The role of spinal cord transmission in the ventilatory response to electrically induced exercise in the anaesthetized dog.

1. The ventilatory response to electrically induced ;exercise' was studied in six chloralose-anaesthetized dogs. The on-transient and steady-state responses to ;exercise' were compared in the same dogs before and after spinal cord transection at T8/9 (dermatome level T6/7) on fifteen occasions.2. Phasic hind limb ;exercise' was induced for periods of 4 min by passing current (2 Hz modulated 50 Hz sine wave) between two needles inserted through the hamstring muscles. The maximum current used was 30 mA. This was below the level previously found to produce an artifactual stimulation of breathing with the cord intact.3. Cord transection produced no significant change in either the resting values of ventilation ( V(I)) and CO(2) production ( V(CO) (2)) or the ventilatory equivalent for CO(2) during ;exercise' ( big up tri, open V(I)/ big up tri, open V(CO) (2)).4. During the steady state of exercise P(a, CO) (2) was on average significantly lower than at rest with the cord intact (mean big up tri, openP(a, CO) (2), - 2.1 mmHg; range - 5.7 to + 1), and higher, though not significantly, with the cord cut (mean P(a, CO) (2), + 1.2 mmHg; range - 1.5 to + 4.3). However, even in the absence of spinal cord transmission, the ventilatory response to exercise could not be accounted for on the basis of CO(2) sensitivity; the big up tri, open V(I)/ big up tri, openP(a,CO) (2) obtained with exercise (apparent sensitivity) was significantly greater than that obtained with CO(2) inhalation (true sensitivity) both before and after cord section.5. V(I) and V(CO) (2) increased more slowly with the cord cut than with the cord intact. This was thought to be due to a slower increase in venous return in the absence of sympathetic innervation of the lower half of the body following cord transection.6. Similar experiments were performed during muscle paralysis (following gallamine triethiodide). Ventilation was maintained with a respirator controlled by phrenic nerve activity. These experiments showed an increase in ventilation, independent of muscle contraction, which was only present when the cord was intact and which was confined to the on-transient. Only in the absence of spinal cord transmission could there be certainty that the dynamics of the ventilatory response to electrically induced ;exercise' was free of artifact.7. It was concluded that spinal cord transmission is not necessary for the steady-state ventilatory response to electrically induced exercise of the hind limbs.8. The dog with spinal cord transection provides a suitable model for the study of the chemical control of breathing during electrically induced exercise.