The burden of subclinical TB in Nigeria

Abstract

SETTING: This study is a retrospective review of a large-scale systematic TB screening project conducted in six states of Nigeria. OBJECTIVE: To determine the magnitude and characteristics of subclinical TB and the relative contributions of bacteriological versus clinical diagnosis in its identification in Nigeria. DESIGN: Data were retrospectively analysed from six states of Nigeria, where parallel screening with any TB symptoms and chest X-ray (CXR) with artificial intelligence (AI) was used for active case finding. Diagnosis of TB among presumptive was confirmed using either bacteriological tests or clinical review of CXR. RESULTS: Out of 8,516 presumptive identified during the project, 172 (2.0%) had no TB symptoms (males: 73.8%, females: 26.2%). The overall prevalence of TB among all presumptive was 21.9% (n = 1,867), including 62 (3.3%) subclinical TB and 1,805 (97.3%) active TB cases. The proportion of clinical diagnosis using CXR was significantly higher in the subclinical TB group than in the active TB group (79.0% vs. 63.5%; P = 0.012, OR = 2.2, 95% CI 1.17–4.03). CONCLUSION: Subclinical TB contributed 3.3% of the large TB burden in this study (22 per 100 presumptive). These cases would have been missed if only symptom-based TB screening had been employed.

Keywords

• latent tuberculosis
• active tuberculosis
• digital x-ray with artificial intelligence
• asymptomatic presumptive