Objective: We aimed to determine whether IP-10 and RANTES plasma levels can be used in diagnosis and monitoring of pulmonary tuberculosis (PTB).
Methods: Plasma levels of cytokines/chemokines were measured using a Bio-Plex® multiplex cytokine assay system in a cohort containing 457 clinically suspected PTB patients including a training set (n = 41)and two independent test sets A (n = 242) and B (n = 174).
Results: Plasma levels of IP-10 and RANTES were significantly higher in PTB patients than healthy controls' in both training and independent test sets (P < 0.05). Compared with other combinations, the combination of IP-10 and RANTES had the best performance with an AUC of 1.0 in training set. The performance characteristic of this model was successfully validated in independent test set A although this combination only resulted in a slightly improvement of AUC value in independent test set B. Plasma IP-10 and RANTES levels were weakly and positively correlated with blood glucose concentrations. Moreover, IP-10 levels were positively correlated with CRP and ESR in PTB patients. Furthermore, in response to therapy, both IP-10 and RANTES levels significantly decreased over the period of 6 months (P < 0.001).
Conclusions: Taken together, combination of IP-10 and RANTES could be potentially used as diagnostic and monitoring biomarker in PTB management.
Keywords: Biomarkers; Diagnosis; IP-10; Pulmonary tuberculosis; RANTES.
Copyright © 2018 Elsevier Ltd. All rights reserved.