INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ANIMALS STUDIES INDICATE THAT the vestibular system contributes to cardiovascular and sympathetic regulation (1, 19). In humans, head-down rotation (HDR), which engages the otolith organs of the vestibular system, elicits pronounced increases in muscle sympathetic nerve activity (MSNA) (3, 7, 11, 16) that produce limb vasoconstriction (7, 16). Recently, using off-vertical axis chair rotation, which activates the otolith organs, Kaufmann et al. (4) also found increases in MSNA. The strength of the vestibulosympathetic reflex is apparent in that this reflex continues to elicit increases in MSNA even in the setting of pronounced sympathoexcitation (e.g., baroreflex unloading and skeletal muscle reflex engagement) (9, 10). Furthermore, experimental data indicate that the reflex interaction between the vestibulosympathetic reflex and both the baroreflex (9) and skeletal muscle reflex (10) on MSNA output in humans is additive.

The interaction between the vestibulosympathetic reflex and the arterial chemoreflex is unknown. Both the arterial chemoreflex and the vestibulosympathetic reflex can produce pronounced increases in MSNA in humans (14-16). Thus the purpose of this study was to determine the interaction between the vestibulosympathetic reflex and the arterial chemoreflex. It was hypothesized that the vestibulosympathetic reflex and arterial chemoreflex interaction would be additive in respect to MSNA. Support for this hypothesis would be consistent with the nature of the interaction between the vestibulosympathetic reflex and the baroreflex and skeletal muscle reflex. The findings of the present study suggest that the neural interaction of the vestibulosympathetic reflex and arterial chemoreflex is additive in respect to their output, but its hemodynamic influence is inhibitory in nature. This inhibitory influence on hemodynamic measures may be a result of the local effects of hypoxia on vascular tone and the heart.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Experimental Design

Protocol 1. The purpose of this experimental protocol was to determine both the MSNA and cardiovascular responses to HDR. Subjects performed HDR in the prone position (16). Briefly, subjects were positioned on an examination table with the head extending over the end of the table, so the head could be rotated downward without interference from the table. Before HDR, the subject's head was supported upright (chin-up neck extended position). After a 3-min baseline period, the chin support was removed and the head was passively rotated downward to the point of maximal rotation. After a 1-min period of HDR, the subject's head was returned to the baseline chin-up position (Fig. 1).

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Experimental timeline for protocols 1, 2, and 3. In protocol 1, head-down rotation (HDR) was performed (1 min) after a baseline period with the head in the chin-up neck-extended position (3 min). A period of recovery (3 min) followed HDR with the head in the baseline position. For protocols 2 and 3, after a baseline period of normoxia (3 min) with the head in the baseline position, hypoxic breathing began. After a period of hypoxia (5 and 13 min for protocols 2 and 3, respectively), HDR was performed (1 min). After this period of HDR, the head was returned to the baseline position for the final minute of hypoxia. After hypoxia, a 3-min period of recovery ensued while subjects breathed room air. ', Minutes.

Protocol 2. The purpose of this experimental protocol was to determine the interaction between the vestibulosympathetic reflex and the arterial chemoreflex. At baseline, the prone subject's head was extended over the edge of the examination table for a 3-min baseline period. This was followed by hypoxic gas inhalation (10% O₂ and 90% N₂) for 7 min. During minute 6 of hypoxia, HDR was performed for 1 min. After this 1-min period, the head was returned to the baseline chin-up position for the final minute of hypoxic exposure (Fig. 1).

Protocol 3. The purpose of this experimental protocol was to determine whether the nature of the interaction between the vestibulosympathetic reflex and the arterial chemoreflex demonstrated in protocol 2 persisted during an extended (15 min) period of hypoxia. As in protocols 1 and 2, the baseline period was obtained with the prone subject's head in the baseline chin-up supported position breathing room air. After a 3-min baseline period, the subject began to breathe a hypoxic gas mixture (10% O₂ and 90% N₂). During minute 14 of hypoxia, HDR was performed for 1 min. After this 1-min period, the head was returned to the baseline chin-up position and hypoxia was maintained for minute 15.

Measurements

During all trials, the subjects breathed through an open-circuit non-rebreathing mouthpiece (Hans Rudolph, Kansas City, MO). During hypoxia, the inspiratory port was switched from room air to a hypoxic gas mixture via a two-way manually operated valve. Responses to hypoxia were determined in the minute preceding HDR (minute 5 and minute 13 for protocol 2 and protocol 3, respectively). Continuous measurements of MAP were made by use of a Finapres photoplethysmograph (Ohmeda, Louisville, CO). Continuous measurements of arterial oxygen saturation were made from the subject's fingertip by using pulse oximetry (Ohmeda). The mean voltage neurogram, HR, MAP, and arterial oxygen saturation were digitized (MacLab, ADInstruments, Milford, MA) and stored on computer for real-time data monitoring and later analyses.

Data Analysis

Responses during the three protocols were compared by using a repeated-measure ANOVA. Because responses during protocols 2 and 3 did not differ significantly, the responses were combined, and data during protocols 2 and 3 are presented as a single trial. Changes in MSNA, MAP, and HR from baseline were calculated. The algebraic sum of the change scores for the individual HDR and the hypoxia periods were compared with the change score when HDR and hypoxia were performed simultaneously by use of a paired t-test. Statistical significance was set at a P value of <0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Baseline levels of MSNA (15 ± 2 vs. 15 ± 2 bursts/min for HDR and hypoxia trials, respectively), MAP (96 ± 4 vs. 97 ± 3 mmHg), HR (69 ± 2 vs. 68 ± 2 beats/min), and oxygen saturation (98 ± 1 vs. 98 ± 1%) were similar before each protocol. Inspiration of the hypoxic gas mixture significantly reduced the level of arterial oxygen saturation (98 ± 1 vs. 76 ± 1% for normoxia and hypoxia, respectively). HDR did not alter arterial oxygen saturation during either normoxia (98 ± 1 vs. 98 ± 1% before and after HDR, respectively) or hypoxia (76 ± 1 vs. 76 ± 1%).

MSNA increased during both HDR [increase ()7 ± 1 bursts/min and 50 ± 11% for burst frequency and total MSNA, respectively; P < 0.05] and hypoxia (6 ± 2 bursts/min and 62 ± 29%; P < 0.05) (Fig. 2). The algebraic sum of the increase in MSNA for HDR and hypoxia (13 ± 1 bursts/min and 115 ± 36%) did not differ significantly from the increases in MSNA when HDR was performed during hypoxia (11 ± 2 bursts/min and 127 ± 57% change from baseline period of normoxia) (Fig. 2). Neurograms during the different intervention from one subject are depicted in Figure 3. MAP was not altered by HDR but was increased by hypoxia (5 ± 1 mmHg; P < 0.05) (Fig. 4). The algebraic sum of the responses in MAP to HDR and hypoxia (5 ± 1 mmHg; P < 0.05) was greater than the increases in MAP when HDR was performed during hypoxia (3 ± 1 mmHg for change from baseline period of normoxia; P < 0.05 compared with the algebraic sum) (Fig. 4). HR increased during both HDR (4 ± 1 beats/min; P < 0.05) and hypoxia (21 ± 2 beats/min; P < 0.05) (Fig. 4). The algebraic sum of these increases (24 ± 2 beats/min) was significantly greater than the increases in HR when HDR was performed during hypoxia (19 ± 21 beats/min for change from baseline period of normoxia) (Fig. 4).

View larger version (23K): [in this window] [in a new window]   Fig. 2.   HDR and hypoxia elicited significant increases () in muscle sympathetic nerve activity. The algebraic sum of the individual responses to HDR and to hypoxia was not different from the increase in muscle sympathetic nerve activity during HDR and hypoxia performed simultaneously. A:  burst activity. B:  total activity. All comparisons, P < 0.05 from baseline. N.S., not significant.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Representative neurograms from 1 subject obtained during baseline, HDR, hypoxia, and the combined period of hypoxia and HDR.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Mean arterial blood pressure (A) and heart rate (B) responses during the experimental interventions. The algebraic sum of the individual responses to hypoxia and to HDR was greater than the response to HDR performed during hypoxia.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The primary finding from this study is that the reflex interaction between the vestibulosympathetic reflex and arterial chemoreflex is additive for MSNA in humans. In contrast, the MAP and HR reflex interaction exhibits an inhibitory interaction. These findings suggest that there is not central integration between the arterial chemoreflex and vestibulosympathetic reflex in regard to reflex control of MSNA output in humans. The inhibitory interaction in the cardiovascular variables may be produced by local influences of systemic hypoxia on vascular tone and HR.

Despite the additive influence of otolithic engagement on hypoxia-induced increases in MSNA, both the MAP and HR responses exhibit an inhibitory reflex interaction. Hypoxia is a complex stimulus. In addition to pronounced increases in MSNA, hypoxia elicits vasodilation and tachycardia. The mechanisms underlying the vasodilation are not understood but likely involve alterations in local endothelial and metabolic products such as adenosine, prostaglandins, local pH, and nitric oxide as well as increased levels of circulating epinephrine (6, 12, 18). Hypoxia attenuates sympathetic vasoconstriction in animals (2), suggesting decreased postjunctional adrenergic receptor sensitivity. Release of these vasodilator substances and decreased sympathetic vasoconstrictor responsiveness are likely to contribute to the local vasodilation. Additionally, it is likely that the increased vasoconstrictor MSNA outflow during hypoxia restrains and obscures the true magnitude of the vasodilatory stimuli (18).

We are not able to determine why the MAP increases during hypoxia and activation of the vestibulosympathetic reflex performed simultaneously was inhibited relative to the individual responses. However, it is likely to involve one or more of the complex mechanisms, initiated by hypoxia, described above. Additionally, the vestibulosympathetic reflex appears to elicit vasodilation in humans (7, 11). This suggestion of the vestibulosympathetic reflex mediating vasodilation is supported by the fact that HDR is associated with maintained arterial blood pressure, despite preserved cardiac output (16) and pronounced limb vasoconstriction (7). The site of this vasodilation remains unknown but may involve the renal or mesenteric vascular beds. Thus it is possible that, despite a similar increase in vasoconstrictor MSNA during hypoxia, the vasodilator contribution of the vestibulosympathetic reflex response is augmented by hypoxia. We speculate that it may play a role in the blunted increase in MAP during simultaneous vestibulosympathetic reflex and arterial chemoreflex engagement. However, the magnitude of this inhibitory interaction (~3 mmHg) is small and thus unlikely to have profound physiological implications.

The mechanism(s) underlying the inhibitory influence of hypoxia and the vestibulosympathetic reflex on HR responses is unclear. Previously, HDR has been demonstrated to reduce a measure of vagal tone in humans (5). Additionally, hypoxia elicits tachycardia. Thus it is possible that the ability to further withdraw vagal tone when HDR is performed in the tachycardic hypoxic state is diminished by the level of vagal tone available to withdraw. In support of this concept, the HR during hypoxia performed alone approached 100 beats/min. At HRs approaching this level, the ability for further vagal withdrawal is nearly exhausted (13), and this may explain in part why the reflex increase in HR was blunted during simultaneous hypoxia and HDR relative to the individual responses. However, the chronotropic responses to acute hypoxia exposure are likely to be mediated by both withdrawal of vagal tone and increased sympathetic nervous system activation directed to the heart. Thus vagal tone may not have been exhausted in the present study, but its reduction may have contributed, in part, to the present findings.

The demonstration of an additive influence of HDR and arterial chemoreflex activation on MSNA is supported by the demonstration that the algebraic sum of the individual increases in MSNA during separately performed HDR and hypoxia are similar in magnitude to MSNA increases during a combined period of hypoxia and HDR. This additive influence on MSNA is similar to that noted when the vestibulosympathetic reflex was engaged during both baroreflex unloading (9) and activation of the skeletal muscle reflex (10). Thus these previous studies as well as the present study provide experimental support for the concept that the vestibulosympathetic reflex is a robust and independent reflex in regards to reflex control of MSNA output in humans.