Low secondary attack rate after prolonged exposure to sputum smear positive miliary tuberculosis in a neonatal unit

Several neonatal intensive care units (NICU) have reported exposure to sputum smear positive tuberculosis (TB). NICE guidelines give support regarding investigation and treatment intervention, but not for contact definitions. Data regarding the reliability of any interferon gamma release assay (IGRA) in infants as a screening test for TB infection is scarce. We report an investigation and management strategy and evaluated the viability of IGRA (T-Spot) in infants and its concordance to the tuberculin skin test (TST).

We performed an outbreak investigation of incident TB infection in a NICU after prolonged exposure to sputum smear positive miliary TB by an infant's mother. We defined individual contact definitions and interventions and assessed secondary attack rates. In addition, we evaluated the technical performance of T-Spot in infants and compared the results with the TST at baseline investigation.

Overall, 72 of 90 (80%) exposed infants were investigated at baseline, in 51 (56.7%) of 54 (60%) infants, follow-up TST at the age of 6 months was performed. No infant in our cohort showed a positive TST or T-Spot at baseline. All blood samples from infants except one responded to phytohemagglutinin (PHA), which was used as a positive control of the T-Spot, demonstrating that cells are viable and react upon stimulation. 149 of 160 (93.1%) exposed health care workers (HCW) were investigated. 1 HCW was tested positive, having no other reason than this exposure for latent TB infection. 5 of 92 (5.5%) exposed primary contacts were tested positive, all coming from countries with high TB incidences. In total, 1 of 342 exposed contacts was newly diagnosed with latent TB infection. The secondary attack rate in this study including pediatric and adult contacts was 0.29%.

This investigation highlighted the low transmission rate of sputum smear positive miliary TB in a particularly highly susceptible population as infants. Our expert definitions and interventions proved to be helpful in terms of the feasibility of a thorough outbreak investigation. Furthermore, we demonstrated concordance of T-Spot and TST. Based on our findings, we assume that T-Spot could be considered a reliable investigation tool to rule out TB infection in infants.