Most dairy cows experience a substantial decrease in feed intake in the 1 to 2 weeks preceding parturition, which has been associated with systemic inflammation (SI). We aimed to assess the association of prepartum decrease in feed intake and the onset of SI, and if present, the ability of anti-inflammatory treatment to mitigate inflammation. Holstein cows (n = 45) were blocked on parity (1.1 ± 1, range 0 to 3) and BCS (3.6 ± 0.3, range 3.25 to 4) and assigned to control (n = 13), untreated feed restriction (FR-U; n = 15), or feed restriction treated with anti-inflammatory (FR-T; n = 17) groups. Daily feed intake was measured from −22 d before expected parturition until 35 d postpartum. All cows received the same total mixed rations during the dry and fresh periods. Control cows were fed ad libitum while FR-U and FR-T cows were reduced to 70% of their average intake for 4 d (−15 to −12 d). FR-T cows received meloxicam (0.5 mg/kg of BW) once/day for 4 d (−13 to −10 d). Blood samples were collected before feeding on d −22, −15, −14, −13, −12, −10, −7, −5, −3, 0, 1, 3, 5, 7, 15, 22, and 35 from calving to measure total protein, albumin, urea, glucose, GGT, AST, GLDH, BHB, NEFA, haptoglobin, and IGF-1. Cows with dystocia or clinical metabolic disease were excluded from the analyses of the postpartum but not from the prepartum period. Treatment effects were assessed by linear regression models treatment, sampling day, and their interactions. All outcomes changed (P < 0.05) with time independent of treatment. In both FR groups, concentration of NEFA in serum gradually increased from the first until the last day of restriction, when it was higher (P = 0.02) than the control group (control 0.28 ± 0.03, FR-U 0.67 ± 0.1, and FR-T 0.67 ± 0.05 mmol/L). FR increased serum urea concentrations on −14 d (P = 0.05), but on −10 d it was lower than the control (P = 0.02). No differences among treatments were observed for the remaining metabolites. This model of feed restriction to mimic that seen in the week before calving produced substantial fat mobilization but not, based on the present indicators, the hypothesized induction of SI.