SUMMARY1. We determined the incidence of exercise-induced arterial hypoxaemia and its determinants in sixteen highly trained, healthy runners who were capable of achieving and sustaining very high metabolic rates (maximal 02 uptake = 72 + 2 ml kg-' min-or 4-81 + 0-131 min'). Arterial blood gases and acidbase status were determined at each load of a progressive short-term exercise test and repeatedly during constant-load treadmill running while breathing air and during inhalation of mildly hypoxic, hyperoxic, and helium-enriched gases.2. 5. In view of the correction of hypoxaemia with mild hyperoxia and the very high ratios of alveolar ventilation to pulmonary blood flow (PA/C = 4-6) achieved during heavy exercise, we think it unlikely that non-uniformity of the VA/QC distribution or veno-arterial shunt could explain the hypoxaemia observed in most of our subjects. By exclusion, we suggest that hypoxaemia may be attributed to a diffusion limitation secondary to very short red cell transit times in at least a portion of the pulmonary circulation. These excessively short transit times may occur at high metabolic rates 6 PHY 355 J. A. DEMPSEY, P. G. HANSON AND K. S. HENDERSON if pulmonary capillary blood volume has achieved its maximum expansion at a time when pulmonary blood flow continues to increase.6. Considerable individual variations were present in the hyperventilatory response to heavy work. These variations were inconsistently related to corresponding levels of metabolic acidosis and in many cases a compensatory hyperventilation was minimal or absent in the face of substantial acidosis and hypoxaemia. The findings that tidal breaths during heavy exercise often exceeded the maximum flow-volume curve and that hyperventilation was consistently obtained with helium inhalation, suggest that the hyperventilatory response to heavy work is determined to a significant extent by the mechanical load imposed on the chest wall secondary to increased pulmonary impedance.

We assessed the contribution of carotid body chemoreceptors to the ventilatory response to specific CNS hypercapnia in eight unanaesthetized, awake dogs. We denervated one carotid body (CB) and used extracorporeal blood perfusion of the reversibly isolated remaining CB to maintain normal CB blood gases (normoxic, normocapnic perfusate), to inhibit (hyperoxic, hypocapnic perfusate) or to stimulate (hypoxic, normocapnic perfusate) the CB chemoreflex, while the systemic circulation, and therefore the CNS and central chemoreceptors, were exposed consecutively to four progressive levels of systemic arterial hypercapnia via increased fractional inspired CO 2 for 7 min at each level. Neither unilateral CB denervation nor CB perfusion, per se, affected breathing. Relative to CB control conditions (normoxic, normocapnic perfusion), we found that CB chemoreflex inhibition decreased the slope of the ventilatory response to CNS hypercapnia in all dogs to an average of 19% of control values (range 0-38%; n = 6), whereas CB chemoreflex stimulation increased the slope of the ventilatory response to CNS hypercapnia in all dogs to an average of 223% of control values (range 204-235%; n = 4). We conclude that the gain of the CNS CO 2 /H + chemoreceptors in dogs is critically dependent on CB afferent activity and that CNS-CB interaction results in hyperadditive ventilatory responses to central hypercapnia.

We assessed the speed of the ventilatory response to square-wave changes in alveolar P(CO2) and the relative gains of the steady-state ventilatory response to CO2 of the central chemoreceptors vs. the carotid body chemoreceptors in intact, unanesthetized dogs. We used extracorporeal perfusion of the reversibly isolated carotid sinus to maintain normal tonic activity of the carotid body chemoreceptor while preventing it from sensing systemic changes in CO2, thereby allowing us to determine the response of the central chemoreceptors alone. We found the following. 1) The ventilatory response of the central chemoreceptors alone is 11.2 (SD = 3.6) s slower than when carotid bodies are allowed to sense CO2 changes. 2) On average, the central chemoreceptors contribute approximately 63% of the gain to steady-state increases in CO2. There was wide dog-to-dog variability in the relative contributions of central vs. carotid body chemoreceptors; the central exceeded the carotid body gain in four of six dogs, but in two dogs carotid body gain exceeded central CO2 gain. If humans respond similarly to dogs, we propose that the slower response of the central chemoreceptors vs. the carotid chemoreceptors prevents the central chemoreceptors from contributing significantly to ventilatory responses to rapid, transient changes in arterial P(CO2) such as those after periods of hypoventilation or hyperventilation ("ventilatory undershoots or overshoots") observed during sleep-disordered breathing. However, the greater average responsiveness of the central chemoreceptors to brain hypercapnia in the steady-state suggests that these receptors may contribute significantly to ventilatory overshoots once unstable/periodic breathing is fully established.

In awake dogs, lactic acid was injected into the phrenic and deep circumflex iliac arteries to elicit the diaphragm and abdominal muscle metaboreflexes, respectively. At rest, injections into the phrenic or deep circumflex iliac arteries significantly increased mean arterial blood pressure 21 +/- 7% and reduced cardiac output 6 +/- 2% and blood flow to the hindlimbs 20 +/- 9%. Simultaneously, total systemic, hindlimb, and abdominal expiratory muscle vascular conductances were reduced. These cardiovascular responses were not accompanied by significant changes in the amplitude or timing of the diaphragm electromyogram. During treadmill exercise that increased cardiac output, hindlimb blood flow, and vascular conductance 159 +/- 106, 276 +/- 309, and 299 +/- 90% above resting values, lactic acid injected into the phrenic or deep circumflex iliac arteries also elicited pressor responses and reduced hindlimb blood flow and vascular conductance. Adrenergic receptor blockade at rest eliminated the cardiovascular effects of the respiratory muscle metaboreflex. We conclude that the cardiovascular effects of respiratory muscle metaboreflex activation are similar to those previously reported for limb muscles. When activated via metabolite production, the respiratory muscle metaboreflex may contribute to the increased sympathetic tone and redistribution of blood flow during exercise.

We assessed the time course of changes in eupneic arterial PCO(2) (Pa(CO(2))) and the ventilatory response to hyperoxic rebreathing after removal of the carotid bodies (CBX) in awake female dogs. Elimination of the ventilatory response to bolus intravenous injections of NaCN was used to confirm CBX status on each day of data collection. Relative to eupneic control (Pa(CO(2)) = 40 +/- 3 Torr), all seven dogs hypoventilated after CBX, reaching a maximum Pa(CO(2)) of 53 +/- 6 Torr by day 3 post-CBX. There was no significant recovery of eupneic Pa(CO(2)) over the ensuing 18 days. Relative to control, the hyperoxic CO(2) ventilatory (change in inspired minute ventilation/change in end-tidal PCO(2)) and tidal volume (change in tidal volume/ change in end-tidal PCO(2)) response slopes were decreased 40 +/- 15 and 35 +/- 20% by day 2 post-CBX. There was no recovery in the ventilatory or tidal volume response slopes to hyperoxic hypercapnia over the ensuing 19 days. We conclude that 1) the carotid bodies contribute approximately 40% of the eupneic drive to breathe and the ventilatory response to hyperoxic hypercapnia and 2) there is no recovery in the eupneic drive to breathe or the ventilatory response to hyperoxic hypercapnia after removal of the carotid chemoreceptors, indicating a lack of central or aortic chemoreceptor plasticity in the adult dog after CBX.