Innovative methods for the diagnosis and treatment of implant-associated infections

Indwelling devices and prosthesis are increasingly used in modern medicine practices. Failures of prosthetic joint devices constitute an important complication. Especially, the management of patients with a septic failure is characterized by difficulties in the diagnosis and by frequent treatment failures, with infection relapses. Thus, the goal of our studies was to establish and evaluate innovative methods for the diagnosis, prophylaxis and treatment of prosthetic infections.
The accurate differentiation between septic and aseptic loosening of the implants is difficult. Nuclear medicine methods are promising, non-invasive procedures aiming to visualize the specific accumulation at sites of infection of an injected radiotracer. In our study, we compared in vitro and in the tissue cage mouse model of implant infection: 99mTc-UBI, 99mTc-ciprofloxacin, 99mTcN-CIPROCS2 and 111In-DTPA-biotin for targeting E. coli and S. aureus. Radiochemical purity of the labeled agents ranged between 94 % and 98 %. Stability in serum was high for all tracers. In vitro binding assays displayed a rapid and non-displaceable bacterial binding for all tested agents. Using the tissue cage mouse model, the tested agents accumulated in vivo in infected sites, being 99mTc-ciprofloxacin and 111In-DTPA-biotin from 4 h p.i. discriminative for both E. coli and S. aureus infections, whereas 99mTc-UBI 29-41 and 99mTcN-CiproCS2 discriminated only for E. coli infected cages. The measured tissue cage to blood (T/NT) ratios remained lower than 3, a finding that in our opinion may constitute a limiting factor for the use of the tested tracers in patients.
Following, we evaluated the potentials for targeting infections of the novel transcobalamin II non-binder 99mTc-labeled derivative of Vitamin B12 (99mTc-PAMA(4)-Cbl). For comparison, we tested in parallel the labeled native vitamin 57Co-Cbl and the 99mTc-PAMA(4)-Cbl derivative. In vitro binding to a S. aureus and an E. coli strain was specific and could be antagonized by addition of unlabeled Cbl. 99mTc-PAMA(4)-Cbl showed lower in vitro binding than 57Co-Cbl to E. coli, but similar binding to S. aureus. In vivo, 57Co-Cbl showed gradual accumulation into the cage fluids, and discriminated only for E. coli infected cages. On the contrary, 99mTc-PAMA(4)-Cbl showed a rapid kinetic, it was rapidly cleared from most tissues and was able to discriminate both S. aureus and E. coli infected from sterile cages at 4 h p.i.. Thus, the new 99mTc-PAMA(4)-Cbl derivative may represent a promising candidate for bacterial imaging in humans.
In addition, we demonstrated the validity of the mouse tissue cage model for screening radiotracers targeting infections. The negative control, 99mTc-DTPA, showed rapid accumulation and clearance from both sterile and infected cages, whereas, the positive control, 67Gallium citrate, accumulated selectively in infected cages between 48 h and 72 h p.i.
Early and accurate detection of methicillin-resistant Staphylococcus aureus (MRSA) is essential in the hospital and the outpatient setting. We established a calorimetry assay for discrimination of MRSA from methicillin susceptible S. aureus (MSSA). The assay consisted of paired heat measurements of batch cultures in pure medium and in medium supplied with cefoxitin. Relative heat was calculated as ratio between the total heat, measured in the presence and absence of cefoxitin. Using a relative heat cutoff of 0.4, 19 of 20 MRSA (95%) and 10 of 10 MSSA (100%) clinical isolates were correctly identified within 5 h. Thus, microcalorimetry may be successfully applied in routine screening for MRSA and potentially be extended to screen resistance patterns of other pathogens and antibiotic agents.
Ga3+ is a semi-metal element competing for iron-binding sites of transporters and enzymes. We investigated the activity of gallium maltolate (GaM), against laboratory and clinical strains of MSSA, MRSA and methicillin susceptible or resistant S. epidermidis (MSSE, MRSE). The MICs of GaM were higher for S. aureus (375-2000 µg/ml) than S. epidermidis (94-200 µg/ml). Minimal biofilm inhibitory concentrations (MBIC) were 3000-≥6000 (S. aureus) and 94-3000 µg/ml (S. epidermidis). In time-kill studies, GaM exhibited a slow and dose-dependent killing mechanism. Sub-inhibitory concentrations of GaM inhibited growth-related heat production measured in a batch calorimeter. The GaM minimal heat inhibitory concentrations (MHIC) correlated well with the MIC values. Thus, GaM exhibited activity against staphylococci, but high concentrations were required. These data supports the potential use of GaM for local application, including treatment of wound infections, MRSA decolonization and implant coating.
Finally, we investigated the efficacy of linezolid, alone and in combination with rifampin (rifampicin), against MRSA in a guinea pig model of foreign-body infection. In vitro, linezolid was bacteriostatic against the tested strain. In time-kill studies, development of rifampin resistance was observed with rifampin alone, but was prevented by the addition of linezolid. After the administration of single intraperitoneal doses, linezolid concentrations into sterile cage fluids remained above the MIC during 12 h. Antimicrobial treatments administered to animals with cage implant infections were given twice daily for 4 days. Linezolid alone reduced planktonic bacteria in cage fluid during treatment. Efficacy in eradication of cage-associated infection was achieved only when linezolid was combined with rifampin, with cure rates being between 50% and 60%. For comparison, the levofloxacin-rifampin combination was tested and demonstrated the highest cure rate (91%). Thus, the linezolid-rifampin combination may be a treatment option for infections caused by quinolone-resistant MRSA.