Arm Cycling Combined with Passive Leg Cycling Enhances VO_2peak in Persons with Spinal Cord Injury Above the Sixth Thoracic Vertebra

Fifteen moderately active individuals with SCI were recruited after informed consent. They were allocated into 2 groups according to level of injury: at or above T6 (SCI-high, n = 8) or below T6 (SCI-low, n = 7).21 Inclusion criteria were stable neurological state with lower motor neurons intact, American Spinal Injury Association Impairment Scale (AIS) A, and ability to FES cycle for a minimum of 1 minute. Exclusion criteria were pacemakers, severe autonomic dysreflexia, and decubitus in any area. Neurological condition according to the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI-AIS)36 and general medical condition were assessed before testing. Body mass was recorded on a SECA digital chair scale 950 (Hamburg, Germany). Individual characteristics are listed in Table 1.

Ventilatory and pulmonary gas exchange parameters were measured using a Metamax II Cortex system (Cortex Biophysik GmbH, Germany). Participants wore a facemask with a volume transducer and tube to collect the gas. The Metamax unit samples gas concentrations every 10 seconds from a mixing chamber. Prior to testing, calibration was performed with a 3-L calibration syringe (Hans Rudolph Jäger GmbH, Germany). For ACE, the ERGOMED 840L (Siemens, Germany) was modified by changing pedals for arm use (Figure 1). The ERGYS 2 Rehabilitation System (Therapeutic Alliances Inc., Ohio) was used for both PLC and FES hybrid cycling (Figure 1).

Hemolyzed blood lactate ([La⁻]_b) was measured by a Lactate Pro LT-1710 Analyzer (Arkray Factory Inc, KDK Corp., Japan) after VO_2peak testing of a capillary blood sample taken from a finger.

Heart rate (HR) was measured with Polar monitors (Polar Electro, Oy, Finland). Peak HR was determined as the highest HR measured at VO_2peak. A rating of perceived exertion (RPE, Borg 6-20) was recorded after each test. Oxygen pulse (O₂ pulse in mL·beat⁻¹) was calculated as the ratio of VO_2peak (mL·kg⁻¹·min⁻¹) to peak HR (beat·min⁻¹).

The 3 exercise tests with separate VO_2peak measurements were performed in random order on separate days. Participants were familiar with ACE but none were competitive wheelchair athletes. For determination criteria, VO_2peak was considered achieved when respiratory exchange ratio (RER) ≥ 1.1, [La⁻]_b ≥ 7, and RPE ≥ 15 (Borg 6-20). In the upper body mode, a VO₂ plateau (stable VO₂ despite increased power output and pulmonary ventilation) is rarely reached, so VO_2peak is used to denote maximal effort. When the aforementioned RER, [La⁻]_b, and RPE criteria were met, the average of the highest VO₂ values within 3 consecutive 10-second measurements was calculated as VO_2peak. All tests were performed with the intention to reach determination criteria for peak aerobic power.

During ACE, the shoulder joint was horizontally aligned with the ACE axis and the elbows slightly flexed at the point of furthest reach. Subjects used their personal rigid frame or electrical wheelchairs, which were placed on a stable platform when required. All participants performed warm-up at their own cadence with zero loading for approximately 5 minutes. Then, 4 minutes of 30 W output on the ACE was followed by progression to peak ACE, where ACE workload was increased by 5 W/min for SCI-high and by 10 W/min for SCI-low subjects.

Participants performed combined ACE-PLC while seated in the ERGYS 2 (Figure 1) to optimize comparison. With legs fixed to the pedals, the test technician manually turned the flywheel at 70 rpm with timing aided by audible cues from a digital metronome. Otherwise, the testing procedure was identical to the ACE protocol described above.

FES hybrid tests began with 5 minutes of warm-up on the ACE, followed by 4 minutes at 30 W on the ACE combined with 2 minutes of FES manual warm-up and then FES hybrid peak testing (Figure 1). No resistance on the flywheel during FES cycling was provided, whereas ACE workload was increased in 1-minute increments by 5 or 10 W (as described) until VO_2peak determination criteria were met. Five participants demonstrated severe muscle fatigue during pretests on the FES (2 SCI-high and 3 SCI-low subjects). Thus, to reach fatigue at approximately the same time for both arms and legs, FES cycling had to be individually set for these 5 participants. Several days prior to the test, the ACE power output (W) performance at 5 minutes before ACE_peak (denoted 5min_peak) and ACE_peak were measured. Then, in the actual FES hybrid test, electrostimulation was started at the predefined 5min_peak. Self-adhesive electrodes were used for electrical stimulation. Impulse frequency was set at 40 Hz and amplitude at 140 mA.21 The other participants performed 2 minutes of manually assisted FES cycling warm-up at an intensity of 50% of the individually set muscle fatigue thresholds. Where necessary, the test personnel provided manual assistance to ensure that pedaling cadence did not drop below a default preset stimulation cut-off speed of 35 rpm and to ensure stimulation current throughout the test.

We certify that all applicable institutional and governmental regulations concerning the ethical study of human volunteers were followed during the course of this research. The research was conducted in accordance with the Declaration of the World Medical Association, and informed consent was obtained from the participants as required.

Tests for multivariate normality and a constant covariance matrix were performed using STATA 2014 statistical software (StataCorp, LP, Texas). Multivariate regression analyses were then used to examine differences in response variables among exercise modalities and between SCI-high and SCI-low groups. Separate multivariate regression analyses were also used to examine within-group differences by level of injury (LOI) in SCI-high subjects. A p value < .05 was considered statistically significant.

Performance on ACE as measured by VO_2peak differed significantly between SCI-high and -low groups (F = 7.03; p = .02). In contrast, groups showed similar VO_2peak on ACE-PLC (F = 2.56; p = .13) and FES hybrid (F = 0.34; p = .57).

In the SCI-high group, mean VO_2peak (SD) during ACE-PLC was 19% higher than during ACE alone [21.0 (3.8) vs 17.6 (5.0) mL·kg⁻¹·min⁻¹; F = 38.10; p = .001; 95% CI, 18.3–23.4 and 14.1–21.2, respectively]. FES hybrid cycling induced an additional VO_2peak increase of 3.5 mL·kg⁻¹·min⁻¹ compared to ACE-PLC [24.4 (4.1) mL·kg⁻¹·min⁻¹; F = 20.68; p = .001; 95% CI, 21.2–27.5] (Figure 2, Table 2 ), although absolute VO_2peak values (L·min⁻¹) did not differ significantly between modalities (Table 2). The mean O₂ pulse was 2 mL·beat⁻¹ higher (24%) during ACE-PLC compared to ACE alone (p = .003; 95% CI, 1.2–4.3) and 7.8% higher during FES hybrid compared to ACE-PLC [11.0 (2.0) vs 10.2 (1.3) mL·kg⁻¹·min⁻¹; p = .01; 95% CI, 0.9–6.0]. Mean PO_peak was 10.5 W higher in the FES hybrid modality compared to ACE alone [83 (31) vs 72.5 (32) W; p = .001; 95% CI, 6–15]. No differences in V_E, [La⁻]_b, RER, HR_peak, RPE, and PO_peak were found between ACE-PLC and ACE alone (Table 2). Regression analysis showed that specific LOI (above T6) did not contribute to these differences among exercise modalities.

In contrast to the SCI-high group, no significant differences in VO_2peak or other related parameters were observed among modalities in the SCI-low group (Figure 2, Table 3).

No signs of clinically relevant autonomic failure, dysreflexia, or related disturbances were observed during the course of this study.

While ACE is a valuable modality to maintain fitness in individuals with SCI, training modalities involving the lower extremities can provide added benefit. This comparison of peak oxygen consumption (VO_2peak) and related metabolic responses during 3 different exercise testing conditions, ACE, ACE-PLC, and FES hybrid, in individuals classified as SCI-high or SCI-low revealed that ACE-PLC significantly increased VO_2peak compared to ACE alone in SCI-high but not SCI-low individuals. However, ACE-PLC did not reach the intensity level of FES hybrid as measured by VO_2peak and PO_peak. Our findings, therefore, indicate that for SCI-low adults, ACE alone is sufficient exercise for cardiovascular fitness as indicated by VO_2peak. The increased VO_2peak in the SCI-high group with additional lower extremity training modalities (ACE-PLC or FES hybrid) may result principally from compensation for neurophysiological regulation deficits, such as SNS dysfunction. Since PSNS pathways do not pass through the spinal cord, high-level SCI leads to unbalanced SNS:PSNS activity, 23,33 specifically reduced SNS efferent activity23 and blunted circulatory responses33 below the LOI. The SCI-high group may have a general decentralization of SNS control (sympathetic vessels + splanchnic vessels) compared to the SCI-low group. Consistent with this notion, exercise responses were enhanced and differently graded in the SCI-high group.

Theoretically, injury-related blunted HR and contractility responses of the heart may be present in the SCI-high but not the SCI-low group. Across training modalities, both HR_peak responses and work output were lower (20 beats·min⁻¹ and 20 W lower) in the SCI-high compared to the SCI-low group. Nash et al19 reported increased SNS activity leading to diffuse noradrenergic discharge affecting HR and withdrawal of vagal tone leading to cardio-acceleration selectively in the SCI-high group. However, the HR_peak responses (Table 2) in SCI-high subjects were moderate and do not suggest widespread SNS influence.

In exercising able-bodied individuals, spinal circuits are responsible for regulating cardiac activity and regional vascular tone to maintain systemic vascular homeostasis.22 However, in individuals with SCI, spinal circuits comprising homeostatic control loops are disrupted. The SCI-high individuals exhibited a superior response to ACE-PLC and FES hybrid compared to the SCI-low group. These differences are unlikely due solely to increased lower leg venous return mediated by PLC or FES because similar responses were found when occlusive thigh cuffs were applied to promote circulatory adjustments.21 Thus, a contribution from “venous pooling” is still unproven. Our findings are supported by those of Thijssen et al,33 who assessed acute changes in blood flow in the brachial artery, portal vein (indirect splanchnic vessels flow), and femoral artery during ACE in SCI-high, SCI-low, and control subjects. A large decrease in splanchnic vessel blood flow was observed in SCI-low individuals and controls, whereas SCI-high individuals exhibited no change. The authors concluded that splanchnic vessels in the SCI-high group respond distinctly from local muscular vessels regarding blood redistribution during ACE testing. A recent study37 found preserved vascular function during PLC in SCI-low individuals. Thus, the more functional SNS system in SCI-low appears to normalize responses among the 3 exercise modalities. A lower leg venous return effect during ACE-PLC appears unlikely as SCI-low subjects exhibited no changes in VO_2peak and related parameters. SCI-high individuals exhibited significantly increased pulmonary ventilation (VE). This probably reflects augmented abdominothoracic muscular work and a diaphragm pumping effect on the lungs. This mechanism alone does not increase VO₂, but may play an important role in promoting central blood redistribution and increasing systolic blood pressure.19 

Individuals with SCI are less physically active and consequently certain diseases are more frequent with increasing age and time since injury.38,39 In this study, the SCI-low individuals were approximately 10 years older than the SCI-high group. Ageing in individuals with SCI most likely contributes to deconditioning,40 which implies reduced exercise tolerance that secondarily could have affected our VO_2peak data.

We also observed a trend towards an inverse relationship between LOI and VO_2peak within the SCI-high group. That is, the higher the LOI, the lower VO_2peak. Again, the lack of a significant correlation in regression analysis may be due to insufficient sample size, so this issue warrants further study.

The sample size is too small to draw general conclusions on the relationship between level of spinal injury and exercise response. However, small sample sizes are often the case in studies on individuals with SCI. The ISNCSCI-AIS score36 does not correspond to the disruption of the autonomic pathways and the functional state of the autonomic nervous system was not evaluated. So contributions due to possible SNS:PNS imbalance or loss of central regulation are still uncertain.

In SCI-high individuals, training modalities in addition to ACE appear necessary to increase training load, while no additional training modality appears necessary to increase the training load in SCI-low individuals. In situations where the more advanced hybrid FES system is not available, PLC during ACE for SCI-high adults appears to facilitate exercise at a higher VO₂ and so may be sufficient to promote cardiovascular fitness.

The authors declare no conflicts of interest. The authors would like to thank all of the individuals with SCI participating in the study.

ACE

Arm crank ergometry is the quantitative measurement and evaluation of physical capacity and cardiopulmonary exercise tolerance (> stress and > strain) in both healthy and diseased subjects as well as trained athletes necessary during arm cranking.

AIS

ASIA Impairment Scale.

ASIA

American Spinal Cord Injury Association.

CVD

Cardiovascular disease includes disorders of the heart, blood vessels, and hemodynamic regulation.

HR

Heart rate is both a basic parameter of the exercise response during ergometry and a reference value for calculation of other parameters assessing cardiac response to stress.

ISNCSCI

International Standards for Neurological Classification after Spinal Cord Injury.

(La⁻)_b

Blood lactate concentration is indicative of lactate (lactic acid) produced within resting and working skeletal muscle.

PO

Power output is the amount of work performed per unit time in Watts (W, J·s⁻¹).

PO_peak

Peak power output is the amount of work at peak performance.

V_E

Pulmonary ventilation is the rate at which gas enters or leaves the lungs. It is defined as tidal volume × respiratory rate (L·min⁻¹).

RER

Respiratory exchange ratio is the ratio of carbon dioxide production to oxygen consumption (VCO₂/VO₂) measured across the mouth. In the steady state, when body CO₂ and O₂ stores are unchanging, this ratio serves as an indicator of gas exchange at the tissue level (ie, respiratory quotient, RQ) and reflects substrate utilization.

RPE

Rating of Perceived Exertion (Borg) reflects the subjective sensation of exertion and corresponds well to exercise capacity.

SCI

Spinal cord injury.

VO_2peak

Peak oxygen uptake is the highest oxygen uptake elicited during an exercise test to exhaustion in the absence of an oxygen uptake plateau. It is expressed either as an absolute rate in liters of oxygen per minute (L/min⁻¹) or as a relative rate in milliliters of oxygen per kilogram of bodyweight per minute (mL·kg⁻¹·min⁻¹).