RESUMEN
Abstract Background There is an obvious need for more prompt and specific biomarkers of bacterial infections in generalized pustular psoriasis patients. Objective The aim of this study was to evaluate the diagnostic properties and define appropriate cut-off values of procalcitonin and C-reactive protein in predicting bacterial infection in generalized pustular psoriasis patients. Methods Sixty-four generalized pustular psoriasis patients hospitalized from June 2014 to May 2017 were included in this retrospective study. The values of procalcitonin, C-reactive protein, details of infection, and other clinical parameters were analyzed. Results Receiver operating characteristic curve analysis generated similar areas (p = 0.051) under the curve for procalcitonin 0.896 (95% CI 0.782-1.000) and C-reactive protein 0.748 (95% CI 0.613-0.883). A cut-off value of 1.50 ng/mL for procalcitonin and 46.75 mg/dL for C-reactive protein gave the best combination of sensitivity (75.0% for procalcitonin, 91.7% for C-reactive protein) and specificity (100% for procalcitonin, 53.8% for C-reactive protein). Procalcitonin was significantly positively correlated with C-reactive protein levels both in the infected (r = 0.843, p = 0.040) and non-infected group (r = 0.799, p = 0.000). Study limitations The sample size and the retrospective design are limitations. Conclusions The serum levels of procalcitonin and C-reactive protein performed equally well to differentiate bacterial infection from non-infection in generalized pustular psoriasis patients. The reference value of procalcitonin and C-reactive protein applied to predicting bacterial infection in most clinical cases may not be suitable for generalized pustular psoriasis patients. C-reactive protein had better diagnostic sensitivity than procalcitonin; however, the specificity of procalcitonin was superior to that of C-reactive protein.