Characterization of Bacterial Isolates Collected From a Sheep Model of Osseointegration

Dustin L. Williams,1, 2 Roy D. Bloebaum,3, 4 James Peter Beck,4 Cathy A. Petti1, 5

1Associated Regional and University Pathologists (ARUP) Laboratories, ARUP Institute for Clinical and Experimental Pathology, Salt Lake City, Utah

2Department of Biomedical Engineering, University of Utah, Salt Lake City, Utah

3Department of Veterans Affairs, Salt Lake City, Utah

4Department of Orthopaedics, University of Utah School of Medicine, Salt Lake City, Utah

5Departments of Medicine and Pathology, University of Utah School of Medicine, Salt Lake City, Utah

Background: Femorally inserted percutaneous osseointegrated implant technology provides an alternative to current socket prosthetics. However, similar to other percutaneous devices, there is no current effective method to prevent infection of these implants. Thus, animal models are being developed to help ascertain infection prevention strategies as well as to understand the natural course of infection related to these implants. These animal models lack a thorough characterization of the infecting pathogens involved.

A sheep model of osseointegration was developed having a unilateral titanium alloy pin inserted in the upper inner aspect of the tibia. The skin/pin interface was treated with a novel cationic steroid antimicrobial-13 (CSA-13) as a topical anti-infection agent (described elsewhere). There were also untreated controls. It was hypothesized that no bacteria would be cultured from the skin/pin interface, tissue, bone or blood of those animals receiving CSA-13 and that the organism(s) collected from the skin/pin interface of control animals would be the infecting organisms collected from tissue, bone and blood following sacrifice.

Methods: Upon the development of infection, bacterial isolates cultured from the skin/pin interface as well as from post-euthanasia sites including tissue, bone and blood were characterized using routine clinical laboratory procedures in conjunction with the automated Phoenix system used for clinical diagnostics.

Results: Forty three total bacterial species were identified. A wide variety of clinical pathogens were characterized including Staphylococci, Streptococci and Pseuodomonas aeruginosa as well as several veterinary pathogens such as Arcanobacterium pyogenes and Pasteurella multocida.

Conclusion: Bacteria with human clinical relevance as well as veterinary pathogens were characterized herein. It is important to characterize bacterial pathogens in percutaneous biomaterials applications to elucidate the mechanisms, rates and paradigms of infection. Here we begin to elucidate the pathogens involved in infection of osseointegrated implants relevant mainly to an animal model, but that may prove clinically relevant in human applications.