Vancomycin-Loaded Polymethylmethacrylate Spacers Fail to Eradicate Periprosthetic Joint Infection in a Clinically Representative Mouse Model

Background: Periprosthetic joint infection (PJI) remains a devastating complication following total joint arthroplasty. Current animal models of PJI do not effectively recreate the clinical condition and thus provide limited help in understanding why treatments fail. We developed a mouse model of the first-stage surgery of a 2-stage revision for PJI involving a 3-dimensionally printed Ti-6Al-4V implant and a mouse-sized cement spacer that elutes vancomycin.

Methods: Vancomycin was mixed with polymethylmethacrylate (PMMA) cement and inserted into custom-made mouse-sized spacer molds. Twenty C57BL/6 mice received a proximal tibial implant and an intra-articular injection of 3 × 105 colony-forming units of Staphylococcus aureus Xen36. At 2 weeks, 9 mice underwent irrigation and debridement of the leg with revision of the implant to an articulating vancomycin-loaded PMMA spacer. Postoperatively, mice underwent radiography and serum inflammatory-marker measurements. Following euthanasia of the mice at 6 weeks, bone and soft tissues were homogenized to quantify bacteria within periprosthetic tissues. Implants and articulating spacers were either sonicated to quantify adherent bacteria or examined under scanning electron microscopy (SEM) to characterize the biofilm.

Results: Vancomycin-loaded PMMA spacers eluted vancomycin for ≤144 hours and retained antimicrobial activity. Control mice had elevated levels of inflammatory markers, radiographic evidence of septic loosening of the implant, and osseous destruction. Mice treated with a vancomycin-loaded PMMA spacer had significantly lower levels of inflammatory markers (p < 0.01), preserved tibial bone, and no intra-articular purulence. Retrieved vancomycin-loaded spacers exhibited significantly lower bacterial counts compared with implants (p < 0.001). However, bacterial counts in periprosthetic tissue did not significantly differ between the groups. SEM identified S. aureus encased within biofilm on control implants, while vancomycin-loaded spacers contained no bacteria.

Conclusions: This animal model is a clinically representative model of PJI treatment. The results suggest that the antimicrobial effects of PMMA spacers are tightly confined to the articular space and must be utilized in conjunction with thorough tissue debridement and systemic antibiotics.

Clinical Relevance: These data provide what we believe to be the first insight into the effect of antibiotic-loaded cement spacers in a clinically relevant animal model and justify the adjunctive use of intravenous antibiotics when performing a 2-stage revision for PJI.