Diagnostic accuracy and safety of Cy-Tb for detection of tuberculosis infection: A multicentric study from India

Background and objectives Only 10% of the individuals infected with Mycobacterium tuberculosis (MTB) manifest active tuberculosis. Interferon-γ release assay (IGRA) and tuberculin skin test using purified protein derivative (PPD) are used to identify tuberculosis (TB) infection. We aimed to study the comparative performance of Cy-Tb (the SIILTIBCY skin test) and PPD with IGRA as the reference standard. Methods In step-I, 30 TB patients and 32 IGRA-negative health controls; in step-II, 2651 participants aged 1 to 85 yr; and in step- III, 647 household contacts of TB patients aged 1-18 years old were enrolled. Blood for IGRA was drawn immediately prior to intra-dermal administration of Cy-Tb (the SIILTIBCY skin test) or PPD. The size of induration of Cy-Tb ≥5 mm or PPD ≥ 10mm after 48 to 72 h were considered positive. Combined Data from step II and step III were analysed for sensitivity, specificity, agreement, and kappa-coefficient with IGRA as the reference standard using Stata version 14.2. Results The step-I results for Cy-Tb showed 90% specificity, 93.8% specificity, and an agreement of 91.9% with IGRA and a Cohen's kappa coefficient of 0.83. Combined analysis of step-II and step-III data revealed sensitivity of 82.6% (95% CI 79.1-85.8), specificity of 71.1% (95% CI 68.2-73.8), agreement of 74.8%, and kappa of 0.5 for Cy-Tb. PPD showed sensitivity and specificity of 71.0% (95% CI 66.7-75.0) and 71.9% (95% CI 69.0-74·6), respectively with 71.6% agreement and kappa of 0.4 with IGRA. Interpretation and conclusions Performance of Cy-Tb was comparable to that of PPD with IGRA as reference standard. Cy-Tb skin test can be used for diagnosis of TB infection.