|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary, Allergy and Critical Care (Dr. Imadojemu), Penn State Heart & Vascular Institute (Drs. Mawji and Leuenberger, Ms. Gray, and Ms. Hogeman), Department of Health Evaluation Sciences (Mr. Kunselman), The Milton S. Hershey Medical Center, The Pennsylvania State University College of Medicine, Hershey, PA.
Correspondence to: Urs A. Leuenberger, MD, Penn State Heart & Vascular Institute, Mail Code H047, 500 University Dr, Hershey, PA 17033; e-mail: uleuenberger{at}psu.edu
Abstract
Background: Sympathetic nerve activity is increased in awake and regularly breathing patients with obstructive sleep apnea (OSA). Over time, repetitive hypoxic stress could alter sympathetic chemoreflex function in OSA.
Methods: We determined the responses to acute hypoxia (fraction of inspired oxygen of 0.1, for 5 min), static handgrip exercise, and the cold pressor test (CPT) in 24 patients with OSA (age, 50 ± 3 years [mean ± SEM]; apnea-hypopnea index, 47 ± 6 events per hour) and in 14 age- and weight-matched nonapneic control subjects. Muscle sympathetic nerve activity (MSNA) [peroneal microneurography], BP, and ventilation were monitored.
Results: Basal MSNA was higher in OSA patients compared to control subjects (45 ± 4 bursts per minute vs 33 ± 4 bursts per minute, respectively; p < 0.05). Furthermore, compared to control subjects, the MSNA responses to hypoxia were markedly enhanced in OSA (p < 0.001). Whereas the ventilatory responses to hypoxia tended to be increased in OSA (p = 0.06), the BP responses did not differ between the groups (p = 0.45). The neurocirculatory reflex responses to handgrip exercise and to the CPT were similar in the two groups (p = not significant). In OSA patients who were retested after 1 to 24 months of continuous positive airway pressure (CPAP) therapy (n = 11), basal MSNA (p < 0.01) and the responses of MSNA to hypoxia (p < 0.01) decreased significantly, whereas the ventilatory responses remained unchanged (p = 0.82).
Conclusion: These data suggest that the sympathetic responses to hypoxic chemoreflex stimulation are enhanced in OSA and may normalize in part following CPAP therapy.
Key Words: autonomic nervous hypoxemia sleep apnea
Sympathetic nerve activity is increased in awake and regularly breathing patients with obstructive sleep apnea (OSA).1234 The cause of chronic sympathoexcitation in OSA is not known. Because BP is elevated in many of these patients,5 it has been postulated that hypertension in OSA is linked to elevated sympathetic activity.6
Obstructive apnea during sleep is associated with transient surges of sympathetic activity147 that are blunted substantially when oxyhemoglobin desaturation is prevented by administration of oxygen.1 Therefore, the surges of sympathetic activity triggered by obstructive apnea are in part due to hypoxia and activation of carotid and/or aortic chemoreceptors. Because OSA is associated with repetitive nocturnal episodes of hypoxia, the sympathetic responses to acute hypoxia may be altered over time. Enhanced chemoreflex activity could play a role in the pathogenesis of chronic sympathoexcitation and hypertension in OSA.8 In support of this concept, in an animal model, chronic intermittent hypoxia has been shown to result in enhanced chemoreflex responses.9 In humans with OSA, isocapnic hypoxia has been reported to be associated with a pressor response.10 Narkiewicz et al11 reported increased ventilatory and BP responses but similar sympathetic nerve responses to acute hypoxia when isocapnia was maintained in patients with OSA compared to control subjects. Because increased BP (via baroreflex deactivation) and enhanced ventilatory responses may attenuate hypoxia-induced sympathetic activation,1213 these findings were consistent with enhanced chemoreflex sensitivity in OSA.11 Whether sympathetic activation would be unmasked during hypoxia accompanied by spontaneous hypocapnia is not known. Continuous positive airway pressure (CPAP) therapy has been shown to reduce daytime sympathetic nervous system activity in patients with OSA.141516 However, whether CPAP therapy normalizes sympathetic reflex function in OSA is not known.
Therefore, the goals of this study were to determine the neurocirculatory and ventilatory responses to a brief exposure to acute hypoxia and to evaluate the effect of CPAP therapy on these responses in patients with OSA. We hypothesized that OSA subjects would exhibit enhanced sympathetic and/or ventilatory responses to hypoxia and that these responses would be improved by CPAP therapy. Our data suggest that sympathetic chemosensitivity to hypoxia is enhanced in OSA and may in part normalize following CPAP therapy.
Materials and Methods
Subjects
Thirty-nine patients with the clinical syndrome of OSA and an apnea-hypopnea-index > 10 events per hour documented by standard overnight polysomnography were recruited through the sleep laboratory at the Penn State University Milton S. Hershey Medical Center. Patients with symptomatic coronary artery disease or intrinsic lung disease and those who had been previously treated with CPAP therapy were excluded. Twenty-one age- and weight-matched, otherwise-healthy control subjects were recruited in the local community and underwent polysomnography to exclude OSA. Twenty-four OSA patients and 14 control subjects in whom technically satisfactory nerve recordings were obtained constituted our study population. Three of the OSA patients and none of the control subjects were smokers. Eight of the OSA patients were known hypertensives, five patients were receiving an angiotensin-converting enzyme inhibitor, two patients were receiving a thiazide diuretic, and one patient was receiving a ß-blocker. Two other OSA patients were receiving oral hypoglycemic agents. One of the control subjects was receiving an angiotensin-converting enzyme inhibitor, and one control subject was receiving a ß-blocker. The drugs were withheld on the morning of the study. The study protocol was approved by the Institutional Review Board, and written informed consent was obtained. The studies were performed in the General Clinical Research Center after overnight fasting, in the morning hours and with the subjects in a supine position.
Hemodynamic Measurements
Beat-by-beat BP was determined via finger photoplethysmography (Finapres; Ohmeda; Madison, WI).17 To prevent finger edema, the device was turned off every 30 min for approximately 1 min. Basal BP was confirmed with an automated sphygmomanometer (Dinamap; Critikon; Tampa, FL). Mean arterial pressure (MAP) was calculated as diastolic pressure plus one third of pulse pressure. Heart rate (HR) was derived via ECG.
Microneurography
Peroneal microneurography was used to determine sympathetic vasoconstrictor nerve traffic directed to skeletal muscle arterioles (muscle sympathetic nerve activity [MSNA]) as described previously.1 MSNA was expressed in bursts per minute and in total amplitude per minute. To account for the variability in gain settings between individuals, in each individual the amplitude of an average burst at baseline was arbitrarily given a value of 100 U.
Ventilatory Parameters
Minute ventilation (
E) and end-tidal carbon dioxide level were determined with a gas monitor (Ohmeda Respiratory Gas Monitor, model 5250; Ohmeda). Arterial oxygen saturation (SaO2) was monitored by ear oximetry (model 3740; Ohmeda).
Experimental Protocols
Acute Hypoxia:
Following instrumentation, a face mask with separate valves for inspiratory and expiratory gas flow was placed and checked for leaks. Following acclimatization to the face mask, MSNA, arterial pressure, and respiratory parameters were recorded for 5 min (baseline). The inspiratory port was then connected to hypoxic gas (fraction of inspired oxygen [FIO2] of 0.1) for 5 min (hypoxia).
Static Handgrip Exercise: After a 30-min rest period, the subjects performed 2 min of static handgrip exercise at 30% of maximum force (Stoelting Dynamometer; Stoelting; Wood Dale, IL). At the end of handgrip, a pneumatic arm cuff was inflated to 250 mm Hg to achieve post-handgrip circulatory arrest for 2 min. By trapping muscle metabolites in the forearm, this maneuver isolates the contribution of muscle metaboreflex activation to the integrated sympathetic response to handgrip exercise.18 The reflex responses to handgrip exercise were examined to assess sympathetic reflex function unrelated to the arterial chemoreflex.
Cold Pressor Test: After a 30-min rest period, the subjects hand was placed in ice-cold water (cold pressor test [CPT]) for 90 s. The CPT was used as a nonspecific stimulus to the sympathetic nervous system.19
CPAP Therapy
Following diagnostic polysomnography, the OSA patients underwent a CPAP trial and were fitted with a CPAP device. After at least approximately 1 month of CPAP therapy, the experimental protocols were repeated.
Data Analysis
Baseline data were averaged over a 5-min period. Data obtained during each intervention (hypoxia, static handgrip/post-handgrip circulatory arrest, and CPT) were averaged per minute. The unpaired t test was used to compare baseline characteristics between OSA and control subjects. One-way repeated-measures analysis of variance was performed to compare within-group responses across the paradigm.20 Post hoc comparisons were made with a Dunnett test. Between-group comparisons were made with two-way repeated-measures analysis of variance or by unpaired t test as appropriate. The statistical analyses were performed using statistical software (SAS version 6.12; SAS Institute; Cary, NC); p < 0.05 was considered statistically significant. The data are presented as mean ± SEM.
Results
Baseline Characteristics
Baseline characteristics of the OSA patients and the control subjects are shown in Table 1
. The two groups were well matched for age, sex, and body mass index. During room air breathing, the discharge rate of sympathetic nerve fascicles (MSNA), MAP, and HR were higher in patients with OSA compared to control subjects (p < 0.05).
|
E, and a mild decrease in end-tidal carbon dioxide. Whereas MAP did not change in the control subjects, it was mildly increased during the last min of hypoxia in the OSA group (Table 2).
|
|
E tended to be greater in OSA than in control subjects (F = 3.8; p = 0.06). In OSA, during the first minute of hypoxia,
E rose by 42 ± 6% but only by 28 ± 3% in control subjects (p < 0.05). In contrast, by two-way analysis of variance, the BP responses to acute hypoxia were similar in the two groups (p = 0.45). While baseline HR was higher in OSA than in control subjects (p < 0.01), the HR increased similarly in the two groups (p = 0.20). Although in OSA oxygen saturation appeared to decrease more promptly during hypoxia, this difference was not statistically significant (p = 0.12). At the end of hypoxia, the decrease of SaO2 was similar in the two groups (OSA vs control subjects, 16 ± 1% vs 17 ± 1%, respectively; p = 0.50).
Responses to Static Handgrip/Post-Handgrip Circulatory Arrest and the CPT in OSA and Control Subjects
The effects of static handgrip and post-handgrip circulatory arrest on BP, HR, and MSNA were similar in OSA and in control subjects (Table 3
). Likewise, the effects of the CPT on BP, HR, and MSNA in OSA and in control subjects were similar (Table 4
).
|
|
E (Fig 2
), and on SaO2, and end-tidal carbon dioxide (data not shown) to hypoxia.
|
There are two important findings in this study. First, compared to control subjects, patients with OSA exhibited a greater increase of sympathetic activity and a trend toward a greater rise of
E but similar BP responses when exposed to acute hypocapnic hypoxia. Because the exercise pressor reflex and the responses to the CPT were similar in OSA and control subjects, this suggests a specific sensitization of the arterial chemoreflex in OSA. Second, intermediate-duration CPAP decreased basal sympathetic activity and in part normalized the sympathetic responses to hypoxia. This suggests that CPAP therapy may at least in part reverse chemoreflex sensitization in OSA.
Arterial chemoreflex sensitivity in OSA is of interest because this reflex plays an important role in mediating the acute neurocirculatory effects of apnea212223 because repetitive exposure to hypoxia may alter chemoreflex sensitivity,9 and because enhanced chemoreflex activity has been implicated in the pathogenesis of hypertension.824 In agreement with prior studies,25262728 we found that in healthy humans, acute hypocapnic hypoxia causes reflex activation of the sympathetic nervous system, a rise in HR and
E, but no significant change in arterial pressure. These effects are the net result of chemoreflex-mediated sympathetic activation, decreased vagal tone, and production of vasodilator metabolites in many vascular beds.2526272829 Compared to control subjects, in patients with OSA the hypoxia-induced increase of sympathetic activity was enhanced and ventilation tended to be increased. The prominent excitatory effect of hypoxia on sympathetic nerve activity in OSA is remarkable because enhanced ventilation is expected to inhibit sympathetic activity.12 Whether the increased neural vasoconstrictor signal in OSA during hypoxia causes greater sympathetic restraint of hypoxia-induced vasodilation remains to be determined.
In contrast to prior reports,1011 the pressor response to hypoxia we observed in OSA was small and was not statistically different from the response in the control subjects. Our findings differ in part from a report by Narkiewicz et al,11 in that we found an increased sympathetic response but only a trend for an augmented ventilatory response to hypoxia in OSA. These investigators11 reported that the ventilatory response to a 3-min bout of isocapnic hypoxia was enhanced in OSA, whereas the increase of sympathetic nerve activity was similar. Even though the responses of sympathetic nerve activity were similar in OSA and control subjects, based on greater increases of
E and BP, which are expected to counteract a rise in sympathetic nerve activity, they concluded that the peripheral chemoreflex was potentiated in OSA.11 Methodologic differences are likely responsible for the discrepancies between these and our results. Specifically, while in these studies an attempt was made to maintain isocapnia, our findings represent the effects of hypocapnic hypoxia of longer duration (5 min). It has been reported that compared to hypocapnic hypoxia, isocapnic hypoxia leads to a greater ventilatory response, a smaller rise in sympathetic activity, and an increase in BP in normal humans.12
Despite identical FIO2 and a trend toward enhanced ventilatory responses compared to the control subjects when exposed to hypoxia, in OSA the oxygen desaturation was not attenuated. This could relate to pulmonary ventilation-perfusion mismatch in patients with OSA but not in the control subjects. Importantly, at the end of hypoxia the level of oxygen desaturation was not different in the two groups, suggesting that the different MSNA responses cannot be explained by differences in the hypoxic stimulus.
In OSA patients and in control subjects alike, MSNA and BP rose during static handgrip and remained elevated during forearm ischemia. This is consistent with activation of the muscle metaboreflex. The magnitude of this response was similar to that reported previously for healthy humans.1830 This suggests that unlike in heart failure, a condition that similar to OSA is characterized by increased sympathetic nerve activity, the muscle metaboreflex is preserved in OSA. In addition, nonspecific sympathetic stimulation elicited by the CPT revealed no OSA-specific abnormality of this reflex. Taken together and in agreement with the report by Narkiewicz et al,11 the enhanced chemoreflex responses but preserved muscle metaboreflex and cold pressor responses in OSA are consistent with a specific sensitization of the arterial chemoreflex.
CPAP is the current therapeutic corner stone for OSA and has been shown to improve symptoms,31 to decrease BP,32 to decrease daytime sympathetic nervous system activity,1533 and to improve vascular function.34 CPAP presumably exerts these beneficial effects by eliminating obstructive apneas during sleep.35 Our data suggest that CPAP therapy improves symptoms of daytime somnolence, reduces basal daytime MSNA, and appears to normalize the sympathetic responses, whereas it has no effect on the ventilatory responses to acute hypoxia. Because tonic activation of excitatory chemoreflex afferents may contribute to increased efferent sympathetic nerve traffic directed to skeletal muscle, prevention of apneic events by CPAP may attenuate sympathetic nerve activity by decreasing the sensitivity and tonic activity of the chemoreflex.11 However, because our sample size was small and the duration of CPAP was limited to a few months in all but two subjects and because all of our subjects were highly compliant, a goal that is not commonly achieved in many patients,36 our data do not allow us to quantify the influence of CPAP duration and compliance on chemoreflex function in OSA.
The mechanism underlying the enhanced sympathetic and ventilatory chemoreflex responses and their anatomic substrate are unclear. It has been postulated that enhanced chemoreflex responses following intermittent hypoxia may be due to changes in brainstem neural circuits involved in the regulation and integration of chemoreflex activity.9 One such potential site may be the rostral ventrolateral medulla, as neurons in this area are known to play a major role in autonomic neural control.37 Alternatively, functional and/or structural changes in the carotid bodies in response to intermittent hypoxia should also be considered.38 For example, enhanced chemosensitivity to hypoxia noted in our study could represent a manifestation of "long-term facilitation," a phenomenon characteristically observed in laboratory models after intermittent but not continuous hypoxia.39 In animal models, intermittent hypoxia has also been shown to result in increased afferent nerve traffic from the carotid bodies and enhanced responses to subsequent episodes of hypoxia.39 It has been postulated that this mechanism may contribute to the chronically elevated sympathetic activity seen in OSA.39 Both scenarios, altered integration of neural inputs at the level of the brainstem or enhanced sensitivity of peripheral chemoreceptors, would be consistent with findings from studies in healthy humans that demonstrate that even short-term intermittent hypoxia raises sympathetic activity404142 and BP.42
In conclusion, our data demonstrate that the sympathetic neural responses to hypoxia but not the reflex responses to handgrip exercise or to cold pressor stimulation were enhanced in OSA. This suggests that arterial chemoreflex sensitivity is increased in OSA. Furthermore, CPAP therapy, which largely eliminates intermittent nocturnal hypoxia in patients with OSA, decreased sympathetic activity, tended to lower arterial pressure and in part normalized the sympathetic responses to hypoxia suggesting that these chemoreflex abnormalities are due to the nocturnal breathing disturbance typical of OSA. These findings provide a mechanistic rationale for the use of CPAP therapy in this disease. The precise anatomic and functional substrates responsible for chemoreflex sensitization in OSA require further investigation.
Acknowledgements
The authors thank the study subjects for participation in this project, the nursing staff in the General Clinical Research Center for assistance with monitoring of the subjects, and Jennie Stoner for secretarial assistance.
Footnotes
Abbreviations: CPAP = continuous positive airway pressure; CPT = cold pressor test; FIO2 = fraction of inspired oxygen; HR = heart rate; MAP = mean arterial pressure; MSNA = muscle sympathetic nerve activity; OSA = obstructive sleep apnea; SaO2 = arterial oxyhemoglobin saturation;
E = minute ventilation
This work was supported in part by grants K23 HL04190 (V.A.I.), R01 HL68699 (U.A.L.), and National Institutes of Health/National Center for Research Resources grant M01 RR10732.
The authors have no conflicts of interest to disclose.
Received for publication October 23, 2006. Accepted for publication December 18, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |