Comparison of accuracy of transabdominal ultrasonography, progesterone and pregnancy-associated glycoproteins tests for discrimination between single and multiple pregnancy in sheep

The aim of the present study was to evaluate and, compare the accuracy of transabdominal ultrasonographic (US) and the progesterone (P4-RIA) and ovine pregnancy-associated glycoprotein (ovPAG-RIA) tests for the discrimination between single and multiple pregnancy in sheep. One hundred pregnant Awassi x Merino ewes were scanned by transabdominal ultrasonography (3.5 MHz linear-array transducer) at Days 43-56 and 81 of these ewes were scanned at Days 76-87 of gestation. The ewes were scanned in dorsal recumbency at the bare area of the inguinal regions (without pre-scanning shaving of the ventral abdominal wall). After each scan, blood samples were withdrawn from the jugular vein to estimate the levels of P4 and ovPAG by radioimmumoassay. At lambing, 61 ewes gave birth to single lambs and 39 ewes gave birth to multiples. The sensitivity of the transabdominal US, the P4-RIA and the ovPAG-RIA tests for determining ewes carrying multiples was 54, 64.1 and 64.1% at Days 43-56. At Days 76-87 of gestation these accuracies were 60.0, 66.7 and 76.6% for the US, P4-RIA and PAG-RIA tests, respectively. The specificity of the transabdominal US, the P4-RIA and the ovPAG-RIA tests for determining ewes carrying singles, was 78.6, 60.7 and 62.3% at Days 43-56 and 78.4, 64.7 and 70.6% at Days 76-87 of gestation, respectively. It is concluded that the accuracy of transabdominal ultrasonographic (without pre-scanning shaving of the ventral abdominal wall), the P4- and the ovPAG-RIA tests for determination of the fetal numbers in Awassi x Merino crossbred ewes is too low to be used in the field. (c) 2005 Elsevier Inc. All rights reserved.