Respiratory dysfunction and management in spinal cord injury

Abstract

Respiratory dysfunction is a major cause of morbidity and mortality in spinal cord injury (SCI), which causes impairment of respiratory muscles, reduced vital capacity, ineffective cough, reduction in lung and chest wall compliance, and excess oxygen cost of breathing due to distortion of the respiratory system. Severely affected individuals may require assisted ventilation, which can cause problems with speech production. Appropriate candidates can sometimes be liberated from mechanical ventilation by phrenic-nerve pacing and pacing of the external intercostal muscles. Partial recovery of respiratory-muscle performance occurs spontaneously. The eventual vital capacity depends on the extent of spontaneous recovery, years since injury, smoking, a history of chest injury or surgery, and maximum inspiratory pressure. Also, respiratory-muscle training and abdominal binders improve performance of the respiratory muscles. For patients on long-term ventilation, speech production is difficult. Often, practitioners are reluctant to deflate the tracheostomy tube cuff to allow speech production. Yet cuff-deflation can be done safely. Standard ventilator settings produce poor speech quality. Recent studies demonstrated vast improvement with long inspiratory time and positive end-expiratory pressure. Abdominal binders improve speech quality in patients with phrenic-nerve pacers. Recent data show that the level and completeness of injury and older age at the time of injury may not be related directly to mortality in SCI, which suggests that the care of SCI has improved. The data indicate that independent predictors of all-cause mortality include diabetes mellitus, heart disease, cigarette smoking, and percent-of-predicted forced expiratory volume in the first second. An important clinical problem in SCI is weak cough, which causes retention of secretions during infections. Methods for secretion clearance include chest physical therapy, spontaneous cough, suctioning, cough assistance by forced compression of the abdomen ("quad cough"), and mechanical insufflation-exsufflation. Recently described but not yet available for general use is activation of the abdominal muscles via an epidural electrode placed at spinal cord level T9-L1.

Fig. 1

Data obtained during tidal breathing (solid lines) and during relaxation (dashed lines) from a representative subject with abdominal cuff deflated versus inflated. The vertical axis represents the cross-sectional area of the (upper or lower) rib cage. The horizontal axis represents the change in lung volume. The vertical lines were obtained when, at or near end-tidal inspiration, the subject relaxed against an occluded valve. With the cuff deflated, the upper-rib-cage area was less at end-inspiration than at functional residual capacity. With the cuff inflated, the lower-rib-cage area increased more per unit change in lung volume than it did with the cuff deflated. This is indicated by the increase in slope of the plot of volume versus lower-rib-cage area. Also, as a result of cuff inflation, upper-rib-cage area increased during tidal inspiration, whereas it had decreased with the cuff deflated. (From Reference , with permission.)

Fig. 2

Phonation relative to pressure and time. The dashed line represents the minimum pressure required for phonation. See text for details. (From Reference , with permission.)

Fig. 3

Changes (from usual) in speaking rate with lengthened inspiratory time (T_I), positive end-expiratory pressure (PEEP), and lengthened T_I plus PEEP. (From Reference , with permission.)

Fig. 4

Tracheal pressure during speaking with a one-way valve (top panel) versus with positive end-expiratory pressure (PEEP) set to 15 cm H₂O (bottom panel). Note that the tracings are almost indistinguishable. See text for details. (From Reference , with permission.)

Fig. 5

Comparison of transdiaphragmatic pressure generated during phrenic-nerve stimulation (PS) and intramuscular diaphragm stimulation (DS) in anesthetized dogs. Supramaximal amplitudes were applied over a 10–40 Hz range during airway occlusion. Each number represents results from a separate animal. (From Reference , with permission.)

Fig. 6

Superimposed maximum changes in inspired volume from separate left, separate right, and bilateral diaphragm stimulation, over the course of the reconditioning period of one subject. Supramaximal stimulus variables (24 mA or 25 mA, 50 Hz) were employed. (From Reference , with permission.)

Fig. 7

Effects of electrical stimulation applied at different spinal cord regions before phrenicotomy (circles) and after phrenicotomy (dots) in a dog anesthetized with pentobarbital. Inspired volume is expressed as a percentage of maximum. Inspired-volume generation was greatest in the vicinity of the T2-T3 spinal-cord region and decreased progressively at spinal levels cephalad and caudad to that region. Stimulation was applied with supramaximal stimulus amplitude and stimulus frequency of 50 Hz. (From Reference , with permission.)

Fig. 8

Maximum inspired volumes with intercostal stimulation at various times during the reconditioning period of one patient. As a result of gradually increasing the pacing duration, there were progressive increments in inspired volume. Because inspiratory time was fixed, increases in inspired volume were achieved by increases in inspiratory flow. The improvement in inspired volume is attributable to the training effect of long-term stimulation and reversal of intercostal muscle atrophy. (From Reference , with permission.)