Introduction

Open fractures are common, and infection a frequent complication. There is still uncertainty regarding the urgency of initial treatment. The majority of animal studies indicate that early irrigation and debridement reduces infection; unfortunately, these studies often do not involve antibiotics. Clinical studies indicate that the timing of initial debridement does not affect the infection rate. These studies are observational and fraught with confounding variables. The purpose of this study was to control for these variables using an animal model incorporating both systemic antibiotics and surgical treatment.

Infection is a common complication of severe open fractures and compromises bone healing. The present standard of care is a two-stage approach comprising of initial placement of antibiotic-impregnated PMMA beads to control infection followed later by bone grafting. Although the systemic antibiotics and PMMA/antibiotic beads control the infection initially, there are often residual bacteria within the wound. After grafting and definitive closure, the implanted graft is placed in an avascular defect and could function as a nidus for infection. Bioactive porous polyurethane (PUR) scaffolds have been shown to improve bone healing by delivering recombinant human bone morphogenetic protein-2 (BMP-2) and reduce infection by delivering antibiotics. The release kinetics of the BMP-2 were an initial burst to recruit cells and sustained release to induce the migrating cells. The Vancomycin (Vanc) release kinetics were designed to protect the graft from contamination until vascularisation by having an initial burst and then remaining over the MIC for Staph. aueus for two months. In this study, PUR+BMP-2+Vanc scaffolds were first tested in a non-infected critical size rat femoral segmental defect and was found to perform comparably to PUR+BMP-2, thus indicating that Vanc did not hinder bone healing. PUR+BMP-2+Vanc scaffolds were subsequently evaluated in an infected critical size rat femoral segmental defect. The dual delivery approach resulted in significantly more new bone formation and infection control than both PUR+BMP-2 and the collagen+BMP-2 treatments. These data indicate that the dual-delivery strategy effectively protects the graft from infection during wound healing and regenerates more bone in contaminated defects. This moderately osteoconductive bone graft is capable of being injected, provides a more sustained release of BMP-2 than the collagen sponge, and can release antibiotics for over 8 weeks. The dual-delivery approach may improve patient outcomes of open fractures by protecting the osteoinductive graft from colonization until vascularization occurs.

Open fractures are common, and infection a frequent complication. There is still uncertainty regarding the urgency of initial treatment. The majority of animal studies indicate that early irrigation and debridement reduces infection; unfortunately, these studies often do not involve antibiotics. Clinical studies indicate that the timing of initial debridement does not affect the infection rate. These studies are observational and fraught with confounding variables. The purpose of this study was to control for these variables using an animal model incorporating both systemic antibiotics and surgical treatment.

This study used a segmental defect rat femur model contaminated with Staphylococcus aureus and treated with a 3 day course of systemic cefazolin (5 mg/Kg 12 hourly) and surgical treatments, both of which were initiated independently at 2, 6 and 24 hour time points. After 14 days bone and hardware was harvested for separate microbiological analysis.

These results show that the earlier systemic antibiotic treatment or surgery is initiated. When antibiotics are started at 2 hours, delaying surgical treatment from 2 to 6 hours significantly increases infection (p=0.047). However, delaying surgery to 24 hours increases infection, but not significantly (p=0.054). The timing of antibiotics had a more significant effect on the proportion of positive samples than earlier surgery. At the 2 and 6 hour treatments, the p value was 0.004 and for the 6 and 24 timings it was 0.003.

Surgery and antibiotics at 2 hours completely eradicates the bacteria, but surgical delay for 6 hours appears to allow the bacteria to form non-susceptible colonies. Delaying antibiotics to 6 or 24 hours had a profound detrimental effect on the infection rate regardless of timing of surgery. These findings are consistent with the concept that bacteria progress from a vulnerable planktonic form to a treatment-resistant biofilm.

Purpose: Previous work demonstrated that negative pressure wound therapy (NPWT) resulted in less Pseudomonas aeruginosa than standard wet-to-dry (WTD) dressings in a complex orthopaedic wound model. Staphylococcus aureus is more clinically relevant in open fractures, and is the most prevalent bacteria in osteomyelitis. The purpose of this study is to determine if S. aureus responds similarly to P. aeruginosa when treated with NPWT.

Methods: A complex musculoskeletal wound was created on the hindlimb of 20 goats and contaminated with S. aureus (lux) bacteria. The bacteria are genetically engineered to emit photons, allowing for quantification with a photon-counting camera system. The wounds were débrided and irrigated with 9 L of normal saline using gravity flow irrigation 6 hours after inoculation. Goats were assigned to two different treatment groups: a control group using WTD dressing changes and an experimental group using NPWT. The wounds were débrided and irrigated every other day for 6 days. Bacteria within the wounds were quantified both before and after each débridement.

Results: There was no difference between treatment groups in amounts of bacteria in the wound at all time points (p≥0.37).

Conclusion: Previous work demonstrated that NPWT resulted in a significant and clinically relevant reduction of P. aeruginosa at all time points in a similar model. We presume that NPWT was effective because it created an environment that allowed the body to ward off this “opportunistic” gram negative. However, as shown in this study, S. aureus is less affected by NPWT and persists within the wound.

Background: Despite aggressive debridement, thorough irrigation, systemic antibiotics, and staged treatment, many open fractures still become infected. A graft that can promote bone regeneration and prevent infection could decrease complications. Polyurethane (PUR) scaffolds have previously been shown in separate studies to be nontoxic, osteoconductive, can promote bone growth by delivering BMP, and prevent infection by the sustained release of an antibiotic. This scaffold can deliver both BMP and vancomycin simultaneously; the purpose of this study is to determine if the co-delivery of the antibiotic inhibits bone formation.

Methods: Using an established critical size defect rat femur model, the amount of bone formation created by PUR scaffolds containing low and high doses of rhBMP-2 (2.4 μg and 22.4 μg respectively) and 0.8 mg vancomycin (8% of graft by weight) were compared to scaffolds that contained rhBMP-2 without antibiotics. After 4 weeks, the femurs were harvested and bone growth was assessed using microCT.

Results: There was no significant difference in bone growth between the groups that had the high dose of rhBMP-2. Surprisingly, the scaffolds that had the low dose of rhBMP-2 and vancomycin promoted more bone formation than scaffolds that had rhBMP-2 and no antibiotics.

Conclusions: The addition and co-delivery of vancomycin to the scaffolds did not inhibit bone growth. The addition of vancomycin to the PUR scaffolds may have altered the release kinetics of the rhBMP-2; this may explain the increase of bone formation in this group. This study demonstrates that incorporation of a therapeutic and a clinically-relevant level of vancomycin does not inhibit bone formation. These results suggest that a dual delivery bone graft has potential to reducing complications associated with open fractures.