Acute Ventilatory Support During Whole-Body Hybrid Rowing in Patients With High-Level Spinal Cord Injury: A Randomized Controlled Crossover Trial

Abstract

Background: High-level spinal cord injury (SCI) results in profound spinal and supraspinal deficits, leading to substantial ventilatory limitations during whole-body hybrid functional electrical stimulation (FES)-rowing, a form of exercise that markedly increases the active muscle mass via electrically induced leg contractions. This study tested the effect of noninvasive ventilation (NIV) on ventilatory and aerobic capacities in SCI.

Methods: This blinded, randomized crossover study enrolled 19 patients with SCI (level of injury ranging from C4 to T8). All patients were familiar with FES-rowing and had plateaued in their training-related increases in aerobic capacity. Patients performed two FES-rowing peak exercise tests with NIV or without NIV (sham).

Results: NIV increased exercise tidal volume (peak, 1.50 ± 0.31 L vs 1.36 ± 0.34 L; P < .05) and reduced breathing frequency (peak, 35 ± 7 beats/min vs 38 ± 6 beats/min; P < .05) compared with the sham test, leading to no change in alveolar ventilation but a trend toward increased oxygen uptake efficiency (P = .06). In those who reached peak oxygen consumption (Vo₂peak) criteria (n = 13), NIV failed to significantly increase Vo₂peak (1.73 ± 0.66 L/min vs 1.78 ± 0.59 L/min); however, the range of responses revealed a correlation between changes in peak alveolar ventilation and Vo₂peak (r = 0.89; P < .05). Furthermore, those with higher level injuries and shorter time since injury exhibited the greatest increases in Vo₂peak.

Conclusions: Acute NIV can successfully improve ventilatory efficiency during FES exercise in SCI but may not improve Vo₂peak in all patients. Those who benefit most seem to be patients with cervical SCI within a shorter time since injury.

Trial registry: ClinicalTrials.gov; Nos.: NCT02865343 and NCT03267212; URL: www.clinicaltrials.gov.

Keywords: exercise testing; noninvasive ventilation; pulmonary rehabilitation; respiratory pattern; spinal cord injury.

Figure 1

Consort diagram study flowchart. Vo₂ = oxygen consumption.

Figure 2

Respiratory pattern during maximal incremental functional electrical stimulation-rowing exercise test with NIV and without NIV (sham). Breathing frequency was significantly reduced over time during exercise with NIV compared to sham, whereas tidal volume was significantly increased over time. *Between-group comparison over time, P < .05. Values are mean ± SD (N = 19). NIV = noninvasive ventilation.

Figure 3

Individual and mean ± SD values of V_Apeak and Vo₂peak during maximal incremental functional electrical stimulation-rowing test in the subgroup of patients who reached the criteria of maximality for Vo₂peak (n = 13). V_Apeak = peak alveolar ventilation; Vo₂peak = peak oxygen consumption. See Figure 1 legend for expansion of other abbreviation.

Figure 4

Correlation between the change in V_Apeak and Vo₂peak between NIV and sham condition in the subgroup of patients who reached the criteria of maximality for Vo₂peak (n = 13). See Figure 2 and 3 legends for expansion of abbreviations.

Figure 5

Individual and mean ± SD values of V_Apeak and OUES during maximal incremental functional electrical stimulation-rowing test in the whole group of patients (N = 19). OUES = oxygen uptake efficiency slope. See Figure 1 and 2 legends for expansion of other abbreviations.

Figure 6

Comparison of change in OUES depending on their response to NIV in V_A. Change in V_Apeak as a discriminant factor for change in OUES. Those in whom NIV increased V_A (change in V_A > 0, n = 12) exhibited an improvement in OUES, whereas those in whom NIV did not increase V_A (change in V_A < 0, n = 7) exhibited a reduction in OUES (P < .05). Box plots presenting the median, 10th, 25th, 75th and 90th percentiles with error bars (n = 19). V_A = alveolar ventilation. See Figure 1, 2, and 3 legends for expansion of other abbreviations.

Figure 7

Comparison of change in OUES depending on TSI (≤ 6 years or > 6 years). Those with TSI ≤ 6 years increased OUES with NIV significantly more than in individuals with TSI > 6 years (P < .05). Box plots presenting the median, 10th, 25th, 75th and 90th percentiles with error bars (N = 19). TSI = time since injury.

Figure 8

Main reason for stopping functional electrical stimulation-rowing incremental exercise test while using sham or NIV on the subgroup of patients who reached Vo₂peak (n = 13). “Arm fatigue” was the most reported reason for stopping exercise in the sham condition, while “exhaustion” was the most reported one in the NIV condition (P = .10). See Figure 1 and 3 legends for expansion of abbreviations.