| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of 1 Neurobiology, 2 Radiology, and 3 Neurology, 4 School of Nursing, University of California, Los Angeles, Los Angeles, California 90095-1763
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The repetitive upper airway muscle atonic episodes and cardiovascular sequelae of obstructive sleep apnea (OSA) suggest dysfunction of specific neural sites that integrate afferent airway signals with autonomic and somatic outflow. We determined neural responses to the Valsalva maneuver by using functional magnetic resonance imaging. Images were collected during a baseline and three Valsalva maneuvers in 8 drug-free OSA patients and 15 controls. Multiple cortical, midbrain, pontine, and medullary regions in both groups showed intensity changes correlated to airway pressure. In OSA subjects, the left inferior parietal cortex, superior temporal gyrus, posterior insular cortex, cerebellar cortex, fastigial nucleus, and hippocampus showed attenuated signal changes compared with controls. Enhanced responses emerged in the left lateral precentral gyrus, left anterior cingulate, and superior frontal cortex of OSA patients. The anterior cingulate, cerebellar cortex, and posterior insula exhibited altered response timing patterns between control and OSA subjects. The response patterns in OSA subjects suggest deficits in particular neural pathways that normally mediate the Valsalva maneuver and compensatory actions in other structures.
functional magnetic resonance imaging; cerebellum; insula; anterior cingulate; hippocampus
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
PATIENTS WITH OBSTRUCTIVE SLEEP APNEA (OSA) exhibit atonia of the upper airway musculature in the presence of continued diaphragmatic efforts during sleep, resulting in cessation of airflow, enhanced venous return, and exaggerated transient rises in arterial pressure, together with a tachycardic-bradycardic sequence (44). The syndrome is associated with a cluster of anatomic features, including obesity, enhanced airway resistance, and micrognathia. However, certain characteristics of OSA suggest that central neural dysfunction, as well as body morphology, contribute to the syndrome. These neurogenic features include impaired autonomic responses to the Valsalva maneuver (48), the occasional emergence of central apnea before obstructive events (3), a propensity for the syndrome to worsen without significant weight gain or change in upper airway anatomy (36), and amelioration of the syndrome with atrial pacing, an effect presumably mediated by central integration of cardiac afferent information with respiratory control mechanisms (10).
The persistence of atonia in the upper airway, despite continued and even enhanced diaphragmatic efforts in OSA, may represent a loss of coordination between peripheral sensory stimulation, peripheral and central chemoreception, and output to upper airway musculature. Evidence of aberrant peripheral afferent processes mediating sensory information from the upper airway in OSA exists (17). Deficient integration of airflow and airway pressure afferent information associated with cessation of airflow within regions that control coordination of motor output may participate in initiation or maintenance of the syndrome. In this context, both inadequate extent of neural response or altered response timing may be etiologic factors.
The potential for dysfunctional brain areas in OSA patients to respond inappropriately to respiratory or cardiovascular challenges should be viewed in the context of findings of significant gray matter loss in afflicted patients (24). The affected brain areas included sites that serve upper airway muscle control, limbic sites involved in initiation of inspiratory effort or affective control of vocalization musculature, and cerebellar sites mediating coordination of motor and cardiovascular control. The gray matter loss may result in specific deficiencies in response to autonomic and breathing challenges, but the functional role in OSA is unknown.
The Valsalva maneuver consists of a prolonged expiratory effort against a high resistance, resulting in increased thoracic and upper airway pressure and a sequence of sympathetic and parasympathetic effects on blood pressure and heart rate that have been useful for autonomic evaluation of a range of clinical syndromes (5, 20, 28). The procedure offers a temporal assessment of participation of neural structures mediating sympathetic and parasympathetic outflow resulting from the decreased venous return and alteration in blood pressure accompanying the maneuver, and, in addition, some indication of processing of airway pressure afferents and voluntary breathing effort.
We used functional magnetic resonance imaging (fMRI) to visualize brain areas recruited by the Valsalva maneuver in OSA and control subjects. The maneuver activates multiple brain sites, as indicated earlier (12, 14), and elicits aberrant autonomic responses in OSA subjects (11). Correlation of the time course of neural activation with physiological responses to the Valsalva maneuver in OSA patients can indicate areas that are deficient in the patient population and suggest regions important for maintaining adequate upper airway patency and appropriate autonomic responses.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Twenty-one healthy male control subjects and 21 male OSA
subjects participated in the study. Subjects were subsequently excluded from the study if they were currently using cardiovascular-altering medications such as 
-blockers, 
-agonists, vasodilators,
angiotensin-converting enzyme inhibitors, cholinergic stimulating
drugs, or mood-altering medication such as serotonin reuptake
inhibitors. Similarly, subjects with syndromes, other than OSA,
characterized by autonomic effects (e.g., diabetes mellitus) were
excluded. Any subject who did not maintain a load pressure of 30 mmHg
for 15 s during the Valsalva challenges was also excluded. Of the
initial 42 subjects, data from 15 control (mean age 45 ± 3 yr;
range 30-58 yr) and 8 OSA subjects (mean age 44 ± 4 yr;
range 31-63 yr) were used for further analysis. Data from 12 of
the 15 control subjects have been reported previously
(14). The body mass index of both OSA subjects and controls was similar (controls, 31 vs. OSA, 27). Values for age, systolic, diastolic and mean blood pressure (BP), body mass index, respiratory disturbance index, disease severity [diagnosed according to standard techniques (40)], continuous positive airway
pressure or mandibular prosthesis use, and time from initial diagnosis to image recording are shown in Table 1.
The study was approved by the Institutional Review Board of the
University of California at Los Angeles. The procedures were conducted
with the understanding and written consent of the subjects.
|
Each subject wore nose clips and breathed via a mouthpiece through a two-way nonrebreather respiratory valve (Hans Rudolph, Kansas City, MO). The ECG was measured by using standard magnetic resonance imaging-compatible surface electrodes and arterial oxygen saturation was acquired by use of a Nonin oximeter. These signals were amplified (32) and transferred externally via magnetic resonance-compatible equipment (13). End-tidal CO2 was monitored through a cannula on the mouthpiece. Airway load pressure and thoracic wall movements were measured via tubing attached to the mouthpiece and to an air-filled bag held in place with a belt on the thoracic wall, respectively. The tubing was fed to pressure transducers external to the scanner room. Arterial pressure (AP) readings were obtained immediately before and after each scanning period. All physiological signals were acquired on digital media (Quatech, Akron, OH) with 12-bit analog-to-digital converters. Peak detection software was used to calculate heart rate (HR), respiratory rate, and mean O2 saturation.
Three Valsalva maneuvers were performed during the challenge period. Subjects were instructed to exhale strongly into a mouthpiece for a period of 18 s, maintaining a mean load pressure (LP) above 30 mmHg. Each effort was followed by an 18-s recovery period. Breathing was restricted by a flow resistor, that allowed a slow air leak, ensuring an open glottis and accurate intrathoracic pressure measures. Before the scanning period, each subject practiced this sequence a minimum of three times until the required load pressure was maintained for an 18-s period. Valsalva ratios (VR = maximum cardiac interbeat interval/minimum cardiac interbeat interval) were calculated for each maneuver for each subject and plotted against age and LP.
Images were collected with a 1.5-T magnetic resonance scanner (General Electric Signa, Milwaukee, WI). Foam pads placed on both sides of the head and masking tape, applied over the forehead to the scanner head holder, were used to minimize head movement. Gradient echo echo-planar imaging using blood oxygen level-dependent contrast was used. A time series of 25 echo-planar image volumes (repetition time = 6 s per volume, echo time = 60 ms, flip angle = 90 degrees, field of view = 30 × 30 cm, no interslice gap, voxel size = 2.3 × 2.3 × 5.0 mm thick), composed of 20 oblique sections, was acquired continuously during a 150-s period, with 60-s baseline (10 volumes) followed by a 90-s challenge period (15 volumes). A series of T1-weighted anatomical images (repetition time = 500 ms, echo time = 9 ms, field of view = 30 × 30 cm, no interslice gap, voxel size = 1.2 × 1.6 × 5.0 mm thick) was collected at the same levels as the functional images.
In four control and five OSA subjects, replicate sessions were collected, resulting in a total of 25 individual scanning sessions for the control group and 17 for the OSA group. The first volume in each series was removed to account for signal saturation. Image volumes were then corrected for slice acquisition timing, and motion was spatially normalized to the Montreal Neurological Institute co-ordinate system, Gaussian smoothed (full-width-at-half-maximum = 8 mm) using SPM99 software (8). All image sets were then intensity normalized by use of in-house software. Multiple trials in a single subject were considered by averaging the image sets for that subject. Thus for each subject, one series of images was used for subsequent statistical analysis.
Cluster Analysis
With the use of SPM99, the intensity normalized images for each subject were analyzed with second level population effects procedures. Each subject's time series was modeled with load pressure (convolved with the hemodynamic response function) as the independent variable. The resulting parameter values for each voxel were stored in a contrast image. A one-sample t-test was performed on the resulting contrast images to determine regions responding in a similar fashion across both groups. For presentation purposes, the resulting statistical maps were thresholded (corrected P < 0.01), color-coded for statistical significance, and overlaid onto spatially-normalized mean T1-weighted anatomical images. The statistical maps were initially overlaid onto a mean functional image set to verify the anatomical localization of the significant voxels. A two-sample t-test was performed to determine regions responding differently between the control and OSA groups. For comparisons between groups, the design comparing contrast images is less powerful, and we used less conservative uncorrected t-tests to identify the empirically derived clusters that represent areas of potential difference in the OSA population. These areas were also thresholded (uncorrected P < 0.05) and overlaid onto a mean anatomical image set. For selected clusters, the average (±SE) signal intensities at each time point were extracted from the intensity normalized images and plotted over time by group; group differences in these time courses were confirmed using repeated measures ANOVA (RMANOVA), a method that accounts for the repeated measures across time.Volumes of Interest Analysis
Volumes of interest (VOI), including those of known cardiorespiratory function, were selected from the functional images using anatomical landmarks, on a subject-by-subject basis. With the use of custom designed software, the average voxel intensity of the VOI in each volume was calculated from the intensity normalized images, resulting in a time trend for each subject. For both VOI and cluster time trends, signal intensity changes were calculated relative to baseline and compared between baseline and challenge periods and between groups at each time point using RMANOVA. Significance was set at P < 0.05. To examine differences in global signal intensity changes between the two groups, the mean signal intensity of each volume was calculated at each time point for each subject and then averaged for each group. Significant differences in global signal intensity between the control and OSA groups at each time point were determined using a standard t-test (P < 0.01). VOI analysis highlighted patterns of response across entire, well-delineated structures, whereas cluster analysis distinguished responses in subregions of one or more structures.| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Physiology
Resting BP was higher in OSA subjects compared with controls (mean BP: controls = 91 ± 3 mmHg, OSA = 102 ± 5 mmHg). During each Valsalva maneuver, control and OSA subjects exerted a similar mean LP (mean LP: control = 42 ± 1 mmHg; OSA = 41 ± 3 mmHg) (Fig. 1A). Although the initial mean HR in the OSA group was significantly higher than that of controls (mean HR: control = 64 ± 1 beats/min; OSA = 74 ± 1 beats/min; t-test, P < 0.01), the patterns of HR responses during each Valsalva maneuver were similar. Heart rate began to increase ~7 s after onset of each effort, reaching a maximum ~3 s after completion of each Valsalva effort (mean HR max: control = 88 ± 5 beats/min; OSA = 92 ± 4 beats/min). Heart rate then declined rapidly below baseline, followed by a slow return toward baseline during the remainder of the 18 s rest period.
|
The VR decreased significantly with age in controls (r = 0.54, df = 44, P = 0.0001) but not in the OSA group (r = 0.12, df = 23, P = 0.59). Age and LP were not correlated in either control (r = 0.02, df = 44, P = 0.88) or OSA subjects (r = 0.04, df = 23, P = 0.87), suggesting that the decline in VR with age did not result from differences in LP (Fig. 1, B and C). The VR, however, did not show a significant group effect (mean VR: controls = 1.89 ± 0.1; OSA = 1.76 ± 0.2; t-test, P > 0.05).
fMRI Signal Changes
The Valsalva maneuver showed very little stimulus-related head motion. In no subject was there any head movement greater than 1.5 mm in the x, y, or z direction, or rotation of >1°. Although in most subjects head movement was greater during the challenge, it did not appear to be directly correlated to the onset or completion of each Valsalva maneuver.Comparable Control and OSA Patterns
The cluster analysis highlighted a number of brain sites that exhibited similar patterns of response in both control and OSA groups. Two regions within the cerebral cortex exhibited significant changes in signal intensity during the Valsalva maneuver: an area of the inferior frontal cortex encompassing Brodmann's areas 45 and 47, corresponding functionally to the supplementary motor cortex supplying upper airway musculature, and a site within the anterior insular cortex (Fig. 2A). Signal intensities within these regions increased significantly during each Valsalva maneuver and returned to below baseline levels in the intervening rest period (Fig. 2B).
|
In deeper brain structures, cluster analysis demonstrated in control
and OSA groups significant changes in signal intensity bilaterally
within the dentate nuclei of the cerebellum. Region-of-interest analysis revealed significant changes in both groups in the dorsal medulla, dorsal pons, midbrain, and amygdala (Figs. 2 and 3). Signal intensities within these regions increased gradually during each
Valsalva maneuver and returned to or below baseline during each rest
period (Fig. 2B). Signal
intensity within the ventral pons also changed significantly, gradually
increasing during the entire challenge period. No significant signal
intensity change emerged within the ventral and medial medullary
regions.
|
Group Differences
There was no significant difference in global signal intensity between the control and OSA groups at any time point. However, cluster analysis revealed multiple brain sites in OSA subjects that showed significantly different patterns of regional signal intensity response to the Valsalva maneuver, compared with controls. Five major regions of difference emerged in the cerebral cortex: 1) an area of the left inferior parietal lobe showed significantly greater increases in signal intensity during each Valsalva maneuver in control, relative to OSA subjects. This region corresponds functionally to Wernicke's area (Brodmann's area 40), which integrates reception of sensory information for expression of speech (Fig. 4, A and B; Ref. 37). 2) A region of the left precentral gyrus showed increased signal change in OSA and no change or a decline in controls during each Valsalva maneuver. This region, corresponding to Brodmann's area 4, supplies musculature of the upper airway and diaphragm (37). 3) Bilateral regions of the anterior superior temporal gyrus (within Brodmann's area 22), showed significantly greater increases in signal intensity in controls compared with OSA subjects. 4) A region of the left frontal cortex, confined to the superior frontal gyrus (Brodmann's area 10), showed signal intensity increases during each Valsalva maneuver that were enhanced in OSA relative to control subjects, although this difference was reduced over subsequent trials. 5) A portion of the posterior insular cortex exhibited significant increases in signal intensity bilaterally in controls but little change in OSA subjects during each maneuver. This difference was greater on the left side than the right (Fig. 5; Table 2).
|
|
|
Regional differences also emerged in deeper brain structures. Within the cerebellum, the fastigial nucleus and a discrete region within the quadrangular lobule of the cerebellar cortex showed significant changes in signal intensity in controls and little change in signal in the OSA subjects during each Valsalva maneuver. Enhanced responses in OSA subjects also appeared in the left anterior cingulate gyrus (Fig. 5), and region-of-interest analysis revealed diminished responses bilaterally in the hippocampus (Fig. 3).
Time course patterns. In a number of regions, signal intensity changes in OSA subjects differed from controls in their time course pattern as well as in magnitude. The response in the cerebellar cortex of OSA subjects was initially inverse to that of the controls and subsequently remained elevated while the controls responded in a cyclic fashion during each Valsalva maneuver (Fig. 5). During the initial Valsalva maneuver, there was no signal change in the posterior insula, after which it followed a pattern similar to controls but with diminished magnitude (Fig. 5). In the anterior cingulate, the OSA response initially led the control response, as did responses from the left precentral gyrus (Figs. 4B and 5).
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Subjects with OSA showed altered neural responses to the Valsalva maneuver in several regions where gray matter loss had been previously demonstrated (24). The affected areas included sites involved in motor control of diaphragmatic and upper airway muscles; sensory integration from the oral airway, cerebellar cortical, and nuclear sites involved in blood pressure and breathing control; and limbic structures that can exert roles in inspiratory onset and blood pressure modulation. These findings suggest ineffective responses from selected structures normally mediating the challenge, combined with recruitment of alternative areas that, we speculate, are attempting to mount an effective physiological output. The functional reorganization of neural mechanisms found in OSA patients responding to the Valsalva maneuver may play a significant role in mediating the cardiovascular and respiratory patterns found during obstructed breathing within sleep.
Physiology
The Valsalva maneuver leads to a number of well-characterized autonomic changes (5, 19). At the beginning of the strain, AP increases transiently due to the increase in intrathoracic pressure (phase I). However, as the increase in intrathoracic pressure continues, atrial filling falls and AP declines, resulting in a baroreceptor mediated increase in HR (phase II); at this time, a significant increase in sympathetic tone and circulating catecholamines moves net arterial pressure toward baseline. At the completion of the maneuver, the sudden decrease in intrathoracic pressure results in a transient fall in AP (phase III), followed by an overshoot in AP as sympathetic tone remains elevated after intrathoracic pressure has returned to normal (phase IV). The magnitude of the increase in HR during phase II compared with the fall in HR during phase III (Valsalva ratio) has been shown to give a measure of autonomic functioning (21). OSA patients exhibit a number of autonomic deficiencies, including significantly lower VR (11), decreased overall HR variability, and abnormal HR variability to an orthostatic maneuver (41). Furthermore, OSA subjects exhibit high sympathetic activity during wakefulness (45) and, possibly as a consequence, have a greater propensity for systemic arterial hypertension (27); the present group of OSA subjects also showed higher AP. The VR of the OSA subjects presented in this study did not decrease with age to a statistically significant degree. Our OSA subjects ranged in age from 28 to 63, a range that would not be expected to show a sufficiently large variation to be significant with only eight subjects (21).fMRI Signal Similarities
The Valsalva maneuver evoked similar signal intensity changes in control and OSA subjects in a number of neural regions. These areas included the dentate nucleus of the cerebellum, dorsal brainstem, amygdala, anterior insular cortex, lateral prefrontal cortex, and lentiform nucleus. These regions all responded in a pattern that followed the LP, and signals increased gradually over the course of the Valsalva maneuver. The ventral pons showed a gradual increase in signal intensity over the entire challenge period without a return to baseline between separate maneuvers. Respiratory and cardiovascular functions of the neural regions responding to the Valsalva maneuver in healthy subjects, including those areas just described, have been discussed elsewhere (14).Decreased Responsiveness in OSA Subjects
Six regions exhibited decreased responsiveness in OSA subjects, relative to controls. These areas were the left inferior parietal cortex, superior temporal gyrus, posterior insular cortex, cerebellar cortex, fastigial nucleus, and hippocampus.The inferior parietal cortex, which partially corresponds to Wernicke's area, integrates sensory input from the head and upper airway. A one-third reduction in two-point discrimination in the oropharynx and an 82% higher vibratory threshold occur in the upper airway of OSA patients compared with controls (17). The diminished recruitment of this upper airway sensory integration area during the Valsalva maneuver may reflect a dysfunction of this region in OSA patients. Impairment of afferent information of mechanical stimulation of the upper airway can lead to diminished airway muscle tone, because stimulation by airflow pressure or vibration has long been known to assist airway patency (38).
The superior temporal gyrus also exhibited decreased responsiveness compared with controls. This cortical area is a complex region involved in a number of functions, including language production and word recognition (15), and the left side has been previously implicated in voluntary expiratory efforts (7), findings comparable to those described here. Swallowing, a task requiring integration of sensory and motor aspects of the upper airway, also activates the superior temporal gyrus (25); a high level of integration is also required by the Valsalva maneuver. A reduction in afferent input or integration within the superior temporal gyrus may contribute to the prolongation of apnea in OSA.
The posterior insular cortex showed diminished responsiveness bilaterally to the challenge in OSA subjects. The insular cortex plays a significant role in autonomic regulation (42). Lesions of the posterior insula interfere with baroreceptor functioning, a result not apparent from anterior insular lesions (54). Parasympathetic and sympathetic modulation appear to be lateralized in the insula, with parasympathetic related control principally sited in the left side and sympathetic in the right (29). Lesions of the left insular cortex of humans enhance basal sympathetic tone (30). The decreased reactivity found during the Valsalva maneuver was greater on the left side. Impaired insular function may underlie the higher sympathetic tone reported in OSA (45), and the higher heart rate found here. The near absence of posterior insular responses in OSA subjects during the first Valsalva maneuver may delay blood pressure changes during apneic periods. The close interaction of blood pressure and breathing patterns provides a potential mechanism to alter the time course of apneic events (47).
The Valsalva maneuver requires considerable respiratory effort; dyspnea can occur near completion of each exertion. Air hunger has been shown to recruit, among other structures, the insular cortex (1, 35). If the insula signal increase during the Valsalva challenge is associated with dyspnea, the signal changes in this region would be expected to increase principally near the completion of each maneuver. Because the signal intensity increases were gradual and occurred over the entire maneuver period, a pattern similar to the HR response, the insular signal changes and the attenuation of the posterior insular response in OSA subjects likely reflect autonomic outflow roles, rather than air hunger.
Although the dentate cerebellar nuclei showed similar signal intensity pattern changes during each Valsalva maneuver in both control and OSA groups, activity in the fastigial nucleus and a region of the right quadrangular lobule in the cerebellar cortex did not. The region of decreased responsiveness in the cerebellar cortex mirrors an area previously described as exhibiting gray matter loss in OSA subjects (24).
The motor regulatory role of the cerebellum extends to the respiratory musculature (52) and modulation of large losses in blood pressure (22), a role particularly related to portions of the fastigial nucleus (23). The cerebellar cortex receives afferent input from climbing fibers that originate in the inferior olivary nucleus. These cerebellar climbing fibers are very sensitive to ischemic damage, and the hypoxia accompanying repeated apnea events during OSA may cause such damage. The unique metabolic characteristics of climbing fiber bundles allow the possibility for highly restricted areas of cerebellar cortex damage (50). Output of the cerebellar cortex is inhibitory and expressed via the deep cerebellar nuclei, and damage would normally disinhibit fastigial nuclei action impinging on thalamic and brainstem sites. The signal changes in the cerebellar cortex were both diminished and phase shifted compared with controls, offering the potential for excitatory action from deep cerebellar nuclei to arrive at inappropriate times on effector sites. The specific role for interactions between cerebellar and sensory and motor sites found to be altered in function in this study awaits further investigation with animal models. The potential for these cerebellar sites to contribute to both the respiratory or autonomic aspects of OSA is substantial.
Areas within the hippocampus that showed muted signals in OSA overlapped areas of gray matter loss (24). The hippocampus contains neurons that discharge in a respiratory cycle-dependent fashion (9) and show marked activity pattern changes to sigh-apnea sequences (39). The contribution from hippocampal sources likely relate to restoration of inspiratory onset after apnea.
Increased Responsiveness in OSA Subjects
The region of the cortical precentral gyrus showing altered responsiveness in OSA subjects serves the diaphragm and upper airway musculature. Not only did the response in this region differ in OSA subjects, but it was the only site in which the direction of signal change was opposed to that of controls. During expiration, the upper airway is fully dilated due to increased intraluminal pressure, and upper airway dilator muscles are largely inactive (51). Similarly, the upper airway is fully dilated during the Valsalva maneuver, and this dilation is enhanced in OSA (43). The decrease in signal intensity in the motor cortex of controls may underlie the decline in upper airway dilator activity during the Valsalva maneuver. The increased signal in the OSA subjects may reflect an overcompensation of the upper airway musculature, resulting in the enhanced upper airway dilation observed in these subjects. Alternatively, the increase in signal intensity may represent an overcompensation of drive to the diaphragm to overcome a motor system that is ineffective during difficult resistive challenges.Substantial evidence demonstrates that the anterior cingulate plays a significant role in respiration and vocalization associated with affect and, in particular, upper airway musculature control (49). Vocalization requires exquisite integration of upper airway and diaphragmatic musculature to obtain a sequence of enhanced thoracic pressure through inspiratory effort followed by expiratory airflow. A critical aspect of OSA is initiation of inspiration after prolonged cessation of airflow. The region of delayed and opposite reactivity of the left anterior cingulate overlapped the area of previously demonstrated gray matter loss in OSA subjects (24); we speculate that the increased responses within this area represent a compensatory effort by neuronal inputs to overcome the anatomic loss in this region.
The anterior cingulate markedly reacts to several different blood pressure challenges, as demonstrated by fMRI evidence in both animals and humans (12, 18, 53); elicits large changes in blood pressure and HR on stimulation (2); and contains single neurons that fire in a discharge pattern related to the respiratory and cardiac cycle (9). The fMRI signal increases in the anterior cingulate of OSA subjects, with relatively little change in control subjects, may reflect an attempt in OSA subjects to overcome loss of normal neural function in other brain regions to mediate an almost-adequate HR response to changes in blood pressure.
The increased responses in the left superior frontal gyrus may be related to the enhanced signal changes found in the anterior cingulate. Transcranial stimulation near the superior frontal gyrus recruits the anterior cingulate (34). A cortical region bordering the superior frontal gyrus responsive site has also been implicated in aspects of working memory (31). It may be the case that the increased responses in the cortical areas reflect an "intention" aspect of recruitment of upper airway muscles, complementing a role that has been proposed for other motor actions from the anterior cingulate (33).
Limitations
Global signal changes. The issue of global signal changes with blood pressure manipulation is frequently overlooked in fMRI studies, especially in studies of neural signal responses to pain; pain of either deep or superficial origin evokes profound changes in AP, much larger than those observed in the Valsalva maneuver. During phases I and IV of the Valsalva maneuver, AP increases 5-10%, and in early phase II and during phase III AP declines ~5% (5). Studies investigating the effects of changes in AP show that changes of 20 to 60 mmHg have no significant effect on regional blood oxygen level-dependent signal changes in a number of selected brain regions (52). Large global changes in signal intensity induced by hypercapnia can be removed by intensity normalization to reveal regional changes related to a simultaneous stimulus (4, 16). It is assumed that these hypercapnia-induced increases result from the action of CO2 on cerebral vasculature. However, AP and sympathetic activity increase substantially during brief periods of hypercapnia. Narkiewicz et al. (26) report that AP increased ~15% during a 3 min period breathing 7% CO2, and Edwards et al. (6) show that this AP increase was up to 30% in women during certain times in the menstrual cycle. Although AP was not measured, given the high levels of CO2 and the long duration exposure, subjects in both the Corfield et al. (4) and Kemna and Posse (16) studies likely experienced substantial increases in AP and sympathetic drive; thus the increases in global cerebral blood flow reported likely result from a combination of increased AP as well as the direct effects of CO2 on cerebral vasculature. The effectiveness of intensity normalization used in those studies is likely shared by procedures in this study and can account for any changes in global intensity from increases in AP and direct effects of CO2 on cerebral vasculature.
The Valsalva maneuver evokes a decline in overall cerebral blood flow (46) and therefore a decline in global signal intensity. If the autoregulatory system differed between the groups, the magnitude of the changes in global signal intensity should also differ. Although OSA is generally associated with cardiovascular dysfunction, in this study, both the control and OSA groups exhibited similar changes in global signal intensity during the Valsalva maneuvers. The effectiveness of intensity normalization is supported by the finding that most of the regional signal intensity changes during the Valsalva maneuver were located in previously well-defined cardiovascular or respiratory related regions, and some were unilaterally represented. Furthermore, because the overall changes in signal intensity did not differ between OSA and control subjects, differences between the two groups can be inferred to result from differences in the regional blood flow resulting from changes in synaptic activity.Temporal resolution. The repetition time in this study precluded an examination of the rapid phases of the Valsalva maneuver, particularly phases I and III. Alterations in sympathetic tone and the circulating catecholamines endure for up to several minutes (21). Our findings almost certainly reflect a temporal summation of central mechanisms of withdrawal of vagal influences, activation of sympathetic drive to adrenals and peripheral vasculature, as well as increased cardiac rate and stroke volume.
We used an analytic model that tracks load pressure, the relevant stimulus for central mediation of the reflex. The load pressure stimulus is inevitably confounded with expiratory muscle activation but is coincident with activation of central nervous systems mediating autonomic responses, even though autonomic responses may take a period of time to develop. Although the humoral and vascular bed responses require a finite period of time to develop into the late phase II of the Valsalva maneuver, central initiation of these responses is likely quite rapid. Therefore, even though the peripheral effects may outlast the intrathoracic pressure and be out of phase with it, many central components of interest in this study could be imaged, despite the coarse temporal resolution and abbreviated phase IV of our protocol. Regions that show enhanced signal that closely follows the time course of the thoracic pressure signal are unlikely to be responding to either the initial transient neuronal responses or the more enduring autonomic features of the reflex. This is the first attempt to measure regional brain changes in an OSA patient population during this common autonomic test; because the sites of participating regions were unknown, it was important to examine the entire brain. Subsequent studies using higher field strength imaging, with improved spatial and temporal resolution, will allow examination of more rapid and transient neural participation in the Valsalva maneuver. In conclusion, although a number of neural sites responded in a similar fashion to the Valsalva maneuver in both OSA and control subjects, particular sites showed differing response patterns in OSA subjects not medicated with cardiovascular or mood-altering agents. In certain regions, signal responses to the Valsalva maneuver increased in OSA over controls; in other areas, responses in OSA subjects were substantially lower or phase-shifted. In several regions, areas of signal pattern differences overlapped sites of previously demonstrated gray matter loss in OSA subjects. The increased signal change in particular structures suggests a compensatory response to the aberrant integrated neural organization in OSA. Areas of altered responsivity have the neuroanatomic and functional properties to mediate significant sensory or motor integration or coordination perspectives. Deficiencies in sensory reception from the airway, deficits in initiating inspiratory efforts of the upper airway, or phase alterations in initiating blood pressure responses may mediate the respiratory muscle patterns found in OSA.| |
ACKNOWLEDGEMENTS |
|---|
We thank Rebecca Harper for technical assistance.
| |
FOOTNOTES |
|---|
This research was supported by National Heart, Lung, and Blood Institute Grant HL-60296.
Address for reprint requests and other correspondence: R. M. Harper, Dept. of Neurobiology, Univ. of California at Los Angeles, Los Angeles, CA 90095-1763 (E-mail: rharper{at}ucla.edu).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published November 1, 2002;10.1152/japplphysiol.00702.2002
Received 30 July 2002; accepted in final form 24 October 2002.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Banzett, RB,
Mulnier HE,
Murphy K,
Rosen SD,
Wise RJ,
and
Adams L.
Breathlessness in humans activates insular cortex.
Neuroreport
11:
2117-2120,
2000[ISI][Medline].
2.
Burns, SM,
and
Wyss JM.
The involvement of the anterior cingulate cortex in blood pressure control.
Brain Res
340:
71-77,
1985[ISI][Medline].
3.
Chervin, RD,
and
Guilleminault C.
Obstructive sleep apnea and related disorders.
Neurol Clin
14:
583-609,
1996[Medline].
4.
Corfield, DR,
Murphy K,
Josephs O,
Adams L,
and
Turner R.
Does hypercapnia-induced cerebral vasodilation modulate the hemodynamic response to neural activation?
Neuroimage
13:
1207-1211,
2001[ISI][Medline].
5.
Dawson, SL,
Panerai RB,
and
Potter JF.
Critical closing pressure explains cerebral hemodynamics during the Valsalva maneuver.
J Appl Physiol
86:
675-680,
1999
6.
Edwards, N,
Wilcox I,
Polo OJ,
and
Sullivan CE.
Hypercapnic blood pressure response is greater during the luteal phase of the menstrual cycle.
J Appl Physiol
81:
2142-2146,
1996
7.
Evans, KC,
Shea SA,
and
Saykin AJ.
Functional MRI localisation of central nervous system regions associated with volitional inspiration in humans.
J Physiol
520:
383-392,
1999
8.
Friston, KJ,
Holmes AP,
Poline JB,
Grasby PJ,
Williams SC,
Frackowiak RS,
and
Turner R.
Analysis of fMRI time-series revisited.
Neuroimage
2:
45-53,
1995[ISI][Medline].
9.
Frysinger, RC,
and
Harper RM.
Cardiac and respiratory relationships with neural discharge in the anterior cingulate cortex during sleep-walking states.
Exp Neurol
94:
247-263,
1986[ISI][Medline].
10.
Garrigue, S,
Bordier P,
Jais P,
Shah DC,
Hocini M,
Raherison C,
Tunon De Lara M,
Haissaguerre M,
and
Clementy J.
Benefit of atrial pacing in sleep apnea syndrome.
N Engl J Med
346:
404-412,
2002
11.
Hanly, PJ,
George CF,
Millar TW,
and
Kryger MH.
Heart rate response to breath-hold, valsalva and Mueller maneuvers in obstructive sleep apnea.
Chest
95:
735-739,
1989
12.
Harper, RM,
Bandler R,
Spriggs D,
and
Alger JR.
Lateralized and widespread brain activation during transient blood pressure elevation revealed by magnetic resonance imaging.
J Comp Neurol
417:
195-204,
2000[ISI][Medline].
13.
Harper, RM,
Parker JM,
Frysinger RC,
Henderson LA,
del Vecchio W,
and
Woo MA.
Infrared transfer of electrophysiologic signals during magnetic resonance imaging.
Sleep Research Online
4:
13-15,
2001.
14.
Henderson, LA,
Macey PM,
Macey KE,
Frysinger RC,
Woo MA,
Harper RK,
Alger JR,
Yan-Go FL,
and
Harper RM.
Brain responses associated with the Valsalva maneuver revealed by functional magnetic resonance imaging.
J Neurophysiol
88:
3477-3486,
2002
15.
Herholz, K,
Pietrzyk U,
Karbe H,
Wurker M,
Wienhard K,
and
Heiss WD.
Individual metabolic anatomy of repeating words demonstrated by MRI-guided positron emission tomography.
Neurosci Lett
182:
47-50,
1994[ISI][Medline].
16.
Kemna, LJ,
and
Posse S.
Effect of respiratory CO2 changes on the temporal dynamics of the hemodynamic response in functional MR imaging.
Neuroimage
14:
642-649,
2001[ISI][Medline].
17.
Kimoff, RJ,
Sforza E,
Champagne V,
Ofiara L,
and
Gendron D.
Upper airway sensation in snoring and obstructive sleep apnea.
Am J Respir Crit Care Med
164:
250-255,
2001
18.
King, AB,
Menon RS,
Hachinski V,
and
Cechetto DF.
Human forebrain activation by visceral stimuli.
J Comp Neurol
413:
572-582,
1999[ISI][Medline].
19.
Levin, AB.
A simple test of cardiac function based on heart rate changes induced by the Valsalva maneuver.
Am J Cardiol
18:
90-99,
1966[ISI][Medline].
20.
Low, PA.
Diabetic autonomic neuropathy.
Semin Neurol
16:
143-151,
1996[ISI][Medline].
21.
Low, PA.
Clinical Autonomic Disorders: Evaluation and Management. Philadelphia, PA: Lippincott-Raven, 1997.
22.
Lutherer, LO,
Lutherer BC,
Dormer KJ,
Janssen HF,
and
Barnes CD.
Bilateral lesions of the fastigial nucleus prevent the recovery of blood pressure following hypotension induced by hemorrhage or administration of endotoxin.
Brain Res
269:
251-257,
1983[ISI][Medline].
23.
Lutherer, LO,
Williams JL,
and
Everse SJ.
Neurons of the rostral fastigial nucleus are responsive to cardiovascular and respiratory challenges.
J Auton Nerv Syst
27:
101-111,
1989[ISI][Medline].
24.
Macey, PM,
Henderson LA,
Macey KE,
Alger JR,
Frysinger RC,
Woo MA,
Harper RK,
Yan-Go FL,
and
Harper RM.
Brain morphology associated with obstructive sleep apnea.
Am J Resp Crit Care Med
166:
1382-1387,
2002
25.
Martin, RE,
Goodyear BG,
Gati JS,
and
Menon RS.
Cerebral cortical representation of automatic and volitional swallowing in humans.
J Neurophysiol
85:
938-950,
2001
26.
Narkiewicz, K,
Pesek CA,
van de Borne PJ,
Kato M,
and
Somers VK.
Enhanced sympathetic and ventilatory responses to central chemoreflex activation in heart failure.
Circulation
100:
262-267,
1999
27.
Narkiewicz, K,
and
Somers VK.
The sympathetic nervous system and obstructive sleep apnea: implications for hypertension.
J Hypertens
15:
1613-1619,
1997[ISI][Medline].
28.
Netten, PM,
de Vos K,
Horstink MW,
and
Hoefnagels WH.
Autonomic dysfunction in Parkinson's disease, tested with a computerized method using a Finapres device.
Clin Auton Res
5:
85-89,
1995[ISI][Medline].
29.
Oppenheimer, SM,
Gelb A,
Girvin JP,
and
Hachinski VC.
Cardiovascular effects of human insular cortex stimulation.
Neurology
42:
1727-1732,
1992
30.
Oppenheimer, SM,
Kedem G,
and
Martin WM.
Left-insular cortex lesions perturb cardiac autonomic tone in humans.
Clin Auton Res
6:
131-140,
1996[ISI][Medline].
31.
Owen, AM.
The role of the lateral frontal cortex in mnemonic processing: the contribution of functional neuroimaging.
Exp Brain Res
133:
33-43,
2000[ISI][Medline].
32.
Parker, JM,
Alger JR,
Woo MA,
Spriggs D,
and
Harper RM.
Acquisition of electrophysiologic signals during magnetic resonance imaging.
Sleep
22:
1125-1126,
1999[ISI][Medline].
33.
Paus, T.
Primate anterior cingulate cortex: where motor control, drive and cognition interface.
Nat Rev Neurosci
2:
417-424,
2001[ISI][Medline].
34.
Paus, T,
Castro-Alamancos MA,
and
Petrides M.
Cortico-cortical connectivity of the human mid-dorsolateral frontal cortex and its modulation by repetitive transcranial magnetic stimulation.
Eur J Neurosci
14:
1405-1411,
2001[ISI][Medline].
35.
Peiffer, C,
Poline JB,
Thivard L,
Aubier M,
and
Samson Y.
Neural substrates for the perception of acutely induced dyspnea.
Am J Respir Crit Care Med
163:
951-957,
2001
36.
Pendlebury, ST,
Pepin JL,
Veale D,
and
Levy P.
Natural evolution of moderate sleep apnoea syndrome: significant progression over a mean of 17 months.
Thorax
52:
872-878,
1997[Abstract].
37.
Penfield, W,
and
Roberts L.
Speech and Brain-Mechanisms. Princeton, NJ: Princeton Univ. Press, 1959.
38.
Pillar, G,
Fogel RB,
Malhotra A,
Beauregard J,
Edwards JK,
Shea SA,
and
White DP.
Genioglossal inspiratory activation: central respiratory vs mechanoreceptive influences.
Respir Physiol
127:
23-38,
2001[ISI][Medline].
39.
Poe, GR,
Kristensen MP,
Rector DM,
and
Harper RM.
Hippocampal activity during transient respiratory events in the freely behaving cat.
Neuroscience
72:
39-48,
1996[ISI][Medline].
40.
Polysomnography Task Force, American Sleep Disorders Association Standards of Practice Committee.
Practice parameters for the indications for polysomnography and related procedures.
Sleep
20:
406-422,
1997[ISI][Medline].
41.
Salo, TM,
Jula AM,
Piha JS,
Kantola IM,
Pelttari L,
Rauhala E,
Metsala TH,
Jalonen JO,
Voipio-Pulkki LM,
and
Viikari JS.
Comparison of autonomic withdrawal in men with obstructive sleep apnea syndrome, systemic hypertension, and neither condition.
Am J Cardiol
85:
232-238,
2000[ISI][Medline].
42.
Saper, CB.
Convergence of autonomic and limbic connections in the insular cortex of the rat.
J Comp Neurol
210:
163-173,
1982[ISI][Medline].
43.
Schwab, RJ,
Gefter WB,
Hoffman EA,
Gupta KB,
and
Pack AI.
Dynamic upper airway imaging during awake respiration in normal subjects and patients with sleep disordered breathing.
Am Rev Respir Dis
148:
1385-1400,
1993[ISI][Medline].
44.
Shepard, JW, Jr.
Cardiopulmonary consequences of obstructive sleep apnea.
Mayo Clin Proc
65:
1250-1259,
1990[ISI][Medline].
45.
Somers, VK,
Dyken ME,
Clary MP,
and
Abboud FM.
Sympathetic neural mechanisms in obstructive sleep apnea.
J Clin Invest
96:
1897-1904,
1995[ISI][Medline].
46.
Tiecks, FP,
Lam AM,
Matta BF,
Strebel S,
Douville C,
and
Newell DW.
Effects of the valsalva maneuver on cerebral circulation in healthy adults. A transcranial Doppler study.
Stroke
26:
1386-1392,
1995
47.
Trelease, RB,
Sieck GC,
Marks JD,
and
Harper RM.
Respiratory inhibition induced by transient hypertension during sleep in unrestrained cats.
Exp Neurol
90:
173-186,
1985[ISI][Medline].
48.
Veale, D,
Pepin JL,
Wuyam B,
and
Levy PA.
Abnormal autonomic stress responses in obstructive sleep apnoea are reversed by nasal continuous positive airway pressure.
Eur Respir J
9:
2122-2126,
1996[Abstract].
49.
Von Cramon, D,
and
Jurgens U.
The anterior cingulate cortex and the phonatory control in monkey and man.
Neurosci Biobehav Rev
7:
423-425,
1983[ISI][Medline].
50.
Welsh, JP,
Yuen G,
Placantonakis DG,
Vu TQ,
Haiss F,
O'Hearn E,
Molliver ME,
and
Aicher SA.
Why do Purkinje cells die so easily after global brain ischemia? Aldolase C, EAAT4, and the cerebellar contribution to posthypoxic myoclonus.
Adv Neurol
89:
331-359,
2002[ISI][Medline].
51.
Wiegand, DA,
Latz B,
Zwillich CW,
and
Wiegand L.
Upper airway resistance and geniohyoid muscle activity in normal men during wakefulness and sleep.
J Appl Physiol
69:
1252-1261,
1990
52.
Xu, F,
and
Frazier DT.
Medullary respiratory neuronal activity modulated by stimulation of the fastigial nucleus of the cerebellum.
Brain Res
705:
53-64,
1995[ISI][Medline].
53.
Yu PL, Henderson LA, Frysinger RC, Galons JP, Bandler R, and Harper
RM. Functional magnetic resonance imaging (fMRI) of feline brain
during depressor challenges. Soc Neurosci Abstr, 2001, A690.8.
54.
Zhang, ZH,
Rashba S,
and
Oppenheimer SM.
Insular cortex lesions alter baroreceptor sensitivity in the urethane-anesthetized rat.
Brain Res
813:
73-81,
1998[ISI][Medline].
This article has been cited by other articles:
|
|
 |
|
  D. S. Kimmerly, D. D. O'Leary, R. S. Menon, J. S. Gati, and J. K. Shoemaker Cortical regions associated with autonomic cardiovascular regulation during lower body negative pressure in humans J. Physiol., November 15, 2005; 569(1): 331 - 345. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. von Leupoldt and B. Dahme Cortical Substrates for the Perception of Dyspnea Chest, July 1, 2005; 128(1): 345 - 354. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Harper, P. M. Macey, M. A. Woo, K. E. Macey, T. G. Keens, D. Gozal, and J. R. Alger Hypercapnic Exposure in Congenital Central Hypoventilation Syndrome Reveals CNS Respiratory Control Mechanisms J Neurophysiol, March 1, 2005; 93(3): 1647 - 1658. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. E. Macey, P. M. Macey, M. A. Woo, R. K. Harper, J. R. Alger, T. G. Keens, and R. M. Harper fMRI signal changes in response to forced expiratory loading in congenital central hypoventilation syndrome J Appl Physiol, November 1, 2004; 97(5): 1897 - 1907. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Woo, P. M. Macey, G. C. Fonarow, M. A. Hamilton, and R. M. Harper Regional brain gray matter loss in heart failure J Appl Physiol, August 1, 2003; 95(2): 677 - 684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Harper, P. M. Macey, L. A. Henderson, M. A. Woo, K. E. Macey, R. C. Frysinger, J. R. Alger, K. P. Nguyen, and F. L. Yan-Go fMRI responses to cold pressor challenges in control and obstructive sleep apnea subjects J Appl Physiol, April 1, 2003; 94(4): 1583 - 1595. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||
significant differences (P < 0.05) between
control and OSA groups.
