gms | German Medical Science

Objectives: Besides surgical intervention, the optimal antibiotic therapy for periprosthetic joint infections (PJI) continues to be a subject of controversial discussion. The combination of a locally effective antibiotic spacer therapy with an intravenously administered antibiotic is the gold standard for the therapy of complicated PJIs. Current research focuses on the composition and especially the dosage of the antibiotics used. While therapy used to be cautious regarding possible nephrotoxic concentrations, recent studies suggest the therapeutic option of significantly increasing the concentration in locally effective spacers without incurring systemic side effects. We investigated the concentration of gentamicin- and vancomycin-coated spacers regarding their local and systemic concentration and their effect on kidney function.

Methods: In a prospective study, we report on 43 patients (12 male, 31 female, average age 70 years, age range 25 to 90 years) who underwent prosthesis removal followed by spacer implantation including gentamycin (Palacos^®R+G) or gentamycin+vancomycin (Copal^®G+V) due to periprosthetic infection at the hip, knee, or shoulder joints. To determine the systemic and local antibiotic concentrations, blood samples were taken on the first and second day after the implantation of the spacer system, and tissue fluid was harvested from the end of inserted drain. The concentration levels were measured using the ELISA method (Enzyme Linked Immunosorbent Assay). Possible effects on the patients’ kidney function were examined by determining the GFR and serum creatinine differences from preoperative and postoperative measurements.

Results: The measurements showed that gentamicin concentrations, both locally and systemically, were well below potential toxicity effects at both measurement points (local: M = 38.14 µg/ml (day 1: M = 41.01 µg/ml/ day 2: M = 34.82 µg/ml)/ systemic: M = 0.16 µg/ml (day 1: M = 0.16 µg/ml/ day 2: M = 0.15 µg/ml). No correlation with possible deterioration of kidney function could be demonstrated.

The investigations of the vancomycin concentrations showed similar results. Here, the concentration was below threshold (local: M = 2.58 µg/ml (day 1: M = 3.04 µg/ml/ day 2: M = 2.00 µg/ml)/ systemic: M = 6.03 µg/ml (day 1: M = 6.23 µg/ml/ day 2: M = 5.80 µg/ml). A negative impact on kidney function could also not be demonstrated.

Conclusion: An acute nephrotoxic effect of the applied antibiotics could not be found in our study. Considering our clinical data and further findings from the literature, an increase in the antibiotic concentration in the spacers for locally administered antibiotic therapy is a possible option to optimize the therapy of PJIs. The kidney is not at risk by using these spacers.