The effect of different antibiotic combinations in calcium sulfate cement on the growth of Cutibacterium acnes and Staphylococcus periprosthetic shoulder infection isolates

Background: Periprosthetic joint infections (PJI) of the shoulder are a devastating complication of shoulder arthroplasty and are commonly caused by Staphylococcus and Cutibacterium acnes . Absorbable calcium sulfate (CS) beads are sometimes used for delivering antibiotics in PJI. This study evaluates the in vitro effect of different combinations of gentamicin, vancomycin, and ertapenem in beads made from CS cement on the growth of C acnes and coagulase-negative Staphylococcus (CNS) strains. Methods: Three strains of C acnes and 5 strains of CNS from clinically proven shoulder PJI were cultured and plated with CS beads containing combinations of vancomycin, gentamicin, and ertapenem. Plates with C acnes were incubated anaerobically while plates with Staphylococcus were incubated aerobically at 37 degrees C. Zones of inhibition were measured at intervals of 3 and 7 days using a modified Kirby Bauer technique, and beads were moved to plates containing freshly streaked bacteria every seventh day. This process was run in triplicate over the course of 56 days. Statistical analysis was conducted using SPSS v. 28 with repeated measures analysis of variance (ANOVA) and pairwise comparisons with Tukey correction. Results: In experiments with C acnes , beads containing ertapenem + vancomycin and vancomycin alone formed the largest zones of inhibition over time (P < .001). In experiments with Staphylococcus , beads containing vancomycin alone formed the largest zones of inhibition over time for all 5 strains (P < .001). Zones of inhibition were 1.4x larger for C acnes than for Staphylococcus with beads containing vancomycin alone. For both C acnes and Staphylococcus , beads containing ertapenem had the strongest initial effect, preventing all bacterial growth in C acnes and almost all growth for Staphylococcus during the first week but dropping substantially by the second week. Beads containing gentamicin alone consistently created smaller zones of inhibition than beads containing vancomycin alone, with vancomycin producing zones 5.3x larger than gentamicin in C acnes and 1.3x larger in Staphylococcus (P < .001). Discussion: These data suggest that for both C acnes and Staphylococcal species, CS beads impregnated with vancomycin were most effective at producing a robust antibiotic effect. Additionally, ertapenem may be a viable supplement in order to create a more potent initial antibiotic effect but is not as effective as vancomycin when used alone. Gentamicin alone was not effective in maintaining consistent and long-term antibiotic effects. These results indicate that amongst the antibiotics currently commercially available to be used with CS, vancomycin is consistently superior to gentamicin in the setting of C. acnes and CNS.