The impact of BCG vaccination on tuberculin skin test responses in children is age dependent: evidence to be considered when screening children for tuberculosis infection

Abstract

Background: Following exposure to TB, contacts are screened to target preventive treatment at those at high risk of developing TB. The UK has recently revised its recommendations for screening and now advises a 5 mm tuberculin skin test (TST) cut-off irrespective of age or BCG status. We sought to evaluate the impact of BCG on TST responses in UK children exposed to TB and the performance of different TST cut-offs to predict interferon γ release assay (IGRA) positivity.

Methods: Children <15 years old were recruited from 11 sites in the UK between January 2011 and December 2014 if exposed in their home to a source case with sputum smear or culture positive TB. Demographic details were collected and TST and IGRA undertaken. The impact of BCG vaccination on TST positivity was evaluated in IGRA-negative children, as was the performance of different TST cut-offs to predict IGRA positivity.

Results: Of 422 children recruited (median age 69 months; IQR: 32-113 months), 300 (71%) had been vaccinated with BCG. BCG vaccination affected the TST response in IGRA-negative children less than 5 years old but not in older children. A 5 mm TST cut-off demonstrated good sensitivity and specificity in BCG-unvaccinated children, and an excellent negative predictive value but was associated with low specificity (62.7%; 95% CI 56.1% to 69.0%) in BCG-vaccinated children. For BCG-vaccinated children, a 10 mm cut-off provided a high negative predictive value (97.7%; 95% CI 94.2% to 99.4%) with the positive predictive value increasing with increasing age of the child.

Discussion: BCG vaccination had little impact on TST size in children over 5 years of age. The revised TST cut-off recommended in the recent revision to the UK TB guidelines demonstrates good sensitivity but is associated with impaired specificity in BCG-vaccinated children.

Keywords: Paediatric Lung Disaese; Respiratory Infection; Tuberculosis.

Figure 1

Receiver operating characteristic curves for tuberculin skin test results of different cut-offs to predict interferon γ release assay positivity. (A) BCG-unvaccinated children <2 years (n=20; AUC=1.00; 95% CI not calculable). (B) BCG-vaccinated children <2 years (n=64; AUC=0.79; 95% CI 0.65 to 0.93). (C) BCG-unvaccinated children aged 2 to <5 years (n=32; AUC=0.97; 95% CI 0.91 to 1.00). (D) BCG-vaccinated children aged 2 to <5 years (n=65; AUC=0.78; 95% CI 0.60 to 0.95). (E) BCG-unvaccinated children aged 5 to <10 years (n=41; AUC=0.95; 95% CI 0.88 to 1.00). (F) BCG-vaccinated children aged 5 to <10 (n=108; AUC=0.92; 95% CI 0.86 to 0.97). (G) BCG-unvaccinated children aged 10 to <15 (n=29; AUC=0.99; 95% CI 0.96 to 1.00). (H) BCG-vaccinated children aged 10 to <15 (n=63; AUC=0.95; 95% CI 0.90 to 1.00).

Figure 1

Receiver operating characteristic curves for tuberculin skin test results of different cut-offs to predict interferon γ release assay positivity. (A) BCG-unvaccinated children <2 years (n=20; AUC=1.00; 95% CI not calculable). (B) BCG-vaccinated children <2 years (n=64; AUC=0.79; 95% CI 0.65 to 0.93). (C) BCG-unvaccinated children aged 2 to <5 years (n=32; AUC=0.97; 95% CI 0.91 to 1.00). (D) BCG-vaccinated children aged 2 to <5 years (n=65; AUC=0.78; 95% CI 0.60 to 0.95). (E) BCG-unvaccinated children aged 5 to <10 years (n=41; AUC=0.95; 95% CI 0.88 to 1.00). (F) BCG-vaccinated children aged 5 to <10 (n=108; AUC=0.92; 95% CI 0.86 to 0.97). (G) BCG-unvaccinated children aged 10 to <15 (n=29; AUC=0.99; 95% CI 0.96 to 1.00). (H) BCG-vaccinated children aged 10 to <15 (n=63; AUC=0.95; 95% CI 0.90 to 1.00).