Martin McNally

Fracture-related infection (FRI) is a common and devastating complication of orthopedic trauma in all settings. Data on the microbiological profile and susceptibility of FRI to antibiotics in low-income countries are scarce. Therefore, this study aimed to investigate the microbial patterns and antimicrobial susceptibility of FRI in a sub-Saharan African setting in order to provide guidance for the formulation of evidence-based empirical antimicrobial regimens. We conducted a retrospective analysis of patients treated for FRI with deep tissue sampling for microbiological culture from January 2016 to August 2023 in four tertiary-level hospitals in Yaoundé, Cameroon. There were 246 infection episodes in 217 patients. Cultures were positive in 209 (84.9%) cases and polymicrobial in 109 (44.3%) cases. A total of 363 microorganisms from 71 different species were identified, of which 239 (65.8%) were Gram-negative. The most commonly isolated pathogens were Staphylococcus aureus (n = 69; 19%), Enterobacter cloacae (n = 43; 11.8%), Klebsiella pneumoniae (n = 35; 9.6%), Escherichia coli (n = 35; 9.6%), and Pseudomonas aeruginosa (n = 27; 7.4%). Coagulase-negative staphylococci (CoNS) were isolated in only 21 (5.9%) cases. Gram-negative bacteria accounted for the majority of the infections in early (70.9%) and delayed (73.2%) FRI, but Gram-positive bacteria were prevalent in late FRI (51.7%) (p < 0.001). Polymicrobial infections were more frequent in the early (55.9%) and delayed (41.9%) groups than in the late group (27.6%) (p < 0.001). Apart from Staphylococcus aureus, there was no significant difference in the proportions of causative pathogens between early, delayed, and late FRI. This study found striking resistance rates of bacteria to commonly used antibiotics. MRSA accounted for 63% of cases. The most effective antibiotics for all Gram-positive bacteria were linezolid (96.4%), vancomycin (92.5%), clindamycin (85.3%), and fucidic acid (89.4%). For Gram-negative bacteria, only three antibiotics displayed a sensitivity >50%: amikacin (80.4%), imipenem (74.4%), and piperacillin + tazobactam (57%). The most effective empirical antibiotic therapy (with local availability) was the combination of vancomycin and amikacin or vancomycin and imipenem. In contrast to the literature from high-resource settings, this study revealed that in a sub-Saharan African context, Gram-negative bacteria are the most common causative microorganisms of FRI. This study revealed striking resistance rates to commonly used antibiotics, which will require urgent action to prevent antimicrobial resistance in low and middle-income countries.

We report microbiological results from a cohort of recurrent bone and joint infection to define the contributions of microbial persistence or replacement. We also investigated for any association between local antibiotic treatment and emerging antimicrobial resistance. Microbiological cultures and antibiotic treatments were reviewed for 125 individuals with recurrent infection (prosthetic joint infection, fracture-related infection, and osteomyelitis) at two UK centres between 2007 and 2021. At re-operation, 48/125 (38.4%) individuals had an organism from the same bacterial species as at their initial operation for infection. In 49/125 (39.2%), only new species were isolated in culture. In 28/125 (22.4%), re-operative cultures were negative. The most commonly persistent species were Staphylococcus aureus (46.3%), coagulase-negative Staphylococci (50.0%), and Pseudomonas aeruginosa (50.0%). Gentamicin non-susceptible organisms were common, identified at index procedure in 51/125 (40.8%) and at re-operation in 40/125 (32%). Gentamicin non-susceptibility at re-operation was not associated with previous local aminoglycoside treatment (21/71 (29.8%) vs. 19/54 (35.2%); p = 0.6). Emergence of new aminoglycoside resistance at recurrence was uncommon and did not differ significantly between those with and without local aminoglycoside treatment (3/71 (4.2%) vs. 4/54 (7.4%); p = 0.7). Culture-based diagnostics identified microbial persistence and replacement at similar rates in patients who re-presented with infection. Treatment for orthopaedic infection with local antibiotics was not associated with the emergence of specific antimicrobial resistance.