Developing a Novel Clinically Representative Biofilm Based Gram-Negative Prosthetic Joint Infection Rat Hip Hemiarthroplasty Model

Abstract

Introduction: Gram-negative prosthetic joint infections (GN-PJI) present unique challenges in management due to their distinct pathogenesis of biofilm formation on implant surfaces. The purpose of this study is to establish a clinically representative GN-PJI model that can reliably recapitulate biofilm formation on titanium implant surface in vivo. I hypothesized that biofilm formation on an implant surface will affect its ability to osseointegrate. Methods: The model was developed using 3D-printed titanium hip implants, to replace the femoral head of male Sprague-Dawley rats using a posterior surgical approach. GN-PJI was induced using two bioluminescent Pseudomonas aeruginosa (PA) strains: a reference strain (PA14-lux) and a mutant strain that is defective in biofilm formation (flgK-lux). Infection was assessed in real-time using the in vivo imaging system (IVIS) and Magnetic Resonance Imaging (MRI) and in vitro by quantifying bacterial loads on collected implants surface and in periprosthetic tissues as well as biofilm visualization using the Field emission scanning electron microscopy (FE-SEM). The implant stability, as an outcome, was directly assessed by quantifying the osseointegration in vitro using microCT scan, and indirectly assessed by identifying the gait pattern changes using DigiGaitTM system in vivo. Results: Bioluminescence detected by IVIS, was focused on the hip region, demonstrating localized-infection, with the ability of PA14-lux to persist in the model compared to flgK-lux defective in biofilm formation. This was corroborated by MRI as the PA14-lux induced relatively larger implant-related abscesses. Biofilm formation at the bone-implant-interface induced by the PA14-lux was visualized using FE-SEM versus defective-biofilm formation by flgK-lux. This could be quantitatively confirmed, by average viable-colony-count of the sonicated implants, 3.77x108CFU/ml versus 3.65x103CFU/ml for PA14-lux and flgK-lux, respectively (p=0.0025; 95%CI: -6.08x108 to -1.45x108). This difference in the ability to persist in the model was reflected significantly on the implant osseointegration with a mean intersection surface 4.1x106μm2 1.99x106 for PA14-lux versus 6.44x106μm2 2.53x106 for flgK-lux and 7.08x106μm2 1.55x106 for non-infected control (p=0.048). Conclusions: To date, the proposed in vivo biofilm-based model is the most clinically representative for GN-PJI since animals can bear weight on the implant and poor osseointegration correlates with biofilm formation. Clinical Relevance: The current model will allow for reliable testing of novel biofilm-targeting therapeutics.