Early and accurate pregnancy diagnosis in dairy cattle is a key component of effective reproductive herd management ⁴. The primary objective of pregnancy monitoring is not only to identify pregnant animals but also to accurately detect non-pregnant individuals ⁵. Early pregnancy diagnosis shortens the open days period, thereby improving the financial efficiency of dairy farms. Timely identification of non-pregnant cows after artificial insemination facilitates faster rebreeding and reduces feeding and management costs ^1,5.

In Türkiye, the most commonly used pregnancy diagnostic methods under field conditions in dairy cattle are rectal palpation and transrectal ultrasonography. However, the use of biochemical-based diagnostic approaches for early pregnancy detection has been increasing ⁶. In addition to rectal palpation and ultrasonography, several biochemical parameters measurable in maternal blood—such as pregnancy-associated glycoproteins (PAG), pregnancy-specific protein B (PSPB), interferon-tau (IFN-τ), and early pregnancy factor (EPF)—are also employed for pregnancy detection ^1,7-10. In particular, commercially available rapid test kits based on the detection of PAG in milk, serum, plasma, and whole blood are gaining increasing attention in field veterinary practice due to their practical applicability, rapid results, user-friendly nature, and cost-effectiveness. The literature indicates that PAG-based lateral flow tests offer significant advantages in terms of reliability, speed, and suitability for field conditions in early pregnancy diagnosis ^6,11,12.

Pregnancy-associated glycoproteins are secreted by trophoblastic binucleate cells of the placenta and enter the maternal circulation, making them reliable biomarkers for early pregnancy diagnosis. PAG can be detected in serum or milk samples and, when analyzed using ELISA-based methods, allow pregnancy detection with approximately 95–98% accuracy as early as days 28–30 of gestation ^2,5,9. Variations in PAG concentrations during early gestation not only assist in pregnancy diagnosis but may also serve as potential indicators for predicting the risk of embryonic loss ¹³. In cases of non-pregnancy, PAG measurements provide highly reliable results, with accuracy exceeding 95%. Negative PAG results obtained after day 40 post-insemination indicate non-pregnancy with very high confidence (>95%) ¹¹. However, PAG consist of multiple isoforms with differing biochemical structures and half-lives, which may allow their persistence in maternal circulation during the late postpartum period, thereby increasing the risk of false-positive results ¹⁴.

The aim of this study was to provide insight into the reliability of serum PAG measurements for early pregnancy diagnosis, to evaluate the advantages and limitations of PAG testing, and to promote its use as a tool to enhance the economic efficiency of dairy farming. Ultimately, the study seeks to contribute to shortening the calving-to-conception interval, increasing national livestock numbers, and supporting the development of future research initiatives in the livestock sector.

Animals: In this study, clinically healthy Holstein cows (n = 20) were used. The animals ranged in age from 1 to 6 years and had an average body weight of approximately 650 kg. Throughout the experimental period, all animals were maintained under identical environmental and management conditions. The cows were housed in group pens equipped with adequate resting areas, proper ventilation, and dry bedding. In accordance with the recommendations of the National Research Council ^15, the animals were fed a total mixed ration (TMR) formulated based on their milk yield. Clean and fresh drinking water was provided ad libitum. At the beginning of the study, the body condition scores (BCS) of all animals were approximately 3 on a 1–5 scale.

Ultrasonographic Pregnancy Diagnosis: All animals received an estrus synchronization protocol followed by artificial insemination. All gynecological examinations were performed using transrectal ultrasonography with a portable B-mode ultrasound device equipped with a 5–10 MHz probe (SIUI CTS-800 Veterinary Ultrasound, Shantou, China). Based on the ultrasonographic findings, a total of 20 cattle were included in the study, comprising 10 heifers confirmed to be at 30–40 days of gestation and 10 cows determined to be non-pregnant at ≥110 days postpartum.

Collecting Samples: Blood samples (8 mL) were collected from all animals (n= 20) via the coccygeal vein into anticoagulant-free tubes and kept on ice for less than 2 hours until serum separation. Serum was obtained by centrifugation at +4°C (4000 × g for 5 minutes), after which the samples were labeled and stored at −20 °C until analysis.

Sample Analysis: The PAG-based rapid pregnancy test kit used in this study is a commercially available lateral flow immunoassay, with an approximate unit cost of 9 USD per test under local market conditions in Türkiye. The test was selected not only for its diagnostic performance but also for its suitability for routine field use and its cost-effectiveness as a practical diagnostic tool. According to the instructions provided in the test kit insert (Anieasy Bovine Pregnancy Rapid Test Kit, colloidal gold, Shenzhen, China) ^16, 120 μL (three drops) of blood, milk, or urine samples were applied to the sample well of the Pregnancy-Associated Glycoprotein rapid (lateral flow) test, followed by the addition of two drops of diluent. After sample application, a 20-minute period was allowed for result development, and the outcomes were documented within the first 10 minutes. The rapid test yields four possible result patterns. A positive reaction was defined by the distinct appearance of both the control and test lines. When the control line was clearly visible but the test line appeared faint, the result was classified as suspected positive. A negative result was indicated by the presence of the control line only. Tests in which the control line failed to appear were considered invalid, regardless of the visibility of the test line.

Serum samples were classified as positive when both the control and test lines were clearly visible. Negative and suspicious results were evaluated according to the instructions provided in the manufacturer's original reference manual. The reported percentage values are consistent with, and proportionally aligned to, the research and development validation data provided by the manufacturer.

It is acknowledged that retesting of serum samples classified as suspicious positive, through repeated examinations (e.g., one week later), could eliminate ambiguous and uncertain cases. However, this confirmatory approach was not applied in the present study, as the primary objective was to compare the diagnostic performance of the rapid test kit directly with ultrasonographic examination as the reference (gold standard) method.

Within this framework, the classifications were defined as follows: true positives referred to cows identified as pregnant by the rapid test kit and subsequently confirmed by ultrasonographic examination. True negatives denoted cows classified as non-pregnant by both the rapid test and ultrasonography. False positives referred to cows that tested positive for pregnancy using the rapid test but were confirmed as non-pregnant by ultrasonography. False negatives were defined as cows that yielded negative results in the rapid test but were diagnosed as pregnant by ultrasonographic evaluation.

Statistical Analysis: The following metrics were used for data analysis: accuracy ((true positive + true negative) / total number of samples), sensitivity (true positive / (true positive + false negative)), specificity (true negative / (true negative + false positive)), positive predictive value (true positive / total positive results), and negative predictive value (true negative / total negative results) (Table 1). The Kappa scoring scale applied in the study was as follows: 0.80–1.00 = very good, 0.60–0.80 = good, 0.40–0.60 = moderate, 0.20–0.40 = fair, and <0.20 = poor (Table 2). Additionally, the relationship between PAG and USG results was evaluated using Pearson correlation analysis. A correlation coefficient (r) was calculated to determine the strength and direction of the linear association between the two variables. Statistical analyses were performed using SPSS version 30.0 (IBM Inc. Chicago, USA) software, and the level of significance was set at p<0.05.

Confidence intervals could not be provided because the serum PAG concentrations (ng/mL) were not quantitatively measured. In addition, the detection thresholds of the PAG antibody bands impregnated on the lateral flow immunochromatographic test strip—specifically, the PAG concentration ranges (ng/mL) required to produce a positive or negative reaction—are not disclosed by the manufacturer. Consequently, quantitative interpretation of PAG levels and threshold-based confidence interval calculations were not feasible.

Büyütmek İçin Tıklayın

Table 1: Calculations od metrics

Büyütmek İçin Tıklayın

Table 2: Kappa scoring scale

In Group 2, consisting of non-pregnant animals at ≥110 days postpartum (n= 10), rapid testing of blood samples showed suspected positive results in 3 animals and negative results in 7 animals. Based on these findings, the PAG rapid test kit applied during the fourth postpartum month produced 30% false results and 70% correct results. These data are presented in Table 3. Overall evaluation of rapid test results from blood samples revealed an accuracy of 85%, sensitivity of 100%, specificity of 70%, positive predictive value (PPV) of 77%, and negative predictive value (NPV) of 100% (Table 3).

All animals were definitively classified as pregnant or non-pregnant by ultrasonographic examination. Animals confirmed as pregnant by ultrasonography were also identified as pregnant by the PAG lateral flow test kit. When the diagnostic accuracy of ultrasonography and the lateral flow test was compared in animals at 30–40 days of gestation, a concordance rate of 100/100 was observed. In contrast, this rate was calculated as 70/100 in animals evaluated during the fourth postpartum month (Table 4). The comparison between ultrasonography and the PAG test kit yielded a Kappa coefficient of 0.70, indicating a good level of agreement between the two methods. Pearson correlation analysis revealed a moderate-to-strong positive correlation between PAG and USG results (r= 0.734, R²= 0.538, p<0.01, Figure 1). This indicates that approximately 53.8% of the variance in USG outcomes was explained by PAG results, suggesting a statistically significant and meaningful linear relationship between the two diagnostic methods. However, given the limited sample size (n= 20), this finding should be interpreted with caution and verified in larger cohort studies.

Büyütmek İçin Tıklayın

Table 3: Accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the glycoprotein test associated with pregnancy in dairy cows

Büyütmek İçin Tıklayın

Table 4: Accuracy results of the lateral flow test/ultrasonographic examination

Büyütmek İçin Tıklayın

Figure 1: Linear correlation between pregnancy associated glycoprotein (PAG) rapid lateral flow test results and ultrasonographic (USG) findings

Kline et al. ¹⁷ reported high false-positive rates in PAG testing during the postpartum period, particularly within the first 42 days, noting that false positivity reached 98.16% at day 42 postpartum using IDEXX lateral flow tests. When considered together with the current study, these results suggest that an adequate postpartum waiting period is essential for reliable PAG testing, as early application may lead to false-positive outcomes. Kline et al. ¹⁷ recommended the use of the IDEXX Alertys OnFarm Pregnancy Test from day 40 of gestation onward, and both studies demonstrated consistent results during early pregnancy. However, while PAG detectability in the postpartum period was limited to 42 days in a study by Kline et al. ^17, the present study identified detectable PAG levels extending up to 120 days postpartum. Agreement between ultrasonography and PAG testing was good in the present study (Kappa= 0.70), whereas Kline et al. ¹⁷ reported very good agreement (Kappa=0.84). These differences may be attributed to variations in analytical sensitivity and antibody characteristics of the test kits used.

In another study, Akköse et al. ¹⁸ reported 100% sensitivity and 97.3% accuracy for pregnancy diagnosis after day 28 of gestation. Both studies emphasize the critical role of early pregnancy detection in effective reproductive herd management. However, the 30% false-positive rate and 70% specificity observed in non-pregnant cows at ≥110 days postpartum in the present study highlight a significant limitation of PAG testing during the postpartum period due to the prolonged persistence of PAG. Although Akköse et al. ¹⁸ did not directly evaluate the postpartum period, the high specificity (94.9%) reported at day 28 of gestation supports the reliability of PAG testing in early pregnancy. Differences in test performance among studies may result from the use of different commercial kits, such as the Bioeasy rapid test used in the present study and the IDEXX ELISA kit used by Akköse et al. ^18, as well as methodological differences. Additionally, while the present study demonstrated good agreement between ultrasonography and PAG testing (Kappa=0.70), Akköse et al. ¹⁸ reported no significant correlation between PAG concentrations and progesterone levels, suggesting variability in diagnostic concordance across methodologies.

Ricci et al. ¹⁹ reported that PAG may take up to 60 days postpartum to be completely cleared from maternal circulation and recommended performing pregnancy tests after this period to avoid false-positive results. In the present study, agreement between ultrasonography and serum-based PAG rapid testing was also classified as good (Kappa = 0.70). In contrast, Ricci et al. ¹⁹ reported Kappa values of 0.84 (very good) for plasma PAG ELISA and 0.77 (good) for milk PAG ELISA. These differences may be explained by methodological variations between rapid tests and laboratory-based ELISA assays, as well as differences in biological sample matrices. Ricci et al. ¹⁹ also reported that plasma PAG concentrations were approximately twice as high as milk PAG levels, highlighting the influence of sample type on PAG detection sensitivity.

Kaya et al. ²⁰ compared the performance of ELISA-based bovine pregnancy tests and IDEXX rapid pregnancy tests using plasma samples. They emphasized the practical advantages of rapid test kits, particularly their ease of use under field conditions without the need for laboratory infrastructure. In the present study, the agreement between ultrasonography and PAG testing was again found to be good (Kappa = 0.70). Although Kaya et al. ²⁰ did not report Kappa statistics, both studies accepted ultrasonography as the gold standard for pregnancy diagnosis.

In a study by Akköse et al. ^21, the IDEXX rapid pregnancy test used in dairy heifers achieved 96.2% sensitivity and 96.6% accuracy at day 32 of gestation. These findings are consistent with the 100% accuracy and sensitivity observed in the present study during days 30–40 of pregnancy. Akköse et al. ²¹ reported that PAG concentrations increase earlier in heifers than in cows, allowing reliable identification of non-pregnant heifers from day 24 of gestation, although confirmation of early positive results was recommended. In contrast, the present study identified a 30% false-positive rate in cows tested at ≥110 days postpartum, which may be related to differences between the Bioeasy rapid test used in this study and IDEXX kits used in other investigations.

In another study conducted by Szelényi et al. ²² evaluated early pregnancy diagnosis in 637 Holstein cows. Ultrasonography identified 42.9% (n= 273) of the animals as pregnant and 57.1% (n= 364) as non-pregnant. Among ultrasonographically confirmed pregnant cows, 1.1% yielded negative results with the IDEXX Alertys OnFarm test, while 11.2% of ultrasonographically non-pregnant cows tested positive. Based on these results, overall accuracy, sensitivity, specificity, PPV, and NPV were reported as 93.1%, 98.9%, 88.7%, 86.8%, and 99.1%, respectively. While true-positive rates were comparable to those of the present study, differences in false-positive rates may be attributed to sample size and the use of different commercial test kits. Taken together, these findings support the high diagnostic value of PAG-based tests for early pregnancy detection while emphasizing that test timing is critical to minimizing false-positive results during the postpartum period.

In a study conducted by Aydın and Dinç ^23, PAG levels were evaluated in milk samples using an ELISA kit. The test demonstrated a sensitivity of 92%, a specificity of 100%, a correct pregnancy detection rate of 100%, a correct non-pregnancy detection rate of 91.7%, and an overall accuracy of 95.8%. Moreover, a statistically excellent agreement was observed between ultrasonography and the ELISA test kit (kappa value >0.8), and the false-positive rate of the milk PAG test was reported as 8.03%. While the rates related to positive pregnancy diagnosis obtained from these two studies show similarity, differences are observed in the false-positive rates. These discrepancies are thought to primarily arise from variations in the number of animals, the measurement methods employed, and the types of samples analyzed.

Kaya et al. ²⁴ evaluated the effectiveness of a commercial ELISA kit based on PAG for early pregnancy diagnosis in Holstein–Friesian heifers and lactating cows. In that study, the overall sensitivity, specificity, and accuracy of the PAG-ELISA test were reported as 95.9%, 94.7%, and 95.2%, respectively, and the test performance was emphasized to be higher in heifers compared to lactating cows. Furthermore, it was stated that false-positive results could be observed in some cows evaluated during the postpartum period, which might be associated with residual PAG levels from previous pregnancies and individual variability. The findings obtained in the present study show partial similarity with the results reported by Kaya et al. ²⁴. Both studies demonstrate that PAG-based tests are reliable during the early stages of pregnancy. However, differences were observed in the false-positive rates during the late postpartum period. These discrepancies in postpartum false-positive rates may be explained by differences in the diagnostic methods used (lateral flow rapid test vs. ELISA), the type of biological samples analyzed (plasma vs. serum), the limited number of animals included in the studies, and variations in the timing of postpartum evaluation.

In the present study, the detection of pregnancies in heifers on days 30–40, confirmed by ultrasonography (the gold standard), using a PAG-based rapid test kit, was conducted to evaluate the reliability, accuracy, and internationally recognized validity of the test. Considering the cost of the rapid test kit, ultrasonographic examination remains a more accurate and cost-effective approach for early pregnancy diagnosis and the detection of embryonic loss. In cattle, previous pregnancies may leave residual PAG molecules that can be detected by the rapid test kit, potentially leading to false-positive results. Therefore, ultrasonography continues to be the indispensable method for early pregnancy diagnosis and the detection of embryonic death in such cases.

1) Barbato O, Menchetti L, Brecchia G, Barile VL. Using pregnancy-associated glycoproteins (PAGs) to improve reproductive management: From dairy cows to other dairy livestock. Animals 2022;12:1-16.rn

2) Filho RVO, Franco GA, Reese ST, et al. Using pregnancy associated glycoproteins (PAG) for pregnancy detection at day 24 of gestation in beef cattle. Theriogenology 2020;141:128-133.rn

3) Saat N, Ozturk B. Sığırlarda hızlı test kitleri ile gebelik teşhisi. In: Goncagül G, Günaydın E. (Editors). Veteriner Hekimlik Alanında Akademik Araştırma ve Derlemeler. Ankara: Platanus Publishing, 2023: 197-210.rn

4) Saat N, Cetinkaya YB, Ozturk T, Ozturk B, Risvanli A. Comparison of pregnancy specific protein-b abortive and pregnant ewes. Large Animal Review (LAR) 2025;31:359-364.rn

5) Balhara AK, Gupta M, Singh S, Mohanty AK, Singh I. Early pregnancy diagnosis in bovines: Current status and future directions. Scientific World Journal 2013;2013:1-10.rn

6) Akköse M. Comparative evaluation of two commercial pregnancy-associated glycoproteins tests for early detection of pregnancy in dairy cattle. Theriogenology 2023;200:11-17.rn

7) Szenci O. Recent possibilities for the diagnosis of early pregnancy and embryonic mortality in dairy cows. Animals 2021;11:1-18.rn

8) Bozkurt T, Gündoğan M, Esen F, Kul S. The use of B mode real-time ultrasonography rectally for early pregnancy diagnosis in sheep. Fırat Üniversitesi Sağlık Bilimleri Veteriner Dergisi 1999;13:185-188.rn

9) Yıldız A, Deveci H. Diagnosis of pregnancy by determining early pregnancy factor (EPF) in cows. Fırat Üniversitesi Sağlık Bilimleri Veteriner Dergisi 2000;14:85-92.rn

10) Saat N, Çetinkaya YB. Sığırlarda doğum sonrası dönemde gebelikle ilişkili glikoprotein hızlı testinin güvenirliği. In: Goncagül G, Günaydın E, (Editors). Veteriner Hekimlik Alanında Akademik Araştırma ve Derlemeler. Ankara: Platanus Publishing, 2023:287-297.rn

11) Safak T, Tümer KÇ, Alp Yİ, Özen TY, Yilmaz-Koc O. Evaluation of the accuracy and performance of two commercial pregnancy-associated glycoprotein tests for early pregnancy detection in cows. Vet Med Sci 2025;11:1-7.rn

12) Benabdelaziz A, Boudjemai S, Abdelli AA, et al. Development and validation of a solid phase radioimmunoassay system for the determination of pregnancy-associated glycoproteins in bovine serum. Kafkas Univ Vet Fak Derg 2024;30:445-453.rn

13) Ealy AD, Seekford ZK. Symposium review: Predicting pregnancy loss in dairy cattle. J Dairy Sci 2019;102:11798-11804.rn

14) Northrop EJ, Rich JJJ, Rhoades JR, Perry GA. Comparison of two bovine serum pregnancy tests in detection of artificial insemination pregnancies and pregnancy loss in beef cattle. PLoS One 2019;14:1-10.rn

15) National Research Council (NRC). Nutrient Requirements of Dairy Cattle. 7th Edition, Washington, DC: National Academies Press, 2001.rn

16) Devmurari YJ, Vala KB, Parikh SS, et al. Comparative efficacy of PAG-based rapid strip test kit, ultrasonography and per rectal palpation for early pregnancy diagnosis in cattle and buffaloes. Indian Journal of Veterinary Sciences and Biotechnology (IJVSBT) 2025;21:13-17.rn

17) Kline AC, Zoca SM, Epperson KM, et al. Evaluation of pregnancy associated glycoproteins assays for on farm determination of pregnancy status in beef cattle. PLoS One 2024;19:1-14.rn

18) Akköse M, Çebi Şen Ç. Sütçü sığırlarda erken gebelik tanısı. J Anim Prod 2019;60:171-179.rn

19) Ricci A, Carvalho PD, Amundson MC, et al. Factors associated with pregnancy-associated glycoprotein (PAG) levels in plasma and milk of Holstein cows during early pregnancy and their effect on the accuracy of pregnancy diagnosis. J Dairy Sci 2015;98:2502-2514.rn

20) Kaya MS, Bademkiran S, Köse M, et al. Sığırlarda gebelikle ilişkili glikoproteinlerin tespitinde kullanılan iki test kitinin karşılaştırılması. Ataturk Universitesi Veteriner Bilimleri Dergisi 2017;12:45-53.rn

21) Akköse M, Kurban M, Gökçe MM, Vural MR. Assessment of the diagnostic accuracy of a pregnancy-associated glycoproteins test for early pregnancy diagnosis in dairy heifers. Reprod Domest Anim 2025;60(3): e70050.rn

22) Szelényi Z, Lipthay I, Sánta A, et al. Pregnancy evaluation with a point-of-care pregnancy test in dairy cattle. Theriogenology 2024;214:201-205.rn

23) Aydın H, Dinç DA. Evaluation of pregnancy-associated glycoprotein (PAG) test in early pregnancy diagnosis in dairy cows. J Ad VetBio Sci Tech 2024;9:183-189.rn

24) Kaya MS, Köse M, Bozkaya F, et al. Early pregnancy diagnosis using a commercial ELISA test based on pregnancy-associated glycoproteins in Holstein-Friesian heifers and lactating cows. Turk J Vet Anim Sci 2016;40:694-699.