Comparison of vaginal examination methods to evaluate urovagina and purulent vaginal discharge in periestrous dairy cows

Abstract

Urovagina and purulent vaginal discharge (PVD) are usually diagnosed using a speculum, Metricheck device, or gloved hand. In periestrous dairy cows, a comparative study of these vaginal examinations for diagnosing urovagina or PVD has not yet been conducted. This study aimed to identify an effective vaginal examination method for periestrous dairy cows to ensure successful artificial insemination. Data were collected from 227 Holstein-Friesian cows during 300 occurrences of spontaneous estrus. Urovagina was evaluated using the speculum and gloved-hand methods and classified as mild, moderate, or severe. Vaginal discharge was evaluated using speculum, Metricheck, and gloved-hand methods and classified as vaginal discharge score (VDS) 0–5, with 2–5 defined as PVD-positive. Sensitivity and specificity of the gloved-hand versus speculum method in diagnosing urovagina, speculum versus Metricheck method, and gloved-hand versus Metricheck method in diagnosing PVD positivity were equivalent. The incidence of urovagina tended to be higher with the gloved-hand than with the speculum method. The incidence of PVD positivity tended to be higher with the gloved-hand than with the speculum and Metricheck methods. To analyze the correlation between pregnancy outcomes and mucus characteristics diagnosed using each method, logistic regression analysis was conducted, and the final models were compared. In this model, urovagina was selected as the explanatory variable and was associated with poor pregnancy. The results indicate that the gloved-hand method would be useful for managing fertility by detecting urovagina and PVD in periestrous dairy cows.

Vaginal diseases, such as urovagina and purulent vaginal discharge (PVD), have a negative impact on the reproductive performance of dairy cows. Urovagina represents the accumulation of urine in the cranial portion of the vagina in cows and mares, also known as urine pooling or vesicovaginal reflux of urine-mixed mucus [1, 2]. More than a quarter of dairy cows has been reported to contract urovagina at least once during the lactation period [3]. Moderate to severe urovagina is thought to cause infertility in dairy cows [3]. PVD, caused by inflammation of the genital ducts (uterus, cervix, and vagina), negatively affects reproductive health [4, 5]. PVD-positive dairy cows diagnosed 2 weeks to 1 month after delivery are less fertile than PVD-negative cows [6,7,8]. Understanding the adverse effects of urovagina and PVD on fertility is crucial for improving reproductive outcomes.

Both urovagina and PVD can be diagnosed by artificial insemination (AI) technicians using vaginal examinations. Generally, three types of vaginal examination methods are used to detect genital inflammation in postpartum dairy cows, which entail using a speculum (vaginoscope), Metricheck device, and gloved hand. Among these, the speculum has been used for the longest time for diagnosing PVD and observing the external uterine orifice around estrus [9, 10]. The Metricheck and gloved-hand methods, which collect vaginal mucus using a rubber cap or a clean gloved hand, respectively, were introduced to diagnose postpartum PVD [11, 12]. Pleticha et al. (2009) compared these three methods in postpartum dairy cows and concluded that all of them could diagnose PVD to the same degree without negatively affecting reproductive performance [7]. No comparative study of these three vaginal examinations for diagnosing PVD in periestrous dairy cows (cows after a voluntary waiting period [VWP] with varying days in milk [DIM]) has yet been reported. Furthermore, a comparative study of these vaginal examinations for diagnosing urovagina in both postpartum and periestrous dairy cows has not been reported [10]. Similar to postpartum vaginal examinations, periestrous vaginal examinations can potentially suggest better choices for farmers to enhance the reproductive performance of dairy cows because these examinations not only determine estrus and appropriate insemination time but also identify animals with and without PVD, allowing appropriate interventions [9, 13]. The conception rate (CR) of AI in cows identified as having a clear vaginal discharge using a Metricheck device was higher than that in cows with an abnormal vaginal discharge [13]. Therefore, evaluation of vaginal discharge with a Metricheck device in estrous dairy cows is considered an effective method to identify cows with severe uterine disease [13]. A speculum can be used to determine the periestrous vaginal and cervical status, thus pinpointing the optimal timing for AI [9]. However, using a speculum is time-consuming and thus not frequently used as a diagnostic tool [14, 15]. Therefore, alternative methods for vaginal examination without the use of special devices, such as the gloved-hand method, require further investigation.

This study aimed to identify an effective vaginal examination method for evaluating urovagina and PVD for successful AI in periestrous dairy cows. We hypothesized that vaginal examinations using a speculum, Metricheck device, and gloved hand could diagnose urovagina and/or PVD with equal efficacy and predict the risk of pregnancy in periestrous dairy cows, as described for postpartum dairy cows [6, 7, 12]. To test this hypothesis, we conducted multiple examinations on dairy cows at the time of periestrus and compared the severity of urovagina and PVD scores obtained for each method. Additionally, we analyzed the relationship between the scores of each vaginal examination and pregnancy using logistic regression analysis.

Materials and Methods

This study was approved by the Animal Research Ethics Committee of Chiba Prefectural Agricultural Mutual Aid Association (Approval number: #CNS170101).

Animals and herds

Data were collected during 300 occurrences of spontaneous estrus from 227 Holstein-Friesian cows (one estrus from 178 cows and more than one estrus from 49 cows) belonging to 32 dairy herds located in the northwestern part of Chiba Prefecture or the southwestern part of Ibaraki Prefecture, Japan, from September 2016 to April 2017. The cows were milked twice daily and fed a typical regional diet consisting of high milk-producing concentrate and hay, with some farms also supplementing with byproducts and silage. This study focused on spontaneous estrus events in cows following VWP of 40 days postpartum, with DIM at the time of estrus observation ranging from 41 to 762. This VWP was used to exclude the postpartum period from the study. The average milk yield per year for cows ranged from approximately 7,000–9,000 kg. Cows from 27 and 5 herds were maintained in tie stalls and loose housing systems, respectively. Estruses that were induced by the intravaginal insert of controlled internal drug release (CIDR, Eazi-Breed, InterAg, Hamilton, New Zealand), postovulation cows diagnosed with hemorrhagic vaginal discharge through at least one vaginal examination method, and estruses of cows diagnosed with ovarian cysts or pyometra by transrectal palpation and ultrasound scan were excluded from this study.

Evaluation of vaginal mucus

Periestrous cows were selected by dairy farmers when they displayed signs of estrus such as restlessness, swelling and reddening of the vulva, viscous vaginal mucus, or standing firm when mounted by another cow. All cows claimed to be in estrus, which had been requested for AI or embryo transfer (ET) by farmers, were evaluated by the three vaginal examination methods using a speculum (Fujihira Industry, Tokyo, Japan), Metricheck device (SimcroTech, Hamilton, New Zealand), and long-sleeved plastic glove (a disposable rectal examination glove) in that order. The severity of urovagina and vaginal discharge score (VDS) were determined using each method. And then, transrectal palpation of the uterus and ovaries was performed before AI or ET.

Before vaginal examination, the vulva was washed with water and wiped with a dry paper towel. The speculum and Metricheck were cleaned and disinfected before use by rinsing with water and wiping with cotton soaked in 70% ethanol. Disposable and non-sterilized rectal gloves were used in the gloved-hand method.

First, a speculum was inserted sufficiently deep into the vagina to observe the external cervical os and vaginal wall. Urovaginal positivity and VDS were evaluated by visual inspection using a flashlight. Cases that showed vaginal contents with urine-mixed mucus (characterized by a yellowish color, watery consistency, and uremic smell) were defined as having urovagina [16,17,18]. The amount of mucus and/or urine in the vagina was observed under a speculum and classified as follows: low, a small amount of mucus only on the floor of the vagina; moderate, an appreciable amount of mucus covering up to half of the external cervical os; and large, a large amount of mucus covering more than half of the external cervical os. The severity of urovagina was determined according to the amount of urine inside the vagina as follows: mild, a low amount of urine or urine-mixed mucus; moderate, a moderate amount of urine or urine-mixed mucus; and severe, a large amount of urine or urine-mixed mucus [8]. The VDS was classified as follows: 0, no discharge; 1, clear mucus; 2, mucus with flecks of pus; 3, mucus containing less than 50% purulent discharge; 4, mucus containing ≥ 50% purulent discharge; and 5, mucus containing ≥ 50% purulent discharge with an odor. VDS 2–5 was defined as PVD-positive, and VDS 0–1 was defined as PVD-negative [12]. VDS 3–5 was defined as intense PVD. Subsequently, a Metricheck device was inserted, advanced into the vaginal fornix, and retracted caudally. Mucus adhering to the surface of the rubber hemisphere was visually observed on the device and urovaginal positivity and VDS were determined as described above. Finally, a gloved-hand method was used. Vaginal mucus was collected using a gloved hand, placed in a clear plastic tube, and visually observed to evaluate urovaginal positivity and VDS. To determine the severity of urovagina in the gloved-hand method, the amount of mucus and/or urine observed with the speculum method was used, as it is difficult to accurately determine the amount using the gloved-hand method.

The decision to perform AI was influenced by the presence and severity of PVD. In cases in which intense PVD was detected, AI was typically not performed. However, the final decision to perform AI was based on the overall condition of the cow as assessed by the first author. AI was performed in four cases of intense PVD with other favorable factors. Meanwhile, the urovagina did not influence the decision to perform AI. This is because endometritis, which is indicated by the presence of PVD, can be treated with antibiotics [14] or nonsteroidal anti-inflammatory drugs [19, 20], whereas treatments for urovagina have not been established. After the vaginal examinations, 55 cows were not inseminated because of high rectal temperature (n = 2), intense PVD (n = 13), sticky or starchy mucus (n = 16), waiting for intense estrous behavior for successful AI (n = 13), or a request for ET by the farmer. AI was performed using the rectovaginal technique with a plastic sheath cover. After removing the data of estrus in which ET was conducted (n = 11), the insemination rate (IR) was calculated by dividing the number of inseminated cows by the number of spontaneously estrous cows.

The incidence of urovagina and PVD was analyzed by categorizing the cows according to DIM (total number of cows, cows with < 200 DIM, and cows with ≥ 200 DIM) to investigate the association between the postpartum period and mucus characteristics. The total number of cows included 300 occurrences of periestrus from 227 cows, cows with < 200 DIM included 198 occurrences of periestrus from 157 cows, and cows with ≥ 200 DIM included 102 occurrences of periestrus from 80 cows. We used the threshold 200 DIM as it is the commonly used time point to evaluate management methods and reproductive performance in dairy cattle [7, 21].

Comparison of three vaginal examination methods

The severity of urovagina determined by the speculum method was compared with that determined by the gloved-hand method using a two-way table with an unweighted kappa statistic as a measure of agreement. The VDS determined by the Metricheck method was compared with that determined by the speculum and gloved-hand methods using the same process. The kappa statistic was interpreted as follows: > 0.8, almost perfect agreement; 0.6–0.8, substantial agreement; 0.4–0.6, moderate agreement; 0.2–0.4, fair agreement; and ≤ 0.20, slight agreement between the two methods [22]. Sensitivity and specificity were calculated using the standard cross-tabulation technique [22], which provisionally sets the VDS determined by the Metricheck method [13] and urovaginal severity determined by the speculum method [3] as the gold standard. The prevalence of urovagina and PVD was statistically analyzed using the Fisher exact test, followed by the Benjamini–Hochberg comparison method. R statistical software (version 4.3.3, R Foundation for Statistical Computing, Vienna, Austria) [23] was used for all analyses. Statistical significance was set at P-values < 0.05. The tendency of statistical significance was set at P-values < 0.1.

Pregnancy confirmation, CR, and pregnancy rate

Pregnancy was confirmed by professional veterinarians using transrectal palpation of the uterus or ultrasound scanning (B-mode, Tringa Linear VET, Esaote S.p.A., Genoa, Italy) 35–65 days after estrus. The CR was calculated by dividing the number of cows diagnosed as pregnant by the number of inseminated cows. The pregnancy rate (PR) was calculated by dividing the number of cows diagnosed as pregnant by the number of cows in spontaneous estrus.

Parameters for the logistic regression analysis

Housing, age, breed, parity, body condition score (BCS), temperature-humidity index (THI), number of previous services, semen type, existence of incomplete fusion of the Müllerian ducts, and state of vaginal mucus on the day of estrus were recorded. Age and breed data were obtained from the National Livestock Breeding Center database of Japan. Data on parity and number of previous services were obtained from the clinical records of the Western Veterinary Clinical Center. BCS was determined using a 5-point scale with a quarter-point system [24] and grouped as follows: BCS group (BCSG) = 1, if BCS was ≤ 2.75; BCSG = 2, if BCS was > 2.75 and ≤ 3.25; and BCSG = 3, if BCS was > 3.25 [21]. The THI on the day of the vaginal examination and AI were also recorded. To calculate the THI, daily ambient temperature (AT) and relative humidity (RH) data were obtained from the Tateno meteorological station, located approximately 30 km from the farms. The THI on the day of AI was calculated based on the equation by Kendall and Webster [25]: THI = (1.8 × AT + 32) − ([0.55 − 0.0055 × RH] × [1.8 × AT − 26]). A mean THI ≥73 was defined as heat stress for reproduction [26]. Data were recorded when the number of previous services was greater than three and sex-sorted semen was used. The genital tract was morphologically diagnosed as normal, mild-type incomplete fusion of the Müllerian ducts, or severe-type incomplete fusion of the Müllerian ducts using a previously described method [27]. Mild-type cows (n = 3) were included in this study because their fertility was reported to be equivalent to that of normal cows [27], while severe-type cows were not found.

Logistic regression analysis

The reproductive records of artificially inseminated cows were used for the logistic regression analysis (n = 235). A multinomial logistic regression analysis was performed using the pregnancy score (pregnant = 1, non-pregnant = 0) as the objective variable and other items, including PVD and urovagina, as explanatory variables. First, univariate analysis (Fisher exact test) was performed to assess the association between CR and each categorical factor, including heat stress, farmer, housing, parity, BCSG, number of previous services, sex-sorted semen, incomplete fusion of the Müllerian ducts, PVD, and urovagina. Age was used as a covariate. Factors with P-values < 0.2 were included in the multiple regression model. In the final models, variables were selected by backward stepwise selection using the Akaike information criterion (AIC) [28]. Adjusted odds ratios (ORs), 95% confidence intervals (CIs), P-values, and Nagelkerke’s R² [29] values were calculated. The three logistic regression models obtained from the results of each vaginal examination method were compared using AIC and Negelkerke’s R² to determine the appropriate method for predicting pregnancy. R statistical software (version 4.3.3) was used for all analyses. Statistical significance was set at P-values < 0.05.

Results

Descriptive analysis

The overall information of 300 periestrous periods from 227 dairy cows, in which the diagnostic results of urovagina (diagnosed using the speculum method) and PVD (diagnosed using the Metricheck method) are presented in Table 1. Primiparous and multiparous cows accounted for 26.7% and 73.3% of the total population, respectively. Mean DIM ± SD was 181.1 ± 120.1, and the median and interquartile range of BCS were 2.75 and 2.75–3.0, respectively. There was no difference between the IR of estrus diagnosed as urovagina-negative by speculum (84.9%) and those diagnosed as urovagina-positive (83.3%, P = 0.74). The IR of estrous cows diagnosed as PVD-negative using the Metricheck method (89.3%) was significantly higher than that of cows diagnosed as PVD-positive (42.9%, P = 4.9 × 10^–8).

Table 1. Overall information on 300 periestrous periods from 227 dairy cows diagnosed using the speculum or Metricheck method.

Diagnosis		n	Distribution of parity (%)		Average DIM	Median BCS (IQR)	Reproductive performance (%)
			1	2–			IR a)	CR b), c)	PR c)
Urovagina-negative d)		282	27.7	72.3	181.7 ± 119.8	2.75 (2.75–3.0)	84.9	38.1	32.2
Urovagina-positive d)		18	11.1	88.9	170.8 ± 127.5	2.75 (2.5–2.75)	83.3	13.3	11.1
PVD-negative e)		272	27.9	72.1	177.4 ± 120.6	2.75 (2.75–3.0)	89.3	36.6	32.5
PVD-positive e)		28	14.3	85.7	217.1 ± 110.7	2.75 (2.5–2.75)	42.9 *	36.4	14.8
	VDS 2	11	0	100	214.4 ± 126.4	2.75 (2.5–2.875)	72.7	50.0	36.4
	VDS 3–5	17	23.5	76.5	218.9 ± 103.4	2.75 (2.5–2.75)	23.5 *	0	0 *
Total		300	26.7	73.3	181.1 ± 120.1	2.75 (2.75–3.0)	84.8	36.6	30.9

^a) Estruses in which ET was conducted (n = 11) were excluded from the analysis. ^b) Estruses in which AI was not performed (n = 44) were excluded. ^c)AIs lacking information on pregnancy (n = 7) were excluded. ^d) Urovagina was diagnosed using the speculum method as the gold standard. ^e) PVD was diagnosed using the Metricheck method as the gold standard. Asterisks indicate statistical significance compared with PVD-negative cows: * P < 0.01. DIM, days in milk; BCS, body condition score; IQR, interquartile range; IR, insemination rate; CR, conception rate; PR, pregnancy rate; PVD, purulent vaginal discharge; VDS, vaginal discharge score.

Comparison of vaginal examination methods

The VDS was successfully evaluated using the speculum, Metricheck, and gloved-hand methods. Urovagina was successfully evaluated using the speculum and gloved-hand methods but could not be diagnosed using the Metricheck method because the black color of the rubber cap of the device made the urovaginal yellow color undetectable. The consistency of urovaginal severity or VDS between the different vaginal examination methods is presented in Supplementary Table 1–3. The measure of agreement between the gloved-hand and speculum methods was substantial when urovaginal severity was assessed (weighted kappa = 0.75, P = 1.4 × 10^–24) (Supplementary Table 1). The measure of agreement between the speculum and Metricheck methods was substantial when VDS was assessed (weighted kappa = 0.69, P = 2.9 × 10^–43) (Supplementary Table 2). The measure of agreement between the gloved-hand method and the Metricheck method was also substantial when the VDS was assessed (weighted kappa = 0.64, P = 1.4 × 10^–34) (Supplementary Table 3). A summary of the comparisons between the methods is presented in Table 2. The sensitivity and specificity of the gloved-hand versus the speculum method for diagnosing urovagina were 100% and 95.0%, respectively (kappa = 0.70). The sensitivity and specificity of the speculum versus Metricheck method for diagnosing PVD positivity were 84.6% and 97.8%, respectively (kappa = 0.80), and those of the gloved-hand versus Metricheck method were 92.9% and 91.9%, respectively (kappa = 0.64).

Table 2. Comparison of sensitivity and specificity of the examination methods for diagnosing PVD and urovagina.

Examination objective	Examination methods	Sensitivity	Specificity	Kappa (95% CI)	P-value
Urovagina	Gloved-hand	100	95.0	0.70	2.5 × 10–20
	(vs. Speculum)			(0.55–0.84)
PVD (VDS 2–5)	Speculum	84.6	97.8	0.80	2.4 × 10–37
	(vs. Metricheck)			(0.67–0.92)
	Gloved hand	92.9	91.9	0.64	2.4 × 10–22
	(vs. Metricheck)			(0.51–0.77)

CI, confidence interval; PVD, purulent vaginal discharge; VDS, vaginal discharge score.

The overall incidence of a urovagina diagnosis tended to be higher using the gloved-hand method (10.7%) than using the speculum method (6.0%) (P = 0.054) (Fig. 1A). The incidence of a urovagina diagnosis tended to be higher in gloved-hand method (12.6%) compared to the speculum method (6.6%) (P = 0.06) in the cows with < 200 DIM, whereas the incidence was comparable between the two methods (P > 0.05) in the cows with ≥ 200 (Fig. 1A). The overall incidence of a PVD diagnosis tended to be higher with the gloved-hand method (16.0%) than with the speculum (10.0%) and Metricheck methods (9.3%) (P = 0.06 for both) (Fig. 1B). The incidence of a PVD diagnosis was significantly higher with the gloved-hand method (13.6%) than with the speculum (6.6%) and Metricheck methods (6.1%) (P < 0.05 for both) in cows with < 200 DIM, whereas the incidence was comparable among the three methods (P > 0.05) in cows with ≥ 200 DIM (Fig. 1B).

Fig. 1.

Incidence of urovagina and purulent vaginal discharge (PVD) in periestrous dairy cows, analyzed by different vaginal examinations. Incidences of urovagina (A) and purulent vaginal discharge (PVD) (B) in all cows, cows at < 200 days in milk (DIM), and cows at ≥ 200 DIM are shown. Urovagina was diagnosed using the speculum and gloved-hand methods. PVD was diagnosed using the speculum, Metricheck, and gloved-hand methods. A vaginal discharge score (VDS) of 2–5 was defined as PVD. The P-values were calculated by the Fisher exact test followed by the Benjamini–Hochberg comparison method.

Logistic regression analysis

Logistic regression analysis was conducted to analyze the correlation between pregnancy and mucus characteristics evaluated using each vaginal examination method. A comparison of the final models is presented in Table 3. In the final models obtained for the gloved-hand and speculum methods, only urovagina was selected by backward stepwise selection using the AIC (Table 3), whereas none of the explanatory variables were selected for the Metricheck method. PVD was not selected as an explanatory variable in this model. The final model that used data from the gloved-hand method presented the highest Negelkerke’s R² and the smallest AIC among the three models, indicating that the mucus characteristics diagnosed by the gloved-hand method predicted pregnancy most effectively. Urovagina (OR = 0.28, P = 0.02) was associated with poor pregnancy in the model that used data from the gloved-hand method.

Table 3. Summary of the final binary logistic regression models for the risk of pregnancy in Holstein-Friesian cows.

Diagnosis method	Variable	Category	OR	95% CI	Nagelkerke’s R2	AIC	P-value
Speculum					0.026	309.3
	Urovagina	Yes	0.24	0.04–0.91			0.06
		No	Reference
Gloved hand					0.038	307.15
	Urovagina	Yes	0.28	0.08–0.75			0.02
		No	Reference

OR, odds ratio; CI, confidence interval; AIC, Akaike information criterion.

Discussion

We compared vaginal examination methods (Metricheck, speculum, and gloved-hand methods) to evaluate urovagina and PVD in periestrous dairy cows to determine whether they were suitable for AI. The results suggested that the gloved-hand method could detect urovagina and PVD at a higher rate and that its sensitivity and specificity were equivalent to those of the other methods. In addition, the risk of pregnancy was most accurately predicted using urovaginal information obtained using the gloved-hand method, as indicated by Negelkerke’s R² and AIC values. It is suggested that the gloved-hand method can diagnose urovagina and PVD simultaneously and accurately and ensures more successful AI than the other methods in periestrous dairy cows. This may be due to the close observation of a large volume of vaginal mucus by AI technicians using the gloved-hand method.

Runciman et al. (2009) reported that the vaginal examination method performed first had little effect on the outcome of the second diagnostic procedure [6]. In contrast, McDougall et al. (2007) suggested that the order of performance of the speculum and Metricheck methods may affect VDS judgment, where a higher rate of PVD positivity was found using the Metricheck method when the speculum method was performed first [12]. We estimated that the gloved-hand method affects the outcomes of subsequent vaginal examinations because it removes a large amount of mucus from the vagina. Therefore, in our study, the examinations were performed in the following order: speculum, Metricheck, and gloved-hand methods.

In this study, there was good agreement between the results of the speculum and gloved-hand methods regarding the evaluation of urovaginal severity and VDS in periestrous dairy cows. To the best of our knowledge, this is the first report to compare different vaginal examination methods for the evaluation of urovagina. The level of agreement between urovaginal severity evaluated by the speculum and gloved-hand methods was substantial, indicating that both methods were compatible for the vaginal examination of periestrous cows. On the other hand, comparison of VDS evaluated by the Metricheck and speculum methods has been well documented in postpartum dairy cows and has indicated that the level of agreement between VDS evaluated by the Metricheck and speculum methods was moderate to almost perfect [6, 12]. Pleticha et al. (2009) reported that the Metricheck method most frequently detected PVD among the three methods in postpartum dairy cows [7]; however, agreement of VDS between the gloved-hand method and the Metricheck or speculum method has not yet been reported. In this study, the agreement of the VDS between Metricheck and speculum or gloved-hand method was substantial, indicating that the three methods could be used interchangeably for vaginal examinations of periestrous dairy cows.

Our study suggests that the ability of vaginal examination methods to diagnose urovagina and PVD in dairy cows varies according to the DIM. Inflammation of reproductive organs after calving decreases with time: the reported incidence of PVD after parturition (36.8% to 69.3%) [6, 7, 12] was much higher than that of PVD at the time of estrus (14.8%) [13] in dairy cows. Pleticha et al. (2009) examined the incidence of PVD in dairy cows at 21–27 DIM using one of the three vaginal examination methods and reported that the Metricheck method detected a significantly larger number of cows with endometritis than the speculum and gloved-hand methods [7]. In contrast, we found that the gloved-hand method diagnosed urovagina and PVD more frequently than the speculum method in periestrous dairy cows at <200 DIM. In periestrous dairy cows at ≥200 DIM, the incidence of urovagina and PVD was comparable among the examination methods. This may be because cows with a long DIM, that were not pregnant for a long period, showed severe PVD; therefore, differences in detection rates among the methods were not detected. These findings suggest that the DIM influences the diagnostic efficiency of the vaginal examination methods.

We detected the yellow color of urovagina using the gloved-hand and speculum methods but not the Metricheck method, suggesting that the former two methods had an advantage over the latter. Furthermore, this study implied that urovagina was more detectable using the gloved-hand method than using the speculum method. This may be because the gloved-hand method uses a transparent tube to observe mucus, making it easy to find the yellow color and odor of urine in the mucus. Gautam and Nakao (2009) reported that the overall lactational incidence of urovagina in Japanese dairy cows was 26.7% [3], which is much higher than the incidence of urovagina diagnosed using a speculum in this study (6.0%). These incidences cannot be directly compared because we examined urovagina at a single point during periestrus, whereas Gautam and Nakao (2009) conducted the examination at the time of monthly reproductive examinations from 2 weeks to 7 months postpartum [3]. In contrast, the incidence of urovagina in the present study was slightly higher than that in Turkish dairy cows diagnosed using a speculum (3.1%) [17]. It should be noted that the order of the methods used for the vaginal examinations might have affected the results of urovaginal diagnosis in this study. It is possible that earlier vaginal examinations stimulated the vagina or bladder and altered the amount of urine in the vagina, which may have affected the diagnosis of urovagina. Thus, potential improper diagnoses due to slight urine contamination of the vagina should be considered, and further analyses on the effect of the order of vaginal examinations are required.

Pregnancy prediction was most accurate when results from the gloved-hand method were used as variables. In the final models that included the results of the vaginal examinations, urovagina was selected as an explanatory variable; however, PVD was not. The significantly higher IR in PVD-negative estrus compared to that in PVD-positive estrus likely lowered the influence of PVD on pregnancy in the logistic regression models, due to which PVD was not selected in the logistic regression models in this study. In contrast, urovagina at estrus had a negative effect on pregnancy in the final model that used data from the gloved-hand method, with an OR of 0.28. Urine has been suggested to negatively affect fertility by causing uterine inflammation and reducing sperm motility and viability [3, 30,31,32]. Effective treatments for urovagina, including surgical extension of the urine orifice [1, 2] and ozone infusion into the uterus on the day of estrus [18], are not prevalent in Japan. Thus, our results were consistent with previous reports that periestrous urovagina negatively affected pregnancy in a severity-dependent manner [3, 18], strongly suggesting that the detection and treatment of urovagina in periestrous dairy cows is important. As a possible method to reduce urovagina, the selection of individuals with a smaller vaginal inclination for breeding may be effective. Cows with a vulval lips angle of > 45º from the vertical line, classified as having a horizontal vulva, were nine times more likely to be diagnosed with urovagina than cows with a vertical vulva [3].

Although vaginal examinations are not often performed in clinical practice, they have several advantages. In this study, we showed the usefulness of vaginal examination, especially the gloved-hand method, in diagnosing urovagina and PVD for a successful pregnancy. The use of a speculum offers an advantage in assessing the periestrous vaginal and cervical status (opening of the external uterine orifice and hyperemia, swelling, and relaxation of the intravaginal part of the uterus), which is critical for successful AI. It has been reported that these changes in intravaginal estrous signs are related to the estradiol peak, luteinizing hormone surge, and ovulation, and that marked relaxation of the intravaginal part of the uterus is the most critical estrous sign that corresponds to the theoretical optimal time for AI [9]. Therefore, a combination of speculum and gloved-hand methods before AI may be a better option to optimize pregnancy outcomes.

In conclusion, this study compared multiple vaginal examination methods (speculum, Metricheck, and gloved-hand methods) to evaluate urovagina and vaginal discharge in periestrous dairy cows. The results suggest that the gloved-hand method is suitable for vaginal examination in periestrous dairy cows, not only for the detection of urovagina and PVD but also for predicting the risk of pregnancy. The gloved-hand method does not require any special tools, except for a long plastic glove and a transparent tube, which means that it is a conducive method for AI technicians to routinely evaluate vaginal discharge before AI.

Conflict of interests

The authors declare no conflicts of interest that could be perceived as prejudicing the impartiality of this research.