Abstract
Background: Chest radiographs (CXR) are key to the diagnosis & management of tuberculosis (TB). End-treatment CXR (E-CXR) provide information on therapy response & risk of post-treatment pulmonary sequelae.
Objectives: 1.To report factors associated with baseline CXR (B-CXR) in pulmonary TB (PTB) & extra-pulmonary TB (EPTB).
2.To document the use of E-CXR in subjects with abnormal (abN) baseline CXR (B-CXR).
Study design: Retrospective data for TB patients at our hospital (1.1.06 - 31.12.08) were obtained from the London TB Register. Patients were classified by TB notification as PTB, EPTB or EPTB with pulmonary involvement (EPTB+PTB). B-CXR (2mths pre- to 1mth post- start TB treatment) & E-CXR (2mths peri-end TB treatment) were reviewed blind by 2 respiratory radiologists.
Results: Of 305 patients, 113 (37.0%) were diagnosed with PTB, 155 (50.8%) EPTB & 37 (12.1%) EPTB+PTB. 285 (93.4%) had B-CXR performed in the correct time period (early/late 5.6%, not performed 1.0%) with no significant difference by TB clinical status. Female patients were less likely to have a B-CXR (F: 89.6%, M: 97.4%, p<0.001) which remained significant in multivariable analysis. 149 (52.3%) had E-CXR: more likely in those with abN B-CXR (63.1% abN, 23.8% normal, p<0.001). 116 of 198 (58.6%) with abN B-CXR had an E-CXR with radiology review available. Of these 24 (20.7%) had abN E-CXR including radiological evidence of active TB (n=14, 12.1%).
Conclusion: B-CXR was performed in 99% cases. E-CXR use was sub-optimal in those with baseline abnormalities, especially as one-fifth of those performed remained abnormal and often suggestive of active disease.
- © 2011 ERS
