Value of serum procalcitonin for the diagnosis of bacterial septic arthritis in daily practice in rheumatology

Introduction/objectivesSeptic arthritis is a diagnostic and therapeutic emergency because of a high morbidity and mortality. Nevertheless, the etiologic diagnosis is often difficult. The aim of our study was to determine if serum procalcitonin was a discriminatory biomarker in case of arthritis of undetermined etiology.MethodPatients were separated in five groups: gouty arthritis, calcium pyrophosphate deposition arthritis, osteoarthritis or post-traumatic arthritis (mechanical arthritis), chronic inflammatory rheumatic arthritis, and septic arthritis. Levels of serum white blood cells, C-reactive protein and procalcitonin were measured.ResultsNinety-eight patients were included: 18 in the gout group, 26 in the calcium pyrophosphate deposition arthritis group, 16 in the mechanical group, 18 in the chronic inflammatory rheumatic group, and 20 in the sepsis group. The area under the receiver operating characteristic curve of white blood cells, C-reactive protein, and procalcitonin levels to diagnose a septic arthritis were 0.69 (IC95% 0.55-0.83), 0.82 (IC95% 0.73-0.91), and 0.87 (IC95% 0.76-0.98) respectively. For a cutoff of 0.5ng/ml, procalcitonin sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio and negative likelihood ratio were 65%, 91%, 65%, 91%, 7.2, and 0.4, respectively. Serum C-reactive protein and procalcitonin levels were correlated, were not different in sepsis or gout groups, and were higher in non-septic arthritis with poly-arthritis than with mono-arthritis (p<0.05).ConclusionsSerum procalcitonin is a useful biomarker in arthritis management with diagnosis performances higher than those of other biomarkers (white blood cells, C-reactive protein).