Tuberculosis among children and adolescents in Rio de Janeiro, Brazil – Focus on extrapulmonary disease

Open AccessPublished:February 14, 2021DOI:https://doi.org/10.1016/j.ijid.2021.02.023

      Highlights

      • Extrapulmonary Tuberculosis (EPTB) is underrepresented among Tuberculosis (TB) patients.
      • Miliary TB and TB meningitis in young children are unrepresented in EPTB patients.
      • EPTB diagnosis is often challenging and mostly performed in hospital settings.
      • Time from sample collection to treatment initiation is longer in EPTB patients.
      • Deaths due to TB are more common among EPTB patients, especially TB meningitis.

      Abstract

      Objective

      To compare the socio-demographic, clinical, and diagnostic characteristics and treatment outcomes between extrapulmonary tuberculosis (EPTB) and pulmonary tuberculosis (PTB) in children and adolescents in Rio de Janeiro, a high TB-burdened Brazilian city.

      Methods

      This retrospective study used data from patients with EPTB and PTB aged 0 – 18 years, notified on two national databases from 2014 to 2016.

      Results

      Among the 1008 patients, 144 (14.2%) had EPTB. Patients with EPTB showed higher odds of hospital-based diagnosis (odds ratio (OR): 6.76 [95% confidence interval (95% CI): 4.62–9.90]; p < 0.001), no laboratory confirmation (OR: 4.9 2.14 [95% CI: 3.07 – 7.85]; p < 0.001), and being <14 years old (OR: 3.13 [95% CI: 2.18–4.49]) than those with PTB. A diagnosis without laboratory investigation was observed among 301/864 (34.8%) patients with PTB, 48/144 (33.3%) with EPTB, and among those aged under five years with EPTB (15/27 [55.6%]). TB deaths were more frequent in patients with EPTB (5/144 [3.5%]) than in those with PTB (4/864[0.5%]) (p = 0.001); 4/5 (80%) TB deaths were due to TB meningitis; 50% died within 14 days of diagnosis.

      Conclusions

      EPTB remains a clinical diagnostic challenge that needs to be addressed to fully benefit from the higher sensitivity laboratory investigations.

      Keywords

      Introduction

      Extrapulmonary tuberculosis (EPTB) has received renewed attention due to an increasing proportion of EPTB among tuberculosis (TB) cases, more pronounced in low-incidence countries (
      • Sandgren A.
      • Hollo V.
      • van der Werf M.J.
      Extrapulmonary tuberculosis in the European union and European economic area, 2002 to 2011.
      ). Epidemiological data regarding EPTB remains limited in children and adolescents; few studies refer to this particular population, which are limited mainly to inpatients, and possibly biased toward complex cases (
      • Wang M.-S.
      • Wang M.-S.
      • Liu X.-J.
      Epidemiological trends in the form of childhood tuberculosis in a referral tuberculosis hospital in Shandong, China.
      ). In addition, larger studies include adolescents in the young adult population (15–24 years) or define children as being under 15 years old (
      • Al-ghafli H.
      • Varghese B.
      • Enani M.
      • Id A.A.
      • Sameera A.l.
      • Albarrak A.
      • et al.
      Demographic risk factors for extrapulmonary tuberculosis among adolescents and adults in Saudi Arabia.
      ,
      • Pang Y.
      • An J.
      • Shu W.
      • Huo F.
      • Chu N.
      • Gao M.
      • et al.
      Epidemiology of extrapulmonary tuberculosis among inpatients, China, 2008-2017.
      ,
      • Tahseen S.
      • Khanzada F.M.
      • Baloch A.Q.
      • Abbas Q.
      • Bhutto M.M.
      • Alizai A.W.
      • et al.
      Extrapulmonary tuberculosis in Pakistan- A nation-wide multicenter retrospective study.
      ). Furthermore, EPTB prevalence varies according to the TB burden at the country level and the definition being used (
      • Kulchavenya E.
      Extrapulmonary tuberculosis: are statistical reports accurate?.
      ,
      • Sandgren A.
      • Hollo V.
      • van der Werf M.J.
      Extrapulmonary tuberculosis in the European union and European economic area, 2002 to 2011.
      ).
      According to the World Health Organization (WHO), Brazil has a high TB and TB-human immunodeficiency virus (HIV) co-infection burden (). EPTB represents 13.4% of TB cases (
      • Gomes T.
      • Reis-Santos B.
      • Bertolde A.
      • Johnson J.L.
      • Riley L.W.
      • Maciel E.L.
      Epidemiology of extrapulmonary tuberculosis in Brazil: a hierarchical model.
      ). In 2019, the incidence rate of TB was 35 cases/100,000 inhabitants; and among 73,584 new TB cases, 1,646 (2.2%) were in children under ten years old (
      • Brazil
      Boletim epidemiológico especial tuberculose.
      ). There were 39 new cases of miliary TB and tuberculous meningitis (TBM) in children under five years old (
      • Brazil
      Boletim epidemiológico especial tuberculose.
      ). This is the first time that the Brazilian government has published a report on national TB data that includes children and adolescents.
      Although this is an important step toward the knowledge of TB national data in this population, analysis of TB among children and adolescents proves challenging due to the report’s age range aggregation. For instance, in children under five years old, only data on miliary and TBM are presented; data on adolescents is included in the 10–64 year age group.
      Rio de Janeiro has the second-highest TB incidence (93.7 cases/100,000 inhabitants) and the fourth-highest TB mortality rate (4.6 deaths/100,000 inhabitants) in Brazil (
      • Brazil
      Boletim epidemiológico especial tuberculose.
      ).
      WHO has called for greater disaggregation of EPTB data by age (
      • World Health Organization
      Global Tuberculosis Report 2020.
      ) to gain a better understanding of the epidemiology of the disease in children and adolescents. The limited amount of published information on EPTB among children and adolescents is largely due to the difficulty of laboratory diagnosis (
      • Pang Y.
      • An J.
      • Shu W.
      • Huo F.
      • Chu N.
      • Gao M.
      • et al.
      Epidemiology of extrapulmonary tuberculosis among inpatients, China, 2008-2017.
      ,
      • Tahseen S.
      • Khanzada F.M.
      • Baloch A.Q.
      • Abbas Q.
      • Bhutto M.M.
      • Alizai A.W.
      • et al.
      Extrapulmonary tuberculosis in Pakistan- A nation-wide multicenter retrospective study.
      ) and the variety of clinical presentations (
      • Maltezou H.C.
      • Spyridis P.
      • Kafetzis D.A.
      Extrapulmonary tuberculosis in children.
      ,
      • Santiago-García B.
      • Blázquez-Gamero D.
      • Baquero-Artigao F.
      • Ruíz-Contreras J.
      • Bellón J.M.
      • Muñoz-Fernández M.A.
      • et al.
      Pediatric extrapulmonary tuberculosis: clinical spectrum, risk factors and diagnostic challenges in a low prevalence region.
      ,
      • Tahseen S.
      • Khanzada F.M.
      • Baloch A.Q.
      • Abbas Q.
      • Bhutto M.M.
      • Alizai A.W.
      • et al.
      Extrapulmonary tuberculosis in Pakistan- A nation-wide multicenter retrospective study.
      ).
      We aimed to compare the socio-demographic, clinical, and diagnostic characteristics and treatment outcomes between EPTB and pulmonary TB (PTB) in children and adolescents under programmatic conditions in Rio de Janeiro.

      Methods

       Study design

      A retrospective study was performed comparing PTB and EPTB forms in children and adolescents (0–18 years) diagnosed in Rio de Janeiro city and notified in the Brazilian Notifiable Diseases Surveillance System (Sistema de Informação de Agravos de Notificação em Tuberculose [SINAN-TB]), from August 1, 2014, to August 1, 2016, during the period of Xpert MTB/RIF (Xpert, Cepheid, Sunnyvale, CA, USA) implementation in Rio de Janeiro as the first laboratory approach for PTB and EPTB diagnosis, replacing sputum smear microscopy (SSM) (
      • Brazil
      Sistema de informação de agravos de notificação em tuberculose (SINAN-TB).
      ). SINAN is a national database that records data on the clinical forms of TB and follow-up. All patients with TB with a laboratory or clinical diagnosis performed at any health care facility (reference centers, local hospitals, or primary health care) are recorded, as this is mandatory for starting anti-TB treatment for PTB, EPTB, or associated forms (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ). Another national database, the Gerenciamento do Ambiente Laboratorial (GAL), is responsible for monitoring and managing laboratory data. GAL was used to check the laboratory results through a manual linkage between the two databases (
      • Jesus R de
      • Guimarães R.P.
      • Bergamo R.
      • Santos LCF dos
      • Matta ASD da
      • Paula Júnior FJ de
      Sistema Gerenciador de Ambiente Laboratorial: relato de experiência de uma ferramenta transformadora para a gestão laboratorial e vigilância em saúde.
      ).

       Definitions

      PTB was defined as TB affecting the lung parenchyma, including intrathoracic or mediastinal lymph nodes (
      • World Health Organization
      Definitions and Reporting Framework for Tuberculosis – 2013 Revision.
      ). EPTB was defined as TB affecting other organs: laryngeal, pleural, peripheric lymph nodes, osteoarticular, TBM, genitourinary, ocular, cutaneous, abdominal, and pericardiac (
      • World Health Organization
      Definitions and Reporting Framework for Tuberculosis – 2013 Revision.
      ). Miliary TB was defined as EPTB according to the SINAN database (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ).
      The types of cases were classified as: new cases, retreatment, and transfer, according to the notification on admission in the SINAN database (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ,
      • World Health Organization
      Definitions and Reporting Framework for Tuberculosis – 2013 Revision.
      ). The “transferred in” category refers to patients identified in a health care unit coming from another one to continue treatment without interruption for more than 30 days (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ). Patients described as “extremely vulnerable” were those with at least one of the following characteristics: homeless, inmate, and/or tobacco, illicit drug, or alcohol use. Laboratory confirmation was defined as any positive result (SSM, Xpert, mycobacterial culture (culture), or histopathology) that confirmed TB.
      According to WHO and the National Tuberculosis Control Program (NTCP) guidelines (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ,
      • World Health Organization
      Definitions and Reporting Framework for Tuberculosis – 2013 Revision.
      ), the outcomes were classified as favorable (cure, treatment completion) and unfavorable (loss to follow-up [LTFU], treatment failed, and death). Drug-resistant TB (DR-TB) was classified separately (
      • World Health Organization
      Definitions and Reporting Framework for Tuberculosis – 2013 Revision.
      ).
      Cure was determined based on two negative SSM: one during the treatment follow-up (any month) and another at the end of the TB treatment. Treatment completion was defined as the patient completing the treatment without evidence of failure and was discharged based on clinical and radiological exams (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ).
      LTFU was defined when the patient interrupted the treatment for 30 consecutive days or more or when the diagnosed patient did not start treatment (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ). Death due to TB was defined when the patient died due to TB as the basic cause. Death due to other causes was defined when the basic cause of death is not TB, although it was associated (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ). Transferred out is defined as when the patient is transferred to another health care unit for treatment completion. Other definitions are presented in the online supplementary material 1.
      The NTCP has guidelines for TB treatment, including the TBM and osteoarticular forms, according to age (children aged <10 and ≥10 years) and weight (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ) following those proposed by WHO (
      • World Health Organization
      Guidance for National Tuberculosis Programmes on the Management of Tuberculosis in Children.
      ).

       Statistical analyses

      Descriptive statistics were used to summarize clinical, epidemiological, and laboratory data. Continuous variables were compared using the Mann–Whitney U test. Categorical variables were compared using Fisher’s exact test and analyzed using the binary logistic regression to calculate the odds ratio (OR) and 95% confidence interval (95% CI) to comparing EPTB and PTB forms. Statistical analyses were performed using the SPSS Statistics program version 24 (IBM Corp. Armonk, NY, USA).

       Ethical issues

      This study was approved by the Research Ethics Committees of the Instituto de Puericultura e Pediatria Martagão Gesteira of the Federal University of Rio de Janeiro and Rio de Janeiro Municipal Health Department under references No. 961,452 (2015) and No. 1,629,126 (2016), respectively.

      Results

      During the study period, a total of 1059 children and adolescents were registered with all TB forms (PTB, EPTB, and PTB associated with EPTB). EPTB patients represented 13.5% (114/1059) of all TB forms.
      Our study population consisted of 1008 patients identified with isolated forms of PTB (864 [85.7%]) and EPTB (144[14.3%]), excluding 51 patients with PTB associated with EPTB. Compared to PTB, EPTB patients showed higher odds of hospital-based diagnosis (OR: 6.76 [95% CI: 4.62–9.90]; p < 0.001), diagnosis outside the residence area (OR: 3.54 [95% CI: 2.42–5.18]; p < 0.001), and no laboratory confirmation (OR: 4.90 [95% CI: 3.07 – 7.85]; p < 0.001). Furthermore, EPTB was associated with patients under 14 years old (OR: 3.13 [95% CI: 2.18–4.49]) (Table 1). The median time from sample collection to treatment initiation was significantly longer in patients with EPTB (nine days [interquartile range [IQR]: 2 – 97]) than in those with PTB (five days [IQR: 2 – 11]). However, we did not observe a significant difference in time from diagnosis to LTFU.
      Table 1Patients characteristics according to Pulmonary and Extrapulmonary Tuberculosis.
      CharacteristicsTuberculosis forms
      Patients with both PTB and EPTB were excluded.
      Total N = 1008 (%)OR
      Odds ratio.
      (95% CI
      Confidence Interval.
      )
      P_value
      Pulmonary N = 864 (%)Extrapulmonary N = 144 (%)
      Gender
      Female411 (47.6)65 (45.1)476 (47.2)reference
      Male453 (52.4)79 (54.9)532 (52.8)0.97 (0.32–2.92)0.958
      Age Range (years old)
      Less than 592 (10.6)27 (18.8)119 (11.8)2.90 (1.75–4.79)<0.001
      5–951 (5.9)20 (13.9)71 (7)3.87 (2.12–6.91)<0.001
      10–14119 (13.8)36 (25)155 (15.4)2.98 (1.89–4.71)<0.001
      15–18602 (69.7)61 (42.4)663 (65.8)reference
      Type of case
      New case805 (93.2)133(92.4)938 (93.1)reference
      Retreatment54 (6.2)8 (5.6)62 (6.2)0.90 (0.42 -1.93)0.78
      Transferred in5 (0.6)3 (2.1)8 (0.8)3.63 (0.85–15.37)0.80
      Health care unit of notification
      Primary health care750 (86.8)71 (49.3)821 (81.4)reference
      Hospital114 (13.2)73 (50.7)187 (18.6)6.76 (4.62–9.90)<0.001
      Same residence and health care unit area
      Yes729 (84.4)87 (59)816 (81)reference
      No135 (15.6)57 (39.6)192 (19)3.54 (2.42–5.18)<0.001
      HIV
      Human immunodeficiency virus.
      test results
      Negative689 (79.7)103 (71.5)792 (78.6)reference
      Positive29 (3.4)4 (2.8)33 (3.3)0.92 (0.32–2.68)0.88
      Results not available/not performed146 (16.9)37 (25.7)183 (18.2)1.69 (1.18–2.57)0.13
      Comorbidities
      No816 (94.4)137 (95.1)953 (94.5)reference
      Yes48 (5.6)7 (4.9)55 (5.5)0.87 (0.38–1.96)0.73
      Extreme vulnerability
      The definition of "extreme vulnerability" can be found in the Methods section.
      No764 (88.4)126 (87.5)890 (88.3)reference
      Yes90 (10.4)2 (1.4)92 (9.1)0.10 (0.05–0.23)<0.001
      Missing data10 (1.2)16 (11.1)26 (2.6)0.14 (0.00–0.69)<0.001
      Laboratory confirmation
      Patients who did not perform laboratory investigation were excluded from this analysis.
      Positive results486 (86.3)54 (56.2)540 (81.9)reference
      Negative results77 (13.7)42 (43.8)119 (18.1)4.90 (3.07–7.85)<0.001
      Outcome
      Patients with missing information were excluded from the analysis.
      Favorable681 (78.8)103 (71.5)784 (77.8)reference
      Unfavorable112 (13)15 (10.4)127 (12.6)0.88 (0.50–1.58) -0.68
      Transferred out33 (3.8)15 (10.4)48 (4.8)3.24 (1.92–5.46)<0.001
      DR-TB
      Drug-resistant tuberculosis.
      20 (2.3)1 (0.7)21 (2.1)0.33 (0.04–2.49)0.28
      Time (days)from sample collection to treatment initiation
      median (IQR
      Interquartile range.
      )
      5 (2–11)9 (2–97)5 (2–12)0.02
      Time (days) from diagnosis to LTFU
      Loss to follow-up.
      Median (IQR
      Interquartile range.
      )
      182 (95–241)198 (145–385)188 (97–425)0.25
      a Patients with both PTB and EPTB were excluded.
      b Odds ratio.
      c Confidence Interval.
      d Human immunodeficiency virus.
      e The definition of "extreme vulnerability" can be found in the Methods section.
      f Patients who did not perform laboratory investigation were excluded from this analysis.
      g Patients with missing information were excluded from the analysis.
      h Drug-resistant tuberculosis.
      i Interquartile range.
      j Loss to follow-up.
      The median days from diagnosis to death due to TBM was 24.5 (IQR 8 – 74.8), and 50% of the patients died within 14 days of the diagnosis. TB treatment success was the most frequent outcome in PTB (681 [78.8%]) and EPTB (103 [71.5%]) (Table 1).
      When looking at the unfavorable outcomes exclusively (Table 2), LTFU was significantly more common in patients with PTB (97 [86.6%] than those with EPTB (10 [66.7%]). Only two adolescents with PTB, aged 18 years, without comorbidities, were LTFU before starting treatment (data not shown).
      Table 2Distribution of the Unfavorable outcomes exclusively according to Tuberculosis forms.
      Unfavorable Outcomes
      From the 1008 patients, 127 presented unfavorable outcomes as presented in Table 1.
      Tuberculosis formsTotal n = 127 (%)P_value
      Pulmonary n = 112 (%)Extrapulmonary n = 15 (%)
      Loss to follow-up97 (86.6)10 (66.7)107 (84.2)0.001
      Death due to TB
      TB – tuberculosis.
      4 (3.6)5 (33.3)9 (7.1)
      Death due to other causes11 (9.8)011 (8.7)
      * From the 1008 patients, 127 presented unfavorable outcomes as presented in Table 1.
      a TB – tuberculosis.
      Death due to TB was significantly higher among patients with EPTB (5 [33.3%]) than those with PTB (4 [3.6%]), and the most frequent unfavorable outcome in EPTB. Four (80%) patients died due to TBM (one death in those under five years old; two in those aged 10–14 years, and one in those aged 15–18 years). Among these patients, none were TB-HIV coinfected, 3/4 (75%) were male, and one was malnourished. The other death occurred in one patient with pleural EPTB (Table 2).
      Pleural forms were more frequent in those aged 10–14 (12/36 [33.3%]) and 15–18 (35/61 [57.4%]), while the peripheric lymph node forms were more frequent in those aged <5 (10/27 [37%]) and 5–9 (9/20 [45%]); TBM was more frequent in those aged <5 (9 [33.3%]) (Figure 1).
      Hospital notification (22/61 [36.1%]) and diagnosis performed outside the residence area were less frequent among those 15–18 years old (17/61 [27.9%]) (Table 3). LTFU was more frequent among those aged 10–14 (2/36 [5.6%]) and 15–18 years (7/61 [11.5%]), while transferred out was more frequent among those under five years old (6/27 [22.2%]). Adolescents were the only patients with comorbidities. The types of comorbidities according to TB forms are presented as online supplementary material 2.
      Table 3Patients characteristics according to age range in EPTB.
      CharacteristicsAge Range (years)Total
      <54–910–1415–18
      N = 27 (%)N = 20 (%)N = 36 (%)N = 61 (%)N = 144 (%)
      Gender
       Female11 (40.7)10 (50)18 (50)26 (42.6)66 (45.1)
       Male16 (59.3)10 (50)18 (50)35 (57.4)79 (54.9)
      Type of case
       New case25 (92.6)20 (100)31 (86.1)57 (93.5)133 (92.4)
       Retreatment1 (3.7)04 (11.1)3 (4.9)8 (5.6)
       Transferred in1 (3.7)01 (2.8)1 (1.6)3 (2)
      Health care unit of notification
       Primary health care12 (44.4)7 (35)13 (36.1)39 (63.9)71 (49.3)
       Hospital15 (55.6)13 (65)23 (63.9)22 (36.1)73 (50.7)
      Same residence and health care unit area
       Yes17 (63)9 (45)17 (47.2)44 (72.1)87 (60.4)
       No10 (37)11 (55)19 (52.8)17 (27.9)57 (39.6)
      Comorbidities
       Yes004 (11.1)3 (4.9)7(4.9)
       No27 (100)20 (100)32 (88.9)58 (95.1)137 (95.1)
      HIV test results
       Positive01 (5)1(2.8)2 (3.3)4 (2.8)
       Negative19 (70.4)11 (55)26 (72.2)47 (77.1)103 (71.5)
       Results not available1 (3.7)01 (2.8)1 (1.6)3 (2.1)
       Not performed7 (25.9)8 (40)8 (22.2)11 (18)34 (23.6)
      Outcome
       Cure/treatment completion19 (70.4)12 (60)25 (69.4)47 (77)103 (71.5)
       Loss to follow-up01 (5)2 (5.6)7 (11.5)10 (6.9)
       Death due to TB1 (3.7)1 (5)2 (5.6)1 (1.6)5 (3.5)
       Transferred out6 (22.2)3 (15)2 (5.6)4 (6.6)15(10.5)
       DR-TB
      DR-TB – drug-resistant tuberculosis.
      001 (2.8)01 (0.7)
       missing data1 (3.7)3 (15)4 (11)2 (3.3)10 (6.9)
      a DR-TB – drug-resistant tuberculosis.
      Diagnosis without laboratory investigation was observed among 301/864 (34.8%) patients with PTB and among 48/144 (33.3%) patients with EPTB. It was particularly high among those under five years old with EPTB (15/27 [55.6%]) (Figure 2A,B) when comparing with the other age ranges: 6/20 (30%), 11/36 (30.6%), and 16/61 (26.2%) among those aged 5–9 years, 10–14 years and 15 – 18 years, respectively.
      Figure 2
      Figure 2Laboratory results according to Tuberculosis Forms.
      The proportion of positive results among EPTB patients with laboratory confirmation was 8/12 (66.7%), 10/14 (71.4%), 11/25 (44%), 25/45 (55.6%), among those aged under five years, 5 – 9 years, 10 – 14 years and 15–18 years, respectively. Across all ages, laboratory confirmation was based on histopathology and culture, except for only 3/144 (2.1%) adolescents: two who had Xpert and another one who had SSM positive results (Figure 2A,B). Conversely, laboratory confirmation with Xpert or SSM was performed in 424/864 (49.1%) patients with PTB, mainly among adolescents (403/424 [95%]).

      Discussion

      Our results show the difficulty in diagnosing and treating EPTB. EPTB diagnosis, when compared to PTB, is more likely to be hospital-based and without laboratory confirmation, consistent with other studies with a mixed population of children and adults with EPTB (
      • Pang Y.
      • An J.
      • Shu W.
      • Huo F.
      • Chu N.
      • Gao M.
      • et al.
      Epidemiology of extrapulmonary tuberculosis among inpatients, China, 2008-2017.
      ,
      • Tahseen S.
      • Khanzada F.M.
      • Baloch A.Q.
      • Abbas Q.
      • Bhutto M.M.
      • Alizai A.W.
      • et al.
      Extrapulmonary tuberculosis in Pakistan- A nation-wide multicenter retrospective study.
      ). We also observed an association between EPTB and diagnoses performed outside the patient's residence area and transferred out patients, suggesting a search for additional resources during treatment (
      • Gomes T.
      • Reis-Santos B.
      • Bertolde A.
      • Johnson J.L.
      • Riley L.W.
      • Maciel E.L.
      Epidemiology of extrapulmonary tuberculosis in Brazil: a hierarchical model.
      ). Both findings may represent economic losses to the family, as they may have to go further away from home to be diagnosed or to complete treatment (
      • Jørstad Md
      • Amus J.
      • Marijani M.
      • Sviland L.
      • Mustafa T.
      Diagnostic delay in extrapulmonary tuberculosis and impact on patient morbidity: a study from Zanzibar.
      ).
      Our findings illustrate how the EPTB disease spectrum varies according to the country’s TB burden. Consistent with other studies from high and middle burden countries, the peripheric lymph node form was more common in patients under 15 years old (
      • Carvalho A.C.C.
      • da Silva Martins P.
      • Cardoso C.A.A.
      • Miceli A.L.
      • Martire T.
      • Sant’Anna M de F.B.P.
      • et al.
      Pediatric tuberculosis in the metropolitan area of Rio de Janeiro.
      ,
      • Tahseen S.
      • Khanzada F.M.
      • Baloch A.Q.
      • Abbas Q.
      • Bhutto M.M.
      • Alizai A.W.
      • et al.
      Extrapulmonary tuberculosis in Pakistan- A nation-wide multicenter retrospective study.
      ) and osteoarticular and pleural forms in those over nine years old (
      • Tahseen S.
      • Khanzada F.M.
      • Baloch A.Q.
      • Abbas Q.
      • Bhutto M.M.
      • Alizai A.W.
      • et al.
      Extrapulmonary tuberculosis in Pakistan- A nation-wide multicenter retrospective study.
      ; World Health Organization, 2019). Moreover, the former had a slight increase in prevalence, and the latter was particularly rare in the first five years. Contrary to our findings, a hospital-based study found that pleural TB was the most common EPTB form among those under 15 years old (
      • Wang M.-S.
      • Wang M.-S.
      • Liu X.-J.
      Epidemiological trends in the form of childhood tuberculosis in a referral tuberculosis hospital in Shandong, China.
      ). It is noteworthy that in some studies, pleural TB is categorized within the PTB definition (
      • Santiago-García B.
      • Blázquez-Gamero D.
      • Baquero-Artigao F.
      • Ruíz-Contreras J.
      • Bellón J.M.
      • Muñoz-Fernández M.A.
      • et al.
      Pediatric extrapulmonary tuberculosis: clinical spectrum, risk factors and diagnostic challenges in a low prevalence region.
      ). Nevertheless, in low and middle burden countries, the peripheric lymph node form is the most common EPTB form, regardless of the patient's age (
      • Al-ghafli H.
      • Varghese B.
      • Enani M.
      • Id A.A.
      • Sameera A.l.
      • Albarrak A.
      • et al.
      Demographic risk factors for extrapulmonary tuberculosis among adolescents and adults in Saudi Arabia.
      ,
      • Maltezou H.C.
      • Spyridis P.
      • Kafetzis D.A.
      Extrapulmonary tuberculosis in children.
      ), and osteoarticular forms are more frequent in infancy (
      • Marais B.J.
      • Gie R.P.
      • Schaaf H.S.
      • Hesseling A.C.
      • Obihara C.C.
      • Nelson L.J.
      • et al.
      The clinical epidemiology of childhood pulmonary tuberculosis: a critical review of literature from the pre-chemotherapy era.
      ,
      • Santiago-García B.
      • Blázquez-Gamero D.
      • Baquero-Artigao F.
      • Ruíz-Contreras J.
      • Bellón J.M.
      • Muñoz-Fernández M.A.
      • et al.
      Pediatric extrapulmonary tuberculosis: clinical spectrum, risk factors and diagnostic challenges in a low prevalence region.
      ).
      In line with other studies, TBM predominantly affected those under five years old (60%) (
      • Bang N.D.
      • Caws M.
      • Truc T.T.
      • Duong T.N.
      • Dung N.H.
      • Ha D.T.M.
      • et al.
      Clinical presentations, diagnosis, mortality and prognostic markers of tuberculous meningitis in Vietnamese children: a prospective descriptive study.
      ,
      • Van Well G.T.J.
      • Paes B.F.
      • Terwee C.B.
      • Springer P.
      • Roord J.J.
      • Donald P.R.
      • et al.
      Twenty years of pediatric tuberculous meningitis: a retrospective cohort study in the western cape of South Africa.
      ). Only one case of miliary TB form was found. Although the NTCP guidelines classify this form as PTB (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ), the SINAN-TB database notifies it as EPTB (
      • Brazil
      Sistema de informação de agravos de notificação em tuberculose (SINAN-TB).
      ), making classification very difficult.
      The low number of TBM patients found in our study may be partially explained by the high coverage of BCG vaccination (strain Moreau-Rio de Janeiro) in Rio de Janeiro city (above 100% from 1995 to 2016) (
      • Secretaria Municipal de Saúde do Rio de Janeiro
      Cobertura vacinal BCG nos anos 2014/2016. Coord Geral Articul Estratégica Vigilância Em Saúde; Coord Vigilância Em Saúde; Coord Do Programa Imunizações.
      ). One study found an increased rate of pediatric hospital admissions due to TBM, during the BCG shortage, in South Africa (
      • du Preez K.
      • Seddon J.A.
      • Schaaf H.S.
      • Hesseling A.C.
      • Starke J.R.
      • Osman M.
      • et al.
      Global shortages of BCG vaccine and tuberculous meningitis in children.
      ). Van Toorn et al. observed that children under five years old who died (3.8%) were critically ill on admission, suggesting that there are undetected cases because children die before admission (
      • van Toorn R.
      • Schaaf H.S.
      • Laubscher J.A.
      • van Elsland S.L.
      • Donald P.R.
      • Schoeman J.F.
      Short intensified treatment in children with drug-susceptible tuberculous meningitis.
      ). Our study also suggests that there were undetected cases, as more than half of the diagnoses in children under five years old was performed without laboratory investigation. This represents a significant gap in case detection and TB surveillance. Conversely, this group has a high mortality risk, and treatment based on clinical diagnosis, although not desirable, is one of the strategies to reduce the mortality and morbidity risk in high-burden countries (
      • Jenkins H.E.
      • Yuen C.M.
      • Rodriguez C.A.
      • Nathavitharana R.R.
      • McLaughlin M.M.
      • Donald P.
      • et al.
      mortality in children diagnosed with tuberculosis: a systematic review and meta-analysis.
      ). This reflects the need for high sensitivity diagnostic tests in children.
      In contrast to other studies (
      • Gomes T.
      • Reis-Santos B.
      • Bertolde A.
      • Johnson J.L.
      • Riley L.W.
      • Maciel E.L.
      Epidemiology of extrapulmonary tuberculosis in Brazil: a hierarchical model.
      ), we could not establish any TB-HIV co-infection associations. The likely reasons are the exclusion of patients with EPTB associated with PTB (
      • Mor Z.
      • Pinsker G.
      • Cedar N.
      • Lidji M.
      • Grotto I.
      Epidemiology of extrapulmonary tuberculosis in Israel, 1999-2010.
      ) and the relatively high proportion of patients without HIV testing. This is concerning as TB-HIV co-infection is a risk that fosters higher morbidity and mortality (
      • Jenkins H.E.
      • Yuen C.M.
      • Rodriguez C.A.
      • Nathavitharana R.R.
      • McLaughlin M.M.
      • Donald P.
      • et al.
      mortality in children diagnosed with tuberculosis: a systematic review and meta-analysis.
      ,
      • World Health Organization
      Global Tuberculosis Report 2020.
      ). However, there is an ongoing debate about the HIV-EPTB association (
      • Naing C.
      • Mak J.W.
      • Maung M.
      • Wong S.F.
      • Kassim A.I.B.M.
      Meta-analysis: the association between HIV infection and extrapulmonary tuberculosis.
      ,
      • Shivakoti R.
      • Sharma D.
      • Mamoon G.
      • Pham K.
      Association of HIV infection with extrapulmonary tuberculosis: a systematic review HHS Public Access.
      ).
      Patients with EPTB had a longer time from sample collection to treatment initiation than those with PTB. This difference is probably due to the higher diagnostic complexity of EPTB, especially considering that laboratory confirmation with Xpert or SSM was found in 2.1% of cases (all adolescents) and 49.1% in patients with PTB (95.1% adolescents). Therefore, treatment was likely initiated based on epidemiological history and clinical examination, mainly in those under ten years old with EPTB.
      Since 2002, the NTPC recommends the scoring system, validated nationally, to diagnose PTB for children and adolescents without laboratory investigation or initially negative (
      • Brazil
      Manual de recomendações para o controle da tuberculose no Brasil.
      ). However, SSM was performed in those under ten years old with PTB and EPTB forms, despite not being a national recommendation and Xpert availability. This exemplifies the challenges and controversies surrounding laboratory confirmation in children.
      Even though Xpert was fully available during the study period, it was performed in only 10.4% of EPTB patients and yielded 14.2% positive results. This may reflect an adaptation period to the new NTCP guidelines regarding the use of Xpert, low confidence of health professionals in the methodology, or even limitations associated with the technique (volume and availability of samples). A recent study showed that Xpert has modest sensitivity, mainly in liquid samples in children (
      • Seo Y.S.
      • Kang J.M.
      • Kim D.S.
      • Ahn J.G.
      Xpert MTB/RIF assay for diagnosis of extrapulmonary tuberculosis in children: a systematic review and meta-analysis.
      ). This underscores the importance of good quality samples in children to achieve positive results, as they are paucibacillary and adequate volumes are difficult, contributing to the difficulty in diagnosis. Thus, well-trained health care professionals and high-quality materials are desirable, though not always available, and are as important as the sensitivity of the test to achieving positive results.
      We observed that the proportion of treatment success among EPTB was lower than that of PTB patients; however, both were below the overall 84% treatment success rate in high-burden countries (
      • World Health Organization
      Global Tuberculosis Report 2020.
      ).
      The time from diagnosis to LTFU was similar probably because patients with PTB and some EPTB forms face the same treatment routine and challenges in health care units, pointing out that TB as a whole is a significant challenge for treatment adherence (
      • Ambresin A.E.
      • Bennett K.
      • Patton G.C.
      • Sanci L.A.
      • Sawyer S.M.
      Assessment of youth-friendly health care: a systematic review of indicators drawn from young people’s perspectives.
      ).
      LTFU and death due to TB were the most prevalent unfavorable outcomes in PTB and EPTB, respectively. This is likely due to the predominance of adolescents in the PTB group, and LTFU is frequent in this group (
      • Enane L.A.
      • Lowenthal E.D.
      • Arscott-Mills T.
      • Matlhare M.
      • Smallcomb L.S.
      • Kgwaadira B.
      • et al.
      Loss to follow-up among adolescents with tuberculosis in Gaborone, Botswana.
      ). Moreover, an extreme vulnerability was more common in PTB because adolescents display more risky behaviors. This is probably the same reason why “death due to other causes” occurred only among patients with PTB. In another study, 53.8% of adolescents died from HIV infection and 23.8% from violence among those with TB (
      • de Oliveira M.C.B.
      • Sant’Anna C.C.
      • Raggio L.R.
      • Kritski A.L.
      Unfavorable outcomes in tuberculosis: multidimensional factors among adolescents in Rio de Janeiro, Brazil.
      ). Taken together, our results underline the relevance of programmatic actions in TB control strategy specific to this population.
      Regarding the EPTB group, 80% of deaths were due to TBM, and 50% of TBM patients died in the first 14 days of diagnosis. Furthermore, TBM outcome is highly associated with the disease stage. Chiang et al. found that the risk of death in stage 3 was 34.1 (95% CI 21.9 – 48.9), in which the patient shows severe neurologic signs, was higher than in stage 1 (1.3; 95% CI 0.1 – 13.1), in which the patient is alert and without focal neurological signs and thus clinical diagnosis is more difficult (
      • Chiang S.S.
      • Khan F.A.
      • Milstein M.B.
      • Tolman A.W.
      • Benedetti A.
      • Starke J.R.
      • et al.
      Treatment outcomes of childhood tuberculous meningitis: a systematic review and meta-analysis.
      ). Therefore, our results suggest a substantial delay in the diagnosis and subsequently in treatment, posing a significant challenge to the End TB strategy of zero deaths (
      • World Health Organization
      Global Tuberculosis Report 2020.
      ).
      The remaining eight patients with TBM faced a significant morbidity risk. Other authors found that 53.9% of the surviving patients suffered from neurological sequelae (
      • Chiang S.S.
      • Khan F.A.
      • Milstein M.B.
      • Tolman A.W.
      • Benedetti A.
      • Starke J.R.
      • et al.
      Treatment outcomes of childhood tuberculous meningitis: a systematic review and meta-analysis.
      ). Thus, the high proportion of treatment success among EPTB patients masks the true burden of death due to TBM (
      • Tahseen S.
      • Khanzada F.M.
      • Baloch A.Q.
      • Abbas Q.
      • Bhutto M.M.
      • Alizai A.W.
      • et al.
      Extrapulmonary tuberculosis in Pakistan- A nation-wide multicenter retrospective study.
      ).
      The strength of our study is the reliability and quality of data retrieved from two national databases reflecting the reality of EPTB in Rio de Janeiro city without selection bias of the health care facility. In Brazil, TB treatment is free of charge and is only offered in the public health system. A notification in the SINAN database is necessary for the patient to receive TB treatment. Thus, our data provide a very reliable picture of the pragmatic TB health care during this timeframe in Rio de Janeiro. In addition, we analyzed data from 144 patients affected only by EPTB without association with PTB, a higher number than other studies that only focused on this population. Moreover, we disaggregated the age as recommended by the WHO guidelines, which may encourage future studies to adopt this data presentation. At the same time, there are limitations concerning the lack of clinical data. Miliary TB patients may have been missed due to misclassification as PTB in the SINAN database. Missing information is also an inherent problem in studies that use secondary data.
      Lastly, EPTB is a challenge for clinical and laboratory diagnoses posing a threat to mortality and morbidity due to delay in treatment initiation (
      • Jørstad Md
      • Amus J.
      • Marijani M.
      • Sviland L.
      • Mustafa T.
      Diagnostic delay in extrapulmonary tuberculosis and impact on patient morbidity: a study from Zanzibar.
      ,
      • Pang Y.
      • An J.
      • Shu W.
      • Huo F.
      • Chu N.
      • Gao M.
      • et al.
      Epidemiology of extrapulmonary tuberculosis among inpatients, China, 2008-2017.
      ). This study highlights the difficulties in TB diagnosis when the presentation is atypical or in patients with paucibacillary forms. Although Xpert Ultra fills an important gap in the laboratory diagnosis of paucibacillary TB forms, it will not address the clinical delay from identifying TB to confirmation and subsequent initiation of adequate treatment.

      Author’s contributions

      All authors have made substantial contributions to all of the following: the conception and design of the study, or acquisition of data, or analysis and interpretation of data, drafting the article or revising it critically for important intellectual content, and final approval of the version to be submitted.

      Declaration of interest

      On behalf of all authors, the corresponding author states that there is no conflict of interest.

      Funding sources

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      Acknowledgements

      ALK and CCS are members of REDE TB (Rede de Pesquisa em Tuberculose) and are supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq).

      Appendix A. Supplementary data

      References

        • Al-ghafli H.
        • Varghese B.
        • Enani M.
        • Id A.A.
        • Sameera A.l.
        • Albarrak A.
        • et al.
        Demographic risk factors for extrapulmonary tuberculosis among adolescents and adults in Saudi Arabia.
        PLoS One. 2019; 14: 1-14https://doi.org/10.1371/journal.pone.0213846
        • Ambresin A.E.
        • Bennett K.
        • Patton G.C.
        • Sanci L.A.
        • Sawyer S.M.
        Assessment of youth-friendly health care: a systematic review of indicators drawn from young people’s perspectives.
        J Adolesc Heal. 2013; 52: 670-681https://doi.org/10.1016/j.jadohealth.2012.12.014
        • Bang N.D.
        • Caws M.
        • Truc T.T.
        • Duong T.N.
        • Dung N.H.
        • Ha D.T.M.
        • et al.
        Clinical presentations, diagnosis, mortality and prognostic markers of tuberculous meningitis in Vietnamese children: a prospective descriptive study.
        BMC Infect Dis. 2016; 16: 1-10https://doi.org/10.1186/s12879-016-1923-2
        • Brazil
        Boletim epidemiológico especial tuberculose.
        Secr Vigilância Em Saúde, Ministério Da Saúde2020: 40
        • Brazil
        Sistema de informação de agravos de notificação em tuberculose (SINAN-TB).
        Ministério Da Saúde, 2020 (accessed October 20, 2020)
        • Brazil
        Manual de recomendações para o controle da tuberculose no Brasil.
        second edi. Ministério da Saúde, Brasilia2019
        • Carvalho A.C.C.
        • da Silva Martins P.
        • Cardoso C.A.A.
        • Miceli A.L.
        • Martire T.
        • Sant’Anna M de F.B.P.
        • et al.
        Pediatric tuberculosis in the metropolitan area of Rio de Janeiro.
        Int J Infect Dis. 2020; 98: 299-304https://doi.org/10.1016/j.ijid.2020.06.070
        • Chiang S.S.
        • Khan F.A.
        • Milstein M.B.
        • Tolman A.W.
        • Benedetti A.
        • Starke J.R.
        • et al.
        Treatment outcomes of childhood tuberculous meningitis: a systematic review and meta-analysis.
        Lancet Infect Dis. 2014; 14: 947-957https://doi.org/10.1016/S1473-3099(14)70852-7
        • Enane L.A.
        • Lowenthal E.D.
        • Arscott-Mills T.
        • Matlhare M.
        • Smallcomb L.S.
        • Kgwaadira B.
        • et al.
        Loss to follow-up among adolescents with tuberculosis in Gaborone, Botswana.
        Int J Tuberc Lung Dis. 2016; 20: 1320-1325https://doi.org/10.5588/ijtld.16.0060
        • Gomes T.
        • Reis-Santos B.
        • Bertolde A.
        • Johnson J.L.
        • Riley L.W.
        • Maciel E.L.
        Epidemiology of extrapulmonary tuberculosis in Brazil: a hierarchical model.
        BMC Infect Dis. 2014; 14: 9https://doi.org/10.1186/1471-2334-14-9
        • Jenkins H.E.
        • Yuen C.M.
        • Rodriguez C.A.
        • Nathavitharana R.R.
        • McLaughlin M.M.
        • Donald P.
        • et al.
        mortality in children diagnosed with tuberculosis: a systematic review and meta-analysis.
        Lancet Infect Dis. 2017; 17: 285-295https://doi.org/10.1016/S1473-3099(16)30474-1
        • Jesus R de
        • Guimarães R.P.
        • Bergamo R.
        • Santos LCF dos
        • Matta ASD da
        • Paula Júnior FJ de
        Sistema Gerenciador de Ambiente Laboratorial: relato de experiência de uma ferramenta transformadora para a gestão laboratorial e vigilância em saúde.
        Epidemiol e Serviços Saúde. 2013; 22: 525-529https://doi.org/10.5123/s1679-49742013000300018
        • Jørstad Md
        • Amus J.
        • Marijani M.
        • Sviland L.
        • Mustafa T.
        Diagnostic delay in extrapulmonary tuberculosis and impact on patient morbidity: a study from Zanzibar.
        PLoS One. 2018; 13: 1-17https://doi.org/10.1371/journal.pone.0203593
        • Kulchavenya E.
        Extrapulmonary tuberculosis: are statistical reports accurate?.
        Ther Adv Infect Dis. 2014; 2: 61-70https://doi.org/10.1177/2049936114528173
        • Maltezou H.C.
        • Spyridis P.
        • Kafetzis D.A.
        Extrapulmonary tuberculosis in children.
        Arch Dis Child. 2000; 83: 342-346https://doi.org/10.1136/adc.83.4.342
        • Marais B.J.
        • Gie R.P.
        • Schaaf H.S.
        • Hesseling A.C.
        • Obihara C.C.
        • Nelson L.J.
        • et al.
        The clinical epidemiology of childhood pulmonary tuberculosis: a critical review of literature from the pre-chemotherapy era.
        Int J Tuberc Lung Dis. 2004; 8: 392-402
        • Mor Z.
        • Pinsker G.
        • Cedar N.
        • Lidji M.
        • Grotto I.
        Epidemiology of extrapulmonary tuberculosis in Israel, 1999-2010.
        Int J Tuberc Lung Dis. 2013; 17: 229-233https://doi.org/10.5588/ijtld.12.0375
        • Naing C.
        • Mak J.W.
        • Maung M.
        • Wong S.F.
        • Kassim A.I.B.M.
        Meta-analysis: the association between HIV infection and extrapulmonary tuberculosis.
        Lung. 2013; 191: 27-34https://doi.org/10.1007/s00408-012-9440-6
        • de Oliveira M.C.B.
        • Sant’Anna C.C.
        • Raggio L.R.
        • Kritski A.L.
        Unfavorable outcomes in tuberculosis: multidimensional factors among adolescents in Rio de Janeiro, Brazil.
        Am J Trop Med Hyg. 2020; 103: 2492-2500https://doi.org/10.4269/ajtmh.20-0209
        • Pang Y.
        • An J.
        • Shu W.
        • Huo F.
        • Chu N.
        • Gao M.
        • et al.
        Epidemiology of extrapulmonary tuberculosis among inpatients, China, 2008-2017.
        Emerg Infect Dis. 2019; 25: 457-464https://doi.org/10.3201/eid2503.180572
        • du Preez K.
        • Seddon J.A.
        • Schaaf H.S.
        • Hesseling A.C.
        • Starke J.R.
        • Osman M.
        • et al.
        Global shortages of BCG vaccine and tuberculous meningitis in children.
        Lancet Glob Heal. 2019; 7: e28-9https://doi.org/10.1016/S2214-109X(18)30474-1
        • Sandgren A.
        • Hollo V.
        • van der Werf M.J.
        Extrapulmonary tuberculosis in the European union and European economic area, 2002 to 2011.
        Eurosurveillance. 2013; 18: 1-9https://doi.org/10.2807/ese.18.12.20431-en
        • Santiago-García B.
        • Blázquez-Gamero D.
        • Baquero-Artigao F.
        • Ruíz-Contreras J.
        • Bellón J.M.
        • Muñoz-Fernández M.A.
        • et al.
        Pediatric extrapulmonary tuberculosis: clinical spectrum, risk factors and diagnostic challenges in a low prevalence region.
        Pediatr Infect Dis J. 2016; 35: 1175-1181https://doi.org/10.1097/INF.0000000000001270
        • Secretaria Municipal de Saúde do Rio de Janeiro
        Cobertura vacinal BCG nos anos 2014/2016. Coord Geral Articul Estratégica Vigilância Em Saúde; Coord Vigilância Em Saúde; Coord Do Programa Imunizações.
        2020
        • Seo Y.S.
        • Kang J.M.
        • Kim D.S.
        • Ahn J.G.
        Xpert MTB/RIF assay for diagnosis of extrapulmonary tuberculosis in children: a systematic review and meta-analysis.
        BMC Infect Dis. 2020; 20: 1-10https://doi.org/10.1186/s12879-019-4745-1
        • Shivakoti R.
        • Sharma D.
        • Mamoon G.
        • Pham K.
        Association of HIV infection with extrapulmonary tuberculosis: a systematic review HHS Public Access.
        Infection. 2017; 45: 11-21https://doi.org/10.1007/s15010-016-0960-5
        • Tahseen S.
        • Khanzada F.M.
        • Baloch A.Q.
        • Abbas Q.
        • Bhutto M.M.
        • Alizai A.W.
        • et al.
        Extrapulmonary tuberculosis in Pakistan- A nation-wide multicenter retrospective study.
        PLoS One. 2020; 15: 1-16https://doi.org/10.1371/journal.pone.0232134
        • van Toorn R.
        • Schaaf H.S.
        • Laubscher J.A.
        • van Elsland S.L.
        • Donald P.R.
        • Schoeman J.F.
        Short intensified treatment in children with drug-susceptible tuberculous meningitis.
        Pediatr Infect Dis J. 2014; 33: 248-252https://doi.org/10.1097/INF.0000000000000065
        • Wang M.-S.
        • Wang M.-S.
        • Liu X.-J.
        Epidemiological trends in the form of childhood tuberculosis in a referral tuberculosis hospital in Shandong, China.
        Biomed Res Int. 2020; (Aug)6142567https://doi.org/10.1155/2020/6142567
        • Van Well G.T.J.
        • Paes B.F.
        • Terwee C.B.
        • Springer P.
        • Roord J.J.
        • Donald P.R.
        • et al.
        Twenty years of pediatric tuberculous meningitis: a retrospective cohort study in the western cape of South Africa.
        Pediatrics. 2009; 123: 1-8https://doi.org/10.1542/peds.2008-1353
        • World Health Organization
        Global Tuberculosis Report 2020.
        WHO, 2020: 232 (accessed October 1, 2020)
        • World Health Organization
        Global Tuberculosis Report.
        WHO, 2019: 297 (accessed June 22, 2020)
        • World Health Organization
        Definitions and Reporting Framework for Tuberculosis – 2013 Revision.
        WHO, 2014: 1-47 (accessed June 13, 2020)
        • World Health Organization
        Guidance for National Tuberculosis Programmes on the Management of Tuberculosis in Children.
        second ed. WHO, 2014 (Accessed June 9, 2020)

      Linked Article