Introduction Interferon-gamma release assays (IGRAs) have been promoted as having a key role in contact tracing for tuberculosis (TB). In immunosuppressed patients T-SPOT is considered to be the test of choice for detection of latent TB infection. We describe a contact tracing exercise carried out on a Clinical Haematology Unit (CHU) where 168 patients had significant nosocomial exposure to a case of smear-positive TB.

Methods A 29-year old female homeless patient with Hodgkin's lymphoma was diagnosed with smear positive pulmonary tuberculosis (TB). In the 6 months prior to diagnosis she had attended the CHU on 31 occasions including several prolonged inpatient stays. The Incident Management Team initiated a retrospective exercise to identify those at risk in this vulnerable population.

Results 485 haematology patients were subdivided into three cohorts of risk, as well as 69 significant staff contacts. 168 patients considered at highest risk were contact traced and offered screening. 149 of these 168 patients underwent T-SPOT testing. 11 patients did not attend and six patients had died from their underlying condition. In two patients it was not performed as they were already receiving anti-TB therapy. One of these patients had no other known exposure prior to the index case. Of the 149 patients tested with T-SPOT, 19 (12.8%) had a positive result, 76 (51.0%) had a negative result, and 54 (36.2%) had an indeterminate result (negative result with a failed positive control). All those with positive and indeterminate results were followed up in clinic. No further cases of TB were identified. Due to the large number of indeterminate results further T-SPOT testing of additional cohorts was not conducted.

Conclusions T-SPOT testing is considered to have superior sensitivity to tuberculin skin testing in certain immunosuppressed populations, where indeterminate T-SPOT test results have been reported to be infrequent. However, the number of indeterminate results in our study was significantly higher than previously reported in other immunosuppressed groups. This episode calls into question the stated utility of T-SPOT testing in this cohort of patients with haematological conditions. More research is required on the efficacy of IGRAs in the severely immunosuppressed.