Master of Science / Department of Clinical Sciences / David E. Anderson / Drug delivery systems for time release of recombinant human bone morphogenetic protein-2 (rhBMP-2) and antibiotics in orthopedic surgeries continue to be developed. Recently, a biodegradable novel polymeric matrix has been developed for this purpose. We hypothesized that impregnation of the matrix with rhBMP-2 would enhance bone healing. The objectives of the study were to characterize elution of rhBMP-2 and two antimicrobials (tigecycline, tobramycin) from the matrix, and bone response to the matrix in the presence or absence of rhBMP-2 and antimicrobials. In vitro elution of tigecycline, tobramycin, and rhBMP-2 from the matrix was investigated. Drug concentration in media were measured on days 1-6, 8, 10, 13, 15, 17, 21, 25, 28, and 30 using high pressure liquid chromatography/mass spectrometry/mass spectrometry (HPLC/MS/MS; antimicrobials) and ELISA (rhBMP-2). In vivo testing was done using a unicortical defect created into each tibia of twenty adult goats. Animals were randomly assigned to one of 5 groups: 1) control (untreated defect); 2) matrix; 3) matrix+ antimicrobials (tigecycline+tobramycin); 4) matrix+rhBMP-2; and 5) matrix+antimicrobials+rhBMP-2. Plasma concentration of tigecycline and tobramycin and serum concentration of rhBMP-2 were measured by the above techniques on days 1-7, 9, 11, 13, 15, 17, 22, 26, and 30. Bone response was assessed on days 0, 14, and 30 using radiographic scoring and dual energy X-ray absorptiometry (bone mineral density [BMD]). After euthanasia on day 30, histomorphologic analyses of the bone defects were done. Categorical variables were analyzed using a generalized linear model, and continuous variables using an ANOVA with P < 0.05 considered significant. In vitro elution was characterized by a rapid release on day 1 followed by a slow release until day 30 for both antimicrobials and rhBMP-2. Plasma antimicrobial concentrations showed continued release throughout the study period. Serum rhBMP-2 concentration, radiographic scores and BMD were not significantly different between groups. Periosteal and endosteal reaction surface areas were significantly greater surrounding the defects in group 4 (matrix+rhBMP-2). There was no significant difference between the groups for the percent of bone filling the defect. The matrix served as an appropriate antimicrobial and rhBMP-2 delivery system and successfully stimulated bone production when rhBMP-2 was present.