INTRODUCTION: Over the last decade, the number of knee arthroplasties performed in the U.S. have tremendously increased and the demand is estimated to grow by 174 percent by 2030. Usually, a total joint arthroplasty is an effective procedure that provides considerable pain relief. However, there has been a large and increasing number of arthroplasty failures mainly due to periprosthetic joint infections (PJI) that have amplified the amount of revision surgeries, which has resulted in increased medical costs and a prolonged rehabilitation process. Due to the increased rates of surgical complications and morbidity associated with revision surgeries, it should be a priority to prolong the lifetime of the primary arthroplasty by preventing infection and increasing the stability of the implant. Bisphosphonates are common medications used to treat osteoporosis by increasing bone density. These medications have also been proposed for off-label use to reduce aseptic loosening and increase implant fixation after arthroplasty. However, perioperative bisphosphonates have previously been identified as a risk factor for PJI and are known to modulate the immune system. As a result, we hypothesize that zoledronic acid (ZA) will increase bacterial burden in a mouse model of PJI.
METHODS: Eight to twelve-week-old C57BL/6 mice were treated on the day of surgery with 500ug/kg of ZA or vehicle in experiment one, or low/medium/high doses (200, 500, 1000ug/kg) or vehicle in experiment two. Additionally, a medium dose of ZA (500ug/kg) or vehicle was administered one week prior to surgery in experiment three. The drug was administered via tail vein injections. For the surgical procedure, the mice were anesthetized via inhalation of isoflurane. An incision over the right knee was made and the distal part of the femur was accessed through a medial parapatellar arthrotomy. A 0.8mm-diameter titanium implant was placed into the distal femur, and the knee was inoculated with 1E+3 CFUs of bioluminescent Staphylococcus aureus or saline. Bacterial burden was monitored with in vivo bioluminescence imaging. To confirm the bioluminescence imaging, bacteria was quantified on the last day of postoperative imaging. The implant and surrounding tissue were harvested for CFU enumeration. Independent sample t-test and two-way ANOVA were used to establish significant differences in the outcome measures.
RESULTS: In experiment one, ZA mice showed significantly higher bacterial burden than controls (Figure 1A; p<0.05). In experiment two, ZA mice showed higher bacterial burden than controls at all time points, with the largest effect observed in the medium dose group (p<0.01) followed by the high dose group (p<0.05). The low dose group was not significantly different from controls. Mean implant and tissue CFU were higher in the ZA group than control mice (Figure 1B, 1C), with CFUs enumerated from 63% (5/8) of implants versus 38% (3/8) of controls (p<0.05). In experiment three, ZA mice showed similar bacterial burden when compared to the control group.
CONCLUSIONS: For the first time, we demonstrate in a validated model of PJI, that perioperative zoledronic acid significantly increases bacterial burden and the rate of PJI compared to controls. This effect is dose dependent, with a possible ceiling effect. The administration of preoperative zoledronic acid did not significantly increase bacterial burden, which may highlight the importance of timing. The ability of zoledronic acid to prevent resorption of the bone by inhibiting osteoclasts may impair healing and extend bacterial contamination due to lack of bone remodeling. Further study is required to describe the effects of dose, bisphosphonate type, timing of administration, and mechanisms of these findings. These preliminary results should be considered when using bisphosphonates in peri-operative arthroplasty patients.