Pulmonary function tests (PFTs) comprise a set of assessments for studying lung function. They may be used in individuals presenting with respiratory symptoms (such as dyspnea or cough) or radiological abnormalities in need of clarification, preoperative evaluation setting or suspected occupational respiratory disease, or for monitoring the disease course and treatment response in patients with known respiratory disease1,2.

Correct interpretation of PFT requires that the American Thoracic Society (ATS) and European Respiratory Society (ERS) technical standards be met, which demands not only adequate equipment, but also minimally acceptable patient cooperation for protocol required manoeuvres3. Thus, obtaining acceptable PFT may prove a challenge, especially in children, older patients, or patients in more severe stages of disease1. On the technician’s side, explaining the procedure, providing examples and verbal encouragement are essential2. However, the type and stage of the respiratory disease in question may potentially influence the difficulty in conducting these tests, which can also lead to patient anxiety and discomfort.

Exertional dyspnea is a common symptom in chronic obstructive airway disease (COPD) and interstitial lung disease (ILD). This is a complex symptom, with several associated mechanisms and closely related to exercise tolerance and quality of life. In both nosological groups, depression and anxiety are common4-8, and may be related with the level of dyspnea/tolerance for exertion, which, in turn, may condition physical and emotional participation in basic activities of daily living6.

In this study, the authors aimed to evaluate and compare dyspnea and anxiety before and after PFT, the level of heart and respiratory rate, and the delay in performing these tests in a group of patients with either fibrotic ILD or obstructive airway disease (COPD and bronchial asthma).


The authors conducted a prospective analytical cross-sectional study, between May 2021 and September 2022, of sequentially enrolled patients who underwent PFT in a lung function laboratory at a level two hospital (hierarchy according to responsibilities and specialties present in the hospital), who had either fibrotic ILD or some form of obstructive airway disease, and agreed to participate in the study.

After obtaining informed consent, a survey (Supplementary file) was applied on the day of the PFT previously requested by the attending pulmonologist. The survey, specially created for this purpose, also evaluated the number of attempts needed to obtain an acceptable output-volume curve, as well as the heart rate and respiratory rate before and immediately after the end of the tests.

Assessment of dyspnea

To assess dyspnea four questionnaires/scales were used:

  • Modified Medical Research Council Dyspnea Scale (mMRC), validated for the Portuguese language, applied before PFT. Originally validated for assessing disability in COPD9, it is currently one of the most used tools for grading exertional dyspnea in chronic respiratory patients.

  • Modified Borg scale, applied before and after the PFT. This scale is widely used to assess perceived exertion10.

  • Visual analogue dyspnea scale, applied before and after PFT.

  • University of California, San Diego Shortness of Breath Questionnaire (UCSDSOBQ), consisting of 24 items for classification (0–5) of the degree of dyspnea according to the limitation imposed on the performance of various daily tasks and activities, completed before PFT11.

Assessment of anxiety

To assess anxiety two scales were used:

  • Hospital Anxiety and Depression Scale (HADS), consisting of 7 related specific questions (scored from 0–3)12, completed immediately before undergoing PFT.

  • Generalized Anxiety Disorder Assessment (GAD-7), comprising 7 items (score: 0–3) portraying various daily situations and reaction frequency13, completed immediately before undergoing PFT.

Statistical analysis

Descriptive analysis of the studied variables was performed by obtaining frequencies for categorical variables, and means/medians and standard deviations/interquartile ranges for quantitative variables. Comparison was made between the obstructive airway disease group and the fibrotic ILD group, and between some predefined subgroups within those: asthma vs COPD, fibrotic ILD with ‘usual interstitial pneumonia’ (UIP) pattern (by histology or identified on high-resolution chest CT) vs ‘non-UIP’ pattern. Comparison of variables between groups was performed and tested using Fisher’s exact test/Pearson’s chi-squared test, Wilcoxon’s test/Welch’s two sample t-test and Kruskal-Wallis’s rank sum test.

We used the total time that the patient remained in the lung function laboratory to complete the examination as a result (dependent variable) of the variables under study and applied multiple linear regression and regression with automatic selection (stepwise). Some correlated variables from the set of independent variables were excluded from the analysis. A stepwise linear regression was used to identify predictors/associations of ‘total time’ with the selected independent variables. At each step, variables were added based on p-values, and the AIC was used to set a limit on the total number of variables included in the final model.

Statistical analysis was performed using R version 4.2.2. The level of significance was set at 0.05. The hospital’s ethics committee approved this study (Ref 44-05-2022).


A total of 80 patients were evaluated, 40 of them with obstructive airway disease and another 40 with some form of fibrotic ILD. In the first group, 45% of the patients had a diagnosis of COPD (with an overall mean FEV1 value of 70.18 ± 21.9% of the predicted value, with half of the patients in GOLD stages 3 or 4), while the rest had a diagnosis of bronchial asthma (77.3% of them in GINA treatment 1–3 and 22.7% in steps 4–5). Regarding the patients included in the fibrotic ILD group, the distribution of individual diagnoses was as follows: 35.0% with fibrotic hypersensitivity pneumonitis, 27.5% with idiopathic pulmonary fibrosis (IPF), 12.5% with idiopathic or secondary fibrotic non-specific interstitial pneumonia (NSIP), 10.0% with UIP secondary to rheumatoid arthritis, 7.5% of unclassifiable fibrotic interstitial pneumonia , and 5.0% of patients with chronic silicosis complicated by progressive massive fibrosis. In this group, mean FVC and DLCO values were 77.1 ± 22.1% and 53.6 ± 19.8% predicted, respectively.

The following variables were evaluated: age, gender, diagnosis, lung function, number of attempts, number of acceptable curves, heart and respiratory rates before and after PFT, mMRC, UCSDSOBQ, Borg scale (dyspnea), visual dyspnea scale, GAD7, HADS – dimension of ‘anxiety’, time spent (minutes) in the room and number of technically acceptable flow-volume curves between the two groups (Table 1).

Table 1

Description of the characteristics studied in the two groups*

CharacteristicsnOverall (N=80)Obstructive airway disease (N=40)ILD (N=40)pa
Female39 (49%)22 (55%)17 (42%)
Male41 (51%)18 (45%)23 (57%)
Age (years)8063 (13)57 (11)69 (11)<0.001
FEV1/FVC8072 (12)66 (11)78 (9)<0.001
FEV18074 (24)70 (22)78 (25)0.12
FVC8079 (21)82 (20)77 (22)0.3
DLCO-SB6860 (21)67 (20)54 (20)0.007
KCO6877 (20)79 (21)76 (18)0.6
Spirometry8042 (53%)18 (45%)24 (62%)0.14
Plethysmography8038 (48%)22 (56%)16 (40%)0.14
BD test prescribed8036 (45%)24 (60%)12 (30%)0.007
DLCO prescribed8069 (86%)29 (72%)40 (100%)<0.001
Number of attempts783.96 (1.62)3.63 (1.05)4.28 (1.97)0.3
Number of acceptable curves780.004
28 (10%)0 (0)8 (20%)
367 (86%)37 (97%)30 (75%)
43 (3.8%)1 (2.6%)2 (5.0%)
HR before PFT8080 (13)81 (12)80 (14)0.9
HR after PFT7783 (13)83 (12)84 (14)0.8
RR before PFT8020.3 (4.5)19.0 (3.0)21.6 (5.4)0.010
RR after PFT7722.2 (5.8)20.2 (3.5)24.3 (6.9)0.002
mMRC801.18 (0.99)0.50 (0.68)1.85 (0.77)<0.001
UCSDSOBQ8022 (23)21 (22)24 (23)0.5
Borg before PFT801.16 (1.48)1.10 (1.57)1.23 (1.39)0.4
Borg after PFT802.62 (2.10)2.36 (1.90)2.88 (2.28)0.4
Visual scale before PFT801.38 (1.60)1.28 (1.57)1.48 (1.65)0.6
Visual scale after PFT803.05 (2.28)2.78 (2.15)3.33 (2.40)0.3
HADS805.0 (3.5)5.6 (3.7)4.5 (3.2)0.2
GAD7806.5 (5.3)7.0 (5.7)5.9 (4.8)0.4
Lab time7825 (8)25 (6)26 (10)0.4

* Data are given either as n (%) or mean (SD).

a Pearson’s chi-squared test. Welch’s two sample t-test. Wilcoxon’s rank sum test. FEV1/FVC: forced expiratory volume/forced vital capacity. DLCO: diffusing capacity of the lungs for carbon monoxide. KCO: carbon monoxide transfer coefficient BD: bronchodilation. HR: heart rate. RR: respiratory rate. mMRC: modified medical research council dyspnea scale. UCSDSOBQ: University of California, San Diego shortness of breath questionnaire. HADS: hospital anxiety and depression scale. GAD7: generalized anxiety disorder assessment.

The group with fibrotic ILD had a lower median age (p<0.001), a lower DLCO (p<0.05) and a higher FEV1/FVC value (p<0.001). Post PFT heart rate, respiratory rate, and severity of dyspnea by visual analogue scale, also tended to be higher in this group.

Regarding dyspnea, the baseline mMRC level was significantly higher in the fibrotic ILD patients (p<0.001). The UCSDSOBQ and Borg scale values also tended to be higher in this group of patients compared to patients with obstructive airway disease. No statistically significant differences were found between subgroups: asthma versus COPD and UIP versus non-UIP (Table 2).

Table 2

Description of variables by subgroups according to diagnosis*

CharacteristicsnOverall (N=80)Non UIP (N=22)UIP (N=18)COPD (N=19)Asthma (N=21)pa
Female39 (49%)12 (55%)5 (28%)7 (37%)15 (71%)
Male41 (51%)10 (45%)13 (72%)12 (63%)6 (29%)
Age (years)8063 (13)65 (12)73 (9)60 (8)54 (12)<0.001
FEV1/FVC8072 (12)75 (9)83 (6)58 (8)73 (8)<0.001
FEV18074 (24)71 (27)88 (18)59 (21)80 (18)<0.001
FVC8079 (21)74 (25)81 (17)75 (18)88 (21)0.10
DLCO-SB6860 (21)57 (19)50 (21)59 (19)75 (19)0.004
KCO6877 (20)78 (15)73 (21)69 (21)89 (17)0.033
Spirometry8042 (53%)11 (50%)13 (76%)6 (32%)12 (57%)0.058
Plethysmography8038 (48%)11 (50%)5 (28%)13 (72%)9 (43%)0.059
BD test prescribed8036 (45%)10 (45%)2 (11%)11 (58%)13 (62%)0.007
DLCO prescribed8069 (86%)22 (100%)18 (100%)15 (79%)14 (67%)<0.001
Number of attempts783.96 (1.62)3.73 (1.08)4.94 (2.58)3.72 (0.89)3.55 (1.19)0.3
Number of acceptable curves782.94 (0.37)2.95 (0.38)2.72 (0.57)3.00 (0.00)3.05 (0.22)0.032
HR before PFT8080 (13)81 (16)79 (11)79 (13)82 (12)>0.9
HR after PFT7783 (13)85 (15)82 (11)82 (14)84 (11)0.9
RR before PFT8020.3 (4.5)23.1 (6.2)19.8 (3.8)20.0 (2.7)18.1 (3.0)0.002
RR after PFT7722.2 (5.8)26.4 (7.8)21.8 (4.7)21.3 (3.5)19.2 (3.3)<0.001
mMRC801.18 (0.99)1.82 (0.73)1.89 (0.83)0.74 (0.73)0.29 (0.56)<0.001
UCSDSOBQ8022 (23)27 (25)19 (21)26 (25)16 (19)0.3
Borg before PFT801.16 (1.48)1.50 (1.22)0.89 (1.53)0.87 (1.14)1.31 (1.89)0.2
Borg after PFT802.62 (2.10)3.36 (2.52)2.28 (1.84)2.32 (1.48)2.40 (2.25)0.5
Visual Scale before PFT801.38 (1.60)1.86 (1.67)1.00 (1.53)1.00 (1.11)1.52 (1.89)0.2
Visual Scale after PFT803.05 (2.28)3.73 (2.60)2.83 (2.09)2.68 (1.42)2.86 (2.69)0.5
HADS805.0 (3.5)5.4 (3.9)3.4 (1.5)5.9 (3.7)5.3 (3.7)0.2
GAD7806.5 (5.3)6.2 (5.6)5.6 (3.6)7.3 (4.8)6.9 (6.5)0.7
Lab time7825 (8)28 (10)24 (9)25 (6)24 (6)0.3

* Data are either given as n (%) or median (interquartile range).

a Pearson’s chi-squared test. Kruskal-Wallis’s rank sum test. Fisher’s exact test.

In the anxiety assessment, there were no statistically significant differences between groups. However, the analysis by subgroups showed a trend towards a higher level of anxiety in the subgroup of patients diagnosed with COPD, compared to the other groups of patients (Table 2).

Total time in the respiratory function laboratory was significantly impacted by the performance of a bronchodilation test (p<0.001), the number of attempts needed to achieve a technically adequate flow-volume curve (p<0.001), the visual analogue dyspnea scale value before the test (p<0.001), UCSDSOBQ score (p<0.05), heart and respiratory rates before the test (p<0.001), and male gender (p<0.05). A diagnosis of fibrotic ILD was also a determining factor (p<0.05) (Table 3).

Table 3

Study of determining variables for time in the respiratory function laboratory

VariablesMultiple linear regressionStepwise linear regression
Beta95% CIpBeta95% CIp
Obstructive disease----
Fibrotic interstitial disease4.1-0.65–8.80.0894.40.94–7.90.014
Male-2.9-6.2–0.490.092-3.2-6.0 – -0.390.026
FEV1/FVC-0.14-0.32–0.040.117-0.17-0.31 – -0.030.020
BD test prescribed9.76.6–13<0.0019.46.6–12<0.001
DLCO prescribed-8.9-22–4.50.189-7.6-18–2.80.148
Number of attempts2.01.1–3.0<0.0012.01.2–2.9<0.001
Number of acceptable curves-3.3-8.2–1.50.173-3.6-7.9–0.830.110
HR before PFT-0.10-0.21–0.000.058-0.11-0.21– -0.010.034
RR before PFT0.45-0.05–0.950.0770.550.19–0.910.004
RR after PFT0.05-0.36–0.460.790
UCSDSOBQ-0.14-0.26 – -0.020.024-0.12-0.20 – -0.040.003
Visual Scale before PFT2.51.1–3.8<0.0012.51.5–3.6<0.001
Visual Scale after PFT0.25-0.60–1.10.553


Differences were found in some objective measures of cardio-respiratory stress, such as HR and RR, immediately after the end of PFT in the group of patients with fibrotic ILD.

Additionally, this group of patients reported a higher level of dyspnea when assessed by the mMRC scale, and tended to have a higher value on both the Borg scale and on the visual analogue dyspnea scale, which aligns with the fact that dyspnea is a common and intrusive symptom in patients with IPF, fibrotic hypersensitivity pneumonitis and ILD secondary to connective tissue disease14-16.

Regarding anxiety, a prevalence of up to 31% has been reported in patients with chronic ILD17 and up to 34% in patients with asthma18. In patients with COPD, there have been reports of prevalence reaching as high as 55%19. In this study, the COPD subgroup also tended to report higher levels of anxiety.

In patients with fibrotic ILD or obstructive airway disease, both dyspnea and anxiety may profoundly impact the natural course of disease, not only by reducing quality of life measures but also by increasing social isolation, depression, lack of adherence to therapy and risk of hospitalizations and exacerbations19.

Patients with fibrotic ILD spent a longer time in the laboratory to complete PFT (p<0.05) and tended to have a lower number of acceptable curves during their performance.

Although PFTs are an invaluable tool for longitudinal monetarization of these patients, in advanced stages of disease the progressive limitation of their ability to collaborate with the performance of serial PFT may lead to greater levels of frustration and anxiety while also impairing the tests’ reliability and reproducibility.

A better awareness by the clinical staff of specific difficulties in performing PFT imposed by certain diseases and the possibility of improving strategies and providing adequate pulmonary laboratory time, may help to reduce anxiety and discomfort in these patients.


Obstructive airway diseases and ILD are eminently different in their pathophysiology, clinical repertoire, and natural history. We sought to assess differences imposed by the type of disease on the difficulty in performing PFT and the level of dyspnea and anxiety hence generated.

Despite limitations related to sample size, there was a greater cardio-respiratory stress in patients with fibrotic ILD when compared with patients with obstructive airway disease, objectively assessed by HR and RR immediately after finishing PFT. Similarly, some indicators suggest a higher level of dyspnea after PFT in the former group. A longer time needed to perform PFT as well as a lower number of acceptable curves were also observed in the ILD subgroup. As for anxiety, the subgroup of patients with COPD tended to present with higher levels. The awareness for these differences can help to anticipate hazards and allow differentiated approaches to these patients.