Accuracy of pregnancy diagnosis and prediction of foetal numbers in sheep with linear‐array real‐time ultrasound scanning

Open Access
Publisher Website
Google Scholar
Abstract
Pregnancy diagnosis was carried out in sheep by means of transabdominal linear‐array real‐time ultrasound scanning. Animals were restrained standing, and the transducer was placed on the hairless area of the ventral abdominal wall just in front of the udder. Of a total of 818 tests, 724 were performed between days 29 and 89 of pregnancy, 598 animals subsequently lambed and 126 were non‐lambing animals. Only 8 of these tests were wrong: 3 false positive and 5 false negative diagnoses. Sensitivity, specificity, positive‐ and negative predictive values for these tests were 99.2%, 97.6%, 99.5%, and 96% respectively. There was evidence to indicate that the three false positive tests were caused by foetal mortality or unobserved abortions that took place after testing. Only 2 of the 5 false negative tests were carried out after day 39 of gestation. Counting of foetal numbers (1, 2 or 3) was performed in only some animals (n = 210) between days 45 and 77 of gestation. Three groups of animals (A: 89 ewes; B. 27 PMSG‐treated ewes; C: 94 ewes) were analyzed separately. Overall accuracy of all predictions was 83.1%, 37.0% and 78.7% for the 3 groups respectively. Animals in group B produced only 3 or more lambs. Sensitivity of the countings of singles, twins and triplets or more were 90.4%, 90.4% and 50% respectively for the animals from group A and 91.9%, 86% and 21.4% for the animals from group C. When only two categories (1 and ≥ 2) were used these figures were 90.4% (1) and 94. 1% (≥2) for animals from group A and 91.9% (1) and 89.5% (≥ 2) for animals from group C. It was concluded that linear‐array real‐time ultrasound scanning by the transabdominal route is a reliable method for (early) pregnancy diagnosis in sheep. In addition the technique enables a fairly accurate separation of ewes carrying a single lamb from as early as 40–70 days after mating.