Pregnancy diagnosis in swine: a comparison of the accuracies of mechanical and endocrine tests with return to estrus.

: The relative accuracies of various commercially available detection procedures in the evaluation of pregnancy status were evaluated in 51 sows of mixed parity. Sows were mated 1 to 3 times during estrus to one or more boars. A blood sample was obtained from each sow between 19 and 23 days after estrus (first day of estrus = day 1) and again between 26 and 30 days and was analyzed by radioimmunoassay for content of progesterone and estrone sulfate, respectively. Sows were considered pregnant if progesterone concentrations were greater than 5 ng/ml and if estrone sulfate concentrations were greater than 0.5 ng/ml. At daily intervals, sows were exposed to mature, sexually aggressive boars and observed for signs of estrus. Between 31 and 35 days of gestation, sows were evaluated for pregnancy by use of the following mechanical devices: a Doppler ultrasound pregnancy detector fitted with an abdominal probe, the Doppler detector with a rectal probe, an amplitude-depth ultrasound pregnancy detector (diode ruler), and another amplitude-depth ultrasound detector from a different manufacturer (oscilloscope). Relative to the proportion of sows that farrowed, the overall accuracies of the techniques were returns to estrus, 98%; estrone sulfate, 92%; progesterone, 92%; Doppler with rectal probe, 86%; Doppler with abdominal probe, 86%; diode ruler amplitude-depth ultrasound, 73%; and oscilloscope amplitude-depth ultrasound, 86%. Estrus detection resulted in the lowest prevalence of sows that were predicted to farrow, but did not subsequently farrow (false-positive diagnoses) and the lowest prevalence of sows that were not predicted to farrow, but subsequently did farrow (false-negative diagnoses).(ABSTRACT TRUNCATED AT 250 WORDS)