Improved diagnostic accuracy by repetitive ultrasonic pregnancy testing in sheep.

Operator effects and instrument accuracy in using the Scanopreg ultrasonic pregnancy detector in sheep bred at synchronized estrus were studied in three experiments. In the first study, four operators tested the same 101 ewes at 60 and 80 days after breeding. The only significant difference among the four operators was that one operator consistently underestimated pregnancy. Operators did not differ in their diagnoses between days 60 and 80. In the second study, there were no differences between two operators who tested 239 ewes 90 days after breeding. In the third study, one operator tested 318 ewes 60, 70 and 90 days after breeding. The accuracy of diagnosis of pregnancy was at least 90% on each day tested; the corresponding diagnoses of nonpregnancy were 52, 76 and 79% correct. Some ewes that were initially diagnosed as nonpregnant were correctly recognized as pregnant when tested later than day 60. Most of the missed pregnancies were in ewes carrying a single lamb. A second Scanopreg test on day 90 of ewes not diagnosed pregnant on day 60 or 70 identified additional ewes as pregnant. Paired tests (days 70 and 90) recognized 99% of the ewes that eventually lambed.