Despite progress in reducing tuberculosis (TB) incidence and mortality in the past 20 years, TB is a top ten cause of death in children under 5 years worldwide. However, childhood TB remains massively underreported and undiagnosed, mostly because of the challenges in confirming its diagnosis due to the paucibacillary nature of the disease and the difficulty in obtaining expectorated sputum in children.

Pneumonia is the leading cause of death in children under the age of 5 years worldwide. There is growing evidence that, in high TB burden settings, TB is common in children with pneumonia, with up to 23% of those admitted to hospital with an initial diagnosis of pneumonia later being diagnosed as TB. However, the current World Health Organization (WHO) standard of care (SOC) for young children with pneumonia considers a diagnosis of TB only if the child has a history of prolonged symptoms or fails to respond to antibiotic treatments. Hence, TB is often under-diagnosed or diagnosed late in children presenting with pneumonia.

In this context, the investigators are proposing to assess the impact on mortality of adding the systematic early detection of TB using Xpert MTB/RIF Ultra, performed on NPAs and stool samples, to the WHO SOC for children with severe pneumonia, followed by immediate initiation of anti-TB treatment in children testing positive on any of the samples.

TB-Speed Pneumonia is a multicentric, stepped wedge diagnostic trial conducted in six countries with high TB incidence: Cote d'Ivoire, Cameroon, Uganda, Mozambique, Zambia and Cambodia.

The sub-study on Covid-19 will assess the prevalence and impact of the Covid-19 in young children hospitalized with severe pneumonia. The sub-study findings are expected to guide policy makers and clinicians on potential specific screening and management measures for these vulnerable groups of children. They are also key to analysing TB-Speed Pneumonia results on mortality in a context of the Covid-19 outbreak and to take into consideration SARS-CoV-2 infection status in the main study analysis.

Pneumonia is the leading cause of death in children under the age of 5 years worldwide.

There is growing evidence that, in high TB burden settings, TB is common in children with pneumonia, with up to 23% of those admitted to hospital with an initial diagnosis of pneumonia later being diagnosed as TB]. This is particularly true in the African and Asian WHO regions, which accounted for 30% and 35% of all paediatric TB cases in 2016 respectively.

In these regions, the case fatality rate for childhood pneumonia associated with TB is high, ranging from 4% to 21% [8], with younger age, malnutrition and HIV infection increasing the risk of death.

The current standard of care (SOC) for young children with pneumonia considers a diagnosis of TB only if the child has a history of prolonged symptoms or fails to respond to antibiotic treatments. Hence, TB is often under-diagnosed or diagnosed late in children presenting with pneumonia. Although TB is a chronic disease in adults, recent data show that the duration of respiratory symptoms before admission can be acute in children with severe pneumonia associated with TB [8]. Hence, Identifying TB cases early and shortening the diagnostic delay to initiate appropriate TB treatment in children with clinical presentation of severe acute pneumonia is likely to reduce mortality.

An improved molecular diagnostic tool for paediatric TB:

Xpert MTB/RIF (Cepheid, USA) is an automated nucleic acid amplification test (NAAT) that simultaneously detects Mycobacterium tuberculosis (MTB) and genes associated with resistance to rifampicin. The assay was a major breakthrough in bringing molecular tests for the diagnosis of TB closer to the community, with performances close to mycobacterial culture. WHO therefore recommended Xpert MTB/RIF as the first test to be used for the diagnosis of TB among populations who may have drug-resistant and/or HIV-associated TB.

In 2013, WHO updated its policy to include Xpert MTB/RIF as the initial test for the diagnosis of TB in children, based on a systematic review and meta-analysis showing a pooled sensitivity of 66% (CI 95% 51-81) and a specificity of 98% (CI 95% 96-99) of Xpert MTB/RIF performed on gastric lavages when compared with culture. WHO recommendations, detailed in the 2014 guidance on paediatric TB, stated that Xpert MTB/RIF may be used instead of smear microscopy in all children and should be used in children with HIV infection or presumptive multidrug-resistant TB.

Although data on the performance of Xpert MTB/RIF in children with pneumonia is limited, in Bangladesh, sensitivity on gastric aspirates or sputum samples compared to culture in this group was equivalent to that reported in other studies.

The next-generation of Xpert MTB/RIF assay, Xpert MTB/RIF Ultra (Ultra), has a limit of detection of 16 colony forming units (CFU)/mL (compared to the current version which detects 130 CFU/mL), representing an approximately 8-fold improvement. This lower threshold is similar to the detection level of culture and would facilitate the rapid diagnosis of paucibacillary TB disease as seen in children. Retrospective analyses on frozen respiratory samples in children have shown a sensitivity of 71% for Ultra versus 47% for Xpert MTB/RIF. A recently published study reveals, however, a lower specificity of U ltra in adults, particularly in those with a previous history of TB, potentially resulting in false diagnoses and overtreatment of TB. Though further prospective studies are needed, the risk of false-positive results could be less significant in children as only a small proportion of children have previously had TB. The estimated clinical impact of Ultra is therefore likely to vary depending on the settings, with a recent modelling exercise finding a larger mortality benefit in patient populations with high TB prevalence, high HIV prevalence, and high case fatality ratios for untreated TB.

The current WHO recommendations for the use of Xpert MTB/RIF also apply to the use of Ultra as the initial diagnostic test for all adults and children with signs and symptoms of TB. An update of the current guidelines for the use of Ultra is planned for 2018.

Alternative specimen collection methods adapted to children:

Young children are frequently unable to expectorate sputum and there is no clear evidence and guidance on which specimen or combination of specimens should be used in order to maximize the probability of bacteriological confirmation of TB in children. At the programmatic level, implementation of gastric aspirates and induced sputum can be challenging].

Our research group and other groups in Africa and Asia have shown that alternative specimen collection methods such as nasopharyngeal aspirates (NPA) and stool samples are easier to be implemented in resource-limited settings and are better tolerated in young and sick children. These methods do not require a child to fast (as for gastric aspirates) and are more suitable than induced sputum in children with severe respiratory deficit. In children with presumptive TB, Xpert MTB/RIF has a sensitivity on NPAs close to the one achieved with induced sputum [28], [35]. Recent studies have shown similar sensitivity of Xpert MTB/RIF on the combination of one stool and one NPA as compared to two induced sputum or two gastric aspirates. Stool testing by Xpert MTB/RIF shows results close to respiratory samples in terms of sensitivity but requires a simplified specimen processing methodology for further field use. The flotation method, based on Sheather's sucrose solution used in the PAANTHER study, showed promising results but relies on centrifugation and other labour-intensive processes. Stool processing will be further optimized in Output 4 of the TB-Speed project to enable implementation at a lower healthcare level.

Most studies on childhood TB diagnostics are early proof-of-concepts or studies that evaluate diagnostic accuracy, collecting and testing multiples samples using existing microbiological tests. Implementation studies, with patient health outcomes as the primary endpoint of interest, are seldom implemented despite the need for such studies to inform policies. The WHO recommendation to use Xpert MTB/RIF in children was based on diagnostic accuracy, but evidence of its impact on TB outcome has not been evaluated. As in adults, its use in children may have a limited impact on outcome in children with a strong suspicion of TB, due to the common use of empirical treatment in these populations [48]-[51]. However, this is likely to be different in children with severe pneumonia presenting with acute symptoms, for which TB is usually suspected only after empiric antibiotic treatment has been shown to be ineffective.

The TB-Speed approach to early detection of TB in children with severe pneumonia:

In line with the strategies advocated by the National Tuberculosis Programmes (NTPs) in participating countries, the TB-Speed project aims not only to contribute to the reduction of TB-associated mortality, but also to initiate an innovative approach for the early detection and treatment of TB in young children with severe pneumonia to enable access to high-quality healthcare in this highly vulnerable group.

Our hypothesis is that in high TB burden countries, testing young children with severe pneumonia for TB and starting those who test positive on anti-TB treatment on the day of presentation, could reduce all-cause mortality through reduction of mortality attributed to TB. In this context, the investigators are proposing a research study to assess the impact on mortality of adding the systematic early detection of TB using Ultra performed on NPAs and stool samples to the WHO SOC for children with severe pneumonia, followed by immediate initiation of anti-TB treatment in children testing positive on any of the samples. If successful, this intervention could be systematically implemented at district hospital level where children with severe pneumonia are referred. Furthermore, the investigators hypothesize that the intervention will raise TB awareness among clinicians and may lead to more empirical TB treatment initiated as compared to the control. This further justifies the stepped-wedge design chosen for this study.

From a health economics perspective, the investigators hypothesize that benefits in terms of survival and Disability Adjusted Life Years (DALYs) will outweigh extra costs incurred by systematic Ultra testing in children with severe pneumonia.

Mortality in children aged 2 to 59 months hospitalized with severe pneumonia and mortality attributable to TB:

There were an estimated 120 million pneumonia episodes in children younger than 5 years in 2011, including 14 million severe episodes. 1.3 million pneumonia episodes led to death. In 2014, WHO revised definitions for pneumonia and severe pneumonia, considering 'chest-indrawing pneumonia' as non-severe and downgrading 'pneumonia with severity signs' from very severe to severe. Ambulatory care and oral antibiotics are now recommended for 'chest-indrawing pneumonia' and now only severe pneumonia justifies referral from primary health centres to district hospitals for intra-venous (IV) antibiotics and oxygen therapy if needed.

Participating countries and added value of the multi-country aspect:

The impact of this innovative approach may vary with TB incidence as well as some geographical and seasonal variability that can affect the prevalence and aetiology of pneumonia in young children. To provide a better basis for the generalizability of results, the project will take place in six countries with different epidemiological and environmental backgrounds, in Sub-Saharan Africa (Cameroon, Cote d'Ivoire, Mozambique, Uganda, and Zambia) and South East Asia (Cambodia). Cambodia, Zambia and Mozambique are among the 30 high TB burden countries according to WHO classification.

The sub-study on Covid-19 will assess the prevalence and impact of the Covid-19 in young children hospitalized with severe pneumonia. The sub-study findings are expected to guide policy makers and clinicians on potential specific screening and management measures for these vulnerable groups of children. They are also key to analysing TB-Speed Pneumonia results on mortality in a context of the Covid-19 outbreak and to take into consideration SARS-CoV-2 infection status in the main study analysis. The sub-study will be implemented in all participating countries except Cambodia.

At the time of enrolment in the TB-Speed Pneumonia main study, patients will be also offered to participate to the Covid ancillary study. Children will be tested for SARS-Cov-2 on the day of enrolment. The duration of enrolment in the sub-study will be of 6 months in total. Duration of follow-up for a particular participant will be 12 weeks, as in the main Pneumonia study.

Inclusion criteria:

1. Children aged 2 to 59 months

2. Newly hospitalized for severe pneumonia defined using WHO criteria as cough or difficulty in breathing with:

   1. Peripheral oxygen saturation < 90% or central cyanosis, or

   2. Severe respiratory distress (e.g. grunting, nasal flaring, very severe chest indrawing), or

   3. Signs of pneumonia, defined as cough or difficulty in breathing with fast breathing (tachypnea) and/or chest indrawing, with any of the following danger signs:

      • Inability to breastfeed or drink,
      • Persistent vomiting
      • Lethargy or reduced level of consciousness
      • Convulsions,
      • Stridor in calm child
      • Severe malnutrition

3. Informed consent signed by parent/guardian

Exclusion criteria:

- Ongoing TB treatment or history of intake of anti-TB drugs in the last 6 months