A comparison of methods for early pregnancy diagnosis

Early identification of both pregnant and nonpregnant cows improves reproductive efficiency and pregnancy rate in cattle by decreasing the between service interval. Early pregnancy detection is only useful when techniques have a high level of accuracy for detection of both pregnant and non-pregnant animals. Pro's and con's for using; rectal palpation, ultrasonography, milk progesterone tests, or the early conception factor (ECF) test for pregnancy diagnosis are presented.

Summarized from: Proceedings 2nd Florida Dairy Road Show (2005)

Early identification of pregnant and nonpregnant cows improves reproductive efficiency and pregnancy rate in cattle by decreasing the interval between services. Many new and old technologies are available to identify pregnant and nonpregnant animals quickly after breeding. Such methods can play a key role in reproductive management to rapidly return open animals to the breeding program. This paper reviews rectal palpation, ultrasonography, milk progesterone tests, and the early conception factor (ECF) test for pregnancy diagnosis.

Measure of Accuracy of Pregnancy Diagnosis

Earlier pregnancy diagnosis can occur with increased errors in diagnosis. The accuracy of the diagnosis is expressed by sensitivity and specificity, and positive and negative predictive values. Table 1 explains these terms and gives an example calculation:

Table 1. Calculation and example of sensitivity, specificity, predictive value positive and predictive value negative (100 cows are diagnosed for pregnancy).

True status of cow Open Open
A = 40
Correct Decision
B = 5
False Negative
 A+B = 45
  Pregnant C = 1
False Positive
A+C = 41 
D = 54
Correct Decision
B+D = 59
C+D = 55
A+B+C+D = 100

Specificity = A/(A+B) = 40/45 = 88.9%
Sensitivity = D/(C+D) = 54/55 = 98.2%
Predictive value negative = A/(A+C) = 40/41 = 97.6%
Predictive value positive = D/(B+D) = 54/59= 91.5%

Specificity is the probability that a cow that is truly open is also diagnosed to be open. Specificity is calculated: A/(A+B) = 40/45 = 88.9%.

Sensitivity is the probability that the diagnosis is pregnant, given that a cow is truly pregnant. Calculation for sensitivity: D/(C+D) = 54/55 = 98.2%.

The predictive value negative is the probability that the cow is truly open, if the diagnosis is open, calculated as: A/(A+C) = 40/41 = 97.6%.

Predictive value positive is the probability that the cow is truly pregnant if the diagnosis is pregnant: D/(B+D) = 54/59 = 91.5%.

A pregnancy diagnosis method should have both a high Predictive Value Negative and high Sensitivity.

Rectal Palpation

Palpation of the uterine contents rectally is probably the most commonly used method for pregnancy diagnosis. Pregnancy diagnosis after insemination can be conducted as early as 30 days in heifers and 35 days in cows, although much practice is necessary in order to determine pregnancy at that stage. Several palpable structures are indicative of pregnancy. Due to accumulation of fluids within the pregnant uterine horn, one of the initial signs of pregnancy is a difference in size of uterine horns (uterine asymmetry). Also, it is possible to feel the slipping of the fetal membrane along the greater curvature within the uterus (membrane/fetal slip). There is a rule of thumb that is quite useful in estimating fetal age based on the size of the fetus in relationship to the size of some well known animals. This rule of thumb is detailed in the following Table 2.

Table 2. Calf fetal size at various stages of pregnancy in relation to the size of some commonly known adult animals. (source: P.J. Hansen)

Stage of pregnancy Calf fetal size in relation to the size of commonly known adult animals
2 months mouse
3 months rat
4 months small cat
5 months large cat
6 months beagle dog

Table 3 presents the uterine position and diameter, as well as structures felt at palpation according to stage of pregnancy. The following describes when specific structures can first be palpated:

Membrane Slip - 30-35 days
Amniotic vesicle - 35-60 days
Fetus – 65+ days

Rectal palpation has the advantage of being an accurate, fast, and relatively cheap method that is less labor intensive as compared to the previous methods. Nonetheless, training is necessary; the exam should be conducted by a veterinarian or by an experienced herdsman. The main disadvantage of rectal palpation is that it cannot be performed until later in gestation than some other methods. Some veterinarians are able to determine pregnancy by palpation as early as 35 days after insemination, but usually rectal examinations take place between 45 and 60 days after insemination to increase the accuracy of the exam.

Table 3. Uterine position/diameter and structures during pregnancy. (source: P.J. Hansen)

Days of gestation Uterine position Uterine diameter Palpable Structures
35-40 Pelvic floor Slightly enlarged Uterine asymmetry/fetal slip
45-50 Pelvic floor 5.0 - 6.5 cm Uterine asymmetry/fetal slip
60 Pelvis/abdomen 6.5 - 7.0 cm Membrane slip
90 Abdomen 8.0 - 10.0 cm Small placentomes/fetus (10-15 cm)
120 Abdomen 12 cm Placentomes/fetus (25-30 cm)/fremitis
150 Abdomen 18 cm Placentomes/fetus (35-40 cm)/fremitis


In the 1980s, real time ultrasonography was developed for use in domestic animals. An ultrasound machine resembles a radar device. A probe is inserted through the rectum and positioned above the uterus. The ultrasound probe acts to send and receive sound waves. When an electric field is applied to the crystals in the probe, they change shape and vibrate like cymbals, creating waves of sound. The ultrasound probe directs these high frequencies, low intensity sound waves toward the tissues. Different proportions of the sound waves emitted are reflected back to the probe, depending on the density of the tissue. The returning sound waves produce pressure on the crystals, generating an electric charge, which is converted to a visual image on the screen.

Fluid, such as blood or follicular fluids, does not reflect sound waves and no image (black) appears on the screen. Bone is the densest tissue and reflects sound waves almost completely depicting white images. Other tissues reflect varying proportions of sound waves and produce images of various shades of gray. The differences in the reflection of sound waves from various tissues or differences in the angle at which sound waves strike tissue surfaces may cause echoes.

Real-time, B-mode ultrasonography has been reported to detect pregnancy in cattle as early as 9 (Boyd et al., 1988) or 12 days (Pierson and Ginther, 1984) into gestation. Other reports, however, have disputed those claims and emphasized that the accuracy of ultrasound diagnosis of pregnancy on Day 10 through 16 is not significantly better than a random guess (<50%). Accuracy of diagnoses improves, however, by Day 18 (85%), 20 (100%) and 22 (100%) of pregnancy (Kastelic et al., 1989). Presence and vitality of the embryo can be confirmed by the detection of a heartbeat as early as 19 to 24 days of gestation. The embryo initially appears as a short line (Day 20-22), later becomes C-shaped (Day 22-30), and finally, by Day 30-32 of gestation assumes an L shape. Although the embryo can first be detected between Days 19 and 24 of gestation

(Curran et al., 1986), it is most practical to scan females which are expected to have embryos that are >26 days of age.

The main advantages of the use of ultrasound for pregnancy diagnosis are 1) the high accuracy of the results and 2) relatively early pregnancy. Main disadvantages of the use of ultrasonography are related to the cost and time involved. Ultrasound machines are relatively expensive. Pregnancy diagnosis with an ultrasound machine takes more time than rectal palpation. Also the training required for proper interpretation of the images can serve as a disadvantage.

Cows diagnosed pregnant at an early ultrasound exam have a greater risk of early embryonic loss and, therefore, must undergo subsequent pregnancy examinations to identify and rebreed cows that experience such loss. If left unidentified, cows experiencing embryonic loss after an early pregnancy diagnosis would actually reduce reproductive efficiency by extending their calving interval (Fricke, 2002).

Milk Progesterone Tests

Progesterone is the hormone known as the pregnancy hormone. The progesterone test is based on the expected changes in the production of this hormone during the reproductive cycles and pregnancy. Progesterone levels elevate during the midcycle of each reproductive cycle and during the entire gestation period.

Since progesterone is produced by the corpus luteum in the ovary, high progesterone levels show a functional corpus luteum. The corpus luteum forms and produces high amounts of progesterone after the cow has been in estrus and ovulated. If the cow is not pregnant, the corpus luteum regresses and progesterone levels decline to low levels about 2 days before the cow comes into heat again. However, if the cow becomes pregnant, the corpus luteum continues to function and progesterone levels remain high throughout gestation.

Therefore, if the cow is not pregnant and has regular estrous cycles, progesterone levels in the milk follow a cyclic pattern, being low from about 2 days before heat until about 4 to 5 days after heat and high during the middle portion of the cycle. Since progesterone is essential for the maintenance of pregnancy, it is continuously high in pregnant cows.

To use milk progesterone as a pregnancy indicator, the milk sample must be collected between 21 to 24 days after the cow was in estrus and inseminated. Low progesterone would indicate that the cow is not pregnant and high progesterone would indicate that the cow has a functional corpus luteum and might be pregnant.

The test is most accurate in determining non-pregnant cows, because if the progesterone levels are low the cow cannot be pregnant. A number of situations can result in high milk progesterone. Reasons milk progesterone levels might be high between 21 and 24 days after insemination include:

  • The cow is pregnant.
  • The cow is in the middle of her estrous cycle but not pregnant due to:
    o an estrous detection error led to breeding at the wrong time
    o a longer than usual estrous cycle.
  • Embryonic mortality. The cow conceived but the embryo died.
  • Abnormalities, such as pyometra (accumulation of pus in the uterus or a dead, mummified fetus).

Numerous studies indicate that the accuracy of early pregnancy diagnosis by milk progesterone is only about 80%. Reasons for this include: variation in estrus cycle length between cows, estrus detection errors, uterine disease (pyometra), ovarian dysfunction (luteal or follicular cysts), and early embryonic mortality, to name a few. In short, the reliability of milk progesterone for the diagnosis of pregnancy is not satisfactory. Any such diagnosis should be confirmed by palpation or ultrasound.

However, with a series of samples taken at day 0 (the date of insemination), 21, and 24, the accuracy for early diagnosis of non-pregnancy approaches 95 to 100% (Nebel et al., 1988).

Therefore, the milk progesterone test is a tool in the determination of the non-pregnant cow. The advantage of this early confirmation of non-pregnancy prevents the further loss of early breeding opportunities. Cow-side milk progesterone assays conducted between 18 and 24 days post AI had an overall reported accuracy of 97.2% for cows identified as nonpregnant (Pennington et al., 1985). This makes milk progesterone the earliest proven method for identifying nonpregnant animals.

The Early Conception Factor (ECF) Test

Recently, a new early pregnancy test has become commercially available for use in cattle. The Early Conception Factor (ECF) test (Concepto Diagnostics, Knoxville, TN) reportedly detects a pregnancy-associated glycoprotein within 48 hours of conception. Two studies have compared results from the ECF test conducted between Days 3 to 7 and Days 11 to 15 post-AI to pregnancy diagnosis using palpation per rectum and ultrasound ranging from 25 to 60 post-AI (Adams and Jardon, 1999; DesCôteaux et al., 2000). Results for test sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 86%, 4%, 49%, 23%, and 46%, respectively.

Although the observed agreement between readers (91% for Test 1; 89% for Test 2) and between tests for the same serum sample (94% for Reader 1; 91% for Reader 2) was high, the overall rates of false positive and false negative results were 96% and 14%, respectively.

Cordoba et al. (2001) concluded that the ECF test, in its present form, is an unreliable method for determining pregnancy status on day six after estrus in dairy cattle. Although the predictive value of a positive ECF test result increases as the conception rate in a herd increases, the predictive value of a negative

ECF test result would be less than 50% in dairy herds exhibiting a conception rate greater than 25%.


The best technique or combination of techniques for a given herd will deliver the desired level of accuracy in pregnancy diagnosis within the confines of availability of experienced labor and current facilities. Table 4 briefly summarizes the strengths and weaknesses of each technique.

Table 4. Comparison of early pregnancy diagnosis techniques.

Pregnancy Diagnosis Technique Early Testing Time Diagnosis Pregnancy Accuracy Diagnosis Non-Pregnancy Accuracy
Rectal palpation . ... ....
Ultrasound .. .... ....
Milk Progesterone ... .. ...
ECF .... . .

Literature Cited

1. Adams, C.S., and P.W. Jardon. 1999. Evaluation of the early conception factor tests in cows 3-7 days post-breeding. Proc. Am. Assoc. Bov. Pract. 32: 240-241.

2. Beal, W.E., R.B. Edwards, III and J.M. Kearnan. 1989. Use of B-mode, linear array ultrasonography for evaluating the technique of bovine artificial insemination. J. Dairy Sci. 72:2198.

3. Boyd, J.S., S.N. Omran and T.R. Ayliffe. 1988. Use of a high frequency transducer with real time B-mode ultrasound scanning to identify early pregnancy in cows. Vet. Rec. 123:8.

4. Cordoba, M.C., R. Sartori, and P.M. Fricke. 2001. Assessment of a commercially available early conception factor (ECF) test for determining pregnancy status of

dairy cattle. J. Dairy Sci. 84:1884.

5. Curran, S., R.A. Pierson and O.J. Ginther. 1986. Ultrasonographic appearance of the bovine conceptus from days 20 through 60. J. Amer. Vet. Med. Assoc.189:1295.

6. DesCôteaux, L., P.D. Carrière, and M. Bigras-Poulin. 2000. Evaluation of the Early Conception Factor (ECF) dipstick test in dairy cows between days 11 and 15 post-breeding. Bov. Pract. 34:87-91.

7. Fricke, P. M. 2002. Scanning the future-ultrasonography as a reproductive management tool for dairy cattle. J. Dairy Sci. 85:1918.

8. Kastelic, J.P., S. Curran, and O.J. Ginther. 1989. Accuracy of ultrasonography for pregnancy diagnosis on days 10 to 22 in heifers. Theriogenology 31:813.

9. Nebel R.L. 1998. On-farm progesterone tests. J. Dairy Sci. 71:1682-1690.

10. Pennington, J.A., L.H. Schultz, and W.F. Hoffman. 1985. Comparison of pregnancy diagnosis by milk progesterone on day 21 and day 24 postbreeding: field study in dairy cattle. J. Dairy Sci. 68:2740–2745. Proceedings 2nd Florida Dairy Road Show (2005) 29

11. Pierson, R.A. and O.J. Ginther. 1984. Ultrasonography for detection of pregnancy and study of embryonic development in heifers. Theriogenology 22:225.

12. Ruiz, F.J., Oltenacu, P.A., Smith, R.D. 1989. Evaluation of on-farm milk progesterone tests to determine non pregnant cows and to prevent insemination errors. J. Dairy Sci. 72:2718-2727. Proceedings 2nd Florida Dairy Road Show (2005) 30


University of Florida

University of Florida