Differential Activation of Respiratory Muscles by CO2 and Hypoxia in Man

Abstract

Sears et al (Nature 1982; 299:728–30) have demonstrated an initial activation of expiratory neurons in response to rising PaCO2 during hypocapnic apnea in anesthetized cats. In contrast, hypoxia gave rise initially to inspiratory activity. These findings suggested a pattern of neuronal activation fundamentally different in response to CO2 than in response to hypoxia. We therefore hypothesized that CO2 would cause a relative expiratory bias, while hypoxia would cause relative inspiratory bias in normal human subjects. To this end, the electromyographic activity of the diaphragm (EMGdi) and the abdominal expiratory muscles (EMGab) was recorded during progressive hypercapnia (HCVR) and hypoxia (HVR) in 5 normal males. The slopes of the regression lines relating log-EMGdi activity to minute volume of ventilation ( V ˙ E ) were steeper during HVR runs than during HCVR runs (mean ± SE, 0.03201 ± 0.00724 vs 0.02729 ± 0.00676, p<0.03). Phasic expiratory EMGab activity was seen in 15 of 15 HCVR runs but in only 6 of 15 HVR runs. The maximum level of VE attained prior to the onset of EMGab activity was significantly lower during HCVR runs than during HVR runs (23.1 ± 2.5 vs 34.8 ± 4.0 L/min, p<0.003). We conclude that in normal humans: (1) the diaphragm contributes to VE to a greater extent in response to hypoxia than to CO2, and (2) CO2 causes a more consistent recruitment at lower V ˙ E of abdominal expiratory activity than does hypoxia. Our findings are consistent with those of Sears et al in showing a differing pattern of respiratory neuromuscular activation in response to CO2 than to hypoxia.