The purpose of this study was to demonstrate a neuromechanical inhibitory effect on respiratory muscle activity during mechanical ventilation and to determine whether upper and lower airway receptors provide this inhibitory feedback. Several protocols were completed during mechanical ventilation: (1) positive and negative pressure changes in the upper airway, (2) airway anesthesia to examine the consequences of receptor blockade on respiratory muscle activity, (3) increasing FRC with positive end-expiratory pressure to study the effect of hyperinflation or stretch on respiratory muscle activity, and (4) use of heart-lung transplant patients to determine the effects of vagal denervation on respiratory muscle activity. All subjects were mechanically hyperventilated with positive pressure until inspiratory muscle activity was undetectable and the end-tidal PCO2 decreased to less than 30 mm Hg. End-tidal PCO2 (PETCO2) was increased by either adding CO2 to the inspired gas or decreasing tidal volume (50 ml/min). The PETCO2 where a change in inspiratory muscle activity occurred was taken as the recruitment threshold (PCO2RT). Neuromechanical feedback caused significant inspiratory muscle inhibition during mechanical ventilation, as evidenced by the difference between PCO2RT and PETCO2 during spontaneous eupnea (45 +/- 4 versus 39 +/- 4 mm Hg) and a lower PCO2RT when tidal volume was reduced with a constant frequency and fraction of inspired CO2. Recruitment threshold was unchanged during positive and negative pressure ventilation, during upper and lower airway anesthesia, and in vagally denervated lung transplant patients. These findings demonstrate that neuromechanical feedback causes highly significant inhibition of inspiratory muscle activity during mechanical ventilation; upper and lower airway receptors do not appear to mediate this effect.