Is it feasible to radiologically monitor the evolution of non-CF bronchiectasis?

Abstract

To the Editor: We read with great interest the manuscript of Park et al. who evaluated the radiological progression of bronchiectasis among 155 adult patients in Korea.1 The authors reported an increase of bronchiectasis radiological severity, evaluated with the Bhalla score, in 36% of the patients over a 7-year follow-up, with an independent association with low BMI and chronic infection with Pseudomonas aeruginosa. We would like to congratulate the authors for their effort to address such an important topic, recently recognized as one of the research priorities in bronchiectasis by the European Multicentre Bronchiectasis Audit and Research Collaboration.2 The evaluation of radiological severity of bronchiectasis suffers from the absence of a score specifically derived and validated in a non-cystic fibrosis (CF) population, and recent experiences have tested scores, such as the Reiff and the Bhalla scores, that have been mainly used in the CF population.3 The implementation of a radiological severity tool, particularly the Bhalla score, to detect the morphological changes of bronchiectasis over time has been reported for the first time in literature by Park et al. Although the authors showed a statistically significant increase of the Bhalla score from 9.5 to 10.1 points, it could be argued that this might not correspond to a ‘radiologically’ significant difference. Furthermore, incidence of 7.7% patients showing a decrease in Bhalla score during follow-up, which is incorrect if we define bronchiectasis as a permanent dilation of the bronchi, leads us to doubt the reliability of this score as the most accurate tool to evaluate bronchiectasis progression. Baseline scores and follow-up computed tomography (CT) in separate time points might be affected by some bias such as intra-observer variability and lack of sensitivity. An alternative approach to radiological scores would be a side-by-side comparison between baseline and follow-up CT scans, especially in case of mild bronchiectasis changes. Stratifying longitudinal CT changes into categories (e.g. stable, worse, improved bronchiectasis individual features) might be more appropriate for this purpose. Further studies should also overcome other limitations of the paper by Park et al., and evaluate bronchiectasis over time using the same CT protocol in different epidemiological scenarios and in an unselected population of bronchiectasis patients (with a lower prevalence of non-tuberculous mycobacteria infection). Finally, another scientific merit of the study by Park et al. is the identification of predictors of radiological progression in bronchiectasis, including chronic infection with P. aeruginosa. This latest finding supports previous literature which showed P. aeruginosa as a crucial variable characterizing a specific clinical phenotype in bronchiectasis, and leading to worse outcomes.4, 5 Clinical phenotypes might guide radiologists in better interpreting of CT scans of bronchiectasis patients with the final aim to not only consolidate the importance of variables already included in radiological scores but also identify new findings associated with disease severity and progression.