This study was approved by the Ethics Review Board of the Clinical Hospital (47710715.9.0000.0068), and all participants provided written informed consent. The subjects with asthma were evaluated on two nonconsecutive days, at least 48 hours apart. On day 1, anthropometric indices, clinical asthma control (Asthma Control Questionnaire, ACQ), and lung function (spirometry) were assessed. After that, the subjects were randomized to perform either a cardiopulmonary exercise test (CPET) or two incremental step tests. On day 2, the subjects performed the other test, the CPET or IST, according to randomization.

All incremental tests were performed using a metabolic gas analyzer. Subjects were instructed to use a bronchodilator (400 μg of salbutamol) 15 minutes before each test to obtain a better performance on the cardiopulmonary exercise.12

Subjects with asthma aged between 18 and 60 years with a body mass index (BMI) between ≥20 and ≤ 40 kg/m2 were recruited from a university hospital during a routine medical consultation. They were diagnosed with moderate or severe persistent asthma according to the Global Initiative for Asthma13 and clinically stable for at least six months (i.e., no hospitalizations, emergency care, or changes in medication in the last 30 days for at least 30 days). All patients received short- and long-acting bronchodilators and inhaled corticosteroids, and none received immunobiological monoclonal therapy. The exclusion criteria were the presence of cardiovascular, musculoskeletal, or other chronic pulmonary diseases; uncontrolled hypertension or diabetes; active cancer; use of oral corticosteroids; smokers or ex-smokers (≥10 packs/year or stopped smoking for a period equal to or longer than 12 months); and pregnancy or breastfeeding.

The ACQ-714 validated for Brazilian Portuguese15 was used. The questionnaire has seven questions for daytime and nocturnal asthma symptoms, activity limitations, dyspnea, wheezing, use of a rescue bronchodilator (short-acting β2-agonist) in the past week and the forced expiratory volume in the 1st second (FEV1, in % of predicted, prebronchodilator).16 The responses are given on a 7-point scale, and the overall score is the mean of the responses (0=totally controlled, 6=severely uncontrolled). Values greater than or equal to 1.5 indicate uncontrolled asthma; values between 0.75 and 1.5 indicate partially controlled asthma, and values less than 0.75 indicate fully controlled asthma.14

Pulmonary function testing was performed according to the current American Thoracic Society (ATS)/European Respiratory Society (ERS) guidelines.17 Forced vital capacity (FVC), FEV1, and the FEV1/FVC ratio absolute values were obtained and expressed as percentages according to the reference values for the Brazilian population.18

The CPET was performed using an electric treadmill ergometer (Jaeger™) linked to a digital exercise evaluation system with a gas analyzer (Vyntus CPX™). The peak oxygen uptake (peak VO2), minute ventilation (VE), carbon dioxide production (VCO2), respiratory exchange ratio (RER), heart rate (HR), modified Borg score for dyspnea (BD), and leg fatigue (BF) at rest and at the end of the exercise test were analyzed. The criteria for exercise interruption before symptom limitation were in accordance with the American Thoracic Society/American College of Chest Physicians guidelines,19 The following conditions were also used as interruptions to the exercise test: loss of leg coordination, mental confusion, and dizziness or fainting, as previously established.19

The ramp protocol was used with a fixed speed and a 2% increase in the slope every minute. Before the test, the subject was asked to choose one of the following speeds: 2.4, 3.6, 4.8, 6.0, or 7.2 km/h; therefore, the protocol was individualized.20-21 Before the exercise test, the subjects were exposed to all speeds for periods between 1 and 3 minutes before choosing the velocity. The criteria for the CPET test interruption included the following: chest pain suggestive of ischemia; complex ectopy; second- or third-degree heart block; a fall in the systolic pressure of 20 mmHg from the highest value; arterial hypertension (250 mmHg systolic; 120 mmHg diastolic); severe desaturation (SpO2 of 80% accompanied by symptoms or signs of severe hypoxemia).

The subjects with asthma stepped up and down on a 20 cm high wooden bench (width 40 cm, depth 60 cm) as previously described.8 An audio signal dictating the stepping rate played on a compact disc. The initial stepping rate was 10 steps/min, with a one-step increment every 30 s up to the tolerance limit. The researcher interrupted the test when the subject was unable to keep the pace for 15 s.22 The tests were performed using a system with a gas analyzer (Vyntus CPX™), and the variables recorded were the same as those for the CPET. The results from IST-1 and IST-2 were used to perform the reliability analyses. Two ISTs were performed on the same day with a rest period of at least 30 minutes between them. The second test was performed when the participant's vital signs had returned to baseline levels to ensure the same clinical conditions for the patient in both tests. The IST with better performance regarding the peak VO2 value was called the best IST (b-IST). The values from the b-IST were used for validity and interpretability analyses.

Statistical analysis was performed using Sigma Stat statistical software, version 2.03 (SPSS Inc., USA). The normality of the data was assessed using the Kolmogorov–Smirnov test. Categorical variables were expressed as an absolute number, percentage, and frequency. A paired t-test was used to compare the variables of the CPET versus the b-IST and IST–1 versus IST–2. The Wilcoxon test was used for nonparametric data. All p values less than 0.05 were considered to be indicative of statistical significance.

The property reliability involves three domains: reliability, error measurement, and internal consistency. However, as this study is about a field test, internal consistency has not been evaluated. The reliability of the IST was analyzed by test-retest scores (IST-1 and IST-2) using intraclass correlation coefficient (ICC) model 2,1 (absolute subtype agreement for single measurements) with 95% confidence intervals (CI95%). The classification adopted was <0.40= low, from 0.40 to 0.75= moderate; from 0.76 to 0.90= substantial and > 0.90= excellent.23

The measurement error is associated with the absolute error of the measurement.24 The test and retest scores for each test were evaluated by the standard error of measurement (SEM) and the minimal detectable change with 90% confidence (MDC90). SEM was analyzed using the SEM=Standard Deviation√1-ICC. The SEM was considered very good if <5% of the total score, good if ≥5% and <10%, doubtful if ≥10% and <20%, and negative if >20%.25 The MDC90 was calculated considering the number of steps in the IST–1 and IST–2 as follows: score ranged in the test, subtracted from the score in the retest, and divided by √2×SEM.26 The limits of agreement and precision of peak VO2 between the b-IST and CPET and between the IST-1 and IST-2 were calculated using the Bland and Altman method.27

The validity was analyzed by the correlation of peak VO2 (l/min) obtained from both the CPET and IST–b. The hypothesis is that the correlation between peak VO2 assessed via the CPET and IST-b was strong and positive. Pearson's correlation test was used. The classification adopted was r<0.30 indicating a weak correlation; 0.30≥ and <0.60 as moderate; and r >0.60 as strong.28

The floor and ceiling effects were analyzed and considered to be present if 15% or more of the individuals reached the minimum or maximum score in the evaluation, respectively.23

Reliability analysis demonstrated an ICC of 0.88 (95% CI 0.80–0.93) for the total number of steps and 0.93 (95% CI 0.88–0.96) for peak VO2 (l/min), showing substantial and excellent reliability. The means of the peak VO2 from IST-1 (1673 ± 0.42 l/min) and IST-2 (1754 ± 0.43 l/min) were similar (p > 0.05) (Table 2). The total number of steps and peak VO2 showed a very good measurement error between test-retest, SEM% 12.4 Total Number of Steps (TNS) (3%) and MDC90 42 TNS (2.5%) (Table 3).

The Bland–Altman plot also showed an average difference in the peak VO2 of ± 93 mL/min between both IST tests (Fig. 1). In addition, a strong linear correlation was observed in the peak VO2 between the IST-1 and IST-2, considering the subjects altogether (r = 0.9; p < 0.0001) (Table 3).

Fig. 1.

Bland–Altman plot for VO2 at the peak of IST-1 and IST-2. The continuous line corresponds to the average difference between the lower and upper limits of agreement. IST-1, first modified incremental step test; IST-2, second modified incremental step test.

(0.07MB).

A ceiling effect was observed in 10% of the subjects, and an appropriate floor effect was observed since none of the subjects presented (0%) (Table 3).

Differences were observed between the CPET and the b-IST when peak VO2 was evaluated in ml/min. In addition, a between-group difference was observed according to the disease control comparing controlled (n = 24) versus uncontrolled asthma (n = 26) (Table 2). However, no difference between those groups was observed [CPET: controlled 1,925 ± 458 vs. uncontrolled 2,198 ± 505 mL/min (p = 0.06); b-IST: controlled 1688 ± 425 vs. uncontrolled 1846 ± 437 mL/min (p = 0.20)]. On the other hand, when the subjects' weight was considered (peak VO2 in mL/kg), no difference was verified intra- or intertest when comparing subjects with controlled versus uncontrolled asthma (p = 0.09).

A lower cardiopulmonary response was observed on the b-IST compared to the CPET, except for perceived exertion in terms of dyspnea and leg fatigue (Table 2). In addition, a strong linear correlation was observed in the peak VO2 between the b-IST (Table 2) and the CEPT (r = 0.85; p < 0.001), and homoscedasticity was observed between the two tests.