Plasma Viscosity Has a Role in the Diagnosis of Prosthetic Joint Infection After Total Knee Arthroplasty.

BACKGROUND

The diagnosis of prosthetic joint infection (PJI) is challenging because no single test has consistently demonstrated an adequate discriminative potential. The combination of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) with adequate thresholds is well established. This study sought to investigate the role of plasma viscosity (PV) in the diagnosis of PJI following painful total knee arthroplasty.

METHODS

The medical notes, and hematological and microbiology results of 310 patients who underwent revision for a painful total knee arthroplasty were evaluated. Infection was confirmed using Musculoskeletal Infection Society criteria in 102 patients (32.9%), whereas 208 patients (67.1%) were classified as noninfected. Serum investigations including ESR, CRP, and PV were analyzed using receiver observer curves and optimal cutoff points identified.

RESULTS

There was a strong correlation between PV and both ESR and CRP. The area under curve was 0.814 for PV and 0.812 for ESR. Statistical analysis showed noninferiority of PV as compared to ESR in diagnosing PJI. A PV value of ≥ 1.81 mPa.s. had the best efficiency of 82.1%. Combining a CRP ≥ 13.5 mg/L with a PV ≥ 1.81 mPa.s. in a serial test approach yielded the highest specificity of 97.9% and positive likelihood ratio of 22.8. Sensitivity was 47.9% and a negative likelihood ratio of 0.53.

CONCLUSION

PV is noninferior to ESR in diagnosing PJI. Its use is justified in clinical practice. It is cheaper, quicker, more efficient, and not influenced by hematocrit levels or medication. In this cohort, a PV value ≥ 1.81 mPa.s. would be an adequate cutoff to diagnose PJI in combination with CRP ≥ 13.5 mg/L.