APNEA FOLLOWING NORMOCAPNIC MECHANICAL VENTILATION IN AWAKE MAMMALS -A DEMONSTRATION OF CONTROL-SYSTEM INERTIA

1. Inhibition of inspiratory muscle activity from volume-related feedb ack during mechanical ventilation has been shown previously. To determ ine if this neuro-mechanical inhibition displays a memory effect, the duration of expiration immediately following cessation of mechanical v entilation was assessed in eight normal subjects. The subjects were pa ssively mechanically ventilated via a nasal mask until the end-tidal C O2 (P(ET,CO2)) was a minimum of 30 mmHg and inspiratory effort was no longer detected, as evidenced by stabilization of mouth pressure and d isappearance of surface diaphragm EMG activity. The ventilator output was held constant at a mean tidal volume (V(T)) of 1.0 l and breath du ration of 4.6 s and P(ET,CO2) was increased 1-1.5 mmHg/min (via increa sed inspired CO2 fraction, F(I,CO2)) until inspiratory muscle activity returned. The P(ET,CO2) at which activation first occurred was define d as the CO2 recruitment threshold (P(CO2,RT)). The mechanical ventila tion protocol was repeated and the P(ET,CO2) increased 1-1.5 mmHg/min until it was a mean of 1.1 mmHg above spontaneous P(ET,CO2) and 3.6 mm Hg below P(CO2,RT). After 4.6 min of mildly hypercapnic mechanical ven tilation, the mechanical ventilation was terminated. 2. Following term ination of mechanical ventilation, the duration of the subsequent apno ea was 14.6 +/- 2.8 s (mean +/- S.E.M.) or 453 +/- 123 % > spontaneous T(E) and 178 +/- 62% > the T(E) chosen by the subject during 'assist control' ventilation at V(T) = 1.0 l. 3. To test the hypothesis that t he apnoea following cessation of mechanical ventilation was due to a v agally mediated memory effect, the study was repeated in five double-l ung transplant patients with similar P(CO2,RT) to normal subjects. The se pulmonary vagally denervated patients also displayed an apnoea (14. 5 +/- 4.0 s) upon cessation of mechanical ventilation (at a P(ET,CO2) 2.0 mmHg > eupnoea and 2.4 mmHg < P(CO2,RT)), that was 367 +/- 162 % > spontaneous T(E). 4. We also found significant apnoea in the awake do g immediately following mildly hypercapnic passive mechanical ventilat ion, and this was similar before and after bilateral vagal blockade (1 5.7 +/- 1.3 and 19.7 +/- 4.7 s, respectively). 5. We conclude that neu romechanical inhibition of inspiratory muscle activity, produced by pa ssive mechanical ventilation at high V(T), exhibits a memory effect re flected in T(E) prolongation, which persists in the face of substantia l increases in chemoreceptor stimuli. This effect is not dependent on vagal feedback from lung receptors. 6. We hypothesize that this persis tent apnoea represents an inherent 'inertia', characteristic of the ve ntilatory control system. This inertia contributes to the prolongation of apnoea, independently of the specific mechanism which initiated th e apnoea and may explain why apnoeas are commonly terminated at higher P(a,CO2)s than those at initiation.