Utility of Chest-Xray follow up programs for tuberculosis—A retrospective clinical audit

Abstract

Introduction/Aim: Tuberculosis (TB) is a relatively uncommon infection in Australia, however immigrants born in high-incidence countries, including people from refugee-like backgrounds, health care workers, and close contacts of active TB patients are identified as high risk for latent TB (LTB). Management options for these groups include chemoprophylaxis or CXR surveillance. CXR surveillance is recommended by a number of guidelines for patients who are unwilling or unsuitable for chemoprophylaxis. The period of observation has been variable, reducing from 5 years, to 18 months in the past, to 2 years currently. There is limited evidence as to how useful this approach is or how long surveillance should continue.

Method: A 5-year retrospective review was performed from October 2014 to October 2019 on patients enrolled in the Cairns TB Control Unit CXR surveillance program. Surveillance CXR was performed at 6 and at 18–24 months. Data was collected on patient demographics, adherence, and outcomes.

Results: Two hundred eight patients were identified. The commonest indications were contact tracing 123(59.1%), refugee health screening 46(23.1%), health care worker screening 19(9.1%) and TB health undertaking 13(6.3%). Fifty-six (29%) patients were born in Australia, including 19(9.1%) Indigenous, 34(16.3%) in Bhutan, 21(10.1%) in Myanmar and 15(7.21%) in Philippines.

The majority of patients 178(85.6%) had a positive Mantoux with an average reading of 15 mm, 3(1.4%) had positive IGRA and 27(12.9%) had a history of previously treated TB. One hundred eighty-three (83%) of patients had a normal CXR at initial screening, 22(10.85%) had changes consistent with old TB. One hundred forty-eight (71.2%) patients completed the screening program.

There were no new CXR findings related to TB at the 6 or 24-month surveillance period. No patients in the study population developed active TB.

Conclusion: CXR surveillance in patients with LTB is labour intensive and not without cost, and seems of low utility in identifying the development of active TB.