Should Synovial Fluid Aspirate Guide Treatment Decisions in PJI?
A study from Hospital for Special Surgery compared the pathogen identified in synovial fluid aspirated preoperatively with the pathogen identified in the intraoperative tissue cultures to determine the degree of concordance/discordance between the 2 types of cultures – and the implications for managing patients with periprosthetic joint infection.
When a total hip or total knee arthroplasty patient is diagnosed with a periprosthetic joint infection (PJI), the surgeon typically develops a treatment plan, including decisions about antibiotic therapy, based on the pathogen identified preoperatively in aspirated synovial fluid.
Is that the right approach? Can the surgeon be assured that the synovial fluid will reliably provide the same results as intraoperative tissue cultures?
The answer depends on the infecting organism, according to a study from Hospital for Special Surgery (HSS), recently presented at the 2020 annual meeting of the American Association of Hip & Knee Surgeons and published online ahead of print by The Journal of Arthroplasty.
A conversation between co-authors Alberto V. Carli, MD, FRCSC, an orthoapaedic surgeon specializing in adult reconstruction, and Andy O. Miller, MD, an infectious disease specialist, was the genesis of the study.
“We were discussing the type of antibiotics to utilize in a cement spacer as part of a 2-stage revision surgery for chronic infection,” Dr. Carli said. “Since the antibiotics we were planning to utilize both in surgery and after surgery were based on joint cultures taken before surgery, I asked Andy how often he had to change his post-surgery antibiotic plan due to new bacteria being identified in the intraoperative cultures. When he gave me an estimate, but admitted that he was not entirely sure, we decided that it would be a great study to undertake.”
Utilizing the HSS longitudinal PJI database, the researchers investigated the diagnostic sensitivity, specificity, and predictive performance of synovial fluid aspirated preoperatively versus tissue cultures taken during surgery in total joint arthroplasty (TJA) patients diagnosed with PJI. This included the degree of concordance/discordance between the 2 types of cultures and the types of organisms most likely to be involved in discordance.
They identified 363 primary TJA patients (189 total hip arthroplasty, 174 total knee arthroplasty) who had treated at HSS for PJI between 2009 and 2017. All patients included in the study had a positive preoperative intra-articular synovial fluid sample within 90 days of a revision procedure during which intraoperative tissue cultures were taken.
Patients were divided into 2 categories: Concordant (same bacteria species detected in the preoperative aspirated synovial fluid and in the intraoperative tissue cultures; same antibiotic resistance profile) and Discordant (different bacteria species detected in the preoperative aspirated synovial fluid and in the intraoperative tissue cultures; different antibiotic resistance profile). The Discordant group was further divided into Discordant Similar and Discordant Different, depending on whether the different organisms could or could not be eradicated by the same antibiotic.
The researchers found that in about three quarters of cases, the organism identified at initial PJI diagnosis was the same as the organism identified in the intraoperative cultures, with culture concordance in 279 patients (76.8%) and culture discordance in 84 patients (23.1%)
Methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-sensitive Staphylococcus aureus had the highest rates of concordance at 100%, with the strongest ability to predict intraoperative culture results. Cutibacterium acnes (C. acnes) had the lowest concordance and a poor diagnostic performance, with low sensitivity, specificity, positive predictive value, and negative predictive value.
The researchers also found that Staphylococcus epidermidis and C. acnes were the most commonly co-infecting organisms in the Discordant Similar group, while Enterococcus species most commonly grew independent from the synovial fluid aspiration and became polymicrobial on tissue culture in the Discordant Different group.
“This study showed that tissue cultures collected in surgery can identify new or completely different bacteria compared with cultures taken before surgery,” Dr. Carli said. “The new bacterial results from the intraoperative cultures have the potential for changing the types of antibiotics we treat patients with after surgery and can even change our estimate for how successful our treatment will be.
“Therefore, the main takeaway from our study is that surgeons should be meticulous in ensuring they collect tissue cultures in infected cases, regardless of what information they have about the infection before surgery. The results can have serious implications for their patients.”
The researchers also found that “in very specific types of infections, such as those involving MRSA, there was remarkable consistency between the culture results collected before surgery and those collected in surgery,” Dr. Carli said. “This means that if a patient has a MRSA infection and is systemically ill, giving antibiotics right away to prevent possible life-threatening sepsis will not compromise intraoperative cultures, as they will most likely yield MRSA as well.”
First author K. Keeley Boyle, MD, received the 2020 James A. Rand Young Investigator’s Award at the AAHKS annual meeting for her work on this study.
Boyle KK, Kapadia M, Henry M, Miller AO, Carli AV. The James A. Rand Young Investigator’s Award. Are Intraoperative Cultures Necessary if the Aspiration Culture Is Positive? A PJI Concordance Study. (Paper 20). Presented at the 30th AAHKS Annual Meeting, November 5-8, 2020, Dallas, Texas.
Boyle KK, Kapadia M, Chiu Y-F, Khilnani T, Miller AO, Henry MW, Lyman S, Carli CV. Are Intraoperative Cultures Necessary If the Aspiration Culture Is Positive? A Concordance Study in Periprosthetic Joint Infection. J Arthroplasty . 2021 Feb 2;S0883-5403(21)00113-3. doi: 10.1016/j.arth.2021.01.073. Online ahead of print.