TB Research

Automated chest radiography and mass systematic screening for tuberculosis

Falak Madhani, Rabia Maniar, Aziza Burfat, Marwa Ahmed, Saira Farooq, Aqsa Sabir, Abdul Khalique Domki, Liesl Page‐Shipp, et al. (12 authors)

The International Journal of Tuberculosis and Lung Disease · 2020-07

Abstract

BACKGROUND: Systematic screening for TB using automated chest radiography (ACR) with computer-aided detection software (CAD4TB) has been implemented at scale in Karachi, Pakistan. Despite evidence supporting the use of ACR as a pre-screen prior to Xpert ® MTB/RIF diagnostic testing in presumptive TB patients, there has been no data published on its use in mass screening in real-world settings. METHOD: Screening was undertaken using mobile digital X-ray vehicles at hospital facilities and community camps. Chest X-rays were offered to individuals aged ≥15 years, regardless of symptoms. Those with a CAD4TB score of ≥70 were offered Xpert testing. The association between Xpert positivity and CAD4TB scores was examined using data collected between 1 January and 30 June 2018 using a custom-built data collection tool. RESULTS: Of the 127 062 individuals screened, 97.2% had a valid CAD4TB score; 11 184 (9.1%) individuals had a CAD4TB score ≥70. Prevalence of Xpert positivity rose from 0.7% in the <50 category to 23.5% in the >90 category. The strong linear association between CAD4TB score and Xpert positivity was found in both community and hospital settings. CONCLUSION: The strong association between CAD4TB scores and Xpert positivity provide evidence that an ACR-based pre-screening performs well when implemented at scale in a high-burden setting.

MeSH terms

  • Medicine
  • Tuberculosis