Lower Thoracic Spinal Cord Stimulation to Restore Cough in Patients With Spinal Cord Injury: Results of a National Institutes of Health–Sponsored Clinical Trial. Part I: Methodology and Effectiveness of Expiratory Muscle Activation

Electrical stimulation system.

Effects of lower thoracic SCS for 1 subject on airflow and airway pressure generation during stimulation at the T9, T11, and L1 spinal levels alone and in combinations at TLC (A) and at FRC (B). Large airflow rates and airway pressures were generated during single site stimulation. These parameters were greater with combined stimulation at any 2 sites. Combined stimulation at all 3 sites did not result in further increases in these parameters.

Mean peak airflow rates (upper panel) and mean airway pressures (lower panel) during SCS at the T9, T11, and L1 spinal levels alone and in combinations at TLC (solid bars) and at FRC (dotted bars). Mean spontaneous peak expiratory flow rates and airway pressure are shown for comparison (empty bars). Large peak airflow rates and airway pressures of similar magnitude were generated during SCS during single site stimulation. Combined stimulation of 2 sites, however, resulted in significantly greater peak airflow rates and airway pressures (P<.05 for each). There were no significant differences in either peak airflow rates or airway pressure generation between any 2 sites. Combined stimulation of 3 sites did not result in further increases in these parameters.

Relationship between stimulus frequency and mean airway pressure generation (expressed as a percent maximum) during single site SCS and combined stimulation of 2 sites at FRC and at TLC. There were progressive increases in airway pressure generation with increases in stimulus frequency. There was a plateau between 40 and 50Hz, because there were only small changes in pressure generation between these stimulus frequencies. There were no significant differences between responses at TLC and FRC.

Relationship between stimulus amplitude and mean airway pressure generation (expressed as a percent maximum) during single site SCS and combined stimulation of 2 sites at FRC and at TLC. There were progressive increases in airway pressure generation with increasing stimulus amplitude. With 2-site stimulation, a plateau developed between 30 and 40V, because there were no significant differences in pressure generation between these amplitude levels (P>.05). There were no significant differences between responses at TLC and FRC.

Relationship between pulse width and mean airway pressure generation (expressed as a percent maximum) during combined stimulation at 2 sites at TLC. There was a significant increase in pressure generation between 100 and 150 μs (P<.05). However, there were no further increases in pressure generation with increasing pulse duration as high as 250μs.

Relationship between airway pressure and peak airflow generation for each subject at FRC and at TLC. There was a highly significant linear relationship between these parameters (P<.01). By this relationship, peak airflow rates could be predicted based on the magnitude of airway pressure generation.

Mean changes in airway pressure (expressed as a percent maximum) with 2-site SCS applied every 1 minute over a 30-minute period. There were no significant decrements in airway pressure generation over this period, indicating no evidence of system fatigue during the chronic application of SCS.