Clinical characteristics and validation of bronchiectasis severity score systems for post-tuberculosis bronchiectasis.

ABSTRACT

INTRODUCTION: Lung damage related to tuberculosis is a major contributor to the etiology of bronchiectasis in China. It is unknown whether bronchiectasis severity score systems are applicable in these cases. OBJECTIVES: To evaluate the clinical characteristics and validation of bronchiectasis severity score systems for post-tuberculosis bronchiectasis. METHODS: The study enrolled 596 bronchiectasis patients in Shanghai Pulmonary Hospital between January 2011 and December 2012. The data for calculating FACED and bronchiectasis severity index (BSI) scores along with mortality, readmission, and exacerbation outcomes were collected and analyzed within a follow-up period with a median length of 48 months (interquartile range 43-54 months). RESULTS: The study enrolled 101 post-tuberculosis bronchiectasis patients and 495 non-tuberculosis bronchiectasis patients. Compared with non-post-tuberculosis bronchiectasis, post-tuberculosis bronchiectasis patients experienced less bilateral bronchiectasis (P = .004), a higher frequency of right upper lobe involvement (P < .001) and showed the cylindrical type more often (P < .001). Follow-up data indicated that both scoring systems were able to predict 48(43-54) month mortality in post-tuberculosis patients as assessed by the area under the receiver operator characteristic curve (AUC) (FACED AUC = 0.81, BSI AUC = 0.70), but they did not predict readmission (FACED and BSI = 0.56) or exacerbation (FACED and BSI = 0.52) well. CONCLUSIONS: There are apparent differences on radiologic features between bronchiectasis patients with and without history of pulmonary tuberculosis. Both FACED and BSI can predict mortality in post-tuberculosis bronchiectasis.