Obstructive sleep apnea (OSA) is associated with oscillations of arterial blood pressure (BP) that occur in phase with irregularities of respiration. To explore the role of the sympathetic nervous system in these responses, we studied muscle sympathetic nerve activity (MSNA; peroneal microneurography), an index of vasoconstrictor nerve traffic, and BP during awake regular breathing and during spontaneous apneas in patients with OSA. To determine the role of the arterial chemoreflex, we also examined the effects of 100% O2 (hyperoxia) on MSNA and BP. In awake regularly breathing patients with OSA (n = 12), resting MSNA was markedly higher than in an age-matched control population (n = 15) [41 +/- 23 (SD) vs. 24 +/- 17 bursts/min; P < 0.05] and was unchanged during hyperoxia (n = 9). Apneas during sleep (n = 8) were associated with surges in MSNA followed by transient rises in BP when breathing resumed. In contrast to room air apneas, hyperoxic apneas of similar duration were associated with attenuated MSNA responses (+82 +/- 84% vs. +5 +/- 25% compared with awake baseline; P < 0.05; n = 6), even though O2 did not affect sleep stage and the occurrence of arousal. Thus the BP oscillations that occur with apnea during sleep may in part be mediated by intermittent surges of sympathetic activity resulting in vasoconstriction. Because the MSNA responses to obstructive apnea are blunted during O2 administration, they appear to be linked to intermittent arterial hypoxemia and stimulation of arterial chemoreceptors.

Forearm vascular responses to intra-arterial infusions of endothelium-dependent andindependent vasodilators have been thoroughly characterized in humans. While the forearm is a well-established experimental model for studying human vascular function, it is of limited consequence to systemic cardiovascular control owing to its small muscle mass and blood flow requirements. In the present study we determined whether these responses could be generalized to the leg. Based upon blood pressure differences between the leg and arm during upright posture, we hypothesized that the responsiveness to endothelium-dependent vasodilators would be greater in the forearm than the leg. Brachial and femoral artery blood flow (Q, ultrasound Doppler) at rest and during intra-arterial infusions of endothelium-dependent (acetylcholine and substance P) and -independent (sodium nitroprusside) vasodilators were measured in eight healthy men (22-27 years old). Resting blood flows in the forearm before infusion of acetylcholine, substance P or sodium nitroprusside were 25 ± 4, 30 ± 7 and 29 ± 5 ml min −1 , respectively, and in the leg were 370 ± 32, 409 ± 62 and 330 ± 30 ml min −1 , respectively. At the highest infusion rate of acetylcholine (16 µg (100 ml tissue) −1 min −1 ) there was a greater (P < 0.05) increase in Q to the forearm (1864 ± 476%) than to the leg (569 ± 86%). Similarly, at the highest infusion rate of substance P (125 pg (100 ml tissue) −1 min −1 ) there was a greater (P < 0.05) increase in Q to the forearm (911 ± 286%) than to the leg (243 ± 58%). The responses to sodium nitroprusside (1 µg (100 ml tissue) −1 min −1 ) were also greater (P < 0.05) in the forearm (925 ± 164%) than in the leg (326 ± 65%). These data indicate that vascular responses to both endothelium-dependent and -independent vasodilator agents are blunted in the leg compared to the forearm.

In heart failure, oxygen has a detrimental effect on cardiac output, stroke volume, pulmonary capillary wedge pressure and systemic vascular resistance. These changes are independent of sympathetic activity and ventilation.