Abstract
Objectives: IFN-γ release assays (IGRAs) are the most accurate diagnostic tests for latent TB. IP-10 is a chemokine expressed in concert with IFN-γ, but in >100 fold higher levels.
The aim of this study was to investigate if IP-10 in plasma from M.tuberculosis specific antigen stimulated whole blood dried and stored on filter paper was stable and could be used for TB diagnosis.
Methods: In Spain, whole blood from 78 patients with culture confirmed TB, and 109 healthy controls were subjected to Quantiferon (QFT) testing. Following, IP-10 was measured in plasma supernatant using ELISA and 25μ aliquots were dried on Whatman903 filter paper.
Filter paper was sent by normal mail to Denmark where 2 discs were cut from the filter paper using a normal office 6mm hole punch. The amount of IP-10 in filter paper was measured using ELISA.
Results: Median P-10 levels in TB patients were 1,073pg/ml and 27.3ng/ml in filter paper and plasma; and in controls 44pg/ml and 1.0ng/ml, respectively (p<0.0001 for both). The correlation between filter paper and plasma IP-10 was very high (r2=0.93). The Area Under the ROC curve was comparable 0.89-0.91 for both IP-10 tests and IFN-γ and the 3 tests had comparable sensitivity and specificity 74-78% and 100-98%, respectively (all n.s. differences)
Conclusion: Plasma from M.tb. antigen stimulated blood can be dried on filter paper and transported over long distances at ambient temperature before analysis. Compared to the currently available IGRAs, the filter paper version allows for high throughput centralized analysis, and could increase the dissemination of specific tests for LTBI in resource restraint settings where IGRAs as we know them today are too complicated to do.
- © 2011 ERS
