Real-time Ultrasonography and diagnosis of pseudopregnancy in swine


W.L. Flowers



The effectiveness of using real-time ultrasonography for diagnosis of pseudopregnant sows was evaluated using 2000 sows on a commercial swine operation in N.C. Sows that were diagnosed as pregnant on day 28 of gestation were randomly assigned to receive either no further evaluation for pregnancy status (control, n=1000) or examination with a 5.0-MHz linear probe between days 65 and 75 of gestation (n=1000). More sows presumed to be pregnant at day 28 failed to farrow in the control treatment (132/905) than in the real-time ultrasound treatment (23/898). Nine of the 23 sows misdiagnosed with real-time ultrasound had fewer than 4 viable fetuses. These results demonstrate that real-time ultrasonography can be used as a management tool to effectively identify pseudopregnant sows.



Pseudopregnancy or “Not-in-Pig” is a condition that occurs when females exhibit  physiological and behavioral signs associated with pregnancy except there are no fetuses present. As a result, many common forms of pregnancy diagnosis including  detection of estrus with a boar and A-mode and Doppler ultrasonography cannot be used to distinguish between pregnant and pseudopregnant females. The timely identification and removal of non-pregnant females from the breeding herd is important because it reduces both non-productive sow days and production costs.


Real-time (B-mode) ultrasonography consists of sound waves emitted from transducers that travel in different patterns. Upon contact with tissues, the waves are reflected back to the transducer, where they are converted into electrical signals and displayed on a monitor as a two dimensional image. Black areas on the screen correspond to fluid-filled structures such as the bladder and uterus, and gray areas represent tissues such as bone and muscle. Because real-time ultrasonography allows for visualization of fetuses within the female reproductive tract, it seems to be ideally suited for identification of pseudopregnant females. Thus, the objective of this experiment was to determine the effectiveness of using real-time ultrasonography for identification of pseudopregnant females.


Materials and Methods

This experiment was conducted on a farrow-to-wean, commercial swine farm in northeastern North Carolina. Weekly breeding groups on this farm consisted of approximately 120 sows. The farm had a history of sows that were diagnosed as pregnant on day 42 of gestation and exhibited subsequent visual signs of pregnancy between 70 and 90 days of gestation, but did not farrow. This phenomenon, referred to as pseudopregnancy or “Not-in-Pig” syndrome, occurred predominantly between July and October. Yorkshire X Landrace X Large White sows (n=2000; mean parity 4.2 + 0.8) that were diagnosed as pregnant between day 22 and 28 of gestation with a 5.0-MHz probe (Ultra Scan 45, Alliance Medical Inc., Montreal, Quebec, Canada) were used in the study. One thousand sows were reexamined with real-time ultrasonography between day 65 and 75 of gestation. Sows in which no fetuses could be seen were sent to a local abattoir, where their pregnancy status was determined via visual observation of the reproductive tract at slaughter. The control treatment consisted of 1,000 sows that were not examined with ultrasonography after day 28 and remained in the herd until their anticipated farrowing date. The experiment was conducted over a 12-week period between July and September. For all ultrasound diagnoses, the probe was placed approximately 3 to 4 inches above the last two pairs of nipples on the right side of the sow.


The proportion of sows diagnosed as pregnant but did not farrow (presumed to be pseudopregnant) between the control and ultrasound treatment was analyzed with chi-square procedures (Snedecor and Cochran, 1989). Breeding group was used as a stratification variable. Sows that aborted, died, exhibited estrus after day 28 of gestation, or were removed from the herd because of lameness or sickness were not included in the statistical analyses.


Results and Discussion

Real-time (B-mode) ultrasonographic images representative of those used for diagnosis of pregnancy between days 65 and 75 of gestation are shown in Figure 1. The outline of the spinal cord and rib cage of fetuses in conjunction, at times, with the observation of a fetal heartbeat were criteria used to classify sows as pregnant. Sows scanned during the same time period whose ultrasound images lacked the definitive outline of fetal skeletons were considered to be not pregnant (pseudopregnant).

Figure 1. Representative images from a nonpregnant/pseudopregnant sow (A) with fluid accumulation (v) and a pregnant sow with viable fetuses (B) on day 70 of gestation using a 5.0-MHz probe. The spinal cord (sp) can be seen in the image from the pregnant animal.

One hundred two (88.7%) of the 115 sows identified as being not pregnant between day 65 and 75 were confirmed as not having any fetuses (Table 1). Nine of the 23 sows misdiagnosed as pregnant with real-time ultrasonography contained 4 or fewer viable fetuses. A smaller (p=0.05) proportion of sows failed to farrow from the group in which real-time images between day 65 and 75 were used as removal criteria compared with a group of their contemporaries in which no diagnostic techniques other than detection of estrus were used after day 28 (23/898 versus 132/192, respectively).

Table 1. Effectiveness of using real-time ultrasonography between days 65 and 75 of gestation for identification of pseudopregnant sows on a commercial swine farm.






Disposition of sows






Number of sows diagnosed as pregnant of day 28

of gestation






Number of sows removed from study before day 65

of gestation




Number of sows diagnosed as pseudopregnant




Proportion of sows remaining in study at day 65 of

gestation that were diagnosed as pregnant but did

not farrow






Proportion of pseudopregnant sows diagnosed correctly






*Numbers in parentheses are percentages.

a,bproportions within the same row with different superscripts are different (P = 0.05).

These results demonstrate that by specifically looking for skeletons of fetuses after day 65 of gestation, sows that aborted after being scanned previously and those that were pseudopregnant during earlier pregnancy diagnoses were accurately identified with real-time ultrasonography. This is not possible with most other forms of ultrasonography or techniques commonly used for pregnancy diagnosis in swine. It is important to recognize that the criteria used for identification of pregnancy after day 65, the presence of the fetal skeleton and heartbeat, are different than those commonly used earlier in gestation, the presence of fluid-filled pockets in the uterus. Failure to make this adjustment will result in the inability to identify pseudopregnant females. This is due to the fact that fluid retention and its subsequent accumulation within the uterus is common during pseudopregnancy.


Although the earliest stage of gestation during which pseudopregnancy can be diagnosed was not examined in the current study, it is doubtful that the accurate identification of this condition via real-time ultrasonography could be achieved much before day 65. Significant calcification of the skeleton usually does not begin until after day 60 in pigs. As a result, delineation between pregnant and pseudopregnant sows before this time period would be difficult.



Real-time ultrasonography between days 65 and 75 of gestation is an accurate way to identify pseudopregnant females.


Literature Cited

Snedecor, G.W. and W.G. Cochran. 1989. Statistical Methods (8th Ed.). Iowa State University Press, Ames, Iowa.