HERD-BASED DIAGNOSIS OF NEOSPORA CANINUM-INDUCED ENDEMIC AND EPIDEMICABORTION IN COWS AND EVIDENCE FOR CONGENITAL AND POSTNATAL TRANSMISSION

Aborting and nonaborting cows and their dams or daughters were studied to determine if herd abortion problems were associated with the prese nce of Neospora caninum antibodies and to estimate when aborting cows may have acquired the infection. Cows were sampled from 20 herds that had experienced an abortion epidemic presumed to have been caused by N . caninum and from 2 herds experiencing endemic abortion. Seroprevalen ce for 14 herds experiencing an epidemic ranged from 7% to 70%, as est imated by enzyme-linked immunosorbent assay. A strong association betw een seropositivity and abortion was found for only 5 of 14 herds with a presumed diagnosis of N. caninum abortion (P less than or equal to 0 .015, lower 95% confidence interval of odds ratio greater than or equa l to 1.2), indicating N. caninum may be overdiagnosed as the cause of an abortion epidemic in some herds. No association was found between d am and daughter seropositivity for herds experiencing an epidemic (P g reater than or equal to 0.17), suggesting that most cows aborting duri ng an epidemic were infected postnatally. For the 2 herds with endemic abortion (A, B), odds of an aborting cow having N. caninum antibodies were 3.4-fold (herd A) and 7.0-fold (herd B) higher than odds for non aborting cows (P less than or equal to 0.05). Cows that aborted a fetu s infected with N. caninum were more likely to have had a previous ser opositive daughter than were nonaborting seronegative cows (P less tha n or equal to 0.0025), suggesting that infection had been acquired bef ore conception of the aborted fetus.