Functional Electrical Stimulation Changes Muscle Oxygenation in Patients with Chronic Obstructive Pulmonary Disease During Moderate-Intensity Exercise: A Secondary Analysis

Abstract

We previously showed that functional electrical stimulation during cycle ergometry (FES-cycling) increased oxygen consumption (VO₂), indicating that metabolism during exercise was increased. However, the effects on muscle oxygenation have never been studied. The aim of this secondary analysis was to analyse changes in muscle oxygenation during an FES-cycling session. Eight patients with chronic obstructive pulmonary disease who were participating in a pulmonary rehabilitation programme were enrolled. Each participant carried out 30 minutes of cycle ergometry with a constant load at 50% of peak oxygen uptake, either (i) with FES or (ii) without (Placebo-FES). Oxygenation of the vastus lateralis (VL) muscle over time was measured using near-infrared spectroscopy (NIRS) during both sessions. External power output on the cycle ergometer was the same in both conditions. There were no differences in dyspnoea between the groups, although the concentrations of deoxygenated haemoglobin and myoglobin (deoxy(Hb + Mb)) in the VL were significantly greater during Placebo-FES than FES-Cycling (respectively +212 ± 65% vs. +84 ± 29%; p < 0.001), as was the decrease in muscle oxygen saturation (StO₂) (p < 0.001). When adjusted for VO₂, there was a greater increase over time in the deoxy(Hb + Mb)/VO₂ ratio during Placebo-FES than FES-cycling (p < 0.0001). FES-cycling could be a useful strategy to decrease muscular deoxy(Hb + Mb) and limit decreases in muscle StO₂, however this should be confirmed in larger studies.

Keywords:

• COPD
• functional electrical stimulation cycling
• muscle oxygenation
• pulmonary rehabilitation

Introduction

In chronic obstructive pulmonary disease (COPD), the loss of respiratory function and associated muscle dysfunction reduce the capacity to carry out activities of daily living in affected individuals. Several authors reported gross reductions in the volume and strength of the quadriceps muscle, the number of capillaries within the muscle and its oxidative capacity, as well as changes in muscle fibre distribution.^1–3 Pulmonary rehabilitation programmes encourage patients to participate in physical exercise to improve physical exercise capacity and reduce exertional dyspnoea. Oxygen therapy and non-invasive ventilation are sometimes used to facilitate fitness training for patients with severe dyspnoea.^4,5

In a previous study, we evaluated a new approach to facilitate physical exercise.⁶ Rather than relieving dyspnoea by providing ventilatory support, we increased muscle work directly through electrical stimulation of the quadriceps, synchronised with pedalling movements on a cycle ergometer (FES-cycling). The results showed that FES-cycling increased oxygen consumption (VO₂)_, indicating that exercise metabolism was increased. We hypothesised that this increase in VO₂ was due to an increased recruitment of the quadriceps via the electrical stimulation. However, to our knowledge, no studies have yet investigated the effects of FES-cycling on muscle in individuals with COPD.

The aim of this secondary analysis was to compare changes in muscle oxygenation during training sessions with FES-cycling and with Placebo-FES. Muscle oxygenation during exertion was measured non-invasively, in real-time using near-infrared spectroscopy (NIRS).

Method

Design

This study involved analysis of secondary data from a study published by Medrinal et al.⁶ The original study was a single-centre, cross-over, controlled, randomised and single-blind trial. Each participant completed two 30-minute cycle ergometer sessions (with functional electrical stimulation (FES-cycling) and without (Placebo-FES)). The session orders were randomised (by a co-worker who was not directly involved in the study) and placed into sealed envelopes for each participant. The investigator did not open these envelopes until just before the first session.⁷

The original study was registered on ClinicalTrials.gov (NCT02594722) and the present work required a modification to the initial protocol, which was granted by our local ethics committee (Comité de Protection des Personnes Nord-Ouest III). The modification involved the addition of muscle oxygenation analysis for each of the last ten participants included in the study. All participants provided written consent, in accordance with the Declaration of Helsinki, and the study was conducted, analysed and reported according to the Consolidation Standards of Reporting Trials (CONSORT) statement for randomised controlled trials.

Population

Participants were recruited from patients enrolled in a pulmonary rehabilitation programme in the Groupe Hospitalier du Havre. The inclusion criterion was stage 2, 3 or 4 COPD, according to the GOLD classification. Exclusion criteria were: exacerbation of respiratory function less than four weeks previously, implanted with a pacemaker, contraindications to either functional electrical stimulation or to undertaking 30 minutes of pedalling, or an inability to pedal.

Interventions

Before participating in the study (during the 48 hours before the first session), a pre experimental exercise test was carried out to titrate the external power output that corresponded to 50% of the peak VO₂ achieved during the incremental exercise testing. This external power output was then used in both conditions.

Each session of cycle ergometry began with a 2-minute warm-up. Participants were then instructed to maintain a frequency of 50–60 rpm for 30 minutes. The cycle ergometer used was a MotoMed Letto II® (RECK-Technik, Germany). A 30-minute break was allowed between the cycle ergometry sessions in order to ensure that cardio-respiratory and muscle function returned to their baseline states.

FES-cycling

Functional electrical stimulation was applied and coordinated with the cycling using a stimulator that was relayed to a computer (RehaMove®, Hasomed, Germany). For all participants, the electrical stimulation was set to a frequency of 35 Hz and impulse length of 300 μs, in accordance with conventional electrotherapy protocols.⁸ Rectangular RehaMove® electrodes (5 × 9 cm) were placed at each extremity of both quadriceps muscles (total of four electrodes). They were used to produce an intermittent, rectangular current with sufficient intensity to induce a visible or palpable muscle contraction that was tolerable for the participant.

Placebo-FES

For the placebo sessions, the conditions were the same as for the FES-cycling: both used the same electrical stimulation device and cycle ergometer and the electrodes remained in situ during both sessions, regardless of the session order. However, in the placebo condition, the electrical stimulation was set to a frequency of 2Hz and impulse length of 300 μs, with an intensity that was too low to induce a contraction.⁹

Measurements

During each cycle ergometry session, muscle oxygenation was measured on the vastus lateralis (VL) muscle using a commercially available NIRS device (Portamon, Artinis Medical Systems, The Netherlands). The Portamon device was a continuous-wave system, which simultaneously used the modified Beer-Lambert and spatially resolved spectroscopy methods. This system was composed of three optodes (positioned 30, 35 and 40 mm from the receptor) that emitted two different wavelengths of light (760 and 850 nm) which were captured by a receptor. The maximum penetration depth for this device is 20 mm. In biological tissue, diffusion is an important phenomenon. The actual length of the optical trajectory may be greater than the physical distance between the emitter and the detector. In order to account for the longer optical trajectory, a correction factor is incorporated. This factor, called differential pathlength factor (DPF), accounts for the increase in optical pathlength due to scattering in the tissue. A DPF value of 4 was used in the present study. The NIRS device was fixed using an elastic bandage to minimise movement during cycling and the bandage was black to avoid light-related errors.¹⁰

The signal emitted from the optodes was captured at a frequency of 1 Hz and measured the concentration of (1) haemoglobin (Hb) and myoglobin (Mb) oxygenated (oxy(Hb + Mb)), (2) Hb + Mb deoxygenated (deoxy(Hb + Mb)) and (3) total haemoglobin and myoglobin (Total(Hb + Mb)) in the area of interrogation. Total(Hb + Mb) is the sum of oxy(Hb + Mb) and deoxy(Hb + Mb). This represents the volume of blood in the part of the muscle evaluated by the NIRS.^11,12 The tissue saturation index provided an estimate of the oxygen saturation of the tissue (StO2) defined as:

StO2 = 100×(oxy(Hb + Mb)/total(Hb + Mb).

The contribution of Hb and Mb in vivo to the NIRS signal is not well known for muscle oxygenation, which made the data difficult to interpret. Absorption by Hb or Mb in large vessels is high. NIRS therefore provides information regarding oxygen status, particularly in the small blood vessels (arterioles, capillaries and venules) and myocytes which absorb NIRS to a lesser extent.

The concentrations of oxy(Hb + Mb), deoxy(Hb + Mb) and total(Hb + Mb) were expressed in arbitrary units (A.U) by the NIRS. To facilitate intra- and inter-participant comparisons, changes in oxy(Hb + Mb), deoxy(Hb + Mb) and total(Hb + Mb) during exertion were converted into percentage change from the participant’s baseline values. StO₂ (ratio of oxy(Hb + Mb) to total(Hb + Mb)) was expressed as an absolute %. StO₂ was expressed as the Δ of change from baseline.

Participants also wore a face mask relayed to an infrared gas analyser to measure breath-by-breath gas exchange (Vmax Vyntus™ CPX, Carefusion) during both sessions.

Primary outcome

The primary outcome measure was the change in oxygenation of the VL muscle over time in comparison to baseline values (ΔDeoxy(Hb + Mb)) during both cycling sessions. We chose to analyse change in deoxy(Hb + Mb) for the following reasons: (1) In contrast with oxy(Hb + Mb) and total(Hb + Mb), changes in deoxy(Hb + Mb) are relatively insensitive to alterations in skin blood volume or increases in flow during exercise. Changes in deoxy(Hb + Mb) are often associated with the change in deoxygenation of the muscle under the sensor. (2) Changes in deoxy(Hb + Mb) are often considered as a non-invasive means of measuring microvascular extraction of O₂.¹³ Nevertheless, this depends on certain factors: first the balance between O₂ delivery (QO₂) and O₂ uptake (VO₂); and second, changes in haematocrit, capillary–venous haem concentration, and/or volume of small blood vessels. Variations in these factors can reduce the association between deoxy(Hb + Mb) and the microvascular extraction of O₂. Nevertheless, it appears that the effect of these factors is reduced during exertion, probably because muscle contractions limit vascular dilatation and microvascular hemoconcentration.¹⁴

Secondary outcomes

Secondary outcome measures were changes in Oxy(Hb + Mb), Total(Hb + Mb) and StO₂ of the VL muscle during exertion (compared with baseline).

The kinetics of deoxy(Hb + Mb) (time constant, amplitude, plateau) were analysed for both cycling sessions. The change in deoxy(Hb + Mb) was measured each second. The time courses were modelled using the following non-linear function^15,16: $$\text{deoxy}\left(\text{Hb}+\text{Mb}\right)(\tau )=\text{deoxy}{\left(\text{Hb}+\text{Mb}\right)}_{\text{rest}}+\left(\text{deoxy}{\left(\text{Hb}+\text{Mb}\right)}_{\text{ss}}-\text{deoxy}{\left(\text{Hb}+\text{Mb}\right)}_{\text{rest}}\right))*\left(1-\mathrm{e}-\mathrm{t}/\tau \right)$$

Here “deoxy(Hb + Mb)_rest” represented the initial level of deoxy(Hb + Mb) at rest, “deoxy(Hb + Mb)_ss” represented the steady state of deoxy(Hb + Mb) during exertion, and τ (time constant) represented the time course of the monoexponential deoxy(Hb + Mb) curve. The amplitude of the deoxy(Hb + Mb) (deoxy(Hb + Mb)_ampl) corresponds to the difference between deoxy(Hb + Mb)_ss and deoxy(Hb + Mb)_rest.

Cardiac frequency (CF), dyspnoea (as measured on the modified Borg scale from 0 to 10), lower limb fatigue (VAS 0–10) and oxygen consumption (VO₂) were also evaluated. The deoxy(Hb + Mb)/VO₂ ratio was calculated to normalise the variations in muscle oxygen as a function of VO_2.¹⁷ This ratio represents a balance between the excess oxygen extracted and O₂ consumption.

Statistical analysis

Data are expressed in numbers and percentages for categorical variables and as means and standard deviations for continuous, quantitative variables. The normality of the distribution of each variable analysed was evaluated using the Standardised Skewness and Kurtosis scores. Mean or median values during exertion were analysed using a paired Student t-test or a Wilcoxon matched pair signed rank test. When significant differences were found, oxygen concentration kinetics were modelled using non-linear models. The time courses of change in the deoxy(Hb + Mb)/VO₂ ratio were compared using a Two-Way Anova. The decline in StO₂ from baseline was analysed at 25, 50, 75 and 100% of exercise time and compared using a Two-Way Anova. All analyses were carried out using GraphPad Prism 5 software. A p value <0.05 was considered significant.

Results

Data for FES-cycling and Placebo-FES sessions from eight participants were analysed. Originally ten subjects were eligible for participation, but one withdrew consent on the day of the test and a technical issue rendered another participant’s NIRS signal uninterpretable.

The mean age of the participants was 60 years, the mean FEV1 was 52% of the predicted value and the mean distance walked on the 6MWT was 81% of the predicted distance (see Table 1).

Table

Patient characteristics (n = 8)	Mean ± SD
Biological sex, F/M	5/3
Age, y	60.2 ± 6.7
BMI, kg/m²	23.6 ± 4.1
Pulmonary function
FVC, L	2.6 ± 0.5
FVC, % predicted	80.6 ± 18
FEV1, L	1.4 ± 0.5
FEV1, % predicted	52.5 ± 18
Gold 2, n (%)	5
Gold 3, n (%)	2
Gold 4, n (%)	1
FEV1/FVC, %	56.6 ± 9.2
IC, L	1.7 [1.4–2.2]
TLC, % predicted	106.3 ± 20
RV, % predicted	150.7 ± 48.7
DLCO, % predicted	48.6 ± 14.4
Cardiopulmonary Exercise Testing
VO2peak (mL/min)	1102 [923–1425]
VO2peak, % predicted	75.8 ± 9.3
HRpeak, bpm	130 ± 25
SBPpeak, mmHg	167 [150–189]
DBPpeak, mmHg	99 ± 15.6
VEpeak, L/min	39 [35–59]
VEpeak/MVV, %	120 ± 20
SpO2peak	96 ± 3.9
6MWT, m	447 ± 25
6MWT, % predicted	81.3 ± 6.6
Work load intensity, W	35 [30–49]

Mean ± SD or Median [Q1–Q3]. Definition of abbreviations: BMI, Body Mass Index; FVC, Forced Vital Capacity; FEV1, Forced Expiratory Volume in one second; IC, Inspiratory Capacity; TLC, Total Lung Capacity; RV, Residual Volume; DLCO, carbon monoxide transfer factor; VO₂peak, peak of oxygen consumption; HR, Heart Rate; SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure; VE, minute ventilation; SpO₂peak, oxygen saturation measured at the VO₂peak; 6MWT, Six-Minute Walk Test. IC, VO₂peak, SBP, VE and work load intensity are expressed as medians [Q1–Q3]

The results for the metabolic parameters, dyspnoea and lower limb fatigue are provided in Table 2.

Table 2. Differences in physiological variables

Variable	Placebo-FES	FES-Cycling	Mean difference between interventions (95%CI)	p value
Leg muscle oxygenation
Δ Oxy(Hb + Mb) (%)	164 ± 32	133 ± 10	30.8 (−24 to 86.3)	0.13
Δ Total(Hb + Mb) (%)	65.9 ± 30	39.4 ± 15.4	26.5 (−9.6 to 62.6)	0.08
Δ Deoxy(Hb + Mb) (%)	212 ± 65	84 ± 29	128 (125 to 130)	<0.0001
Δ StO2 (%)	−15.5 ± 6.3	−7.7 ± 8.1	−7.7 (−8.3 to 7.3)	<0.0001
Muscle deoxygenation kinetics
Δ Deoxy(Hb + Mb)ss (%)	242.8 ± 1	95.3 ± 0.7	147.5 (146 to 148.2)	<0.01
Δ Deoxy(Hb + Mb)ampl (%)	293.9 ± 19	112.7 ± 13	181 (178 to 184)	<0.001
Δ Deoxy(Hb + Mb) (τ) (s)	149.4 ± 16	146.8 ± 27	2.6 (−7.2 to 3.5)	0.87
Metabolic
VO2 (mL/min)	655 ± 18	704 ± 5.1	−48.6 (−81.5 to 15.7)	0.02
Heart rate (b/min)	97.3 ± 7	100 ± 9.7	−2.3 (−3 to 7.8)	0.33
Subjective
Breathlessness (Borg)	2 ± 0.9	1.9 ± 1.1	0.12 (−0.9 to 0.7)	0.73
Leg fatigue (VAS)	2.1 ± 1.1	2.1 ± 1.2	0 (−0.9 to 0.9)	1

Values are means (SD). Breathlessness and leg fatigue were measured using the Borg scale (0–10) and a visual analogue scale (VAS) (0–10) respectively, with higher scores indicating more severe symptoms. Δ Oxy(Hb + Mb), Δ Deoxy(Hb + Mb) and Δ Total(Hb + Mb) values are expressed as the percentage change from baseline. Oxy(Hb + Mb), muscle haemoglobin and myoglobin oxygenated; Deoxy(Hb + Mb), muscle haemoglobin and myoglobin deoxygenated; Total(Hb + Mb), total muscle haemoglobin and myoglobin (changes in local blood volume); VO₂, oxygen consumption. Deoxy(Hb + Mb)_rest represented the initial level of Deoxy(Hb + Mb) at rest, Deoxy(Hb + Mb)_ss represented the steady state of Deoxy(Hb + Mb) during exertion, and τ (time constant) represented the time course of the monoexponential Deoxy(Hb + Mb) curve. The amplitude of the Deoxy(Hb + Mb) (Deoxy(Hb + Mb)_ampl) corresponds to the difference between Deoxy(Hb + Mb)_ss and Deoxy(Hb + Mb)_rest.

The Δdeoxy(Hb + Mb) was significantly greater for the placebo intervention compared with FES-Cycling (respectively +212 ± 65% vs. +84 ± 29%; p < 0.001).

The kinetics of deoxy(Hb + Mb) concentrations are shown by a non-linear regression in Figure 1. Changes in amplitude of the deoxy(Hb + Mb) over time were significantly greater for the placebo intervention compared with FES-Cycling (deoxy(Hb + Mb)_ampl placebo vs. deoxy(Hb + Mb)_ampl FES, respectively 294 (95% CI 284–304) % vs. 113 (95% CI 106–119) %; p < 0.01). No significant differences were found for the kinetics time constant of deoxy(Hb + Mb) between the interventions (see Table 2). Individual patient data and R square values are presented in Supplemental Data Tables 1 and 2.

Figure 1. Comparison of the kinetics of oxygen extraction between FES-cycling (grey line) and Placebo-FES (black line). There was no significant difference in Deoxy (Hb + Mb) time constant kinetics between groups. There was a significant difference between the amplitude of oxygen extraction (p < 0.01); Change in deoxy(Hb + Mb) over time is modelled by a non-linear regression).

When adjusted for VO₂, the two-way ANOVA revealed a greater increase over time in the deoxy(Hb + Mb)/VO₂ ratio for the Placebo-FES intervention than for FES-cycling (p < 0.0001) (see Figure 2).

Figure 2. Comparison of Deoxy(Hb + Mb) as a function of oxygen consumption between FES-Cycling (solid line) and Placebo-FES (dashed line). (A) The grey area represents the warm-up period and the white area shows the active phase (50% of the VO₂ peak). (B) The Deoxy(Hb + Mb)/VO₂ ratio was analysed at 25, 50, 75 and 100% of exercise time. The significant increase in the ratio in the Placebo-FES group shows an excess in oxygen extraction for a given VO₂ ((p < 0.0001); the difference between conditions over time was analysed using two-way ANOVA).

With regard to the other NIRS variables, there was no significant difference between the interventions for the Δtotal(Hb + Mb) of Δoxy(Hb + Mb) during exertion, however StO₂ decreased significantly during the placebo intervention (p < 0.01) (see Figure 3).

Figure 3. Comparison of tissue saturation (StO₂) of the VL muscle during FES-Cycling (dashed line) and Placebo-FES (solid line). (A) The grey area represents the warm-up period and the white area shows the active phase (50% of the VO₂ peak). (B) The decline from baseline of StO₂ was analysed at 25, 50, 75 and 100% of exercise time. ΔStO₂ of the VL during exertion was significantly lower during Placebo-FES (p < 0.0001). The difference between the two interventions over time was analysed by a two-way ANOVA).

Discussion

This study is the first to evaluate the effects of FES-cycling on skeletal muscle oxygenation in individuals with COPD. Muscle and tissue deoxygenation (deoxy(Hb + Mb)) occurred to a significantly lower extent during the FES-Cycling sessions compared with the FES-Placebo (+84 ± 29% vs +212 ± 65%; respectively, p < 0.001), although there was no concomitant change in kinetics.

This smaller increase in deoxy(Hb + Mb) could reflect a lower level of voluntary muscle activity during the FES-cycling sessions. This idea is supported by several facts: firstly, the VL muscle (from which the NIRS measurements were taken) is mainly active during the extension phase of cycling (1st quadrant).¹⁸ Changes in deoxy(Hb + Mb) are often associated with the microvascular extraction of O₂, thus this is suggestive of less voluntary muscle activity.¹³ Secondly, there was a significant decrease in StO₂ over time during the placebo intervention, indicating that the availability of oxygen to the muscles was reduced and oxygen consumption was increased.^19–21

No differences were found in muscle fatigue between the two interventions. Paillard et al. showed that supplementation of voluntary contractions by electrical stimulation reduces muscle fatigue (measured by an isokinetic dynamometer).²² It is possible that a reduced level of voluntary muscle activity could decrease the energetic cost and oxygen demand, thus reducing muscle fatigue. In the present study, muscle fatigue was measured on a VAS, which is subjective, rather than on an isokinetic dynamometer, which is more objective.²³

The deoxy(Hb + Mb)/VO₂ ratio was significantly lower during the FES-Cycling than during the placebo intervention (p < 0.0001). This could be attributed to two factors. (1) We previously evoked the fact that there was significantly less deoxy(Hb + Mb) under the sensor in the FES-Cycling group. This decrease in deoxy(Hb + Mb) is partially responsible for the decrease in the deoxy(Hb + Mb)/VO₂ ratio. (2) There was a significant increase in VO₂ in the FES-Cycling group. This further amplified the decrease in the deoxy(Hb + Mb)/VO₂ ratio. These results are surprising since theoretically, if we consider the hypothesis that deoxy(Hb + Mb) is associated with the microvascular extraction of O₂ under the sensor during exertion, an increase in VO₂ could lead to an increase in deoxy(Hb + Mb). In the present case, this did not occur. It must be noted that we only measured variations in muscle oxygenation of the VL with NIRS. It is likely that the decrease in deoxy(Hb + Mb) in the FES-cycling group is due to a smaller degree of activity of the VL (reflecting voluntary activity). Furthermore, we propose that the significant increase in VO₂ was due to the increase in the number of motor units recruited in the rectus femoris muscle by the electrical stimulation, which was superimposed on the voluntary contractions. We cannot confirm this hypothesis since the NIRS system used only measured voluntary VL activity, and not electrically-induced rectus femoris activity. Nevertheless, these results suggest that the metabolic load of the quadriceps muscle was increased by the FES, without an associated increase in dyspnoea. This indicates that FES-cycling could be an appropriate method for patients with severe COPD who have difficulty initiating a pulmonary rehabilitation programme.

Muscle and tissue oxy(Hb + Mb)) and muscle and tissue blood volume (total(Hb + Mb)) both increased over time, with no differences in these variables between the FES-cycling and Placebo interventions. The increase in these factors over time during both interventions could be the result of local vasodilatation induced by the exercise and the recruitment of arterioles and capillaries. These results are in agreement with data from a study carried out in healthy subjects during moderate intensity exercise (40% and 60% of VO_2peak).^24,25

This work, a secondary analysis of a randomised, controlled cross-over study, provided new information regarding muscle activity during a novel training method: FES-cycling. The design limited confounding factors relating to the NIRS and successfully demonstrated that NIRS is an easy-to-use and non-invasive technique that is suitable for the continuous measurement of muscle activity and exercise intensity. Participants were evaluated under the same conditions and on the same day, but with sufficient rest time in between sessions; there was no evidence to suggest that there was an effect of session order or a carry-over effect.⁶ The NIRS hardware was well fixed to the participant to limit data artefacts being recorded during the exercise and, in addition, the NIRS was not removed between the sessions.

This study has several limitations: (1) only eight participants underwent NIRS measurements, (2) the contribution of Hb and Mb in vivo to the NIRS signal is not well known for muscle oxygenation which made data difficult to interpret. Despite this, the Mb desaturation apparently contributes to less than 20% of the total NIRS desaturation signal during exercise,²⁶ (3) catheters were not used to precisely measure changes in femoral arterial and venous oxygen, although StO₂ results were linearly related to femoral venous saturation indicating that NIRS could be used as a measure of oxygen consumption as a function of local blood flow,²⁷ (4) changes in oxygenation of the rectus femoris muscle were not measured, which might have shown greater activity in this muscle during FES-cycling due to the combined electrical stimulation and voluntary activity and (5) changes in deoxy(Hb + Mb) can be influenced by certain factors (QO₂/VO₂) such as changes in haematocrit, capillary–venous haem concentration, and/or volume of small blood vessels). Although the influence of these factors is reduced during exertion, we wish to highlight the complexity of interpreting changes in deoxy(Hb + Mb).¹⁴

Conclusions

The results of this secondary analysis suggest oxygen extraction was greater during the Placebo-FES intervention than during FES-cycling, possibly due to a higher level of voluntary muscle activity with no electrical stimulation. Nevertheless, VO₂ was greater in the FES-cycling group. This was probably due to a greater recruitment of muscle fibres in the whole of the quadriceps muscle with the supplementation of voluntary muscle activity with electrical stimulation, resulting in an increase in the metabolic load of the quadriceps with no increase in dyspnoea. The results of this secondary analysis of our previous study suggest FES-cycling could decrease muscle deoxy(Hb + Mb) and limit decreases in muscle StO₂, thus opening the door to future research in this new type of training.

Disclosure of interest

The authors report no conflict of interest

Acknowledgments

We thank Johanna Robertson and Jennifer Dandrea Palethorpe for translation and constructive criticism.

Additional information

Funding

This work was supported by ADIR Association (Rouen, France) and GHAHR Association (Le Havre, France).