SEARCH

SEARCH BY CITATION

Keywords:

  • fMRI;
  • microneurography;
  • muscle sympathetic nerve activity;
  • baroreflex;
  • medulla

Abstract

  1. Top of page
  2. Abstract
  3. BAROREFLEX CIRCUITRY IN EXPERIMENTAL ANIMALS
  4. BAROREFLEX CIRCUITRY IN AWAKE HUMANS
  5. CONCURRENT MSNA AND FMRI RECORDING
  6. BRAINSTEM ACTIVITY CHANGES DURING SPONTANEOUS FLUCTUATIONS IN MSNA
  7. CORTICAL SITES REGULATING MSNA IN AWAKE HUMANS
  8. LITERATURE CITED

The sympathetic division of the nervous system is critical for maintaining both resting arterial pressure and for producing changes in regional perfusion required by behavioral state changes. A primary determinant of arterial pressure is the level of vasoconstriction within skeletal muscle. It is well established that there is a tight relationship between dynamic changes in arterial pressure and muscle sympathetic nerve activity (MSNA) through the workings of the baroreflex. While the central circuitry underlying the baroreflex has been extensively investigated in anesthetized experimental animals, few studies have investigated the central circuitry responsible for the baroreflex in awake human subjects. Recently we were the first to record concurrently MSNA (using microneurography) and brain activity (using functional magnetic resonance imaging) in awake humans in a series of experiments designed to determine the central circuitry underlying the baroreflex in humans. We confirmed that the baroreflex involves activity changes within the nucleus tractus solitarius, caudal ventrolateral, and rostral ventrolateral medulla. Because conditions such as essential hypertension, obesity, and obstructive sleep apnea are all characterized by significant increases in resting MSNA, it is important to understand both brainstem and cortical sites involved in regulating resting levels of MSNA. Future investigations which define cortical sites involved in generating and modulating MSNA are important if we are to understand the underlying mechanisms of many conditions characterized by hypertension. Anat Rec, 2012. © 2012 Wiley Periodicals, Inc.

The sympathetic division of the nervous system controls a diverse range of organs which are vital for homeostasis. In addition to being critical for maintaining homeostasis, the sympathetic nervous system is mobilized during behavioral changes such as during fight/flight defensive behaviors. A primary determinant of arterial pressure is the level of sympathetically mediated vasoconstriction within skeletal muscle. Indeed, the level of muscle sympathetic nerve activity (MSNA) is extremely consistent, that is, microneurographic studies have shown that resting MSNA, measured as either burst frequency or burst incidence is seemingly identical in a given individual from day to day, year to year (Fagius and Wallin,1993). Although consistent within an individual, between individuals, MSNA can vary greatly, ranging from ∼5 to over 100 bursts per 100 heart beats in some individuals (Fagius and Wallin,1993; Charkoudian et al.,2005).

Whilst resting MSNA is significantly related to cardiac output (Charkoudian et al.,2005), MSNA is not tightly related to resting arterial pressure. Hence the level of MSNA cannot be predicted from an individual's resting arterial pressure or heart rate alone (Joyner et al.,2008). This may result from the fact that the relationship between MSNA and arterial pressure is influenced by a number of factors, including age, hypoxia, exercise and hormones (Ng et al.,1993; Minson et al.,2000; Halliwill and Minson,2002). Although there is not a significant relationship between resting MSNA and resting arterial pressure, it is well established that there is a tight relationship between dynamic changes in arterial pressure and MSNA, that is, through the workings of the baroreflex. Spontaneous variations in arterial pressure are detected by high-pressure arterial baroreceptors in the aortic arch and carotid sinuses, which then evoke opposing changes in MSNA, which act to cushion the original change in pressure. This homeostatic mechanism is designed to maintain resting arterial pressure at a stable level and prevent sudden, sizeable changes in arterial pressure, such as those that would occur, for example, during sitting and standing.

While the central circuitry underlying the baroreflex has been extensively investigated in anesthetized experimental animals, few studies have investigated the central circuitry responsible for the baroreflex in awake human subjects. Some investigators have indirectly measured human baroreflex functioning by using maneuvers such as lower body negative pressure (Kimmerly et al.,2005) and maximal inspiratory apneas (Macefield et al.,2006). Although these investigations are valuable, they did not measure brain activity and MSNA concurrently, and further, did not investigate brain circuits responsible for the small fluctuations in resting MSNA that result from baroreceptor activation. We were the first to record concurrently MSNA (using microneurography) and brain activity (using functional magnetic resonance imaging) in awake humans in a series of experiments designed to determine the central circuitry underlying the baroreflex in humans (Macefield and Henderson,2010). Because conditions such as essential hypertension (Grassi et al.,1998), congestive heart failure (Floras,2003), and obstructive sleep apnea (Carlson et al.,1993) are all characterized by significant increases in resting MSNA, it is important to understand both brainstem and cortical sites involved in regulating resting levels of MSNA.

BAROREFLEX CIRCUITRY IN EXPERIMENTAL ANIMALS

  1. Top of page
  2. Abstract
  3. BAROREFLEX CIRCUITRY IN EXPERIMENTAL ANIMALS
  4. BAROREFLEX CIRCUITRY IN AWAKE HUMANS
  5. CONCURRENT MSNA AND FMRI RECORDING
  6. BRAINSTEM ACTIVITY CHANGES DURING SPONTANEOUS FLUCTUATIONS IN MSNA
  7. CORTICAL SITES REGULATING MSNA IN AWAKE HUMANS
  8. LITERATURE CITED

The neural circuitry responsible for the baroreflex has been extensively investigated over many decades in anesthetized experimental animals. Early work in anesthetized cats revealed a region in the rostroventrolateral medulla (RVLM) that, upon activation, evoked significant increases in muscle or cutaneous sympathetic drive. Excitation of some RVLM sites evoked increases in muscle sympathetic nerve activity and profound increases in arterial pressure, whilst activation at other RVLM sites evoked increases in cutaneous sympathetic nerve activity without changes in arterial pressure (Dampney and McAllen,1988; McAllen et al.,1995). Furthermore, activity of the RVLM appears to be critical for the maintenance of resting vasomotor tone and arterial pressure, since RVLM destruction results in precipitous falls in resting arterial pressure (Dampney and Moon,1980). In addition to maintaining basal tone and arterial pressure, intracellular recordings from RVLM neurons in anesthetized experimental animals have revealed that they are critical for the expression of the baroreceptor reflex. The firing rates of RVLM neurons decrease with “peripherally induced” increases in arterial pressure, for example following intravenous injections of phenylepherine (Agarwal et al.,1990). Furthermore, the inhibitory response of RVLM neurons to phenylepherine is blocked by chemical inactivation of a region caudal to the RVLM, that is, the caudal ventrolateral medulla (CVLM). In addition, chemical inhibition of the CVLM attenuates the inhibitory response of RVLM neurons to stimulation of a third medullary region, the nucleus tractus solitarius (NTS) (Agarwal et al.,1990).

These early experiments, along with many subsequent investigations in anesthetized experimental animals have defined what is now known as the baroreflex circuit. This circuit consists of primary afferent axons from high-pressure baroreceptors located in the aortic arch and carotid sinus baroreceptors, projecting to the caudal region of the NTS. Within the caudal NTS, these primary afferents synapse onto second-order neurons which send excitatory (glutamatergic) projections onto GABAergic neurons within the region of the CVLM. These CVLM GABAergic neurons synapse directly onto excitatory RVLM neurons and serve to inhibit the spontaneous activity of RVLM premotor sympathetic neurons (Dampney,1994). In addition to this well-defined circuit, it is now known that the nucleus ambiguus and dorsal motor nucleus of the vagus within the lateral and dorsal medulla, respectively, also receive glutamatergic projections from NTS (Wang et al.,2001). This projection acts to influence heart rate via vagal cardiac efferent's which, upon activation, result in a cholinergically mediated slowing of the heart. Studies in conscious experimental animals, using c-fos expression as a marker of neuronal activation, have confirmed the operation of the NTS-CVLM-RVLM pathway during maneuvers that alter arterial pressure (Minson et al.,1997; Dampney et al.,2003).

In addition to these components of the basic baroreflex arc, it is known that other brainstem and higher brain regions can influence the sensitivity of the baroreflex. For example, during defensive behaviors such as fight/flight, increases in arterial pressure and heart rate are essential to support the required changes in muscle activity. In this instance, the brain region responsible for expression of flight/fight behaviors, the midbrain periaqueductal gray matter, also inhibits baroreflex activity so that the required cardiovascular changes are not constantly reduced by baroreflex activity (Nosaka et al.,1993; Bandler and Keay,1996). This relatively simple example highlights the fact that the operational midpoint, as well as the sensitivity of the baroreflex, is being constantly monitored and altered in response to changes in an animal's behavior.

BAROREFLEX CIRCUITRY IN AWAKE HUMANS

  1. Top of page
  2. Abstract
  3. BAROREFLEX CIRCUITRY IN EXPERIMENTAL ANIMALS
  4. BAROREFLEX CIRCUITRY IN AWAKE HUMANS
  5. CONCURRENT MSNA AND FMRI RECORDING
  6. BRAINSTEM ACTIVITY CHANGES DURING SPONTANEOUS FLUCTUATIONS IN MSNA
  7. CORTICAL SITES REGULATING MSNA IN AWAKE HUMANS
  8. LITERATURE CITED

While in the anesthetized cat, fMRI has revealed signal intensity changes within the baroreflex circuitry (Henderson et al.,2004), relatively few studies have applied fMRI to investigate the role of the brainstem in human cardiovascular control. Some of the first studies to explore cardiovascular-related brain regions in humans using fMRI used the well-described Valsalva maneuver, which evokes a sustained increase in MSNA and arterial pressure. These initial investigations revealed that the Valsalva maneuver is associated with increased signal intensity within multiple brain regions, including the dorsal pons and medulla (Harper et al.,1998; Henderson et al.,2002). Whilst these studies are important, the results need to be interpreted in the context that the skeletal motor activity required to perform the Valsalva maneuver itself must be responsible for at least some of the reported brain activation.

In an attempt to reduce the degree of brain activation due to skeletal muscle activity, we recently used a maximal inspiratory breath-hold, a maneuver that requires only the glottis to be closed (through active constriction), while the inspiratory muscles are relaxed, to explore regions underlying increased MSNA in humans. The increase in MSNA evoked by this maneuver is believed to result from unloading of the low-pressure baroreceptors (Macefield,1998). We found that a maximal inspiratory breath-hold caused significant signal intensity changes in three discrete regions of the medulla: robust signal increases occurred in the region of the RVLM and signal decreases in the regions of the CVLM and NTS (Macefield et al.,2006). Brain activation patterns have also been assessed during lower body negative pressure, a maneuver that activates the baroreflex and does not require the subject to perform any task. Although the authors did not report medullary changes, they did report changes in higher brain regions such as the insular, anterior cingulate and orbitofrontal cortices, as well as in the midbrain and thalamus (Kimmerly et al.,2005).

Although these studies have gone some way to determining the central circuits responsible for producing changes in MSNA, these efforts relied on investigating brain activity changes during evoked, sustained changes in MSNA. Furthermore, the MSNA changes were not recorded at the same time as the changes in brain activity and were, for the most part, taken from separate investigations performed in a separate location, often on different days and sometimes in a different group of individuals. With this in mind, we recently began to record brain activity (using fMRI) and MSNA (using microneurography) concurrently, to assess brain circuitry activated during small, spontaneous fluctuations in MSNA in awake humans.

CONCURRENT MSNA AND FMRI RECORDING

  1. Top of page
  2. Abstract
  3. BAROREFLEX CIRCUITRY IN EXPERIMENTAL ANIMALS
  4. BAROREFLEX CIRCUITRY IN AWAKE HUMANS
  5. CONCURRENT MSNA AND FMRI RECORDING
  6. BRAINSTEM ACTIVITY CHANGES DURING SPONTANEOUS FLUCTUATIONS IN MSNA
  7. CORTICAL SITES REGULATING MSNA IN AWAKE HUMANS
  8. LITERATURE CITED

The microneurographic technique was developed over 50 years ago and can be used to directly measure action potentials in unmyelinated postganglionic sympathetic fibers in awake human subjects (Vallbo et al.,2004). The technique involves the insertion of a tungsten microelectrode into an appropriate peripheral nerve, which for sympathetic activity is most often the common peroneal nerve. In general, multiunit sympathetic nerve activity is recorded, and it occurs as bursts of impulses separated by silent periods. Figure 1 shows a typical microneurographic recording in a laboratory setting. Multiunit MSNA was recorded from the peroneal nerve using a low-noise headstage, the signal was amplified (2 × 104) and filtered (0.3–3.0 kHz). Heart rate was measured via standard Ag–AgCl electrocardiogram chest electrodes, continuous pulsatile arterial pressure using noninvasive pulse plethysmography placed on an index finger of the subject's nondominant hand and the calibrating cuff placed around the contralateral upper arm, and finally respiration using a strain gauge transducer attached to a strap around the chest. It is clear that each time pulsatile arterial pressure falls, there are multiple MSNA bursts.

thumbnail image

Figure 1. A typical microneurographic recording in a laboratory setting. Multiunit MSNA was recorded from the peroneal nerve. Note that when arterial pressure (AP) decreases, muscle sympathetic nerve activity (MSNA) increases. bpm: beats per minute; ECG: electrocardiogram; RMS: root mean square.

Download figure to PowerPoint

Although microneurographic recording in a laboratory setting has become a widely used technique, it is still difficult to perform given the extremely small signals, which need to be amplified ∼20,000 times. The equipment must be adequately shielded from external interference and the subject must remain relative still so that the nerve being recorded is not lost. Adding a high (3 Tesla) magnetic field (100,000× the earth's magnetic field) and high-intensity electromagnetic (radio frequency) pulses, as occurs in the fMRI environment, makes the recording of MSNA in awake humans even more challenging.

In a recent series of experiments, we managed to successfully record MSNA at the same time as measuring brain activity using fMRI. In these studies we used a scan repetition time of 8 sec and implemented a 4-sec image collection period followed by a 4-sec rest period during which MSNA was recorded. This protocol was chosen to take advantage of the temporal delays inherent in blood oxygen level dependent (BOLD) imaging. It is known that the microvascular responses to an increase in neuronal activity lag by ∼5 sec (Logothetis et al.,2001), and that we need to allow ∼1 sec for a muscle vasoconstrictor volley to travel from the brainstem to the peripheral recording site at the knee (Fagius and Wallin,1980). Accordingly, we reasoned that changes in BOLD signal intensity would reflect changes in neural activity associated with emission of sympathetic volleys recorded in the previous 4-sec epoch. Furthermore, we collected axial slices sequentially from caudal to rostral so that the timing of each slice could be related to the MSNA recording. Figure 2 shows a recording of spontaneous raw and integrated MSNA nerve activity. Note the single MSNA burst during the second rest period.

thumbnail image

Figure 2. A typical microneurographic recording during concurrent functional magnetic resonance imaging (fMRI). Brain images were collected during 4 sec at which time muscle sympathetic nerve activity (MSNA) was not distinguishable. However, due to the fMRI hemodynamic delay (∼5 sec) and the delay for MSNA traffic to travel from the brain to the recording electrode (∼1 sec), brain activity during the MSNA collection period was reflected in signal intensity changes during the following 4-sec period. Note the MSNA burst during the second MSNA collection period. RMS: root mean square.

Download figure to PowerPoint

BRAINSTEM ACTIVITY CHANGES DURING SPONTANEOUS FLUCTUATIONS IN MSNA

  1. Top of page
  2. Abstract
  3. BAROREFLEX CIRCUITRY IN EXPERIMENTAL ANIMALS
  4. BAROREFLEX CIRCUITRY IN AWAKE HUMANS
  5. CONCURRENT MSNA AND FMRI RECORDING
  6. BRAINSTEM ACTIVITY CHANGES DURING SPONTANEOUS FLUCTUATIONS IN MSNA
  7. CORTICAL SITES REGULATING MSNA IN AWAKE HUMANS
  8. LITERATURE CITED

Using the protocol described above, we explored the medullary circuits responsible for the baroreflex by measuring regional brainstem activity changes during small, spontaneous changes in MSNA in awake humans. Significant correlations between regional brainstem signal intensity and MSNA were found to occur in those regions previously described as being responsible for baroreflex functioning in anesthetized experimental animal, the RVLM, CVLM, and NTS (Macefield and Henderson,2010). More specifically, we found that small increases in MSNA were associated with fMRI signal intensity increases in the RVLM and signal decreases within the regions of the CVLM and NTS (Fig. 3).

thumbnail image

Figure 3. Functional magnetic resonance imaging (fMRI) signal intensity changes correlated to spontaneous fluctuations in muscle sympathetic nerve activity (MSNA) in eight subjects. Increases in signal intensity with increases in MSNA are coded by the hot color scale and signal decreases with the cool color scale and are overlaid onto a series of axial fMRI slices from an individual subject. CVLM: caudal ventrolateral medulla; NTS: nucleus tractus solitaries; RVLM: rostral ventrolateral medulla; SI: signal intensity.

Download figure to PowerPoint

It remains unclear exactly what fMRI signal intensity changes actually represent, with some suggesting that it reflects underlying synaptic transmission and others neuronal firing (Logothetis et al.,2001; Mukamel et al.,2005; Viswanathan and Freeman,2007). We found that during increases in MNSA (in response to spontaneous falls in arterial pressure), signal intensity within the NTS decreased—which is consistent with this region receiving lower baroreceptor afferent drive—that is, a withdrawal of excitatory drive. Lower NTS activity would in turn result in decreased CVLM activity, which would result in a reduction in its tonic inhibitory drive onto promoter sympathetic neurons within the RVLM. Consistent with these hypotheses, we found decreases in CVLM and increases in RVLM signal intensity during small falls in arterial pressure (increases in MSNA). These results suggest that BOLD signal intensity changes within the brainstem reflect changes in neuronal firing rather than synaptic events per se, given that the synaptic events within RVLM are related to the active release of the inhibitory neurotransmitter GABA.

Figure 4 shows the fMRI signal intensity changes and MSNA changes within the RVLM and CVLM of two subjects. Note that within the RVLM, signal intensity increases were significantly correlated to increases in MSNA. Alternatively, signal intensity decreases were significantly correlated to increases in MSNA within the region of the CVLM. Also note the relatively dorsal location of the RVLM and CVLM compared to its location in experimental animals. In humans, the RVLM and CVLM are displaced dorsally by the large inferior olivary nuclei (Allen et al.,1998). The functional and anatomical localization of the human RVLM is illustrated in Fig. 5. The left panel shows fMRI signal intensity increases in the RVLM region during spontaneous and apnea-evoked MSNA increases (Macefield et al.,2006; Macefield and Henderson,2010). The right panel shows the anatomical location of the human RVLM based on immunohistochemical identification (Allen et al.,1998).

thumbnail image

Figure 4. Functional magnetic resonance imaging (fMRI) signal intensity changes correlated to spontaneous fluctuations in muscle sympathetic nerve activity (MSNA) in two individual subjects overlaid onto axial fMRI image slices. Signal intensity increases (hot color scale) within rostral ventrolateral medulla (RVLM) are on the left and signal decreases (cool color scale) within the caudal ventrolateral medulla (CVLM) are on the right. Below each axial slice are plots of the mean raw signal intensity during low MSNA (blue) and high MSNA (red). In addition, plots of MSNA levels against signal intensity are also shown. Note the significant correlation between MSNA and fMRI signal intensity. SI: signal intensity. * represents a significant difference in signal intensity during high and low MSNA conditions. # following each R-value indicates a significant negative correlation between signal intensity and MSNA.

Download figure to PowerPoint

thumbnail image

Figure 5. Functional and anatomical localization of the human rostroventrolateral medulla (RVLM). The left panel shows a bilateral increase in RVLM functional magnetic resonance imaging (fMRI) signal intensity during a maximal inspiratory breath-hold [data from Macefield et al. (2006)] and fMRI signal intensity increases during small spontaneous increases in MSNA [data from Macefield and Henderson (2010)]. The right panel shows the anatomical identification if the RVLM using the binding of Angiotensin II receptors [reproduced from Macefield and Henderson (2010) courtesy of AM Allen].

Download figure to PowerPoint

CORTICAL SITES REGULATING MSNA IN AWAKE HUMANS

  1. Top of page
  2. Abstract
  3. BAROREFLEX CIRCUITRY IN EXPERIMENTAL ANIMALS
  4. BAROREFLEX CIRCUITRY IN AWAKE HUMANS
  5. CONCURRENT MSNA AND FMRI RECORDING
  6. BRAINSTEM ACTIVITY CHANGES DURING SPONTANEOUS FLUCTUATIONS IN MSNA
  7. CORTICAL SITES REGULATING MSNA IN AWAKE HUMANS
  8. LITERATURE CITED

As mentioned in the Introduction, the sympathetic nervous system is mobilized during behavioral changes such as during exercise and defensive activities. In addition, cortically driven states such as chronic mental stress can cause a sustained increase in MSNA, resulting in hypertension (Esler et al.,2008). Although a great deal is known about brainstem mechanisms controlling MSNA, relatively little is known about the role of higher centers in cardiovascular control. Some investigations in humans using fMRI have revealed cortical sites that respond to challenges that evoke increases in MSNA, such as the Valsalva maneuver, inspiratory capacity apnea and lower body negative pressure.

In previous studies we have shown that Valsalva maneuvers and maximal capacity apneas evoke fMRI signal intensity increases in cortical regions such as the anterior insula, anterior cingulate and cerebellar cortices as well as the deep cerebellar nuclei (Henderson et al.,2002; Macefield et al.,2006) (Fig. 6). Similarly, Kimmerly et al. (2005) revealed that lower body negative pressure evoked fMRI signal changes in the anterior insula and anterior cingulate cortices. It is well known that the insula is an important region for regulating cardiovascular control. In rodents, electrical stimulation of the insula can evoke increases in arterial pressure, heart rate and renal sympathetic nerve activity (Butcher and Cechetto,1995) and insula lesions can interfere with baroreceptor functioning (Zhang et al.,1998). In humans, lesions of the left insular cortex results in enhanced basal sympathetic tone (Oppenheimer et al.,1996). Although the insula does not project directly to the RVLM (Saper,1982), electrical stimulation of the insula does excite some RVLM sympathoexcitatory neurons, presumably via an indirect pathway (Sun,1992). Similar to the insula, electrical stimulation of the anterior cingulate cortex elicits large changes in arterial pressure and heart rate upon stimulation (Burns and Wyss,1985). The anterior cingulate also contains neurons that fire in a discharge pattern related to the respiratory and cardiac cycles (Frysinger and Harper,1986).

thumbnail image

Figure 6. Changes in functional magnetic resonance (fMRI) signal intensity during increases in muscle sympathetic nerve activity (MSNA) evoked by Valsalva maneuvers and inspiratory capacity apneas overlaid in a mean T1-weighted anatomical image set. Although there were some regional differences, during both Valsalva maneuvers and inspiratory capacity apneas, increased MSNA was associated with signal intensity increase (hot color scales) within the anterior insular, anterior cingulate, and cerebellar cortices as well as in the deep cerebellar nuclei. The slice locations in Montreal Neurological Institute space are indicated at the lower right of each brain slice. SI: signal intensity.

Download figure to PowerPoint

In contrast to the accepted idea that cortical regions, such as the cingulate and insula are involved in cardiovascular regulation, most investigations consider the cerebellum to be almost exclusively involved in movement control. However, over 60 years ago, the cerebellum was described as having a role in autonomic control (Moruzzi,1940). Electrical and chemical stimulation of the deep cerebellar nuclei can modify arterial pressure, heart rate and vasopressin release (Del Bo et al.,1983; Bradley et al.,1987). Investigations in experimental animals have also revealed roles for other areas in the control of cardiovascular function, such as the midbrain periaqueductal gray and hypothalamus.

Although these previous investigations have begun to delineate the role of various higher brain regions in the control of sympathetic activity, they have not explored directly the role of these regions in mediating spontaneous fluctuations in MSNA. Because the baseline increases in MSNA that characterize conditions of chronic stress, obstructive sleep apnea and essential hypertension likely involve changes in the on-going regulation of the baroreflex circuitry by higher brain centers, it is important to determine the role of higher centers in the baseline modulation of sympathetic drive. We have recently begun a series of experiments in which we are recording MSNA and cortical fMRI activity concurrently to answer this question. So, rather than limiting our scans to the brainstem, we are assessing the whole brain. By knowing the caudo-rostral sequence of axial slices obtained during scanning, we can identify the regions in the brain corresponding to the arrival of sympathetic bursts at the recording microelectrode 4 sec later. In this way, we can examine areas of increases or decreases in signal intensity that covary with the peripherally recorded sympathetic bursts. Once we have identified the cortical and subcortical areas that correlate with MSNA in healthy individuals we can then use these results to explore the functional changes in the brain associated with pathophysiological increases in MSNA, such as heart failure and obstructive sleep apnea.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. BAROREFLEX CIRCUITRY IN EXPERIMENTAL ANIMALS
  4. BAROREFLEX CIRCUITRY IN AWAKE HUMANS
  5. CONCURRENT MSNA AND FMRI RECORDING
  6. BRAINSTEM ACTIVITY CHANGES DURING SPONTANEOUS FLUCTUATIONS IN MSNA
  7. CORTICAL SITES REGULATING MSNA IN AWAKE HUMANS
  8. LITERATURE CITED