HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rowley, J. A. Articles by Badr, M. S. Search for Related Content PubMed PubMed Citation Articles by Rowley, J. A. Articles by Badr, M. S.

The effect of rapid eye movement (REM) sleep on upper airway mechanics in normal human subjects

James A. Rowley, Brian R. Zahn, Mark A. Babcock and M. Safwan Badr

Received 6 January 1998; accepted after revision 16 April 1998.

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

1. It has been proposed that the upper airway is more compliant during rapid eye movement (REM) sleep than during non-rapid eye movement (NREM) sleep. The purpose of this study was to test this hypothesis in a group of subjects without sleep-disordered breathing.

2. On the first night, the effect of sleep stage on the relationship of retropalatal cross-sectional area (CSA; visualized with a fibre-optic scope) to pharyngeal pressure (P_PH) measured at the soft palate during eupnoeic breathing was studied. Breaths during REM sleep were divided into phasic (associated with eye movements) and tonic (not associated with eye movements). There was a significant decrease in pharyngeal CSA during NREM sleep compared with wakefulness. There was no further decrease observed during either tonic or phasic REM sleep. Pharyngeal compliance, defined as the slope of the regression CSA versus P_PH, was significantly increased during NREM sleep compared with wakefulness and REM sleep, with the compliance during both tonic and phasic REM sleep being similar to that observed in wakefulness.

3. On the second night, the effect of sleep stage on pressure-flow relationships of the upper airway was investigated. There was a trend towards the upper airway resistance being highest in NREM sleep compared with wakefulness and REM sleep.

4. We conclude that the upper airway is stiffer and less compliant during REM sleep than during NREM sleep. We postulate that this difference is secondary to differences in upper airway vascular perfusion between REM and NREM sleep.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

Rapid eye movement (REM) sleep is a distinct neurophysiological state associated with significant changes in breathing pattern and ventilatory control as compared with both wakefulness and non-rapid-eye movement (NREM) sleep (Douglas, 1994). In particular, changes in tidal volume, minute ventilation (Millman, Knight, Kline, Short, Chung & Pack, 1988) and the ventilatory responses to hypercapnia (Sullivan, Murphy, Kozar & Phillipson, 1979) have been shown to occur in association with the bursts of phasic eye movements that characterize REM sleep. Given the influence of REM sleep on respiration and ventilatory control, it can be hypothesized that upper airway physiology would also be influenced by REM sleep. This hypothesis is supported by clinical data that show that obstructive apnoeas are more frequent and longer during REM sleep (Findley, Wilhoit & Suratt, 1985; Series, Cormier & LeForge, 1990). However, there is a paucity of data on upper airway physiology during REM sleep. The most studied measure of upper airway patency, upper airway resistance, has been studied during REM sleep and has been found to be the same or decreased compared with NREM sleep (Hudgel, Martin, Johnson & Hill, 1984; Wiegand, Zwillich & White, 1989; Wiegand, Latz, Zwillich & Wiegand, 1990a). No previous work has investigated upper airway compliance during REM sleep. Thus, it is unclear if the upper airway behaves differently during REM and NREM sleep.

Recently, it has been shown that fibre-optic imaging can be used to visualize the upper airway during sleep (Badr, Toiber, Skatrud & Dempsey, 1995; Morrell & Badr, 1998). Previous work from our laboratory showed that there are significant within-breath changes in the cross-sectional area of the retropalatal airway during sleep, particularly in those subjects with evidence of sleep-disordered breathing (Morrell & Badr, 1998). In this study, we have combined measures of cross-sectional area with measures of pharyngeal pressure in order to measure the compliance of the upper airway (Kuna, Bedi & Ryckman, 1988; Isono, Morrison, Launois, Feroah, Whitelaw & Remmers, 1993; Isono, Remmers, Tanaka, Sho, Sato & Nishino, 1997) and to study the hypothesis that the upper airway is more compliant during REM sleep than during NREM sleep. Thus, in this work we investigated the effect of REM sleep on retropalatal cross-sectional area (CSA), within-breath changes in CSA, upper airway compliance, and upper airway resistance in a group of subjects without significant sleep-disordered breathing.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects

The experimental protocol was performed in accordance with the Declaration of Helsinki and was approved by the Human Investigation Committee of the Wayne State University School of Medicine and the Detroit Veterans Affairs Medical Center. Informed written consent was obtained from all subjects. We studied eight subjects who were recruited from the general population. The anthropometric, polysomnographic and clinical data are summarized in Table 1.

Table 1. Anthropometric, polysomnographic and clinical data for each subject

REM enhancement

A major reason for the paucity of data during REM sleep is the difficulty in achieving stable periods of REM sleep of sufficient length in heavily instrumented subjects, whose sleep was frequently interrupted for experimental manipulations. We overcame this problem in several ways. First, we severely restricted the sleep of our subjects prior to their study. Subjects were instructed to sleep no more than 5 h the night prior to the study and were then kept awake the night of the study until approximately 05.00-06.00 h. Thus, all subjects had been awake for approximately 24 h prior to the study. Second, we studied the subjects starting at 05.00 h, which is the time of the day at which the majority of people have the highest propensity for REM sleep (Carskadon & Dement, 1994). Finally, we kept interruptions to a minimum after the initial data collection for wakefulness and NREM sleep.

Measurements

Electroencephalograms (EEG), electro-oculograms (EOG), and chin electromyograms (EMG) were recorded (model 7-B, Grass Inc.) using the international 10-20 system of electrode placement (EEG: C3-A2 and C4-A1; EOG: F7-A2 and F8-A2). Airflow () was measured by a pneumotachometer (model 3700A, Hans Rudolph Inc., Kansas City, MO, USA) attached to a nasal mask. Tidal volume (V_T) was obtained from the integrated airflow signal. Airway pressures were measured using a pressure-tipped catheter (model TC-500XG, Millar Co., Houston, TX, USA), which was threaded though the mask (for positioning see Protocols). Ribcage and abdominal movements were monitored using DC-coupled respiratory inductance plethysmography (Respitrace, Ambulatory Monitoring, Ardsley, NY, USA). Arterial oxygen saturation (S_a,O2) was obtained using a pulse oximeter with an ear probe (Biox 3700, Ohmeda, Louisville, CO, USA).

The pharyngeal lumen was visualized using a paediatric fibre-optic bronchoscope (FB10X, Pentax Precision Instrument Co., Orangeburg, NY, USA). Topical anaesthesia was applied as follows: first, 2 % lidocaine was atomized into the pharynx through the mouth; second, 10 % lidocaine spray was used to anaesthetize both nostrils; finally, a 2 % lidocaine jelly was used to provide both lubrication and anaesthesia to the nostril through which the scope was passed. The position of the scope was standardized across subjects by advancing the tip to touch the end of the soft palate and then withdrawing it 2-3 cm. Slight variation of the orientation of the scope among subjects ensured clear visualization of the retropalatal lumen. Once the fibre-optic scope was positioned, it was secured using soft putty around the hole of the nasal mask through which it was passed. A continuous image of the retropalatal lumen was obtained from a closed-circuit video camera (Endovision 3000, Pentax Precision Instrument Co.) connected to the scope. The video image and the respiratory signals were digitized at 5 frames s^-1 and 25 Hz, respectively, using specially developed software. The images were also recorded onto videotape, along with the airflow signal which was modulated (FM-1 mod/demod, Wolfe Industries, San Marino, CA, USA) and recorded onto an audio track of the videotape.

Protocols

Night 1: airway visualization protocol. The subjects were sleep restricted as described above and were asked to report to the laboratory at 22.00 h. All subjects were instructed to use a decongestant, oxymetazoline hydrochloride, 0·05 % (Goldline Laboratories, Fort Lauderdale, FL, USA) 12 h before the study start time. An additional dose was given prior to the start of the study if the subject had subjective nasal stuffiness. At approximately 05.00 h, sleep staging electrodes and respiratory bands were attached and the subjects then laid supine in the bed. Local anaesthesia was given and the pressure catheter passed through one nostril. The fibre-optic scope was then passed through the opposite nostril and positioned as described above. Using the fibre-optic scope, the pressure catheter tip was positioned at the level of the retropalatal rim to measure pharyngeal pressure (P_PH). The nasal mask was then carefully lowered onto the face and secured. At this point, the exact position of the fibre-optic scope was adjusted and the scope plus the attached video camera were placed in a clamp suspended above the subject's head. The mask was carefully sealed, including the hole through which the scope was inserted. A check for air leakage around the mask was made by occluding the airflow during an attempted inspiration and expiration. The remaining transducers were then attached and further fine adjustments to the orientation of the scope were made. Following a period of wakefulness during which 3-5 min of data were collected for analysis (see below) the subject was allowed to go to sleep. During the sleep period the subject's head position was fixed with the use of sand-filled bolsters.

All variables were continuously monitored throughout the study. The fibre-optic image and the respiratory signals were acquired to the computer on-line during wakefulness, stage 2 sleep, slow wave sleep and REM sleep. Data were acquired only during periods in which the retropalatal lumen was clearly visible (i.e. no secretions obscuring the image). The study was terminated after a period of stable REM sleep was achieved during which there was sufficient time to collect both clear fibre-optic images and respiratory signals.

Night 2: pressure-flow relationship protocol. This was a split night study: during the first half, data were obtained to create upper airway pressure-flow loops; the second half was a standard polysomnogram. For this study, the subjects were requested to restrict their sleep to 6 h the night prior to the study and the subject was studied beginning at 23.00 h after approximately 18 h of sleep deprivation. Sleep staging electrodes and respiratory bands were attached. Supraglottic pressure (PSG) was measured by positioning the pressure catheter tip in the hypopharynx by observing the tip of the catheter disappear behind the tongue. The subject was laid supine on the bed and the nasal mask was placed on the face and secured. A check for air leakage around the mask was made. The remaining transducers were then attached. Following a period of wakefulness during which 3-5 min of data were collected for analysis, the subject was allowed to fall asleep. During this study, the subject's head position was not fixed. All variables were monitored continuously throughout the study and respiratory signals were acquired onto the computer on-line during wakefulness, stage 2 sleep, slow wave sleep, and REM sleep.

After data had been collected from all stages of sleep, the subject was awakened and the mask, pneumotachometer and pressure catheter removed. A nasal-oral thermistor was put in place and the subject was allowed to fall back asleep. The subject was allowed to sleep in any position comfortable for him or her. All variables were monitored and recorded on paper for another 3-4 h.

Data analysis

Night 1. Wakefulness/sleep stage was scored according to standardized criteria (Rechtstaffen & Kales, 1968). Inspired tidal volume (V_T), inspiration time (T_I), expiration time (T_E), breathing frequency (f) and minute ventilation (V_E) were calculated breath by breath during a period of wakefulness, stage 2 sleep, slow wave sleep, and REM sleep. Each breath during REM sleep was scored as either tonic REM (TREM) or phasic REM (PREM). PREM breaths were scored if a rapid EOG deflection > 50 µV in amplitude occurred within the 1 s preceding inspiration or any time during the inspiratory cycle. TREM breaths were scored during periods of ocular quiescence between the phasic eye bursts. For each of the five stages, approximately six to twelve breaths were analysed. Breaths for analysis were selected during a period of time in which there was no arousal from sleep or any increase in EEG frequency and during which the retropalatal lumen was clearly visible. For the wakefulness and NREM stages, the breaths were obtained during a period where the breathing was regular and stable and the breaths were consecutive. For the REM stages, breaths were selected in relation to eye bursts being present (PREM) or absent (TREM). In six of eight subjects, the wakefulness period analysed was from the beginning of the study; in the remaining two subjects, a post-sleep period was analysed as the subject could not stay awake at the beginning of the study long enough to collect adequate data.

The retropalatal cross-sectional area (CSA) was obtained for each digitized frame (5 frames per second) by manually outlining the retropalatal lumen using computer software (SigmaScan, Jandel). During this process, the investigator was blinded to the phase of respiration. The reproducibility of this technique has previously been validated by our laboratory (Taha, Werkheiser, Morrel & Badr, 1997). For each image, the scanning software provided an area in pixels. We converted these relative areas to absolute areas by using the dimensions of the pressure catheter as a reference (Launois, Remsburg, Yang & Weiss, 1996).

The cross-sectional areas obtained using this method were further analysed as follows. First, the CSA was measured at fixed points within inspiration and expiration (0, 25, 50, 75 and 100 % of inspiratory and expiratory duration); 0 % of inspiration was defined as the first frame after which flow crossed from negative to positive, while 0 % of expiration was defined as the first frame after which flow crossed from positive to negative. Second, we plotted the CSA of the digitized frames for each breath (both inspiration and expiration) against the P_PH that corresponded to each image of that breath. We defined the compliance of the upper airway (C_UA) as the slope of the regression line, CSA versus P_PH. To illustrate better these curves for the group, we created idealized pressure-CSA graphs for wakefulness and each stage of sleep using the CSA at the beginning of inspiration (CSA_I) as the intercept and the C_UA as the slope, plotting the ideal line across the range of P_PH most commonly encountered during the study (-1 to +1 cmH₂O).

Night 2. The wakefulness/sleep stage was scored according to standardized criteria (Rechtstaffen & Kales, 1968). Inspired tidal volume (V_T), inspiration time (T_I), total breath duration (T_TOT), and minute ventilation (V_E) were calculated for ten to fifteen breaths during a period of wakefulness, stage 2 sleep, slow wave sleep, PREM and TREM sleep as described for night 1. A pressure-flow loop was plotted for each breath. All trials were averaged and a composite pressure-flow loop was plotted for each subject. To generate a composite pressure-flow plot of breaths of different duration, pressure and flow were sampled at equally distributed points in both inspiration and expiration. Inspiratory resistance (R_UA) at a fixed flow of 0·1 l s^-1 was computed from each loop as a numeric representation of the linear part of the pressure-flow loop.

The apnoea-hypopnoea index (AHI) was determined for each subject by analysing the standard polysomnography portion of the night. An apnoea was defined as an absence of airflow for at least 10 s. A hypopnoea was defined as a 50 % decrease in airflow for at least 10 s in association with an arousal or a 3 % decrease in oxygen saturation. The AHI was defined as the number of apnoeas and hypopnoeas divided by the total sleep time.

Statistical analysis

Night 1. All statistical analyses were performed on the complete data set of eight subjects. Comparisons of the mean values T_I, T_E, V_T, V_E, f, C_UA and CSA_I were carried out using a one-way analysis of variance (ANOVA) with repeated measures, with sleep stage as the factor. To determine if there were within-breath changes in CSA for a given stage, a one-way ANOVA with repeated measures was performed, with the percentage of breath duration as the factor. For comparisons that reached significance (P < 0·05), post hoc analysis was performed using the Student-Newman-Keuls method.

Night 2. All statistical analyses were performed on the complete data set of five subjects. Comparison of the mean values T_I, T_E, V_T, V_E and R_UA were carried out using a one-way ANOVA with repeated measures, with sleep stage as the factor. For comparisons that reached significance (P < 0·05), post hoc analysis was performed using the Student-Newman-Keuls method.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

All eight subjects underwent the airway visualization night. Only five of eight subjects completed the second night; subject MV moved before he could be restudied and subjects JS and FM refused further study. None of the subjects had symptoms of a sleep disorder and, as can be seen in Table 1, all had an AHI < 5 h^-1. We do not have AHI data for MV, JS or FM because they did not undergo the second night's study; however, we do not believe that they had significant sleep-disordered breathing as none complained of any sleep disorder or snoring and none showed evidence of sleep-disordered breathing during the initial study night.

In six of the eight subjects, we were able to obtain ventilatory data and clear images in all stages of sleep, including REM. Table 1 presents the REM latency and number of minutes of REM sleep for each subject. Two subjects (AC and AB) had slow wave sleep during the course of their study but did not have clear images during this stage. Because of the similarities seen between stage 2 and slow wave sleep on preliminary analysis and because we were most interested in comparing REM sleep with NREM sleep, we combined the data for stage 2 and slow wave sleep into one category, NREM sleep, before the final statistical analyses were performed.

Effect of sleep stage on ventilatory data

The group mean levels for each respiratory variable during wakefulness and sleep stages for each of the two nights are shown in Table 2. For night 1, wakefulness was associated with a higher V_E (6·3 ± 1·2 l min^-1) compared with each of the three sleep stages, NREM (5·2 ± 1·0 l min^-1), TREM (5·3 ± 1·1 l min^-1) and PREM (5·3 ± 0·8 l min^-1). There was no difference in any of the other respiratory variables between the different stages of sleep. For night 2, the only significant difference between respiratory variables was in the T_I, which was smaller in wakefulness (1·3 ± 0·1 s) than in the three sleep stages (NREM, 1·6 ± 0·2 s; TREM, 1·7 ± 0·3 s; PREM, 1·6 ± 0·2 s).

Table 2. Respiratory variables during wakefulness and sleep

Effect of REM sleep on CSA

The effect of sleep stage on the CSA of the airway lumen at the beginning of inspiration (CSA_I) is shown in Fig. 1. A one-way ANOVA with repeated measures indicated that sleep stage had a significant effect on CSA_I (P < 0·001). Post hoc analysis showed that the mean CSA_I during wakefulness (117·4 ± 30·5 mm², P < 0·05) was significantly larger than the CSA_I in NREM (84·0 ± 31·5 mm²), TREM (79·7 ± 29·5 mm²) and PREM sleep (78·1 ± 27·8 mm²). There was no significant difference in the CSA_I between the three different sleep stages.

		View larger version [in this window] [in a new window]
Figure 1. The effect of sleep stage on baseline cross-sectional area The baseline pharyngeal cross-sectional area (CSAI) for each subject is plotted for wakefulness (Wake), NREM, tonic REM (TREM) and phasic REM (PREM) sleep. The mean ± S.D. CSAI is also shown for each stage (). The CSAI during wakefulness is significantly larger than the CSAI in NREM or tonic or phasic REM sleep (* P < 0·05).

Effect of REM sleep on within-breath changes in CSA

Examples of the changes in retropalatal CSA during wakefulness and sleep in a representative subject are shown in Figs 2 and 3. Figure 2 illustrates single breaths from wakefulness and each of the sleep stages from one subject. In this subject, there is significant narrowing of the airway at baseline with sleep onset. There was a significant within-breath decrease in CSA during NREM sleep but not during either tonic or phasic REM sleep. The mean CSA, P_PH and at defined points during inspiration and expiration for each stage for this subject are shown in Fig. 3A. The figure again shows that there was significant narrowing of the airway during sleep and further narrowing during the respiratory cycle during NREM sleep. However, there are no within-breath changes during tonic or phasic REM sleep despite similar changes in P_PH and .

		View larger version [in this window] [in a new window]
Figure 2. Examples of CSA, PPH and V in one subject Examples of the retropalatal cross-sectional area (CSA; ), pharyngeal pressure (PPH; ) and airflow (V; ) for one breath in each stage of wakefulness (Wake) and sleep for one subject. Fibre-optic images illustrate the retropalatal CSA at different phases of the respiratory cycle.

		View larger version [in this window] [in a new window]
Figure 3. The effect of sleep stage on CSA, PPH and V for one subject A, mean pharyngeal CSA (CSA), pharyngeal pressure (PPH) and airflow (V) measured at fixed percentages within inspiration and expiration (percentage breath duration) for all breaths analysed is shown for wakefulness (), NREM (), tonic REM () and phasic REM sleep () for one representative subject. In this subject, CSAI (the CSA at 0 % inspiration) decreases during NREM and REM sleep. There are no within-breath changes in CSA during wakefulness, and tonic and phasic REM sleep while there is significant inspiratory narrowing observed during NREM sleep. B, pharyngeal CSA is plotted against pharyngeal pressure (PPH) for wakefulness (--), NREM (- -), tonic REM (- · -) and phasic REM sleep (······). The pharyngeal compliance is defined as the slope of the relationship CSA versus PPH. The graph illustrates that the pharyngeal compliance is significantly increased during NREM sleep compared with wakefulness and tonic and phasic REM sleep.

Figure 4 illustrates the within-breath changes in retropalatal CSA for the group as a whole. For all subjects, there were no significant within-breath changes in CSA during wakefulness. In contrast, during NREM sleep the retropalatal CSA narrowed significantly during late inspiration (area at 0 % inspiration, 84·0 ± 31·5 mm²; area at 75 % inspiration, 63·5 ± 35·4 mm²; P < 0·05 compared with 0 % inspiration) and had returned to its initial size by the beginning of expiration (area at 0 % expiration, 80·2 ± 36·4 mm²; P < 0·05 compared with 75 % of inspiration, not significant compared with 0 % inspiration). In tonic REM sleep, the retropalatal CSA narrowed significantly in late inspiration (area at 0 % inspiration, 79·6 ± 29·5 mm²; area at 75 % inspiration, 70·3 ± 26·1 mm²; area at 100 % inspiration, 70·5 ± 26·1 mm²; both P < 0·05 compared with 0 % inspiration), returning to initial size by early expiration (area at 25 % expiration, 81·2 ± 29·3 mm²; P < 0·05 compared with 75 % and 100 % inspiration and not significant compared with 0 % inspiration). It should be noted, however, that the changes during tonic REM sleep may not be clinically significant as changes in CSA of less than 20 % must be interpreted with caution (see Discussion). There were no significant within-breath changes in retropalatal CSA during phasic REM sleep.

		View larger version [in this window] [in a new window]
Figure 4 The effect of sleep stage on within-breath changes in CSA, PPH and V Group mean pharyngeal CSA (CSA), pharyngeal pressure (PPH) and airflow (V) measured at fixed percentages within inspiration and expiration (percentage breath duration) for all breaths analysed is shown for wakefulness (), NREM (), tonic REM () and phasic REM sleep (). There were no significant within-breath changes in CSA during wakefulness or phasic REM sleep. There were significant mid-inspiratory decreases in CSA during NREM and tonic REM sleep with the airway returning to baseline CSA by early expiration.

Effect of REM sleep on upper airway compliance

Examples of the CSA-P_PH curves for one subject are illustrated in Fig. 3B. The upper airway compliance was defined as the slope of the regression CSA versus P_PH. Therefore, the figure illustrates that for this subject the airway compliance increased during NREM sleep compared with wakefulness. However, the compliance decreased during tonic and phasic REM sleep and was similar to the compliance measured during wakefulness. The group mean CSA-P_PH curves are illustrated in Fig. 5A and the idealized CSA-P_PH curves are shown in Fig. 5B. For the group, upper airway compliance increased during NREM sleep compared with wakefulness. Upper airway compliance then decreased during tonic and phasic REM sleep to a level similar to that measured during wakefulness.

		View larger version [in this window] [in a new window]
Figure 5. The effect of sleep stage on upper airway compliance A, group mean pharyngeal cross-sectional area (CSA) is plotted against pharyngeal pressure (PPH) for wakefulness (--), NREM (- -), tonic REM (- · -) and phasic REM sleep (······). For the group as a whole, pharyngeal compliance (defined as the slope of the relationship CSA versus PPH) is significantly increased during NREM sleep compared with wakefulness, tonic and phasic REM sleep (P < 0·05). B, idealized cross-sectional area (CSA) versus pharyngeal pressure (PPH) curves for wakefulness (--), NREM (- -), tonic REM (- · -) and phasic REM sleep (······). These idealized curves illustrate that CSA decreases with sleep and that the pharyngeal compliance is greatest during NREM sleep (P < 0·05).

Individual and group mean upper airway compliances (C_UA) for wakefulness and each stage of sleep are shown in Fig. 6. For the group as a whole there was a significant effect of sleep stage on C_UA (P = 0·032). Post hoc analysis showed that the C_UA during NREM sleep (13·2 ± 14·4 mm² cmH₂O^-1; P <0·05) was significantly larger than the C_UA during wakefulness (5·8 ± 8·0 mm² cmH₂O^-1), TREM (4·0 ± 6·0 mm² cmH₂O^-1) and PREM sleep (2·5 ± 4·6 mm² cmH₂O^-1). There was no significant difference in the C_UA during wakefulness compared with TREM or PREM sleep or the C_UA during TREM compared with PREM sleep.

		View larger version [in this window] [in a new window]
Figure 6. The effect of sleep stage on upper airway compliance The upper airway compliance (CUA) for each subject is plotted for wakefulness (Wake), NREM, tonic REM (TREM) and phasic REM (PREM) sleep. The mean ± S.D. CUA is also shown for each stage (). The CUA during NREM is significantly larger than the CUA during wakefulness or tonic or phasic REM sleep (* P < 0·05).

Effect of REM sleep on pressure-flow relationships

The pressure-flow relationships of the five subjects who were studied on the second night are presented in Fig. 7. The pressure-flow relationships for subject AB showed no differences between wakefulness and any sleep stage. In two subjects (AC and HB), the maximal flow and the slope were clearly decreased in NREM sleep compared with REM sleep, which in turn was similar to wakefulness. In the two remaining subjects (RE and RS), there was decreased maximal flow and a decreased slope in both NREM and REM sleep, but there was significant hysteresis in NREM, indicating that there was a difference in the behaviour of the upper airway during NREM sleep. Inspiratory R_UA (at a fixed flow of 0·1 l s^-1) was increased during NREM sleep compared with either tonic or phasic REM sleep in four of five subjects. For the group, there was a trend for increased R_UA during NREM sleep (3·5 ± 4·4 cmH₂O l^-1 s^-1) compared with wakefulness (0·7 ± 0·8 cmH₂O l^-1 s^-1), tonic REM (0·8 ± 0·4 cmH₂O l^-1 s^-1), or phasic REM sleep (0·8 ± 0·5 cmH₂O l^-1 s^-1; P = 0·17).

		View larger version [in this window] [in a new window]
Figure 7. The effect of sleep stage on upper airway pressure-flow relationships Pressure (PSG)-flow (V) relationships for wakefulness (--), NREM (- -), tonic REM (- · -) and phasic REM sleep (·······) are illustrated for each of the 5 subjects in whom this measurement was made.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The aim of the present study was to investigate the effect of REM sleep on upper airway cross-sectional area (CSA) and compliance in a group of subjects without significant sleep-disordered breathing. There were four main findings from our study. First, upper airway CSA was significantly smaller during NREM sleep compared with wakefulness but did not decrease further in either tonic or phasic REM sleep. Second, there were significant within-breath changes in CSA during NREM sleep and tonic REM sleep such that the airway narrowed during inspiration and returned to baseline calibre during expiration. These changes were not seen during wakefulness or phasic REM sleep. Third, upper airway compliance, defined as the slope of the regression line for CSA plotted against pharyngeal pressure, was unchanged during either tonic or phasic REM sleep as compared with wakefulness. This was in contrast to the large increase in compliance seen during NREM sleep. Fourth, there was a trend towards a higher upper airway resistance during NREM sleep compared with REM sleep. These findings indicate that the upper airway was less compliant during REM sleep than during NREM sleep.

Validity of techniques and methodology

Fibre-optic endoscopy has several limitations which need to be considered when interpreting the findings. First, movement of the fibre-optic scope could have influenced the measurements of CSA. To prevent this, the orientation of the scope was fixed by anchoring it to the mask, and the position of the subject's head was fixed by the sand pillow in which the head was positioned prior to sleep. In addition, all scope adjustments made during a study (usually to help clear the airway of secretions) were performed by one investigator (J. A. R.). Finally, images were analysed only if there was judged to be no change in the relationship of the scope to anatomical landmarks in different planes (e.g. soft palate, epiglottis).

Second, an important consideration is the ability to detect accurately and reproducibly the edge of the airway lumen. While this process is somewhat subjective, we believe that the edge can be visualized with reasonable precision. Breaths in which the airway lumen was not clearly visible were rejected for analysis. Previous work from our laboratory has shown that the coefficient of variation is 10 % when outlining an image of known diameter, suggesting that changes of less than 20 % should be interpreted with caution (Taha et al. 1997). However, the magnitude of CSA changes seen during this study was greater than this in most cases. The position of the pharyngeal catheter is important if the diameter of this catheter is to be used to provide a factor to convert the areas from the number of pixels to an absolute area. At all times the position of the catheter remained vertical to the pharyngeal lumen, allowing measurement of the diameter at the same plane as the CSA measurement. The diameter measurement was made by one investigator in all subjects (J. A. R.).

Third, only six to twelve images were analysed per stage of sleep. Reasons for the limited number of breaths included the need to outline manually the pharyngeal lumen of each image; poor visualization of the pharyngeal lumen; and for REM sleep, a limited number of breaths available to analyse because of the criteria used to choose phasic and tonic breaths and because we had only one REM sleep period in the majority of subjects. Despite the small numbers, we believe that the images analysed are representative. During NREM sleep, we analysed consecutive breaths when the breathing pattern was regular and stable. REM breaths were chosen from periods which met the accepted polysomnographic definition of REM sleep in all subjects. In addition, we believe that the breaths chosen are representative because of the similarity of findings between the eight subjects.

Finally, fibre-optic endoscopy does not allow simultaneous visualization of multiple anatomical levels. Previous studies have indicated that the retropalatal airway is the site of maximum narrowing in patients with sleep apnoea and in normal subjects (Horner, Shea, McIvor & Guz, 1989; Morrison, Launois, Isono, Feroah, Whitelaw & Remmers, 1993; Isono et al. 1997). All our subjects demonstrated narrowing at this level with sleep onset and thus, we believe that the study of upper patency and compliance is appropriately made by analysing images at this level.

The measurement of pharyngeal compliance requires several assumptions. First, we are not measuring true pharyngeal airway compliance, defined as changes in pharyngeal volume per change in pharyngeal pressure. While pharyngeal volume has been measured in the human (Rolfe, Olson & Saunders, 1991) and could be estimated by measuring the pharyngeal CSA at multiple anatomical levels, it would not be practical to do so on sleeping, spontaneously breathing subjects. We believe that pharyngeal CSA is a reasonable substitute, as have others who have measured upper airway compliance (Kuna et al. 1988; Isono et al. 1993). Second, an accurate compliance measurement using area instead of volume requires that the pressure measured be at the same level as the changes in area, as in this study. Third, the measurement of compliance assumes that the pressure being measured is a transmural pressure as the extraluminal pressure is assumed to be constant (Kuna et al. 1988; Olson, Fouke, Hoekje & Strohl, 1988). Finally, the relative contributions to the transmural pressure (Rowley, Permutt, Willey, Smith & Schwartz, 1996), such as the pressure intrinsic to the airway wall (attributed mostly to the upper airway muscles) and the pressure surrounding the airway (attributed to structures such as the tongue, tonsils and pharyngeal fat pads) cannot be ascertained with this method, primarily because it is difficult to measure the pressure surrounding the airway in a human.

Finally, it must be noted that we made pharyngeal compliance measurements over a small range of pharyngeal pressures (2 cmH₂O) because we were specifically interested in studying the effect of REM sleep on compliance during eupnoeic breathing. Thus, the compliance measurements may be different from those made by experimentally manipulating the pharyngeal pressure generally with externally applied pressure over a larger range of pressures (10-20 cmH₂O) (Kuna et al. 1988; Isono et al. 1997). By measuring compliance in this fashion in the static upper airway, it has been observed that the pharyngeal compliance changes with the pharyngeal CSA (Isono et al. 1993, 1997). However, we cannot make direct comparison of our work with this previous work since we measured compliance in a dynamic airway. We are not aware of previous work that has investigated upper airway compliance in a dynamic airway or on a breath-by-breath basis. Since the upper airway is a dynamic structure, we believe that the measurement of upper airway compliance is best made during eupnoeic breathing and that this approach allows us to make unique observations on the effect of sleep stage on the upper airway.

The effect of REM sleep on the upper airway

In this study, we observed pharyngeal narrowing at the level of the soft palate in subjects without sleep-disordered breathing and we observed that baseline retropalatal CSA decreases by approximately 25 % during NREM sleep and does not decrease further during either tonic or phasic REM sleep. Thus, there was no additional effect of REM sleep on baseline retropalatal dimensions. This finding is corroborated by previous studies which have found that upper airway resistance during REM sleep was the same (Hudgel et al. 1984; Wiegand et al. 1989) or less than (Wiegand et al. 1990a) that seen during NREM sleep. Our study also showed a similar finding, with a trend for the R_UA at a fixed flow of 0·1 l s^-1 to be less during REM sleep than during NREM sleep.

We also found that upper airway compliance during tonic and phasic REM sleep was similar to the compliance measured during wakefulness. This was in marked contrast to the significantly increased compliance observed during NREM sleep. Thus, the airway is less compliant (more stable) during REM sleep than during NREM sleep. There are several possible explanations for the finding that upper airway compliance is decreased during REM sleep compared with NREM sleep. First, this could be an artifact related to our techniques of measuring CSA and pharyngeal pressure. This is unlikely since the fibre-optic scope placement and orientation were tightly controlled during the study and only those images that clearly showed the retropalatal lumen were analysed. The pharyngeal catheter was also carefully placed to measure the intraluminal pressure at the site of narrowing. In addition, the pharyngeal pressure- flow loops indicate a trend towards decreased upper airway resistance during REM sleep compared with NREM sleep, corroborating a difference in upper airway physiology between REM and NREM sleep. Finally, the observed changes in baseline measurements of CSA during NREM sleep were in agreement with those of previous investigators (Horner et al. 1989; Morrell & Badr, 1998). Thus, we do not believe that our findings represent an artifact related to technique or methodology.

Second, it is possible that our assumption that intraluminal pressure reflects transmural pressure is incorrect. This assumption is correct only if the relevant extraluminal pressure is atmospheric pressure (Olson et al. 1988). However, there is evidence that the extraluminal pressure may be above atmospheric pressure in certain circumstances, as evidenced by the occurrence of upper airway obstruction during central apnoea (Badr et al. 1995). Nevertheless, there is no evidence that the surrounding pressure would be influenced by the stage of sleep. In addition, it is important to again note that we were measuring the pharyngeal pressure at the site of narrowing not a downstream or collapsing pressure. Thus, any differences in the pharyngeal pressure between sleep stages cannot explain the difference in compliance.

Finally, the change in compliance could be secondary to a change in the intrinsic properties of the airway wall. Previous work has shown that airway wall stiffness is related to both the upper airway neuromuscular activity and the non-neuromuscular properties of the airway wall (Isono & Remmers, 1994; Rowley et al. 1996; Rowley, Williams, Smith & Schwartz, 1997). Previous work on the effect of REM sleep on upper airway neuromuscular activity has clearly shown that the neuromuscular activity to upper airway muscles was either decreased (Sauerland & Harper, 1976; Sauerland, Orr & Hairston, 1981; Wiegand, Zwillich, Wiegand & White, 1991) or unchanged (Wiegand, Latz, Zwillich & Wiegand, 1990b) during REM sleep. Thus, REM sleep should be associated with either increased or similar airway compliance compared with NREM sleep. Since we found a decreased compliance during REM sleep, the change in the compliance cannot be secondary to changes in the phasic activation of upper airway muscles.

Non-neuromuscular determinants of airway wall stiffness include the intrinsic stiffness of the airway wall muscles and connective tissue (Olson et al. 1988; Rowley et al. 1996; Isono et al. 1997). Caudal tracheal traction (van de Graaff, 1988; Rowley et al. 1996) has been shown to increase airway wall stiffness and thus may alter airway wall properties on a dynamic within-breath basis. Caudal tracheal traction is increased with increases in inspiratory activity and lung volumes (van de Graaff, 1988). In our study, REM sleep is not associated with increased lung volumes, as evidenced by similar tidal volumes between the two stages. Thus, we do not believe that differences in tracheal traction can explain the decreased compliance in REM sleep.

It has been proposed that the vascular perfusion of the upper airway is a determinant of wall stiffness (Olson et al. 1988). Direct evidence for this relationship has been demonstrated in cardiac papillary muscle, which becomes stiffer with increased perfusion (Kitabatake & Suga, 1978). Indirect evidence for an effect of vascular perfusion on upper airway wall stiffness comes from both animal and human experiments. In the cat, pharmacological vasoconstriction has been shown to increase airway stiffness (Mayor et al. 1996). In the human, an increased central venous pressure attenuates the changes in pharyngeal CSA between functional residual capacity and end-inspiratory volume indicative of a decreased compliance of the pharyngeal wall (Shepard, Pevernagie, Stanson, Daniels & Sheedy, 1996). While a difference in upper airway vascular perfusion has not been demonstrated for REM and NREM sleep, it has been shown that coronary and cerebral perfusion is higher during REM sleep than in NREM sleep (Kirby & Verrier, 1989; Madsen et al. 1991). Thus, it is plausible that vasodilatation and increased vascular perfusion also occur in vessels supplying the pharyngeal airway during REM sleep leading to a decreased airway compliance. It should be noted, however, that in other studies the changes in vascular tone were associated with changes in baseline pharyngeal CSA (Wasicko, Hutt, Parisi, Neubauer, Mezrich & Edelman, 1990; Shepard et al. 1996), which we did not observe. Thus, we propose that the differences in compliance between REM and NREM sleep were secondary to an increased perfusion of the upper airway during REM sleep.

Tonic versus phasic REM sleep

In contrast to previous investigations (Millman et al. 1988; Wiegand et al. 1991), we noted that pharyngeal calibre and compliance were not different between tonic and phasic REM sleep. We considered several possible explanations for the difference between our findings and others. First, there may not be a differential response of the upper airway to phasic REM sleep as there is for the respiratory system. This would be unusual as the upper and lower airways are synchronized in normal, spontaneously breathing subjects. Second, changes in upper airway physiology may have been found if we had investigated the oropharynx, where the genioglossus has a larger influence on airway patency and which has been shown to have decreased EMG activity during phasic REM sleep (Sauerland & Harper, 1976; Wiegand et al. 1991). Third, the definition we used to separate phasic from tonic breaths may not be sufficiently discriminating. For instance, it has been shown that ventilatory events are best associated with characteristic pontine-geniculate-occipital spikes, which are not detected with standard EEG techniques (Orem, 1980). However, the definition we used has been used by previous investigators who did find differences between phasic and tonic REM sleep (Wiegand et al. 1991).

Implications

Since we did not study subjects with known obstructive sleep apnoea (OSA), our findings do not directly elucidate the mechanisms of airflow obstruction. Nonetheless, the major finding of this study, that REM sleep was associated with a lower pharyngeal compliance that NREM sleep despite a similar CSA, has several important implications regarding the determinants of upper airway compliance in the sleeping human. First, sleep stage is an important determinant of upper airway compliance. Second, as discussed above, if the assumption is that the intraluminal pressure reflects the transmural pressure, a decreased compliance during REM sleep would have to be associated with increased neuromuscular activity or increased intrinsic stiffness of the airway wall. Since increased neuromuscular activity has not been observed (Sauerland & Harper, 1976; Sauerland et al. 1981; Wiegand et al. 1991), non-neuromuscular properties of the upper airway, not neuromuscular activity, may be the critical determinants of compliance during sleep. In other words, in the absence of significant neuromuscular activity, as is seen in REM sleep, changes in compliance are secondary to changes in the non-neuromuscular properties of the airway wall. Since REM sleep was associated with a decreased compliance compared with NREM sleep, a distinguishing feature of REM sleep may be an increase in the stiffness of the airway wall independent of neuromuscular activity and possibly related to changes in the vascular perfusion of the airway wall. Since neuromuscular activity is also decreased during NREM sleep compared with wakefulness, the non-neuromuscular properties of the upper airway are probably an important determinant of upper airway compliance during NREM sleep as well, albeit not to the same degree as in REM sleep.

Third, it has been observed that in patients with OSA, obstructive apnoeas appear to be more common during REM sleep than NREM sleep (Findley et al. 1985; Series et al. 1990). However, our data in normal subjects indicate that the upper airway is less compliant during REM sleep. Thus, if the clinical observation of increased apnoeas in REM sleep is correct, REM sleep in apnoeics would be associated with an increased compliance. This would indicate that the non-neuromuscular properties of the upper airway in apnoeics favour a collapsing airway while in normal subjects they favour a stable airway. Alternatively, since we studied only the nasopharynx, we could have missed a segment of even greater narrowing in the oropharynx, where the genioglossus is the primary dilator muscle. Genioglossus activity has clearly been shown to be decreased during REM sleep (Sauerland & Harper, 1976; Sauerland et al. 1981; Wiegand et al. 1991) and it is conceivable that REM sleep is characterized by greater collapse in the oropharynx than the nasopharynx. However, this is speculative as all the previous studies on the site of upper airway narrowing in normal subjects and apnoeics have been performed during either NREM sleep or general anaesthesia (Isono et al. 1993, 1997; Morrison et al. 1993).

In conclusion, our study has shown that baseline pharyngeal CSA is unchanged between NREM and REM sleep and that eupnoeic breathing during REM sleep is associated with a decreased compliance of the upper airway compared with NREM sleep. Our data suggest that differences in compliance are secondary to differences in the intrinsic properties of the airway wall independent of changes in the neuromuscular activity. We put forward the idea that the non-neuromuscular properties of the airway wall are a more important determinant of upper airway stiffness during sleep than is upper airway neuromuscular activity. Future research should be directed at elucidating which mechanical properties are critical to determining upper airway compliance.

	REFERENCES

Top Abstract Introduction Methods Results Discussion References

Badr, M. S., Toiber, F., Skatrud, J. B. & Dempsey, J. (1995). Pharyngeal narrowing/occlusion during central sleep apnea. Journal of Applied Physiology 78, 1806-1815	[Medline]
Carskadon, M. A. & Dement, W. C. (1994). Normal human sleep: An overview. In The Principles and Practice of Sleep Medicine, ed. Dement, W. C., Kryger, M. H. & Roth, T., pp. 16-25. W. B. Saunders Company, Philadelphia.
Douglas, N. J. (1994). Control of ventilation during sleep. In The Principles and Practice of Sleep Medicine, ed. Dement, W. C., Kryger, M. H. & Roth, T., pp. 204-211. W. B. Saunders Company, Philadelphia.
Findley, L. J., Wilhoit, S. C. & Suratt, P. M. (1985). Apnea duration and hypoxemia during REM sleep in patients with obstructive sleep apnea. Chest 87, 432-436	[Abstract]
Horner, R. L., Shea, S. A., McIvor, J. & Guz, A. (1989). Pharyngeal size and shape during wakefulness and sleep in patients with obstructive sleep apnea. Quarterly Journal of Medicine 268, 719-735.
Hudgel, D. W., Martin, R. J., Johnson, B. & Hill, P. (1984). Mechanics of the respiratory system and breathing pattern during sleep in normal humans. Journal Applied of Physiology 56, 133-137.
Isono, S., Morrison, D. L., Launois, S. H., Feroah, T. R., Whitelaw, W. A. & Remmers, J. E. (1993). Static mechanics of the velopharynx of patients with obstructive sleep apnea. Journal of Applied Physiology 75, 148-154	[Medline]
Isono, S. & Remmers, J. E. (1994). Anatomy and physiology of upper airway obstruction. In The Principles and Practice of Sleep Medicine, ed. Dement, W. C., Kryger, M. H. & Roth, T., pp. 642-656. W. B. Saunders Company, Philadelphia.
Isono, S., Remmers, J. E., Tanaka, A., Sho, Y., Sato, J. & Nishino, T. (1997). Anatomy of pharynx in patients with obstructive sleep apnea and in normal subjects. Journal of Applied Physiology 82, 1319-1326	[Abstract/Full Text]
Kirby, D. A. & Verrier, R. L. (1989). Differential effects of sleep stage on coronary hemodynamic function. American Journal of Physiology 256, H1378-1383	[Medline]
Kitabatake, A. & Suga, H. (1978). Diastolic stress-strain relation of nonexcised blood-perfused canine papillary muscle. American Journal of Physiology 234, H416-420	[Medline]
Kuna, S. T., Bedi, D. G. & Ryckman, C. (1988). Effect of nasal airway positive pressure on upper airway size and configuration. American Review of Respiratory Disease 138, 969-978	[Medline]
Launois, S. H., Remsburg, S., Yang, W. J. & Weiss, J. W. (1996). Relationship between velopharyngeal dimensions and palatal EMG during progressive hypercapnia. Journal of Applied Physiology 80, 478-485	[Medline]
Madsen, P. L., Schmidt, J. F., Wildschiodtz, G., Friberg, L., Holm, S., Vorstrup, S. & Lassen, N. A. (1991). Cerebral O2 metabolism and cerebral blood flow in humans during deep and rapid-eye-movement sleep. Journal of Applied Physiology 70, 2597-2601	[Medline]
Mayor, A., Schwartz, A. R., Rowley, J. A., Willey, S. J., Gillespie, M. B., Smith, P. L. & Robotham, J. (1996). Effect of blood pressure changes on air flow dynamics in the upper airway of the decerebrate cat. Anesthesiology 84, 128-134	[Medline]
Millman, R. P., Knight, H., Kline, L. R., Shore, E. T., Chung, D. C. & Pack, A. I. (1988). Changes in compartmental ventilation in association with eye movements during REM sleep. Journal of Applied Physiology 65, 1196-1202	[Medline]
Morrell, M. J. & Badr, M. S. (1998). Effects of NREM sleep on dynamic within-breath changes in upper airway patency in humans. Journal of Applied Physiology 84, 190-199	[Abstract/Full Text]
Morrison, D. L., Launois, S. H., Isono, S., Feroah, T. R., Whitelaw, W. A. & Remmers, J. E. (1993). Pharyngeal narrowing and closing pressures in patients with obstructive sleep apnea. American Review of Respiratory Disease 148, 606-611	[Medline]
Olson, L. G., Fouke, J. M., Hoekje, P. L. & Strohl, K. P. (1988). A biomechanical view of upper airway function. In Respiratory Function of the Upper Airway, ed. Mathew, O. P. & Sant'Ambrogio, G., pp. 359-389. Marcel Dekker, New York.
Orem, J. (1980). Medullary respiratory neuron activity: relationship to tonic and phasic REM sleep. Journal Applied of Physiology 48, 54-65.
Rechtstaffen, A. & Kales, A. (1968). A Manual of Standardized Terminology Techniques and Scoring System for Sleep Stages of Human Subjects. National Institutes of Health, Washington, DC.
Rolfe, I., Olson, L. G. & Saunders, N. A. (1991). Pressure-volume properties of the upper airway in man. Respiration Physiology 86, 15-23	[Medline]
Rowley, J. A., Permutt, S., Willey, S. J., Smith, P. L. & Schwartz, A. R. (1996). Effect of tracheal and tongue displacement on upper airway airflow dynamics. Journal of Applied Physiology 80, 2171-2178	[Medline]
Rowley, J. A., Williams, B. C., Smith, P. L. & Schwartz, A. R. (1997). Neuromuscular activity and upper airway collapsibility: mechanisms of action in the decerebrate cat. American Journal of Respiratory and Critical Care Medicine 156, 515-521	[Abstract/Full Text]
Sauerland, E. K. & Harper, R. M. (1976). The human tongue during sleep: electromyographic activity of the genioglossus muscle. Experimental Neurology 51, 160-170	[Medline]
Sauerland, E. K., Orr, W. C. & Hairston, L. E. (1981). EMG patterns of oropharyngeal muscles during respiration in wakefulness and sleep. Electromyography and Clinical Neurophysiology 21, 307-316.	[Medline]
Series, F., Cormier, Y. & LaForge, J. (1990). Influence of apnea type and sleep stage on nocturnal postapneic desaturation. American Review of Respiratory Disease 141, 1522-1526	[Medline]
Shepard, J. W., Pevernagie, D. A., Stanson, A. W., Daniels, B. K. & Sheedy, P. F. (1996). Effects of changes in central venous pressure on upper airway size in patients with obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine 153, 250-254	[Abstract]
Sullivan, C. E., Murphy, E., Kozar, L. F. & Phillipson, E. A. (1979). Ventilatory responses to CO2 and lung inflation in tonic versus phasic REM sleep. Journal of Applied Physiology 47, 1304-1310.
Taha, B. H., Werkheiser, M., Morrell, M. J. & Badr, M. S. (1997). Simultaneous digitization of upper airway fiber-optic images and respiratory signals. Sleep 10, 883-890.
van de Graaff, W. B. (1988). Thoracic influence on upper airway patency. Journal of Applied Physiology 65, 2124-2131	[Medline]
Wasicko, M. J., Hutt, D. A., Parisi, R. A., Neubauer, J. A., Mezrich, R. & Edelman, N. H. (1990). The role of vascular tone in the control of upper airway collapsibility. American Review of Respiratory Disease 14, 1569-1577.
Wiegand, D. A., Latz, B., Zwillich, C. W. & Wiegand, L. (1990a). Upper airway resistance and geniohyoid muscle activity in normal men during wakefulness and sleep. Journal of Applied Physiology 69, 1252-1261	[Medline]
Wiegand, D. A., Latz, B., Zwillich, C. W. & Wiegand, L. (1990b). Geniohyoid muscle activity in normal men during wakefulness and sleep. Journal of Applied Physiology 69, 1262-1269	[Medline]
Wiegand, L., Zwillich, C. W. & White, D. P. (1989). Collapsibility of the human upper airway during normal sleep. Journal of Applied Physiology 66, 1800-1808	[Medline]
Wiegand, L., Zwillich, C. W., Wiegand, D. & White, D. P. (1991). Changes in upper airway muscle activation and ventilation during phasic REM sleep in normal men. Journal of Applied Physiology 71, 488-497	[Medline]

Acknowledgements

This work was supported by the Department of Veterans Affairs and the National Heart, Lung, and Blood Institute. M. S. B. is a career investigator of the American Lung Association.

Corresponding author

J. A. Rowley: Wayne State University School of Medicine, Sleep Disorders Center, 4707 St Antoine, 1 Center, Detroit, MI 48201, USA.

Email: jrowley{at}oncgate.roc.wayne.edu

This article has been cited by other articles:

				J. A. Rowley, I. Deebajah, S. Parikh, A. Najar, R. Saha, and M. S. Badr The influence of episodic hypoxia on upper airway collapsibility in subjects with obstructive sleep apnea J Appl Physiol, September 1, 2007; 103(3): 911 - 916. [Abstract] [Full Text] [PDF]

				K. F. Mansour, J. A. Rowley, and M. S. Badr Measurement of pharyngeal cross-sectional area by finite element analysis J Appl Physiol, January 1, 2006; 100(1): 294 - 303. [Abstract] [Full Text] [PDF]

				J. A. Rowley, C. S. Sanders, B. R. Zahn, and M. S. Badr Gender differences in upper airway compliance during NREM sleep: role of neck circumference J Appl Physiol, June 1, 2002; 92(6): 2535 - 2541. [Abstract] [Full Text] [PDF]

				J. A. Rowley, X. Zhou, I. Vergine, M. A. Shkoukani, and M. S. Badr Influence of gender on upper airway mechanics: upper airway resistance and Pcrit J Appl Physiol, November 1, 2001; 91(5): 2248 - 2254. [Abstract] [Full Text] [PDF]

				J. A. Rowley, C. S. Sanders, B. R. Zahn, and M. S. Badr Effect of REM sleep on retroglossal cross-sectional area and compliance in normal subjects J Appl Physiol, July 1, 2001; 91(1): 239 - 248. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rowley, J. A. Articles by Badr, M. S. Search for Related Content PubMed PubMed Citation Articles by Rowley, J. A. Articles by Badr, M. S.