Does pre-operative sampling predict intra-operative cultures and antibiotic sensitivities in knee replacements revised for infection?: a study using the NJR dataset
Holleyman, R.J., Deehan, D.J., Charlett, A. et al.Knee
Purpose
This study examined a cohort of primary knee arthroplasties revised for peri-prosthetic infection to (1) determine whether the microorganisms isolated by sampling (up to 6 months pre-operatively) are representative of the microorganisms encountered at the time of knee revision, (2) determine whether the antibiotic sensitivity profiles of matching organisms correlate and (3) determine whether the proportion of organism and antibiotic sensitivity matches is influenced by the type of infection, timing of sample collection and the type of microorganism isolated.
Methods
From the National Joint Registry database for England and Wales, a consecutive series of primary knee arthroplasties performed between 2003 and 2014 that went on to have a revision for peri-prosthetic infection were identified. Each case was then linked to a national microbiology database held by Public Health England. Following data linkage, intra-operative and pre-operative cultures were identified and compared in a group of 75 patients.
Results
Pre-operative genus matched intra-operative genus in 56 of 75 cases (75 %). Of those 56, the corresponding antimicrobial sensitivities matched in 37 (66 % of cases). Overall, 37 of 75 cases (49 %) matched for both microorganism and antimicrobial sensitivity. The proportion of matches was highest in the pure Staphylococcal genus infections (genus match 88 %, sensitivity match 62 %, genus and sensitivity match 55 %) and lowest in those patients with mixed organism infections (genus match 29 %, sensitivity match 14 %).
Conclusion
Pre-operative joint sampling obtained by either aspiration or tissue biopsy is advocated in cases where peri-prosthetic infection is suspected and provides the only means of determining infecting organism prior to revision. Overall, only half of all pre-operative cultures matched completely the findings of intra-operative samples in terms of both the microorganism(s) encountered and their corresponding antimicrobial sensitivity profile. Clinicians should be vigilant regarding the possibility of the patient acquiring further microorganisms or the development of new antibiotic resistance by the time of revision, especially in historic pre-operative cultures. Broad spectrum antibiotic therapies should therefore be employed until final microorganisms and antimicrobial sensitivities are confirmed from intra-operative samples.
Level of evidence
IV.
Link to article