Seropositive animals were assigned on day 120 of gestation to a Control (n = 33) or Treatment (n = 34) group. Treatment consisted of progesterone supplementation by fitting a progesterone releasing intravaginal device containing 1.55 g of progesterone for 28 days. Mean Neospora caninum antibody titres tested by ELISA on days 40, 90 and 120 of gestation were used to classify the cows as showing high (30 units) or low (<30 units) titration. Abortion rate showed a significant increase (P < 0.01) in the Treatment group (35%) compared to the Control (9%). Using logistic regression procedures, there was a significant (P < 0.0001) interaction between progesterone treatment and antibody titration. Progesterone-treated cows with a high antibody titre showed a likelihood of abortion 14.3 times higher than those not treated, while in cows with low antibody titres, exogenous progesterone had not effect. Our results do not support our starting hypothesis and suggest that progesterone supplementation in cows with high N. caninum antibody titres increases the risk of abortion possibly by affecting cell-mediated immune responses.