Mathematical modeling of the expected effectiveness of empirical antibiotic therapy for peri-prosthetic infection

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Periprosthetic infection (PJI) is a serious problem in orthopedic practice. In the case of an unknown pathogen, empirical antibiotic therapy is based on local epidemiological data. The aim of this study is to calculate the expected effectiveness of empirical antibiotic therapy based on microbiological monitoring of PJI pathogens using mathematical modeling. Material and methods. A retrospective analysis of 379 bacterial isolates isolated in the PJI of the hip, knee and shoulder joints (2018–2024) was carried out. Microbiological identification and sensitivity testing were performed using standard methods. The indicator of expected empirical eff ectiveness (IEEE) of antibiotics was calculated. Results . The incidence of coagulase-negative staphylococci (CoNS) was 39.6%, Staphylococcus aureus was 28.5%, gram-negative bacteria was 10.3%, Streptococcus was 8.7%, and Enterococcus was 5.3%. Antibiotics with consistently high IEEE (≥ 80% in all periods) were identifi ed: vancomycin, linezolid, and tigecycline; with an average IEEE (50–79%) — include ciprofl oxacin/levofl oxacin, clindamycin, tetracycline, rifampicin, aminoglycosides, trimethoprim-sulfamethoxazole, and meropenem; with a low IEEE (≤ 49%) — include erythromycin, aztreonam, and tobramycin. A negative trend of decreasing the sensitivity of CoNS to fl uoroquinolones (from 70.0 to 47.3%; p =0.014), co-trimoxazole (from 82.0 to 55.9%; p =0.007) was revealed. The optimal combination for the starting empiric therapy is the scheme «vancomycin+meropenem » with the estimated calculated effi ciency ≥ 87.4% against more than 90% of the pathogens of PJI. Сonclusion. Mathematical modeling of the effectiveness of antibiotic therapy based on local data on pathogens and their sensitivity is a promising tool for optimizing empiric therapy for PJI.