Evaluation of the use of sonication combined with enzymatic treatment for biofilm removal in the microbiological diagnosis of prosthetic joint infection

Abstract

ABSTRACT Sonicating explanted prosthetic implants to physically remove biofilms is a recognized method for improving the microbiological diagnosis of prosthetic joint infection (PJI); however, chemical and enzymatic treatments have been investigated as alternative biofilm removal methods. We compared the biofilm dislodging efficacy of sonication followed by the addition of enzyme cocktails with different activity spectra in the diagnosis of PJI with that of the sonication of fluid cultures alone. Consecutive patients who underwent prosthesis explantation due to infection at our institution were prospectively enrolled for 1 year. The diagnostic procedure included the collection of five intraoperative tissue cultures, sonication of the removed devices, and conventional culture of the sonication fluid. The resulting sonication fluid was also treated with an enzyme cocktail consisting of homemade dispersin B (0.04 µg/mL) and proteinase K (Sigma; 100 µg/mL) for 45 minutes at 37°C. The resulting sonication (S) and sonication with subsequent enzymatic treatment (SE) fluids were plated for aerobic and anaerobic culture broth for 7 days (aerobic) or 14 days (anaerobic). Identification was performed by matrix-assisted laser desorption ionization–time-of-flight mass spectrometry (Bruker). We included 107 patients from whom a prosthetic implant had been removed, among which PJI was diagnosed in 36 (34%). The sensitivity of S alone was significantly greater than that of SE alone (82% vs 71%; P < 0.05). Four patients with PJI were positive after sonication alone but negative after sonication plus enzymatic treatment. The four microorganisms missed after the addition of the enzyme cocktail were Staphylococcus aureus , two coagulase-negative Staphylococci, and Cutibacterium acnes . In conclusion, sonication alone was more sensitive than sonication followed by enzymatic treatment. The combination of these two methods had no synergistic effect; in contrast, the results suggest that the combination of both dislodgment methods affects the viability of gram-positive microorganisms. IMPORTANCE While the potential of sonication and enzymes as biofilm dispersal agents has been previously described, the originality of our work resides in the combination of both methods, which is hypothesized to enhance the ability to remove biofilm and, therefore, improve the microbiological diagnosis of PJI