Six-minute Stepper Test to Set Pulmonary Rehabilitation Intensity in Patients with COPD – A Retrospective Study

ABSTRACT

Pulmonary rehabilitation (PR) improves outcomes in patients with chronic obstructive pulmonary disease (COPD). Optimal assessment includes cardiopulmonary exercise testing (CPET), but consultations are limited. Field tests could be used to individualize PR instead of CPET. The six-minute stepper test (6MST) is easy to set up and its sensitivity and reproducibility have previously been reported in patients with COPD. The aim of this study was to develop a prediction equation to set intensity in patients attending PR, based on the 6MST. The following relationships were analyzed: mean heart rate (HR) during the first (HR_1–3) and last (HR_4–6) 3 minutes of the 6MST and HR at the ventilatory threshold (HRvt) from CPET; step count at the end of the 6MST and workload at the Ventilatory threshold (VT) (Wvt); and forced expiratory volume in 1 second and step count during the 6MST. This retrospective study included patients with COPD referred for PR who underwent CPET, pulmonary function evaluations and the 6MST. Twenty-four patients were included. Prediction equations were HRvt = 0.7887 × HR_1–3 + 20.83 and HRvt = 0.6180 × HR_4–6 + 30.77. There was a strong correlation between HR_1–3 and HR_4–6 and HRvt (r = 0.69, p < 0.001 and r = 0.57, p < 0.01 respectively). A significant correlation was also found between step count and LogWvt (r = 0.63, p < 0.01). The prediction equation was LogWvt = 0.001722 × step count + 1.248. The 6MST could be used to individualize aerobic training in patients with COPD. Further prospective studies are needed to confirm these results.

KEYWORDS:

• Cardiopulmonary exercise testing
• COPD
• field test
• pulmonary rehabilitation
• six-minute stepper test

Introduction

Chronic obstructive pulmonary disease (COPD) is a leading cause of disability and mortality worldwide (1,2). The most significant symptoms are dyspnea and exercise limitation, leading to physical inactivity and muscle wasting (3). Most treatments address these symptoms and pulmonary rehabilitation (PR) has been shown to be effective in improving dyspnea, exercise capacity, and quality of life (4). Moreover, PR helps to reduce the number of exacerbations, length of hospital stay after an exacerbation, and mortality following an exacerbation (5).

Despite these benefits, only a small number of patients with COPD undergo PR (6). The main factor limiting access to these programs is a lack of PR centers. Thus, PR could be carried out of hospital or pulmonary centers. Another factor limiting access to these programs is a lack of assessment centers (7). Optimal assessment should include cardiopulmonary exercise testing (CPET) to determine both the optimal training settings as well as any cardiopulmonary contraindications to PR. However, this is not available in most centers and when it is, consultations are limited. Therefore, PR is often delayed for several weeks and patients can lose motivation.

In order to promote PR, CPET could be replaced by field tests to assess exercise capacity and individualize PR in patients with COPD. The 6-minute walk test (6MWT) is currently the gold standard to assess exercise capacity and can accurately predict a target heart rate (HR) for training in patients with COPD (8,9). However, the test has some disadvantages such as environmental constraint (requires a 30m-long corridor to be performed) which can limit it use in some centers and in ambulatory settings.

The 6-minute stepper test (6MST) is a new field tool. It consists of performing the largest number of steps on a stepper for 6 minutes. Its sensitivity and reproducibility have previously been reported in patients with COPD (10–13). It is easy to set up in the clinical setting and could be used to individualize PR in patients with COPD.

The aim of this study was to develop a prediction equation to set rehabilitation intensity for patients attending PR, with the use of a simple, readily available field test.

Methods

Study design and patient selection

This retrospective study included patients with COPD referred for PR from September 2015 to August 2016 to the ADIR Association, Bois-Guillaume, France.

Inclusion criteria

Patients were included if they had a clinical diagnosis of COPD (ratio between forced expiratory volume in one second [FEV1] and forced vital capacity [FCV] <0.70). The severity of airflow limitation was assessed according to the GOLD classification (14). They had to be at least 18 years old and have undergone both CPET and pulmonary testing, and a 6MST during their first session of PR. They had to be below or equal to 90 kg (maximum weight supported by the stepper).

Non-inclusion criteria

Patients were not included if they had any heart rate modulating treatment (i.e. beta blockers, pacemakers), orthopedic disorders limiting the achievement of the 6MST, or had experienced a pulmonary exacerbation between the CPET and 6MST.

Exclusion criteria

Patients were excluded from the analysis if data from the 6MST were uninterpretable due to technical problems or if CPET had not determined the ventilatory threshold (i.e. CPET with oxygen or submaximal exercise), or suspected cardiac disorders on electrocardiogram during CPET.

Data extraction

Data regarding age, gender, height, weight, body mass index (BMI), pulmonary function, exercise capacity (CPET and 6MST), cardiovascular comorbidities, use of domiciliary noninvasive ventilation and long-term oxygen were extracted though a retrospective chart review.

Assessment

6MST

Patients performed two 6MSTs (Athlitec, GoSport, Sassenage, France) separated by a rest period of at least 20 minutes. The second test began when the HR and transcutaneous oxygen saturation (SpO₂), values returned to baseline values. Performance on the second test was often better because (1) the hydraulic jacks used in this study were more flexible once they had warmed-up and (2) there is a probable “learning” effect. (10–12) Standardization of the instructions for the 6MST was based on the ATS guidelines for the 6MWT (15), as previously described by Borel et al. (10):

“The object of this test is to make the highest number of strokes you can during six minutes duration. Six minutes is a long time, so you will be exerting yourself. You will probably get out of breath or become exhausted. You are permitted to slow down, to stop, and to rest as necessary. You may lean against the wall while resting, but you have to resume exercise as soon as you are able. The correct movement is the one: you have to stretch the bent leg until the step has touched the stepper base. Then do the same movement with the other leg.”

The test was performed in an isolated room in order to avoid noise or external stimuli which can affect performance. The stepper was placed near a door and the patient was allowed to put a hand on it if out of balance or exhausted. The height of the step was fixed to 20 cm. (10) A step was defined as the rise and lowering of one foot. The patient was informed of the time each minute. No other encouragement was given. HR and SpO₂ were continuously recorded by pulse oximetry (Oxymontre NONIN 3150, Nonin Medical Inc., Plymouth, MN) and then extracted by Nvision software (Henrotch, Aartselaar, Belgium).

CPET

CPET was performed on an electromagnetic ergometer (Ergoselect 200, Ergoline, Bitz, Germany) with an incremental protocol. Following a 3 minutes warm-up period, incremental ramp exercise (5–20 W/min) was maintained up to until exhaustion. A face mask (Hans Rudolph, Inc., Kansas city, MO, USA), pneumotach, and gaz analyzer (Ergocard, Medisoft, Louvain, Belgium) were used to measure gases (oxygen comsuption (VO₂) and carbon dioxide production (VCO₂)) breath by breath. Ventilatory threshold was manually determined as the average of 4 methods: (1) first break, (2) raise in the Minute ventilation (VE)/VO₂ ratio without modification of the VE/VCO₂ ratio (Wasserman's method), (3) raise in the raise in the expired carbonic gaz (PetCO₂) and (4) Beaver's method (16,17). HR was continuously monitored with a 12-lead electrocardiogram (Ergocard, Medisoft, Louvain Belgium).

Pulmonary function

Pulmonary function tests were carried out according American Thoracic Society (ATS) and European Respiratory Society (ERS) guidelines with plethysmography (Masterscreen, Jaeger, Wittsburg, Germany). Values were expressed as percentages of established theoretical values for European populations.

Outcome

Primary outcome was the development of the regression équations to predict the HR at the ventilatory threshold (HRvt) from the mean HR during the first (HR1-3) and last (HR4-6) 3 minutes of the 6MST. HR during these periods was averaged from the considered 3 minutes with a sampling frequency of 1 Hz and Nvision software.

Secondary outcomes were (1) the correlation between mean HR_1–3 and HR_4–6 from the 6MST, and the HRvt from CPET, (2) the relationship between step count at the end of the 6MST and the workload at the ventilatory threshold (Wvt), and (3) the relationship between forced expiratory volume in 1 second (FEV1) and the number of steps carried out during the 6MST.

Statistical analysis

Continuous data were expressed as means (SD) or medians (25th–75th percentile) as appropriate. Normality of the distributions was assessed using the Shapiro–Wilk test. The relationship between HR during the different tests as well as step count and Wvt were determined using Pearson or Spearman correlation tests according to the normality of the data distribution. Single linear regressions were performed when appropriate. Since the data for Wvt were not normally distributed, they were normalized using a log-transformation. A predictive equation using linear regression was then derived. Comparison between mean HR_1–3 and HR_4–6 was carried out using a paired t-test. Values from the second 6MST were used for the analysis. A p-value <0.05 was considered statistically significant. Prism 5 software was used for all analyses.

Results

Patients

Among the 132 patients referred for PR during the study period (i.e. all types of respiratory diseases included), 37 met the inclusion criteria. 4 were excluded from the analysis due to technical problems during the recording of HR and SpO₂ during the 6MST. A further 9 patients were excluded from the analysis because the ventilatory threshold was not determined (i.e. patients exercising with oxygen or submaximal CPET). Thus data from 24 patients were analyzed.

Patient characteristics are presented in Table 1. Briefly, 25% were female, mean age was 61.5 (SD 8.7) years, mean BMI was 23.6 (SD 4.2) kg/m², median FEV1 was 45 (range 31.8–55) %, mean step count during the 6MST was 186.8 (SD 59.4) steps, and median power at ventilatory threshold was 37.5 (range 30–47.5) W. Seven patients (29.2%) used long-term oxygen and 2 patients (8.5%) used domiciliary non-invasive ventilation.

Table 1. Patient characteristics (n = 24).

Variable (units)
Female (n)^a	6 (25)
Age (years)^b	61.5 (8.7)
Height (cm)^b	167.2 (7.4)
Body mass (kg)^b	65.9 (12.2)
BMI (kg/m²)^b	23.6 (4.2)
FEV1 (L)^b	1.23 (0.4)
FEV1 (%)^c	45 (31.8–55)
FVC (L)^b	2.73 (0.7)
FVC (%)^b	79.8 (17.1)
FEV1/FVC (% ratio)^b	45.4 (12.7)
COPD stage
GOLD I^a	1 (4)
GOLD II^a	10 (42)
GOLD III^a	9 (38)
GOLD IV^a	4 (17)
6MST (steps)^b	186.8 (59.4)
HR1–3^b	107.1 (14,3)
HR4–6^b	120.6 (15)
Power-vt (W)	37.5 (30–47.5)
Long-term oxygen (n)^a	7 (29)
Non-invasive ventilation (n)^a	2 (8)
Peripheral arterial disease (n)^a	3 (13)
Hypertension	7 (29)
Hypercholesterolemia (n)^a	3 (13)
Diabetes (n)^a	1 (4)
Myocardial infarction (n)^a	2 (8)

^a Values expressed as numbers (%).

^b Values expressed as means (SD).

^c Values expressed as medians (25th–75th percentile).

FEV1/FVC is expressed as percentage ratio.

BMI: body mass index; FEV1: forced expiratory volume in one second; FCV: forced vital capacity; 6MST: 6-minute stepper test; HR_1–3: mean HR during the first 3 minutes of the 6MST; HR_4–6: mean HR during the last 3 minutes of the 6MST; Power-vt: Power at the ventilatory threshold.

HR_1–3 (101.1 (SD 14.3) bpm) was significantly lower than HR_4–6 (114.3 (SD 15) bpm, p < 0.0001).

Regression equations for HR from 6MST and HR from CPET

HR from CPET could be predicted, respectively for HR_1–3 (Figure 1A) and HR_4–6 (Figure 1B) by the following equations:

‐	HRvt = 0.7887 × HR1–3 + 20.83;
‐	HRvt = 0.6180 × HR4–6 + 30.77.

Figure 1. Relationship between mean HR during the first (A) and last (B) 3 minutes of the 6MST, and HR at the ventilatory threshold measured by CPET. HR: heart rate; HRvt: heart rate at the ventilatory threshold; HR_1–3: mean HR during the first 3 minutes of the 6MST; HR_4–6: mean HR during the last 3 minutes of the 6MST; CPET: cardiopulmonary exercise testing; 6MST: 6-minute stepper test.

Correlation between HR from 6MST and HR from CPET

There was a strong positive correlation between mean HR_1–3 and HRvt from CPET (r = 0.69, p < 0.001).

There was also a strong positive correlation between mean HR_4–6 and the HRvt from CPET (r = 0.57, p < 0.01).

Regression and correlation between step count and Wvt

There was a strong positive correlation between step count and Wvt (r = 0.59, p < 0.01). Since data were non-normally distributed, Wvt was normalized using a log-transformation to derive a predictive equation using linear regression. There was a strong positive correlation between the step count and LogWvt (r = 0.63, p < 0.01). LogWvt from the CPET could be predicted by the following equation:

‐	LogWvt = 0.001722 × step count + 1.248.

Correlation between step count and FEV1

There was no significant relationship between step count and FEV1 expressed either as liters (r = −0.1, p = 0.657) or as a percentage of the theoretical norm (r = −0.05, p = 0.8).

Discussion

The results of this study showed that there were strong relationships between the means of both HR_1–3 and HR_4–6 from the 6MST and HRvt (r = 0.65 and r = 0.53 respectively). Thus, target heart rate for training in patients with COPD may be predicted by the following equations:

‐	HRvt = 0.7887 × HR1–3 + 20.83;
‐	HRvt = 0.6180 × HR4–6 + 30.77.

The mean HR from the last 3 minutes was chosen for analysis based on a previous report by Bonnet et al. (9). They used the mean HR from the last 3 minutes of the 6MWT (also called “straight HR”) to predict HRvt with the following equation:

‐	HRvt = (0.75 × HRstraight)—(0.03 × distance)—(0.32 × age) + 64.4.

Moreover, Borel et al. (10) showed that there was no difference between the mean HR of the last 3 minutes of each test (respectively 118.2 (SD 18.7) bpm for the 6MST and 120.8 (SD 12.6) bpm for the 6MWT). However, HR kinetics may differ between tests. Pichon et al. reported a significantly higher HR at the end of the 6MST than at the end of the 6MWT (13). This was also suggested in patients with interstitial lung disease (18). Based on these findings and on our personal experience, we also chose to evaluate the first 3 minutes of the 6MST. As expected, the correlation between HR_1–3 and HRvt was stronger than between HR_4–6 and HRvt. This could be attributed to the significant difference between HR_1–3 and HR_4–6 (i.e lower HR during the first part of the test). These results are in accordance with findings in patients with interstitial lung disease during the 6MST, showing a gradual increase in HR throughout the test, while in the 6MWT, HR stabilized around the 3rd minute. HR was also significantly higher during the 6MST than during the 6MWT (18). This might suggest that the first part of the 6MST, during which HR was lower, may reflect aerobic exercise whereas the last part of the 6MST may reflect an exertion that gradually reached maximal oxygen consumption, as suggested in patients with interstitial lung disease during a 6-minute step test (19). Nevertheless, this assumption should be specifically assessed in future studies. The stronger correlation of mean HR_1–3 with HRvt and lower HR suggests that mean HR_1–3 should be used to individualize exercise intensity in patients with COPD.

To our knowledge, this is the first study to compare HR during the 6MST and HRvt. This is important clinically because it has direct implications for the determination of the level of aerobic training for each individual. Moreover, CPET is considered as the optimal assessment for patients with COPD prior to PR. However, it is probably not necessary for every patient, particularly for those in the earlier stages of the disease. Thus, a medical and cardiac exercise evaluation, in association with a field exercise capacity test could be sufficient to determine an appropriate PR program for some patients with COPD.

These results are in agreement with those reported by Grobois et al. (12). They also found a relationship between Wvt and the step count during 6MST. The workload intensity for training in patients with COPD could be predicted by the following equation: LogWvt = 0.001722 × step count + 1.248. For example, a patient who performed 190 steps (approximately the mean observed in this study) could initially be trained at 35–40 W. However, it is important to remember that whether the prescription of intensity is based on CPET or derived from a field test, it simply provides an indication of the appropriate intensity at which to start training, and should then be adapted to patient's tolerance (dyspnea or muscle fatigue) (20, 21). Unfortunately, we could not compare this equation with the data from Grobois et al. because they did not perform linear regression and did not discuss this outcome. Nevertheless, it has important clinical implications and further studies, should be performed to validate the use of this equation for the implementation of optimal PR.

This study compared effort on a stepper and on an ergometer. Since the effort provided on each is substantially different, some parameters could have influenced the results, including muscle mass, weight, and balance. As weight and balance were both controlled by the inclusion criteria (weight <90 kg and ability to perform the 6MST), it is unlikely that they strongly influenced the results. The type of exercise also influences cardiopulmonary parameters. For example, Pichon et al. found a significantly higher level of desaturation during the 6MWT than during the 6MST. They attributed this to differences in the biomechanical and metabolic requirements of the tests (13). It could be hypothesized that similar differences exist between the effort carried out on a stepper and on an ergometer. Moreover, variations in individual dynamic hyperinflation and lung mechanics may also occur in response to a specific exercise.

Finally, the results showed no significant relationship between step count and respiratory function (FEV1). This is not surprising because there is only a weak-to-moderate positive association between FEV1 and objectively measured physical activity in patients with COPD (22).

The utility of equations for exercise prescription has recently been discussed by Kirkham et al. (23). They assessed the errors associated with exercise prescription based on estimated values from 11 equations (derived from the distance walked during the 6MWT) identified through their research. They concluded that the use of these equations for exercise prescription would result in a significant error (from 12 to 47%, depending on the method used) and should not be used to prescribe exercise intensity. A key point highlighted by the authors is that some of these equations were derived from data collected in the research setting, which may differ from data collected in clinical setting and thus limit their use for clinical purpose. As the present study was a retrospective clinical study, all assessments and data collections were carried out in the clinical setting. This likely strengthens the clinical relevance of our results. Nevertheless, the reliability of the equations found in this study need be tested in a new sample of patients. Furthermore, Kirkham et al. (23) discussed the utility of other methods of exercise intensity prescription, rather than the distance during the 6MWT, including the use of heart rate, which was used by Bonnet et al. (9) and in the present study.

This study has several limitations. First, we cannot rule out some bias due to it retrospective design. However, all assessments were standardized according to current guidelines and data were collected in the clinical setting, strengthening the potential relevance of the determined equations (see above). Secondly, the device used to monitor HR may have been influenced by movement artifacts and electrocardiogram telemetry might have been more accurate. Equally, the accuracy and reliability of the device to measure heart rate during the 6MST has never been assessed and compared with telemetry. However, the specification and technical information (from the manufacturer) for the Nonin 3150 states an accuracy of more or a less 3 bpm with or without motion, and in the range of 40–240 bpm in conditions of low perfusion. This study was based in the clinical setting in which telemetry is not available but pulse oxymetry is commonly used. Pulse oxymetry is frequently used in research to monitor HR during field tests, particularly during the 6MST (10–12). Moreover, patients were in different stages of the disease, the sample was small and addressed a specific population undergoing PR. For example, patients were excluded if they weighed more than 90 kg, which is the weight limit supported by the device. Furthermore, they were mostly male, non-oxygen users, and were able to achieve the ventilatory threshold during CPET which might not reflect the whole population referred for PR. However, CPET is mostly necessary in the case of severe COPD and, ideally, should not be replaced by field tests in that population. Finally, the VT was selected by human observers. This has recently been questioned in patients with COPD and could have introduced some inter-individual error in HR at the VT in more severe patients (24). Thus, further studies should prospectively investigate the relationship between the 6MST and CPET in patients in the early stages of COPD (i.e. patients who can achieve the ventilatory threshold) in order to individualize their aerobic training.

Conclusion

The 6MST could be used to individualize aerobic training in patients with COPD in an ambulatory setting. Further prospective studies are needed to confirm these results and could focus on patients in the early stages of COPD.

Acknowledgments

This work was supported by ADIR Association. We also thank Gwenaëlle Leteurtre for her support during data collection and Johanna Robertson for revision of the English.

Declaration of interest

The authors state that they have no conflicts of interest.

Funding support

We wish to thank the Union des Kinésithérapeutes Respiratoires (UKR) for financial support with the submission process.