The Journal of Arthroplasty, Volume 34, Issue 11, 2744 - 2748

Reevaluating Current Cutoffs for Acute Periprosthetic Joint Infection: Current Thresholds Are Insensitive

Xu, Chi et al.
Hip Knee

Background

Diagnosing acute periprosthetic joint infection remains a challenge. Several studies have proposed different acute cutoffs resulting in the International Consensus Meeting recommending a cutoff of 100 mg/L, 10,000 cell/μL and 90% for serum C-reactive protein (CRP), synovial white blood cell count (WBC), and polymorphonuclear percentage (PMN%), respectively. However, establishing cutoffs are difficult as the control group is limited to rare early aseptic revisions, and performing aspiration in asymptomatic patients is difficult because of a fear of seeding a well-functioning joint arthroplasty. This study (1) assessed the sensitivity of current thresholds for acute periprosthetic joint infection (PJI) and (2) identified associated factors for false negatives.

Methods

We retrospectively reviewed patients with acute PJIs (n = 218), defined as less than 6 weeks from index arthroplasty, treated between 2000 and 2017. Diagnosis of PJI was based on 2 positive cultures of the same pathogen from the periprosthetic tissue or synovial fluid samples. Sensitivities of International Consensus Meeting cutoff values of CRP, synovial WBC, and PMN% were evaluated according to organism type. Multiple logistic regression analysis was performed to determine associated factors for false negatives.

Results

Overall, the sensitivity of CRP, synovial WBC, and PMN% for acute PJI was 55.3%, 59.6%, and 50.5%, respectively. Coagulase-negative Staphylococcus (CNS) demonstrated the lowest sensitivity for both CRP (37.5%) and WBC (55.6%). CNS infection was identified as an independent risk factor for false-negative CRP.

Conclusions

Current thresholds for acute PJI may be missing approximately half of PJIs. Low virulent organisms, such as CNS, may be responsible for these false negatives. Current thresholds for acute PJI must be reexamined.

Download article