Poor Physical Capacity in Bronchiectasis Patients Is Correlated with Poor Quality of Life

To study the risk factors associated with quality of life (QoL) in a cohort of Finnish non-cystic brosis bronchiectasis (BE) patients. We aimed to evaluate which of the clinical characteristics were risk factors for poor quality of life, how patients with frequent exacerbations differed from those with only few exacerbations and if QoL symptom domains were correlated with dyspnoea or severity of BE. Methods A cross-sectional study and part of the EMBARC study including questionnaire data and medical record data. Study participants were recruited between August 2016 and March 2018 from three different pulmonary clinics in Helsinki University Central Hospital (HUH) catchment area, Finland. The study included 95 adult patients with (mean age was 69 (SD± 13) years). A Finnish translation of the disease-specic quality of life-bronchiectasis (QoL-B) questionnaire was applied, and scores in the lowest quarter (25%) of the scale were considered to indicate poor QoL. The bronchiectasis severity index (BSI) and FACED (including FEV1, age, pulmonary bacterial colonization, affected lobes and dyspnoea) score were used. The severity of dyspnoea was examined using the modied Medical Research Council (mMRC) dyspnoea scale. to compare means and mean ranks, respectively. Pearson correlations were calculated to study associations between mMRC score and BSI with different symptom domains. Proportions were compared by χ²-test and Fisher's exact -test. Scores in the lowest quarter (25%) of the scale were considered to indicate poor QoL. Risk factors for poor quality of life in BE were analysed by logistic regression analysis, with age, BMI, FEV1 (%) and FVC (%) considered as continuous variables. Statistical analyses were performed with Statistical Package for (SPSS) (IBM Armonk NY,


Strengths And Limitations Of The Study
Strength of this study is the multifaceted perspective to symptoms by comparing QoL questionnaire with mMRC, BSI and FACED score results.
Further, there were no missing data for QoL questions or for mMRC results and thus the results are valid for this study population.
Limitations of this study are a cross-sectional setting of the analysis and the relatively small number of study participants which is re ective of prevalence of BE but also the relatively small number of populations in Finland and in Helsinki and Uusimaa Hospital area.

Background
Irreversible destruction and enlargement of bronchi is a typical feature of bronchiectasis (BE). Chronic intrabronchial neutrophilic in ammation and potential bacterial colonisation cause chronic daily airway symptoms such as excessive sputum production and recurrent airway infections. The impact of morbidity burden and diminished quality of life (QoL) from this vicious cycle is obviously unfavorable. BE is a long-term condition with rarely curative treatment, for which the aim is to prevent exacerbations and further deterioration of BE and maintain or improve QoL. According to data, not only severe infections but also chronic in ammation can lead to changes in the architecture of the airways [1].
Characterisation of the disease has been challenging in the past due to a lack of validated tools. For this reason, tools to optimise individual treatments and improve clinical care are urgently needed, in addition to better tools to assess the effect of treatment and disease progression. If limited disease is not associated with a chronic condition such as immunode ciency or connective tissue disease, it may be cured with surgery. However, BE is often di cult to cure completely so current treatment options aim to limit disease progression and to prevent exacerbations [2]. It is important to assess the subgroups that would gain the most bene t from a speci c intervention.
Commonly used measures for different lung diseases have serious limitations in clinical trials [3]. For example, spirometry provides a good assessment of lung function but does not assess the speci c health status or severity of the disease [4]. Similarly, radiographic examinations, such as high resolution computed tomography (HRCT) scans, have been used to describe the severity of BE, but the ndings do not correlate well with the clinical features of the disease [5]. Tests to estimate dyspnoea and physical capability (e.g., 6-minutes walking test) are useful and standardised in routine praxis. However, a study by Lee et al. [4] demonstrated that this test was not appropriate to describe the severity of BE.
The rst of the questionnaires developed for respiratory patients, especially those with BE, was the St George's Respiratory Questionnaire (SGRQ), which was followed by a number of disease-speci c questionnaires for chronic obstructive pulmonary disease (COPD) and pulmonary brosis. Wilson et al. [6] validated the SGRQ for BE in 1997 and showed a good correlation of the SGRQ score with disease severity. Another patient-reported questionnaire for BE, the Leicester Cough Questionnaire, inadequately covers the respiratory symptoms [7].
Modern clinical assessment tools for BE have only been used in the past few years. Chalmers et al. [8] validated the bronchiectasis severity index (BSI) in 2014, and Martinez-Garcia et al. [9] established the FACED score in the same year. These tools are exceptional for the assessment of symptomatic or frequently exacerbated patients with severe disease, enabling focused and intensive treatment.
Nowadays, both the BSI and FACED have signi cant roles in assessing the severity of BE.
Quittner et al. [10] validated the disease-speci c quality of life questionnaire-bronchiectasis (QoL-B) version 3.0 in 2014, and con rmed its applicability for use in clinical trials and routine clinical practice.
They assumed the relevance for evaluating BE as a whole.
In the current study, we studied the risk factors for poor QoL in a Finnish BE cohort using clinical assessment tools.

Participants
Altogether, 95 of 205 non-cystic brosis BE patients were included in this cross-sectional study, which focused on assessing quality of life of BE patients. Patients were recruited between August 2016 and March 2018 from the Helsinki University Hospital (HUH) district. BE was diagnosed by high-resolution computed tomography (HRCT) and was assessed by a radiologist. Only those over 18 years of age and with non-cystic brosis BE were recruited. Details of this cohort have been reported previously [11].

Quality of life in bronchiectasis questionnaire
The QoL-B (version 3.0) is a multidimensional disease-speci c questionnaire for patients with BE.
Patient's self-report their current condition over 37 questions, which are categorised into eight different domains (respiratory symptoms, physical activity, role, emotional and social functioning, vitality, health perceptions and treatment burden). The score for each domain ranges from 0 to 100, with higher scores representing fewer symptoms or better functioning. The QoL-B has been reported to give reliable and consistent results if re-tested and has high convergent validity [10] [12]. The QoL-B has been translated into Finnish, which was the version applied in the current study.

Modi ed Medical Research Council scale
In patients with respiratory diseases, the modi ed Medical Research Council (mMRC) dyspnoea scale is a reliable tool to assess functional disability due to dyspnoea. Dyspnoea only on strenuous exercise on level ground is graded 0 points, shortness of breath when hurrying on level ground or walking up a slight hill is graded 1 point, walking slower than people of the same age on level ground due to breathlessness or the need to stop to catch breath when walking at their own pace is graded 2 points, stopping for breath after walking 100 m or after a few minutes on level ground is graded 3 points, and breathless when dressing is graded 4 points. Higher grades on the mMRC scale indicate a higher severity of BE [13] [14].

Bronchiectasis Severity Index
The Bronchiectasis Severity Index (BSI) is a composite disease-speci c prognostic index that was developed to aid clinical decision-making related to BE. Speci cally, the BSI was developed to predict mortality, severe exacerbations, frequency of exacerbations and QoL [8]. The BSI combines the clinical, radiological and microbiological features of BE, and includes the following parameters: body mass index (BMI), forced expiratory volume in one second (FEV1; % of predicted), previous hospital admissions, number of exacerbations in the previous year, mMRC breathlessness score, pseudomonas colonisation or colonisation with other microorganisms, and radiological severity. The grading system of the BSI is mild (0-4 points), moderate (5-8 points) and severe (9 or more points) [8]. Extensive disease was determined as four or more lobes affected by bronchiectasis radiological ndings. The lingula was considered to be an independent lobe.

FACED score
The FACED score is another disease-speci c prognostic index that aims to assess the probability of allcause mortality over 5 or 15 years of follow-up [8][9] [15]. FACED assesses different factors related to BE including lung function (FEV1), age, pulmonary bacterial colonisation, number of lobes affected by BE, and dyspnoea. FACED scoring is graded as mild (0-2 points), moderate (3-4 points) or severe (5-7 points).

Patient and Public Involvement:
This study is the part of EMBARC. The data were collected into the EMBARC database anonymously. Patients or the public were not involved in the design, conduct, or reporting, of our research.
Dissemination of our research will be offered through the Finnish Organisation for Respiratory Health and through the Finnish Allergy, Skin and Asthma Federation.

Statistical analyses
Independent samples t-test and Mann-Whitney U test were used to compare means and mean ranks, respectively. Pearson correlations were calculated to study associations between mMRC score and BSI with different symptom domains. Proportions were compared by χ²-test and Fisher's exact -test. Scores in the lowest quarter (25%) of the scale were considered to indicate poor QoL. Risk factors for poor quality of life in BE were analysed by logistic regression analysis, with age, BMI, FEV1 (%) and FVC (%) considered as continuous variables. Statistical analyses were performed with Statistical Package for Social Sciences (SPSS) program, version 22 (IBM corporation, Armonk NY, USA).

Results
There were altogether 23 patients with poor QoL, 69 patients with more preserved disease speci c QoL and missing information for three BE patients for QoL (Table 1). The overall QoL results (median values) for different symptom domains were 60 (SD ±31.9) for physical functioning, 60 (±27.9) for role, 50 (±20.5) for vitality, 75 (±22.2) for emotional functioning, 58 (±24.4) for social functioning, 77.8 (±24.8) for treatment burden, 41.7 (±21.1) for health perceptions, 57 (±19.7) for respiratory symptoms and 450 (±151.8) for all domains combined (where 0 was the minimum and 800 the maximum). A score in the lowest quarter of the scale, less than 25%, represented a poor QoL.
In the total BE cohort, dyspnoea assessed by mMRC had a mean value of 1.72 (±1.1) and median value of 2 (±1.1), with 18.9% of patients having a mMRC score of 3 and 6.3% having a score of 4.
Overall, the BE cohort had a mean BSI value of 7.7 (±4.2) and median value of 7 (±2.0). Of the BE patients, 22% had mild disease, 44% had moderate disease and 34.1% had severe disease.
When BE severity was estimated according to the FACED index, 42.1% of patients had mild BE, 46.3% had moderate BE and 11.6% had severe BE. The mean score was 2.6 (±1.5) and the median was 3.0 (±1.5).
The BE patients with a poor QoL had a trend for lower lung function (FEV1 84.3 ± 31.4%) than those with preserved QoL (FEV1 88.9 ± 21.4%, p ns) (Table 1). Further, patients with poor QoL had more severe dyspnoea (mMRC mean scores of 2.4 vs. 1.5) and more severe BE disease (BSI of 10.5 vs. 6.9), a greater number of exacerbations (four vs. two over the past 12 months), extensive disease, and more frequent bacterial colonisation and cystic changes.
The mMRC dyspnoea score had a moderate negative correlation with physical domain (r = -5.72, p < 0.01) in the QoL questionnaire (Table 2). Similarly, BSI had a moderate negative correlation with physical domain (r = -5.96, p < 0.01) in the QoL questionnaire (Table 2).
When analysing the risk factors for poor QoL, exacerbations, extensive disease, frequent yearly exacerbations, higher score of BSI and mMRC were found to be important. When the analysis was adjusted for age and gender, exacerbations (OR 1.7), higher score of BSI (OR 1.3), higher score of mMRC (OR 3.3), extensive disease (OR 3.7) and frequent yearly exacerbations (OR 4.9) increased the risk for a poor QoL (Table 4).

Discussion
Exacerbations of BE, extensive radiological disease and poor physical tolerance signi cantly increase the risk of a poor QoL. In particular, recurrent yearly exacerbations were found to increase the risk of poor QoL by ve-fold. In this study, both effort tolerance (mMRC) and disease severity (BSI) were negatively correlated with physical domain which is formed of the questions assessing physical capability associated with BE in the quality of life questionnaire. However, in clinical practice, the mMRC dyspnoea score is simpler and easier to use than the BSI [16] [17]. The BSI is a complex index that requires lung function parameters, radiological features, exacerbation, information on hospitalisation and microbiological ndings to calculate the score [8]. On the other hand, the mMRC is a four-point grading system for self-reported dyspnoea in everyday life [17]. Although BSI has been shown to accurately predict hospitalisations, exacerbations and mortality of non-cystic brosis BE patients, it is more challenging to use and is therefore better suited to scienti c purposes than clinical practice [8] [18]. The mMRC and QoL scores have been used to demonstrate the effect of new treatment options for BE [18].
Health (median score of 41.7), vitality (median score of 50) and respiration (median score of 57) were the domains with the poorest self-reported results in the QoL questionnaire. It has previously been demonstrated that these scores are stable if measured during a stable phase of BE, and that they can be used to differentiate between mild, moderate and severe BE patients [9]. However, in a 2-year follow-up study of 19 BE patients by Magge et al. [19], physical functioning, role functioning and health perceptions were shown to improve with treatment at a specialised care center. These results are similar to ours for health perceptions (mean score 41.4) and vitality (mean score 45.6), but not for respiration (mean score 62.5).
Our nding of signi cantly reduced vitality score and mMRC score correlating with the physical domain is in accordance with previous studies that show reduced physical activity of BE patients compared to gender-and age-matched individuals without BE [20]. In our study, the mMRC scale offered practical information on the physical functioning of patients.
The identi cation of patients at high risk for exacerbations may be valuable to guide clinical decision making with regard to factors such as the frequency and intensity of follow-up and the use of long-term antibiotic therapy. The scoring systems used to assess the severity of BE gave different results for severe disease, as 34.1% of patients were estimated to have severe BE with BSI scoring while only 11.6% of patients were classi ed as having severe BE with FACED grading. Similar results were described by Visser et al. 2019 in their Australian cohort, as 58% of the cohort had severe disease according to the BSI, whereas 17% had FACED-de ned severe disease. Our results are also in good agreement with the observations of a large multidimensional severity assessment conducted by McDonnell et al. 2016, who reported that the BSI is superior to FACED in predicting overall clinically important disease-related outcomes. The BSI includes parameters for previous hospital admissions and the number of exacerbations in the previous year, whereas FACED includes a parameter for age [8][9] [15]. Although, they have been shown to have similar capacities for predicting long-term mortality, in our cross-sectional study more patients were classi ed as having severe BE by BSI scoring than by FACED scoring [15]. This highlights the importance of patients' previous history of exacerbations.
When phenotyping different BE patients, BE caused by connective tissue disease has been associated with a poor prognosis and rapid disease progression [21]. Immunode ciency, COPD and allergic bronchopulmonary aspergillosis (ABPA) have been associated with recurrent exacerbations [21]. In the current study, we found poor QoL regardless of background aetiology in BE patients with exacerbations, and especially in those with frequent exacerbations. Furthermore, colonisation of the bronchi with proteobacteria Pseudomonas aeruginosa or Haemophilus in uenza is associated with chronic neutrophilic in ammation and exacerbations [21].
Anti-IL5 and anti-IL5R therapies have been suggested as an add-on therapy for BE [22]. A study that included BE patients with severe asthma, ABPA, idiopathic BE, combined asthma/COPD, EGPA and PCD reported an increase in FEV1 and QoL, as well as a reduction in exacerbations and mMRC scores [22].

Terpstra et al. 2019 reported signi cantly worse QoL in BE patients with COPD than in other BE patients.
In their study population of 200 BE patients (aged 69.5 years), the mean number of exacerbations was one per year, with zero hospitalisations and a mean mMRC score of 2.68 (SD 1.09). The mean mMRC score in our population was 2.0, but there was a median of two exacerbations per year. Previously, BSI was proposed as a tool for scoring the severity of BE and FACED as a tool for predicting mortality [23] [24].
The limitations of the current study include the relatively small size of the cohort and the cross-sectional setting of the QoL analysis. The small sample size is due to the limited number of subjects recruited in the original study. A strength of the study is that there were no missing data for QoL questions and mMRC scale results.
An earlier study found that previous severe exacerbations (OR 2.6) or an exacerbation in the past 12 months (OR from 1.38 to 3.90) is predictive of future exacerbations [8]. We found that exacerbations (OR 1.7), frequent yearly exacerbations (OR 4.9), poor physical tolerance (OR 3.3) or extensive disease (OR 3.7) increased the risk of poor QoL (Table 4) in the analysis adjusted for age and gender.
Similarly, diminished lung function (analysed as FEV1) has been reported to be a risk factor for future exacerbations [8]. Although those patients with poor QoL had reduced lung function (FEV1 84%) compared to the others (FEV1 89%), we did not identify lung function as a risk factor for future exacerbations [8]. This may be due to the relatively small study population, together with the relatively well-preserved lung function in our study population.
When estimating the effect of exacerbations and frequent exacerbations on poor QoL, it should be noted that in the study by Brill et al. 2015, the median duration of an exacerbation was 16 days, and 16% of patients had at least one exacerbation with a duration of more than 35 days. Thus, the association of poor QoL with exacerbations is crucial. Despite best practice guidelines, exacerbation will occur at some point, so the secondary treatment goals are to prevent deterioration of lung function and BE radiologic ndings and to maintain a good QoL. This should be considered when assessing which BE patients bene t from follow-up at a specialised care center [19]. The current guidelines focus on recommending special care for those with three or more yearly exacerbations and for those on long-term antimicrobial therapy [25][2].

Conclusions
We found that exacerbations, especially frequent exacerbations, increased the risk of poor QoL in patients with BE. Neither comorbidities nor the aetiology of BE appeared to affect the QoL. In addition, more extensive disease or greater severity of BE increased the risk of poor QoL. Furthermore, the mMRC score for dyspnoea and BSI for severity of BE showed the strongest (negative) correlations with physical functioning in BE. When looking for strategies to improve QoL in BE, the focus should be on preventing exacerbations and supporting physical tness to reduce exercise-induced dyspnoea and support patients' ability to cope in daily life activities and also when at an advanced disease status.

ACKNOWLEDGMENTS
The authors would like to acknowledge all the patients who participated in the study. Competing Interest: In relation to this study, the authors declare that there are no con icts of interest.
Contributorship statement: PK was responsible for the design, analysis, interpretation, drafting and nalizing of the manuscript. WM was involved in the analysis, interpretation, drafting and nalizing of the manuscript. JM was responsible for the data collection, analysis, interpretation, drafting and nalizing of the manuscript. TT was involved in the data collection, drafting and nalizing of the manuscript. PB provided statistical assistance and was involved in nalizing of the manuscript.
Data sharing statement: This is part of the EMBARC study and data can be obtained from EMBARC register or from the authors by an approved application.
Disclaimer: The funders of the study had no role in the study design, data collection, data analysis, data interpretation or writing of the report.