Disseminated tuberculosis is associated with more false negative interferon gamma release assay results compared to localized tuberculosis

<b>Background:</b> Active tuberculosis has been associated with false negative interferon gamma release assay (IGRA) results. This may relate to a suppression of the interferon gamma response associated with a more extensive Mycobacterium (M.) tuberculosis infection. <b>Objective:</b> To determine whether disseminated M. tuberculosis infection is associated with a higher rate of false negative IGRA results compared to localized tuberculosis. <b>Methods:</b> In a systematic review of published individual patient data we compared IGRA results in patients with disseminated tuberculosis as evident from a diagnosis of miliary and/or central nervous system (CNS) tuberculosis with interferon gamma release assay results in patients with lymphnode tuberculosis. <b>Results:</b> We identified 38 reports from Pubmed, EMBASE and Cochrane Library data bases containing individual patient data on 68 patients, 19 with lymph node tuberculosis and 49 with miliary and/or CNS tuberculosis. There was do difference in mean age, immunosuppressive co-morbidity or medication or confirmation by culture or PCR testing between the two groups. In patients with miliary and/or CNS tuberculosis 26.5% had a false negatie IGRA result compared to 5.2% of patients with lymph node tuberculosis (p&lt;0.05). There was no difference in indeterminate results between groups. <b>Conclusions:</b> Disseminated tuberculosis may be associated with an increased rate of false negative interferon gamma release assay results compared to localized tuberculosis.