This study investigated and compared the pharmacokinetics of intra-articular (IA) administration of dexamethasone sodium phosphate (DSP) into three equine joints, femoropatellar (IAS), radiocarpal (IAC), and metacarpophalangeal (IAF), and the intramuscular (IM), oral (PO) and intravenous (IV) administrations. No significant differences in the pharmacokinetic estimates between the three joints were observed with the exception of maximum concentration (Cmax ) and time to maximum concentration (Tmax ). Median (range) Cmax for the IAC, IAF, and IAS were 16.9 (14.6-35.4), 23.4 (13.5-73.0), and 46.9 (24.0-72.1) ng/mL, respectively. The Tmax for IAC, IAF, and IAS were 1.0 (0.75-4.0), 0.62 (0.5-1.0), and 0.25 (0.08-0.25) h, respectively. Median (range) elimination half-lives for IA and IM administrations were 3.6 (3.0-4.6) h and 3.4 (2.9-3.7) h, respectively. A 3-compartment model was fitted to the plasma dexamethasone concentration-time curve following the IV administration of DSP; alpha, beta, and gamma half-lives were 0.03 (0.01-0.05), 1.8 (0.34-2.3), and 5.1 (3.3-5.6) h, respectively. Following the PO administration, the median absorption and elimination half-lives were 0.34 (0.29-1.6) and 3.4 (3.1-4.7) h, respectively. Endogenous hydrocortisone plasma concentrations declined from a baseline of 103.8 ± 29.1-3.1 ± 1.3 ng/mL at 20.0 ± 2.7 h following the administration of DSP and recovered to baseline values between 96 and 120 h for IV, IA, and IM administrations and at 72 h for the PO.