Sahar Nejat describes for SIIC his article published in Acta Paediatrica 105(6):671-675, June 2016

Estocolmo, Suecia (special for SIIC)
During the last 25 years, Sweden has received increasing numbers of migrants from countries where tuberculosis (TB) is endemic. We previously showed that the incidence of TB among foreign-born children in Stockholm, Sweden, varies widely depending on background country, from 450/105 person-years among Somalians to 2/105 among those from Eastern Europe/former Soviet Union. We knew from a previous survey that the prevalence of positive Mantoux tuberculin skin tests (TSTs) did not differ much between migrants from these countries (27% vs. 21%), and wanted to find out if interferon-gamma release assays (IGRAs) were better correlated to TB. Such knowledge would inform planning for future screening.

New entrants were screened with a Mantoux TST in primary care or school health departments, and those with an induration = 10 mm were referred to our pediatric TB clinic at Astrid Lindgren Children's Hospital, Karolinska University Hospital, Stockholm, Sweden. We clinically examined and obtained a chest X-ray of all the children, and evaluated most of them with an interferon-gamma release assay (IGRA), usually QuantiFERON-TB Gold® (QFT^®). We used T-SPOT.TB^® in the youngest children and if the QFT gave an indeterminate result. If a child with a Mantoux reading > 15 mm came from a high TB incidence country, and therefore had a high likelihood for TB infection, QFT was often not obtained. Conversely, if a child had only a weakly positive Mantoux test, was BCG vaccinated, came from a country where TB is not highly endemic and there were no other risk factors for TB exposure, he or she could be considered uninfected without obtaining a QFT. We classified the children into country groups depending on WHO estimates of TB incidence in these countries. Group I contained children from Somalia, Group II children from other high TB incidence (> 100/105) countries, Group III medium incidence (20-100/105, mainly Middle East and Northern African) countries, Group IV eastern Europe/former Soviet Union countries, where TST negative school children are frequently revaccinated with BCG. We used our previous knowledge on TST positivity to estimate the population from which the referred children were drawn. Those with clinical or radiological signs of TB disease were admitted to the hospital to obtain material for mycobacterial cultures, and additional examinations were performed if indicated.

During 2008-2014, our TB clinic received 943 TST positive referrals from new entrant TB screening, with 325, 415, 109 and 94 in Groups I-IV, respectively. An IGRA was obtained in 557, (138, 255, 84 and 80 in Groups I-IV). 327 (59%) were negative (36%, 62%, 68% and 80% in Groups I-IV). There were only three QFTs with an indeterminate result. We used T-SPOT.TB in these and another eight children of less than three years of age. The estimated population-level prevalence of QFT positivity was 18.8%, 8.4%, 2.9% and 4.2% in Groups I - IV, which was significantly correlated to the occurrence of TB in foreign-born Stockholm children from the same countries. TST positivity showed no such correlation. Of the 943 referred children, we considered 417 (44%) uninfected. We diagnosed 506 with LTBI, of whom 428 (85%) started preventive therapy, usually with isoniazid for 6 months. We found 20 cases of TB disease, of which 6 confirmed by culture. During the study period, another 10 new entrants presented in primary care or emergency clinics with symptomatic TB before being screened. Of these, 8 were culture confirmed.

As expected, the rate of QFT positivity was high in TST positive children from high TB incidence countries, whereas it was low in those from the Middle East, Northern Africa, Eastern Europe, and the former Soviet Union. Indeed, our results argue against TST screening of children from these latter regions, since the positive predictive value was low (20%-30%). The selection bias described above as to which children were tested with a QFT incurs a possibility that we underestimated the population rates of QFT positivity for Groups I-II. However, our results are consistent with results from other studies.

In conclusion, our study supports the use of IGRAs for TB screening I high-resource settings. Based on our findings, from 2015 TB screening with QFT is offered to those from high incidence countries, and to those who come from other countries but have additional risk factors for TB exposure. It is important to remember that the use of IGRAs, in addition to being costly, demands high quality blood sampling, transport, and laboratory facilities. Sensitivity is not 100% and may be lower in low-resource settings. Recently infected cases may be missed, and severe cases of active TB may present within weeks after immigration.