Augmenting exercise capacity with noninvasive ventilation in high-level spinal cord injury

Abstract

High-level spinal cord injury (SCI) results in a very limited innervated skeletal muscle mass that strongly reduces exercise capacity. Our recent work showed that when adding functional electrical stimulation (FES) of the paralyzed legs (hybrid FES-exercise) to produce higher exercise capacity, peak ventilation became a limiting factor to training-induced improvement in aerobic capacity. Our assumption was that the systemic adaptations to exercise training are delimited by the maximal ventilation that can be achieved. However, herein, we present a case showing an acute increase in aerobic capacity when using noninvasive ventilatory support (NIV) during FES-rowing test in an individual who had previously experimented a plateau in his aerobic capacity for 18 mo. An 18-yr-old man with C5 SCI trained with arms-only rowing for 6 mo and subsequently trained with hybrid FES-rowing for 18 mo. Peak minute ventilation (V̇e_peak) and peak oxygen consumption (V̇o_2peak) were increased after arms-only training and increased further with 6 mo of hybrid FES-row training. Despite continued intense and frequent, hybrid FES-row training, neither V̇e_peak nor V̇o_2peak increased further over the next year (1.94 and 66.0 l/min). However, when this individual performed a FES-rowing V̇o_2peak test with the addition of NIV, V̇e_peak increased by 5 l/min, resulting in an improved V̇o_2peak (2.23 l/min, +12%). This case demonstrates that noninvasive ventilation can overcome limitations to ventilation in high-level SCI and improve aerobic capacity during hybrid FES-exercise to a level not otherwise achievable. In addition, it broadly illustrates the intimate role of pulmonary function in determining the capacity to perform exercise.

INTRODUCTION

Spinal cord injuries (SCI) occur in ~500,000 individuals yearly with ~60% at high levels. Motor, sensory, and autonomic deficits are markedly more significant with cervical and high thoracic (above T3) than low-level injuries (10). One consequence is a huge reduction in exercise capacity in this subpopulation. Indeed, aerobic capacity decreases by 5% with each level of injury from T11 to C4 such that high-level injuries result in aerobic capacities <40% of able-bodied persons (15). Hence, these individuals cannot achieve intensities required to reduce cardiometabolic risk (14). Interestingly, our (8) work employing functional electrical stimulation (FES) of the paralyzed legs with arm exercise to produce higher exercise capacity (hybrid FES-exercise) revealed that peak ventilation may be a key limitation to training-induced adaptations in those with high-level injuries. Indeed, in comparison with those with low-level injuries, despite regular training, those with high-level SCI have lesser increases in aerobic capacity after training with only modest increases in ventilatory capacity (8). Moreover, we have anecdotally noted that after initial increases in aerobic capacity with 6 mo of training with hybrid FES-exercise (rowing), those with high-level injuries demonstrate no further increases despite continued, high-effort training, unlike those with low-level injuries (8). This suggests that the profound pulmonary muscle denervation in these patients (11) restricts the aerobic adaptations to training with hybrid FES-exercise.

Recently, we began a study to build on this observation and on prior work in obese individuals treated for obstructive sleep apnea with respiratory restriction. We reasoned that respiratory restriction to exercise adaptations in those with high-level SCI may be similar to that demonstrated in those with high body mass index (6, 17). Indeed, in both, reduced functional residual capacity and lesser chest compliance induce rapid and shallow breathing during exercise (7, 9), limiting inspiratory capacity. Our prior work in patients with obesity-related restrictive pattern of breathing during exercise showed greater improvements in aerobic capacity when noninvasive ventilatory support (NIV) is used during training (16), which were likely to be related to a reduction in the work of breathing (3). Given that the respiratory response to exercise may be similarly reduced in those with high-level SCI (1), we began a trial of NIV during hybrid FES-exercise training. We hypothesized that NIV-induced increase in exercise tidal volume would allow for further training-induced improvements in aerobic capacity. However, we did not hypothesize that acutely increasing exercise tidal volume via NIV would impact aerobic capacity after prior hybrid FES-exercise training. Our assumption was that the systemic adaptations to exercise training would have been delimited by maximal ventilation and that acutely increasing peak ventilation would have no effect on aerobic capacity. However, herein, we present a case showing a marked acute increase in aerobic capacity via NIV in an individual with high-level SCI who had trained vigorously with hybrid FES-exercise over the prior 18 mo. This exemplifies the important determinant role of peak ventilation for aerobic capacity evidenced during this form of exercise in those with high-level SCI.

PRESENT CASE

The study was approved by the Institutional Review Board at Spaulding Rehabilitation Hospital, and all subjects gave written, informed consent. An 18-yr-old man with no significant past medical history sustained a C5 compression fracture after a classmate landed on his head while playing in a pool. Three months later, on discharge from rehabilitation, his injury was classified as C8 American Spinal Injury Association (ASIA) Motor Incomplete (C). Forced vital capacity and forced expiratory volume in 1 s were reduced (4.01 liters, 75% predicted and 3.80 liters, 82% predicted) and increased slightly over the following 24 mo (4.37 liters, 79% predicted and 4.23 liters, 89% predicted).

Six months after injury, the patient enrolled in an exercise training study (NCT02139436) and was randomized to 6 mo of arms-only rowing exercise. Training sessions were prescribed as three per week at 70–85% of peak capacity for 30–40 min. The patient was highly compliant, averaging seven sessions per month at intensities >80% of peak. Peak exercise tests were performed before and after training. Baseline peak oxygen consumption (V̇o_2peak) was low (1.18 l/min or 16.7 ml·kg⁻¹·min⁻¹) with a peak minute ventilation (V̇e_peak) of 39.6 l/min. Six months of arms-only exercise increased V̇o_2peak by 11% to 1.31 l/min (17.8 ml·kg⁻¹·min⁻¹) accompanied by a V̇e_peak increase of 28% to 50.9 l/min (Fig. 1).

Fig. 1.

Peak aerobic capacity (V̇o_2peak) and peak minute ventilation (V̇e_peak) across 24 mo of training. Reported data represent the 30-s average of the peak values for each test (each time point representing a single test), expressed as a percentage of improvement from baseline (Jan 2015). Testing was initially with arms-only exercise, before and after training (medium gray bars), followed by testing with hybrid functional electrical stimulation (FES)-row exercise, before and after 6, 12, and 18 mo of training (black bars), and, finally, testing with hybrid FES-row exercise and sham or noninvasive ventilatory support (NIV; dark gray bars; no training). The cumulative percentage of increase from baseline in aerobic capacity was 11% after arms-only training, +52% after hybrid FES-row exercise was introduced, +64% after hybrid FES-row training, and then +89% after the introduction of NIV in addition to hybrid FES-row exercise. Aug, August; Feb, February; Jan, January; Jun, June.

Subsequently, the patient was crossed over to hybrid FES-exercise (rowing). FES consisted on a four-channel electrical stimulator (Odstock, Salisbury, United Kingdom) that activates the quadriceps and hamstrings in drive and recovery phases of the rowing cycle, respectively (pulse width of 450 μs, frequency of 40 Hz, maximal intensity of 51 mA, no ramp to allow an instant contraction of the muscle, and no fixed duration to allow the control of the stimulation timing via a button on the rowing handle; Ref. 13). A peak exercise test of hybrid FES-rowing was performed (13), followed by 6 mo of training at the same frequency, intensity, and duration as for arms-only training. The patient was highly compliant. At the end of the 6-mo training period, another peak FES-row test was performed. The introduction of FES to the rowing exercise increased V̇o_2peak by 37% to 1.79 l/min (23.9 ml·kg⁻¹·min⁻¹) and V̇e_peak by 27% to 64.7 l/min (Fig. 1). Six months of FES-row training resulted in a V̇o_2peak of 1.94 l/min (24.8 ml·kg⁻¹·min⁻¹), representing an 8% increase from before training and a cumulative increase of 64% from baseline. In contrast, there were no further increases in V̇e_peak (66.0 l/min).

After completing the study, the patient enrolled in the Spaulding Exercise for Persons with Disabilities (ExPD) program and continued FES-row training for 1 yr with peak exercise tests every 6 mo. The ExPD program functions like a community gym, allowing voluntary participation in FES-rowing for up to five sessions weekly. However, despite continued FES-row training over the next 12 mo, neither V̇o_2peak nor V̇e_peak increased further (Fig. 1).

At this time (i.e., after 18 mo of FES-row training), the patient enrolled in our study employing NIV during FES-rowing in high-level SCI (NCT02865343). Before training began, two peak exercise tests were performed within a period of 48 h, one with NIV (inspiratory positive airway pressure = 23 cmH₂O, expiratory positive airway pressure = 3 cmH₂O; Astral 150; ResMed) and one with sham NIV (inspiratory positive airway pressure = 5 cmH₂O, expiratory positive airway pressure = 3 cmH₂O). The addition of NIV during the FES-row exercise test increased peak tidal volume by 0.5 liter per breath up to 2.3 liters per breath compared with sham NIV (1.8 liters per breath) without alterations in breathing frequency (34.7 vs. 35.5 breaths per minute). As a result, V̇e_peak was increased by 5 l/min, resulting in a 12% increase in V̇o_2peak (2.23 l/min; Fig. 1). This acute increase in V̇o_2peak was as large as prior increases achieved after 6 mo of either arms-only or FES-exercise training (Fig. 1). The higher V̇o_2peak occurred at a similar workload (69 W). However, V̇o₂ plateaued during the sham test (increased workload without increased V̇o₂), suggesting that the test with NIV may not have achieved the true peak workload and possibly V̇o_2peak despite all other criteria being met.

DISCUSSION

This case suggests limitations to ventilation in high-level SCI can constrain aerobic capacity. In this patient, long-term (2 yr), intense (>80% maximum heart rate), frequent (>2 times weekly) exercise training led to increased aerobic capacity that exceeded ventilatory capacity. Only after the addition of NIV to increase peak minute ventilation was the magnitude of improvement in V̇o_2peak fully realized. We were not surprised to find that the addition of FES increased V̇o_2peak and V̇e_peak given the larger muscle mass engaged (13). However, we were surprised to find a marked improvement in V̇o_2peak with the acute addition of NIV (+12%). This may be analogous to training-related adaptations in highly fit, able-bodied individuals. Johnson et al. (5) showed that trained athletes can approach mechanical limits for respiration, contributing to an arterial hypoxemia and limiting maximal aerobic capacity. That is, in highly fit athletes, peripheral muscle and central cardiac capacities exceed pulmonary capacity. In the present case, skeletal muscle and cardiac adaptations with FES-exercise training seem to have outstripped ventilatory capacity, but this was only evidenced with application of NIV (2). Alternatively, NIV may have unloaded accessory respiratory muscles and reduced diaphragmatic work (4). This would have reduced the potential impact of the respiratory metaboreflex. Indeed, by reducing the work of breathing, a lower amount of blood flow would be required for the respiratory muscles, and, therefore, a greater blood flow would be available to the active muscles. Hence, the oxygen cost of ventilation, which can potentially “steal” flow from locomotor muscles, may have been reduced, leading to lesser fatigue and improved performance (4). Although the mechanisms require elucidation and more studies are needed to confirm this observation, this case showed that increased aerobic capacity with whole body exercise training can be restricted by ventilatory capacity in those with SCI and that ventilatory support was required to realize the true increase in exercise capacity. This broadly illustrates the intimate role of pulmonary function in determining the capacity to perform exercise.

GRANTS

Support for this study was provided by NIH Grants R01-HL-117037 and R21-HD-088891 and Fondation pour la Recherche Médicale, France.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.A.T. and I.V. conceived and designed research; E.F. and I.V. performed experiments; J.W.M. and I.V. analyzed data; J.W.M., J.A.T., and I.V. interpreted results of experiments; J.W.M. and I.V. prepared figures; J.W.M. drafted manuscript; J.W.M., E.F., J.A.T., and I.V. edited and revised manuscript; J.W.M., E.F., J.A.T., and I.V. approved final version of manuscript.