Risk factors for the development of bronchiectasis in patients with asthma

Though asthma and bronchiectasis are two different diseases, their coexistence has been demonstrated in many patients. The aim of the present study is to compare the characteristics of asthmatic patients with and without bronchiectasis and to assess risk factors for the development of this condition. Two hundred and twenty-four moderate-severe asthmatic patients were included. The severity of bronchiectasis was assessed by Reiff and FACED parameters. Logistic regression was used to identify independent factors associated with bronchiectasis. Bronchiectasis was identified in 78 asthma patients. In severe asthma patients, its prevalence was 56.9%. Bronchiectasis was defined as mild in81% of patients using modified Reiff criteria and in 74% using FACED criteria. Asthmatic patients with bronchiectasis had decreasing FEV1, FVC and FEV1/FVC (p = 0.002, 0.005 and 0.014 respectively), presented more frequent asthma exacerbations (p < 0.001) and worse asthma control (ACT 21 vs 16pts, p < 0.001). Factors independently associated with bronchiectasis were older age (42–65 years: OR, 3.99; 95% CI 1.60 to 9.95, P = 0.003; ≥ 65 years: OR, 2.91; 95% CI 1.06 to 8.04, P = 0.039), severe asthma grade (OR, 8.91; 95% CI 3.69 to 21.49; P < 0.001) and frequency of asthma exacerbations (OR, 4.43; 95% CI 1.78 to 11.05; P < 0.001). In patients with severe asthma, age of asthma onset (OR, 1.02; 95% CI 1.01 to 1.04; P = 0.015) and asthma exacerbations (OR, 4.88; 95% CI 1.98 to 12.03; P = 0.001) were independently associated with the development of bronchiectasis. The prevalence of bronchiectasis in severe asthmatic patients is high. Age of asthma onset and exacerbations were independent factors associated with the occurrence of bronchiectasis.

Scientific Reports | (2021) 11:22820 | https://doi.org/10.1038/s41598-021-02332-w www.nature.com/scientificreports/ Asthma diagnosis and severity. The diagnosis of asthma was made according to GINA 16 guidelines and based on clinical symptoms plus one or more complementary tests. All patients had shown reversible airway obstruction on bronchodilator testing, or a positive methacholine challenge test, or variability > 20% in the peak flow recording. Moderate and severe asthma were defined based on the GINA guidelines, which corresponded to patients taking steps 3, 4, and 5 of medication with at least 200 μg of inhaled corticosteroids (budesonide or equivalent doses) plus a long-acting beta-2 agonist 16 .
Pulmonary function test. Spirometry was performed using a MasterLab instrument (MasterLab, Jaeger, Germany), according to European Respiratory Society (ERS) and American Thoracic Society (ATS) guidelines 17 .
The reference values used were those proposed by the ERS 18 .
Bronchiectasis diagnosis and severity. Bronchiectasis was identified in accordance with the Spanish Society of Pulmonology and Thoracic Surgery (SEPAR) recommendations, by comparing the internal bronchial lumen diameter with the adjacent artery calibre in HRCT images 8 . Morphological characteristics of BQ, including bronchodilation, bronchial wall thickening, dilatation type and lobe extension were reviewed and assessed according to the modified Reiff score 19 in six lobes (18 points in total): 1-6 mild, 7-12 moderate, and ≥ 13 severe. The FACED parameter 20 , which incorporates variables such as FEV1, age, pseudomonas colonization, lobe extension and dyspnea, was used for the clinical estimation of patients' status..
HRCT was performed with 1 mm cuts at 10 mm intervals in maximum inspiration. All the imaging features of BQ were interpreted by two pulmonologists working independently. For controversial images, an expert radiologist was consulted and the final decision was made. The degree of the inter-observer agreement was assessed by the Kappa statistic 21 . Atopy and smoking status. Patients were considered atopic if they had at least one positive prick test for any of the common environmental allergens. Non-smokers were patients who had never smoked, and ex-smokers were those who had not smoked for at least six months. The number of pack-years was calculated in all cases.

Definition of exacerbations.
Episodes of asthma exacerbations during the previous year were recorded.
Asthma exacerbations were defined according to GINA guidelines 16 which specify asthma attacks or acute asthma exacerbation as episodes of progressive increases in shortness of breath, cough, wheezing, or chest tightness, or some combination of these symptoms, accompanied by reductions in expiratory airflow.
Statistical analysis. Categorical variables were presented as frequencies and percentages. Statistical differences were analysed using the Chi-square test (or Fisher's exact test when appropriate). Continuous variables were presented as means and standard deviations (SD), or as medians with interquartile ranges (IQR) when data were not normally distributed. The Shapiro-Wilk test was used to analyse the distribution of variables. A logistic regression model was used to determine the factors that were independently associated with the outcome, in this case, the presence of BQ. Variables that presented statistically significant differences (p < 0.05) in the bivariate analysis or were considered to be of clinical interest (such as gender) were included as independent variables in the first step. Variables associated with disease management and asthma control test score were not included. A forward stepwise technique (Wald test; removal threshold, p > 0.10) was used to perform this analysis. ORs and 95% CIs were calculated for independent variables. Tolerance and the Variance Inflation Factor (VIF) were examined to avoid multicollinearity among variables. Statistical significance was defined as a two-tailed p ≤ 0.05. The statistical analyses were performed using SPSS (version 25, Chicago, IL).

Ethics approval and consent to participate. The study was approved by the hospital Vall d'Hebron
Ethics Committee (Reference number: PR(AG)50/2019). All patients provided written informed consent prior to participating. All methods were performed in accordance with the relevant guidelines and regulations.
Informed consent. All patients provided written informed consent to participate in the study.
The odds ratios and 95% confidence intervals of variables related to BQ in all asthma patients and in severe asthma patients are shown in Table 4 and in Figs. 3 and 4 respectively. In the entire group, age, severe asthma grade and asthma exacerbations (≥ 3) were independently associated with the presence of BQ, and in the severe asthma group, age of asthma onset and ≥ 3 exacerbations were independently associated with this outcome.

Discussion
The present study found a high percentage of BQ in asthma patients, especially in those with severe asthma. In severe asthma patients, age of asthma onset and exacerbations were independently associated with the occurrence of BQ.   22 reported the presence of BQ in 28.6% of persistent moderate asthma patients. In patients with persistent severe asthma, the presence of BQ is higher, ranging between 35%-80% according to the study 10,11,[23][24][25][26][27] . In agreement with our results, in a large cohort of asthma patients Oguzulgen et al. 9 recently reported that most asthma patients with BQ had severe persistent asthma (49.0%). In paediatric   www.nature.com/scientificreports/ populations with severe asthma, approximately one-third of children had BQ, which was especially frequent in older children 28 . The differences observed between the studies may have been due to the study design or to the heterogeneity of the populations and comorbidities. Moreover, the widespread use of HRCT has increased the rate of identification of radiological bronchial dilatation and other airway abnormalities. Indeed, data from the US suggest that the number of annual HRCT procedures performed rose from 3 million in 1980 to 81.2 million in 2014 29 .
In the present study, asthmatic patients with bronchiectasis presented more frequent asthma exacerbations and worse asthma control. As we mentioned above, the coexistence of BQ and asthma has been observed in many patients, but few studies have been undertaken to investigate the relationship between the two diseases and the effect of BQ on exacerbations 30 . One such study, by Kang et al. 31 , found that BQ may be a risk factor for asthma exacerbation; these authors suggested that HRCT could be considered to search for concurrent BQ in uncontrolled asthma patients. In fact, BQ may be a terminal status in some patients with chronic airway inflammation, and early detection of the condition is recommended when patients present these features. Crimi et al. 32 also advocate the use of HRCT to rule out BQ in patients with severe asthma and frequent exacerbations. A 5-year follow-up study reported a mortality rate of 20% in adults with BQ 33 . Bronchiectasis challenges asthma control and may aggravate the evolution of the condition.
Although asthma severity is an independent risk factor for BQ, the question of whether it is a cause of BQ remains unclear. In mild to moderate predominant BQ, asthma was the cause in 5.4% of cases 34 . Moreover, a Finnish study suggested that one-fourth of cases with BQ could be attributed to asthma 13 . The British guidelines also recommend considering asthma as the cause of BQ in the absence of any other etiological explanation 35 .
The present study found that, in the overall group, the factors independently associated with the occurrence of BQ were older age, asthma severity and frequent asthma exacerbations, while in patients with severe asthma the independently associated factors were age of asthma onset and frequent asthma exacerbations. In line with the results of the present study, Kang et al. 31 found that annual incidence of asthma exacerbations, and emergency room visits due to asthma exacerbations were higher in patients with both asthma and BQ than in those with asthma alone, although the fact that the study population comprised mostly mild to moderate asthma patients may have influenced the results. In a cohort of moderate to severe asthma patients using the NOPES score (NOPES: nitric oxide, pneumonia, expectoration, and severity), Padilla-Galo et al. found these risk factors to be independently associated with BQ 36 .
We found that patients with asthma and BQ were more likely to present with sinusitis and nasal polyps. The association of sinusitis, nasal polyps and intolerance to aspirin, known as Samter's Triad Syndrome 37 , is a common phenomenon in patients with asthma and a relationship between asthma severity and nasal polyps has been reported 38 , but the pathology of BQ and its interaction with other nasal airway diseases requires further exploration. Nonetheless, in patients with BQ, nasal polyps and chronic sinusitis are quite frequent and are also characterized by early onset 39,40 . In fact, chronic sputum expectoration is common in patients with BQ, and mucus hypersecretion has been associated with rhinosinusitis and nasal polyps 41 . The present study also found that asthma patients with BQ were less atopic. In other research, atopy was significantly more prevalent in patients with asthma than in patients with chronic rhinosinusitis and nasal polyps 37 . Likewise, atopic dermatitis is less frequent in subjects with BQ than in those without (2% vs 11%, OR 0.188) 10 .
Bronchial wall thickening and bronchial dilatation are very common in severe asthma patients [42][43][44] , but few studies provide systematic quantification scores of these abnormalities in asthma patients. The present study describes the characteristics and severity of BQ, mainly finding its grade to be mild according to the criteria of Reiff 19 and FACED 20 . Our results are consistent with the study by Padilla-Galo et al. 36 with regard to bronchiectasis severity, since those authors recorded a mean FACED score of 1.45 pts, which also corresponds to mild grade disease. In contrast, they observed a mild predominant grade of bronchial wall thickening, whereas we found a notable proportion of patients with moderate predominant wall thickening. This difference may be attributed to the different degrees of asthma severity analysed in the two studies. In our study, 82% of patients presented cylindrical dilatation, similar to the rate reported by Padilla-Galo et al. (92.9%).
This study has several limitations. First, is a single-center study carried out at a clinic specialized in severe asthma. Larger studies should now be carried out in different settings to corroborate the data obtained. Secondly, the fact that the HRCT was carried out in some patients during the last three years may have introduced a bias. Third, in HRCT, low-dose volumetric acquisition protocol, with 1 mm cuts at 10 mm intervals in maximum www.nature.com/scientificreports/ inspiration was performed. In this sense, continuous 1 mm slices have a higher sensitivity than conventional HRCT images. Moreover, the bronchoarterial ratio could be influenced by aging. In this sense, the normal bronchoarterial ratio in up to 20% of healthy subjects older than 65 years overlaps with the ratio considered to represent bronchiectasis 8 . In our series, 53 patients (24%) were older than 65 years and in this group this radiological criterion may be present, which could alter the results. Finally, the characteristics of the study design do not allow us to differentiate whether the cause of the reduced FEV1 is asthma or the presence of BQ. However, we observed that patients with asthma and BQ have a lower FEV1 than those without BQ, which could mean that there is a synergistic effect between both conditions.

Conclusions
Our results suggest that, in severe asthma patients, the presence of bronchiectasis is associated with age of asthma onset and the number of exacerbations. Our findings draw attention to the impact of BQ in asthma patients and should encourage clinical professionals to improve early detection measures and interventions for BQ in severe asthma patients, in order to improve patients' quality of life and reduce the economic burden of this disease. Further studies are needed to confirm our results and to focus on the underlying mechanisms, for example via an assessment of the proinflammatory cytokines involved in this complex entity, and thus to be able to offer immunophenotype-based precision treatment.