Articles
Mortality in children diagnosed with tuberculosis: a systematic review and meta-analysis

https://doi.org/10.1016/S1473-3099(16)30474-1Get rights and content

Summary

Background

Case fatality ratios in children with tuberculosis are poorly understood—particularly those among children with HIV and children not receiving tuberculosis treatment. We did a systematic review of published work to identify studies of population-representative samples of paediatric (ie, <15 years) tuberculosis cases.

Methods

We searched PubMed and Embase for reports published in English, French, Portuguese, or Spanish before Aug 12, 2016, that included terms related to tuberculosis, children, mortality, and population representativeness. We also reviewed our own files and reference lists of articles identified by this search. We screened titles and abstracts for inclusion, excluding studies in which outcomes were unknown for 10% or more of the children and publications detailing non-representative samples. We used random-effects meta-analysis to produce pooled estimates of case fatality ratios from the included studies, which we divided into three eras: the pre-treatment era (ie, studies before 1946), the middle era (1946–80), and the recent era (after 1980). We stratified our analyses by whether or not children received tuberculosis treatment, age (0–4 years, 5–14 years), and HIV status.

Findings

We identified 31 papers comprising 35 datasets representing 82 436 children with tuberculosis disease, of whom 9274 died. Among children with tuberculosis included in studies in the pre-treatment era, the pooled case fatality ratio was 21·9% (95% CI 18·1–26·4) overall. The pooled case fatality ratio was significantly higher in children aged 0–4 years (43·6%, 95% CI 36·8–50·6) than in those aged 5–14 years (14·9%, 11·5–19·1). In studies in the recent era, when most children had tuberculosis treatment, the pooled case fatality ratio was 0·9% (95% CI 0·5–1·6). US surveillance data suggest that the case fatality ratio is substantially higher in children with HIV receiving treatment for tuberculosis (especially without antiretroviral therapy) than in those without HIV.

Interpretation

Without adequate treatment, children with tuberculosis, especially those younger than 5 years, are at high risk of death. Children with HIV have an increased mortality risk, even when receiving tuberculosis treatment.

Funding

US National Institutes of Health, Janssen Global Public Health.

Introduction

Increased attention is being paid to the long-neglected global epidemic of childhood tuberculosis, and there is a growing consensus that far more children have tuberculosis than was previously thought, most of whom are undiagnosed and thus untreated.1, 2, 3 Around 1 million children are estimated to develop tuberculosis annually, of whom only 36% are notified.3 The greatest case detection failure occurs in children younger than 5 years, who probably account for around half of all paediatric cases.2 The mortality associated with these more than 500 000 new cases of tuberculosis each year is largely preventable and could contribute more than previously thought to global mortality in children younger than 5 years.4

Given that so many children with tuberculosis are never diagnosed, estimation of childhood tuberculosis mortality presents unique challenges. Cohort studies from the past three decades have included very few children who did not receive tuberculosis treatment. Mortality estimates (including the WHO global estimate of 136 000 annual tuberculosis deaths among children3) often rely on vital registration data,3 but attribution of deaths to undiagnosed tuberculosis posthumously is hampered by autopsy costs and the unreliability of vital records and verbal autopsy.5 Because symptoms in children are generally non-specific, many deaths caused by tuberculosis could be erroneously attributed to more common diseases, such as pneumonia.6 An alternative source of information on children who did not receive tuberculosis treatment is the scientific literature published before the discovery of antituberculosis chemotherapy.

Quantification of the mortality impact of the global failure to diagnose and treat childhood tuberculosis necessitates a better understanding of the mortality risks associated with the disease in relevant age groups, with and without access to appropriate treatment. Therefore, we did a systematic review and meta-analysis to estimate case fatality ratios—ie, the proportion of children with tuberculosis who die—stratified by treatment, age group, and HIV infection.

Section snippets

Search strategy

We first searched for systematic reviews of tuberculosis mortality or outcomes in children with all forms of tuberculosis. We found only systematic reviews of outcomes among children with tuberculous meningitis7 and children with multidrug-resistant tuberculosis (ie, strains of tuberculosis that are resistant to isoniazid and rifampicin).8 In both of these reviews, only cohorts receiving treatment were included. We then searched for studies in which deaths among children with tuberculosis were

Results

Of 1087 unique citations found through the database search, 253 studies were eligible for full-text review, 22 of which were included in our analysis (figure 1). Three of these papers were included only after we received additional information from their authors. We also reviewed 89 papers identified in reference lists of reviewed papers or obtained from authors' personal libraries, and included nine of these papers in our analysis.

The 31 included papers10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20

Discussion

To our knowledge, ours is the first comprehensive analysis of case fatality ratios among children with tuberculosis. We found that children receiving adequate treatment had a low mortality risk. By contrast, before the widespread availability of antituberculosis chemotherapy and the BCG vaccination, children with tuberculosis had far worse outcomes; one in five died overall, and 40% of children younger than 5 years old died. Case fatality ratios in young children (aged 0–4 years) were

References (62)

  • BJ Marais et al.

    Interrupted BCG vaccination is a major threat to global child health

    Lancet Respir Med

    (2016)
  • CM Yuen et al.

    Turning off the tap: stopping tuberculosis transmission through active case-finding and prompt effective treatment

    Lancet

    (2015)
  • MX Rangaka et al.

    Controlling the seedbeds of tuberculosis: diagnosis and treatment of tuberculosis infection

    Lancet

    (2015)
  • RA Oberhelman et al.

    Diagnostic approaches for paediatric tuberculosis by use of different specimen types, culture methods, and PCR: a prospective case-control study

    Lancet Infect Dis

    (2010)
  • SS Chiang et al.

    New diagnostics for childhood tuberculosis

    Infect Dis Clin North Am

    (2015)
  • Global Tuberculosis Report 2015

    (2015)
  • PW Setel et al.

    Validity of verbal autopsy procedures for determining cause of death in Tanzania

    Trop Med Int Health

    (2006)
  • PR Donald et al.

    Requirements for the clinical evaluation of new anti-tuberculosis agents in children

    Int J Tuberc Lung Dis

    (2013)
  • Tuberculosis in Kenya: follow-up of the second (1974) national sampling survey and a comparison with the follow-up data from the first (1964) national sampling survey. An East African and British Medical Research Council co-operative investigation

    Tubercle

    (1979)
  • Tuberculosis in Tanzania: a follow-up of a national sampling survey of drug resistance and other factors

    Tubercle

    (1977)
  • J Minion et al.

    Multidrug and extensively drug-resistant tuberculosis in Canada 1997–2008: demographic and disease characteristics

    PLoS One

    (2013)
  • G Abreu et al.

    Cuba's strategy for childhood tuberculosis control, 1995–2005

    MEDICC Rev

    (2011)
  • TC Rodwell et al.

    Tuberculosis from Mycobacterium bovis in binational communities, United States

    Emerg Infect Dis

    (2008)
  • NS Shah et al.

    Epidemiology of smear-negative pulmonary tuberculosis in the United States, 1993–2008

    Int J Tuberc Lung Dis

    (2012)
  • LT Nguyen et al.

    Mortality before or during treatment among tuberculosis patients in North Carolina, 1993–2003

    Int J Tuberc Lung Dis

    (2011)
  • SR Rosenthal et al.

    BCG vaccination in tuberculous households

    Am Rev Respir Dis

    (1961)
  • JG Townsend et al.

    Tuberculosis control among the North American Indians

    Am Rev Tuberc

    (1942)
  • G Drolet

    Present trend of case fatality rates in tuberculosis

    Am Rev Tuberc

    (1938)
  • RR Puffer et al.

    Tuberculosis attack and death rates of household associates; the influence of age, sex, race, and relationship

    Am Rev Tuberc

    (1952)
  • JW Hopkins

    BCG vaccination in Montreal: statistical analysis of the results of research by Dr JA Baudouin on BCG vaccination in Montreal

    Am Rev Tuberc

    (1941)
  • JA Myers

    The natural history of tuberculosis in the human body. I. The demonstrable primary pulmonary infiltrate

    Am Rev Tuberc

    (1959)
  • Cited by (175)

    View all citing articles on Scopus

    These authors contributed equally to this work

    View full text