Subclinical Pulmonary Tuberculosis: Prevalence Among Adults in Community Settings

In countries such as those in Africa and Asia known to have a high burden of tuberculosis, majorities of individuals with undiagnosed pulmonary tuberculosis do not report coughing, according to new findings.¹

These findings suggest that diagnostic pathways based exclusively on cough as the initiating symptom may end up delaying patients’ diagnosis and eventual treatment, potentially leading to worse outcomes. This research was the result of 12 nationally-representative surveys and a subsequent meta-analysis to more clearly define subclinical pulmonary tuberculosis and its prevalence.²

The research was led by Logan Stuck, PhD, from the department of global health at Amsterdam University Medical Centers in the Netherlands.

“We aimed to investigate the prevalence of subclinical pulmonary tuberculosis according to different case definitions and explore its association with demographic characteristics, sputum bacterial load, and selected tuberculosis risk factors,” Stuck and colleagues wrote.

Background and Methods

The investigators looked at the national tuberculosis prevalence surveys which had been given out to participants in the period from 2007 - 2019 in collaboration with the World Health Organization (WHO). They adhered to the standardized protocols and methods required for data collection.

The research team’s choice of surveys was based upon a WHO report that had listed national tuberculosis prevalence surveys given to participants in the period from 2007 - 2016. The datasets that the team evaluated were then harmonized and consolidated into what would become a unified set of data.

The investigators implemented screenings of chest x-rays and symptoms, with diagnostic criteria being consistent across the surveys and tuberculosis being defined through positive Mycobacterium tuberculosis sputum culture. The teams’ analysis of the data maintained a focus on the prevalence of pulmonary tuberculosis among subjects in the age range of 15 years and older.

The investigators utilized 3 case definitions for their estimation of subclinical tuberculosis proportions:

• The absence of persistent cough (≥2 weeks);
• Absence of cough;
• Absence of symptoms (lack of fever, cough, in chest pain, night sweats, and loss of weight)

The research team assessed these case definitions using both unadjusted and adjusted formats and they accounted for false-negative chest x-rays and inconclusive culture findings.

Findings

Overall, the investigators looked at 34 total surveys, and 31 of these met their criteria for inclusion in the research. Individual participant data was drawn from 12 surveys, and this included data on 620,682 subjects in 8 African countries and 4 Asian countries. There ended up being 1,944 individuals diagnosed with tuberculosis.

The research team found that the unadjusted prevalence of subclinical tuberculosis among subjects that had no persistent cough was shown to be 59.1% (95% CI 55.8–62.3). The unadjusted prevalence of the condition, without any cough of any type of time duration, was found to be 39.8% (95% CI 36.6–43.0).

Upon adjustment, the proportions changed to 82.8% (95% CI 78.6–86.6) for no persistent cough and 62.5% (95% CI 56.6–68.7) for no cough at all. This, they determined, indicated a higher prevalence of subclinical TB than previously estimated.

The investigators found that, in a subset of 4 of the surveys, there was a 20.3% proportion of subjects with tuberculosis without any symptoms (95% CI 15.5–25.1) prior to adjustment and 27.7% (95% CI 21.0–36.4) following adjustment. In 1 notable finding, the team reported that tuberculosis without cough, regardless of the type of duration, was found to be more common among females (adjusted OR: 0.79, 95% CI 0.63–0.97 for non-existent persistent cough; adjusted OR 0.76, 95% CI 0.62–0.93 for no cough), suggesting potential gender-related differences in TB presentation.

Among subjects that had tuberculosis, 29.1% (95% CI 25.2–33.3) of those that did not have persistent cough and 23.1% (95% CI 18.8–27.4) of those that had no cough at all were found by the research team to have positive smear examinations. This, the team determined, suggested the value of considering diagnostic methods beyond exclusively symptoms of cough.

“Diagnostic pathways only based on cough as the initiating symptom will delay diagnosis and treatment, potentially resulting in avoidable morbidity, mortality, and M tuberculosis transmission,” they wrote. “Alternative case finding approaches should be expanded urgently.”

References

1. ^{Stuck L, Klinkenberg E, Abdelgadir Ali N, et al. Prevalence of subclinical pulmonary tuberculosis in adults in community settings: an individual participant data meta-analysis. Lancet Infect Dis. Published online March 12, 2024. doi:10.1016/S1473-3099(24)00011-2.}
2. ^{WHO National tuberculosis prevalence surveys 2007–2016. https://www.who.int/publications-detail-redirect/9789240022430. Date accessed: October 16, 2023.}