Cardiorespiratory Responses between One-legged and Two-legged Cycling in Patients with Idiopathic Pulmonary Fibrosis

To the Editor:

International guidelines recommend exercise training within pulmonary rehabilitation (PR) for adults with idiopathic pulmonary fibrosis (IPF) (1). However, the magnitude of benefits of PR in IPF may be less than in chronic obstructive pulmonary disease (COPD) (2) and are not sustained (3). Partitioned muscle training has been investigated for other chronic diseases where a central limitation to exercise dominates (4–6). One-legged cycling partitions the targeted exercising muscle, thereby reducing the total ventilatory burden for the same muscle-specific power. In ventilatory-limited patients with COPD, partitioned training increases cardiorespiratory fitness (4, 7) measured by peak oxygen uptake ($\dot{\mathrm{V}}$o₂pk) greater than that achieved with conventional two-legged cycle training.

We hypothesized that patients with IPF would increase their tolerable exercise time of a leg exercising alone (one-legged cycling) compared with two-legged cycling so that the total work would be doubled (the primary outcome). We also aimed to quantify peripheral muscle aerobic capacity relative to the central capacity by determining the ratio of $\dot{\mathrm{V}}$o₂pk achieved during one- versus two-legged cycling.

Methods

Participants

Patients with IPF (1) and an Medical Research Council (MRC) dyspnea score of 2 or greater were recruited from a tertiary referral center between January and May 2015. Exclusion criteria included the requirement of supplementary oxygen at rest, significant other lung diseases or comorbidities contributing to breathlessness, and conditions that precluded cycling. The study was approved by Yorkshire and The Humber–South Yorkshire Research and Ethics Committee (14/YH/1152). Written informed consent was obtained from all participants before starting the study.

Study design

We used a prospective nonrandomized crossover design. Participants completed four exercise tests (8), separated by at least 48 hours, on an electrically braked ergometer (9) (modified Corival Recumbent; LODE BV) with “fixed” wheel, thereby preserving a natural cycling pattern when cycling with one leg (5). The inactive leg rested a safe distance from the free rotating pedal.

Ventilatory parameters were measured breath-by-breath using a calibrated metabolic system (Quark-CPET; COSMED Ltd.) connected to a mask (V2 Mask; Hans Rudolph Inc.). Electrocardiography and pulse oximetry were monitored continuously. Leg effort and breathlessness scores were assessed at 2-minute intervals using the Borg and modified Borg scales (10), respectively.

Two incremental power tests (IPTs) were performed: first, two-legged cycling (2L-IPT) using an individualized ramp protocol aiming for a duration between 8 and 12 minutes; and second, one-legged cycling (1L-IPT) using the same ramp used for the 2L-IPT.

Two constant power tests (CPTs) that simulated high-intensity aerobic training sessions were performed. The two-legged test (2L-CPT) used 70% of the peak power achieved on the individual’s 2L-IPT, expecting intolerance would occur between 3 and 15 minutes. If the 2L-CPT was >15 min, a second 2L-CPT was performed at an appropriately adjusted higher intensity. The 1L-CPT was performed at half the power of the acceptable 2L-CPT. A maximum duration of 1 hour was decided a priori.

Statistical analysis was performed using SPPS v.22. The normality of the data distribution was tested using the Shapiro-Wilk test. To determine whether there was a significant effect between the conditions of two-legged versus one-legged cycling, paired t tests were used. To demonstrate a large effect (Cohen’s d ≥ 1.0) on the work (kilojoules) achieved (the primary outcome) between the IL-CPT and the 2L-CPT, a sample of 10 was needed with statistical power of 0.80 and α of 0.05 using a two-tailed paired t test.

Results

Fifteen patients were enrolled and 12 patients completed the study (11 male; mean [standard deviation] age, 61 [6] y; forced vital capacity, 72 [20] % predicted; transfer factor for carbon monoxide (TL_CO), 46 [11] % predicted; resting oxygen saturation via pulse oximetry, 98 [1] %). The majority had a GAP (Gender, Age and Physiology) index of stage II; two were prescribed pirfenidone (at the time of recruitment pirfenidone had only recently been licensed in the United Kingdom and was the only antifibrotic agent available) and one oral corticosteroids. Three patients did not complete the protocol for reasons unrelated to the study.

A ventilatory limitation to 2L-IPT was observed in 10 participants. One-legged cycling was well tolerated by all participants. A comparison between the two IPTs is shown in Table 1. Participants achieved 84% of 2L-IPT $\dot{\mathrm{V}}$o₂pk during the IL-IPT, with similar cardiorespiratory responses but less oxygen desaturation throughout IL-IPT.

Table 1.

Comparison of the peak values achieved on the incremental and constant power tests between one- and two-legged cycling

	Incremental Power Tests		Constant Power Tests
	2L-IPT	1L-IPT	2L-CPT	1L-CPT
Power, W	93 ± 32	65 ± 26*	69 ± 24	34 ± 12
Duration, min	N/A	N/A	6.1 ± 3.7	22.7 ± 15.0*
Work, kJ	N/A	N/A	26.7 ± 20.6	53.4 ± 48.3*
$\dot{\mathrm{V}}$o2, ml/min/kg	16.7 ± 3.3	14.0 ± 3.1*	17.0 ± 3.3	14.2 ± 3.3*
$\dot{\mathrm{V}}$o2, % predicted	74 ± 21	N/A	N/A	N/A
$\dot{\mathrm{V}}$co2, ml/min	1,507 ± 375	1,267 ± 400*	1,489 ± 389	1,139 ± 323*
RER	1.02 ± 0.10	1.03 ± 0.11	1.03 ± 0.11	1.08 ± 0.29
$\dot{\mathrm{V}}$e, L/min	68.6 ± 18.1	59.3 ± 18.6*	70.0 ± 23.0	61.6 ± 28.1*
% MVV	92 ± 14	80 ± 17	N/A	N/A
fb, breaths/min	49 ± 8	47 ± 10	49 ± 10	51 ± 14
Heart rate, beats/min	119 ± 22	107 ± 23*	118 ± 20	108 ± 20*
End SpO2, %	90 ± 6	92 ± 7	87 ± 7	89 ± 6*
Nadir SpO2, %	88 ± 7	92 ± 6*	87 ± 7	89 ± 6*
Dyspnea, 0–10, median (IQR)	5 (3–7)	4 (4–5)	6 (5–7)	5 (4–7)
LE, 6–20, median (IQR)	17 (15–18)	14 (13–15)	15 (15–17)	17 (15–19)

Definition of abbreviations: 1L = one-legged; 2L = two-legged; CPT = constant power test; IPT = incremental power test; IQR = interquartile range; fb = frequency of breathing; LE = leg effort; MVV = maximal voluntary ventilation; N/A = not applicable; RER = respiratory exchange ratio; Sp_O2 = oxygen saturation via pulse oximetry; $\dot{\mathrm{V}}$co₂ = carbon dioxide output; $\dot{\mathrm{V}}$e = minute ventilation; $\dot{\mathrm{V}}$o₂ = oxygen uptake.

Data presented as mean ± standard deviation unless otherwise noted.

^*Significant difference between one- and two-legged cycling conditions (P < 0.05).

Three initial 2L-CPTs were >15 minutes and repeated at a higher power. A comparison between the 2L-CPT and 1L-CPT is shown in Table 1. For the primary outcome, there was a large effect (Cohen’s d = 1.3), as almost twice the total work (mean difference [95% confidence interval (CI)], 26.7 [5.4–48.1] kJ) was performed during the 1L-CPT (mean [SD], 53.4 [48.3] kJ) compared with the 2L-CPT (mean [SD], 26.7 [20.6] kJ). The increased endurance between the 1L-CPT compared with 2L-CPT (mean difference [95% CI], 16.7 [8.3–25.0] min) was achieved with evidence of reduced central cardiopulmonary demand (Figure 1) as well as less breathlessness and increased leg effort compared with 2LCPT (Figure 2).

Figure 1.

The physiological response to two-legged (open circles) and one-legged (solid circles) constant power exercise tests. Values are mean and standard error. Sp_O2 = oxygen saturation via pulse oximetry.

Figure 2.

The relationship between leg effort and breathlessness during constant power exercise when cycling with two legs (open circles) compared with one leg (solid circles). Insets show leg effort (top left) and breathlessness (bottom right) as a function of time. Values are mean and standard error.

Discussion

Patients with IPF demonstrate significant muscle reserve relative to their central ventilatory capacity, evidenced by the high percentage of 2L-IPT $\dot{\mathrm{V}}$o₂pk achieved on the 1L-IPT. One-legged cycling was well tolerated and, at the same muscle-specific power as required during two-legged cycling, enabled patients with IPF to achieve double the work on an exercise test that simulated a high-intensity aerobic training session. In practice, one would confine exercise to 20 to 30 min (10 to 15 min each leg) by increasing the power, thereby optimizing the training stimulus.

Work was increased on a 1L-CPT by reducing the total ventilatory burden imposed by conventional two-legged cycling; patients cycled for almost four times the duration with less desaturation throughout and a slower progression to similar breathlessness at the end of the exercise. Therefore, one-legged cycling shows promise to be an effective training strategy for patients with IPF. Our observations in patients with IPF are similar to observations in patients with COPD that led to positive efficacy and effectiveness trials of one-legged cycling in that population (4, 7, 9).

Many patients with IPF stop exercise because of severe oxygen desaturation despite the provision of supplemental oxygen. By limiting aerobic exercise to one-legged cycling, desaturation is lessened by reducing total oxygen extraction and substantially decreasing the pulmonary gas exchange requirements to maintain arterial oxygen. It is reasonable to speculate that less supplemental oxygen would be needed during training.

The results may not be directly extrapolated to patients requiring supplementary oxygen at rest, although there is a need to improve the exercise training component of PR offered to this group. We excluded these patients because of the technical challenge of accurately measuring $\dot{\mathrm{V}}$o₂ inspiring a hyperoxic mixture and chose to perform a proof-of-concept study supported by physiological measures in all participants. Last, we acknowledge the potential for a positive effect of performing three short exercise tests before the IL-CPT, but this effect, if any, would likely only contribute a relatively small part to the large effect observed.

In conclusion, one-legged cycling increases muscle-specific work without increasing breathlessness in patients with IPF. Our study provides justification for a randomized controlled trial of aerobic exercise training using one- compared with two-legged cycling in patients with IPF, with the aim of improving the quality of PR.