Clinical Decision Guidelines
Background: The diagnosis of septic arthritis may not be clear until a culture has resulted. While the sensitivity of the classic value of 50000 WBC in synovial fluid has been called into question, clinicians may need to rely on other diagnostic markers to assist in making a diagnosis. The C-reactive protein has been suggested as a sensitive marker when considering a diagnosis of septic arthritis. One study has suggested a mean CRP value of septic arthritis of 13. Another study suggested a cutoff of 20 with a sensitivity of 92%. Our analysis set out to confirm the usefulness of CRP in aiding a clinician with a patient who may have septic arthritis.
Methods: This study was a single-center, retrospective, chart review at a quaternary academic teaching hospital. Charts were pulled from a two-year period from January 2016 to December 2017 for patients who had a joint aspiration performed in the emergency department. Patients had to be 18 years or older. 179 charts met these criteria and of those 102 had culture results. A positive culture served as a gold standard for septic arthritis. The mean C-reactive protein was recorded for both negative and positive cultures to assess for a variation in these patients.
Results: Of the 102 charts with culture results, 95 had a C-reactive protein value recorded. An overall mean of 5.1 was determined with a mean of 3.9 for culture negative and a mean value of 13.7 for culture positive (p < 0.0001). Using a previously suggested cutoff value of 20, a sensitivity of 39% and a specificity of 95% was obtained. When an arbitrarily chosen cutoff value of 8 was applied, the sensitivity increased to 74% but the specificity decreased to 23%.
Conclusions: This study obtained a median CRP value of 13.7, which is similar to a previously suggested value of 13. The value determined in this study suggests an elevated CRP could assist a clinician in making a diagnosis of septic arthritis. This analysis could not confirm the sensitivity previously proposed with a cutoff value of 20. This suggests a lower cutoff value may be necessary. Future studies should look to determine a highly sensitive and specific cutoff value to increase the usefulness of this marker. Furthermore, randomized clinical control trials need to be conducted to determine if clinical outcomes vary with the use of a CRP cutoff.