Background: Bovine mastitis causes major economic losses for milk producers by reducing the quantity and the quality of the milk or even leading to the complete loss of the mammary gland secretory capacity. During the transition period, dairy cows are susceptible to infectious diseases; therefore, markers that allow early identification of cows in higher risk of developing diseases are especially useful at this time. Therefore, the aim of this study was to evaluate serum markers in the pre and postpartum of multiparous dairy cows with clinical mastitis and with health condition in the postpartum period in a semi-extensive management system.Materials, Methods & Results: Thirty-Six Holstein cows were monitored daily during milking until 59 days postpartum and were categorized according to the pre-milking strip cup test into clinical mastitis (mastitis group (MG)) and absence of symptoms (control group (CG)) that were negative to the test, representing the health cows. All cows were reared as one group and maintained in a semi-extensive pasture-based system. Blood samples were collected weekly after morning milking via venipuncture of the coccinea vein into tubes without anticoagulant and grouped for prepartum (-21 to 0 days from calving), early postpartum (0 to 30 days from calving), and late postpartum (30 to 59 days from calving) periods. Milk production was recorded daily. The serum markers albumin, aspartate aminotransferase (AST), phosphorus, gammaglutamyltransferase (GGT) and non-esterified fatty acids (NEFA) were measured. Statistical analyses were performed using SAS®. The cases of clinical mastitis occurred on average at 37.2 ± 4.9 days postpartum. Health cows (CG) had higher milk production compared to the mastitis group (MG) only in the late postpartum period (P < 0.05). There was no difference among groups for albumin and NEFA concentrations in all periods evaluated (P > 0.05). In the early postpartum period the AST activity was higher in CG than in MG (P = 0.02). The GGT enzyme tended to be more concentrated in the CG than in the MG during the early (P = 0.06) and late (P = 0.08) postpartum periods. Late postpartum phosphorus concentration was lower for MG than CG (P = 0.04). In the prepartum and early postpartum periods, there was no difference among groups for phosphorus concentration (P > 0.05).Discussion: A decrease in milk production in MG compared to CG observed in late postpartum period was due to the incidence of mastitis observed around 37 days postpartum. Cows that presented clinical mastitis in the postpartum period did not differ in the blood concentration of NEFA in the prepartum period. In the late postpartum period higher concentration of phosphorus was observed in the CG than in MG, indicating that animals affected by mastitis may be in the weakest energy status. Regarding liver health, the concentration of AST was higher in the recent postpartum period for CG, in disagreement with previous studies that related AST to tissue injury caused by mastitis. The GGT enzyme tended to had higher concentrations in CG than MG during the whole postpartum period and may be related to increased hepatic metabolism due to higher production. There were no changes in albumin levels among healthy and mastitis cows, indicating that this marker can not be used to predict clinical mastitis. There were no metabolic alterations in the prepartum period related to the occurrence of postpartum mastitis in multiparous cows in a semi-extensive management system.