Summary. Coagulase-negative staphylococci, including S. epidermidis, have emerged as the leading pathogens of hospital-acquired biomaterial-related infections. These infections can be clinically indolent and challenging also for diagnostic imaging. In the current model of catheter-related infections, . 68. Ga-labeled Siglec-9 PET/CT imaging was able to detect peri-implant S. epidermidis bone infections. Introduction. Coagulase-negative staphylococci, including S. epidermidis, have emerged as the leading pathogen of nosocomial (hospital-acquired) biomaterial-related infections, including periprosthetic infections and intravascular catheter-related bloodstream infections. Pathogenic S. epidermidis strains exhibit robust attachment to implant surfaces and subsequent biofilm formation. By nature, the clinical picture of periprosthetic S. epidermidis infections can be indolent with vague signs of infection. These infections are also highly challenging for diagnostic imaging and microbiologic studies. Our recent experimental study of . 18. F-FDG-PET/CT confirmed that subacute peri-implant S. epidermidis infections, reflecting limited inflammatory reaction, are characterised by low . 18. F-FDG uptake. Vascular adhesion protein-1 (VAP-1) is an inflammation inducible endothelial protein, which controls leukocyte migration to sites of inflammation and infection. Siglec-9 is a leukocyte ligand of VAP-1. We hypothesised that . 68. Ga-labeled Siglec-9, developed for PET imaging of inflammation and cancer, could be a novel tracer also for early defection of S. epidermidis peri-implant bone infections. Material & Methods. Thirty adult male Sprague-Dawley rats were randomised into three groups (n=10/group). A clinical intravenous polymer catheter was introduced into the medullary cavity of the left tibia followed by injections of a clinical isolate of S. epidermidis (T-54580, 3 × 10. 8. CFU/mL) and an adjunct sodium morrhuate. In the positive control group, a clinical isolate of S. aureus (52/52A/80, 3 × 10. 5. CFU/mL) with sodium morrhuate was injected. In the negative control group, equal amount of sterile saline was injected via the catheter. The catheter, cut at the level of tibial tuberosity, was left in situ to serve as the implant. Two weeks after surgery, PET imaging with . 68. Ga-DOTA-Siglec-9 was performed with quantitative analysis of the standardised uptake value (SUV) in the region of interests both in vivo and ex vivo. SUV ratio between the operated and contralateral intact tibia was calculated. The presence of infections and the absence of contamination in the negative control group were verified by separate microbiological analyses of bone samples and retrieved implants. The presence of microbial biofilms on catheters was verified ex vivo with fluorescence microscope. Histologic inflammatory reaction was graded using a scoring system. Intergroup differences were tested by means of ANOVA with a post-hoc test. Results. Both staphylococcal strains caused histologically acute osteomyelitic changes. In . 68. Ga-DOTA-Siglec-9 PET/CT imaging of the negative control group, there was a significant difference (29.5%, p<0.001) in the SUV ratio of the operated and contralateral tibia, demonstrating aseptic inflammatory reaction to catheter implantation. The corresponding SUV ratio values were 58.1% in the S. epidermidis group and 41.7% in the S. aureus group. The uptake in the S. epidermidis group was significantly (p=0.009) higher than in the negative control group. Discussion/Conclusion. The animal model was reproducible in creation of culture-positive biomaterial-related infections. . 68. Ga-labeled Siglec-9 PET/CT imaging was able to demonstrate aseptic inflammation in the negative control group and the tracer also detected peri-implant bone infections caused by S. epidermidis

The treatment of chronic osteomyelitis often includes surgical debridement and filling the resultant void with antibiotic-loaded polymethylmethacrylate cement, bone grafts or bone substitutes. Recently, the use of bioactive glass to treat bone defects in infections has been reported in a limited series of patients. However, no direct comparison between this biomaterial and antibiotic-loaded bone substitute has been performed.

In this retrospective study, we compared the safety and efficacy of surgical debridement and local application of the bioactive glass S53P4 in a series of 27 patients affected by chronic osteomyelitis of the long bones (Group A) with two other series, treated respectively with an antibiotic-loaded hydroxyapatite and calcium sulphate compound (Group B; n = 27) or a mixture of tricalcium phosphate and an antibiotic-loaded demineralised bone matrix (Group C; n = 22). Systemic antibiotics were also used in all groups.

After comparable periods of follow-up, the control of infection was similar in the three groups. In particular, 25 out of 27 (92.6%) patients of Group A, 24 out of 27 (88.9%) in Group B and 19 out of 22 (86.3%) in Group C showed no infection recurrence at means of 21.8 (12 to 36), 22.1 (12 to 36) and 21.5 (12 to 36) months follow-up, respectively, while Group A showed a reduced wound complication rate.

Our results show that patients treated with a bioactive glass without local antibiotics achieved similar eradication of infection and less drainage than those treated with two different antibiotic-loaded calcium-based bone substitutes.

Cite this article: Bone Joint J 2014; 96-B:845–50.

Critical size defects in ovine tibiae, stabilised with intramedullary interlocking nails, were used to assess whether the addition of carboxymethylcellulose to the standard osteogenic protein-1 (OP-1/BMP-7) implant would affect the implant’s efficacy for bone regeneration. The biomaterial carriers were a ‘putty’ carrier of carboxymethylcellulose and bovine-derived type-I collagen (OPP) or the standard with collagen alone (OPC). These two treatments were also compared to “ungrafted” negative controls. Efficacy of regeneration was determined using radiological, biomechanical and histological evaluations after four months of healing. The defects, filled with OPP and OPC, demonstrated radiodense material spanning the defect after one month of healing, with radiographic evidence of recorticalisation and remodelling by two months. The OPP and OPC treatment groups had equivalent structural and material properties that were significantly greater than those in the ungrafted controls. The structural properties of the OPP- and OPC-treated limbs were equivalent to those of the contralateral untreated limb (p > 0.05), yet material properties were inferior (p < 0.05). Histopathology revealed no residual inflammatory response to the biomaterial carriers or OP-1. The OPP- and OPC-treated animals had 60% to 85% lamellar bone within the defect, and less than 25% of the regenerate was composed of fibrous tissue. The defects in the untreated control animals contained less than 40% lamellar bone and more than 60% was fibrous tissue, creating full cortical thickness defects. In our studies carboxymethylcellulose did not adversely affect the capacity of the standard OP-1 implant for regenerating bone.