| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Otorhinolaryngology, Head and Neck Surgery (Drs. Boudewyns and Van de Heyning) and Pulmonary Medicine (Dr. De Backer), University Hospital, Antwerp, Belgium; and the Department of Pulmonary Medicine (Drs. Punjabi, O’Donnell, Schneider, Smith, and Schwartz), Johns Hopkins University, Baltimore, MD.
Correspondence to: An Boudewyns, MD, PhD, University Hospital Antwerp, Department of Otorhinolaryngology, Head and Neck Surgery, Wilrijkstraat 10, 2650 Edegem, Belgium; e-mail: an.boudewijns{at}uza.uia.ac.be
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Design: Pressure-flow relationships were constructed in the supine and lateral recumbent positions (nonrapid eye movement [NREM] sleep, n = 10) and in the supine position (rapid eye movement [REM] sleep, n = 5).
Setting: University Hospital Antwerp, Belgium.
Patients: Ten obese patients (body mass index, 32.0 ± 5.6 kg/m2) with severe OSA (respiratory disturbance index, 63.0 ± 14.6 events/h) were studied.
Interventions: Pressure-flow relationships were constructed from breaths obtained during a series of step decreases in nasal pressure (34.1 ± 6.5 runs over 3.6 ± 1.2 h) in NREM sleep and during 7.8 ± 2.2 runs over 0.8 ± 0.6 h in REM sleep.
Results: Maximal inspiratory airflow reached a steady state in the third through fifth breaths following a decrease in nasal pressure. Analysis of pressure-flow relationships derived from these breaths showed that Pcrit fell from 1.8 (95% CI, -0.1 to 2.7) cm H2O in the supine position to -1.1 cm H2O (95% CI, -1.8 to 0.4 cm H2O; p = 0.009) in the lateral recumbent position, whereas RN did not change significantly. In contrast, no significant effect of sleep stage was found on either Pcrit or RN.
Conclusions: Our methods for delineating upper airway pressure-flow relationships during sleep allow for multiple determinations of Pcrit within a single night from which small yet significant differences can be discerned between study conditions.
Key Words: collapsibility • sleep apnea • upper airway
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our previous studies, a pressure-flow relationship was constructed during sleep from which upper airway collapsibility and resistance upstream to the site of collapse could be inferred. This pressure-flow relationship was used to define upper airway collapsibility as the level of nasal mask pressure (Pn) below which the upper airway closed (critical closing pressure [Pcrit]) and the upstream resistance (RN) as the inverse of the slope of the pressure-flow relationship.3 Using these methods, we subsequently demonstrated that patients experienced an improvement in their apnea whenever Pcrit declined sufficiently after weight loss or uvulopalatopharyngoplasty.4 5 These observations led us to suggest that quantitative measurements of Pcrit and its response to specific therapeutic maneuvers could help predict the treatment response in polysomnographic indexes of apnea severity.
Nevertheless, assessing the upper airway response to therapeutic maneuvers remained difficult because Pcrit and RN had to be inferred from pressure-flow relationships generated during periods of stable sleep.6 Our previous studies used a lengthy protocol to establish a pressure-flow relationship from multiple measurements of tidal airflow taken over a wide range of Pn during prolonged periods of stable sleep. Clearly, without methods for rapidly constructing such a pressure-flow relationship, it is not possible to determine Pcrit and RN repeatedly under various test conditions within a single night. Before this approach could be put into clinical practice, therefore, it became necessary to streamline the methods for acquiring pressure-flow data and for assessing Pcrit and RN with precision.
We developed an abbreviated method for generating upper airway pressure-flow relationships from multiple breaths during sleep.7 This method involved assessing responses in tidal airflow to brief rather than prolonged perturbations in Pn. In our initial study, however, we found that airflow did not reach a steady-state level within the first three breaths after an abrupt reduction in Pn. In the present study, therefore, we allowed more time to establish steady-state levels of airflow by lowering Pn for up to six breaths. With this protocol, we hypothesized that (1) airflow would reach a steady-state level within a specific number of breaths; (2) steady-state airflow levels could be used to construct a pressure-flow relationship from which Pcrit and RN could be determined; and (3) these variables could be estimated with sufficiently narrow confidence intervals (CIs) to discern differences in upper airway function between different body positions and sleep stages.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Study Design
Each patient underwent baseline full-night polysomnography (PSG)
to characterize the severity of sleep-disordered breathing during NREM
and rapid eye movement (REM) sleep (see "Baseline PSG" below).
Thereafter, patients underwent an additional night of PSG to assess
upper airway function during sleep (see "Experimental Protocol"
below).
Baseline PSG
Recording Methods: Standard PSG techniques were used
to characterize patients at baseline and during the experimental
protocol. In brief, physiologic variables were continuously recorded
including EEG (C4/A1,
C3/A2), electro-oculogram,
chin muscle electromyogram and electromyogram of anterior tibialis
muscle, and ECG. Snoring was measured with a microphone placed at the
suprasternal notch. A body position sensor attached to a thoracic belt
was used to monitor body position. Oxygen saturation was measured by
pulse oximetry (Palco Laboratories; Santa Cruz, CA). Tidal airflow was
monitored with a thermocouple or full face mask connected to a
pneumotachometer (279331; Hamilton Medical; Reno, NV). A balloon probe
(Medtronic Upper Airway; Maastricht, The Netherlands) connected to a
pressure transducer (Response III; Medtronic Upper Airway; Minneapolis,
MN) measured esophageal pressure as previously described.8
The signals from the esophageal balloon were used to measure
respiratory effort. All physiologic signals were digitized at a
frequency of 100 Hz, and stored for further analysis (Windaq/200; Dataq
Instruments; Akron, OH).
Analysis: Sleep stage analysis of nocturnal
PSGs was performed visually according to the criteria outlined by
Rechtschaffen and Kales.9
A 3-s definition for arousal was
used, as per the American Academy of Sleep Medicine.10
Apnea was defined by the complete absence of oronasal airflow for at
least 10 s. Apneas were classified as obstructive, mixed, or
central according to standard criteria.11
Hypopnea was
defined as a > 50% decrease in oronasal airflow accompanied by a
4% drop in oxygen saturation from baseline or an arousal from
sleep. The RDI was calculated as the total number of apneas and
hypopneas per hour of sleep for both NREM and REM sleep, and separately
for the time that the patient slept supine and in the lateral recumbent
position.
Upper Airway Assessment
An additional full-night PSG was performed for the purpose of
assessing upper airway function during sleep as described in the
experimental protocol below. For this protocol, patients were fitted
with a nasal mask (nasal continuous positive airway pressure [CPAP]
mask; Respironics Inc; Murraysville, PA) as above. A chin strap was
used to minimize leakage of air through the mouth when necessary.
Airflow was measured with a pneumotachometer that was connected to the
nasal mask and a pressure transducer (Sefam; Vandoeuvre-les-Nancy,
France). Pn was measured with a similar pressure transducer connected
to a port in the nasal mask. The nasal mask was connected via a
breathing circuit and a bidirectional valve to a positive pressure
source (Tranquility Plus; Healthdyne Technologies; Marietta, GA) and a
negative pressure source (modified Rem-Star unit; Respironics).
Pn was set within the positive range by altering the flow delivered by
the CPAP device through the breathing circuit. When Pn was lowered into
the negative pressure range, the level of subatmospheric pressure was
preset in the negative pressure source. The bidirectional valve was
then switched, thereby connecting the negative pressure source to the
breathing circuit. Thereafter, the valve was switched back to restore a
positive Pn.
Experimental Protocol
During the experimental protocol, each patient was allowed to
initiate sleep while lying either supine or in the lateral recumbent
position with a single pillow placed under the head. During sleep, Pn
was increased stepwise every 5 min as previously
described3
until inspiratory airflow limitation (see
definition below) was abolished. The holding pressure was then defined
as the lowest level of Pn required to eliminate inspiratory airflow
limitation. This pressure corresponded to the minimally "effective
CPAP pressure"12
and to the "upper airway opening
pressure,"13
as previously described by other
investigators. This level of holding Pn was then maintained throughout
the protocol.
During periods of stable NREM (stage II to IV) or REM sleep, Pn was
lowered abruptly during an inspiration, as illustrated in Figure 1
. Pn was then maintained at this level until 30 s had elapsed or an
arousal from sleep had occurred. Thereafter, Pn was raised to holding
pressure, and lowered repeatedly every 1 to 2 min to discrete levels
encompassing the level at which airflow became zero. If arousal
occurred, the patient was allowed to reestablish stable sleep before
continuing the experimental protocol. Runs spanned a range of
approximately 6 cm H2O range in Pn and a range of
zero to approximately 400 mL/s in maximal inspiratory airflow
(
Imax).
|
Data Analysis
With each step decrease in Pn, up to six consecutive breaths
were analyzed (Fig 1)
. Initially, PSG signals were examined to detect
any arousal from sleep during each run. Runs were excluded from
analysis when they precipitated an arousal or sleep stage transition
within the first 20 s of lowering the Pn or occurred during stage
I NREM sleep. Otherwise, all breaths after reductions in Pn were
analyzed until either six breaths or an arousal had occurred. Each of
these breaths was then assessed for the presence or absence of
inspiratory airflow limitation. For each flow-limited breath,
Imax and Pn were measured. The level of
Imax was computed as the difference in airflow
between the onset of the inspiratory effort (see negative deflections
in esophageal pressure in Fig 1
) and the inspiratory flow maximum.
Pressure and flow values were tabulated and used for all subsequent
analysis. Data were also categorized based on body position and sleep
stage (NREM and REM).
Definition of Flow Limitation: Initially, the
inspiratory airflow and esophageal pressure signals were examined to
determine whether inspiratory airflow limitation had occurred.
Inspiratory airflow limitation was defined by the development of a
Imax as the pressure gradient across the airway
continued to increase. In practice, we determined this gradient to be
increasing whenever esophageal pressure continued to fall progressively
beyond the point of Imax. We recognized, however,
that the esophageal pressure swing may have overestimated the driving
pressure across the upper airway by an amount equal to the change in
lung elastic recoil pressure during lung inflation. A change in recoil
pressure of approximately 2.5 cm H2O was
estimated from end-expiration to end-inspiration if a tidal volume of
approximately 0.5 L and lung compliance of approximately 0.2 L/cm
H2O was assumed. Allowing for variations in tidal
volume and lung compliance between breaths and patients, respectively,
airflow limitation were considered to be present for a given
inspiration whenever the esophageal pressure fell 5 cm
H2O beyond the point of Imax.
This definition of inspiratory airflow limitation was then taken to be
the criterion for establishing the presence of airflow limitation in
all flow-limited inspirations that were analyzed during periods of
reduced Pn.
Breath-to-Breath Analysis: The breath-to-breath
response in Imax to step decreases in Pn was
analyzed with a two-factor analysis of variance (ANOVA; repeated
measures design for responses to breath number and Pn level). Separate
ANOVAs were performed for data obtained in the supine and lateral
recumbent positions within NREM sleep. For these ANOVAs, breath number
and Pn level were considered as within subject (repeated measures)
factors for six breath numbers and five Pn levels, respectively.
Post hoc analysis (Scheffé test) was then used to
establish statistical significance between levels within factors.
Additional post hoc comparisons were made to investigate the
interaction between breath number and Pn.
Pressure-Flow Relationships (Pcrit and
RN): In each patient, the relationship between
Imax and Pn was examined for specific breaths after
step reductions in Pn. Least squares linear regression was computed to
examine the response in Imax to changes in Pn for
each experimental condition (ie, body position and sleep
stage). The regression equation was then solved for Pcrit, which was
defined as the Pn below which Imax became
zero. RN was also calculated as the reciprocal of
the slope of this regression equation, as previously
described.3
The CIs for Pcrit and RN were calculated as follows. The
inverse regression method14
was used for generating 95%
confidence bands around the Imax vs Pn regression
line. The intersections of the upper and lower confidence bands with
the Pn axis (where Imax becomes zero) provide the
lower and upper 95% fiducial limits for Pcrit, respectively. The
95% CIs for RN were computed as the reciprocal of the
upper and lower limits of the 95% CI for the slope of the regression
equation, Imax vs Pn (Fig 2
). The upper or lower bounds of a CI were considered indeterminate and
not reported when these bounds were > 5 cm H2O
and 50 cm H2O/L/s away from point estimates in
Pcrit and RN, respectively.
|
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Imax Responses to Step
Decreases in Pn
After step decreases in Pn during NREM sleep, we observed significant
alterations in Imax within the six ensuing breaths
for supine (F = 11.5; degrees of freedom [df], = 5;
p < 0.001) and lateral recumbent (F = 9.9; df = 5;
p < 0.001) positions. Post hoc analysis revealed no
significant differences in Imax for the
third through fifth breaths in either the supine or lateral recumbent
position. Rather, we found that Imax for
these breaths was significantly lower than that for the first and
second breaths in the supine position, and lower than the first and
sixth breaths in the lateral recumbent position. Stated otherwise, our
findings indicate that Imax fell to a
quasisteady-state level for the third through fifth breaths after step
decreases in Pn in both body positions (Table 2
).
|
Imax
(F = 7.3; df = 4; p < 0.001; and F = 8.2;
df = 4; p < 0.001) for supine and lateral recumbent positions,
respectively. Finally, a significant interaction between Pn and breath
number for both the NREM supine and lateral recumbent conditions
(F = 5.2; df = 20; p < 0.001; and
F = 3.0; df = 20; p < 0.001) was demonstrated.
Specifically, we found that the breath-to-breath differences in
Imax were most pronounced at lower Pn levels. We
conclude from the analysis that breath-to-breath changes in
Imax occurred in response to step reductions in Pn,
particularly at the lowest pressure levels, but that
Imax in the third through fifth breaths reached a
quasisteady state.
Variables of Upper Airway Function
Effect of Body Position: Having demonstrated stable
levels of Imax for the third through fifth breaths,
we then constructed Imax vs Pn relationships from
data obtained in the supine and lateral recumbent positions, as
illustrated in Figure 2
for a representative patient. From these
relationships, the Pcrit and RN, and the CIs around these
variables, were calculated, as represented in Table 3
.
|
|
|
|
Imax from 0.0 ± 0.0 to 88.2 ± 106.2 mL/s when
these patients were breathing at atmospheric Pn in the supine and
lateral recumbent positions, respectively. Finally, the RDI, apnea
index, and hypopnea index did not change in patients whose Pcrit
remained above atmospheric in the lateral recumbent position (n = 4).
These findings are consistent with the notion that the response in RDI
and its component apnea and hypopnea indexes depend on the level to
which Pcrit falls (see "Discussion"). Regarding sleep stage–related changes in Pcrit, we found no significant change in RDI between NREM and REM sleep (60.8 ± 18.2 episodes/h vs 54.6 ± 8.6 episodes/h, respectively; n = 5). Moreover, no significant correlation existed between changes in Pcrit and RDI in response to alterations in either body position or sleep stage. Rather, a significant correlation between the lateral recumbent Pcrit and RDI (p = 0.04) and between the supine Pcrit and BMI (p = 0.001) was found.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Imax fell to a relatively
low level within the first three breaths and remained stable at this
level through the fifth breath. Having demonstrated a quasisteady-state
response in airflow for breaths three through five, we then used these
breaths to construct pressure-flow relationships for NREM (supine and
lateral recumbent position) and REM (supine only) sleep. From these
relationships, we characterized two variables of upper airway
function—Pcrit and RN—both of which determine the
severity of airflow obstruction during sleep. With repeated
measurements of these variables, our methods allowed us to discern
small decreases in Pcrit as patients moved from supine to lateral
recumbent position. When determinations of CI around Pcrit were
examined, we found that a RN, H2O/L/s
significant positional difference in Pcrit was evident within 4 of 10
of our patients. In contrast, no consistent positional change in
RN was detected, and little significant sleep
stage–related change in either Pcrit or RN was
discerned in the patients who were able to maintain REM sleep during
the protocol. We conclude that our methods for delineating upper airway
pressure-flow relationships during sleep allow for multiple precise
NREM determinations of Pcrit within a single night under different test
conditions.
Our approach to studying upper airway function during sleep builds on
previous studies of upper airway pressure-flow relationships. In the
previous work, a single pressure-flow relationship was constructed from
multiple breaths obtained at several levels of Pn during prolonged
periods of stable sleep.15
Although this approach allowed
investigators to delineate differences in Pcrit among groups with
varying degrees of upper airway obstruction during sleep, it did not
afford them the opportunity to make repeated measurements of Pcrit and
RN under several experimental conditions in one patient
within a single night. To investigate the variability in upper airway
function within a single night, it was necessary to develop precise
methods for constructing several pressure-flow relationships during
different sleep stages and body positions. Such comparisons required
that data acquisition be streamlined; yet the data had to reflect
steady-state responses of the upper airway to specific test conditions.
Initially, an abbreviated method was piloted in which Pn was lowered
repeatedly from a holding pressure for three consecutive
breaths.7
In this study, we found that
Imax decreased steadily over these three breaths,
leading us to conclude that a steady-state pressure-flow relationship
could not be constructed from these breaths. We therefore extended the
period during which Pn was lowered, and found that a quasisteady state
in Imax could be achieved between the third and
fifth breaths after a sudden drop in Pn. These breaths were then used
to construct multiple pressure-flow relationships from which Pcrit and
RN were estimated in different body positions and sleep
stages. Although our methods permitted two estimates of these variables
in NREM sleep (supine and lateral recumbent data), it provided REM data
in only half our patients.
There are certain limitations of our methods for determining Pcrit and RN. These limitations derive from the fact that Pcrit is a variable inferred from linear regression of the pressure-flow relationship. A natural consequence of this fact is that the precision of our estimate is determined by factors influencing the regression analysis itself. The first factor influencing the precision relates to the scatter of the data points around the regression line. When significant scatter exists, we expect the CI to widen substantially. Under these circumstances, we interpret such widening to be a reflection of the biologic variability in the Pcrit over time. Although such variability may be related to changes in sleep stage over time,13 16 such differences would be minimized with our protocol, which shortens the elapsed time required to acquire pressure-flow data.
Another factor influencing the size of the CI around Pcrit relates to
the distribution of data points along the regression axes. When few
data exist in the low range of pressure and flow
(Imax of 0 to 100 mL/s), the CI around Pcrit would
widen substantially, particularly if back-extrapolation of the
pressure-flow relationship to the x-axis is required.
Although our protocol stipulated that low-flow data be collected for
each patient in the present study, the CI was either indeterminate (see
patients 1, 2, and 9 in Fig 4
) or excessively wide (patient 5) because
premature arousals precluded adequate sampling of low-flow data. It is
therefore important that Pn be lowered repeatedly into the range
associated with near-zero levels of airflow to minimize statistical
uncertainty in estimating the Pcrit. Thus, low-flow data are required
to obtain CI around Pcrit, which reflect the underlying biologic
variability rather than statistical uncertainty of this estimate.
CIs for RN, however, are determined by the precision with
which the inverse of the slope of the regression equation can be
defined. This variable is best estimated when data are collected over a
widely dispersed range of Pn. In practice, data should encompass the
range of Pn extending from Pcrit to approximately 6 cm
H2O above Pcrit. The upper limit of this range
corresponds to the "upper airway opening pressure" as described by
Issa and Sullivan13
or the "effective CPAP pressure"
described by Condos et al.12
This upper limit cannot be
extended beyond this range because Imax cannot be
determined at higher pressures, which abolish inspiratory airflow
limitation. In our protocol, a stringent criterion for establishing the
presence of airflow limitation was also established (see "Materials
and Methods"), thereby further reducing the range of Pn over which
data were collected. We also constrained the range of Pn applied by
adopting a holding pressure equal to the minimum Pn required to
eliminate inspiratory airflow limitation. As a result of efforts to
standardize the holding pressure and constrain the distribution of data
over the Pn range, our CIs for RN are relatively large,
suggesting that our method for estimating RN is somewhat
less precise than that for Pcrit.
Our estimates of upper airway critical pressures during NREM
sleep are consistent with those previously generated for the supine
condition in NREM sleep.2
Specifically, our Pcrit of
1.8 ± 1.5 cm H2O was nearly identical to
previous estimates by us of 2.5 ± 1.5 cm H2O
(Gleadhill et al)2
and others 2.1 ± 0.1 cm
H2O.16
Of note, the earlier
estimates were derived from pressure-flow relationships generated
during prolonged periods of sleep at various levels of Pn. The fact
that our results with an abbreviated method compare favorably to those
obtained with a steady-state method suggests that our method accurately
characterizes upper airway collapsibility in apneic patients during
sleep. We recognize, however, that reflex responses may alter results
obtained from this abbreviated method,17
as evidenced by
the increase in Imax that we detected in the sixth
breath after step decreases in Pn in the supine position. The
relationship between upper airway neuromuscular activity and Pcrit in
sleeping humans, however, is still unclear inasmuch as no differences
in Pcrit were detected between NREM and REM sleep despite large
differences in genioglossal activity.7
18
Moreover, the
two methods have not yet been compared head-to-head, and the accuracy
of the newer method is not yet proven in asymptomatic snorers and
normal individuals with subatmospheric levels of critical pressure.
With repeated measurements of Pcrit within a single night, we can now examine the sources of variability in upper airway function throughout the night. In the present study, our methods allowed us to discern a statistically significant, albeit modest, decrease in Pcrit between the supine and lateral recumbent positions for a group as a whole. Our findings regarding positional changes are consistent with previous studies13 19 demonstrating a similar direction and magnitude of change in Pcrit. In addition, our methods have allowed us to extend the previous findings in two ways. First, we have established that a lessening in the severity of upper airway obstruction in the lateral recumbent position can be attributed to alterations in collapsibility (Pcrit) rather than RN. Second, our calculations of CI have now allowed for comparisons of Pcrit within individuals. When CI did not overlap, we took this to indicate that Pcrit decreased significantly in the lateral recumbent position. Thus, our methods have helped to establish the precise mechanism for relief of upper airway obstruction (Pcrit vs RN) in patients with OSA as well as the impact of positional changes on the severity of upper airway obstruction in each patient.
It is perhaps surprising that the RDI did not fall when patients slept in the lateral recumbent position, despite the significant decrease in Pcrit with this maneuver. Although a number of possibilities might account for this finding,20 we believe that the Pcrit levels in each body position can explain the lack of response in RDI as follows. In previous studies, we observed that little decrease in RDI occurred unless the Pcrit fell below approximately -4 cm H2O.4 5 Indeed, reductions in Pcrit to this level would account for positional responses in RDI in previous reports.21 22 In our patients, however, the supine Pcrit was above atmospheric pressure. With only modest reductions in Pcrit in the lateral recumbent position, Pcrit still exceeded the -4 cm H2O threshold in all but one patient. With Pcrit remaining only minimally negative, the RDI did not fall significantly. Rather, a shift in the distribution of apneas and hypopneas was found in those patients whose Pcrit fell below atmospheric, consistent with findings from an earlier cross-sectional study in our laboratory.2 Thus, it appears that the Pcrit response was too small for the RDI to decrease in our patients, and might be predicted to lower Pcrit sufficiently only in selected patients whose Pcrit is already subatmospheric in the supine position. We would also predict that partial responses in RDI would occur in selected patients in whom the CI overlaps the -4 cm H2O threshold in the lateral recumbent position. We therefore propose that the Pcrit CI be determined in the lateral recumbent position to target selected patients for positional therapy.6
In summary, our protocol incorporates a number of new features that optimize the estimation of Pcrit and RN from the upper airway pressure-flow relationship. We now believe that rapid, reliable estimates of these variables can be obtained within a single sleep cycle if such a standardized approach is taken to establish both the holding pressure and the range over which the Pn is lowered. Using this protocol, we have also demonstrated the variability in Pcrit related to body position. These latter findings lead us to conclude that better estimates will be obtained when the acquisition of pressure-flow data is restricted to a specific sleep stage and body position. In further studies, our methods may help investigate clinical and physiologic factors influencing the severity of upper airway obstruction during sleep, and may serve to guide clinicians in their selection of therapy for these patients.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Imax = maximal inspiratory airflow Supported by HL 503781, HL 37379, and the Fund of Scientific Research Flanders (FWO), Belgium.
Received for publication November 18, 1999. Accepted for publication May 3, 2000.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. S. Litman, J. M. McDonough, C. L. Marcus, A. R. Schwartz, and D. S. Ward Upper airway collapsibility in anesthetized children. Anesth. Analg., March 1, 2006; 102(3): 750 - 754. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Jordan, A. Wellman, J. K. Edwards, K. Schory, L. Dover, M. MacDonald, S. R. Patel, R. B. Fogel, A. Malhotra, and D. P. White Respiratory control stability and upper airway collapsibility in men and women with obstructive sleep apnea J Appl Physiol, November 1, 2005; 99(5): 2020 - 2027. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Inazawa, T. Ayuse, S. Kurata, I. Okayasu, E. Sakamoto, K. Oi, H. Schneider, and A.R. Schwartz Effect of Mandibular Position on Upper Airway Collapsibility and Resistance J. Dent. Res., June 1, 2005; 84(6): 554 - 558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Farre, J.M. Montserrat, and D. Navajas Noninvasive monitoring of respiratory mechanics during sleep Eur. Respir. J., December 1, 2004; 24(6): 1052 - 1060. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Wellman, A. S. Jordan, A. Malhotra, R. B. Fogel, E. S. Katz, K. Schory, J. K. Edwards, and D. P. White Ventilatory Control and Airway Anatomy in Obstructive Sleep Apnea Am. J. Respir. Crit. Care Med., December 1, 2004; 170(11): 1225 - 1232. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ayuse, T. Inazawa, S. Kurata, I. Okayasu, E. Sakamoto, K. Oi, H. Schneider, and A.R. Schwartz Mouth-opening Increases Upper-airway Collapsibility without Changing Resistance during Midazolam Sedation J. Dent. Res., September 1, 2004; 83(9): 718 - 722. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Marcus, L. B. Fernandes Do Prado, J. Lutz, E. S. Katz, C. A. Black, P. Galster, and K. A. Carson Developmental changes in upper airway dynamics J Appl Physiol, July 1, 2004; 97(1): 98 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P. Patil, N. M. Punjabi, H. Schneider, C. P. O'Donnell, P. L. Smith, and A. R. Schwartz A Simplified Method for Measuring Critical Pressures during Sleep in the Clinical Setting Am. J. Respir. Crit. Care Med., July 1, 2004; 170(1): 86 - 93. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Schwartz, H. Schneider, and P. L. Smith Upper airway surface tension: is it a significant cause of airflow obstruction during sleep? J Appl Physiol, November 1, 2003; 95(5): 1759 - 1760. [Full Text] [PDF]
|
|
|
|
|
|
A. Oliven, D. J. O'Hearn, A. Boudewyns, M. Odeh, W. De Backer, P. van de Heyning, P. L. Smith, D. W. Eisele, L. Allan, H. Schneider, et al. Upper airway response to electrical stimulation of the genioglossus in obstructive sleep apnea J Appl Physiol, November 1, 2003; 95(5): 2023 - 2029. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Kirkness, M. Madronio, R. Stavrinou, J. R. Wheatley, and T. C. Amis Relationship between surface tension of upper airway lining liquid and upper airway collapsibility during sleep in obstructive sleep apnea hypopnea syndrome J Appl Physiol, November 1, 2003; 95(5): 1761 - 1766. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Younes Contributions of Upper Airway Mechanics and Control Mechanisms to Severity of Obstructive Apnea Am. J. Respir. Crit. Care Med., September 15, 2003; 168(6): 645 - 658. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Farre, J. Rigau, J. M. Montserrat, L. Buscemi, E. Ballester, and D. Navajas Static and Dynamic Upper Airway Obstruction in Sleep Apnea: Role of the Breathing Gas Properties Am. J. Respir. Crit. Care Med., September 15, 2003; 168(6): 659 - 663. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Schneider, S. P. Patil, S. Canisius, E. A. Gladmon, A. R. Schwartz, C. P. O'Donnell, P. L. Smith, and C. G. Tankersley Hypercapnic duty cycle is an intermediate physiological phenotype linked to mouse chromosome 5 J Appl Physiol, July 1, 2003; 95(1): 11 - 19. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Kirkness, P. R. Eastwood, I. Szollosi, P. R. Platt, J. R. Wheatley, T. C. Amis, and D. R. Hillman Effect of surface tension of mucosal lining liquid on upper airway mechanics in anesthetized humans J Appl Physiol, July 1, 2003; 95(1): 357 - 363. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. L. Smith, C. P. O'Donnell, L. Allan, and A. R. Schwartz A Physiologic Comparison of Nasal and Oral Positive Airway Pressure Chest, March 1, 2003; 123(3): 689 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Schneider, A. Boudewyns, P. L. Smith, C. P. O'Donnell, S. Canisius, A. Stammnitz, L. Allan, and A. R. Schwartz Modulation of upper airway collapsibility during sleep: influence of respiratory phase and flow regimen J Appl Physiol, October 1, 2002; 93(4): 1365 - 1376. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Gold, C. L. Marcus, F. Dipalo, and M. S. Gold Upper Airway Collapsibility During Sleep in Upper Airway Resistance Syndrome* Chest, May 1, 2002; 121(5): 1531 - 1540. [Abstract] [Full Text] [PDF]
|
