Skip to main content
NCBI home page
Sleep logoLink to Sleep
. 2009 Sep 1;32(9):1173–1181. doi: 10.1093/sleep/32.9.1173

Upper Airway Collapsibility During REM Sleep in Children with the Obstructive Sleep Apnea Syndrome

Jingtao Huang 1, Laurie R Karamessinis 1, Michelle E Pepe 1, Stephen M Glinka 1, John M Samuel 1, Paul R Gallagher 2, Carole L Marcus 1,✉
PMCID: PMC2737575  PMID: 19750922

Abstract

Study Objectives:

In children, most obstructive events occur during rapid eye movement (REM) sleep. We hypothesized that children with the obstructive sleep apnea syndrome (OSAS), in contrast to age-matched control subjects, would not maintain airflow in the face of an upper airway inspiratory pressure drop during REM sleep.

Design:

During slow wave sleep (SWS) and REM sleep, we measured airflow, inspiratory time, inspiratory time/total respiratory cycle time, respiratory rate, tidal volume, and minute ventilation at a holding pressure at which flow limitation occurred and at 5 cm H2O below the holding pressure in children with OSAS and in control subjects.

Setting:

Sleep laboratory.

Participants:

Fourteen children with OSAS and 23 normal control subjects.

Results:

In both sleep states, control subjects were able to maintain airflow, whereas subjects with OSAS preserved airflow in SWS but had a significant decrease in airflow during REM sleep (change in airflow of 18.58 ± 12.41 mL/s for control subjects vs −44.33 ± 14.09 mL/s for children with OSAS, P = 0.002). Although tidal volume decreased, patients with OSAS were able to maintain minute ventilation by increasing the respiratory rate and also had an increase in inspiratory time and inspiratory time per total respiratory cycle time

Conclusion:

Children with OSAS do not maintain airflow in the face of upper-airway inspiratory-pressure drops during REM sleep, indicating a more collapsible upper airway, compared with that of control subjects during REM sleep. However, compensatory mechanisms exist to maintain minute ventilation. Local reflexes, central control mechanisms, or both reflexes and control mechanisms need to be further explored to better understand the pathophysiology of this abnormality and the compensation mechanism.

Citation:

Huang J; Karamessinis LR; Pepe ME; Glinka SM; Samuel JM; Gallagher PR; Marcus CL. Upper airway collapsibility during REM sleep in children with the obstructive sleep apnea syndrome. SLEEP 2009;32(9):1173-1181.

Keywords: Critical pressure, slow wave sleep, sleep disordered breathing

RAPID EYE MOVEMENT (REM) SLEEP IS THE SLEEP STAGE DURING WHICH MOST OBSTRUCTIVE EVENTS OCCUR IN CHILDREN.1,2 HOWEVER, THE REASON WHY obstructive sleep apnea syndrome (OSAS) in children is more severe during REM than during non-REM (NREM) sleep is not fully understood. REM sleep in children with OSAS is associated with decreased pharyngeal muscle tone and blunted arousal responses to hypoxia, hypercapnia, and inspiratory resistive loading.3–6 Given the influence of REM sleep on upper airway muscle activity and ventilatory control, it is thought that children with OSAS may have a more collapsible upper airway during REM as compared with NREM sleep.

We hypothesized that children with OSAS, in contrast with age-matched control subjects, would not maintain airflow in the face of an inspiratory pressure drop during REM sleep.

Traditionally, upper airway collapsibility during sleep has been studied in children by assessing flow changes in response to incremental changes in nasal pressure.7–9 In our experience, this technique has been unsuccessful during REM because the children tended to arouse from sleep. Therefore, in the current study, we measured ventilation in response to a controlled upper airway pressure drop during REM and slow wave sleep (SWS) in children with OSAS and compared the results with those of control subjects.

Some of the results of this study have been previously reported in the form of an abstract.10

METHODS

Children with OSAS and control subjects were studied. Baseline polysomnography was performed in all subjects to evaluate them for the presence of OSAS. On a separate night, airflow, inspiratory time (Ti), total respiratory time (Tt), respiratory rate (RR), tidal volume (Vt), and expired volume per minute (V̇e) changes in response to drops in upper airway pressure were measured during SWS and REM sleep. The Institutional Review Board at the Children's Hospital of Philadelphia approved the study. Informed consent was obtained from the parents or legal guardians of the subjects, and assent was obtained from subjects who were 7 years and older.

Study Group

Fourteen subjects with OSAS and 25 control subjects, aged 5 to 12 years, were recruited. The upper age limit was chosen to limit the study to primarily prepubertal or early pubertal children,11 and the lower age limit to exclude those who were too young to cooperate with the face mask and other aspects of the protocol. To limit overlap between groups, subjects with OSAS were included only if their apnea-hypopnea index was at least 5 per hour, and control subjects were included only if their apnea-hypopnea index was less than 1.5 per hour.12–15 Both obese and nonobese subjects were included. Obesity was defined as body mass index greater than 95th percentile for age, race, and sex.16

Subjects with OSAS

Subjects with OSAS were recruited from those referred to the Sleep Center at the Children's Hospital of Philadelphia. Subjects were eligible for this study if they had no previous upper airway surgery, had no lower respiratory tract diseases except for mild asthma, and had no significant medical conditions (such as craniofacial anomalies or neuromuscular disease) other than OSAS.

Normal Control Subjects

Healthy subjects were recruited from the general community by means of advertisements. Subjects with a history of any combination of nightly snoring, adenoidectomy, or tonsillectomy or with medical conditions requiring daily medications were excluded.

Baseline Polysomnography

Subjects with OSAS and control subjects underwent baseline polysomnography.17–20 During the study, a Rembrandt polysomnography system (Embla, Broomfield, CO) recorded the following parameters: electroencephalogram (EEG)(C3/A2, C4/A1, O1/A2, O2/A1), left and right electrooculograms, submental electromyogram (EMG), tibial EMG, electrocardiogram, oronasal airflow with a 3-pronged thermistor (Pro-Tech Services, Inc., Mukilteo, WA), nasal pressure with a pressure transducer (Pro-Tech Services, Inc.), rib cage and abdominal wall motion using respiratory inductance plethysmography (Viasys Healthcare, Yorba Linda, CA), end-tidal PCO2 (Novametrix Medical Systems, Inc., Wallingford, CT) and arterial oxygen saturation (SpO2) with pulse waveform (Masimo, Irvine, CA). Subjects were also recorded on digital video. Sleep architecture, respiratory events, and arousals were scored using standard criteria.18,20–22

Steady-State Pressure-Flow Measurements

The subjects underwent overnight polysomnography on a separate night to determine the ventilatory response to drops in upper airway pressure. The methods were similar to those used in the baseline sleep study described above, with the following exceptions: Subjects wore a full face mask (Respironics, Murrysville, PA) attached to a heated pneumotachometer (Hans Rudolph, Inc., Kansas City, MO) with a differential pressure transducer (ADInstruments, Colorado Springs, CO). The pneumotachometer was then connected to a continuous positive airway pressure device customized by Respironics to provide both positive and negative pressure.9 Nasal pressure was measured at the mask using a pressure transducer with a demodulator (Validyne Engineering Corp., Northridge, CA). Signals were acquired on a PowerLab system (ADInstruments) and simultaneously displayed on a Rembrandt polysomnography system.

During SWS and REM, flow limitation was determined by the characteristic waveform pattern, consisting of increasing inspiratory flow followed by a midinspiratory plateau,7,23 rather than by using invasive esophageal pressure measurements, in these young children.24 Nasal pressure was titrated until flow limitation disappeared. The nasal pressure was then decreased slightly until flow limitation reappeared. That pressure was termed the holding pressure. The holding pressure was reevaluated before each trial. Studies were initiated with the subject breathing at the holding pressure for approximately 30 seconds. Nasal pressure was then decreased abruptly by approximately 5 cm H2O for 30 seconds. After each trial, nasal pressure was adjusted back to the holding pressure. Multiple trials were performed in SWS and REM for each subject, as tolerated; trials without a sleep-stage match were excluded. Trials with arousals or change of sleep state were also excluded.

Data Analysis

Vt was obtained by integrating the airflow signal. For each pair of trials in SWS and REM, inspiratory airflow, nasal pressure, inspiratory time and total respiratory time were measured for each flow-limited breath for 30 seconds prior to the pressure drop and averaged to obtain mean baseline inspiratory airflow ( V̇a), nasal pressure (Pa), Ti, Tt, RR, Vt, and V̇e. Similarly, for the pressure-drop period, inspiratory airflow ( V̇b), nasal pressure (Pb), Ti, Tt, RR, Vt, and V̇e were obtained. The average percentage change in airflow (ΔV̇ ) for the pressure drop was calculated as

graphic file with name aasm.32.9.1173m1.jpg

Previous studies have shown that, in children, the application of acute subatmospheric upper airway pressure during sleep results in reflex activation of the upper airway muscles after 3 to 5 breaths, with a resultant increase in airflow.3,8,25 Therefore, the analysis of the change in airflow ( V̇b - V̇a) and ΔV̇ was also performed utilizing only the first 3 breaths after the pressure drop and for the breaths during the pressure-drop period, excluding the first 3 breaths.

Statistical Analysis

Histograms and 1-sample Kolmogorov-Smirnov tests of the continuous outcome variables were examined to determine the extent of normalcy of distribution.

For demographic and polysomnographic data, means and standard deviations were reported by group (OSAS and control) (Table 1). Frequency counts and percentages were used for the categorical variable sex. Continuous data were analyzed using 2-sample t-tests, and categorical data were analyzed using the Fisher exact test.

Table 1.

Demographic and Polysomnographic Data

OSAS n = 14 Controls n = 23 P Value
Age, y 8 ± 2 9 ± 2 0.22
Girls, no. (%) 9 (64) 8 (35) 0.10
BMI Z-score 1.9 ± 0.9 1.0 ± 0.9 0.010
AHI, no./h 15.5 ± 9.1 0.3 ± 0.4 <0.0005
SaO2 nadir, % 83 ± 9 93 ± 2 0.001
Peak ETCO2, mm Hg 57 ± 5 50 ± 4 <0.0005
Open in a new tab

Continuous data are presented as mean ± SD; categorical data, as number (%). For continuous data, P values are from 2-sample t-tests; for categorical data, the P value is from the Fisher exact test. OSAS refers to subjects with obstructive sleep apnea syndrome; BMI, body mass index; AHI, apnea-hypopnea index; SaO2, arterial oxygen saturation; ETCO2, end-tidal CO2.

For respiratory mechanics, to account for the repeated trials within subjects, estimated least-squares means and standard errors from mixed-effects models were presented for each of the 4 group-by-sleep stage groupings. Primary outcome variables included (a) change in airflow and percentage change in airflow due to the pressure drop over the entire 30 seconds (V̇b - V̇a and ΔV̇); (b) change in airflow and percentage change in airflow (V̇b - V̇a) of the first 3 breaths after the pressure drop (V̇b - V̇a and ΔV̇, first 3 breaths only); (c) change in airflow and percentage change in airflow due to the pressure drop, excluding the first 3 breaths after the pressure drop (V̇ = V̇b - V̇a and ΔV̇, omit first 3 breaths).

The effects of group (OSAS vs control), sleep stage (SWS vs REM), and replication on the various outcomes were analyzed based on a mixed-effects approach, thereby accounting for the correlations among measurements made on the same subject. Various interaction terms were examined, with group-by-sleep-stage interaction being of primary interest. When effects of replication or interaction terms were nonsignificant, they were omitted from the final model. Contrast statements within the models were used to examine differences between sleep stages for each group separately and to examine differences between groups for each sleep stage separately. Additional analyses involved the inclusion of obesity status and subatmospheric Pb status in the model. A P value of less than 0.05 was considered significant.

RESULTS

Study Population

Fourteen subjects with OSAS and 25 control subjects were recruited. One control subject elected not to undergo the second night of study, and another had an abnormal baseline polysomnogram. Therefore, 23 control subjects completed the study. Subject characteristics are shown in Table 1. As expected, subjects with OSAS had worse breathing during sleep than did control subjects. Subjects with OSAS also had a higher body mass index.

Seven of the subjects with OSAS and 5 of the control subjects had a varying number of replicate (paired SWS and REM) trials, ranging from a second to a fourth replicate. Thus, the 14 subjects with OSAS provided 27 paired (SWS + REM) measurements, and the 23 control subjects provided 29 paired measurements.

Airflow at Baseline and during Pressure Drop

Typical examples of airflow change during pressure drops (V̇b - V̇a) for control subjects and subjects with OSAS in SWS and REM are shown in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Typical examples of airflow changes during pressure drop for control subjects and subjects with obstructive sleep apnea syndrome (OSAS) in slow wave sleep (SWS) and rapid eye movement (REM) sleep are shown. Epochs shown have a time frame of 30 seconds, showing a portion of a trial. The scale was −1000 to 1000 mL/s for the airflow channel and −8 to 5 cm h2o for the nasal pressure channel. During inspiration, airflow and pressure signal tracing have a negative deflection. Initially, the airflow was at the holding pressure. The arrow designates the time point at which nasal pressure was dropped. Control subjects maintained airflow in the face of pressure drop during both SWS and REM sleep, whereas subjects with OSAS did not maintain airflow. The baseline of the airflow tracing moved upwards during pressure drop due to the change of bias flow.

The difference in airflow during the pressure drop compared to baseline (i.e., V̇b - V̇a) was analyzed using a mixed-effects model (Table 2). Group had a statistically significant effect (P = 0.011), but sleep stage was not significant (P = 0.84). However, the group-by-sleep-stage interaction effect was significant (P = 0.048), indicating that the group effect was different for each sleep stage. Contrasts between SWS and REM were examined for each group separately using contrast statements within the mixed-effects models. The effect of sleep stage remained nonsignificant (P = 0.13 and 0.19, for OSAS and control subjects, respectively). Contrasts between OSAS and control subjects for each sleep stage separately indicated that the effect of group was nonsignificant in SWS stage (P = 0.28) but was significant in REM sleep (P = 0.002). Estimated least-squares means and standard errors (SE) for OSAS SWS, OSAS REM, control SWS, and control REM groupings were −21.03 ± 14.06, −44.31 ± 14.06, −0.56 ± 12.34, and 18.56 ± 12.34 mL/s, respectively. The greatest difference between airflow during pressure drop and airflow at baseline was a drop in airflow for patients with OSAS during REM sleep (-44.31 ± 14.06 mL/s), whereas the control subjects in REM sleep had an increase in airflow (18.56 ± 12.34 mL/s). In summary, during REM sleep, V̇b - V̇a increased in control subjects but decreased in patients with OSAS; in SWS, no difference in V̇b - V̇a was observed between the 2 groups.

Table 2.

Estimated Least-Squares Means and Standard Errors and P Values for Pairwise Contrasts for Each of the Outcome Measures, from Mixed-Effects Models

Outcome SWS vs REM
 OSAS vs Control
OSAS Controls SWS REM
V̇b - V̇a, mL/s -21.03 ± 14.06 vs -0.56 ± 12.34 vs -21.03 ± 14.06 vs -44.31 ± 14.06 vs
    Total -44.31 ± 14.06 18.56 ± 12.34 -0.56 ± 12.34 18.56 ± 12.34
P = 0.13 P = 0.19 P = 0.28 P = 0.002
V̇b - V̇a, mL/s -59.58 ± 14.97 vs -16.94 ± 12.90 vs -59.58 ± 14.97 vs -51.44 ± 14.97 vs
    First 3 breaths only -51.44 ± 14.97 26.02 ± 12.71 -16.94 ± 12.90 26.02 ± 12.71
P = 0.54 P = 0.002 P = 0.038 P = 0.0004
V̇b - V̇a, mL/s -3.17 ± 15.69 vs 2.85 ± 14.01 vs -3.17 ± 15.69 vs -41.47 ± 15.69 vs
    Without first 3 breaths -41.47 ± 15.69 20.63 ± 14.44 2.85 ± 14.01 20.63 ± 14.44
P = 0.036 P = 0.31 P = 0.78 P = 0.006
ΔV̇, % -9.85 ± 7.81 vs 2.11 ± 6.89 vs -9.85 ± 7.81 vs -21.07 ± 7.81 vs
    Total -21.07 ± 7.81 20.77 ± 6.89 2.11 ± 6.89 20.77 ± 6.89
P = 0.19 P = 0.027 P = 0.26 P = 0.0003
ΔV̇, % -26.26 ± 9.46 vs -2.59 ± 8.19 vs -26.26 ± 9.46 vs -20.78 ± 9.46 vs
    First 3 breaths only -20.78 ± 9.46 28.41 ± 8.06 -2.59 ± 8.19 28.41 ± 8.06
P = 0.53 P = 0.0007 P = 0.067 P = 0.0004
ΔV̇, % -1.98 ± 8.02 vs 3.17 ± 7.15 vs -1.98 ± 8.02 vs -21.86 ± 8.02 vs
    Withoutfirst 3 breaths -21.86 ± 8.02 22.78 ± 7.37 3.17 ± 7.15 22.78 ± 7.37
P = 0.032 P = 0.032 P = 0.63 P = 0.0002
Open in a new tab

OSAS refers to obstructive sleep apnea syndrome; SWS, slow wave sleep; REM, rapid eye movement sleep. See text for other abbreviations.

V̇b - V̇a was analyzed for the first 3 breaths after the pressure drop. There were statistically significant effects of group (P = 0.002) and sleep stage (P = 0.009) and a marginal, albeit nonsignificant, group-by-sleep-stage interaction (P = 0.067). Airflow decreased during those 3 breaths, compared with baseline, for both groups in SWS and for the OSAS group in REM sleep but increased in REM sleep for the control group. In SWS, the decrement in airflow was more prominent in the OSAS group than in the control group. This is consistent with the results of previous studies demonstrating a hypotonic upper airway during the first 3 breaths after a pressure drop.8,25

V̇b - V̇a was also analyzed excluding the first 3 breaths after the pressure drop. Results from V̇b - V̇a excluding the first 3 breaths and from the overall flow difference including all breaths reached essentially the same conclusions: in REM sleep, V̇b - V̇a increased in control subjects but decreased in patients with OSAS.

Average Percentage Change in Airflow

Analysis of ΔV̇ indicated a statistically significant effect of group (P = 0.003), a nonsignificant effect of sleep stage (P = 0.49), and a significant group-by-sleep-stage interaction effect (P = 0.026) (Table 2 and Figure 2). Contrasts between SWS and REM were examined for each group separately. The effect of sleep stage remained nonsignificant for the OSAS group but was significant for control subjects (P = 0.27 and 0.036, for OSAS and control subjects, respectively). Contrasts between OSAS and control subjects for each sleep stage separately found that the effect of group was nonsignificant in SWS stage (P = 0.20) but was highly significant in REM sleep (P = 0.0004). Estimated least-squares means and SEs of ΔV̇ for OSAS SWS, OSAS REM, control SWS, and control REM groupings were −2.58 ± 1.68, −4.63 ± 1.68, 0.37 ± 1.48, and 4.20 ± 1.48 %, respectively. From flow at baseline to flow during pressure drops, there was a drop in flow for OSAS patients during REM sleep (-4.63 ± 1.68 %) and an increase in flow for control subjects during REM sleep (4.20 ± 1.48 %). In summary, during REM sleep, ΔV̇ in.creased in control subjects but decreased in patients with OSAS; during SWS, no difference in ΔV̇ was observed between the 2 groups.

Figure 2.

Figure 2

Open in a new tab

Boxplots of ΔV̇ omitting the first 3 breaths. The box represents the interquartile range that contains 50% of the values. The line across the box indicates the median. The whiskers extend from the box to the highest and lowest values, excluding outliers (O), which are defined as cases with values between 1.5 and 3 box lengths from either end of the box. No difference was observed between patients with the obstructive sleep apnea syndrome (OSAS) and control subjects in slow wave sleep (SWS), but ΔV̇ increased in control subjects and decreased in children with OSAS in rapid eye movement sleep (REM).

Analysis of ΔV̇ for the first 3 breaths after the pressure drop had results essentially identical to those of the analysis of V̇b - V̇a for the first 3 breaths. ΔV̇ decreased during the first 3 breaths, compared with baseline, for both groups in SWS and for the OSAS group in REM sleep but increased in REM sleep for the control group. In SWS, the decrement in airflow was more prominent in the OSAS group than in the control group.

Analysis of ΔV̇ excluding the first 3 breaths after pressure drop (Figure 2) still indicated that ΔV̇ increased in control subjects but decreased in patients with OSAS during REM sleep; no difference in ΔV̇ was observed between the 2 groups during SWS.

In summary, further analysis of V̇b - V̇a and ΔV̇ showed that, in OSAS during both sleep stages and in control subjects during SWS, there was decreased average airflow for the first 3 breaths after the pressure drop compared with the average airflow at the holding pressure, whereas, in control subjects during REM sleep, there was an increased average flow. In SWS after the first 3 breaths, both groups resumed airflow close to the level at the holding pressure, and, overall, there was no difference in the change in airflow between the 2 groups. In REM sleep, in contrast with the control group, airflow in the OSAS group remained below the level at the holding pressure.

Effects of a Negative Pb

All of the trials had a positive Pa; however, all of the trials from control subjects had a negative Pb during sleep except for 1 during REM sleep, whereas only 4 pairs of trials from subjects with OSAS had a negative Pb. The effects of a negative Pb were therefore evaluated. When Pbwas included in the model as either a categorical variable or a continuous covariate, it had no statistically significant effects on V̇b - V̇a with or without the first 3 breaths after pressure drop.

Effect of Obesity

Because most of the subjects with OSAS but only a few of the control subjects were obese, the effect of obesity on the pressure-flow relationship was evaluated. When the status of obesity (yes/no) was included in the mixed-effects models, no effect of obesity status was observed for any of the outcomes involving flow difference, and the overall results did not change. When contrast statements were used to examine pairwise differences between obese and nonobese groups (obese versus nonobese in control subjects, in subjects with OSAS, and in both groups combined), none were found in any of the outcomes involving flow difference, neither in SWS nor in REM. Regardless of obesity status, when challenged with upper airway pressure drops during REM sleep, patients with OSAS still demonstrated decreased airflow, whereas control subjects showed increased airflow. For instance, for V̇b - V̇a (mL/s) during REM sleep, least-squared means ± SE of the obese OSAS group was −46.96 ± 16.04 versus −29.22 ± 32.30 for the nonobese OSAS group; least-squared means ± SE of obese control subjects was 14.29 ± 28.80 versus 20.04 ± 13.96 for nonobese control subjects.

Changes in Ti, Ti/Tt, RR, Vt and V̇e

Changes in Ti, Ti/Tt, RR, Vt and V̇e, from baseline to pressure drop (i.e., measurement during pressure drop minus measurement at baseline) were analyzed using mixed effects models. Table 3 contains estimated least-squares means and standard errors for each of the 4 group-by-sleep-stage groupings.

Table 3.

Estimated Least-Squares Means and Standard Errors for Each of the 4 Group-by-Sleep-Stage Groupings, Based on Mixed-Effects Models Examining the Effects of Group, Sleep Stage, and Group-by-Sleep-Stage Interaction

Outcome OSAS SWS OSAS REM Control SWS Control REM
ΔTi, s 0.27 ± 0.06 0.40 ± 0.06 0.18 ± 0.05 0.16 ± 0.05
ΔTi/Tt 0.13 ± 0.02 0.12 ± 0.02 0.09 ± 0.01 0.05 ± 0.01
ΔRR, /min 2.07 ± 0.61 0.38 ± 0.61 1.00 ± 0.48 0.68 ± 0.48
ΔVt, mL -18.21 ± 22.43 -0.67 ± 22.43 28.63 ± 18.95 50.81 ± 18.95
ΔV̇e, L/min 0.56 ± 0.44 0.16 ± 0.44 0.86 ± 0.37 0.98 ± 0.37
Open in a new tab

Abbreviations: OSAS refers to obstructive sleep apnea syndrome; SWS, slow wave sleep; REM, rapid eye movement sleep; Ti, inspiratory time; Tt, total respiratory time; RR, respiration rate; Vt, tidal volume; V̇e, minute ventilation.

For changes in Ti and Ti/Tt, there was a significant effect of group (P = 0.022 and 0.020, for Ti and Ti/Tt, respectively) but no effect of sleep stage. Increases in both Ti and Ti/Tt were higher in patients with OSAS as compared with control subjects. For change in RR, there was a significant effect of sleep stage (P = 0.032) but no effect of group. The increment in RR was higher in SWS than in REM.

For change in Vt, there was a significant effect of group (P = 0.037) but no effect of sleep stage. Vt increased in control subjects but decreased in patients with OSAS. Although Vt decreased in patients with OSAS overall, the decrease was considerably lower in patients with OSAS during SWS (-18.21 ± 22.43), as compared with patients with OSAS during REM sleep (-0.67 ± 22.43). Likewise, the Vt increase in control subjects was seemingly more pronounced during REM sleep (50.81 ± 18.95), as opposed to during SWS (28.63 ± 18.95).

For change in V̇e, neither group nor sleep stage had a statistically significant effect. There were no significant group-by-sleep-stage interaction effects in any of these models.

In summary, control subjects were able to maintain Vt and V̇e in response to the pressure drops by increasing Ti, Ti/Tt, and RR in both sleep stages. In patients with OSAS, although Vt decreased during both SWS and REM in response to the mild pressure drops, V̇e was maintained by increasing the RR. The increments in Ti and Ti/Tt were larger in patients with OSAS than in control subjects.

DISCUSSION

This study investigated ventilatory changes in response to upper airway pressure drops during SWS and REM sleep in subjects with OSAS and control subjects. In both sleep states, control subjects were able to preserve airflow, Vt, and V̇e; in subjects with OSAS, airflow was maintained during SWS but decreased significantly during REM sleep. There was a highly significant decrease in the change in airflow during REM sleep in subjects with OSAS compared with control subjects, whereas no significant changes were observed between the groups during SWS. In patients with OSAS, Vt decreased in response to the pressure drop during sleep despite an increased Ti that was not sufficient to maintain the Vt. However, patients with OSAS were able to maintain V̇e by increasing RR.

Technique for Measuring Upper Airway Collapsibility

In children, upper airway collapsibility during NREM sleep has traditionally been evaluated by assessing flow changes in response to stepwise decreases in nasal pressure and determining the critical closing pressure (Pcrit) and slope of this relationship.7,9 However, this method is difficult to apply during REM sleep because it requires measurement of airflow at multiple pressure levels over a relatively long time (a few minutes), resulting in arousal during REM sleep. Also, typically only 2 to 3 breaths are selected and analyzed at each pressure level, but breathing in REM sleep is very variable, so that selected breaths may not represent the ventilation in REM sleep. The method used in the current study required nasal pressure to be decreased for only 30 seconds so that all subjects tolerated the challenge. For this method, change in airflow rather than critical closing pressure was the main outcome parameter because only 2 pressure levels were evaluated.

Upper Airway Muscle Activity during Sleep

REM sleep in children is characterized by phasic muscle twitches, irregular breathing, and hypotonia of skeletal muscles other than the diaphragm. During REM sleep, pharyngeal muscle tone decreases, as compared with during NREM sleep.4 Previous studies during NREM sleep have shown that genioglossal EMG activity increases 3 to 5 breaths after an upper airway pressure drop, with a resultant increase in airflow.3,8,25 Thus, the first 3 breaths after a pressure drop reflect a relatively hypotonic upper airway. The current study did not utilize EMG measurements. Nevertheless, the decrease in airflow during the first 3 breaths after the pressure drop, in both OSAS and control subjects, and during both SWS and REM sleep is consistent with the previous finding of relative hypotonia of the upper airway muscles during abrupt pressure changes. Subsequent to these first 3 breaths, subjects with OSAS displayed a more collapsible upper airway during REM sleep than did control subjects. These results suggest that, in REM sleep, normal control subjects still have sufficient pharyngeal muscle activity and active reflexes to maintain airflow, whereas subjects with OSAS do not.

Adults vs Children

Previous studies in children have shown that most obstructive events occur during REM sleep.1 In contrast, in adults, obstructive events occur frequently during NREM sleep.26 Only a few studies have investigated pressure-flow relationships during REM sleep in adults,27–29 and none have investigated these relationships in children. Boudewyns et al.27 studied pressure-flow relationships in 10 obese adults with OSAS by stepwise decreasing the nasal pressure and found no difference in upper airway collapsibility between NREM and REM sleep. Rowley et al.28 studied spontaneous breathing during wakefulness, NREM sleep, and REM sleep in 5 normal adults and observed the relationship between airflow and supraglottic pressure changes. They found that the upper airway was more collapsible during NREM than REM sleep. In a separate study, they investigated the relationship between oropharyngeal cross-sectional area and pharyngeal pressure in normal adults during wakefulness and sleep.29 They found no differences in these parameters among wakefulness, NREM and REM sleep. The above studies suggest that, for adult control subjects and subjects with OSAS, upper airway collapsibility does not increase when subjects are in REM sleep. The relatively constant upper airway collapsibility during both NREM and REM sleep may explain why obstructive events do not occur predominantly during REM sleep in many adults. In contrast, our study suggests that, in REM but not in SWS, children with OSAS have a more collapsible upper airway compared with that of control subjects, predisposing them to developing obstructive apnea predominantly during REM sleep.

Minute Ventilation

In this study, although patients with OSAS had reduced airflow and tidal volumes in response to the mild airway pressure drops, they maintained V̇e by increasing their RR. Previous studies of normal children and adults during NREM sleep have shown an increased Ti/Tt in response to inspiratory resistance loading.6,30,31 Children also increase their RR but, nevertheless have a decrease in Vt and e.6,30,31 The response to inspiratory loading during sleep has not been studied previously in children with OSAS. In the current study, airflow and Vt were maintained in control subjects but not in patients with OSAS during sleep, although increases in Ti and Ti/Tt and maintenance of e were observed in both groups. The discrepancies in the results of this study and the prior pediatric study6 may be due to the different stimulus modalities applied or to the different sizes of the load. Our speculation is that, although the neuromuscular activity of the upper airway is compromised in patients with OSAS, compensatory mechanisms exist that help maintain V̇e.

Pa and Pb

In this study, both control subjects and subjects with OSAS had a positive holding pressure, although the holding pressure of subjects with OSAS was higher than that of the control subjects. Also, all the trials from the control subjects except one had a negative Pb during REM sleep, whereas most of the trials from subjects with OSAS had a positive Pb. Positive nasal pressure is known to result in upper airway relaxation, manifested by reduced genioglossal EMG activity, in instrumented animals32 and in children.4 In adults, the results are controversial: studies have shown that patients with OSAS demonstrate both increased and decreased genioglossal EMG activity in the face of positive airway pressure.33,34 To determine whether the current study results were affected by differences in pressure levels between the groups, we included Pb in the model as either a categorical variable or a continuous covariate and found no statistically significant effects on airflow. Therefore, we believe that the differences in upper airway responses in REM sleep between the 2 groups cannot be explained by the different levels of applied pressure (Pb) but, rather, by different upper airway collapsibility in REM sleep between the 2 groups.

Limitations and Future Directions

This study has several limitations. First, upper airway EMG was not measured and, thus, assumptions about upper airway hypotonia are based on previous studies that utilized EMG.3,8 Second, the changes in airflow in response to inspiratory pressure drops during sleep were not reevaluated after successful treatment of the OSAS. For example, weight loss may reverse OSAS in a manner that minimally changes the upper airway physiology. Comparing results before and after reversal of OSAS by weight loss may help establish whether there is a cause-effect relationship between responses to airway pressure changes and OSAS in children.

Future studies examining the mechanisms of the changes in airflow in response to pressure drops during REM sleep in OSAS are needed. In particular, measurements of lung volumes would be useful because reductions in lung volume result in a decreased upper airway cross-sectional area.35,36 REM sleep is associated with lower lung volumes, including a lower functional residual capacity,37–39 particularly in children who have a very compliant chest wall. It is possible that the decreased functional residual capacity during REM sleep contributed to decreased upper airway size and that children with OSAS (unlike control subjects) could not compensate for this change. Further investigation into lung volume and the activity of genioglossus, the major upper airway dilation muscle, will lead to a better understanding of the difference in change in the airflow between the 2 groups during REM sleep. In addition, esophageal pressure measurements would help determine whether airflow changes were mediated, at least in part, by a change in central ventilatory drive.

In this study, changes in airflow in response to pressure changes were not associated with EEG evidence of cortical arousal. However, evaluation of more subtle changes in central nervous system activity may help explain why there are periods of relative stability and instability in breathing in patients with OSAS.40 Several methods have been applied to investigate subtle EEG changes during sleep, including respiratory-related evoked potentials19,41 and respiratory cycle-related EEG changes.42,43 Using these techniques, future studies may further explore central mechanisms associated with compensatory responses to airway pressure changes during sleep.

CONCLUSION

The results of the present study support the hypothesis that children with OSAS do not maintain upper airway airflow in the face of an inspiratory pressure drop during REM sleep, indicating a more collapsible upper airway, compared with control subjects, during REM sleep. Interestingly, in this study, airflow changed in response to pressure drops without visible arousal or EEG change. Also, patients with OSAS were able to maintain their V̇e by increasing their RR. Local reflexes, central control mechanisms for this phenomenon, or both local reflexes and central control mechanisms need to be further explored. More-detailed EEG analysis, such as frequency-specific EEG power analysis, may shed light on this mystery.43

DISCLOSURE STATEMENT

This was not an industry supported study. Dr. Marcus has participated in a roundtable discussion for and has received research support from Respironics. The other authors have indicated no financial conflicts of interest.

ACKNOWLEDGMENTS

This study was supported by NIH grants M01 RR000240, UL1 RR024134, and R01 HL58585. Respironics, Inc. provided the airway pressure device and unrestricted research support that paid for a research technician. The authors thank all of the Children's Hospital of Philadelphia sleep laboratory technologists who helped conduct this study. We thank Joseph McDonough, MS, biomedical engineer for Pulmonary Division at the Children's Hospital of Philadelphia, for his technical support. We are grateful to the children and their families for their enthusiastic participation in this study.

REFERENCES

  • 1.Goh DY, Galster P, Marcus CL. Sleep architecture and respiratory disturbances in children with obstructive sleep apnea. Am J Respir Crit Care Med. 2000;162:682–6. doi: 10.1164/ajrccm.162.2.9908058. [DOI] [PubMed] [Google Scholar]
  • 2.Morielli A, Ladan S, Ducharme FM, Brouillette RT. Can sleep and wakefulness be distinguished in children by cardiorespiratory and videotape recordings? Chest. 1996;109:680–7. doi: 10.1378/chest.109.3.680. [DOI] [PubMed] [Google Scholar]
  • 3.Katz ES, Marcus CL, White DP. Influence of airway pressure on genioglossus activity during sleep in normal children. Am J Respir Crit Care Med. 2006;173:902–9. doi: 10.1164/rccm.200509-1450OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Katz ES, White DP. Genioglossus activity during sleep in normal control subjects and children with obstructive sleep apnea. Am J Respir Crit Care Med. 2004;170:553–60. doi: 10.1164/rccm.200403-262OC. [DOI] [PubMed] [Google Scholar]
  • 5.Marcus CL, Lutz J, Carroll JL, Bamford O. Arousal and ventilatory responses during sleep in children with obstructive sleep apnea. J Appl Physiol. 1998;84:1926–36. doi: 10.1152/jappl.1998.84.6.1926. [DOI] [PubMed] [Google Scholar]
  • 6.Marcus CL, Moreira GA, Bamford O, Lutz J. Response to inspiratory resistive loading during sleep in normal children and children with obstructive apnea. J Appl Physiol. 1999;87:1448–54. doi: 10.1152/jappl.1999.87.4.1448. [DOI] [PubMed] [Google Scholar]
  • 7.Marcus CL, McColley SA, Carroll JL, Loughlin GM, Smith PL, Schwartz AR. Upper airway collapsibility in children with obstructive sleep apnea syndrome. J Appl Physiol. 1994;77:918–24. doi: 10.1152/jappl.1994.77.2.918. [DOI] [PubMed] [Google Scholar]
  • 8.Marcus CL, Katz ES, Lutz J, Black CA, Galster P, Carson KA. Upper airway dynamic responses in children with the obstructive sleep apnea syndrome. Pediatr Res. 2005;57:99–107. doi: 10.1203/01.PDR.0000147565.74947.14. [DOI] [PubMed] [Google Scholar]
  • 9.Bandla P, Huang J, Karamessinis L, et al. Puberty and upper airway dynamics during sleep. Sleep. 2008;31:534–41. doi: 10.1093/sleep/31.4.534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Huang J, Samuel J, Pepe M, et al. Upper airway collapsibility during REM sleep in children. Proceedings of the American Thoracic Society. 2006;3:A555. [Google Scholar]
  • 11.Lee PA, Guo SS, Kulin HE. Age of puberty: data from the United States of America. APMIS. 2001;109:81–8. doi: 10.1034/j.1600-0463.2001.d01-107.x. [DOI] [PubMed] [Google Scholar]
  • 12.Marcus CL, Omlin KJ, Basinki DJ, et al. Normal polysomnographic values for children and adolescents. Am Rev Respir Dis. 1992;146:1235–9. doi: 10.1164/ajrccm/146.5_Pt_1.1235. [DOI] [PubMed] [Google Scholar]
  • 13.Witmans MB, Keens TG, Davidson Ward SL, Marcus CL. Obstructive hypopneas in children and adolescents: normal values. Am J Respir Crit Care Med. 2003;168:1540. doi: 10.1164/ajrccm.168.12.954. [DOI] [PubMed] [Google Scholar]
  • 14.Uliel S, Tauman R, Greenfeld M, Sivan Y. Normal polysomnographic respiratory values in children and adolescents. Chest. 2004;125:872–8. doi: 10.1378/chest.125.3.872. [DOI] [PubMed] [Google Scholar]
  • 15.Traeger N, Schultz B, Pollock AN, Mason T, Marcus CL, Arens R. Polysomnographic values in children 2-9 years old: additional data and review of the literature. Pediatr Pulmonol. 2005;40:22–30. doi: 10.1002/ppul.20236. [DOI] [PubMed] [Google Scholar]
  • 16.Rosner B, Prineas R, Loggie J, Daniels SR. Percentiles for body mass index in U.S. children 5 to 17 years of age. J Pediatr. 1998;132:211–22. doi: 10.1016/s0022-3476(98)70434-2. [DOI] [PubMed] [Google Scholar]
  • 17.American Thoracic Society. Standards and indications for cardiopulmonary sleep studies in children. Am J Respir Crit Care Med. 1996;153:866–78. doi: 10.1164/ajrccm.153.2.8564147. [DOI] [PubMed] [Google Scholar]
  • 18.EEG arousals: scoring rules and examples: a preliminary report from the Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. Sleep. 1992;15:173–84. [PubMed] [Google Scholar]
  • 19.Huang J, Colrain IM, Melendres MC, et al. Cortical processing of respiratory afferent stimuli during sleep in children with the obstructive sleep apnea syndrome. Sleep. 2008;31:403–10. doi: 10.1093/sleep/31.3.403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Rechtschaffen A, Kales A. A manual of standardized terminology: Techniques and scoring systems for sleep stages of human subjects. Bethesda, MD: NINDB Neurological Information Network (US); 1968. [Google Scholar]
  • 21.American Academy of Sleep Medicine Task Force. Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research. The Report of an American Academy of Sleep Medicine Task Force. Sleep. 1999;22:667–89. [PubMed] [Google Scholar]
  • 22.American Thoracic Society. Standards and indications for cardiopulmonary sleep studies in children. Am J Respir Crit Care Med. 1996;153:866–78. doi: 10.1164/ajrccm.153.2.8564147. [DOI] [PubMed] [Google Scholar]
  • 23.Condos R, Norman RG, Krishnasamy I, Peduzzi N, Goldring RM, Rapoport DM. Flow limitation as a noninvasive assessment of residual upper-airway resistance during continuous positive airway pressure therapy of obstructive sleep apnea. Am J Respir Crit Care Med. 1994;150:475–80. doi: 10.1164/ajrccm.150.2.8049832. [DOI] [PubMed] [Google Scholar]
  • 24.Chervin RD, Ruzicka DL, Wiebelhaus JL, et al. Tolerance of esophageal pressure monitoring during polysomnography in children. Sleep. 2003;26:1022–6. doi: 10.1093/sleep/26.8.1022. [DOI] [PubMed] [Google Scholar]
  • 25.Marcus CL, Fernandes Do Prado LB, Lutz J, et al. Developmental changes in upper airway dynamics. J Appl Physiol. 2004;97:98–108. doi: 10.1152/japplphysiol.00462.2003. [DOI] [PubMed] [Google Scholar]
  • 26.Charbonneau M, Marin JM, Olha A, Kimoff RJ, Levy RD, Cosio MG. Changes in obstructive sleep apnea characteristics through the night. Chest. 1994;106:1695–701. doi: 10.1378/chest.106.6.1695. [DOI] [PubMed] [Google Scholar]
  • 27.Boudewyns A, Punjabi N, Van de Heyning PH, et al. Abbreviated method for assessing upper airway function in obstructive sleep apnea. Chest. 2000;118:1031–41. doi: 10.1378/chest.118.4.1031. [DOI] [PubMed] [Google Scholar]
  • 28.Rowley JA, Zahn BR, Babcock MA, Badr MS. The effect of rapid eye movement (REM) sleep on upper airway mechanics in normal human subjects. J Physiol. 1998;510(Pt 3):963–76. doi: 10.1111/j.1469-7793.1998.00963.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Rowley JA, Sanders CS, Zahn BR, Badr MS. Effect of REM sleep on retroglossal cross-sectional area and compliance in normal subjects. J Appl Physiol. 2001;91:239–48. doi: 10.1152/jappl.2001.91.1.239. [DOI] [PubMed] [Google Scholar]
  • 30.Wiegand L, Zwillich CW, White DP. Sleep and the ventilatory response to resistive loading in normal men. J Appl Physiol. 1988;64:1186–95. doi: 10.1152/jappl.1988.64.3.1186. [DOI] [PubMed] [Google Scholar]
  • 31.Badr MS, Skatrud JB, Dempsey JA, Begle RL. Effect of mechanical loading on expiratory and inspiratory muscle activity during NREM sleep. J Appl Physiol. 1990;68:1195–202. doi: 10.1152/jappl.1990.68.3.1195. [DOI] [PubMed] [Google Scholar]
  • 32.Mathew OP, Abu-Osba YK, Thach BT. Influence of upper airway pressure changes on respiratory frequency. Respir Physiol. 1982;49:223–33. doi: 10.1016/0034-5687(82)90075-5. [DOI] [PubMed] [Google Scholar]
  • 33.Rapoport DM, Garay SM, Goldring RM. Nasal CPAP in obstructive sleep apnea: mechanisms of action. Bull Eur Physiopathol Respir. 1983;19:616–20. [PubMed] [Google Scholar]
  • 34.Strohl KP, Redline S. Nasal CPAP therapy, upper airway muscle activation, and obstructive sleep apnea. Am Rev Respir Dis. 1986;134:555–8. doi: 10.1164/arrd.1986.134.3.555. [DOI] [PubMed] [Google Scholar]
  • 35.Hoffstein V, Zamel N, Phillipson EA. Lung volume dependence of pharyngeal cross-sectional area in patients with obstructive sleep apnea. Am Rev Respir Dis. 1984;130:175–8. doi: 10.1164/arrd.1984.130.2.175. [DOI] [PubMed] [Google Scholar]
  • 36.Stanchina ML, Malhotra A, Fogel RB, et al. The influence of lung volume on pharyngeal mechanics, collapsibility, and genioglossus muscle activation during sleep. Sleep. 2003;26:851–6. doi: 10.1093/sleep/26.7.851. [DOI] [PubMed] [Google Scholar]
  • 37.Hudgel DW, Devadatta P. Decrease in functional residual capacity during sleep in normal humans. J Appl Physiol. 1984;57:1319–22. doi: 10.1152/jappl.1984.57.5.1319. [DOI] [PubMed] [Google Scholar]
  • 38.Ballard RD, Irvin CG, Martin RJ, Pak J, Pandey R, White DP. Influence of sleep on lung volume in asthmatic patients and normal subjects. J Appl Physiol. 1990;68:2034–41. doi: 10.1152/jappl.1990.68.5.2034. [DOI] [PubMed] [Google Scholar]
  • 39.Henderson-Smart DJ, Read DJ. Reduced lung volume during behavioral active sleep in the newborn. J Appl Physiol. 1979;46:1081–5. doi: 10.1152/jappl.1979.46.6.1081. [DOI] [PubMed] [Google Scholar]
  • 40.Younes M. Contributions of upper airway mechanics and control mechanisms to severity of obstructive apnea. Am J Respir Crit Care Med. 2003;168:645–58. doi: 10.1164/rccm.200302-201OC. [DOI] [PubMed] [Google Scholar]
  • 41.Melendres M, Marcus CL, Abi-Raad R, Trescher WH, Lutz JM, Colrain IM. Respiratory-related evoked potentials during sleep in children. Sleep. 2008;31:55–61. doi: 10.1093/sleep/31.1.55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Chervin RD, Burns JW, Subotic NS, Roussi C, Thelen B, Ruzicka DL. Correlates of respiratory cycle-related EEG changes in children with sleep-disordered breathing. Sleep. 2004;27:116–21. doi: 10.1093/sleep/27.1.116. [DOI] [PubMed] [Google Scholar]
  • 43.Chervin RD, Burns JW, Subotic NS, Roussi C, Thelen B, Ruzicka DL. Method for detection of respiratory cycle-related EEG changes in sleep-disordered breathing. Sleep. 2004;27:110–5. doi: 10.1093/sleep/27.1.110. [DOI] [PubMed] [Google Scholar]

RESOURCES