Current cough and sputum assessed by the cough and sputum assessment-questionnaire (CASA-Q) is associated with quality of life impairment in cystic fibrosis

The patients were prospectively recruited from the Department of Respiratory Diseases - Reims University Hospital, France between November 2016 and December 2019 and included in the RINNOPARI cross-sectional study (Recherche et INNOvation en PAthologie Respiratoire Inflammatoire), an observational cohort of inflammatory chronic lung diseases (NCT02924818, first posted on 5th October 2016). Each patient signed a written informed consent form.

Consecutive patients followed up for CF in the Department of Respiratory Diseases at Reims University Hospital were considered for inclusion if they were at least 18 years old. Exclusion criteria were previous or planned lung transplantation and patients requiring an urgent visit or any ongoing or recent medical condition in the last 4 weeks, including pulmonary exacerbations. We recorded and registered demography, clinical characteristics, pulmonary function test results, exercise tolerance assessed by the 6-minute walking test results (6MWT), sputum microbiological data and thoracic computerized tomography (CT) scan features on an electronic case report form.

For microbiological analysis, sputum samples at inclusion were liquefied by N-acetylcysteine and 1/1000 dilutions were cultured in Columbia blood and chocolate agar (Thermo Fisher Scientific) at 37 °C for 48 h. Colonies were identified by MALDI-TOF mass spectrometry (MALDI Biotyper®, Bruker Daltonics). Chronic infection by Pseudomonas aeruginosa was defined according to Leeds criteria [11]. A similar definition was used to record chronic infections by Staphylococcus aureus, Stenotrophomonas maltophilia, Burkolderia cepacia and Achromobacter xylosoxidans.

Thoracic CT scans performed within 6 months either before or after the inclusion in the study were analyzed by two independent investigators (SD, GD) with a final consensus interpretation. All CT scans were performed with the patient in the supine position at end-inspiratory volume using multidetector CT scanners. One- to 5- mm-thick slices at 5- to 10-mm intervals were analyzed from the lung apices to the lung bases. The diagnosis of bronchiectasis was defined according to the Fleischner Society as dilated bronchial lumen relative to the adjacent pulmonary artery, absence of bronchial tapering, and visualization of bronchi within 1 cm of the pleural surface [12]. The extent of bronchiectasis (0–3), the thickness of the bronchial wall (0–3) and small airways abnormalities (0–1), were quantified for each lobe (lingula was considered as a separated lobe) and a total score (sum of the score of each lobe) was calculated according to Ooi et al. score. The total score ranked from 0 to 42, a higher score suggesting a higher impairment [13].

The CASA-Q is a self-administered questionnaire assessing cough and sputum symptoms, and their impact on daily activities over the 7 previous days [6]. The score contains four domains: cough symptoms (3 items), cough impact (8 items), sputum symptoms (3 items), and sputum impact (6 items). Each item is answered on a scale ranging from “never” to “always” for frequency and from “not at all” to “a lot/extremely” for intensity. For each domain, items were summed and rescaled to obtain a score ranging from 0 to 100, with a higher score reflecting milder respiratory impairment. As previously described, the data were analysed for each domain separately and no total score (sum of the scores of the four CASA-Q domains) was calculated [6, 14].

The health-related quality of life of CF patients was assessed using two questionnaires: (1) the CFQ-R, a CF-specific questionnaire that includes 50 items and provides insight on respiratory and digestive functions and overall welfare during the last 2 weeks. Each score is standardized from 0 to 100, a higher score reflecting a better quality of life [15]; (2) the SGRQ which measures the impact of lung disease on quality of life and well-being and consists of 50 questions assessing symptoms, and their impact on activity and quality of life over the last 4 weeks. The total score ranges from 0 to 100, a higher score indicating a worse quality of life [5].

The primary endpoint was to describe the CASA-Q results in CF adult patients. Secondary endpoints were to evaluate the associations between the CASA-Q and clinical, pulmonary function, exercise tolerance, microbiological and CT-scan features, and also health-related quality of life.

The Cronbach’s alpha value for CASA-Q and each quality-of-life scores (CFQ-R, SGRQ) was calculated.

A cluster analysis was performed to characterise the clinical phenotype associated with highly impaired CASA-Q. We used the two-step clustering function provided by IBM-SPSS (version 27). This classification method automatically identified subgroups of patients using the scores from the 4 domains of the CASA-Q: cough symptoms, cough impact, sputum symptoms, sputum impact. At the first step, the log-likelihood distance was used to assign participants to the cluster leading to the largest log-likelihood. At the second step, the Bayesian Information Criterion (BIC) was used to assess multiple cluster solutions and automatically determine the optimum number of clusters [16].

Demographic, clinical, and functional characteristics are detailed in Table 1. 69% of patients were men. The mean age was 27.8 years with a mean body mass index of 21.8 kg/m². Mean FEV₁% was 64% of the predicted value. Only two patients (4%) were active smokers. No patient was treated with elexacaftor/tezecaftor/ivacaftor.

Chronic infections by S. aureus, P. aeruginosa, A. xylosoxidans and S. maltophilia were present in 74% (n = 37), 40% (n = 20), 4% (n = 2), and 2% (n = 1) of the patients, respectively. Microbiological analysis of sputum at inclusion, available for 42 patients, identified S. aureus, P. aeruginosa and A. xylosoxidans in 76% (n = 32), 40% (n = 17), and 2% (n = 1) respectively. Thoracic CT scans were available for 14 patients. The mean CT-scan score was 25.7 ± 10.6.

The Cronbach’s alpha value for CASA-Q was 0.966. The mean values for the four domains of the CASA-Q were 58 ± 23 [8–100] for cough symptoms, 77 ± 24 [16–100] for cough impact, 62 ± 25 [8.3–100] for sputum symptoms, and 84 ± 21 [8.3–100] for sputum impact.

The associations between CASA-Q domains, and clinical, functional, microbiological, and CT-scan data are detailed in Table 2. Cough symptoms and impact scores from CASA-Q were significantly more impaired in females (47 ± 22 vs. 63 ± 22, p = 0.029; 65 ± 29 vs. 83 ± 19, p = 0.014; respectively). Analysis of CFTR mutations identified that patients with ΔF508 heterozygous mutation were more impaired on the sputum impact score than homozygous patients (76 ± 27 vs. 93 ± 8, p = 0.008). Each score of the four domains of the CASA-Q was significantly more impaired in patients with a dyspnea score ≥ 2 on mMRC scale (cough symptoms: 39 ± 22 vs. 63 ± 20, p = 0.002; cough impact: 54 ± 27 vs. 84 ± 17, p < 0.001; sputum symptoms: 42 ± 26 vs. 68 ± 21, p = 0.002 and sputum impact: 69 ± 29 vs. 90 ± 12, p = 0.001). Exacerbation in the last year was significantly associated with impairment of cough and sputum symptoms scores (53 ± 23 vs. 68 ± 20, p = 0.03; 56 ± 23 vs. 76 ± 23 respectively, p = 0.01). Distance on the 6-minute walking test was associated with a better score on CASA-Q cough (p = 0.027, r² = 0.461) and sputum impact (p = 0.029, r² = 0.456). To summarize, significantly impaired CASA-Q in CF was associated with female gender, ΔF508 heterozygous mutation, significant dyspnea on exercise, exercise capacity and exacerbation history, but not with lung function and bronchiectasis extend on CT-scan.

To further characterise the clinical phenotype associated with highly impaired CASA-Q, we undertook a 2-step clustering of patients depending on scores in the 4 domains of the CASA-Q. This analysis resulted in 2 clusters: a “low score” cluster (n = 9, 18.8%) and a “high score” cluster (n = 39, 81.2%) (Fig. 1). The low score cluster (reflecting the most severely impaired subjects regarding CASA-Q) was characterised by more frequent female patients, more frequent ΔF508 heterozygous patients and dyspnea mMRC score ≥ 2.

The Cronbach’s alpha value for CFQ-R and SGRQ was 0.602 and 0.911 respectively. Detailed results of quality of life scales are shown in Table S1. On the CFQ-R, “vitality” was the most impaired domain (57 ± 24), and “eating disturbance” was the less impaired domain (89 ± 18). The mean SGRQ score was 25 ± 17 with a predominant impairment in the “symptoms” domain (45 ± 21).

The correlation between the CASA-Q and the quality of life scales CFQ-R and SGRQ are detailed in Table 3. The total score and all domains of the SGRQ correlated with each domain of the CASA-Q (p < 0.001) suggesting a close association between the 2 scores (Fig. 2). Except for “role perception”, “treatment burden” and “weight”, domains of CFQ-R significantly correlated with all or many domains of CASA-Q score.

The cluster analysis revealed that the patients in the low score cluster had a global impairment in all domains of the SGRQ and CFQ-R (Fig. 1).