Comments
Abstract
Rationale: Latent infection with Mycobacterium tuberculosis is defined by a positive IFN-γ release assay (IGRA) result in the absence of active tuberculosis. Only few, mostly monocentric studies have evaluated the role of IGRAs to predict the development of tuberculosis in recent contacts in low-incidence countries of tuberculosis.
Objectives: To analyze IGRA results and the effect of preventive chemotherapy on tuberculosis progression rates among recent contacts.
Methods: Results from contact investigations at 26 centers in 10 European countries including testing for latent infection with M. tuberculosis by the QuantiFERON-TB Gold In-Tube (QFT) test or the T-SPOT.TB (TSPOT) were prospectively collected and analyzed.
Measurements and Main Results: Among 5,020 contacts of 1,023 index cases, 25 prevalent secondary cases were identified at screening. Twenty-four incident cases occurred among 4,513 contacts during 12,326 years of cumulative follow-up. In those with a positive IGRA result, tuberculosis incidence was 0.2 (QFT) and 0 (TSPOT) per 100 patient-years when contacts received preventive chemotherapy versus 1.2 (QFT) and 0.8 (TSPOT) per 100 patient-years in those not treated (38 and 37 patients needed to be treated to prevent one case, respectively). Positive and negative predictive values were 1.9% (95% confidence interval [CI], 1.1–3.0) and 99.9% (95% CI, 99.7–100) for the QFT and 0.7% (95% CI, 0.1–2.6) and 99.7% (95% CI, 99.1–99.9) for the TSPOT.
Conclusions: Tuberculosis rarely developed among contacts, and preventive chemotherapy effectively reduced the tuberculosis risk among IGRA-positive contacts. Although the negative predictive value of IGRAs is high, the risk for the development of tuberculosis is poorly predicted by these assays.
IFN-γ release assays (IGRAs) are routinely used for the diagnosis of latent infection with Mycobacterium tuberculosis, but only few, mostly monocentric studies have evaluated the impact of IGRAs to predict the development of tuberculosis in recent contacts in low-incidence countries of tuberculosis.
Progression toward tuberculosis is generally low in contacts of patients with tuberculosis in Western Europe and poorly predicted by either IGRA. The number needed to treat to prevent one case of incident tuberculosis with chemotherapy among IGRA-positive close contacts was 37 with the T-SPOT.TB test and 38 with the QuantiFERON-TB Gold In-Tube test, respectively.
Contact investigation is an important part of the global policy for the control of tuberculosis (1). Especially in geographic areas of the world with a low tuberculosis incidence (e.g., Western Europe), contact investigation is part of the procedures recommended by national guidelines for tuberculosis control (2). Contact investigation detects secondary cases of tuberculosis who developed the disease in the same population. In addition, it aims to identify contact persons latently infected with Mycobacterium tuberculosis who are at risk for later development of tuberculosis and who may benefit from a preventive treatment and/or close clinical surveillance (3).
In healthy individuals who are latently infected with M. tuberculosis direct detection of viable mycobacteria is not possible by current routine microbiologic culture or molecular methods. Therefore, latent infection with M. tuberculosis is ascertained indirectly by the evaluation of the presence of an adaptive immune response to antigens of M. tuberculosis in the absence of active disease (4). Adaptive immune response is classically assayed by the tuberculin skin test (5–7), and more recently by IFN-γ release assays (IGRAs) (QuantiFERON-TB Gold In-Tube [QFT] test; Qiagen, Hilden, Germany) (T-SPOT.TB [TSPOT] test; OxfordImmunotec, Abingdon, UK) (8).
A large number of publications describe the use of IGRAs for the diagnosis of latent infection with M. tuberculosis. However, only a few studies have evaluated their role in predicting the development of tuberculosis in recent contacts in low tuberculosis incidence countries (9–12). Because most of these studies were monocentric, their results may not be representative.
To gain more generalizable data and to ascertain the role of IGRAs in predicting the risk for tuberculosis in recent contacts the Tuberculosis Network European Trials group (www.tb-net.org) (13) performed an international, multicenter, observational cohort study in Western Europe.
Contacts of tuberculosis index cases for whom an IGRA (QFT or TSPOT) was performed as part of the routine screening procedure according to national guidelines (see Table E1 in the online supplement) were consecutively recruited between April 2009 and March 2011 from European healthcare facilities and followed up for the occurrence of tuberculosis until March 2013. Contacts living in the same household as the index patient were defined as “close contacts/relatives.” Contacts with short but intensive contact not living in the same household were defined as “close contacts/not relatives.” Those not belonging to these categories but having contact to an index patient during work, school activities, leisure, or travel for more than 8 hours were defined as “prolonged contacts.” When contact investigation was considered to be necessary by the responsible public health authorities, although the intensity and duration of exposure could not be verified, contacts were defined as “other contacts.”
Data on demographic, epidemiologic, and clinical parameters, data related to the conditions of M. tuberculosis exposure, IGRA test results, and the decision for preventive chemotherapy were electronically recorded on a structured questionnaire, pseudonymized, and transmitted to the coordinating center (Research Center Borstel, Borstel, Germany), where potential inconsistencies or missing entries were assessed. Completion of follow-up data included information whether the patient was still alive or died, was lost to follow-up, was transferred out, or whether she or he developed tuberculosis after 24 months following enrollment in the study. Follow-up information on the occurrence of tuberculosis after initial testing was actively performed by the participating centers; however, the proportion of contacts completing preventive chemotherapy was not recorded.
When tuberculosis cases were notified among the contacts before the median interval between index case notification and contact case notification (i.e., 81 d) they were considered to be prevalent; the others were considered to be incident. The ethics committee of the coordinating center (University of Lübeck, Lübeck, Germany) approved the study.
According to the IGRA results (i.e., positive, negative, and indeterminate) demographic, epidemiologic, and clinical variables were analyzed using relative frequencies (percentages). The risk of development of tuberculosis during the follow-up in patients exposed and not exposed to preventive chemotherapy was measured by computing the incidence density and its 95% confidence intervals (CI).
The diagnostic test performances of QFT and TSPOT test were evaluated by calculating sensitivity, specificity, positive and negative likelihood ratios, and positive and negative predictive values. Logistic regression analyses were performed using different cut-offs to prove the best value for tuberculosis diagnosis for both QFT (0.35, 0.7, 1.0, and 4.0 IU/ml IFN-γ) and TSPOT (5, 9, and 20 spot-forming cells [sfc] per 50,000).
Statistical analysis was performed using STATA13 (StataCorp, College Station, TX).
Twenty-six centers in 10 European countries (see Table E2) were enrolled in the study and collected data on 1,023 index cases and 5,020 contacts (Figure 1). Mean age of the index cases was 41.0 years (SD, 19.5). Five hundred ninety-nine (58.6%) of the index cases were male. The median (interquartile range) number of contacts per index case was two (one to four). For 3,895 (77.6%) of the contacts IGRA testing was performed by QFT and in 1,125 (22.4%) by TSPOT. Among all contacts, 1,367 of 5,020 (27.2%) had a positive IGRA test result (1,068 of 3,895, 27.4%, for QFT; 299 of 1,125, 26.6%, for TSPOT). The proportion of positive test results was highest in the age group 5–14 years (43.1%), among contacts of smear-positive index cases (41.8%) and among close relatives (45.4%). Indeterminate test results were observed in only 27 of 5,020 tests (0.5%) (Figure 1, Table 1).
Figure 1. Flow chart of test results and occurrence of secondary cases with TB in TB contacts included in the study. Secondary cases were defined as prevalent if they were detected within the median interval between index case notification and contact case notification of 81 days found in this study. QFT = QuantiFERON-TB Gold In-Tube test; TB = tuberculosis; TSPOT = T-SPOT.TB test.
[More] [Minimize]| IGRA Negative | IGRA Positive | IGRA Indeterminate | Total [n (%)] | |
|---|---|---|---|---|
| Total, n (%) | 3,626 (72.2) | 1,367 (27.2) | 27 (0.5) | 5,020 (100.0) |
| Sex of contact, n (col %) (row %) | ||||
| Female | 2,027 (55.9) (74.5) | 678 (49.6) (24.9) | 16 (59.3) (0.6) | 2,721 (54.2) |
| Male | 1,599 (44.1) (69.5) | 689 (50.4) (30.0) | 11 (40.7) (0.5) | 2,299 (45.8) |
| Age group, n (col %) (row %) | ||||
| 0–4 yr | 73 (2.0) (76.0) | 23 (1.7) (24.0) | 0 (0.0) (0.0) | 96 (1.9) |
| 5–14 yr | 226 (6.2) (56.6) | 172 (12.6) (43.1) | 1 (3.7) (0.3) | 399 (7.9) |
| 15–34 yr | 1,561 (43.0) (75.9) | 481 (35.1) (23.4) | 15 (55.6) (0.7) | 2,057 (41.0) |
| 35–64 yr | 1,613 (44.5) (73.0) | 585 (42.8) (26.5) | 9 (33.3) (0.4) | 2,207 (44.0) |
| ≥65 yr | 138 (3.8) (58.2) | 97 (7.1) (40.9) | 2 (7.4) (0.8) | 237 (4.7) |
| Unknown | 15 (0.4) (62.5) | 9 (0.7) (37.5) | 0 (0.0) (0.0) | 24 (0.5) |
| Type of contact, n (col %) (row %) | ||||
| Close relatives | 549 (15.1) (53.7) | 464 (33.9) (45.4) | 9 (33.3) (0.9) | 1,022 (20.4) |
| Close not related but intensive | 711 (19.6) (70.3) | 296 (21.7) (29.3) | 5 (18.5) (0.5) | 1,012 (20.2) |
| Prolonged work | 1,792 (49.4) (77.8) | 502 (36.7) (21.8) | 9 (33.3) (0.4) | 2,303 (45.9) |
| Other | 573 (15.8) (84.1) | 104 (7.6) (15.3) | 4 (14.8) (0.6) | 681 (13.6) |
| Unknown | 1 (0.0) (50.0) | 1 (0.1) (50.0) | 0 (0.0) (0.0) | 2 (0.0) |
| Place of birth of contacts, n (col %) (row %) | ||||
| Africa | 120 (3.3) (48.8) | 124 (9.1) (50.4) | 2 (7.4) (0.1) | 246 (4.9) |
| Americas | 219 (6.0) (61.3) | 134 (9.8) (37.5) | 4 (14.8) (1.1) | 357 (7.1) |
| Asia | 299 (8.3) (64.4) | 161 (11.8) (34.7) | 4 (14.8) (0.9) | 464 (9.2) |
| Europe | 2,938 (81.0) (75.6) | 931 (68.1) (24.0) | 16 (59.3) (0.4) | 3,885 (77.4) |
| Oceania | 2 (0.1) (100.0) | 0 (0.0) (0.0) | 0 (0.0) (0.0) | 2 (0.0) |
| Unknown | 48 (1.3) (72.7) | 17 (1.2) (25.8) | 1 (3.7) (1.5) | 66 (1.3) |
| Screening test, n (col %) (row %) | ||||
| QFT | 2,803 (77.3) (71.9) | 1,068 (78.1) (27.4) | 24 (88.9) (0.6) | 3,895 (77.6) |
| TSPOT | 823 (22.7) (73.1) | 299 (21.9) (26.6) | 3 (11.1) (0.3) | 1,125 (22.4) |
| Preventive chemotherapy, n (col %) (row %) | ||||
| Not initiated | 3,127 (86.2) (88.5) | 383 (28.0) (10.8) | 23 (82.1) (0.7) | 3,533 (70.4) |
| Initiated | 499 (13.8) (33.6) | 982 (71.8) (66.1) | 4 (14.8) (0.3) | 1,485 (29.6) |
| Unknown | 0 (0.0) (0.0) | 2 (0.2) (100.0) | 0 (0.0) (0.0) | 2 (0.0) |
| Preventive therapy, n (col %) (row %) | ||||
| Not received | 3,414 (94.2) (85.1) | 574 (42.0) (14.3) | 25 (92.6) (0.6) | 4,013 (79.9) |
| Received | 176 (4.9) (18.1) | 793 (58.0) (81.7) | 2 (7.4) (0.2) | 971 (19.3) |
| Unknown | 36 (1.0) (100.0) | 0 (0.0) (0.0) | 0 (0.0) (0.0) | 36 (0.7) |
| Prevalence TB, n (col %) (row %) | ||||
| Nonprevalent cases | 3,624 (99.9) (72.5) | 1,344 (98.2) (26.9) | 27 (100.0) (0.6) | 4,995 (99.5) |
| Prevalent cases | 3 (0.1) (12.5) | 22 (1.8) (87.5) | 0 (0.0) (0.0) | 25 (0.5) |
The median interval between index case notification and contact case notification was 81 days. Within this interval of contact screening, 25 of 5,020 (0.5%) prevalent cases of tuberculosis were identified. Of those 18 of 23 (78.3%) had symptoms or an abnormal chest radiograph at the time of screening (for two cases information was not available). In addition, 24 of 4,513 (0.5%) cases of secondary incident tuberculosis occurred during the follow-up period, with a range of 84 to 968 days after the screening of the index case (Table 2) among the 89.9% contacts for whom complete follow-up data were available (Figure 1; see Table E3). Thirty-one cases of tuberculosis occurred among 1,022 close relatives (3%) against 18 among 3,998 other contacts (0.5%). Twenty-one cases occurred among 495 contacts aged 0–14 years (4.2%), against 28 among 4,525 older contacts (0.6%).
| Time Since Index Case Notification (d) | Age (yr) | Sex | Country of Origin | Country of Residence | Contact Type | Index Case | PT | QFT (IU/ml) | TSPOT (sfc/250,000) |
|---|---|---|---|---|---|---|---|---|---|
| 84 | 8 | Female | Spain | Spain | Relative | Sputum smear+ | No | 2.17 | — |
| 84 | 43 | Female | UK | UK | Relative | Sputum smear+ | No | 0.38 | — |
| 87 | 21 | Male | Latvia | UK | Work | Sputum smear+ | No | 13.99 | — |
| 99 | 1 | Female | Poland | Poland | Relative | Sputum smear+ | No | 21.94 | — |
| 112 | 14 | Female | UK | UK | Relative | Sputum smear+ | Yes | 12.07 | — |
| 125 | 47 | Male | Nicaragua | Spain | Work | Sputum smear+ | Yes | 11.05 | — |
| 133 | 2 | Male | UK | UK | Close | Sputum smear+ | Yes | 0.24 | — |
| 155 | 56 | Male | Poland | Poland | Close | Sputum smear−/culture+ | No | 5.66 | — |
| 182 | 49 | Male | Switzerland | Switzerland | Relative | Sputum smear+ | No | — | 0 |
| 231 | 11 | Male | China | Spain | Close | Sputum smear+ | No | 3.67 | — |
| 261 | 37 | Female | Sri Lanka | Switzerland | Relative | Sputum smear+ | No | — | 8 |
| 331 | 13 | Female | Pakistan | Spain | Relative | Sputum smear+ | No | 1.46 | — |
| 332 | 11 | Female | Pakistan | Spain | Relative | Sputum smear+ | No | 13.24 | — |
| 343 | 70 | Female | Bangladesh | UK | Relative | Sputum smear+ | No | 0.08 | — |
| 344 | 29 | Female | Poland | Poland | Relative | Sputum smear−/culture+ | No | 21.94 | — |
| 372 | 60 | Male | Poland | Poland | Work | Sputum smear+ | No | 0.8 | — |
| 394 | 15 | Female | Pakistan | Spain | Relative | Sputum smear+ | No | 14.79 | — |
| 400 | 36 | Male | UK | UK | Other | Sputum smear+ | No | 2.045 | — |
| 427 | 31 | Female | Poland | Poland | Relative | Sputum smear−/culture+ | No | 0.52 | — |
| 444 | 26 | Female | Somalia | UK | Other | Sputum smear−/culture− | No | 0.07 | — |
| 476 | 24 | Female | Pakistan | Spain | Relative | Sputum smear+ | No | 3.09 | — |
| 810 | 25 | Male | Eritrea | Switzerland | Relative | Sputum smear+ | No | — | 50 |
| 882 | 17 | Male | Bolivia | Spain | Work | Sputum smear+ | No | 1.08 | — |
| 968 | 8 | Female | France | Spain | Close | Sputum smear+ | No | — | 1 |
Preventive chemotherapy was offered to 1,485 contacts and initiated in 971 (793 of 1,367, 58.0%, with a positive IGRA; 176 of 3,653, 4.8%, with a negative IGRA). Contacts with a negative IGRA result were more likely to receive preventive chemotherapy when they were younger than 15 years of age (P < 0.0001), closely related to the index case (P < 0.0001), or foreign born (P < 0.003).
The progression rate to tuberculosis was generally low (Table 3). Overall 14 of 421 (3.3%) QFT-positive contacts and 2 of 73 (2.7%) TSPOT-positive contacts without preventive chemotherapy developed tuberculosis during follow-up of a median of 2.5 years (interquartile range, 1.9–3.5). Among 481 contacts with a positive QFT test at screening who started preventive chemotherapy, 3 cases (0.6%) of tuberculosis occurred, after 47, 95, and 125 days of therapy, respectively. In contrast, 3 of 2,419 (0.1%) and 2 of 722 (0.3%) cases of tuberculosis occurred among QFT- and TSPOT-negative individuals in the absence of preventive chemotherapy. No cases of tuberculosis occurred during follow-up in QFT- (n = 104) or TSPOT-negative (n = 58) individuals and TSPOT-positive (n = 208) individuals who received preventive chemotherapy.
| Test Result | n | Prophylaxis | TB Cases | Progression Rate (%) | Person Time (yr) | Incidence/100 Patient-Years | Number Needed to Treat | |
|---|---|---|---|---|---|---|---|---|
| QFT | Negative | 2,419 | No | 3 | 0.12 | 6,349.8 | 0.047 | 807 |
| Negative | 104 | Yes | 0 | 0 | 326.4 | 0 | ||
| Positive | 421 | No | 14 | 3.33 | 1,169.1 | 1.198 | 38 | |
| Positive | 481 | Yes | 3 | 0.62 | 1,296.5 | 0.231 | ||
| TSPOT | Negative | 722 | No | 2 | 0.28 | 1,790.1 | 0.112 | 361 |
| Negative | 58 | Yes | 0 | 0 | 316.1 | 0 | ||
| Positive | 73 | No | 2 | 2.73 | 247.8 | 0.807 | 37 | |
| Positive | 208 | Yes | 0 | 0 | 829.7 | 0 |
Although the negative predictive values were 99.9% (95% CI, 99.7–100.0) for the QFT and 99.7% (95% CI, 99.1–99.9) for the TSPOT, positive predictive values were only 1.9% (95% CI, 1.1–3.0) for the QFT and 0.7% (95% CI, 0.1–2.6) for the TSPOT (Table 4). Both IGRAs poorly predicted the development of tuberculosis.
| Value | 95% CI | |
|---|---|---|
| QFT | ||
| Sensitivity, % | 85.0 | 62.1–96.6 |
| Specificity, % | 74.0 | 72.5–75.5 |
| Positive likelihood ratio | 3.3 | 2.7–4.0 |
| Negative likelihood ratio | 0.2 | 0.1–0.6 |
| Positive predictive value, % | 1.9 | 1.1–3.0 |
| Negative predictive value, % | 99.9 | 99.7–100.0 |
| TSPOT | ||
| Sensitivity, % | 50.0 | 8.3–91.7 |
| Specificity, % | 73.6 | 70.8–76.2 |
| Positive likelihood ratio | 1.9 | 0.7–5.1 |
| Negative likelihood ratio | 0.7 | 0.3–1.8 |
| Positive predictive value, % | 0.7 | 0.1–2.6 |
| Negative predictive value, % | 99.7 | 99.1–99.9 |
| n | Incident TB Cases | Ratio | |
|---|---|---|---|
| QFT: IFN-γ concentration, IU/ml | |||
| <0.35 | 2,410 | 3 | 0.001 |
| >0.35 | 1,015 | 17 | 0.017 |
| >0.7 | 796 | 15 | 0.019 |
| >1.0 | 740 | 14 | 0.019 |
| >1.4 | 677 | 13 | 0.019 |
| >2.0 | 615 | 12 | 0.020 |
| >5.0 | 441 | 8 | 0.018 |
| >10.0 | 287 | 7 | 0.024 |
| TSPOT: spot-forming cells | |||
| <5 | 784 | 2 | 0.003 |
| >4 | 305 | 2 | 0.007 |
| >7 | 271 | 2 | 0.007 |
| >10 | 246 | 1 | 0.008 |
| >20 | 204 | 1 | 0.005 |
| >50 | 102 | 1 | 0.010 |
| >100 | 40 | 0 | — |
| >200 | 1 | 0 | — |
The number needed to treat to prevent a case of tuberculosis in case of a positive IGRA was 38 and 37 for QFT and the TSPOT, respectively; in case of a negative IGRA the number needed to treat was 807 and 361 for QFT and the TSPOT, respectively (Table 3).
We ascertained whether the cut-offs for the concentration of IFN-γ in the QFT or of the number of spot-forming cells in the TSPOT that define a positive test result had an influence on the ability to predict tuberculosis (Table 5). The incidence ratios varied between 0.017 and 0.024 when cut-offs between the 0.35 and greater than 10 IU/ml for the IFN-γ concentration in the QFT were tested. The incidence ratio for a QFT test result below 0.35 IU/ml IFN-γ was 0.001.
Similarly, various cut-offs between greater than 4 sfc/250,000 and greater than 50 sfc/250,000 in the TSPOT had comparable incidence ratios for tuberculosis, ranging between 0.007 and 0.01, whereas the incidence ratio for tuberculosis for a cut-off less than 5 sfc/250,000 was 0.003. Logistic regression analysis demonstrated that the current QFT cut-off is the best predictor for progression to active tuberculosis, showing an area under the curve of 0.779 (P < 0.0001). The TSPOT cut-off of 5 sfc/250,000 showed the best area under the curve (i.e., 0.615), although no TSPOT cut-offs were statistically significant (see Table E4).
We evaluated the role of IGRAs in predicting the risk for tuberculosis in contacts in an international, multicenter prospective, observational cohort study performed in Western Europe. The main findings from this study are that the risk of tuberculosis is comparatively low in tuberculosis contacts in this geographic region irrespective of the IGRA test result, that preventive chemotherapy is highly effective to reduce the risk of tuberculosis in contacts with a positive IGRA result, and that IGRAs predict poorly the development of tuberculosis during the first 2 years after screening. Our results highlight the limitations of IGRAs to function as a basis for the decision to initiate preventive chemotherapy in tuberculosis contacts or other populations at risk for the development of tuberculosis in low tuberculosis incidence countries (14–17).
This prospective observation of 5,020 tuberculosis contacts in 26 centers in 10 European countries demonstrated that the proportion of infected contacts, defined by a positive IGRA test result, is approximately one-quarter of the population screened but that the occurrence of tuberculosis during the follow-up time of 2.5 years on average is actually a rare event. Prevalent and incident tuberculosis were observed more frequently among close contacts, among contacts of smear-positive cases, and among contacts younger than age 14 years. More than 96.5% of contacts with a positive IGRA result did not develop tuberculosis, even in the absence of preventive chemotherapy. However, contacts with a positive IGRA result have a relatively higher risk of tuberculosis than contacts with a negative result. Despite the limited impact, immunodiagnostic testing for latent infection with M. tuberculosis is still the most efficient way to reduce the number needed to treat to prevent a case of tuberculosis in contacts (18). The prescription of preventive treatment in contacts decreased the risk of subsequent tuberculosis (from 1.20 to 0.23 per 100 person-years for contacts with a positive QFT and from 0.81 to 0 per 100 person-years for contacts with a positive TSPOT); however, confounding by indication could not be excluded by the study design.
Several studies assessed the proportion of tuberculosis and latent infection with M. tuberculosis among contacts of index cases. A large metaanalysis concluded that the proportion of prevalent cases of tuberculosis among contacts was 4.5% and the proportion of latent infection (defined by a positive tuberculin skin test) was 51.4% (19). This metaanalysis did not evaluate the number of incident cases who were diagnosed after screening. A more recent metaanalysis found similar rates of tuberculosis and latent infection in contacts living in low-income (and high-incidence) countries, with 3.1% and 51.5% prevalent tuberculosis and latent infection, respectively, but lower rates in high-income (and low-prevalence) countries, with 1.4% prevalent cases of tuberculosis and 28.1% latent infection (20). The proportion of prevalent tuberculosis and latent infection was higher among contacts of smear-positive index cases (20). Studies from single centers or regions in countries with low/intermediate incidence of tuberculosis demonstrated comparable rates of prevalent tuberculosis and latent infection among contacts with our study (1.5%/35.0% in Australia [21]; 0.8%/18.1% in Portugal [22]; and 1.0%/27.8% [23] and 2.0%/36.0% in the United States [24], respectively).
We found a similar proportion of prevalent (0.5%) and incident (0.6%) secondary tuberculosis. The distribution of prevalent and incident cases detected in the course of contact tracing depends on the interval definition of prevalence following screening and the observation period. In a recent study from the Netherlands prevalence for secondary cases was defined as the occurrence within 180 days of screening (11). Applying that definition to the secondary cases in our cohort, the number of prevalent cases would have been more than two times the number of incident cases (33 prevalent cases and 16 incident cases). The decrease in the number of incident cases among contacts with a positive IGRA result who received preventive treatment (from 3.2 to 0.4%) is comparable with the decrease observed in a study from the United States (from 1.5 to 0.4%) (23).
The contacts of extra-European origin, born in countries with a prevalence of tuberculosis higher than in Western Europe, had a higher rate of IGRA positivity (up to 50% among Africans), indicating that a large proportion of the positive test results among the contacts may be the consequence of a remote exposure in the country of birth and not the result of the current contact. A study among U.S. army recruits demonstrated that subjects born in a country with a high incidence of tuberculosis had a much higher frequency of positive tuberculin skin test (23.8%) and IGRA (9.5% for QFT and 11.1% for TSPOT) than recruits born in a low-incidence country (1.1., 1.6, and 1.4%, respectively) (25). This highlights the difficulty in the interpretation of test results among migrants, who may represent a large proportion of index cases and contacts in some regions in Europe (26, 27).
This study underlines the fact that investigation of the close contacts of a case of tuberculosis reveals additional cases of active tuberculosis who will benefit from a full course of treatment. Using a higher cut-off for the definition of latent infection with M. tuberculosis did not substantially change the risk for the development of tuberculosis in this study. This is an important finding, which contrasts with previous results from Germany suggesting that raising the cut-off of the QFT could result in an improvement in the prognostic value of this test (9, 12).
Because half of the tuberculosis cases observed among contacts (25 of 49) were detected during the first 81 days following the notification of the index case, contact investigation has an important role in detecting cases who are already active and may further spread M. tuberculosis. For the detection of such cases, the screening with IGRA is of limited sensitivity and specificity and one must rely on the history of complaints, the clinical examination, and the chest radiograph of suspected cases. In some settings, where the contact investigation was performed several weeks or months after the notification of the index case, some incident cases were detected at screening. In such cases the screening with IGRA was not helpful but the detection of tuberculosis relied on clinical and radiologic signs. Screening with IGRAs seems only helpful in contacts without symptoms and with a normal chest radiograph at the time of screening, to identify those with recent latent infection who have the greatest benefit from a preventive therapy. This also highlights the fact that screening contacts for latent infection should be performed soon after the notification of index cases, preferably 8 weeks following the last contact to allow generation of an adaptive M. tuberculosis–specific immune response (28). We suspect that in some cases, late screening was performed because some contacts developed symptoms of tuberculosis in the meantime.
In practice, among contacts with a positive IGRA, the risk of future tuberculosis is low but can be further reduced by preventive treatment. Among contacts with a negative IGRA at screening without additional risk factors for tuberculosis (such as young age or immunosuppression), one may confidently assume a negligible risk of future tuberculosis, unless they are reexposed and infected later. In our study, among five contacts with a negative IGRA result at screening who developed tuberculosis, two were young children (2 and 8 yr) and one adult was immunocompromised. Among the five cases, two adults and one child were diagnosed with tuberculosis between 1 and 3 years after the notification of the index case, so that we cannot exclude another exposure or an additional risk factor.
The risk for tuberculosis observed in this study was lower than in some previous monocentric studies (9) but similar to others (10). Globally, the percentage of cases observed among contacts (1% within 2 years) was lower than the traditionally mentioned figures of 10% (29), half of it during the initial period after screening. The risk observed among contacts with a positive IGRA but no active tuberculosis at screening and who did not receive preventive chemotherapy (3.2% for the QFT and 2.7 for the TSPOT) is similar to those reported before (30). We may speculate that in some studies the population of contacts included more migrants from high-prevalence countries with a corresponding higher rate of latent infection and higher risk of reactivation of tuberculosis (27), whereas in our study most contacts (3,885 of 5,020) were of European origin. A recent study from a single center in the Netherlands reported a global cumulative risk of tuberculosis (prevalent and incident) of 9.5% among 739 infected contacts within 5 years or a risk of incident tuberculosis of 2.4% for contacts with latent tuberculosis infection at screening who did not start preventive chemotherapy (11). Progression rates to tuberculosis in that study were very similar to our data, although the impact of preventive treatment on the risk for tuberculosis was less obvious.
Our results confirm previous studies demonstrating that prescribing preventive treatment in contacts with latent infection with M. tuberculosis decreased the risk of future tuberculosis (31–33), despite a comparatively high number needed to treat to prevent one case. Because we do not know for all individuals if the preventive treatment was really followed until the scheduled end in the contacts who did not develop tuberculosis, this assumption is probably an underestimation and the protective effect may be higher. Only 3 of 24 contacts who developed incident tuberculosis were prescribed a preventive treatment, which was followed for 47, 95, and 125 days, respectively, and was likely not sufficient to offer a significant protective effect. The other 21 contacts who developed tuberculosis did not receive preventive chemotherapy despite a positive IGRA test result. This clearly underlines the fact that persons with a documented tuberculosis contact and a positive IGRA result may benefit from a preventive treatment.
Recent guidelines from the World Health Organization propose to offer a preventive treatment to persons living in countries with a low or medium incidence of tuberculosis who have latent tuberculosis infection and additional risk factors for the development of tuberculosis, such as HIV infection, recent contact with a case of pulmonary tuberculosis, tumor necrosis factor antagonists treatment, dialysis, silicosis, or organ or hematologic transplantation (1, 34). The fact that only 58% of contacts with recent exposure and latent infection with M. tuberculosis received a preventive treatment illustrates poor adherence to previous recommendations. Recent surveys from Germany demonstrated that in some federal states of this country less than 30% of contacts with a positive IGRA test result received preventive chemotherapy (9, 12, 35), whereas in other settings the adherence after prescription may reach up to 80% (36). It is hoped that rates of preventive treatment for latent infection with M. tuberculosis will increase now that new guidelines have been issued by the World Health Organization (1).
This study has several limitations. One is the fact that the procedures for contact investigations followed different national recommendations and were not standardized. Therefore, the results are not fully comparable among the centers and rather represent operational data. Similarly, we could not control for the laboratory procedures and cannot exclude technical differences in the IGRA procedure among the centers. Furthermore, although the centers were instructed to register all contacts for each index case, we cannot exclude a selection bias in that some centers probably tended to include priority contacts with a positive IGRA result. This may have induced an overestimation of the risk of subsequent tuberculosis but does not change the magnitude of the effect of preventive treatment. Although 26 centers in 10 different European countries participated in this study, centers were selected on their agreement to participate in the study and may not be representative for their country or other countries in Western Europe.
The strength of this study is its multicentric and multinational character that allows representativeness beyond single centers or regions, the large population of contacts, and the high proportion of contacts with follow-ups (despite the fact that many contacts were foreign-born migrants who may move from one region to another during the follow-up).
In conclusion, we found that progression toward tuberculosis is generally low in contacts of tuberculosis patients in Western Europe and poorly predicted by either IGRA, and that in contacts without signs of infection this risk is even lower. Although preventive chemotherapy effectively reduced the risk for tuberculosis among IGRA-positive contacts, identification of individuals at risk for the development of tuberculosis could be improved substantially by more predictive biomarkers.
The authors thank Martina Sester, Ph.D., Homburg, Germany, for a critical manuscript review.
Additional Contributors in the Tuberculosis Network European Trials Study Group: Myriam Aeby, Swiss Lung Association Fribourg, Fribourg, Switzerland; Gilda Cuzzi, Enrio Girardi, Valentina Vanini, and Eleonora Vecchi, Department of Epidemiology and Preclinical Research, and Francesco Nicola Lauria and Marco Vecchi, Clinical Department, L. Spallanzani National Institute for Infectious Diseases, Rome, Italy; Maurizio Ferrarese and Alice Repossi, both Regional TB Reference Centre, Villa Marelli Institute, Niguarda Ca'Granda Hospital, Milan, Italy; Silke Gerdes, Public Health Service Hannover, Hannover, Germany; Martina Haller, Swiss Lung Association Aargau, Aarau, Switzerland; Andrea Glaewe and Lenka Krabbe, Center for Clinical Studies, Research Center Borstel, Borstel, Germany; Ole Hilberg, Department of Respiratory Diseases, Aarhus University Hospital, Aarhus, Denmark; Jean-Paul Janssens, Centre Antituberculeux, Geneva, Switzerland; Christof Rübsamen, Public Health Service Wilhelmshaven, Wilhelmshaven, Germany; Stephan Schlösser, Department of Occupational Medicine, University Hospital Bergmannsheil, Bochum, Germany; and Manuel-Angel Villanueva-Montes, S. Neumología, Hospital San Agustín, Avilés, Asturias, Spain.
| 1. | World Health Organization. Guidelines on the management of latent tuberculosis infection. Geneva, Switzerland: WHO; 2014. |
| 2. | Bothamley GH, Ditiu L, Migliori GB, Lange C; TBNET contributors. Active case finding of tuberculosis in Europe: a Tuberculosis Network European Trials Group (TBNET) survey. Eur Respir J 2008;32:1023–1030. |
| 3. | Erkens CG, Kamphorst M, Abubakar I, Bothamley GH, Chemtob D, Haas W, Migliori GB, Rieder HL, Zellweger JP, Lange C. Tuberculosis contact investigation in low prevalence countries: a European consensus. Eur Respir J 2010;36:925–949. |
| 4. | Mack U, Migliori GB, Sester M, Rieder HL, Ehlers S, Goletti D, Bossink A, Magdorf K, Hölscher C, Kampmann B, et al.; C. Lange; TBNET. LTBI: latent tuberculosis infection or lasting immune responses to M. tuberculosis? A TBNET consensus statement. Eur Respir J 2009;33:956–973. |
| 5. | von Pirquet C. Die Allergieprobe zur Diagnose der Tuberkulose im Kindesalter. Wien Med Wochenschr 1907;57:1370–1374. |
| 6. | Mendel F. Die von Pirquetsche Hautreaktion und die intravenöse Tuberkulinbehandlung. Med Klin 1908;(12):402–404. |
| 7. | Mantoux C. L'intradermo-reaction de la tuberculine et son interpretation clinique. Presse Med 1910;(2):10–13. |
| 8. | Pai M, Denkinger CM, Kik SV, Rangaka MX, Zwerling A, Oxlade O, Metcalfe JZ, Cattamanchi A, Dowdy DW, Dheda K, et al. Gamma interferon release assays for detection of Mycobacterium tuberculosis infection. Clin Microbiol Rev 2014;27:3–20. |
| 9. | Diel R, Loddenkemper R, Niemann S, Meywald-Walter K, Nienhaus A. Negative and positive predictive value of a whole-blood interferon-γ release assay for developing active tuberculosis: an update. Am J Respir Crit Care Med 2011;183:88–95. |
| 10. | Haldar P, Thuraisingam H, Patel H, Pereira N, Free RC, Entwisle J, Wiselka M, Hoskyns EW, Monk P, Barer MR, et al. Single-step QuantiFERON screening of adult contacts: a prospective cohort study of tuberculosis risk. Thorax 2013;68:240–246. |
| 11. | Sloot R, Schim van der Loeff MF, Kouw PM, Borgdorff MW. Risk of tuberculosis after recent exposure: a 10-year follow-up study of contacts in Amsterdam. Am J Respir Crit Care Med 2014;190:1044–1052. |
| 12. | Geis S, Bettge-Weller G, Goetsch U, Bellinger O, Ballmann G, Hauri AM. How can we achieve better prevention of progression to tuberculosis among contacts? Eur Respir J 2013;42:1743–1746. |
| 13. | Giehl C, Lange C, Duarte R, Bothamley G, Gerlach C, Cirillo DM, Wagner D, Kampmann B, Goletti D, Juers T, et al. TBNET - Collaborative research on tuberculosis in Europe. Eur J Microbiol Immunol (Bp) 2012;2:264–274. |
| 14. | Sester M, van Leth F, Bruchfeld J, Bumbacea D, Cirillo DM, Dilektasli AG, Domínguez J, Duarte R, Ernst M, Eyuboglu FO, et al.; TBNET. Risk assessment of tuberculosis in immunocompromised patients. A TBNET study. Am J Respir Crit Care Med 2014;190:1168–1176. |
| 15. | Dorman SE, Belknap R, Graviss EA, Reves R, Schluger N, Weinfurter P, Wang Y, Cronin W, Hirsch-Moverman Y, Teeter LD, et al.; Tuberculosis Epidemiologic Studies Consortium. Interferon-γ release assays and tuberculin skin testing for diagnosis of latent tuberculosis infection in healthcare workers in the United States. Am J Respir Crit Care Med 2014;189:77–87. |
| 16. | Slater ML, Welland G, Pai M, Parsonnet J, Banaei N. Challenges with QuantiFERON-TB Gold assay for large-scale, routine screening of U.S. healthcare workers. Am J Respir Crit Care Med 2013;188:1005–1010. |
| 17. | Gutsfeld C, Olaru ID, Vollrath O, Lange C. Attitudes about tuberculosis prevention in the elimination phase: a survey among physicians in Germany. PLoS ONE 2014;9:e112681. |
| 18. | Diel R, Wrighton-Smith P, Zellweger JP. Cost-effectiveness of interferon-gamma release assay testing for the treatment of latent tuberculosis. Eur Respir J 2007;30:321–332. |
| 19. | Morrison J, Pai M, Hopewell PC. Tuberculosis and latent tuberculosis infection in close contacts of people with pulmonary tuberculosis in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Infect Dis 2008;8:359–368. |
| 20. | Fox GJ, Barry SE, Britton WJ, Marks GB. Contact investigation for tuberculosis: a systematic review and meta-analysis. Eur Respir J 2013;41:140–156. |
| 21. | Dobler CC, Marks GB. Risk of tuberculosis among contacts in a low-incidence setting. Eur Respir J 2013;41:1459–1461. |
| 22. | Mendes MA, Gaio R, Reis R, Duarte R. Contact screening in tuberculosis: can we identify those with higher risk? Eur Respir J 2013;41:758–760. |
| 23. | Anger HA, Proops D, Harris TG, Li J, Kreiswirth BN, Shashkina E, Ahuja SD. Active case finding and prevention of tuberculosis among a cohort of contacts exposed to infectious tuberculosis cases in New York City. Clin Infect Dis 2012;54:1287–1295. |
| 24. | Marks SM, Taylor Z, Qualls NL, Shrestha-Kuwahara RJ, Wilce MA, Nguyen CH. Outcomes of contact investigations of infectious tuberculosis patients. Am J Respir Crit Care Med 2000;162:2033–2038. |
| 25. | Mancuso JD, Tribble D, Mazurek GH, Li Y, Olsen C, Aronson NE, Geiter L, Goodwin D, Keep LW. Impact of targeted testing for latent tuberculosis infection using commercially available diagnostics. Clin Infect Dis 2011;53:234–244. |
| 26. | Kik SV, Franken WPJ, Arend SM, Mensen M, Cobelens FGJ, Kamphorst M, van Dissel JT, Borgdorff MW, Verver S. Interferon-gamma release assays in immigrant contacts and effect of remote exposure to Mycobacterium tuberculosis. Int J Tuberc Lung Dis 2009;13:820–828. |
| 27. | Pareek M, Bond M, Shorey J, Seneviratne S, Guy M, White P, Lalvani A, Kon OM. Community-based evaluation of immigrant tuberculosis screening using interferon γ release assays and tuberculin skin testing: observational study and economic analysis. Thorax 2013;68:230–239. |
| 28. | Lee SW, Oh DK, Lee SH, Kang HY, Lee CT, Yim JJ. Time interval to conversion of interferon-gamma release assay after exposure to tuberculosis. Eur Respir J 2011;37:1447–1452. |
| 29. | Comstock GW, Livesay VT, Woolpert SF. The prognosis of a positive tuberculin reaction in childhood and adolescence. Am J Epidemiol 1974;99:131–138. |
| 30. | Rangaka MX, Wilkinson KA, Glynn JR, Ling D, Menzies D, Mwansa-Kambafwile J, Fielding K, Wilkinson RJ, Pai M. Predictive value of interferon-γ release assays for incident active tuberculosis: a systematic review and meta-analysis. Lancet Infect Dis 2012;12:45–55. |
| 31. | Ferebee SH. Controlled chemoprophylaxis trials in tuberculosis: a general review. Bibl Tuberc 1970;26:28–106. |
| 32. | Landry J, Menzies D. Preventive chemotherapy. Where has it got us? Where to go next? Int J Tuberc Lung Dis 2008;12:1352–1364. |
| 33. | Leung CC, Rieder HL, Lange C, Yew WW. Treatment of latent infection with Mycobacterium tuberculosis: update 2010. Eur Respir J 2011;37:690–711. |
| 34. | Goletti D, Sanduzzi A, Delogu G. Performance of the tuberculin skin test and interferon-γ release assays: an update on the accuracy, cutoff stratification, and new potential immune-based approaches. J Rheumatol Suppl 2014;91:24–31. |
| 35. | Robert-Koch-InstitutAnwendung und Akzeptanz der präventiven Behandlung bei Kindern mit Kontakt zu Tuberkulose-Erkrankten. Epidemiol Bull 2013;12:99–102. |
| 36. | Sarivalasis A, Bodenmann P, Langenskiold E, Lutchmaya-Flick C, Daher O, Zellweger JP. High rate of completion of preventive therapy for latent tuberculosis infection among asylum seekers in a Swiss Canton. Swiss Med Wkly 2013;143:w13860. |
*Present address: Servei de Pneumologia, Hospital Universitari Arnau de Vilanova, Lleida, Spain.
†Present address: Molecular Microbiology Department, Hospital Universitari Sant Joan de Déu, University of Barcelona, Barcelona, Spain.
‡A list of additional contributors in the Tuberculosis Network European Trials study group can be found before the beginning of the references.
Supported by the German Ministry of Education and Research (reference 01KI0784), the German Center of Infection Research (C.L.), and the German Center for Lung Research (F.C.R.).
Author Contributions: J.-P.Z. initiated the study. J.-P.Z. and C.L. made substantial contributions to the conception and design of the work and to the acquisition, analysis, and interpretation of data for the work; wrote the draft of the manuscript; critically revised the manuscript for important intellectual content; and gave final approval of the current version to be published. G.S. and S.D. made substantial contributions to the interpretation of data for the work, performed statistical analysis, wrote the manuscript, critically revised the manuscript for important intellectual content, and gave final approval of the current version to be published. M.B. made a substantial contribution to the conception and design of the work and the management of the data, critically revised the manuscript for important intellectual content, and gave final approval of the current version to be published. All other authors made a contribution to the acquisition of the data for the work, critically revised the manuscript for important intellectual content, and gave final approval of the current version to be published. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201502-0232OC on March 12, 2015
Author disclosures are available with the text of this article at www.atsjournals.org.
