Tuberculosis Case Detection and Guideline Adherence among Child Contacts in Switzerland: A Retrospective Observational Study Open Access

Children exposed to a patient with tuberculosis (TB) disease, are at risk of TB infection and TB disease. Younger children are particularly vulnerable and at increased risk for rapid progression to severe forms of TB disease [1‒3]. Most European countries include contact investigations in their national TB strategy and follow the recommendations of the World Health Organization (WHO) [4‒7]. Timely investigation of children contacts allows the detection of TB infection, subclinical or early symptomatic TB disease, and the initiation of treatment to prevent the progression to TB disease and death [3].

In countries with a low TB incidence like Switzerland (TB disease incidence 2019: 5.1/100,000 inhabitants), exposure of a child to a TB index case is a rare event [8]. With decreasing incidence rates, the experience of health care providers in the management of TB infection and disease is at risk of waning [9]. A recently published survey among health care providers in Germany, Austria, and Switzerland showed that there is substantial heterogeneity in management of children exposed to a TB index case and concluded that ongoing education and up-to-date national guidelines are key to ensure the quality of care [10].

So far, no study is available that describes TB infection and disease detection in paediatric contacts in Switzerland and the adherence to current guidelines. The aim of this study was to describe the management of children exposed to a TB index case and the proportion of children with TB infection and disease detected through contact investigations in Switzerland. Further, we aimed to describe the timely course of the examination, guideline adherence in diagnostic pathways contacts, and to assess the risk factors for development of TB infection and disease in children contacts.

This retrospective observational study describes contact investigations and their results for children exposed to TB index cases in Switzerland. In most regions of Switzerland, contact investigations are coordinated through the local offices of the Swiss Lung Associations (SLA). Overall, in 18/26 cantons (representing 46% of the Swiss population in 2020), TB contacts are registered in RespiGo, the electronic database of the SLA. For this study, we extracted data of all children aged ≤16 years registered between January 01, 2019 and July 31, 2021. Contacts who did not show up for examination or without a final diagnosis were excluded. The extracted data included the results of the screening, with immunodiagnostic test results and findings in chest radiography, and the prescribed treatment.

The Swiss guidelines recommend that children at the age of 0–4 year who were exposed to TB receive a prompt examination (within 2 weeks after the notification of the index case) with an immunodiagnostic test such as a tuberculin skin test (TST) or an interferon-γ release assay (IGRA) (QuantiFERON or T-SPOT.TB) and a chest radiograph [11]. In case of normal findings, a prophylactic treatment with isoniazid is recommended until the second examination 2 months after last contact was done. For children aged 5–11 years, a prompt first examination with an immunodiagnostic test is recommended followed by a second examination 2 months after last contact. For asymptomatic children aged 12–16 years, a first immunodiagnostic test is recommended 2 months after last contact. Children with an abnormal immunodiagnostic test should be clinically evaluated and examined with a chest radiograph. The procedure for screening and the management of contacts by age groups as defined in the Swiss Guidance are reproduced in the supplementary Figure 1 (for all online suppl. material, see https://doi.org/10.1159/000534362). The current national Swiss guidelines do not specify which immunodiagnostic test should be done, but in general, three testing approaches are used (1) IGRA alone, (2) TST alone, or (3) a TST followed by an IGRA as confirmatory test.

According to the national Swiss guidelines, children are defined to have TB infection if a positive TST at cut-off >5 mm (usually 2 TU of RT23 tuberculin), or a positive IGRA (QuantiFERON or T-SPOT.TB) test is present [11]. In case of a positive TST and negative IGRA, the children were classified, according to the final diagnosis of the treating physician for this study. Children were defined to have TB disease based on the diagnosis of the treating physician. For treatment of TB disease, the standard regimen is a triple or quadruple treatment with isoniazid, rifampicin, pyrazinamide, and ethambutol. For children with TB infection, a treatment with daily isoniazid for 9 months, rifampicin for 4 months, or a combination of isoniazid and rifampicin for 3 months is recommended. Prophylactic treatment with isoniazid is recommended for children aged 0–4 years at least until the second examination 2 months after last contact excludes a TB infection or disease. The analysis for this study was stratified by age groups 0–4 years, 5–11 years, and 12–16 years in accordance with the recommended different screening approaches and diagnostic pathways for these age groups.

Contacts are investigated in Switzerland if they have had at least 8 h of close contact with the sputum smear positive index case or >40 h of close contact with a sputum smear negative index case with pulmonary TB. For this study, the intensity of the contact was categorised into three groups: 8–40 h of contact in the same household or settings, >40 h of contact in the same household or settings, or very close contact as sharing the bedroom. Index cases were grouped according to their contagiousness in sputum smear positive and sputum smear negative cases (all index cases were confirmed by culture or molecular assay).

Comparisons between groups were made with a Mann-Whitney U test for continuous variables and a χ² test for comparisons of proportions. To assess risk factors associated with TB transmission, we compared children with TB infection or disease to uninfected children in uni- and multivariable logistic regression models. Statistical analysis was done with R (version 4.0.5 using R studio version 2021.09.0).

A total of 1,557 contacts of 111 TB index cases (all drug-sensitive and under treatment for one to 2 weeks) were notified to the local TB offices of the SLA. Of those, 401 were children aged ≤16 years. Fourteen were excluded because of incomplete data and seven because they did not show up for the examination. Overall, 380 (95%) children were included in the analysis. Of those, 7 (2%) had TB disease, 35 (9%) had TB infection, and 338 (89%) were not infected (Table 1). The median age of included children was 13 (interquartile range [IQR]: 8 to 15) years and 180 (52%) were male. Child contacts with TB disease were with a median age of 3 (IQR: 1 to 6) years younger compared to those with TB infection (median age 10 [IQR: 7 to 15] years) and TB non-infected children (median age 13 [IQR: 8 to 15] years, p = 0.008) (Fig. 1). Overall, 194 (51%) children had contact to the TB index case at school and 145 (38%) at home. Further characteristics of the included children are shown in Table 1.

Of the 61 children aged 0–4 years, 49 (80%) were screened for TB with an immunodiagnostic test within 2 weeks after contact (26 [43%] by TST only, 18 [30%] by TST and IGRA, and 5 [8%] by IGRA only) and 45 (73%) were again screened with an immunodiagnostic test in a second examination (Table 2, Fig. 2). A chest radiograph was performed in 40 (65%) children; of which 21 (34%) were done in within 2 weeks and 19 (31%) later than 2 weeks after exposure to the index case. A chest radiograph was performed in 15 (25%) children at the second examination after 2 months. Overall, in 21 (34%) children, no chest radiography was performed.

Five children (8%) were diagnosed with TB disease (four of them within the first 2 weeks) and four with TB infection (7%). Of the children with TB disease, three (60%) had an abnormal chest radiograph and two (40%) had a positive immunodiagnostic test at their first examination. Of the children with TB infection, three (75%) had a positive immunodiagnostics test at the first examination and one child converted from negative IGRA in the first examination to a positive IGRA in the second examination.

Of the 108 children aged 5–11 years, 63 (64%) were screened for TB with an immunodiagnostic test in the first examination after contact (16 [16%] by TST, 10 [10%] by TST and IGRA, and 37 [38%] by IGRA) (Table 2; Fig. 2). In a second examination, 73 (75%) were screened with an immunodiagnostic test. A chest radiography was done in 15 (15%) and 3 (3%) children in the first and second examinations, respectively.

One child (1%) was diagnosed with TB disease and 16 (16%) with TB infection, all of whom received an appropriate treatment. Of the 16 children diagnosed with TB infection, 12 (75%) had a positive immunodiagnostic test (4 [25%] a positive TST only, 3 [19%] a positive TST and IGRA, and 5 [31%] a positive IGRA only) in the first examination, two had a positive immunodiagnostic test in the second examination whereby the first examination was not reported or not done, and the remaining two had no immunodiagnostic test results reported.

Of the 221 children aged 12–16 years, 88 (40%) were screened for TB with an immunodiagnostic test within 2 weeks after contact (Table 2; Fig. 2). Two months after the last contact, 134 (61%) were screened with immunodiagnostic test (63 [29%] by TST only, 2 [1%] by TST and IGRA, and 69 [31%] by IGRA only). A chest radiograph was done in 3 (1%) and 5 (2%) children within 2 weeks and at 2 months after last contact, respectively.

One child (1%) was diagnosed with TB disease and 15 (17%) with TB infection. Of the 15 children diagnosed with TB infection, 7 (47%) had a positive immunodiagnostic test promptly after exposure and 8 (53%) had a positive immunodiagnostic test in the examination 2 months after last contact. The child with TB disease was treated accordingly. All but two children diagnosed with TB infection were treated accordingly.

We compared the examinations done in children with TB contacts with the national Swiss guidelines for all three defined age groups (Table 2; Fig. 2). In total, 51 (83%) children aged 0–4 years were examined with an immunodiagnostic test and/or chest radiograph with 2 weeks after contact. Of the children in this age group, 40 (66%) received a prophylactic treatment until the second examination or were treated for TB infection or disease. Among the contacts aged 5–11 years, 63 (64%) were tested with an immunodiagnostic test in a first examination and 73 (75%) were tested in a second examination. Among contacts aged 12–16 years, 88 (40%) were tested within 2 weeks after contact, and 134 (61%) were tested 2 months after last contact.

Seven children were diagnosed with TB disease and received TB disease treatment and 35 were diagnosed with TB infection. Of the latter, 33 (94%) received TB infection treatment. Of children 0–4 years, 31/61 (51%) received a prophylactic treatment until TB infection was ruled out at the second examination 2 months after contact. Of the children aged 0–4 years, 21/61 (34%) did not receive any treatment (not prophylactic and not for TB infection).

Children who were diagnosed with TB disease or TB infection were further characterised to understand risk factors associated with TB transmission. Of those, the proportion of children exposed to a TB index case in the same household was 100% (7/7) in children with TB disease, 74% (26/35) in children with TB infection, and 33% (112/338) in non-infected children (Fig. 3). Children with TB disease had the highest rate of contact to a sputum smear-positive index case (7/7, 100%), followed by children with TB infection (32/35, 91%), and non-infected children (179/338, 53%, p < 0.001). The risk factors of TB transmission were estimated by comparison of children with TB infection or disease with non-infected children: in univariable logistic regression, the odds ratio for having TB infection or disease was decreasing with older age, increasing with contact intensity, and increasing with contagiousness of the index case (Table 3). In a multivariable logistic regression, risk for TB transmission was increasing with contact intensity (Fig. 3) and with contagiousness of the index case, after correction for age (Table 3).

In this study, a considerable proportion of children who had contact to a TB index case were diagnosed with TB disease or infection. Of the children diagnosed with TB disease, most were younger than 5 years of age. A comparison of the examinations done with the recommendations in the Swiss guidelines revealed deviations from those guidelines in all age groups.

Only few studies describing TB contact investigations in the European low-incidence settings are available [3]. The proportion of children diagnosed with TB disease and infection through contact investigations was comparable to other study results in low-incidence countries. The proportion of children exposed to TB index case who developed TB disease and infection were 4.6% and 12.8% in household contacts below the age of 5 years in a Finnish study [12], 3.5% and 17.1% in a Swedish study [13], and 1.4% and 13% in a French study [14], respectively. Part of the heterogeneity of these results might be explained by the use of different cut-offs for a positive TST. Studies done in countries with a high TB incidence tend to observe higher proportions of TB disease and infection among contacts [15]. Particularly among children below the age of 5 years, the proportion of children contacts with TB disease was estimated at 19% (95% confidence interval: 8.4–37.4%) in a systematic review and meta-analysis [3]. In our study, most children diagnosed with TB infection or disease were exposed in the same household, and the risk of transmission increased with the closeness of the contact and contagiousness of the index case. This is in line with findings of other studies.

Most children with TB disease in our study were diagnosed at the first examination, shortly after the last TB contact. This underlines the former observation that the risk of TB disease in exposed children under the age of 5 years is high, and that the progression to TB disease occurs relatively rapidly [1, 3, 8]. Our study thus supports the recommendation of a prompt screening of children exposed to TB for incipient TB disease or TB infection. Treatment of TB disease is always recommended in children with signs or symptoms of TB disease. Children with TB infection are also recommended to be treated, given the relatively high risk of progression to TB disease with estimated 5-year cumulative TB disease incidence of 15% in untreated children [2]. In children aged 0–4 years without signs of TB infection or disease, a prophylactic treatment is recommended until TB infection or disease can be ruled out 2 months after their last contact with the index case. A second immunodiagnostic test after 2 months is recommended since TST and IGRA require a few weeks to become positive after exposure [16‒18].

In our study, we noted some deviations from the recommendations in the national Swiss guidelines in the management of children exposed to a TB index case. Only about one-third of the children aged 0–4 years were examined with a chest radiograph at the first examination (within 2 weeks) and only about 80% had an immunodiagnostic test performed. About a third of these children did not receive a prophylactic treatment with isoniazid until TB infection could be ruled out at a second examination 2 months after last contact. Among the children aged 5–11 years, only two-thirds were tested promptly. About 40% of the children aged 12–16 were tested promptly (although this is not recommended in Switzerland) and only 60% were retested 2 months after last contact. Since no exact date of the immunodiagnostic test was captured in our database, we cannot exclude that the results of some immunodiagnostic tests performed 2 months after the last contact were registered as a prompt examination.

Most children 0–4 years of age were tested with a TST and in case of a positive result, confirmation was sought with an IGRA. Half of the children 5–11 years of age and the majority of children 12–16 years of age were tested with an IGRA. IGRA is considered more specific in children with previous BCG immunisation but sensitivity is comparable as shown in a recent multicentre study from Europe [19]. As known from other Swiss studies, most of the children with TB disease in Switzerland are of foreign origin [8] and many of them have been vaccinated with Bacillus Calmette-Guérin, which is associated with the risk of a false-positive TST [20]. Therefore, the use of an IGRA is recommended in those cases [21].

Surprisingly, many children did not receive a chest radiograph as part of the screening. The chest radiograph is essential to screening as it is required to exclude TB disease and may be abnormal in children with no or very few symptoms [21, 22].

According to the latest Global TB Report of the WHO, there is still a need to catch up in the implementation of contact investigation and prophylactic treatment, as only one quarter of children below the age of 5 years exposed to TB receive a prophylactic treatment and less than 2% of contacts over the age of 5 years [23]. Policies on the pragmatic management of TB infection are still not adequately implemented [24]. A French study confirmed that a large proportion of cases of TB in children could have been prevented by timely screening after exposure and appropriate prescription of a preventive treatment [25]. According to a mathematical modelling study, implementation of systematic contact investigation for exposed children and prescription of prophylactic treatment if indicated could prevent more than 150,000 cases of TB disease and 100,000 deaths from TB in children below the age of 15 years each year, worldwide [26]. However, in high-TB incidence setting, this approach may be difficult because contact investigations require extensive human and laboratory resources that may not be available. Recently, a call for action was issued by the WHO, emphasizing that prevention of TB disease by screening for TB infection and adequate treatment is necessary to speed up the desired decline of TB incidence [27].

Our study has some limitations. Despite the fact that the registration and initiation of the contact investigation process are coordinated by the local TB offices some examinations were delegated to general practitioners, who may not have sent back the results of the examination. This explains a part of our missing results in the data. In addition, some of the children were lost to follow-up, since they did not attend scheduled appointments. As all cases of contagious TB disease are notified to the local TB offices in Switzerland, we assume that the studied population is representative for the Swiss settings, despite the fact that only a part of Switzerland was captured by the data in our study. Another limitation of our study is that we do not have a follow-up after more than 2 months. We are thus not able to assess if children with positive or negative immunodiagnostic test results who were not treated developed TB disease later. Contacts with a negative immunodiagnostic test result have a very low risk of developing TB disease during the 2 years following screening [28, 29] and we thus assume that this does not influence our findings. A further limitation of our study is that clinical signs and symptoms, and the results of bacteriological examinations were not available in our database. We are thus unable to state the proportion of bacteriologically confirmed TB cases in our study population.

Screening exposed children for TB disease or infection is key, as children are at high risk of TB infection and, if infected, at high risk of rapid progression to TB disease. Even though our study shows deviations from the recommended guidelines in Switzerland, the overall coverage with contact investigations was better than the worldwide average. However, more effort should be made to evaluate exposed children as comprehensively and timely as possible and ensure that children at risk receive adequate treatment.

The authors thank the staff of the local offices of the Swiss Lung Association who performed the contact investigations.

This study was approved by the ethical commission of the canton of Bern, Project-ID 2022-00692. As the study was re-using existing data without patient contact, no informed consent form was deemed necessary (decision of the ethical commission, Art. 34 HFG, Art. 37-40 HFV).

The authors have no conflict of interest to declare.

No funding was necessary for this study, which was performed within the routine work schedule of the authors.

J.P.Z. initiated the study and wrote the draft of the manuscript, N.G. and N.F. performed the data analysis and contributed to the manuscript, N.G. and J.M.E. prepared and supervised the data collection, and N.R. and O.S. contributed to the writing of the manuscript and literature research. All authors read and approved the final version of the manuscript.

Nathalie Gasser and Nora Fritschi contributed equally to this work.

The data used for the preparation of the annual report and for the planned study are extracted from the routine database used by the local TB offices, either the database RespiGO or other solutions that are used in the cantonal settings. They are under the protection of the SLA and can be accessed only by the computer team and members of the TB competence centre. Further enquiries can be directed to the corresponding author.