Antibiotic-loaded synthetic calcium sulfate beads for prevention of bacterial colonization and biofilm formation in periprosthetic infections

Abstract

Periprosthetic infection (PI) causes significant morbidity and mortality after fixation and joint arthroplasty and has been extensively linked to the formation of bacterial biofilms. Poly(methyl methacrylate) (PMMA), as a cement or as beads, is commonly used for antibiotic release to the site of infection but displays variable elution kinetics and also represents a potential nidus for infection, therefore requiring surgical removal once antibiotics have eluted. Absorbable cements have shown improved elution of a wider range of antibiotics and, crucially, complete biodegradation, but limited data exist as to their antimicrobial and antibiofilm efficacy. Synthetic calcium sulfate beads loaded with tobramycin, vancomycin, or vancomycin-tobramycin dual treatment (in a 1:0.24 [wt/wt] ratio) were assessed for their abilities to eradicate planktonic methicillin-resistant Staphylococcus aureus (MRSA) and Staphylococcus epidermidis relative to that of PMMA beads. The ability of the calcium sulfate beads to prevent biofilm formation over multiple days and to eradicate preformed biofilms was studied using a combination of viable cell counts, confocal microscopy, and scanning electron microscopy of the bead surface. Biofilm bacteria displayed a greater tolerance to the antibiotics than their planktonic counterparts. Antibiotic-loaded beads were able to kill planktonic cultures of 10(6) CFU/ml, prevent bacterial colonization, and significantly reduce biofilm formation over multiple days. However, established biofilms were harder to eradicate. These data further demonstrate the difficulty in clearing established biofilms; therefore, early preventive measures are key to reducing the risk of PI. Synthetic calcium sulfate loaded with antibiotics has the potential to reduce or eliminate biofilm formation on adjacent periprosthetic tissue and prosthesis material and, thus, to reduce the rates of periprosthetic infection.

FIG 1

Preparation of PG-CSH beads: production of the beads from a smooth paste (A) and removal of the set beads from the mold (B).

FIG 2

Repeat modified Kirby-Bauer assays for assessing zones of inhibition (ZOI) of EMRSA-16 NCTC 13143 and S. epidermidis ATCC 35984 over a period of 40 days. Images are representative photographs of the ZOI of the two bacterial strains observed on agar plates at days 1, 3, 5, 10, 20, 30, and 40 of vancomycin-tobramycin-loaded PG-CSH beads. Graphs show the sizes of the ZOI (in cm²) over time, as calculated using ImageJ. Assays were performed in triplicate, and data are expressed as the means of 3 data points with standard error bars.

FIG 3

EMRSA-16 NCTC 13143 biofilm formation over time in the presence of antibiotic-loaded synthetic calcium sulfate beads as determined by CLSM images and CFU counts (per cm²). Data are expressed as the means of 15 data points (5 data points per experimental repeat) with standard error bars. Bars, 25 μm.

FIG 4

Biofilm formation of EMRSA-16 NCTC 13143 on vancomycin- and tobramycin-loaded beads relative to that on unloaded beads over time as determined by SEM. White arrows indicate bacterial colonization and biofilm formation. Bars, 25 μm unless otherwise stated.

FIG 5

S. epidermidis ATCC 35984 biofilm formation over time in the presence of antibiotic-loaded synthetic calcium sulfate beads, as determined by CLSM images and CFU counts (per cm²). Data are expressed as the means of 15 data points (5 data points per experimental repeat) with standard error bars. Bars, 25 μm.

FIG 6

Biofilm formation of S. epidermidis ATCC 35984 on vancomycin- and tobramycin-loaded beads relative to unloaded beads over time as determined by SEM. White arrows indicate bacterial colonization and biofilm formation. Scale bars equal 25 μm unless otherwise stated.

FIG 7

Effects of antibiotic-loaded synthetic calcium sulfate beads with 24-h and 72-h contact times on preexisting biofilms of EMRSA-16 NCTC 13143 and S. epidermidis ATCC 35984. Data are expressed as log reductions of the means of 15 data points (5 points per experimental repeat, with each assay performed in triplicate) relative to those of the unloaded treatment group.