The objective of this study was to identify alterations in innate immunity reactants, carbohydrate and lipid metabolites in the blood of transition dairy cows with subclinical mastitis (SM). Multiparous Holstein cows (n = 100) were involved in the study and the experimental period lasted 16 wk starting at −8 wk before until +8 wk postpartum. Health status, feed intake, and milk yield and composition were monitored for each cow during the whole experimental period. Blood samples were collected from the coccygeal vein once/week before the morning feeding. Six healthy cows (CON) and 6 cows with SM were selected for intensive blood analyses. Serum concentrations of lactate, nonesterified fatty acids (NEFA), β-hydroxybutyrate (BHBA), interleukin-1 (IL-1), interleukin-6 (IL-6), tumor necrosis factor (TNF), haptoglobin (Hp), and serum amyloid A (SAA) were determined. Feed intake, milk production and composition also were determined. Data were processed using the MIXED procedure of SAS 9.2. Results indicated that concentrations of lactate (3,043 vs 2,243 ± 149 μmol/L, P < 0.01), TNF (0.52 vs 0.19 ± 0.04 ng/mL, P < 0.01) and SAA (23,915 vs 8,514 ± 2,518 μg/mL, P < 0.01) were greater in cows with SM than CTR cows. Moreover, serum lactate (3,478 vs 2,455 μmol/L at −8 wk, P = 0.03; 3,467 vs 2,162 μmol/L at −4 wk, P = 0.04) and TNF (1.29 vs 0.27 ng/mL at −4 wk, P < 0.01) in cows with SM were different from CTR cows starting at about −8 and −4 wks before diagnosis of disease. Overall feed intake (36.18 vs 39.81 ± 1.55 kg/d, P = 0.01) and milk production (33.76 vs 42.16 ± 2.52 kg/d, P = 0.04) was lower in SM-affected cows. Additionally, milk fat (3.32 vs 5.08 g/kg, P = 0.02) and fat:protein ratio (1.11 vs 1.69, P < 0.01) were lower, whereas SCC (1,867,000 vs 28,330 cells/mL, P = 0.02) and milk urea N (18.70 vs 15.39 mg/dL, P = 0.03) were greater in SM cows at diagnosis of disease vs CON. In conclusion cows affected by SM showed enhanced concentrations of indicators of innate immunity and metabolites related to carbohydrate metabolism weeks before diagnosis of SM. More research is warranted to validate these data and better understand the reasons for activation of innate immune responses to SM in transition dairy cows.