The Six-minute Step Test as an Exercise Outcome in Chronic Obstructive Pulmonary Disease

To the Editor:

Exercise capacity predicts adverse outcomes, identifies exercise-induced oxygen desaturation, and facilitates exercise prescription in chronic obstructive pulmonary disease (COPD). Established tests of exercise capacity such as cardiopulmonary exercise testing (CPET) or field walking tests (6-minute walking test [6MWT] and incremental shuttle walking test [ISWT]) can be limited by equipment, time, or space and are not feasible in some settings (e.g., the home). Field walking tests also require a practice walk to account for the learning effect.

The 6-minute step test (6MST) is a simple, functional test adapted from the 6MWT (1). It is reproducible in COPD (2), correlates with 6MWT distance (3), is responsive to a physical training program (4), and identifies exercise-induced desaturation in interstitial lung disease (5). However, it remains unclear whether the 6MST generates a similar cardiorespiratory response to more established maximal tests of exercise capacity, identifies exercise-induced desaturation in COPD, or can be used for exercise prescription.

We aimed to: 1) determine the association between 6MST with ISWT and CPET; 2) compare cardiorespiratory responses between 6MST, ISWT, and CPET; 3) generate equations on the basis of 6MST for exercise prescription; and 4) compare the responsiveness of 6MST and ISWT to pulmonary rehabilitation (PR).

Methods

Participants were prospectively recruited from those attending PR assessment clinics. Inclusion criteria were: a diagnosis of COPD, an MRC (Medical Research Council) dyspnea score of 2 or higher, and the ability to walk 5 minutes independently. The study was approved by the London Riverside Research Ethics Committee (reference 17/LO/1830). All participants provided informed consent.

CPET, ISWT, and 6MST were performed in random order, with the assessor blinded to the order and the results of the other assessments. For all tests, heart rate, oxygen saturation as measured by pulse oximetry (Sp_O2), and breathlessness scores using a modified Borg CR-10 scale were measured at baseline, 1-minute intervals during exercise, and end of the exercise.

The 6MST was self-paced and performed using a 20-cm high, single-step platform as previously described (5). Standardized encouragement was given at the end of each minute per 6MWT technical standards. The total number of steps achieved in 6 minutes was recorded. The ISWT was performed as previously described (6). The CPET was performed using a cycle ergometer (Ergoselect 200, Ergoline GmbH) with a metabolic cart (Ultima Cardio₂ gas exchange analysis system, MCG Diagnostics) (7).

Participants attended an 8-week, twice-weekly, supervised PR program conducted according to British Thoracic Society Quality Standards (8).

To demonstrate a strong correlation (r > 0.7) between 6MST with ISWT distance, peak oxygen consumption ($\dot{\text{V}}$O_2peak) and peak workload against the null hypothesis (r = 0) with 95% power at a P value threshold of 0.05 required a minimum of 20 participants.

The relationship between variables was analyzed using Pearson’s correlation coefficient or Spearman’s rank correlation, depending on the distribution of data. Changes between baseline and peak heart rate, Sp_O2 nadir, and peak Borg scores were compared between 6MST, ISWT, and CPET using paired sample t tests or Wilcoxon matched-pairs signed-rank tests. Responses of the 6MST and ISWT to PR were assessed using a paired t test, and the standardized mean responses calculated.

On the basis of the ISWT and CPET, we calculated exercise intensity prescription for walking speed (75% of peak walking speed) and cycling work rate (60% of peak cycle work rate) as per the Lung Foundation Australia Pulmonary Rehabilitation Toolkit (9). Multivariable regression was performed to generate predictive equations for walking speed and cycling work rate prescriptions on the basis of 6MST step count. Bland-Altman plots were constructed to demonstrate agreement between the predictive equation derived and the actual prescribed walking speed and cycling work rate by plotting the mean difference between the two measures against the mean of the two measures.

Results

Twenty-four participants (Table 1) completed 6MST and ISWT, with 23 also performing CPET. 6MST correlated with ISWT distance, $\dot{\text{V}}$O_2peak, and peak work rate (r = 0.90, 0.77, and 0.72, respectively; all P < 0.05).

Table 1.

Baseline characteristics

Variable (N = 24)
Age (yr), median (25th, 75th centiles)	72.5 (70, 74.75)
Sex (male), mean (SD)	8 (33)
BMI (kgm−2), mean (SD)	28.7 (6.7)
MRC dyspnea score, median (25th, 75th centiles)	4 (2, 4)
FEV1% predicted, mean (SD)	54.0 (19.8)
FVC% predicted, mean (SD)	91.7 (19.8)
Resting SpO2 (%), mean (SD)	98 (98, 99)
Resting heart rate (beats/min), mean (SD)	79 (10)
ISWT (meters), mean (SD)	330 (140)
Change SpO2	−5 (−8, −3)
Change HR	37 (17)
Change Borg	5 (2)
6MST (steps), mean (SD)	70 (23)
Change SpO2	−5 (4)
Change HR	29 (2)
Change Borg	4 (2)
CPET (n = 23), mean (SD)
$\dot{\text{V}}$O2 peak ml/kg/min	12.9 (3.4)
Peak watts	60.5 (23)
Change SpO2	−2 (−4, −2)
Change HR	32 (20)
Change Borg	4 (2)

Definition of abbreviations: 6MST = 6-minute step test; BMI = body mass index; CPET = cardiopulmonary exercise test; FEV₁ = forced expiratory volume in 1 second; FVC = forced vital capacity; HR = heart rate; ISWT = incremental shuttle walk test; MRC = Medical Research Council dyspnea score; SD = standard deviation; Sp_O2 = oxygen saturation as measured by pulse oximetry.

Median (25th, 75th centile) change in Sp_O2 during 6MST was at least similar or more pronounced to that seen during ISWT or CPET (Figure 1). Changes in heart rate and Borg were also similar across tests (Figure 1).

Figure 1.

Box and whisker plots to show a change in (A) Sp_O2, (B) heart rate, and (C) Borg dyspnea score during each test: CPET, 6MST, and ISWT. Error bars show 95% confidence intervals. * = p < 0.05. 6MST = 6-minute step test; CPET = cardiopulmonary exercise test; HR = heart rate; ISWT = incremental shuttle walk test; ns = not significant; Sp_O2 = oxygen saturation as measured by pulse oximetry.

Multivariable linear regression equations incorporating 6MST to predict prescribed walking speed (75% of maximum walking speed, taken from ISWT) and cycling work rate (60% of peak cycle work rate from CPET) prescriptions were:

Initial walking speed prescription $\left(\mathrm{km}/\mathrm{h}\right)\hspace{0.17em}=\hspace{0.17em}1.866\hspace{0.17em}+\hspace{0.17em}(0.028\hspace{0.17em}\times \hspace{0.17em}\mathrm{6MST})$

and

Initial work rate prescription $\left(\mathrm{W}\right)\hspace{0.17em}=\hspace{0.17em}-4.388([0.707\hspace{0.17em}\times \hspace{0.17em}\mathrm{body} \mathrm{mass}\mathrm{index}]\hspace{0.17em}+\hspace{0.17em}[-3.204\hspace{0.17em}\times \hspace{0.17em}\mathrm{MRC}]\hspace{0.17em}+\hspace{0.17em}[0.430\hspace{0.17em}\times \hspace{0.17em}\mathrm{6MST}]).$

These correlated strongly with actual prescribed walking speed and cycling work rate (r = 0.876 and 0.773, respectively). Bland-Altman plots showed good agreement between the predicted and prescribed walking speeds and cycling work rates, with data scatter lying primarily within the limits of agreement (Figure 2).

Figure 2.

Bland-Altman plot of mean difference against the mean of predicted prescribed speed/watts from 6-minute step test versus prescribed speed/watts from (A) incremental shuttle walk test and (B) cardiopulmonary exercise test.

After pulmonary rehabilitation, mean (standard deviation) 6MST increased from 56 (30) steps to 69 (28) steps, with a mean (95% confidence interval [CI]) change of 12 (5–19) steps and a standardized mean response of 1.22. The ISWT showed a trend toward improvement (mean [95% CI] change of 21 [−5 meters to 48 meters]) with a standardized mean response of 0.38.

Discussion

Simple, functional tests require little space or equipment and may be particularly helpful for home-based assessments for PR (10). Although functional tests such as sit-to-stand and gait speed have been shown to be responsive to exercise-based intervention in COPD (11, 12), they are not maximal tests, nor have they been shown to be appropriate for exercise prescription (13).

Climbing steps or stairs is a familiar physical activity to patients with chronic respiratory disease, and the use of the 6MST has been previously described in COPD, ILD, and after coronavirus disease (COVID-19) (2–5, 14). We have demonstrated that the 6MST correlates strongly with ISWT distance, $\dot{\text{V}}$O_2peak, and peak work rate with similar cardiorespiratory responses, providing evidence that the 6MST is a near-maximal test. Furthermore, we have generated equations on the basis of the 6MST to help exercise practitioners prescribe initial exercise intensity (walking speed or cycling work rate) for a PR program. We also demonstrate that the 6MST is as responsive to PR as the ISWT.

Despite adequate power, this was a single-center study in a cohort of symptomatic but stable patients referred for PR, so our data require corroboration in larger cohorts and other patients with COPD (e.g., less symptomatic or those unwell with an acute exacerbation). The exercise testing was also supervised at a PR center, so our data cannot necessarily be extrapolated to the home setting nor for remotely supervised tests. Future work should include the validation and safety testing of remotely supervised step tests in patients with COPD, as demonstrated in a small feasibility study of adults with cystic fibrosis (15).

Conclusions

We propose that the 6MST is a simple field functional test that could be used as a surrogate of exercise capacity to identify oxygen desaturation and determine responsiveness to exercise training in patients with COPD, particularly when more established exercise tests are not possible (e.g., during a home-based assessment).