of cellular material on both slides. Before staining, smearswere air-dried. In total, 408 slides including 204 from CT and 204 from the CB first duplicate stained with Diff-Quik (Fisher Diagnostics, Newark, DE, USA), according to the manufacturer’s instructions. A subset (n ¼ 114) of the second duplicates of the CB sampleswas randomly selected to be stained with the CIAE method, for the second experiment. In the end, 522 slides were stained, 204 CT Diff-Quik (CT-DQ), 204 CB Diff-Quik (CB-DQ), and 114 CB CIAE (CB-CIAE) (Fig. 1).

A novel cytologic sampling technique to diagnose subclinical endometritis and comparison of staining methods for endometrial cytology samples in dairy cows - Image 1

Fig. 1. Brief overview of the study protocol for both experiments. CB-DQ, CB Diff-Quik; CT-DQ, CT Diff-Quik; PMN, polymorphonuclear cell.

Fig. 2. Image showing how the cytology samples were taken (from top to bottom). First, once the cytology device was manipulated through the cervix and reached the uterine lumen, the cytotape (CT) was released from the Sani-Shield Rod. After rotating twice, the cytobrush was released and rotated at approximately the same location where the CT sample had been taken. In this way, two cytology samples were taken at the same time.

For the CIAE staining, two stock solutions were prepared: a substrate solution and a hexasodium solution. For the substrate solution, 3.58 mg of naphthol-AS-Dchloroacetate (Sigma, ref. no. N0758, St. Louis, USA) was diluted in 0.9 mL of dimethyl sulfoxide (Sigma, ref. no. D5879, St. Louis, USA) and 0.1 mL Triton X-100 (Sigma, ref. no. X100, St. Louis, USA), becoming a light yellow solution. For preparing the hexasodium solution, first, a sodium nitrite solution 1 mol/L was prepared by diluting 345 mg of sodium nitrite (Carl-Roth, ref. no. 8604.1, Karlsruhe, Germany) in 5 mL of distilled water. Once the sodium nitrite solution was prepared, pararosaniline hydrochloride (Sigma, ref. no. P3750, St. Louis, USA) was diluted in 3mL of 1 mol/L HCl (Chem-lab, ref. no. CL05.0311.1000, Zedelgem, Belgium), acquiring a dark brown color. In the end, 0.5 mL of nitrite solution was added to the pararosaniline-HCl solution, turning this into a light brown color. Both the substrate and the hexasodium solution had to rest for 5 minutes before use to reach stabilization. To prepare the final solution, 1 mL of substrate solution and 0.5 mL of hexasodium solution were added to 100 mL of a phosphate-buffered saline solution (pH ¼ 7.2), on a vortex, until a light pink color appeared. Next, slides were incubated for 90 minutes at 37 _C in the CIAE solution. After incubation, slides were rinsed for 2 minutes with tap wáter and 5 minutes with distilled water. For the counterstaining, slides were submerged in a hemaluin Gill staining solution for 7 minutes.

Finally, they were rinsed with tap water for 10 minutes and distilled water for 5 minutes.

Once dry, cover slips (Marienfeld, Lauda-Königshofen, Germany) were mounted on the stained glass slides using Eukitt (O.Kindler GmbH, Freiburg, Germany) as the mounting medium.

Table 1. Concordance correlation coefficient (CCC) values achieved between both sampling techniques and staining methods.
A novel cytologic sampling technique to diagnose subclinical endometritis and comparison of staining methods for endometrial cytology samples in dairy cows - Image 3

Also included: standard error (SE) and variance components.

Table 2. Quality, total cellularity, and RBCs contamination in samples taken by cytobrush (CB) and cytotape (CT).

2.4. Cytologic evaluation.

All slides (CB-DQ, CT-DQ, and CB-CIAE) were evaluated by light microscopy (Kyowa Optical, Tokyo, Japan) using _ 100 and _ 400 magnifications. For each slide, a total of 300 cells were counted by one observer, and the PMN-epithelial cell ratio was assessed [20]. In the CT-DQ and CB-DQ slides also, the total cellularity, quality of the harvested cells, and the background content were assessed.

Total cellularity, quality, and red blood cell (RBC) contamination were assessed after evaluating 10 high power fields at _ 100 magnification ^[10,11,28] and results averaged. To evaluate total cellularity, the number of cells was estimated and classified as no cells, low (<50 cells), moderate (50–100 cells), and high (>100 cells). Quality was evaluated by estimating the percentage of intact cells leading to a categorization in three different groups: very good (>75% intact cells), good (50%–75% intact cells), or bad (<50% intact cells). Finally, the RBC contamination was evaluated by assessing the quantity of erythrocytes and subsequent classification as no RBCs, low (disperse erythrocytes), moderate (high number of erythrocytes), and high RBCs (strong background of erythrocytes).

2.5. Statistical analysis.

Comparison between CB and CT was made for PMNs %, total cellularity, quality, and RBC contamination. Data from all sampled cowswere exported from data capture forms to an Excel (Microsoft Corporation, Seattle, USA) spreadsheet file. Statistical analyses were carried out using SAS software (SAS Institute Inc., Cary, USA). First, descriptive statistical analyses were executed using PROC FREQ and PROC MEANS. To assess agreement among continuous variables (PMNs %) between diagnostic methods, CT versus CB as gold standard, the concordance correlation coefficient (CCC) test was performed using the SAS macro reported by Crawford et al. ^[29]. Briefly, for CCC interpretation, values are between _1 and 1, in which _1 means complete disagreement, 0 translates to an independent situation, and 1 indicates a perfect agreement^[30]. The CCCs are reported with a 95% confidence interval. Enhanced Bland-Altman plots were created to visualize agreement between methods. Comparisons between categorical variables were made using contingency tables and Pearson chi-square tests. The level of significance was set at P value less than 0.05 ^[31]. To compare the agreement of PMNs % between Diff-Quik versus CIAE staining, the SAS macro CCC test was used, and an enhanced Bland-Altman plot was made to picture the agreement between both staining techniques.

3. Results.

Two hundred four Holstein-Friesian cows were included in the present study. In total, 522 slides were analyzed all of which were readable and acceptable for further analysis. In cows sampled at the slaughterhouse (n ¼ 64), 10 cows (15.6%) presented PVD at the moment of examination: VD-0, 54 of 64 (84.4%); VD-1, 8 of 64 (12.5%); VD-2, 2 of 64 (3.1%); and VD-3, 0 of 64 (0%). Eight of the 64 cows (12.5%) showed uterine content, and 41 (64%) had a CL (>2 cm) on at least one of the ovaries. From the cows sampled at the farm (n ¼ 140), 95 of 140 (67.8%) showed VD-0, 28 of 140 (20%) VD-1, 17 of 140 (12.1%) VD-2, whereas no cows were found with VD-3. Sixteen of the 140 cows (11.4%) had uterine content (>0.5 cm), whereas 77 of them (55%) presented a CL (>2 cm) at the moment of examination. In general, from the 204 cows, 149 showed no VD (73%, no PVD), 24 (17.7%) presented uterine content (>0.5 cm) and 118 (57.8%) presented a CL (>2 cm).

The CCC between CT-DQ and CB-DQ concerning the percentage of PMNs is summarized in Table 1. However, three outliers were noticed in the enhanced Bland-Altman Plot (Fig. 3). Re-checking the data revealed that the strong differences between CT-DQ and CB-DQ PMNs % in these three samples were probably due to contamination during cervical manipulation, as these three cows suffered from PVD. Omitting these three outliers and re-performing the CCC test showed an agreement between the two sampling techniques of rc ¼ 0.93 (0.91, 0.94) and a standard error (SE) of 1%. The variance components such as subject variance, random error variance, and method effect values were analyzed to be 2.7%, 0.1%, and 0.01%, respectively.

Fig. 3. Enhanced Bland-Altman plot illustrating the agreement in polymorphonuclear cell (PMN) % between the cytotape (CT) and cytobrush (CB) sampling methods. In red, three outliers are notable. These correspond to a strong difference between CT and CB PMNs %. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4. Cytology smears obtained by cytotape and cytobrush, stained by Diff-Quick, observed by light microscope (_400). (A) Cytotape sample, (B) Cytobrush sample with moderate RBCs contamination. (C) Cytobrush sample with fragmented cells and moderate RBC contamination. (D) Cytobrush sample with high RBC contamination.

Separately, CCCs were calculated based on the presence and/or absence of a CL (>2 cm) and in cows suffering from PVD at the moment of sampling. The CCC in cows with CL was rc ¼ 0.85 (0.79, 0.89), whereas in cows bearing no CL on either of the ovaries, it was rc ¼ 0.82 (0.73, 0.87). In PVD positive cows, the CCC was rc ¼ 0.81 (0.70, 0.88), whereas in PVD negative cows, it was rc ¼ 0.83 (0.77, 0.87).

The quality, total cellularity, and RBCs’ contamination of the slides harvested by CT-DQ and CB-DQ are in detail summarized in Table 2.

For the second experiment, the CCC value between the CB-DQ and CB-CIAE staining was rc ¼ 0.84 (0.78, 0.89) with an SE of 2%. Variance components related to the subject was 4%, being much greater than the random error 0.6% and method effect 0.08% (Table 1).

4. Discussion.

The principal aim of the present study was to validate a new diagnostic technique to take endometrial cytology samples to diagnose SCE using the CB as the gold standard. The secondary goal was to compare the percentage of PMNs in twin cytology samples stained with Diff-Quik versus CIAE.

Fig. 5. Cytobrush with different amounts of blood contamination after endometrial sampling.

Fig. 6. Cytology smears stained by CIAE, observed by light microscope (_ 400). (A and B) Polymorphonuclear cells appearing in bright red. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

The high prevalence, its relatively difficult diagnosis and the economic impact of the condition, all make SCE one of the most important and challenging diseases in the modern dairy industry. Despite its undeniable importance, standardization of SCE diagnosis by cytology has not been fully established yet. Although it is well accepted that the best method to diagnose SCE is by measuring the proportion of PMNs in cytology samples ^[4], controversy remains because of the wide range of cutoff values used to define SCE^[20], DPP at the sampling moment ^[16] and cytology technique implemented to acquire samples ^[8–11,32]. To standardize SCE diagnosis, we developed an innovative device that allows to use the AI gun to take cytology samples. The latter would facilitate significantly sampling during insemination, which offers further advantages in terms of labor efficiency and knowledge of uterine health at the moment of fertilization. The concept of taking samples during AI has three main advantages in comparison to the more conventional sampling techniques: the standardization of the moment of sampling, the usage of ordinary materials, and avoiding “extra” handling of animals to obtain the samples. Furthermore, sampling at the time of insemination allows for a more direct examination of the effect of the counts of PMNs in the uterus on the success of pregnancy.

Central parameter used to evaluate the agreement between the here-presented innovative technique and the well-accepted CB was PMNs %. To obtain a wide distribution of PMNs % in the cytology smears, samples were harvested from cows in both early postpartum (31–37 DPP) and presumably late postpartum. Indeed, PMNs% ranged from 0% to 75% for both techniques, with a mean of 11% for CT and 8% for CB (Table 3). Agreement between both techniques concerning the PMNs % was good with a small SE. Interestingly, the variance components analysis found that the subject variance was significantly higher than the random error and the method effect. The latter implies that it is more likely to have a higher difference in PMNs % between samples when different cows are evaluated than that the disagreement is related to differences in reading the slides or to a method effect (CT or CB) related to the appreciation of the slides by the observer. The presence of PVD did not influence the CCC result. In both situations (PVD negative or positive), the CCC was at least “good”. While inspecting the enhanced Bland-Altman plot, three outliers were noticed presumably because the CT was contaminated during cervical manipulation in PVD positive cows. The agreement became “high” when these three samples were omitted. Therefore, we can conclude that presence of PVD does not affect the agreement between CT and CB, but a practical recommendation would be to use a sanitary sheet ^[33] in cows which are not in heat (closed cervix) during the sampling. In 57.8% of the sampled cows, a CL (>2 cm)was detected by ultrasonography, whereas the other 42.2% were in anestrus or around the time of ovulation (no CL or CL < 2 cm). The CCC between CT and CB were furthermore “good” both in cows bearing and not bearing an active CL. In resume, CT is a sampling technique that achieves cytology samples with a similar number of PMNs than CB and can be used at any stage of the estrous cycle.

Table 3. Total number of samples achieved and stained by each technique.

Minimum, maximum, mean, and standard deviation of the PMNs % obtained by each sampling technique and staining method.

Secondary parameters evaluated between CT and CB were total cellularity, quality, and RBC contamination (Fig. 4). Cellularity is a critical element in cytologic specimens and is closely correlated with the threshold used to define inflammation ^[9,34]. In the present investigation, both methods showed a similar total cellularity. In previous publications in which endometrial cytology methods were compared, CB yielded significantly more cells than LVL or CS^[11,32,35]. It is clear that in LVL, cells can be diluted ^[9]. The most probable reason why CT yielded a similar cellularity as CB is because the paper tape (on the top of the catheter) does not have absorbent properties as the CS, and most of the cellular material collected was adhered to the glass slide when the tape was rolled on it. In contrast with cellularity, both quality and RBC contamination were found to be significantly better for CT in comparison to CB. More intact cells were found in CT slides, resulting in significantly more CT samples being categorized in the category “very good”. In human medicine, the phenomenon of cellular distortion and fragmentation was described when the CS was used ^[10,11]. This distortion and fragmentation are mainly related to the firm adherence of cells to the cotton fibers and the consequent pressure needed to take the samples and to roll them onto the glass slide^[28,35]. The presumable reason why the CT slides yielded a better quality in comparison to the CB slides is the rigidity of the brush bristles and the fragmentation of cells when the brush is rolled on the microscope slide^[11]. Also, the endometrial material is not strongly adhered to the CT; consequently, cells are easily detached from the tape and scattered on the glass slides resulting in a small percentage of fragmented-distorted cells.

Fig. 7. Enhanced Bland-Altman plot illustrating the agreement between the Diff-Quick and CIAE staining. PMN, polymorphonuclear cell.

Remarkably, when samples are taken by CB, there is a significantly higher risk to have a bloody smear in comparison to CT sampling. Presence of RBCs in endometrial cytology samples was already described earlier, demonstrating higher numbers of RBCs in LVL than in CB ^[9] (in cows) and more bloody samples in CB versus CS ^[28], in mares.We found a robust difference represented by around 35% of CB samples with a high or moderate contamination of RBCs (Fig. 5). Rigid fibers of the CB might be responsible for high amount of RBCs in the samples ^[11], and this blood contamination should not be regarded as a negligible observation. Most of SCE publications lacked control groups when CB or LVL was performed. Only in one study, a control group was implemented ^[19]. Regarding the high amount of bloody samples when using the CB, one might argue damaging the uterine epithelium and the concomitant production of lesions in the endometrium, which opportunistic bacteria might use as an entrance gateway to cause a secondary infection. Additionally, bloody contamination of the samples implies the presence of 1% or 2% of PMNs ^[36]; hence, when a low threshold of PMNs is used to diagnose SCE^[37,38], these PMNs could interfere with the diagnosis.

It is mandatory to develop a cow-side diagnostic test to diagnose SCE with fair accuracy and at a low cost, under field conditions ^[39]. Both, CB and LVL are acceptable cytology techniques to diagnose SCE in dairy cows, but the principal reason why CB is preferred by practitioners is due to its practicability and cytologic quality^[8,9]. The main advantage of LVL is that samples are taken from a larger endometrial surface and tend to be more representative for the health of the complete endometrium^{[10,12,28,40,41]}. However, more studies are needed to prove this hypothesis. Major disadvantage of LVL is related to the difficulty to recover fluid after the infusion (17% of the cases), damage due to manipulation during sampling, irritation that flushing liquid can cause to the endometrium and a higher percentage of distorted cells, in comparison to CB smears^{[8–12,32,42–44]}. Main CB disadvantage is the requirement of specialized equipment^[9]. Conversely, CT is not a 100% cow-side diagnostic test because after sampling, slides need to be stained and analyzed under a microscope. However, CT is one step forward to an easier cytologic SCE diagnosis because it does not require special material (paper tape and double guard sheet), and it can be used with the AI gun achieving high-quality endometrial samples. Consequently, CT offers similar characteristics as the CB but requires less specialized material and can be used during AI.

A high-quality staining method is mandatory to yield an objective and accurate evaluation of the PMN-to-epithelial cell ratio in endometrial cytology samples. Modified Wright-Giemsa (Diff-Quik) is currently the most widely used staining to evaluate endometrial cytology slides from dairy cows. To assess the capacity of Diff-Quik to stain PMNs, it was compared with an enzyme histochemical staining in which PMNs appear bright red after staining (CIAE; Fig. 6), and which is therefore often used as the gold standard for identification and counting of these cells^[22,23]. Because the CIAE staining method requires several preparatory steps including a time-consuming incubation, it is considered time demanding and relatively expensive. Therefore, this technique is not recommended for routine use under field circumstances. Concordance correlation coefficient between Diff-Quik and CIAE PMNs’ % was good. Nevertheless, when the number of PMNs increases, the divergence between both staining techniques is also growing. Major differences between CIAE and Diff-Quik are only considerable when very high amounts of PMNs have to be evaluated, and the threshold for SCE is by far exceeded (Fig. 7).

4.1. Conclusions.

In summary, taking endometrial cytology samples with CB and CT yields similar results regarding parameters like PMNs % and total cellularity. However, techniques significantly differ in quality parameters and RBCs contamination in favor of the CT. When samples are taken by CT, less distorted–fragmented cells and a significantly lower contamination with RBCs were reported. On top of these advantages, the CT offers a technique that can be applied at the moment of insemination, by adhering the tape on the insemination pipette. The latter might allow field studies on a large scale to find a correlation between the number of PMNs in the uterine lumen and the conception result. Furthermore, this will allowto determine a straightforward cutoff value at a standardized moment (i.e., during insemination) above which the number of PMNs is associated with a reduced conception rate. Finally, modified Wright- Giemsa (Diff-Quik) is a fast, easy, and high-quality technique to stain endometrial cytology samples.

Acknowledgments.

The authors gratefully thank the family of Van Ranst for allowing part of the sampling at their dairy farm. Also, all personnel involved in helping for the sampling at both the dairy farm and the slaughterhouse are cordially thanked. The authors also thank Delphine Ameye and Christian Puttevils for the technical support with the CIAE staining.

References.

[1] Bondurant R. Inflammation in the bovine female reproductive tract. J Anim Sci 1999;77:101–10.

[2] Kasimanickam R, Duffield T, Foster R, Gartley C, Leslie K, Walton J, et al. Endometrial cytology and ultrasonography for the detection of subclinical endometritis in postpartum dairy cows. Theriogenology 2004;62:9–23.

[3] Gilbert RO, Shin ST, Guard CL, Erb HN, Frajblat M. Prevalence of endometritis and its effects on reproductive performance of dairy cows. Theriogenology 2005;64:1879–88.

[4] Sheldon IM, Lewis GS, LeBlanc S, Gilbert RO. Defining postpartum uterine disease in cattle. Theriogenology 2006;65:1516–30.

[5] Dubuc J, Duffield T, Leslie K, Walton J, LeBlanc S. Definitions and diagnosis of postpartum endometritis in dairy cows. J Dairy Sci 2010;93:5225–33.

[6] Madoz L, Giuliodori M, Jaureguiberry M, Plöntzke J, Drillich M, De la Sota R. The relationship between endometrial cytology during estrous cycle and cutoff points for the diagnosis of subclinical endometritis in grazing dairy cows. J Dairy Sci 2013;96:4333–9.

[7] De Boer M, LeBlanc S, Dubuc J, Meier S, Heuwieser W, Arlt S, et al. Invited review: systematic review of diagnostic tests for reproductive-tract infection and inflammation in dairy cows. J Dairy Sci 2014;97:3983–99.

[8] Barlund C, Carruthers T, Waldner C, Palmer C. A comparison of diagnostic techniques for postpartum endometritis in dairy cattle. Theriogenology 2008;69:714–23.

[9] Kasimanickam R, Duffield TF, Foster RA, Gartley CJ, Leslie KE, Walton JS, et al. A comparison of the cytobrush and uterine lavage techniques to evaluate endometrial cytology in clinically normal postpartum dairy cows. Can Vet J 2005;46:255.

[10] Bohn AA, Ferris RA, McCue PM. Comparison of equine endometrial cytology samples collectedwith uterine swab, uterine brush, and lowvolume lavage fromhealthy mares. Vet Clin Pathol 2014;43:594–600.

[11] Cocchia N, Paciello O, Auletta L, Uccello V, Silvestro L, Mallardo K, et al. Comparison of the cytobrush, cottonswab, and low-volume uterine flush techniques to evaluate endometrial cytology for diagnosing endometritis in chronically infertile mares. Theriogenology 2012;77:89–98.

[12] Roszel J, Freeman K. Equine endometrial cytology. Vet Clin North Am Equine Pract 1988;4:247–62.

[13] Hammon D, Evjen I, Dhiman T, Goff J, Walters J. Neutrophil function and energy status in Holstein cows with uterine health disorders. Vet Immunol Immunopathol 2006;113:21–9.

[14] LeBlanc S, Duffield T, Leslie K, Bateman K, Keefe GP, Walton J, et al. Defining and diagnosing postpartum clinical endometritis and its impact on reproductive performance in dairy cows. J Dairy Sci 2002; 85:2223–36.

[15] LeBlanc SJ. Reproductive tract inflammatory disease in postpartum dairy cows. Animal 2014;8(Suppl 1):54–63.

[16] Bara_nski W, Podhalicz-Dzie?gielewska M, Zdu_nczyk S, Janowski T. The diagnosis and prevalence of subclinical endometritis in cows evaluated by different cytologic thresholds. Theriogenology 2012; 78:1939–47.

[17] Pak H, Yokota S, Teplitz R, Shaw S, Werner J. Rapid staining techniques employed in fine needle aspirations of the lung. Acta Cytol 1980;25:178–84.

[18] Yang G, Alvarez II. Ultrafast Papanicolaou stain. An alternative preparation for fine needle aspiration cytology. Acta Cytol 1994;39: 55–60.

[19] Kaufmann T, Drillich M, Tenhagen B-A, Forderung D, Heuwieser W. Prevalence of bovine subclinical endometritis 4h after insemination and its effects on first service conception rate. Theriogenology 2009; 71:385–91.

[20] Melcher Y, Prunner I, Drillich M. Degree of variation and reproducibility of different methods for the diagnosis of subclinical endometritis. Theriogenology 2014;82:57–63.

[21] Plöntzke J, Madoz L, De la Sota R, Drillich M, Heuwieser W. Subclinical endometritis and its impact on reproductive performance in grazing dairy cattle in Argentina. Anim Reprod Sci 2010;122:52–7.

[22] Leder L-D. Diagnostic experiences with the naphthol AS-D chloroacetate esterase reaction. Blut 1970;21:1–8.

[23] Overbeck W, Jäger K, Schoon H-A, Witte T. Comparison of cytological and histological examinations in different locations of the equine uterusdan in vitro study. Theriogenology 2013;79:1262– 8.

[24] Ferguson JD, Galligan DT, Thomsen N. Principal descriptors of body condition score in Holstein cows. J Dairy Sci 1994;77:2695–703.

[25] Pleticha S, Drillich M, Heuwieser W. Evaluation of the metricheck device and the gloved hand for the diagnosis of clinical endometritis in dairy cows. J Dairy Sci 2009;92:5429–35.

[26] Williams EJ, Fischer DP, Pfeiffer DU, England GC, Noakes DE, Dobson H, et al. Clinical evaluation of postpartum vaginal mucus reflects uterine bacterial infection and the immune response in cattle. Theriogenology 2005;63:102–17.

[27] Veronesi M, Gabai G, Battocchio M, Mollo A, Soldano F, Bono G, et al. Ultrasonographic appearance of tissue is a better indicator of CL function than CL diameter measurement in dairy cows. Theriogenology 2002;58:61–8.

[28] Bourke M, Mills JN, Barnes A. Collection of endometrial cells in the mare. Aust Vet J 1997;75:755–8.

[29] Crawford SB, Kosinski AS, Lin H-M, Williamson JM, Barnhart HX. Computer programs for the concordance correlation coefficient. Comput Methods Programs Biomed 2007;88:62–74.

[30] Carrasco JL, Phillips BR, Puig-Martinez J, King TS, Chinchilli VM. Estimation of the concordance correlation coefficient for repeated measures using SAS and R. Comput Methods Programs Biomed 2013;109:293–304.

[31] Dohoo I, Martin S, Stryhn H. Veterinary epidemiological research. Charlottetown, Prince Edward Island, Canada: VER Inc; 2009.

[32] Overbeck W, Witte T, Heuwieser W. Comparison of three diagnostic methods to identify subclinical endometritis in mares. Theriogenology 2011;75:1311–8.

[33] Aguilar J, Hanks M, Shaw DJ, Else R, Watson E. Importance of using guarded techniques for the preparation of endometrial cytology smears in mares. Theriogenology 2006;66:423–30.

[34] Martin-Hirsch P, Jarvis G, Kitchener H, Lilford R. Collection devices for obtaining cervical cytology samples. Cochrane Database Syst Rev 2000;3:CD001036.

[35] Trimbos J, Arentz N. The efficiency of the cytobrush versus the cotton swab in the collection of endocervical cells in cervical smears. Acta Cytol 1985;30:261–3.

[36] Kahn C.M, Line S, Allen D.G. The Merck veterinary manual, 2005, Merck; Whitehouse Station, NJ

[37] Salasel B, Mokhtari A, Taktaz T. Prevalence, risk factors for and impact of subclinical endometritis in repeat breeder dairy cows. Theriogenology 2010;74:1271–8.

[38] Pothmann H, Prunner I, Wagener K, Jaureguiberry M, de la Sota R, Erber R, et al. The prevalence of subclinical endometritis and intrauterine infections in repeat breeder cows. Theriogenology 2015; 83:1249–53.

[39] LeBlanc SJ. Postpartum uterine disease and dairy herd reproductive performance: a review. Vet J 2008;176:102–14.

[40] Bonnett BN, Miller RB, Etherington WG, Martin SW, Johnson WH. Endometrial biopsy in Holstein-Friesian dairy cows. I. Technique, histological criteria and results. Can J Vet Res 1991;55:155.

[41] Miller H, Kimsey P, Kendrick J, Darien B, Doering L, Franti C, et al. Endometritis of dairy cattle: diagnosis, treatment and fertility. Bovine Pract 1980;15:13–23.

[42] Ball B, Shin S, Patten V, Lein D, Woods G. Use of a low-volume uterine flush for microbiologic and cytologic examination of the mare’s endometrium. Theriogenology 1988;29:1269–83.

[43] Brook D. Uterine cytology. Equine Reproduction Lea and Febiger. Philadelphia 1993. p. 246–54.

[44] Dini P, Farhoodi M, Hostens M, Van Eetvelde M, Pascottini OB, Fazeli M, et al. Effect of uterine lavage on neutrophil counts in postpartum dairy cows. Anim Reprod Sci 2015;158: 25–30.