Background In a previous study we demonstrated the potential usefulness of the procalcitonine serum measure in order to diagnose septic arthritis. This utility is quite important on the differential diagnosis of knee arthritis in patients with known gout or other inflammatory joint disease which can express as a monoarthritis.

Objectives Due to that, this study aims to determine sensibility, specificity, positive predictor value and negative predictor value of procalcinonine sereum measure (PCT) in patients with knee arthritis for the diagnosis of septic arthritis.

Methods We reviewed the registries of patients who consulted due to knee acute monoarthritis between 2013 and 2015 in which a PCT was obtained. Registries were grouped according to the final diagnosis (gouty arthritis only (group I) and septic arthritis with or without gout (group II). Based on a previous study, PCT value of 1.47ng/mL was considered the cut point. Validation tests were applied fixed to demographic and clinical specific scenarios.

Results Registries of 121 patients with gout and 47 patients with septic arthritis (SA) were included. From the 47 patients with SA, 9 were previously diagnosed by gout. All diagnosed were based on guidelines and clinical recomendations for SA and gout (Coakley et al. Rheumatology, 2006 and Zhang et al. Ann Rheum Dis, 2006).

Using the cut point of 1.47 ng/mL, test validation results were as follows: sensibility of 87.2%, specificity of 92.5%, PPV of 82.0% and NPV of 94.9%. Excluding patients with less than 48 hours of onset, there were 44 registries of patients with SA and 83 with gout. In this scenario the results of the validation tests were as follows: Sensibility of 93.1%, specificity of 96.3%, PPV of 93.1% and NPV of 96.38%. Excluding all patients with body temperature above 37.5°C there remained 11 with SA and 81 with gout. In this scenario the results of the validation tests were as follows: Sensibility of 72.7%, specificity of 95.1%, PPV of 66.6% and NPV of 96.2%.

Conclusions This is the first study aimed to validate previous observations about the usefulness of PCT determination in patients with acute knee monoarthritis. Altough the global results in the validation test are a bit inferior that our original observations in a shorter pilot study in terms of sensibility, the NPV remains over 95% in scenarios where the differential diagnosis could be harder to establish such as recent onset flares or cases without an associated febrile syndrome.

Disclosure of Interest None declared