INTRODUCTION: Juvenile idiopathic arthritis (JIA) often causes inflammation of the temporomandibular joint (TMJ) and has been treated with both systemic and intra-articular steroids, with concerns about effects on growing bones. In this study, we evaluated the impact of a macromolecular prodrug of dexamethasone (P-DEX) with inflammation-targeting potential applied systemically or directly to the TMJ. METHODS: Joint inflammation was initiated by injecting two doses of complete Freund's adjuvant (CFA) at 1-month intervals into the right TMJs of 24 growing Sprague-Dawley male rats (controls on left side). Four additional rats were not manipulated. With the second CFA injection, animals received (1) 5 mg of P-DEX intra-articularly (n = 9), (2) 15 mg of P-DEX into the tail vein (n = 7), or (3) nothing in addition to CFA (n = 8). The rats were killed 28 days later and measured by radiography for ramus height (condylar superior to gonion inferior [CsGoInf]), by micro-computed tomography for condylar width (CW) and bone volume/standardized condylar volume (BV/CV), and by histology for retrodiscal inflammatory cells. Inflammation targeting of systemic P-DEX was confirmed by IVIS infrared dye imaging. Inflammation and bone growth were compared between groups using analysis of variance and Pearson's correlations. RESULTS: CFA caused a significant reduction in CsGoInf (p < 0.05), but neither route of P-DEX administration had an effect on CsGoInf or CW at CFA injection sites. BV/CV was significantly reduced in both inflamed and control condyles as a result of either steroid application (p < 0.05). The inflammatory infiltrate was overwhelmingly lymphocytic, comprising 16.4 ± 1.3 % of the field in CFA alone vs. <0.01 % lymphocytes in contralateral controls (p < 0.0001). Both P-DEX TMJ (10.1 ± 1.2 %) and systemic P-DEX (8.9 ± 1.7 %) reduced lymphocytes (p < 0.002). The total area of inflammatory infiltrate was significantly less in the systemic injection group than in the group that received CFA injections alone (2.6 ± 1.5 mm(2) vs. 8.0 ± 1.3 mm(2); p = 0.009), but not in the group that received intra-articular P-DEX (8.8 ± 1.2 mm(2)). CONCLUSIONS: High-dose systemic administration of inflammation-targeting P-DEX is more effective than an intra-articular injection in reducing TMJ inflammation, but both routes may affect TMJ bone density.