INT J TUBERC LUNG DIS 16(12):1594–1599 © 2012 The Union http://dx.doi.org/10.5588/ijtld.12.0389
Risk factors for Mycobacterium tuberculosis infection in Indonesian children living with a sputum smear-positive case M. E. Rutherford,* P. C. Hill,* W. Maharani,† L. Apriani,† H. Sampurno,‡ R. van Crevel,§ R. Ruslami† * Centre for International Health, University of Otago, Dunedin, New Zealand; † Health Research Unit, Faculty of Medicine, Universitas Padjadjaran, Bandung, ‡ Bandung Community Lung Clinic, Bandung, Indonesia; § Department of Medicine, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands SUMMARY S E T T I N G A N D O B J E C T I V E S : Young children living with infectious tuberculosis (TB) cases are at high risk of infection and disease, and screening is recommended. This is rarely conducted in resource-limited settings. Identifying children most at risk of infection may be useful for setting practical screening policies. D E S I G N : Child contacts of smear-positive adult TB patients were invited for Mycobacterium tuberculosis infection and disease screening by symptoms, tuberculin skin test (TST), QuantiFERON®-TB Gold In-Tube assay (QFT-GIT) and chest X-ray. Risk factors for infection were collected using a questionnaire and were calculated separately for TST, for QFT-GIT and for both tests combined. R E S U LT S : Of 304 screened children 145/302 (48%)
were positive using TST, 152/299 (51%) by QFT-GIT and 180/304 (59%) were positive using either or both tests. Positivity for both tests was associated with index case infectivity (acid-fast bacilli [AFB] 3+ vs. AFB 1+: TST OR 2.93, 95%CI 1.59–5.39; QFT-GIT OR 2.28, 95%CI 1.06– 4.90) and exposure (child contact’s parent is the index case: TST OR 7.04, 95%CI 2.23–22.28; QFT-GIT OR 4.30, 95%CI 1.48–12.45). C O N C L U S I O N : M. tuberculosis infection according to either test was high, supporting screening and preventive treatment. Children of smear-positive TB cases who accompany their parents to the clinic should be prioritised for immediate screening. K E Y W O R D S : QuantiFERON®-TB Gold In-Tube assay; tuberculin skin test; child contacts
APPROXIMATELY 180 MILLION children are affected by tuberculosis (TB) disease every year, resulting in several hundred thousand preventable deaths.1 Risk of infection with Mycobacterium tuberculosis is strongly associated with infectious load of and exposure to a TB case.2 Infected children are at high risk of disease progression.3 The World Health Organization (WHO) therefore recommends that all children aged ⩽5 years living with a smear-positive TB patient be screened. Those cleared of disease should be prescribed 6 months of isoniazid preventive therapy (IPT).4 Although screening of children (child contacts) living with a TB patient is incorporated into most national TB programmes, it is rarely conducted in endemic, resource-limited settings.5,6 Investigation of infection levels in child contacts and identifying risk factors for infection within TB-affected households may be of value for informing screening policies where resources are scarce.7 The diagnosis of M. tuberculosis infection remains a barrier to screening and IPT.8 The majority of high-
burden countries continue to use the tuberculin skin test (TST), despite well-known limitations.8 New interferon-gamma release assays (IGRAs) are becoming popular; however, their utility in high-burden settings remains controversial.9 Few studies have investigated risk factors for the positivity of the new generation IGRA, the QuantiFERON®-TB Gold In-Tube assay (QFT-GIT; Cellestis, Carnegie, VIC, Australia) or compared multiple risk factors for QFT-GIT positivity with TST positivity. This study, conducted in Indonesia, a country with the fifth-highest caseload of TB globally,10 aimed to quantify M. tuberculosis infection in children living with a smear-positive adult TB case and identify risk factors for TST and QFT-GIT positivity.
METHODS This study was conducted over 16 months at an urban community lung clinic in Bandung, West Java. This out-patient-based clinic acts as a referral for respiratory illnesses seen in primary health centres and private
Correspondence to: Merrin Rutherford, Centre of International Health, University of Otago, PO Box 913, Dunedin 9013, New Zealand. Tel: (+62) 221 396 613. Fax: (+62) 221 396 613. e-mail:
[email protected] Article submitted 28 May 2012. Final version accepted 24 July 2012.
Infection risk factors in Indonesian children
clinics. Around 2000 adults are diagnosed yearly with TB; approximately 50% are smear-positive. For this study, newly diagnosed adult TB cases living with a child(ren) (6 months–9 years) for more than 3 months were invited to bring their child contacts to the clinic for M. tuberculosis infection and disease screening. Eligible TB cases (index case) were smear and chest X-ray (CXR) positive. Smears were prepared using Ziehl-Neelsen staining; positivity was graded according to WHO guidelines.11 TB cases who had taken anti-tuberculosis medications for >1 month, were retreatment cases or who lived outside of Bandung, were excluded. Child contacts who had received a diagnosis of TB disease within the past year or who were aged 0 0 > Z > –1.0 –1.1 > Z > –2.0 –2 > Z > –3.0 TB case’s relation to child Parent Aunt/uncle Other TB case smear positivity Scanty or 1+ 2+ 3+ TB case has a productive cough Yes No
n (%) 150 (49.3) 58 [31–81] 284 (93.7) 19 (6.3) 221 (73.2) 52 (17.1) 30 (9.9) 228 (75.0) 76 (25.0) 186 (71.8) 58 (22.4) 9 (3.5) 6 (2.3) 231 (76.0) 28 (9.2) 45 (14.8) 105 (34.7) 72 (23.7) 126 (41.6) 242 (79.9) 61 (20.1)
IQR = interquartile range; BCG = bacille Calmette-Guérin; TB = tuberculosis.
TST positivity was associated with high infectious load (smear positivity 3+) of the index case (Appendix Table A.1). For exposure variables, the index case being the child contact’s parent and having fewer than 10 people living in the household were associated with TST positivity. A higher household income index was protective against TST positivity. Age and BCG scar status of the child contact were not associated with TST positivity. Infectivity of the index case (acid-fast bacilli [AFB] 3+ odds ratio [OR] 2.93, 95%CI 1.59–5.39) and the index case being the child contact’s parent (OR 7.04, 95%CI 2.23–22.28) remained significant in multivariate analysis. High infectious load of the index case was also significantly associated with QFT-GIT positivity (Appendix Table A.2). For exposure variables, increased sleeping proximity of the index case to the child contact increased the risk of QFT-GIT positivity, as did the index case being the child contact’s parent and having fewer than 10 people living in the household. Unlike the TST, QFT-GIT positivity was associated with older age of the child contact. In multivariate analysis, case infectivity (AFB 3+, OR 2.28, 95%CI 1.06– 4.90), older age (OR 2.58, 95%CI 1.53– 4.34) and the index case being the child contact’s parent (OR 4.30, 95%CI 1.48–12.45) remained significant. Strong trends across gradients of exposure and infectious load variables were seen for both tests. Dif-
ferences in the proportions of TST-positives vs. QFTGIT-positives were not significant at any exposure level (data not shown). High infectious load was significantly associated with increased CT positivity (Table 2). For exposure variables, the index case being the child contact’s parent, having fewer than 10 people living in the household, close sleeping proximity between the index case and child contact, and older age of the child contact were significantly or borderline significantly associated with increased CT positivity. Crowding within the household was not a significant risk factor. In multivariate analysis, the infectious load of the index case (AFB 3+, OR 2.44, 95%CI 1.24–4.80), the index case being the child contact’s parent (OR 4.08, 95%CI 1.75–9.49) and number of people in the household (compared to >10 people: 5–10 people OR 5.09, 95%CI 1.64–15.80, 1–4 people OR 4.87, 95%CI 1.44–16.46) remained significant predictors of CT positivity.
DISCUSSION In this setting, M. tuberculosis infection by either test was high in children living with a smear-positive TB case. Test positivity was driven by high index case infectivity levels and intimacy of exposure (if the index case was the child contact’s parent). These findings support screening and IPT programmes for child contacts and, in resource-poor settings, provide insight into the special importance of targeting those children whose parent is the index case or who are living with a highly infectious TB patient. Levels of infection and disease among child contacts in this study are similar to those reported elsewhere. A recent review of contact tracing studies from South-East Asia found infection among child contacts aged 0 0 > Z > –1.0 –1.1 > Z > –2.0 Z > –2.1 Child with symptoms No Yes Characteristics of TB case Smear positivity Scanty and 1+ 2+ 3+ Relationship to child Other Aunt/uncle Parent Sleeping proximity to child Different room Same room Same bed Time spent with child, h/day 8 Smoking status Smoker Non-smoker Duration of cough, months 12 Productive cough No Yes Duration of productive cough, months 1 Characteristics of household and household head Number of people in the household >10 5–10 1– 4 Number of people/sleeping room 4 Household income, IRP, 000s** 0–500 >500–1000 >1000–2000 >2000
CTCTpositive* negative Per cent (n = 180) (n = 124) positive n n %
aOR (95%CI)
85 95
65 59
56 62
1 1.26 (0.78–2.05)
50 130
48 76
51 63
1 1.62 (0.98–2.68)
168 12
116 7
59 63
1 1.17 (0.42–3.28)
128 37 15
94 15 15
58 71 50
1 1.79 (0.90–3.56) 0.73 (0.33–1.59)
0.600
117 32 6 3
69 26 3 3
63 55 66 50
1 0.80 (0.40–1.61) 1.20 (0.29–5.00) 0.60 (0.12–3.04)
0.328
134 46
94 30
59 61
1 1.13 (0.65–1.96)
47 43 89
58 29 37
45 60 71
1 1.77 (0.89–3.47) 2.83 (1.43–5.62)
1 1.57 (0.78–3.12)§ 2.44 (1.24– 4.80)
0.000
13 12 155
32 16 76
29 43 67
1 1.70 (0.66–5.86) 5.09 (2.24–13.12)
1 1.38 (0.46– 4.16)¶ 4.08 (1.75–9.49)
0.000
55 17 107
54 11 59
50 61 65
1 1.64 (0.60– 4.47) 1.73 (0.98–3.03)
1 1.10 (0.35–3.46)# 1.15 (0.57–2.35)
0.027
31 57 91
17 39 68
65 59 57
1 0.82 (0.34–1.96) 0.74 (0.33–1.66)
74 104
49 75
60 58
1 0.90 (0.52–1.57)
37 133 9
28 94 2
57 59 82
1 1.00 (0.54–1.84) 2.46 (0.45–13.45)
35 144
26 98
57 60
1 1.15 (0.61–2.15)
102 32
74 25
58 56
1 0.83 (0.40–1.72)
3 115 61
15 76 33
17 60 65
1 7.22 (2.28–22.88) 8.34 (2.49–27.94)
64 98 17
39 71 14
62 85 55
1 0.84 (0.48–1.47) 0.73 (0.29–1.87)
0.395
46 71 20 9
22 53 19 7
68 57 51 56
1 0.64 (0.31–1.31) 0.50 (0.20–1.27) 0.61 (0.21–1.84)
0.069
1 1.49 (0.89–2.48)‡
0.060
0.578
0.181
0.892
1 5.09 (1.64–15.80)§ 4.87 (1.44–16.46)
0.007
(continued )
1598
The International Journal of Tuberculosis and Lung Disease
Table 2
(Continued )
Variable Characteristics of household and household head (continued ) Household income index, IRP, 000s†† 100–200 >200–500 >500 Cooking done inside the house Yes No Any smoker in the household Yes No Relationship to household head Father Mother Other Household head marital status Married Other Occupation of household head Privately employed Self-employed Other
CTCTpositive* negative Per cent (n = 180) (n = 124) positive n n %
P for trend†
OR (95%CI)
aOR (95%CI)
1 0.78 (0.24–2.47)# 0.64 (0.21–1.94) 0.48 (0.13–1.70)
0.044
1 0.33 (0.09–1.21)‡‡ 0.85 (0.16– 4.40)
0.058
29 43 61 12
12 33 40 16
71 57 60 43
1 0.58 (0.21–1.59) 0.60 (0.24–1.51) 0.31 (0.09–1.02)
125 24
88 17
57 58
1 0.93 (0.43–1.99)
125 23
93 12
57 66
1 1.33 (0.59–3.04)
138 7 4
87 15 3
61 32 57
1 0.34 (0.12–1.01) 0.85 (0.14– 4.95)
142 7
103 2
58 78
1 2.51 (0.51–12.39)
0.136
47 79 23
41 48 16
53 62 59
1 1.29 (0.69–2.42) 1.11 (0.43–2.82)
0.242
* Positive by either or both QFT-GIT and TST. † P values were calculated for variables with gradients using an extension of the Wilcoxon rank-sum test. Confounders: ‡ number of persons in the house; § TB case’s relationship to child; ¶ smear positivity; # TB case’s relationship to child, smear positivity; ‡‡ marital status of household head, smear positivity, TB case’s relationship to child. ** IRP = Indonesian rupiah (1 USD = 10 000 IRP). †† Household income index = total monthly household income/total number of people supported by household income. TST = tuberculin skin test; QFT-GIT = QuantiFERON®-TB Gold In-Tube; CT = combined test; OR = odds ratio; CI = confidence interval; aOR = adjusted OR; BCG = bacille Calmette-Guérin; TB = tuberculosis.
and in Turkey, child contacts aged ⩽16 years were 2.2 (95%CI 1.6–2.9) times as likely to be infected if the index case was a parent.19 This association likely reflects the increased intimacy of the relationship between child and parent. In two African studies, where women do the majority of child rearing, increased risk of M. tuberculosis infection was observed if the index case was the mother compared to the father.7,20 The increased risk of infection when the index case is a parent may also be explained by genetic factors. A Gambian study reported that children were less likely to be TST-positive if the index case was not of firstdegree genetic proximity (OR 0.72, 95%CI 0.54– 0.97). The authors suggested that, while closeness of contact was the overriding factor in infection risk, genetic factors may play a role.17 Crowding, reflecting the number of people in relation to space in a house, has long been thought to influence M. tuberculosis infection rates.21 This study and others have not found crowding to be significantly related to infection in child contacts.2,7,18 It has been suggested that the lack of association is due to a saturation effect, whereby above a certain number of people per room, no dose-response relationship between infection levels and crowding levels is observed.22 This study found an inverse relationship between number of people per sleeping room and risk of infection. A study from The Gambia reported similar
findings,2 and the authors suggest that it is intimacy and occurrence of contact with the index case that drive transmission, and not the number of persons in contact with the TB case in a limited area. Poverty, as reflected by household income, was associated with increased TST positivity. Low socioeconomic status (SES) has been suggested to impact infection levels through poor nutrition, crowding, urbanisation and alcoholism.21 While we did not find an association with crowding or poor nutrition, other factors associated with SES may well be important in this setting. These may be amenable to interventions, and further investigation is required. Comparisons have shown little difference between the performance of TST and QFT-GIT in high-burden countries.9 Both the TST and QFT-GIT responded as expected to most hypothesised risk factors, and neither test performed significantly better than the other along any of the gradients. A recent meta-analysis of studies conducted in high-burden settings showed that the regression lines for TST and QFT-GIT positivity over exposure gradients were similar, with overlapping CIs.9 This study has several limitations. We cannot be sure that other TB cases within the household were reported, potentially compromising the evaluation of some risk factors. However, only around 2% of household members would be expected to have co-prevalent
Infection risk factors in Indonesian children
bacteriologically confirmed TB.23 The study’s crosssectional design prevents temporal associations from being made. HIV infection status, a potential risk factor,24 was not established. However, as HIV prevalence is low in Indonesia,12 it is likely that too few participants would have been HIV-positive to show a significant association, and a requirement for testing would have increased the numbers of refusals to participate. Furthermore, truly M. tuberculosis-infected children may have been missed if exposure was 0 0 > Z > –1.0 –1.1 > Z > –2.0 Z > –2.1 Child with symptoms No Yes Characteristics of TB case Smear positivity Scanty and 1+ 2+ 3+ Relationship to child Other Aunt/uncle Parent Sleeping proximity to child Different room Same room Same bed Time spent with child, h/day 8 Smoking status Smoker Non-smoker Duration of cough, months 12 Productive cough No Yes Duration of productive cough, months 1 Characteristics of household and household head Number of household members >10 5–10 1– 4 Number of persons/sleeping room 4 Household income, IRP, 000s¶ 0–500 >500–1000 >1000–2000 >2000
TSTTSTpositive negative Per cent (n = 145) (n = 157) positive n n %
OR (95%CI)
aOR (95%CI)
P for trend*
71 74
79 78
47 49
1 1.09 (0.68–1.74)
40 105
57 100
41 51
1 1.46 (0.88–2.41)
133 12
149 7
47 63
1 1.93 (0.70–5.30)
105 29 11
116 22 19
48 57 37
1 1.46 (0.74–2.88) 0.64 (0.28–1.46)
0.933
96 25 4 3
89 33 5 3
52 43 50
1 0.77 (0.39–1.51) 0.77 (0.20–3.00) 0.96 (0.19– 4.83)
0.349
109 36
117 40
48 47
1 1.0 (0.59–1.69)
33 34 78
71 38 48
32 47 62
1 1.80 (0.89–3.63) 3.35 (1.81–6.21)
1 1.60 (0.78–3.25)† 2.93 (1.59–5.39)
0.000
8 9 128
37 19 101
18 32 56
1 2.31 (0.77–6.79) 5.85 (2.56–13.38)
1 2.36 (0.61–9.13)‡ 7.04 (2.23–22.28)
0.000
47 13 85
62 15 80
43 46 52
1 1.21 (0.41–3.53) 1.35 (0.79–2.32)
0.186
28 42 75
20 54 83
58 44 47
1 0.55 (0.24–1.24) 0.64 (0.31–1.36)
0.494
60 84
63 94
49 47
1 0.94 (0.56–1.58)
25 112 8
40 114 3
38 89 73
1 1.49 (0.82–2.73) 3.04 (0.62–14.88)
23 122
38 119
38 51
1 1.79 (0.95–3.35)
82 31
93 26
47 54
1 1.22 (0.60–2.52)
3 91 51
15 100 42
17 48 55
1 4.28 (1.36–13.50) 5.44 (1.63–18.18)
50 84 11
52 85 20
49 50 35
1 1.03 (0.60–1.76) 0.57 (0.22–1.50)
42 56 17 6
26 68 22 9
62 45 44 40
1 0.50 (0.25–0.99) 0.48 (0.20–1.17) 0.42 (0.13–1.28)
0.120
0.011
1 1.94 (0.99–3.81)§ 0.357
1 2.78 (0.90–8.61)† 2.96 (0.86–10.12)
0.021
0.315
1 0.65 (0.30–1.38)§ 0.50 (0.19–1.31) 0.78 (0.28–2.16)
0.023
(continued )
Infection risk factors in Indonesian children
Table A.1
iii
(Continued )
Variable Characteristics of household and household head (continued ) Household income index, IRP, 000s# 100–200 >200–500 >500 Cooking done inside the house Yes No Any smoker in the household Yes No Relationship to household head Father Mother Marital status of household head Married Other Occupation of household head Privately employed Self-employed Other
TSTTSTpositive negative Per cent (n = 145) (n = 157) positive n n %
OR (95%CI)
aOR (95%CI)
1 0.73 (0.25–2.14)§ 0.54 (0.19–1.54) 0.39 (0.11–1.34)
P for trend*
26 36 49 8
15 40 52 19
63 47 49 30
1 0.55 (0.21–1.40) 0.51 (0.21–1.23) 0.24 (0.08–0.78)
101 22
111 19
48 54
1 1.21 (0.55–2.63)
101 21
117 13
46 62
1 1.77 (0.78– 4.02)
113 7
111 15
50 32
1 0.54 (0.18–1.59)
0.178
116 7
128 2
48 78
1 3.88 (0.78–19.16)
0.032
35 67 21
53 59 18
40 53 54
1 1.59 (0.87–2.91) 1.60 (0.65–3.92)
0.035
* P values were calculated for variables with gradients using an extension of the Wilcoxon rank-sum test. Confounders: † TB case’s relationship to child; ‡marital status and smear positivity of household head; § TB case’s relationship to child, smear positivity. ¶ IRP = Indonesian rupiah (1 USD = 10 000 IRP). # Income index = total monthly household income/total number of people supported by household income. TST = tuberculin skin test; OR = odds ratio; CI = confidence interval; aOR = adjusted OR; BCG = bacille Calmette-Guérin; TB = tuberculosis.
0.011
iv
The International Journal of Tuberculosis and Lung Disease
Table A.2
Risk factors for QFT-GIT assay positivity in child contacts
Variable Characteristics of child Sex Male Female Age, months 0–36 37–120 Ethnicity Sundanese Other BCG scar Yes No Unknown Z score Z>0 0 > Z > –1.0 –1.1 > Z > –2.0 Z > –2.1 Child with symptoms No Yes Characteristics of TB case Smear positivity Scanty and 1+ 2+ 3+ Relationship to child Other Aunt/uncle Parent Sleeping proximity to child Different room Same room Same bed Time spent with child, h/day 8 Smoking status Smoker Non-smoker Duration of cough, months 12 Productive cough No Yes Duration of productive cough, months 1 Characteristics of household and household head Number of household members >10 5–10 1– 4 Number of persons/sleeping room 4 Household income, IRP, 000s# 0–500 >500–1000 >1000–2000 >2000
QFT-GIT- QFT-GITpositive negative Per cent (n = 152) (n = 138) positive n n %
OR (95%CI)
aOR (95%CI)
P for trend*
74 78
72 66
51 54
1 1.19 (0.74–1.91)
36 116
56 82
39 59
1 2.26 (1.37–3.72)
144 8
127 10
53 44
1 0.68 (0.23–2.03)
104 34 14
105 17 16
50 67 47
1 1.94 (1.00–3.77) 0.85 (0.40–1.80)
103 29 4 3
74 28 4 2
58 51 50 60
1 0.78 (0.40–1.55) 0.70 (0.17–2.96) 1.05 (0.17–6.50)
117 35
102 36
53 49
1 0.91 (0.52–1.59)
40 36 75
59 34 45
40 51 63
1 1.56 (0.78–3.11) 2.43 (1.21– 4.86)
1 1.37 (0.63–2.99)‡ 2.28 (1.06– 4.90)
0.001
10 8 134
34 19 85
23 30 61
1 1.51 (0.44–5.17) 5.61 (2.40–13.12)
1 1.41 (0.33–5.92)§ 4.30 (1.48–12.45)
0.000
43 15 93
62 12 64
41 56 59
1 1.87 (0.70–5.02) 2.01 (1.12–3.61)
1 1173 (0.42–3.27)¶ 1.45 (0.70–2.99)
0.006
27 46 78
20 46 72
57 50 52
1 0.78 (0.33–1.80) 0.83 (0.38–1.79)
65 85
56 82
54 51
1 0.88 (0.50–1.55)
31 112 8
29 106 3
52 51 73
1 0.94 (0.50–1.76) 3.05 (0.56–16.70)
30 121
29 109
51 53
1 1.07 (0.56–2.04)
85 27
81 28
51 49
1 0.88 (0.43–1.83)
2 96 53
16 85 37
11 53 59
1 8.80 (1.57– 49.42) 10.83 (1.87–62.64)
49 88 14
47 75 16
51 54 47
1 1.11 (0.63–2.00) 0.80 (0.30–2.13)
0.899
40 58 17 8
27 62 18 8
60 48 49 50
1 0.66 (0.32–1.38) 0.64 (0.25–1.65) 0.68 (0.23–1.99)
0.154
1 2.58 (1.53– 4.34)†
0.004
1 1.65 (0.72–3.81)‡ 0.71 (0.32–1.57)
0.272
0.451
0.948
0.285
0.876
1 4.31 (0.73–25.59)‡ 4.38 (0.70–27.29)
0.006
(continued )
Infection risk factors in Indonesian children
Table A.2
v
(Continued )
Variable Characteristics of household and household head (continued ) Household income index, IRP, 000s** 100–200 >200–500 >500 Cooking done inside the house Yes No Any smoker in the household Yes No Relationship to household head Father Mother Other Marital status of household head Married Other Occupation of household head Privately employed Self-employed Other
QFT-GIT- QFT-GITpositive negative Per cent (n = 152) (n = 138) positive n n %
OR (95%CI)
aOR (95%CI)
P for trend*
25 38 48 11
15 37 46 17
63 51 51 39
1 0.65 (0.24–1.81) 0.64 (0.25–1.63) 0.39 (0.10–1.45)
0.053
106 20
100 19
51 51
1 1.01 (0.44–2.33)
108 17
104 15
51 53
1 1.13 (0.52–2.49)
117 6 3
99 16 4
54 27 43
1 0.35 (0.09–1.38) 0.62 (0.12–3.28)
0.046
119 7
117 2
50 78
1 3.32 (0.67–16.40)
0.053
41 67 18
44 57 18
48 54 50
1 1.18 (0.63–2.21) 1.06 (0.34–3.32)
0.447
* P values were calculated for variables with gradients using an extension of the Wilcoxon rank-sum test. Confounders: † sleeping proximity to TB case; ‡ TB case’s relationship to child, marital status of household head; § marital status of household head, smear positivity; ¶ TB case’s relationship to child, age of child, smear positivity. # IRP = Indonesian rupiah (1 USD = 10 000 IRP). ** Household income index = total monthly household income/total number of people supported by household income. QFT-GIT = QuantiFERON®-TB Gold In-Tube; OR = odds ratio; CI = confidence interval; aOR = adjusted OR; BCG = bacille Calmette-Guérin; TB = tuberculosis.
vi
The International Journal of Tuberculosis and Lung Disease
RÉSUMÉ C O N T E X T E E T O B J E C T I F : Les jeunes enfants vivant avec des sujets atteints de tuberculose (TB) contagieuse encourent un risque élevé d’infection et de maladie. Le dépistage, qui est recommandé, ne se réalise que rarement dans les contextes à ressources limitées. L’identification des enfants encourant le risque le plus élevé d’infection peut être utile pour l’élaboration de politiques pratiques de dépistage. S C H É M A : On a invité les enfants-contact de patients adultes atteints de TB à bacilloscopie positive des frottis pour le dépistage de l’infection et de la maladie à Mycobacterium tuberculosis au moyen des symptômes, d’un test tuberculinique cutané (TST), du test QuantiFERON®-TB Gold In-Tube (QFT-GIT) et du cliché thoracique. On a déterminé les facteurs de risque d’infection par questionnaire et on les a calculés séparément pour le TST et le QFT-GIT ainsi que pour la combinaison des deux.
Sur 304 enfants dépistés, le TST était positif chez 145/302 (48%), le QFT-GIT était positif chez 152/299 (51%), et un test ou les deux tests étaient positifs chez 180/304 (59%). La positivité pour les deux tests est en association avec la contagiosité du cas-index (AFB 3+ comparé à AFB 1+ : OR TST 2,93 ; IC95% 1,59– 5,39 ; OR QGT-GIT 2,28 ; IC95% 1,06– 4,90), ainsi qu’avec l’exposition (lorsque le parent de l’enfant-contact est le cas-index OR TST 7,04 ; IC95% 2,23–22,28 ; OR QFT-GIT 4,30 ; IC95% 1,48–12,45). C O N C L U S I O N : Une infection par M. tuberculosis décelée par n’importe lequel de ces tests plaide fortement en faveur d’un dépistage et d’un traitement préventif. Un dépistage immédiat constitue une priorité élevée chez les enfants de cas atteints de TB à bacilloscopie positive qui accompagnent leurs parents au dispensaire. R É S U LTAT S :
RESUMEN M A R C O D E R E F E R E N C I A Y O B J E T I V O S : Los niños de corta edad que viven con casos contagiosos de tuberculosis (TB) presentan un alto riesgo de contraer la infección y la enfermedad tuberculosa; por esta razón se recomienda practicar en ellos la detección sistemática. Esta medida se practica rara vez en los entornos con recursos limitados. Conocer a los niños que se encuentran en mayor riesgo de contraer la infección podría ser útil en el momento de establecer las normas prácticas de detección sistemática. M É T O D O : Se convocó a los niños contactos de pacientes adultos tuberculosos con baciloscopia positiva a fin de investigar la presencia de infección y enfermedad tuberculosa con base en los síntomas, el resultado de la prueba tuberculínica (TST), de la prueba QuantiFERON®-TB Gold En Tubo (QFT-GIT) y de la radiografía de tórax. Se recogieron datos sobre los factores de riesgo de infección mediante cuestionarios; se calculó el riesgo con cada prueba de manera independiente y con todas las pruebas combinadas.
De 304 niños, 145/302 (48%) tuvieron una prueba TST positiva, en 152/299 (51%) el resultado de la QFT-GIT fue positivo y en 180/304 (59%) el resultado fue positivo en una o en ambas pruebas. La positividad de ambas pruebas se correlacionó con el grado de contagiosidad del caso inicial (baciloscopia 3+ comparado con 1+; para TST: OR 2,93; IC95% 1,59– 5,39; para QFT-GIT: OR 2,28; IC95% 1,06– 4,90) y se correlacionó también con el grado de exposición (uno de los padres del niño como caso inicial: para TST OR 7,04; IC95% 2,23–22,28 y para QFT-GIT OR 4,30; IC95% 1,48–12,45). C O N C L U S I Ó N : La infección por M. tuberculosis detectada con cualquiera de las pruebas constituye un fundamento de peso en favor de la investigación diagnóstica y el tratamiento preventivo. Los hijos pequeños de los casos de TB con baciloscopia positiva que acompañan a sus padres al consultorio son candidatos de alta prioridad para la detección sistemática inmediata. R E S U LTA D O S :
