In this single center study, Pseudomonas aeruginosa was the commonest bacteria colonizing the airways of patients with non-CF bronchiectasis. Our study identified that, other than previously reported risk factors, longer duration of bronchiectasis and PPI use were independent risk factors for Pseudomonas aeruginosa colonization, which have important clinical implications. While advanced age is a known poor prognostic factor, those who had longer duration of bronchiectasis had disease onset at younger age. These patients did not only have symptoms for longer duration and worse quality of life, they were also at increased risk of Pseudomonas aeruginosa colonization, which was associated with worse prognosis in bronchiectasis. Patients have Pseudomonas aeruginosa colonization at young age, it would be important to consider Pseudomonas aeruginosa eradication once it is isolated in sputum, as once Pseudomonas aeruginosa colonized the airway, the chance of successful eradication will be low. The European Respiratory Society (ERS) guidelines suggested eradication of new Pseudomonas aeruginosa infection (16). Eradication of Pseudomonas aeruginosa can prevent the consequences from Pseudomonas aeruginosa colonization, which are greater decline in lung function and more frequent exacerbations, especially in younger patients. (16) It would also be essential to monitor and follow up these patients more closely and regularly.
PPI was another risk factor for Pseudomonas aeruginosa colonization identified in this study. In our cohort, there were total 67 patients on long term PPI. 37 (55.2%) were on PPI for primary prophylaxis, such as given with aspirin or non-steroidal inflammatory drugs; 19 (28.4%) were on PPI for gastro-esophageal reflux disease and 11 (16.4%) were on PPI for secondary prophylaxis after an episode of gastric ulcer. Previous studies suggested that PPI usage was associated with bronchiectasis exacerbation requiring hospitalization.(17) The association between PPI and Pseudomonas aeruginosa colonization may be mediated through gastric acid suppression, as previous studies suggested the association between chronic achlorhydria and risks of community acquired pneumonia (18) and acute gastroenteritis. (19) Gastric acid suppression in patients with bronchiectasis may lead to chronic hypochlorhydria or achlorhydria, promoting the growth of colonization in the lower airway, and hence Pseudomonas aeruginosa colonization. Both PPI and Pseudomonas aeruginosa colonization are independent risk factors for bronchiectasis exacerbation, their synergistic effects on bronchiectasis exacerbation should not be under-estimated. As most other risk factors for Pseudomonas aeruginosa colonization and bronchiectasis exacerbation are non-modifiable, avoiding PPI usage should be considered in patients with low risk of gastrointestinal bleeding, and switching to low dose PPI or histamine-2 receptor antagonist should be considered. From our result, more than 80% of the patients on PPI did not have absolute indication for PPI use and might be able to step down their PPI dose or even switch to histamine-2 receptor antagonist which have less propensity in causing achlorhydria. For patients with bronchiectasis, especially who have more severe disease, the indication and dose of PPI should be carefully reviewed and consider weaker gastric acid suppressant if not contraindicated. Shorter course of PPI or PPI use on as needed basis would be another possible strategy. This may have impact on the sputum microbiology in terms of Pseudomonas aeruginosa colonization, as well as bronchiectasis exacerbation risk.
Our study also suggested that the Pseudomonas aeruginosa colonization was associated with more severe disease, with more extensive involvement and increased risk of exacerbation requiring hospitalization. (2, 12–14) Our results were also in line with previous findings in Hong Kong that Pseudomonas aeruginosa colonization was associated with higher sputum volume and worse airflow obstruction with lower FEV1/FVC ratio. (2) The current study also observed that for patients with Pseudomonas aeruginosa colonization, apart from the lower FEV1/FVC ratio to suggest airflow obstruction, these patients also have lower FEV1 and lower FVC in terms of their percentages of predicted values. The lower FEV1 and lower FVC in percentage of predicted value would suggest these patients have worse lung function, which is an important poor prognostic factor as reported by BSI and FACED score. In one report by Martinez-Garcia et al., use of hypertonic saline for sputum induction and use of inhalation antibiotics were reported to be risk factors of Pseudomonas aeruginosa colonization. While the authors in the study did not specify the indications for sputum induction and use of inhalation antibiotics for their patients, it is reasonable to surmise that the use of therapies may reflect that the patients had more severe bronchiectasis requiring such treatments. In our cohort, no patient was prescribed hypertonic saline for sputum induction and few patients were on inhalation antibiotics. This is because long-term home nebulization therapy is difficult for most patients in Hong Kong, partly due to living environment that is considered to be sub-optimal. On the other hand, we have more patients with regular intravenous antibiotics. Patients on regular intravenous antibiotics in our centre were those who had more severe disease with Pseudomonas aeruginosa colonization and repeated exacerbations despite other measures. From our finding and what has been reported in the literature, patients with Pseudomonas aeruginosa colonization deserve more aggressive treatment to preserve lung function and to prevent exacerbation.
There are a few limitations in our study. First, this study involved only a single centre. However, being a tertiary medical centre, the respiratory unit receives referrals from all other health care sources. Patients diagnosed with bronchiectasis were managed in a designated bronchiectasis clinic in our centre. Second, the lung function tests were done at different time for the patients within this cohort. Despite this, the results from our study is consistent with what was previously reported in the literature. Some of the patients did not have lung function test done. As such, leastwise deletion was adopted which can provide less biased estimates and conservative results. (20) In order to understand the clinical course of patients with non-cystic fibrosis bronchiectasis, a well-designed study with long duration of follow up will be worth-while.