J Neurophysiol 93: 1647–1658, 2005;
doi:10.1152/jn.00863.2004.
Hypercapnic Exposure in Congenital Central Hypoventilation Syndrome
Reveals CNS Respiratory Control Mechanisms
R. M. Harper,1,4 P. M. Macey,1 M. A. Woo,3 K. E. Macey,1 T. G. Keens,5 D. Gozal,6 and J. R. Alger2,4
1
Department of Neurobiology, 2Department of Radiology, 3School of Nursing, and the 4Brain Research Institute, University of California
at Los Angeles and 5Childrens Hospital Los Angeles, Los Angeles, California; and 6Kosair Children’s Hospital Research Institute,
Department of Pediatrics, University of Louisville School of Medicine, Louisville, Kentucky
Submitted 23 August 2004; accepted in final form 2 November 2004
INTRODUCTION
Congenital central hypoventilation syndrome (CCHS) provides an “experiment of nature” that can differentiate brain
mechanisms controlling breathing. Affected children exhibit
deficient ventilatory responses to hypercapnia and hypoxia and
chronic respiratory insufficiency in the absence of major pulmonary, cardiac, neuromuscular, or chest wall disease, and
show a diminished drive to breathe during sleep (Commare et
al. 1993; Haddad et al. 1978; Oren et al. 1987; Paton et al.
1989). Deficiencies also include a loss of dyspnea, i.e., the
perception of air hunger or discomfort for the need to breathe
to either hypoxic or hypercapnic exposures (Paton et al. 1989;
Shea et al. 1993). Furthermore, CCHS patients display reduced
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: [email protected]).
www.jn.org
influences of breathing on cardiac rate variation (Woo et al.
1992) and a range of both sympathetic and parasympathetic
nervous system control deficits, including profuse sweating
and a proclivity for syncope (Vanderlaan et al. 2004; WeeseMayer et al. 1992). Despite reduced ventilatory responses to
hypercapnia or hypoxia, peripheral chemoreceptor responses
are partially preserved, particularly in those children who are
able to sustain near-adequate ventilatory output during wakefulness (Gozal et al. 1993). Affected patients increase ventilation to exercise and to passive motion of the extremities, even
during sleep (Gozal and Simakajornboon 2000; Gozal et al.
1996), indicating that integration of respiratory motor output
with cyclic locomotor pattern systems (Dejours 1959) are
preserved, even with central chemoreceptor integrative failure.
Although ventilatory responses to chemoreceptor stimulation
are impaired, affected children do arouse from sleep to high
CO2 (Marcus et al. 1991), suggesting that central afferent
processes are largely intact and that breathing deficits in CCHS
result from diminished integration of respiratory motor processes with sensory input.
The combination of intact aspects of breathing control with
selective deficiencies in CCHS provides a unique opportunity
to examine central control of discrete components of breathing
regulation. Despite the range of respiratory deficiencies, affected patients apparently possess a relatively unimpaired afferent system for central chemoreception (since arousal to
hypercapnia is preserved), functioning motor output systems,
and integrity of respiratory motor/locomotor patterning organization. We used functional MRI (fMRI) procedures to evaluate responses to hypercapnia throughout the brain in CCHS
patients and control subjects. We hypothesized that brain
structures implicated in mediating chemoreception and responsive to air hunger, the latter including the cingulum, insula, and
amygdala (Banzett et al. 2000; Evans et al. 2002; Peiffer et al.
2001), would be less responsive in CCHS, and that motor
coordination areas of the cerebellum underlying rapid adjustment of breathing muscle action and compensatory cardiovascular responses would react abnormally in the syndrome.
METHODS
Fourteen children with a CCHS diagnosis, based on standard
criteria (American Thoracic Society 1999) (7 males and 7 females)
and 14 controls (8 males and 6 females) participated. The groups were
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.
0022-3077/05 $8.00 Copyright © 2005 The American Physiological Society
1647
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
Harper, R. M., 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 93: 1647–1658, 2005; doi:10.1152/
jn.00863.2004. Congenital central hypoventilation syndrome (CCHS)
patients show impaired ventilatory responses and loss of breathlessness to hypercapnia, yet arouse from sleep to high CO2, suggesting
intact chemoreceptor afferents. The syndrome provides a means to
differentiate brain areas controlling aspects of breathing. We used
functional magnetic resonance imaging to determine brain structures
responding to inspired 5% CO2-95% O2 in 14 CCHS patients and 14
controls. Global signal changes induced by the challenge were removed on a voxel-by-voxel basis. A priori– defined volume-of-interest time trends (assessed with repeated measures ANOVA) and cluster
analysis based on modeling each subject to a step function (individual
model parameter estimates evaluated with t-test, corrected for multiple comparisons) revealed three large response clusters to hypercapnia
distinguishing the two groups, extending from the 1) posterior thalamus through the medial midbrain to the dorsolateral pons, 2) right
caudate nucleus, ventrolaterally through the putamen and ventral
insula to the mid-hippocampus, and 3) deep cerebellar nuclei to the
dorsolateral cerebellar cortex bilaterally. Smaller clusters and defined
areas of group signal differences in the midline dorsal medulla,
amygdala bilaterally, right dorsal-posterior temporal cortex, and left
anterior insula also emerged. In most sites, early transient or sustained
responses developed in controls, with little, or inverse change in
CCHS subjects. Limbic and medullary structures regulating responses
to hypercapnia differed from those previously shown to mediate
loaded breathing ventilatory response processing. The findings show
the significant roles of cerebellar and basal ganglia sites in responding
to hypercapnia and the thalamic and midbrain participation in breathing control.
1648
HARPER ET AL.
J Neurophysiol • VOL
time trends, including early, transient patterns, which did not necessarily follow an ON-OFF sequence. For that reason, we implemented
VOI analysis using custom routines that evaluated participation of a
priori– defined structures without assumptions about the response
pattern and were unaffected by potential variation in spatial normalization, since each VOI was outlined on a subject-by-subject basis.
Voxel intensities were averaged for each subject at each time-point,
and trends of these VOI analyses were plotted. RMANOVA evaluated
differences from baseline for each group and response differences
between groups (Littell et al. 1996). The VOI analysis involved
averaging all voxel values within the VOI prior to statistical analysis,
and therefore there were no multiple comparisons across voxels.
Additionally, the Tukey-Fisher criterion for multiple comparisons was
applied, namely the overall effect of the model was tested for
significance (P ⬍ 0.05) prior to testing individual time-points for
within- or between-group effects.
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 and
parents or guardians.
RESULTS
Physiology
Respiratory and cardiac rate and variability responses to
hypercapnia have been described earlier (Macey et al. 2004c).
Briefly, breathing and heart rates rose during the hypercapnic
challenge in both control and CCHS groups, but a transient
decline to baseline breathing rates after 60 s in control subjects
did not occur in CCHS patients. A remarkable augmentation of
heart rate variability, contributed largely by respiratory-related
variation, increased only transiently and to a much lesser extent
in CCHS cases. ETCO2, measured in a subset of subjects, was
higher in CCHS compared with control subjects during baseline and challenge periods, and increased in both groups during
hypercapnia.
Global BOLD signal
The global BOLD signal increased in both groups 20 s after
challenge onset (Macey et al. 2003), but the increase was
greater in control subjects from 30 s onward. Detrending
removed all global effects (Macey et al. 2004b).
Regional response patterns
Figure 1 shows regions of significant response in the control
group based on the cluster analysis. Most patterns appeared
early and transiently, with rapid increases bilaterally in the
dorsal cerebellar cortex and deep nuclei (Fig. 1, A and E, see
FN) and in the posterior right insula (Fig. 1B). Rapid, but
largely transient, signal decreases occurred in a region encompassing the dorsal pons, including the parabrachial region,
midbrain, the posterior thalamus, and hypothalamus (Fig. 1, A
and C, time trend), a portion of the right hippocampus and
overlying cortex (Fig. 1D), the caudate nuclei bilaterally (Fig.
1, E, time trend, and F), and left putamen (Fig. 1F).
Several clusters of response differences were apparent in
CCHS patients. These clusters included a region extending
from the caudate nucleus through the putamen and ventral
insula and extended to the amygdala and dentate and CA3
areas of the hippocampus and surrounding cortex on the
superior-lateral surface of the hippocampus (Fig. 2A, 1, 2, 4,
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
age-matched [age: mean, 11 ⫾ 2 (SD) yr; range, 8 –15 yr]. Patients
breathed spontaneously during waking, but were ventilated via a
tracheostomy during sleep; all showed depressed ventilatory responses to
hypercapnia. Patients with Hirschsprung’s disease were excluded, as
were patients with indications of cardiac dysfunction or any additional
indication of primary pulmonary or neuromuscular disease.
Subjects lay supine in a 1.5-T MRI scanner (General Electric Signa,
Milwaukee, WI) and breathed through a two-way, nonrebreather
valve. Tracheostomy openings were closed throughout the studies,
and subjects wore noseclips. Head movement was minimized by
application of masking tape over the forehead and foam pads on either
side of the head. Airflow and the ECG (lead I) were recorded together
with fMRI images (Macey et al. 2004c). End-tidal CO2 (ETCO2) and
O2 saturation were measured in a subset of 10 control and 7 CCHS
subjects, respectively. Each subject underwent two scanning periods,
separated by ⱖ8 min: the first consisting of a 150-s baseline breathing
room air, and the second period with a 30-s baseline followed without
pause by a 120-s challenge. The gas mixture (5% CO2-95% O2) was
delivered via the inspiratory arm of the two-way valve throughout the
120-s challenge period. The high level of O2 was used to suppress
peripheral chemoreceptor afferent activity.
Each 150-s scanning period consisted of 25 volumes of 20 oblique
image slices, collected using a gradient-echo echo-planar imaging
(EPI) protocol [repetition time (TR) ⫽ 6 s, time to echo (TE) ⫽ 60
ms, flip angle ⫽ 90°, field of view (FOV) ⫽ 30 ⫻ 30 cm, no interslice
gap, and voxel size ⫽ 2.3 ⫻ 2.3 ⫻ 5 mm]. Blood oxygen level–
dependent (BOLD) intrinsic contrast was used to evaluate neural
responses to the challenge. Spin-echo T1-weighted images (TR ⫽ 500
ms, TE ⫽ 9 ms, FOV ⫽ 30 ⫻ 30 cm, no interslice gap, voxel size ⫽
1.2 ⫻ 1.2 ⫻ 5 mm) were collected at comparable orientation and
positioned to assist anatomical identification.
The images were evaluated with a statistical parametric mapping
package, SPM (Friston et al. 1995), and custom software. Preprocessing included correction for slice timing and motion, and volumes were
spatially normalized. Gray matter was segmented from white matter
and cerebrospinal fluid (CSF) to create a mask of regions where the
probability of gray matter was ⬎0.5 (Ashburner and Friston 1997);
after application of the mask to the spatially normalized images, the
resultant images were smoothed. Global signal changes, induced by
overall perfusion changes from the challenges or other sources, were
removed (Macey et al. 2004b), and significant regional changes in signals
were assessed.
Two analytic approaches were used: a cluster analysis and a volume
of interest (VOI) procedure. Cluster analysis provided an overall
whole brain search on a voxel-by-voxel basis to determine what brain
areas were activated by hypercapnia, assuming a step function of OFF
to ON response pattern (termed “boxcar”), of generally increased or
decreased signal to the challenge. The time course of each voxel was
matched using a linear model of a boxcar pattern convolved with a
standard hemodynamic response function, and the resulting parameter
estimates for each subject were saved as one “volume.” These estimates were compared within the control group using a one-sample
t-test (a population, or random effects procedure; P ⬍ 0.05, false
discovery rate correction for multiple comparisons) and across groups
using a two-sample t-test, and “clusters” of adjacent significant voxels
were mapped to show an overview of responsive sites. Time courses
of all significant voxels within selected clusters were extracted,
averaged across all voxels to avoid multiple comparisons issues,
plotted for the two groups, and verified using repeated measures
ANOVA (RMANOVA; P ⬍ 0.05). The RMANOVA procedure is
specifically designed to account for multiple comparisons across time,
and although the technique can be considered an extension of standard
ANOVA, it is implemented using a linear model (Littell et al. 1996).
Even though clusters often encompass more than one structure, the
time course of the cluster response is representative of the response
time course of all areas within that cluster. Physiological changes to
hypercapnia described earlier (Macey et al. 2004c) showed unique
RESPIRATORY CONTROL IN HYPERCAPNIA
1649
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
FIG. 1. Coronal, sagittal, and axial views of significant responses in the control subjects. Clusters are overlaid onto a single anatomical image derived from
one mid-age range subject. Regions of signal increase are color-coded in yellow-red scale, and regions of reduced signal are color-coded in blue-green (P ⬍ 0.05
threshold; t-statistic scale at bottom of figure). Signal increases occurred in (A) cerebellar cortex, bilaterally and (B) right posterior insula. Signal decreases
occurred in (C) dorsal pons, including parabrachial region through the midbrain to the posterior thalamus and hypothalamus; (D) right hippocampus; (E) caudate
nucleus, fastigial nucleus (FN); and (F) right putamen. Time trends of the average value of all voxels within the clusters are shown for the group with SE bars;
for A–F, white arrows on overlays indicate the cluster to which the time trend corresponds.
and 5); signal changes in CCHS patients were larger in this
cluster (blue-coded areas), with values rising in CCHS, but
showing an initial transient decline in control subjects (Fig.
2B). The pattern was primarily unilateral, with the largest
J Neurophysiol • VOL
cluster on the right side, although a smaller cluster in the left
basal ganglia was also apparent (Fig. 2A3).
A second extensive cluster of enhanced signals in CCHS
patients ranged from the posterior-dorsal, medial, and ventral
93 • MARCH 2005 •
www.jn.org
1650
HARPER ET AL.
thalamus (Fig. 3A, 1 and 2) through the medial midbrain (Fig.
3A, 1–3) and, with a short interruption, terminated in the
dorsolateral pons (Fig. 3A4). As in the more rostral cluster of
Fig. 2, an initial transient decline emerged in control subjects,
with a gradual return to baseline, but CCHS patients showed
only a late rise (Fig. 3B).
A third large cluster extended from the deep cerebellar
nuclei bilaterally to the dorsal cerebellar cortex (Fig. 4A, 1– 4)
in a defined columnar arrangement (Fig. 4A, 2 and 3); in this
cerebellar cluster, signals were greater in control subjects
(red-orange-white coding), with pattern differences emerging
early in the challenge (Fig. 4B).
Increased signals in control over CCHS subjects were also
apparent in a region encompassing the dorsal medulla, within
which lies in the solitary tract nucleus (NTS; Fig. 5A, 1–3). The
corresponding trend plot (Fig. 5A, right) shows an early and
J Neurophysiol • VOL
sustained rise in control subjects but an initial transient decline
in CCHS cases. A bilateral cluster in the amygdala, extending
to nearby cortex (Fig. 5B, 1–3) showed initial, transient divergent responses in the two groups, a decline in CCHS, and an
increase in control subjects (Fig. 5B, right). In contrast to the
transient decline in the control response in the right insula, a
sustained rise in a cluster in the left anterior insula appeared,
with no change in CCHS (Fig. 5C). Signals in CCHS patients
were larger in a cluster over the posterior-superior temporal
cortex, on the right side only (Fig. 5D, 1–3); an early transient
decline in the control group appeared (Fig. 5D, right).
The response time trends of defined VOI are outlined in Fig.
6, and Table 1 represents a summary of those changes, The
large, early, transient responses found in Figs. 2–5 for multiple
structures also appeared in VOI for the dentate nucleus, hippocampus, caudate head, left anterior insula, medial medulla,
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
FIG. 2. A: coronal, sagittal, and axial views of significant
differences in responses between congenital central hypoventilation syndrome (CCHS) and control groups of a cluster
extending from the right head of the caudate nucleus through
the putamen and globus pallidus to the insula (1, 2, and 4)
and to the dentate and CA3 areas of the hippocampus (3); the
region is smaller on the left (3) (overlays onto a mean
anatomical image, derived from all subjects to show variability of structures, are placed in the web-based data supplement1). Regions of greater signal in control than CCHS
subjects are color-coded in yellow-red scale, and regions of
reduced signal in control relative to CCHS subjects are
color-coded in blue-green (P ⬍ 0.05). B: time trends from
clusters of significant differences in response between CCHS
and control groups for regions defined in the group cluster,
plotted as group mean with SE bars.
RESPIRATORY CONTROL IN HYPERCAPNIA
1651
and amygdala (Fig. 6), with more modest changes in CCHS or
changes in the opposite direction (dorsal and ventral midbrain,
hippocampus, amygdala). A nadir emerged in control subjects
in the hippocampus 60 s after onset of the challenge, which
temporally corresponded to a respiratory rate difference between groups (Macey et al. 2004c). The VOI for the lentiform
nuclei, ventral medulla, and ventral pons showed significant
responses to the challenge but no group differences.
A data supplement includes the same clusters and time
trends as Figs. 2–5, but with the clusters overlaid onto a mean
of all subjects’ T1-weighted anatomical images collected at the
same slice locations as the fMRI data. The supplementary
figures are provided to give a visual indication of the data
variation due to spatial registration procedures.1
1
The Supplementary Material for this article (four figures) is available
online at http://jn.physiology.org/cgi/content/full/00863.2004/DC1.
J Neurophysiol • VOL
DISCUSSION
The most remarkable aspect of the regional fMRI signal
responses to hypercapnia, assessed after correction for global
changes, consisted of pronounced differences expressed in
regions not classically associated with breathing control, but
traditionally related to affect, autonomic regulation, or motor
coordination. A second noteworthy finding in control subjects
was the early transient nature of the responses, especially in
motor regions. The early reactions were muted or absent in
cerebellar, thalamic, and basal ganglia sites in CCHS, whereas
limbic sites, principally the hippocampus and nearby cortex,
amygdala, insula, and dorsal and ventral midbrain showed
responses that were delayed or were in the opposite direction
compared with control subjects. A column of deficient responses extending from the dorsal, medial, and ventral thalamus through the midbrain to the dorsolateral pons was unex-
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
FIG. 3. A: axial, sagittal, and coronal views of significant
response differences between CCHS and control groups in a
cluster extending from the posterior and medial thalamus (1
and 2) to the medial midbrain (3), and with minimal separation, to the dorsolateral pons (4). B: trends of values for
this significant cluster. Trace, color coding, and labeling
attributes as in Fig. 1.
1652
HARPER ET AL.
pected. Aberrant responses also appeared in some, but not all,
sites presumably targeted by PHOX2B expression, mutations
of which have been found in a high proportion of CCHS
patients (Amiel et al. 2003; Weese-Mayer et al. 2003); these
areas included the dorsal medulla and other sites that should
play a role in the modulation of the hypercapnia challenge
(Dauger et al. 2003).
fMRI signal interpretation
Images collected during fMRI procedures are sensitive to
levels of deoxyhemoglobin in the blood, with higher levels of
deoxyhemoglobin leading to lower signal intensities; these
signal intensities are termed the BOLD signal. Neural activation results in an inflow of oxygenated blood that decreases
deoxyhemoglobin concentration and increases the BOLD sigJ Neurophysiol • VOL
nal (Ogawa et al. 1990). There is a strong correlation between
changes in BOLD signal and local field potentials, and therefore changes in the BOLD signal reflect input and intracortical
processing (Logothetis et al. 2001).
Control responses
The findings confirm earlier fMRI studies in the adult that
noted cerebellar and more rostral involvement in mediating
hypercapnia (Brannan et al. 2001; Corfield et al. 1995; Gozal
et al. 1994; Harper et al. 1998; Kastrup et al. 1999; Parsons et
al. 2001), as well as animal studies, indicating a significant role
for deep cerebellar nuclei in such challenges (Xu and Frazier
1997, 2002). Several of these studies showed more substantial
changes within multiple sites in the adult than appeared here in
children. However, we adopted very conservative correction
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
FIG. 4. A: Sagittal, coronal, and axial views of significant
response differences between CCHS and control groups in a
cluster extending from the deep cerebellar nuclei (A1) to the
dorsal cerebellar cortex (A1–A4), bilaterally. B: trends of values
for this significant cluster. Trace, color coding, and labeling
attributes as in Fig. 1.
RESPIRATORY CONTROL IN HYPERCAPNIA
1653
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
FIG. 5. Left: axial, sagittal, and coronal views of regions of significant difference in response between CCHS and control groups from several clusters, overlaid
onto a mean anatomical image, with display conventions and color coding as in Fig. 1. A1–A3: amygdala. B1–B3: dorsal medulla. C1–C3: left insula. D1–D3:
posterior temporal cortex. Right: time trends from clusters of numbered images on the left, plotted as group mean with SE bars.
procedures to avoid global signal contributions from CO2
effects on the vasculature, and these procedures may have
reduced the extent of signal change in several areas (Macey et
al. 2004b). Specifically, we partitioned gray matter and only
examined effects therein, since gray and white matter have
J Neurophysiol • VOL
differing patterns of global signal change, and we removed all
signal patterns matching the overall global pattern within gray
matter (to view global signal effects, see Macey et al. 2003). It
is useful to extend the adult findings to children, but of perhaps
greater importance are the time courses of responses in affected
93 • MARCH 2005 •
www.jn.org
1654
HARPER ET AL.
structures. The most common response was an early transient
rise or fall in signal in cerebellar and basal ganglia structures,
with an intermediate-duration response in thalamic-midbrain
and hippocampal sites, and a gradual onset and more-prolonged response in the insula. The longer-duration reactions in
limbic structures likely represent autonomic regulatory control
aspects, whereas the transient cerebellar and basal ganglia
patterns presumably reflect faster motor components of the
hypercapnic challenge, i.e., respiratory efforts required to increase tidal volume.
Overall reactivity in CCHS
We earlier found that cerebral vascular reactivity seems to
be altered in CCHS (Macey et al. 2003); this alteration may
reduce BOLD signal changes assumed to represent neural
J Neurophysiol • VOL
activity variation between the two groups, given equivalent
levels of neural activity change. Many of the group differences
here seemed to be of an attenuated nature, with a missing early
transient or muted late response to hypercapnia. The extent of
overall reduction of early and late responses could reflect
global vascular reactivity deficiencies. This possibility, however, does not explain all the abnormal patterns in CCHS
subjects, given that other group differences were very robust,
with signals that differentiated the two groups moving in
opposite directions. These latter patterns appeared in the dorsal
and ventral midbrain, the amygdala, and the hippocampus (Fig.
6). The different response patterns included both transient
(dorsal and ventral midbrain) and more prolonged (amygdala,
hippocampus) signals. The presence of several regions with
responses in the opposite direction is an unlikely scenario for
overall muting by impaired cerebral reactivity processes.
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
FIG. 6. Time trends of volume of interest (VOI) from control and CCHS subjects during baseline and challenge periods. *Time-points of group difference
(RMAVONA, P ⬍ 0.05); time-points of significant signal increase or decrease relative to baseline within each group are indicated by bars above or below plots
(key at bottom).
RESPIRATORY CONTROL IN HYPERCAPNIA
TABLE
1. Summary of VOI analysis
Area Name of VOI [X Y Z] mm MNI
Coordinates of Mean Center of VOI
CCHS
⫺⫹⫺
⫹
⫹
⫺
⫺
⫺
⫺⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫹
⫺⫹
⫺⫹
⫹
⫹
⫹⫹
⫹⫹
⫹
⫹
⫹
⫹
⫹
⫹
*
*
⫹⫹
⫹
⫹
⫹
⫹
⫹
*
*
*
*
*
*
*
Significance was tested using RMANOVA (P ⬍ 0.05). Significant signal
increase (⫹), decrease (⫺), or no significant difference over ⬎ 1 time-point (0)
is indicated for each group, relative to baseline. Group differences are indicated by an asterisk (*). Where both groups increased and a group effect was
noted, the group with the greater extent of signal change is indicated by “⫹⫹.”
Where a significant decrease was followed by a significant increase, “⫺⫹” is
used, and in the control subjects for the head of caudate, “⫺⫹⫺” represents an
initial decrease followed by an increase and a later decrease. The coordinates
in MNI space (X Y Z, in mm) of the average center across subjects of each VOI
are shown. VOI, volume of interest; RMANOVA, repeated measures
ANOVA; MNI, Montreal Neurological Institute; CCHS, congenital central
hypoventilation syndrome.
Moreover, CCHS patients, although showing specialized cognitive and affective deficits, perform adequately in a number of
other motoric and cognitive tasks (Chen and Keens 2004;
Vanderlaan et al. 2004), an unexpected outcome for a generalized insufficiency in vascular reactivity. It may be the case
that deficits in vascular reactivity contribute to some of the
muted responses found here, but other regions most likely
responded with inappropriate neural activation or deactivation,
either from syndrome-specific dysfunction or by “release”
from other muted sites.
Cerebellar structures
The cerebellar cortex and deep nuclei showed large bilateral
areas of response to hypercapnia in control infants, as has been
shown earlier in adults (Gozal et al. 1994); these areas exhibited marked deficits in CCHS patients. The normal cerebellar
response is an early, short-lasting increase, suggestive of a
“signaling” role. The signal is likely that of chemoreception, a
role well-shown by others in animal studies of the fastigial
nucleus (Xu and Frazier 1997). The fastigial nucleus also
assists regulation of large blood pressure changes (Lutherer et
al. 1989). Deep cerebellar nuclei project to multiple reticular
and rostral sites, including the basal ganglia, with the fastigial
nucleus projecting to ventrolateral and posterior thalamic areas, in addition to pontine reticular regions (for review, see
J Neurophysiol • VOL
Carpenter and Batton 1982; Person et al. 1986). The findings of
altered responses in CCHS, particularly in dorsal thalamic,
caudate, and lentiform nuclei, may be secondary to the cerebellar deficiencies.
Cerebellar Purkinje cells normally inhibit the deep cerebellar
nuclei (Ohtsuka 1988), which showed response deficits in
CCHS. Purkinje cells of the cerebellum are exceptionally
sensitive to ischemia-induced damage (O’Hearn and Molliver
1997; Schadé and McMenemey 1963); a portion of this damage seems to be mediated via excitotoxic processes through
fibers from the inferior olive, a major afferent input to Purkinje
cells (Welsh et al. 2002). CCHS patients are often exposed to
hypoxia by virtue of inadequate assisted ventilation during
sleep or during waking periods associated with infection or
rest. Thus some portion of the cerebellar damage may be
secondary to repetitive hypoxic exposure, although intrinsic
damage from the syndrome cannot be excluded.
An exaggerated increase in respiratory-related heart rate
variation accompanied hypercapnia in control subjects; this
rise was not found in CCHS patients, who showed minimal
variation from respiratory sources even during baseline (Macey
et al. 2004c). The reduced respiratory-related arrhythmia points
to a loss of integration of rapid response respiratory and cardiac
interactive elements, a role the cerebellum serves, among other
motor coordination tasks (Lutherer and Williams 1986; Lutherer et al. 1989; Xu and Frazier 1994, 1997, 2002).
Limbic structures
The control subjects showed recruitment of several limbic
structures, and the CCHS patients showed temporal, magnitude, and directional differences in these responses. The limbic
areas likely serve multiple functions in mediation of CO2
responses, as suggested by animal and human studies, including chemoreception, perception of air hunger, and autonomic
aspects of the breathing challenge.
The perception of respiratory discomfort to hypoxia or
hypercapnia is a powerful drive to inspiratory effort (Moosavi
et al. 2003; Simon et al. 1989) and is lacking in CCHS (Paton
et al. 1989; Shea et al. 1993). The level of CO2 used in this
study (5%), together with delivery of a hyperoxic (95% O2)
balance (a combination that minimizes air hunger; Moosavi et
al. 2003), was unlikely to induce breathlessness in the control
subjects. Nevertheless, fMRI differences in limbic structures in
this study may relate to the neural systems that incorporate
sensation of high CO2 to elicit air hunger, because limbic
regions mediate affective responses to high CO2 or hypoxia.
Limbic sites, such as the amygdala and insula, have long been
implicated in distress processing (Amaral 2003) and are recruited in mediating restricted-breathing tasks that induce dyspnea (Evans et al. 2002). Several limbic sites, particularly the
insula and amygdala, revealed responses to hypercapnia in
control subjects that differed from CCHS patients, but not the
cingulate gyrus, a structure that is recruited during extreme
loaded and restricted breathing in adults (Evans et al. 2002;
Peiffer et al. 2001) and shows substantial response differences
in CCHS from control subjects in expiratory loading sufficient
to induce respiratory discomfort (Macey et al. 2004a). The
cingulate gyrus showed increased responses to hypercapnia,
but the increase was comparable in both groups. At least part
of the perception of breathlessness from excessive exertion is
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
Basal ganglia
Head of caudate [0 16 7]
Lentiform [1 5 4]
Limbic regions
Amygdala (left) [⫺21 ⫺6 ⫺18]
Amygdala (right) [22 ⫺6 ⫺18]
Hippocampus [0 ⫺20 ⫺17]
Insula (left anterior) [⫺22 ⫺3 ⫺18]
Insula (right anterior) [22 ⫺2 ⫺18]
Insula (left posterior) [⫺21 ⫺10 ⫺19]
Insula (right posterior) [22 ⫺9 ⫺18]
Brainstem regions
Dorsal midbrain [0 ⫺33 ⫺15]
Ventral midbrain [0 ⫺25 ⫺15]
Dorsal pons [0 ⫺37 ⫺35]
Ventral pons [0 ⫺28 ⫺35]
Dorsal medulla [0 ⫺46 ⫺54]
Medial medulla [0 ⫺40 ⫺53]
Ventral medulla [0 ⫺41 ⫺54]
Cerebellar regions
Dentate nucleus [0 ⫺61 ⫺38]
Fastigial nucleus [⫺1 ⫺39 ⫺22]
Vermis [0 ⫺63 ⫺25]
Group
Difference
Control
1655
1656
HARPER ET AL.
Thalamic and midbrain areas
The cluster of deficient responses from the posterior dorsal,
ventral, and medial thalamus through the medial midbrain and
extending to an area of the dorsal lateral pons in which the
locus coeruleus is sited likely relates to chemoresponsive roles.
The locus coeruleus contains chemosensitive neurons (Andrzejewski et al. 2001; Oyamada et al. 1998). Although the role
in CO2 modulation is unclear, appropriate respiratory responses to low O2 in fetal sheep are dependent on dorsal
thalamic and midbrain structures (Koos et al. 1998), and the
posterior thalamus shows c-fos expression to hypoxia (Sica et
al. 2000). Deficient responses to hypoxia also emerged in the
posterior thalamus, midbrain, and dorsal pons of CCHS cases
(Macey et al. 2005b). Mediation of both low oxygen and
hypercapnia apparently are affected in diencephalic and midbrain areas in the syndrome.
PHOX2B targets
The dorsal medulla, which contains among other structures
the NTS, showed deficient responses to hypercapnia and would
be expected to be affected by mutations in PHOX2B transcription, a mutation found in a high proportion of CCHS patients
J Neurophysiol • VOL
(Amiel et al. 2003; Weese-Mayer et al. 2003). The dorsal
medulla, however, showed no change from a control pattern to
a cold pressor challenge (Macey et al. 2005a), which elicited
substantial cardiovascular and breathing responses, both of
which are deficient in CCHS (Kim et al. 2002). Collectively,
the findings suggest that the syndrome may result from multiple failures in genetic expression; other genetic processes in
central breathing control are currently being explored (Rhee et
al. 2004).
Temporal patterns
Among physiological responses, an increase in respiratoryrelated heart rate variation began immediately, in temporal
association with the altered difference in signal in the cerebellum and associated motor structures. An early onset was also
noted in the dorsal medulla, insula, and amygdala, the latter
structure presumably related to affective components of the
signal. A 60-s nadir in response in the hippocampus for control
over CCHS subjects overlapped a decline in respiratory rate
found in controls, but not found in CCHS patients. The hippocampus has been previously implicated in aspects of respiratory timing and may contribute to conditions for inspiratory
drive (Poe et al. 1996).
Limitations
When these data were collected, the sample period (⬃2 min)
was the maximum available due to scanner constraints. Ideally,
longer scanning times, with multiple stimulation periods,
would be used. Additionally, the results showed early and
transient responses, and therefore higher temporal resolution
would have allowed for detailed examination of timing patterns. Recently developed scanners have the capabilities to
implement improved protocols.
The analyses revealed several regions where early, but
transient responses occurred, patterns consistent with rapid
chemoreceptor responses. The cluster analysis was performed
using a pattern (step function) which highlighted regions differing between the groups either throughout the challenge or at
specific periods. Ideally, further cluster analyses would be
performed using other models, e.g., an initial rapid but transient response to more specifically highlight areas responding
in such a pattern. However, the difficulties associated with
specifying a complex model or of using several separate
models were such that we opted for the simpler step function
pattern, with the additional aspect of presenting time trends
from clusters to allow for determination of response timing.
Global signal changes
Hypercapnia, by inducing substantial cerebral vasodilation
(Kety and Schmidt 1948), results in large global BOLD signal
changes, on which regional changes are superimposed. Others
have evaluated regional responses to large global changes,
including primary visual cortex responses to photic stimulation
after hypercapnic challenges (Corfield et al. 2001; Li et al.
2000) and corrected for such overall effects by intensity normalization procedures. The global signal here increased in both
groups 20 s after challenge onset and increased further in
controls later in the challenge (Macey et al. 2003). Differences
appeared between gray and white matter levels of response, so
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
retained in CCHS, despite the loss of dyspnea to high CO2
(Shea et al. 1993). Thus CCHS subjects are capable of experiencing breathlessness, but only do so in response to exercise,
not chemoreceptor stimulation. It may be the case that characteristics of air hunger are mediated by different structures,
depending on the afferent source, with exertion-related signaling of affect controlled by cingulate-related areas in conjunction with insular and amygdala structures, while chemoreceptor-related affective components are mediated primarily by
insular and amygdala sites.
The deficient responses in the amygdala may contribute to
the loss of air hunger in CCHS and the reduction of respiratory
drive. The amygdala has direct projections to the rostral ventrolateral medullary respiratory group of neurons (Gaytan and
Pasaro 1998) and contains chemosensitive neurons (Teppema
and Dahan 2005).
Both the left and right insula showed group response differences: the right ventral insula as part of a larger response
cluster, with CCHS signals larger and missing an early transient decline found in control subjects, and the left as an
anterior cluster showing a sustained increase in the control
subjects, but no change in CCHS patients. The insula shows
lateralized autonomic properties, with the right insula principally modulating sympathetic characteristics, and the left side
controlling parasympathetic aspects (Cechetto and Chen 1990;
Oppenheimer et al. 1992). Since the enhanced respiratoryrelated influences on heart rate resulting from hypercapnia are
dependent on vagal outflow (to rapidly correct heart rate), a
component of the diminished heart rate variation to the challenge in CCHS may derive from reduced parasympathetic
control mediated by the left anterior insula. The right insula
response characteristics provide a basis for the exaggerated
sympathetic responses encountered in CCHS (Weese-Mayer et
al. 1992), since the right insula often provides an inhibitory or
disfacilitatory action on sympathetic action (Henderson et al.
2003).
RESPIRATORY CONTROL IN HYPERCAPNIA
global trends in gray matter were separated from those in fibers
to preclude differential overall effect contributions for the later
detrending analysis. The overall signals were removed to
examine regional responses using a voxel-by-voxel detrending
procedure (Macey et al. 2004b), a more conservative method
than intensity normalization. The use of detrending procedures,
together with response trends that showed closely adjacent
areas with differing patterns of response, argue that the findings result from differential recruitment of brain areas and not
from global changes.
Summary
ACKNOWLEDGMENTS
The authors thank Dr. Munawar Saeed, A. Kim, R. K. Harper, and C.
Valderama for technical support and Dr. Rajesh Kumar for editorial assistance.
GRANTS
This research was supported by National Institute of Child Health and
Human Development Grant HD-22695.
REFERENCES
Amaral DG. The amygdala, social behavior, and danger detection. Ann NY
Acad Sci 1000: 337–347, 2003.
American Thoracic Society. Idiopathic congenital central hypoventilation
syndrome: diagnosis and management. Am J Respir Crit Care Med 160:
368 –373, 1999.
Amiel J, Laudier B, Attie-Bitach T, Trang H, de Pontual L, Gener B,
Trochet D, Etchevers H, Ray P, Simonneau M, Vekemans M, Munnich
A, Gaultier C, and Lyonnet S. Polyalanine expansion and frameshift
mutations of the paired-like homeobox gene PHOX2B in congenital central
hypoventilation syndrome. Nat Genet 33: 459 – 461, 2003.
Andrzejewski M, Muckenhoff K, Scheid P, and Ballantyne D. Synchronized rhythms in chemosensitive neurons of the locus coeruleus in the
absence of chemical synaptic transmission. Respir Physiol 129: 123–140,
2001.
Ashburner J and Friston K. Multimodal image coregistration and partitioning—a unified framework. NeuroImage 6: 209 –217, 1997.
J Neurophysiol • VOL
Banzett RB, Mulnier HE, Murphy K, Rosen SD, Wise RJ, and Adams L.
Breathlessness in humans activates insular cortex. Neuroreport 11: 2117–
2120, 2000.
Brannan S, Liotti M, Egan G, Shade R, Madden L, Robillard R, Abplanalp B, Stofer K, Denton D, and Fox PT. Neuroimaging of cerebral
activations and deactivations associated with hypercapnia and hunger for air.
Proc Natl Acad Sci USA 98: 2029 –2034, 2001.
Carpenter M and Batton R. Connections of the fastigial nucleus in the cat
and monkey. In: The Cerebellum—New Vistas, edited by Palay S and
Chan-Palay V. Berlin: Springer-Verlag, 1982, p. 250 –295.
Cechetto DF and Chen SJ. Subcortical sites mediating sympathetic responses
from insular cortex in rats. Am J Physiol 258: R245–R255, 1990.
Chen ML and Keens TG. Congenital central hypoventilation syndrome: not
just another rare disorder. Paediatr Respir Rev 5: 182–189, 2004.
Commare MC, Francois B, Estournet B, and Barois A. Ondine’s curse: a
discussion of five cases. Neuropediatrics 24: 313–318, 1993.
Corfield DR, Fink GR, Ramsay SC, Murphy K, Harty HR, Watson JD,
Adams L, Frackowiak RS, and Guz A. Evidence for limbic system
activation during CO2-stimulated breathing in man. J Physiol 488: 77– 84,
1995.
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.
Dauger S, Pattyn A, Lofaso F, Gaultier C, Goridis C, Gallego J, and
Brunet JF. PHOX2B controls the development of peripheral chemoreceptors and afferent visceral pathways. Development 130: 6635– 6642, 2003.
Dejours P. The regulation of ventilation during muscular exercise in man.
J Physiol (Paris) 51: 929 –935, 1959.
Evans KC, Banzett RB, Adams L, McKay L, Frackowiak RSJ, and
Corfield DR. BOLD fMRI identifies limbic, paralimbic, and cerebellar
activation during air hunger. J Neurophysiol 88: 1500 –1511, 2002.
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.
Gaytan SP and Pasaro R. Connections of the rostral ventral respiratory
neuronal cell group: an anterograde and retrograde tracing study in the rat.
Brain Res Bull 47: 625– 642, 1998.
Gozal D, Hathout GM, Kirlew KA, Tang H, Woo MS, Zhang J, Lufkin
RB, and Harper RM. Localization of putative neural respiratory regions in
the human by functional magnetic resonance imaging. J Appl Physiol 76:
2076 –2083, 1994.
Gozal D, Marcus CL, Shoseyov D, and Keens TG. Peripheral chemoreceptor
function in children with the congenital central hypoventilation syndrome.
J Appl Physiol 74: 379 –387, 1993.
Gozal D, Marcus CL, Ward SL, and Keens TG. Ventilatory responses to
passive leg motion in children with congenital central hypoventilation
syndrome. Am J Respir Crit Care Med 153: 761–768, 1996.
Gozal D and Simakajornboon N. Passive motion of the extremities modifies
alveolar ventilation during sleep in patients with congenital central hypoventilation syndrome. Am J Respir Crit Care Med 162: 1747–1751, 2000.
Haddad GG, Mazza NM, Defendini R, Blanc WA, Driscoll JM, Epstein
MA, Epstein RA, and Mellins RB. Congenital failure of automatic control
of ventilation, gastrointestinal motility and heart rate. Medicine (Baltimore)
57: 517–526, 1978.
Harper RM, Gozal D, Bandler R, Spriggs D, Lee J, and Alger J. Regional
brain activation in humans during respiratory and blood pressure challenges.
Clin Exp Pharmacol Physiol 25: 483– 486, 1998.
Henderson LA, Woo MA, Macey PM, Macey KE, Frysinger RC, Alger
JR, Yan-Go F, and Harper RM. Neural responses during Valsalva
maneuvers in obstructive sleep apnea syndrome. J Appl Physiol 94: 1063–
1074, 2003.
Kastrup A, Kruger G, Glover GH, Neumann-Haefelin T, and Moseley
ME. Regional variability of cerebral blood oxygenation response to hypercapnia. NeuroImage 10: 675– 681, 1999.
Kety S and Schmidt C. The effects of altered arterial tensions of carbon
dioxide and oxygen on cerebral blood flow and cerebral oxygen consumption of normal young men. J Clin Invest 27: 484 – 492, 1948.
Kim AH, Macey PM, Woo MA, Yu PL, Keen TG, Gozal D, and Harper
RM. Cardiac responses to pressor challenges in congenital central hypoventilation syndrome. Somnologie 6: 109 –115, 2002.
Koos BJ, Chau A, Matsuura M, Punla O, and Kruger L. Thalamic locus
mediates hypoxic inhibition of breathing in fetal sheep. J Neurophysiol 79:
2383–2393, 1998.
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
The control responses and the fMRI response deficits in
CCHS revealed neural processes for hypercapnia that included
brain regions not usually considered to mediate respiratory
drive. Since chemoreceptor signals reach at least some central
sites in CCHS, and components of motor output are intact,
major breathing deficits must lie in integrating input and output
processes. That integration loss may derive from failure of
cerebellar and diencephalic sites to adequately mediate chemosensory signals and likely includes several deficient processes:
1) inadequate reception or translation of signals from chemosensitive regions in cerebellar deep nuclei to appropriate motoric control structures in more rostral areas, 2) disruption of
posterior thalamic, midbrain, and dorsal pontine areas in integrating sensory information to pontine arousal sites, 3) failure
to recruit the discomfort of breathlessness in limbic sites (the
discomfort in controls may be below the level of conscious
awareness), and 4) failure of limbic (insula, amygdala) and
cerebellar sites to appropriately regulate autonomic outflow.
These neural deficits may result from specific syndrome-related damage to limbic sites whose development may have
been affected by the defective targeting resulting from altered
PHOX2B or other gene expression, and a portion of the deficits
may arise from syndrome-related pathology in overall vascular
reactivity in the brain.
1657
1658
HARPER ET AL.
J Neurophysiol • VOL
cating cerebellum in the experience of hypercapnia and hunger for air. Proc
Natl Acad Sci USA 98: 2041–2046, 2001.
Paton JY, Swaminathan S, Sargent CW, and Keens TG. Hypoxic and
hypercapnic ventilatory responses in awake children with congenital central
hypoventilation syndrome. Am Rev Respir Dis 140: 368 –372, 1989.
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.
Person RJ, Andrezik JA, Dormer KJ, and Foreman RD. Fastigial nucleus
projections in the midbrain and thalamus in dogs. Neuroscience 18: 105–
120, 1986.
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.
Rhee JW, Arata A, Selleri L, Jacobs Y, Arata S, Onimaru H, and Cleary
ML. Pbx3 deficiency results in central hypoventilation. Am J Pathol 165:
1343–1350, 2004.
Schadé J and McMenemey W. Selective Vulnerability of the Brain in
Hypoxaemia. Philadelphia, PA: A. A. Davis Co., 1963.
Shea SA, Andres LP, Shannon DC, Guz A, and Banzett RB. Respiratory
sensations in subjects who lack a ventilatory response to CO2. Respir
Physiol 93: 203–219, 1993.
Sica AL, Greenberg HE, Scharf SM, and Ruggiero DA. Chronic-intermittent hypoxia induces immediate early gene expression in the midline
thalamus and epithalamus. Brain Res 883: 224 –228, 2000.
Simon PM, Schwartzstein RM, Weiss JW, Lahive K, Fencl V, Teghtsoonian M, and Weinberger SE. Distinguishable sensations of breathlessness
induced in normal volunteers. Am Rev Respir Dis 140: 1021–1027, 1989.
Teppema LJ and Dahan A. Central chemoreceptors. In: Pharmacology and
Pathophysiology of the Control of Breathing, edited by Ward DS, Dahan A,
and Teppema LJ. New York: Marcel Dekker, 2005.
Vanderlaan M, Holbrook CR, Wang M, Tuell A, and Gozal D. Epidemiologic survey of 196 patients with congenital central hypoventilation syndrome. Pediatr Pulmonol 37: 217–229, 2004.
Weese-Mayer DE, Berry-Kravis EM, Zhou L, Maher BS, Silvestri JM,
Curran ME, and Marazita ML. Idiopathic congenital central hypoventilation syndrome: analysis of genes pertinent to early autonomic nervous
system embryologic development and identification of mutations in
PHOX2b. Am J Med Genet 123A: 267–278, 2003.
Weese-Mayer DE, Silvestri JM, Menzies LJ, Morrow-Kenny AS, Hunt
CE, and Hauptman SA. Congenital central hypoventilation syndrome:
diagnosis, management, and long-term outcome in thirty-two children.
J Pediatr 120: 381–387, 1992.
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.
Woo MS, Woo MA, Gozal D, Jansen MT, Keens TG, and Harper RM.
Heart rate variability in congenital central hypoventilation syndrome. Pediatr Res 31: 291–296, 1992.
Xu F and Frazier DT. Cerebellar role in the load-compensating response of
expiratory muscle. J Appl Physiol 77: 1232–1238, 1994.
Xu F and Frazier DT. Respiratory-related neurons of the fastigial nucleus in
response to chemical and mechanical challenges. J Appl Physiol 82: 1177–
1184, 1997.
Xu F and Frazier DT. Role of the cerebellar deep nuclei in respiratory
modulation. Cerebellum 1: 35– 40, 2002.
93 • MARCH 2005 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.4 on June 17, 2017
Li TQ, Kastrup A, Moseley ME, and Glover GH. Changes in baseline
cerebral blood flow in humans do not influence regional cerebral blood flow
response to photic stimulation. J Magn Reson Imaging 12: 757–762, 2000.
Littell RC, Milliken GA, Stroup WW, and Wolfinger RD. SAS System for
Mixed Models. Cary, NC: SAS Institute, 1996.
Logothetis NK, Pauls J, Augath M, Trinath T, and Oeltermann A.
Neurophysiological investigation of the basis of the fMRI signal. Nature
412: 150 –157, 2001.
Lutherer LO and Williams JL. Stimulating fastigial nucleus pressor region
elicits patterned respiratory responses. Am J Physiol 250: R418 –R426,
1986.
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.
Macey KE, Macey PM, Woo MA, Harper RK, Alger JR, Keen TG, and
Harper RM. fMRI signal changes to forced expiratory loading in congenital
central hypoventilation syndrome. J Appl Physiol 97: 1897–1907, 2004a.
Macey PM, Alger JR, Kumar R, Macey KE, Woo MA, and Harper RM.
Global BOLD MRI changes to ventilatory challenges in congenital central
hypoventilation syndrome. Respir Physiol Neurobiol 139: 41–50, 2003.
Macey PM, Macey KE, Kumar R, and Harper RM. A method for removal
of global effects from fMRI time series. Neuroimage 22: 360 –366, 2004b.
Macey PM, Macey KE, Woo MA, Keens TG, and Harper RM. Aberrant
neural responses to cold pressor challenges in congenital central hypoventilation syndrome. Pediatr Res In press (a).
Macey PM, Valderama C, Kim AH, Woo MA, Gozal D, Keens TG,
Harper RK, and Harper RM. Temporal trends of cardiac and respiratory
responses to ventilatory challenges in congenital central hypoventilation
syndrome. Pediatr Res 55: 953–959, 2004c.
Macey PM, Woo MA, Macey KE, Keen TG, Saeed MM, Alger J, and
Harper RM. Hypoxia reveals posterior thalamic, cerebellar, midbrain and
limbic deficits in congenital central hypoventilation syndrome. J Appl
Physiol In press (b).
Marcus CL, Bautista DB, Amihyia A, Ward SL, and Keens TG. Hypercapneic arousal responses in children with congenital central hypoventilation syndrome. Pediatrics 88: 993–998, 1991.
Moosavi SH, Golestanian E, Binks AP, Lansing RW, Brown R, and
Banzett RB. Hypoxic and hypercapnic drives to breathe generate equivalent
levels of air hunger in humans. J Appl Physiol 94: 141–154, 2003.
Ogawa S, Lee TM, Kay AR, and Tank DW. Brain magnetic resonance
imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci
USA 87: 9868 –9872, 1990.
O’Hearn E, and Molliver ME. The olivocerebellar projection mediates
ibogaine-induced degeneration of Purkinje cells: a model of indirect, transsynaptic excitotoxicity. J Neurosci 17: 8828 – 8841, 1997.
Ohtsuka K. Inhibitory action of Purkinje cells in the posterior vermis on
fastigio-prepositus circuit of the cat. Brain Res 455: 153–156, 1988.
Oppenheimer SM, Gelb A, Girvin JP, and Hachinski VC. Cardiovascular
effects of human insular cortex stimulation. Neurology 42: 1727–1732,
1992.
Oren J, Kelly DH, and Shannon DC. Long-term follow-up of children with
congenital central hypoventilation syndrome. Pediatrics 80: 375–380, 1987.
Oyamada Y, Ballantyne D, Muckenhoff K, and Scheid P. Respirationmodulated membrane potential and chemosensitivity of locus coeruleus
neurones in the in vitro brainstem-spinal cord of the neonatal rat. J Physiol
513: 381–398, 1998.
Parsons LM, Egan G, Liotti M, Brannan S, Denton D, Shade R, Robillard
R, Madden L, Abplanalp B, and Fox PT. Neuroimaging evidence impli-