Journal of Applied Physiology Ad Instruments
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 103: 1929-1941, 2007. First published September 6, 2007; doi:10.1152/japplphysiol.00561.2007
8750-7587/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
103/6/1929    most recent
00561.2007v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Younes, M.
Right arrow Articles by Hanly, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Younes, M.
Right arrow Articles by Hanly, P.

Mechanisms of breathing instability in patients with obstructive sleep apnea

Magdy Younes, Michele Ostrowski, Raj Atkar, John Laprairie, AnDrea Siemens, and Patrick Hanly

Sleep Centre, Foothills Medical Centre, University of Calgary, Calgary, Alberta, Canada

Submitted 23 May 2007 ; accepted in final form 5 September 2007


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
The response to chemical stimuli (chemical responsiveness) and the increases in respiratory drive required for arousal (arousal threshold) and for opening the airway without arousal (effective recruitment threshold) are important determinants of ventilatory instability and, hence, severity of obstructive apnea. We measured these variables in 21 obstructive apnea patients (apnea-hypopnea index 91 ± 24 h–1) while on continuous-positive-airway pressure. During sleep, pressure was intermittently reduced (dial down) to induce severe hypopneas. Dial downs were done on room air and following {approx}30 s of breathing hypercapneic and/or hypoxic mixtures, which induced a range of ventilatory stimulation before dial down. Ventilation just before dial down and flow during dial down were measured. Chemical responsiveness, estimated as the percent increase in ventilation during the 5th breath following administration of 6% CO2 combined with {approx}4% desaturation, was large (187 ± 117%). Arousal threshold, estimated as the percent increase in ventilation associated with a 50% probability of arousal, ranged from 40% to >268% and was <120% in 12/21 patients, indicating that in many patients arousal occurs with modest changes in chemical drive. Effective recruitment threshold, estimated as percent increase in pre-dial-down ventilation associated with a significant increase in dial-down flow, ranged from zero to >174% and was <110% in 12/21 patients, indicating that in many patients reflex dilatation occurs with modest increases in drive. The two thresholds were not correlated. In most OSA patients, airway patency may be maintained with only modest increases in chemical drive, but instability results because of a low arousal threshold and a brisk increase in drive following brief reduction in alveolar ventilation.

mechanisms; dial down; effective recruitment threshold; arousal threshold

UPPER AIRWAY OPENING at the end of obstructive apneas and hypopneas usually consists of an abrupt increase in flow that is almost always associated with cortical arousal. The "suddenness" of the flow response along with an accompanying arousal have led to the enduring notion that arousal is the mechanism responsible for opening (38). Indeed, arousal has been considered an essential survival mechanism in this disorder (33). If arousal is necessary for upper airway opening, then the mechanism of ventilatory instability is fairly simple; obstruction continues until arousal occurs and recurs on resumption of sleep.

Recently, Younes proposed that most obstructive sleep apnea (OSA) patients could open their airway without arousal (52, 53). This was based on several observations including the following. 1) Most patients develop periods of stable breathing, and sleep, in the same body position and sleep state in which they experience OSA (52), thereby signifying that upper airway dilator activity can be increased sufficiently to maintain patency in most patients during sleep. 2) In many obstructive events, airway opening preceded arousal, and the temporal relation between arousal and airway opening suggested an incidental association (53). 3) Upper airway opening occurred at the same time regardless of whether arousal occurred before or after opening or did not occur at all (53). Younes postulated that chemical drive must increase a threshold amount before the pharyngeal dilator muscles can reflexly open the airway (henceforth called effective recruitment threshold; TER). Recurrent obstructive events develop if a ventilatory overshoot occurs at the end of the event and reduces chemical drive below TER, thereby removing the very stimulus required to maintain arousal-free airway patency. According to this proposition, factors that increase the likelihood and magnitude of postevent overshoots promote recurrent obstruction (i.e., OSA). Little is known about the operation of these factors in patients with OSA.

In theory, an excessive overshoot may result from the following: 1) a high TER, since, by definition, chemical drive must increase a large amount before the airway opens reflexly [this would increase the likelihood of arousal occurring before reflex opening, and arousals greatly increase the postevent overshoot (53)]; 2) a low arousal threshold (TA), since this would trigger overshoot-augmenting arousals even if TER were low; or 3) a large, fast respiratory motor response to transient changes in alveolar ventilation (high dynamic controller gain). The obligate circulatory delay between lung and chemoreceptors dictates that chemical drive must continue to increase for a finite period after airway opening. With large and fast chemical responses, the postevent increase in drive would be large, increasing the likelihood of postevent arousal and promoting more complete opening of the airway, two factors that increase the overshoot.

In this study, TER was determined in OSA patients by transiently dialing down continuous positive airway pressure (CPAP) to a level associated with severe flow limitation at resting chemical drive (dial down from air breathing) and examining the effect, on dial-down flow, of increasing chemical drive to different levels before dial down by briefly changing the inspired gas mixture. To determine the relation between TER and arousal threshold, the increase in drive associated with arousal (TA) was also determined in each patient. Finally, to evaluate the potential for postevent increase in drive, the rate of increase in drive during the brief periods of hypercapnia and/or hypoxia was determined along with measurement of lung-to-carotid circulatory delay.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
As per usual practice in Calgary, patients referred to the Sleep Center for evaluation of OSA had overnight monitoring at home using an oxyhemoglobin-saturation based monitor (Remmers Sleep Recorder model 4.2, Saga Tech Electronic, Calgary, Canada) (22, 43). Patients who had a respiratory disturbance index >15 h–1 during the home study, and who did not have any of the exclusion criteria, were invited to participate. Exclusion criteria included significant comorbidities (dialysis-dependent renal failure, congestive heart failure, severe chronic obstructive pulmonary disease, previous stroke) or use of sedatives or antidepressants. The Conjoint Health Research Ethics Board at the University of Calgary approved the study protocol, and all subjects gave written informed consent to participate. Twenty-one patients were studied.

Patients underwent two studies on separate nights, a diagnostic polysomnography study followed by the dial-down study.

Diagnostic Polysomnography

The diagnosis and severity of OSA was confirmed by overnight, attended polysomnography, during which we obtained a three-channel electroencephalogram (EEG), an electro-oculogram, and a submental electromyogram using surface electrodes. Airflow was measured by monitoring nasal pressure through nasal cannulae (Ultima Dual Pressure Sensor 0585, Braebon Medical, Carp, Ontario, Canada). Respiratory effort was measured by inductance plethysmography with transducers placed on the chest and abdomen (Respitrace, Ambulatory Monitoring, Ardsley, NY). Arterial oxyhemoglobin saturation was recorded with a pulse oximeter (Biox 3740, Ohmeda, Boulder, CO). All variables were recorded continuously by a computerized data-acquisition system and stored electronically for later analysis (Sandman, Tyco Healthcare, Kanata, Ontario, Canada). Certified polysomnographic technologists scored sleep, arousals and the presence and type of obstructive events using standard criteria (1, 1a, 37). Apnea-hypopnea index (AHI), average and minimum O2 saturation, and number of respiratory events with arousal were calculated for the supine and lateral positions in non-rapid eye movement (NREM) sleep.

Dial-Down Study

This was done 9.4 ± 13.3 days after diagnostic polysomnography, before institution of CPAP. CPAP was applied via nasal mask connected to a multipurpose ventilator research prototype, described previously (52–54), that allowed reduction of CPAP to 1.0 cmH2O. The variables recorded were the same as in the clinical polysomnography study except for respiratory flow, which was recorded from a pneumotachograph inserted in the hose of the ventilator. Mask pressure was recorded from a side port in the mask. All signals were sampled at 120 Hz and recorded using a Windaq data-acquisition system (DATAQ Instruments, Akron, OH).

Studies were initiated in the supine position except in four patients who could not sleep in this position. Patients were encouraged to remain in the supine position throughout the night, but many unconsciously turned to the lateral position. When this happened, they were left in that position until they awoke spontaneously and were instructed to resume the supine position. Pressure was titrated to the first level associated with no snoring or flow limitation. This level was maintained thereafter (holding pressure). If flow limitation appeared during ventilatory stimulation, holding pressure was increased appropriately.

Preliminary testing.   During stable sleep and breathing room air, CPAP was reduced in steps each lasting three breaths (dial downs). Three to four levels of dial-down pressure were selected that spanned the range between holding pressure and the pressure associated with complete obstruction, or 1 cmH2O, if only a hypopnea was observed at 1 cmH2O. Each level was administered two to four times. The purpose of this preliminary step was to determine the dial-down pressure at which the airway just closes (PCLOSE). Typically, at this dial-down pressure some dial downs resulted in complete obstruction, while others resulted in severe flow limitation.

Tests with altered inspired gas concentrations.   Subsequently, dial downs were done only at PCLOSE. If complete obstruction was not observed at 1 cmH2O, the subsequent dial-down pressure was 1 cmH2O. With some observations, dial down was applied while the patient breathed room air (air dial downs, e.g., Fig. 1A). Alternately, inspired gas was altered for {approx}30 s before dial down (gas dial down, e.g., Fig. 1, B and C). This was done using a pair of proportional solenoid valves, one connected to a nitrogen tank and the other to a tank containing 25% CO2, 21% O2, balance nitrogen. Flow through the solenoid valves was made proportional to flow through the ventilator's hose (obtained from the flowmeter in the CPAP hose) so that the inspired concentrations of CO2 and O2 were fairly independent of airflow. The proportionality dial used by the technologists was calibrated to result, on applying a manual switch, in CO2 concentrations between 0 and 10% and O2 concentrations between 11 and 21%. Five different inspired gas mixtures were typically applied at random before dial downs. These were 3, 6, and 9% CO2 in air; 3% CO2 with either 15% or 11% O2; and 1% CO2 with either 15% or 11% O2. The choice of whether to use 15% or 11% O2 was based on the magnitude of oxyhemoglobin desaturation; reduction in arterial O2 saturation (SaO2) below 80% was avoided. In patients who experienced frequent arousals with the 3% CO2 challenge, the CO2 concentrations used were 2, 4, and 6% instead of 3, 6, and 9%. In two patients the highest CO2 delivered was 3%; at higher levels the patient aroused before dial downs with nearly every test. One patient did not receive hypoxic mixtures because the N2 solenoid was inoperative.


Figure 1
View larger version (83K):
[in this window]
[in a new window]

  		Fig. 1. Tracings obtained from the same patient showing progressive increase in flow during dial downs as a function of ventilation level before dial down (A–C). EEG, electroencephalogram; PAW, airway pressure; VE, minute ventilation in last breath before dial down. Dial down began at the arrows. Note the severe flow limitation during dial down from air (A). Note also that with the higher CO2 challenge (6%; C) flow during the first dial-down breath is similar to the flow on continuous positive-airway pressure (CPAP), indicating no flow limitation despite the much lower CPAP pressure. Note also that dial-down flow in C decreased rapidly within 2 breaths after returning inspired gas to room air (the switch was at the time of dial down; the CO2 signal is delayed relative to flow).

 
A change in gas mixture was maintained for {approx}30 s unless arousal occurred earlier. If sleep continued for 30 s, dial down was applied and maintained for three breaths, unless arousal occurred earlier. Enough time was allowed between tests for stable sleep to resume.

Analyses

Analysis was limited to periods of non-rapid eye movement (NREM) sleep since the large breath-by-breath variability typical of rapid eye movement (REM) sleep precluded confident estimates of drive at any given breath. Except for the four patients who could not sleep in the supine position, analysis was limited to periods spent in the supine position; because of deliberate efforts to maintain the patient in that position, periods in the lateral position were usually brief and did not provide an adequate amount of data.

Leak during holding pressure was estimated from a 30-s moving average of flow. Leak level during dial downs was estimated by selecting a breath during which Respitrace end-expiratory volume did not change and determining the flow offset that would result in equal inspired and expired volumes in the integrated flow signal in that breath. Leak levels were subtracted from the primary flow signal at the appropriate times to obtain patient flow.

Ventilatory measurements during periods on holding pressure.   Tidal volume (VT), respiratory rate (RR), minute ventilation (VE = VT x RR), mean inspiratory flow (VT/inspiratory time), and peak inspiratory flow were determined breath by breath from the corrected flow signal. Baseline ventilatory data are reported as the average of the 10 breaths preceding an air dial down or preceding a change in inspired gas concentration. For tests with gas challenges, breath-by-breath ventilatory data between the onset of gas delivery and dial down, or arousal, whichever occurred first, were tabulated and grouped according to the gas mixture used. Data for the last breath preceding the dial down or, if an arousal occurred first, preceding arousal were noted ("last breath"). The level of chemical drive achieved before dial down or arousal is reported as the difference between values in this last breath and the corresponding baseline values. This difference is expressed in absolute units (e.g., l/min, l/s) and as percentage of baseline.

Calculation of lung-carotid delay.   VE at breaths 2, 3, and 4 following a gas change were compared with VE of the first exposed breath by paired t-test. Time interval between the onset of first inspiration receiving an altered gas mixture and the first breath to show a significant increase in VE was taken as the circulatory delay. End-tidal PCO2 (PETCO2) was measured in the expiratory phase during which CPAP was reduced at the onset of the dial down. This expiratory phase invariably provided a technically adequate expiratory CO2 trace from which to estimate PETCO2. Oxygen saturation at baseline was noted. To determine SaO2 at the chemoreceptor during altered gas tests, the interval between upper airway opening at the termination of an obstructive apnea and the corresponding increase in the oximeter signal was determined. This duration reflects the delay between a change in alveolar PO2 and first indication of this change in the oximeter signal (lung-to-oximeter delay). Carotid-to-oximeter delay was calculated from this lung-to-oximeter delay minus the lung-to-carotid delay, estimated as described above. Oxygen saturation at the chemoreceptor just before dial down, or just before arousal, was estimated from the oximeter signal at a time corresponding to the carotid-to-oximeter delay beyond the time point of interest.

Dial-down measurements.   A senior certified polysomnography technologist (M. Ostrowski) determined whether a cortical arousal occurred during the dial down according to standard criteria (1a). A cortical arousal was identified if there was a visible shift to higher frequencies in any of the three EEG leads. If a cortical arousal was detected, she identified its onset to the nearest 0.1 s. Measurement of inspiratory flow during the dial down (dial-down flow) was limited to periods preceding identified arousals, if any. If no cortical arousal was present during the first two breaths, the higher of the inspiratory flow values in these two breaths is reported. If arousal occurred during breath 2, flow during breath 1 is reported. The absence of cortical or autonomic arousal at the times of measurement of dial-down flow (i.e., values to be reported here) was further confirmed by Fourier analysis and by beat-to-beat changes in heart rate (see below).

Estimation of TER.   The objective of this analysis was to determine whether increasing chemical drive before dial down results in a higher inspiratory flow during the dial down in the absence of arousal and, if so, the level of increase in drive at which dial-down flow begins to increase relative to the value obtained at resting drive (air dial downs). An increase in dial-down flow at the same dial-down pressure without arousal would signify that upper airway dilators can be effectively recruited during sleep when chemical drive increases (see DISCUSSION for the meaning of "effective recruitment" and mechanisms by which chemical drive may result in dilator recruitment).

Some gas challenges resulted in arousal before dial down. These were not used for determining TER. In others, dial down was possible before arousal (Fig. 1). Dial downs during which at least one breath occurred before arousal were used to determine TER as follows. For each eligible dial down, VE during the last breath before dial down (pre-dial-down VE) and dial-down flow were noted. Dial-down flow was plotted against pre-dial-down VE for all dial downs obtained in the patient, including those preceded by air breathing (e.g., Fig. 2, A–C) . Mean (±SD) of dial-down flow for all air dial downs was calculated. Effective recruitment was deemed to have occurred if flow during gas dial downs exceeded mean + 2SD of flow in air dial downs in at least two gas dial downs (e.g., Fig. 2, A–C). Where there were only a few instances in which dial-down flow was significantly increased, TER is reported as the lowest pre-dial-down VE associated with a significant increase in flow (Fig. 2C). Where enough points with significantly increased flow were available, a regression was obtained for pre-dial-down VE vs. dial-down flow over the VE range containing the significant points, and TER is reported as VE at the intersection of the slope and mean dial-down flow in air tests (Fig. 2, A and B).


Figure 2
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 2. A–C: relation between minute ventilation in the last breath before dial down and dial-down flow in 3 patients. Each point represents the result of one dial down, and each panel contains all the data obtained in the same body position throughout the night in 1 patient. {circ}, Dial down while breathing room air; bullet, dial down {approx}30 s after changing inspired gas to different CO2 and O2 mixtures; solid horizontal line, average flow in air dial downs; dotted horizontal line, average + 2SD of flow in air dial downs. Diagonal lines in A and B are regression lines over the ventilation range containing significantly higher flow. In such cases, effective recruitment threshold (TER) is ventilation at the intersection between the regression line and average flow in air dial downs. Note that in C there is a range of ventilation over which flow is not increased above air dial downs. At higher ventilation levels, some dial downs are associated with higher flow, while others are not. D–F: Kaplan-Meier plots of the probability of having no arousal at different levels of ventilatory stimulation in the same 3 patients shown in A–C. TA50, ventilation level at which arousal occurred half the time; TA0, ventilation level that could not be exceeded without arousal, which defines the outer limit of the ventilation range over which TER can be explored. A and D: patient 18. B and E: patient 8. C and F: patient 10.

 
Estimation of arousal threshold (TA).   VE values preceding dial downs or preceding arousal, if arousal occurred, for all tests in each patient were arranged in ascending order. VE values associated with arousal were identified. A Kaplan-Meier survival curve (6) was generated that reflected the probability of having no arousal as a function of VE reached (e.g., Fig. 2, D–F). TA is reported as the VE associated with 50% probability of arousal (TA50, Fig. 2) and also the value that could not be exceeded without arousal (TA0, Fig. 2). TA0 defines the outer limit of the ventilatory range over which TER can be explored.

Confirmation of lack of arousal at the time of measurement of dial-down flow.   An important question is whether observations in which dial-down flow increased significantly with the increase in chemical drive were associated with cortical arousals that were too subtle to be detected by the naked eye. To address this issue, we identified the two observations with the greatest increase in dial-down flow (i.e., upper airway opened during the dial down) in each patient who demonstrated an increase in dial-down flow with chemical stimulation (no. of patients = 12; no. of observations = 24). With each of these observations, we performed Fourier analysis on successive 3-s epochs of a central EEG channel. Analysis was performed in the 3-s epoch centered around the time of upper airway opening during dial down and for the preceding 60 s (20 epochs). This encompassed all the time during which inspired gas composition was altered and 10 epochs (30 s) of baseline. For each observation, we calculated mean ± SD of EEG power in the alpha/sigma range (7.3–14.0 Hz) and in the beta range (14.0–35.0 Hz) in the 10 baseline epochs, and we determined whether the power in these two frequency ranges was significantly higher (i.e., >mean + 2SD of baseline) in the epoch containing upper airway opening. The average power in these two frequency ranges in the 24 epochs containing upper airway opening was also compared with the corresponding values during baseline and during the last epoch before dial down by ANOVA for repeated measures (ANOVA-R). In each of these 12 patients, a similar analysis was performed on the two observations with the greatest increase in pre-dial-down VE that were not associated with a significant increase in dial-down flow.

We also determined whether there was a significant increase in heart rate that may suggest autonomic/subcortical arousal at the time of upper airway opening. In the same 24 observations demonstrating airway opening during gas dial downs, we calculated beat-by-beat heart rate for the beat encompassing upper airway opening and for the preceding 60 s. To filter out sinus arrhythmia, a five-beat moving average of heart rate was calculated up to, but not including, the beat encompassing upper airway opening. Heart rate in the latter beat was compared with average heart rate at baseline and just before dial down by ANOVA-R. A similar analysis was performed in the 24 corresponding observations with the greatest increase in pre-dial-down VE that were not associated with a significant increase in dial-down flow.

In patients who received 2, 4, and 6% CO2, instead of 3, 6, and 9%, the values obtained with 2 and 4% were averaged and are reported as results with 3% CO2. Values obtained under similar conditions in a given patient were averaged (e.g., VE during administration of 6% CO2) so that each patient is represented only once in any group data. Group data are reported as means ± SD, unless otherwise indicated.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Table 1 shows polysomnography data. All patients had severe OSA in the body position in which dial-down results are reported, with the AHI ranging from 41 to 123 min–1.


View this table:
[in this window]
[in a new window]

  		Table 1. Patient characteristics and clinical polysomnography data

 
Apnea or severe flow limitation, recognized by the characteristic flat or gently down-sloping flow contour (10), was evident in all patients during air dial downs (e.g., Fig. 1A). In two patients, severe hypopneas, but no apneas, were observed at 1 cmH2O. In these two patients, the estimated PCLOSE (by back extrapolation) was 0 and –2 cmH2O. In all others, PCLOSE, and hence dial-down pressure used, was ≥1 cmH2O (3.3 ± 2.3 cmH2O). There were 9.1 ± 3.4 air dial downs per patient. Flow during air dial downs was 0.058 ± 0.052 l/s, representing 13.9 ± 11.8% of peak flow on CPAP.

Seventy-four percent of all observations occurred in stage 2 sleep. Slow-wave sleep occurred sporadically in several patients. Fraction of observations in slow-wave sleep ranged from 0 (7 patients) to 82% (26 ± 26%). The inconsistency of presence and percentage of observations in slow-wave sleep within patients precluded a systematic analysis of effect of sleep stage (i.e., stage 2 vs. slow-wave sleep) on the results. Accordingly, the data from all NREM observations were pooled.

Ventilatory Responses to Chemical Stimulation Before Dial Down

Table 2 shows baseline ventilatory data. The first significant increase in VE following a gas challenge occurred during the next breath (breath 2) in four patients, during breath 3, in 14 patients, and during breath 4 in 3 patients. Estimated lung-to-carotid delay was 9.9 ± 2.4 s.


View this table:
[in this window]
[in a new window]

  		Table 2. Baseline ventilatory data on therapeutic CPAP

 
Respiratory rate of the "last breath" was 14.8 ± 2.5 min–1, which was not different from baseline (14.7 ± 2.3 min–1). Changes in VE, VT, and mean inspiratory flow were accordingly similar in proportion. Only changes in VE are reported.

Table 3 shows average breath-by-breath changes in VE following the change in inspired gas, and the corresponding changes in PETCO2 and O2 saturation during the last breath. The five mixtures resulted in a wide range of stimulation before dial down (last breath, Table 3). Oxyhemoglobin desaturation associated with the hypoxic mixtures was modest ({approx}4%). During hypoxia in 1% CO2, the ventilatory response was essentially isocapnic ({Delta}PETCO2 = –0.7 ± 1.3; P < 0.05). Hypoxia in 3% CO2 resulted in mild hypercapnia ({Delta}PETCO2 = 3.3 ± 2.2; P < 0.0001). The hypercapnic mixtures were associated with progressively higher PETCO2 with little changes in O2 saturation. With all five gas mixtures, the ventilatory response began in earnest by breath 3 and progressed rapidly through breath 5 (Table 3). With pure hypoxia (hypoxia in 1% CO2), the ventilatory response had leveled off before dial down or arousal (penultimate {Delta}VE, Table 3). Where hypercapnia was produced, VE continued to increase beyond the 5th breath, albeit at a slower rate.


View this table:
[in this window]
[in a new window]

  		Table 3. Ventilatory responses to changes in inspired gas composition

 
The last data set in Table 3 is an estimate of what the ventilatory response might be if the pulmonary gas changes were similar to those expected during an obstructive apnea or moderate hypopnea in which O2 saturation decreased by {approx}4% and VT decreases below the anatomic dead space [{approx}1 ml/pound ideal body wt (9)]. Under these conditions, there is essentially no alveolar ventilation, and the situation is analogous to the patient rebreathing his or her own expired gas. It was derived by adding the response to 6% CO2, which was essentially isoxic (Table 3), and the response to hypoxia in 1% CO2, which was essentially isocapnic, and multiplying the product by 1.3 to account for the multiplicative relation between hypoxic and hypercapnic responses (see APPENDIX for justification). The estimated increase in respiratory drive during an obstructive apnea or severe hypopnea was, on average, very large, reaching a {Delta}VE of 187% baseline by breath 5 (Table 3, hypoxia in 6% CO2). The response was, however, highly variable among patients; the 5th breath response [{Delta}VE5 (%)] ranged from 56 to 477% (Table 4).


View this table:
[in this window]
[in a new window]

  		Table 4. Individual patient characteristics and results of chemical responsiveness dVE5, effective recruitment threshold (TER) and arousal threshold (TA50)

 
Arousal in Response to Increased Respiratory Drive (TA50 and TA0)

The frequency of cortical arousals before dial down (i.e., within 30 s of altering gas mixture) was 10.5%, 23.4%, and 59.4% for the 3%, 6%, and 9% CO2 mixtures, respectively. It was 7.5% and 20.3% for the hypoxia with 1% CO2 and hypoxia with 3% CO2, respectively. In two patients, arousal occurred infrequently, before or after dial down, even with the highest level of stimulation ({Delta}VE 174% and 268% baseline). In these patients TA could not be determined. In the other 19 patients, TA50, determined taking into account arousals before and during dial down, ranged from 40% to 282% increase in VE over baseline (112 ± 58%, Table 4). TA0 ranged from 90% to 348% baseline (168 ± 72%).

Response of Dial-Down Flow to Increased Respiratory Drive in the Absence of Arousals

The range of chemical stimulation over which dial-down flow could be assessed was limited by the highest VE that could be obtained without arousal (TA0). There were 17.0 ± 9.9 gas challenges per patient (Table 4). Not all these challenges were followed by dial downs because of the frequent occurrence of arousal before 30 s had elapsed. The number of dial downs that were preceded by some degree of chemical stimulation was 12.1 ± 8.1 (range 1–32, Table 4). Figure 1 shows tracings from a patient who demonstrated a clear increase in dial-down flow in response to stimulation (patient 8, Table 4). Figure 2B shows the relation between last-breath VE and dial-down flow in this patient. In 12 patients a significant increase in dial-down flow was observed in at least two observations following chemical stimulation (Table 4). The pattern of response varied among patients. In four patients significant responses occurred over a wide range of pre-dial-down VE, and the response was orderly such that a significant correlation was obtained between last-breath VE and dial-down flow (e.g., Fig. 2, A and B). In these four patients, TER ranged from 0 to 2.2 l/min (0 to 31%) above baseline VE. In the other eight patients the relation was not orderly. Typically, there was a range of last-breath VE over which dial-down flow was consistently not different from air dial downs, and a higher range in which some dial downs were associated with a significant increase in flow, while in others flow was not different from baseline (e.g., Fig. 2C). In these patients TER ranged from 2.2 to 8.0 l/min above baseline (5.0 ± 2.2 l/min), representing 33 to 109% of baseline VE (74 ± 28% baseline).

In nine patients there was no increase in dial-down flow over the entire range of pre-dial-down VE (Table 4). An example is shown in Fig. 3. Although in these patients the precise value of TER could not be determined, it was clearly higher than the highest chemical drive that can be tolerated without arousal. Six of these nine had no response despite >100% increase in VE (Fig. 4A, Table 4). In three patients arousal threshold (TA0) was quite low, severely limiting the range of chemical drive that could be tested, and making it impossible to determine whether failure of response was due to a high TER (Fig. 4B).


Figure 3
View larger version (69K):
[in this window]
[in a new window]

  		Fig. 3. Tracings in a patient showing no increase in dial-down flow despite a substantial increase in ventilation before dial-down. VE, minute ventilation in last breath before dial down. A: air breathing. Note the severe flow limitation during dial down. B: dial down following a period of hypercapnia.

 

Figure 4
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 4. Plots similar to Fig. 2 in 2 additional patients. A: patient shows no increase in dial-down flow in response to substantial increases in pre-dial-down ventilation made possible by a very high arousal threshold. In this patient (patient 16), both TER and TA were very high and indeterminate. B: a patient with very low arousal threshold (patient 11), making it impossible to determine TER.

 
Relation Between TER, TA, and Chemical Responsiveness

Table 4 shows the values of chemical responsiveness [%increase in ventilation at breath 5 (dVE5; %)], TER, and TA in individual patients, along with the corresponding PCLOSE. Figure 5 summarizes the relation between TER and TA50 in the 21 patients. With two exceptions, all patients demonstrated a TER > 0, indicating the need for a threshold increase in drive before effective reflex dilatation of the airway can take place. In at least 8 of 21 patients (38%), TER was >100%. In patients where both thresholds were available (solid circles, n = 12) there was no correlation between TER and TA50 (r2 = 0.03, P > 0.5).


Figure 5
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 5. Relation between TER and TA in 21 patients. Data with horizontal arrows are those in which TER was higher than the highest chemical drive that could be tolerated without arousal (e.g., the 2 patients in Fig. 4). Data with vertical arrows are those in which TA was indeterminate but greater than the indicated value (e.g., patient in Fig. 4A). Note no correlation between TER and TA when both were known (solid circles).

 
Figure 6 shows the relation between chemical responsiveness [dVE5 (%)] and arousal threshold (TA50) in the 20 patients in whom asphyxic responses could be estimated. As indicated earlier, TA could not be determined in two patients (vertical arrows), while in two patients the maximum CO2 received was 3% (horizontal arrows). There was a weak but significant correlation between the two variables in the remaining 16 patients (r2 = 0.44, P < 0.005).


Figure 6
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 6. Relation between arousal threshold (ordinate) and chemical responsiveness, expressed as the estimated percent increase in respiratory drive during the 5th breath following the onset of rebreathing, in combination with mild desaturation. See text. Data with vertical arrows are those in which TA was indeterminate but greater than the indicated value (e.g., the patient in Fig. 4A). Data with horizontal arrows are from 2 patients in whom there were no data with 6% CO2. The abscissa values in these 2 patients represent the actual response observed with 3% CO2 in 15% N2 and, hence, underestimate the response with 6% CO2 in hypoxia. r2 is for regression between the 2 variables where both could be determined (bullet, n = 16).

 
There was no correlation between chemical responsiveness and TER in 10 patients in whom both variables were available (r2 = 0.05, P > 0.5). There was also no correlation between the severity of mechanical abnormality (PCLOSE) and chemical responsiveness (r2 = 0.03, P < 0.5), TER (r2 = 0.02, P < 0.7), or TA50 (r2 = 0.09, P < 0.3).

Evidence for Lack of Arousal at Upper Airway Opening

In agreement with visual assessment of lack of cortical arousal, neither alpha/sigma nor beta EEG power showed a significant increase at the time of upper airway opening in any of the 24 analyzed observations (i.e., those with the greatest increase in dial-down flow in the 12 responding patients). Likewise, the average alpha/sigma and beta powers at opening in the 24 observations were not increased relative to baseline or to the power just before dial down (Table 5). Heart rate increased slightly but significantly with the increase in chemical drive (just before dial down vs. baseline, Table 5), but there was no further increase at the time of upper airway opening. There was also no difference in the magnitude of increase in heart rate between observations with and without upper airway opening (Table 5). Peak inspiratory flow during dial downs associated with upper airway opening was significantly higher than at baseline but lower than just before dial down, indicating that opening was not complete.


View this table:
[in this window]
[in a new window]

  		Table 5. High-frequency EEG power and heart rate at different times during chemical stimulation

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
We have described an approach to evaluating the various mechanisms responsible for breathing instability in OSA patients. The results confirm that in the majority of patients, even in those with severe OSA, modest increases in chemical drive can result in upper airway opening without arousal and suggest that in these patients instability develops because 1) arousals also occur, on average, with only modest increases in respiratory drive, and 2) the dynamic response of chemoreceptors to the pulmonary gas changes associated with obstructive events is, on average, sufficiently brisk to result in considerable postevent increase in respiratory drive even in the face of a normal circulatory delay. This study also suggests that the mechanisms responsible for instability may vary considerably among patients (Table 4, Figs. 5 and 6).

Critique/Limitations of Methods

  1. Several studies have documented the progressive increase in inspiratory pressure during naturally occurring obstructive apneas and reported the pressure level at which the airway opens (3, 4, 25, 30, 38, 40, 46, 49, 56). Since arousal is usually present when the airway opens, and given the strong belief that arousal is responsible for opening, the threshold increase in drive required for airway opening became synonymous with arousal threshold (3, 4, 25, 30, 38, 40, 46, 49, 56).
    Increases in inspiratory effort in response to mechanical loading during sleep reflect changes in blood gas tensions (51). As judged by the rate of increase in inspiratory pressure, chemical drive usually changes rapidly in the course of obstructive events (3, 4, 25, 30, 38, 40, 46, 49, 56), particularly just before opening (2a). Furthermore, the circulatory delay inherent in these chemical responses dictates that respiratory drive will continue to increase for a finite period beyond airway opening. The rapid increase in drive during natural obstructive events, and the inherent inertia of chemical responses, would make it difficult to identify a separate threshold increase in drive at which the airway may open without arousal (TER) since both TER and arousal threshold may be crossed almost simultaneously (53). A special methodological feature of the present study made it possible to identify the two thresholds separately. Here, we imposed a range of increases in chemical drive, but at each level the increase in drive was arrested, or nearly so, before the obstruction was induced ("penultimate {Delta}VE", Table 3). With this arrangement, if TER is below arousal threshold a range of pre-dial-down chemical drive would be found where the airway would remain open during the dial down, at least for the period of circulatory delay, despite lack of arousal.
  2. We studied the effect of chemical drive on dial-down flow when airway pressure was near PCLOSE. Under these conditions any increase in net dilating force should increase dial-down flow. This approach increases the sensitivity of detecting a change in net airway distending pressure. However, TER, so determined, may underestimate the increase in drive required to keep the airway open without CPAP if PCLOSE is >0. Under these conditions, drive must increase by TER plus an additional amount to overcome PCLOSE. We feel this is an acceptable limitation since once TER is reached the relation between chemical drive and extra distending force appears to be steep (see Relation Between Chemical Drive and Maximum Flow Beyond TER) so that the error may not be so large. Nonetheless, it is important to appreciate that when PCLOSE is >0, TER underestimates the increase in drive required to open the airway in the absence of CPAP.
  3. Changes in drive were superimposed on optimal CPAP. Thus the increase in respiratory drive, as well as TER and TA, could be defined by reference to the drive the individual patient has during sleep with a normal mechanical load (thereafter called eupneic drive). For example, one may say that net dilatation, or arousal, occurred when drive increased by 60% above eupnea. Defining chemical responses, TER and TA in the same units and by reference to eupneic drive, is of considerable value (see Interrelation of Chemical Responsiveness, TER, and TA).
  4. The present results were obtained from patients before initiation of CPAP treatment. It is, therefore, not possible to tell whether the destabilizing influences identified here were premorbid or a consequence of long-standing OSA. A reevaluation of the responses examined here after CPAP treatment would be of considerable interest.
  5. Only one value per patient is given for TER, TA, and ventilatory responses (Table 4). It is clearly possible that these responses vary from night to night, or even at different times during the same night. The study was not designed to address the consistency of these responses in the same patient, and efforts to address consistency within the same night were not successful; most data reported here were obtained from a 1- to 2-h period during which body position was stable in NREM sleep and it was possible to test four to five inspired gas mixtures on at least one occasion, but usually more (Table 4). At other times, the patient may have been in a different position or spent a long time awake between observations so that we could not obtain enough data to compare results at different times of the night. Additional studies are needed to address this important issue as it may explain within- or between-night variability in the severity of OSA.

Effective Recruitment Threshold (TER)

Magnitude.   TER ranged widely from zero to >175% (Fig. 5). In 12/21 patients (57%), and possibly 16/21 patients (76%), TER was less than 110% of eupneic drive (Fig. 5). Although this percent increase may appear large, it is in fact very small when measured against the capacity of the respiratory system to increase its output [e.g., >1,000% during moderate exercise (47)], or when one considers how little changes in blood gas tensions are typically required to increase drive by this amount. Thus, given the normal steady-state response to CO2 [{approx}2 l·min–1·mmHg–1 (36)], resting ventilation can be doubled through a mere 3-mmHg increase in arterial PCO2 (PaCO2). It is, therefore, possible that in many, but certainly not all, patients the airway may remain open if respiratory drive were allowed to increase modestly, and remain increased. This finding further emphasizes the important role an excessive postapneic ventilatory overshoot, which reduces drive to eupneic levels or even lower, plays in perpetuating the obstructive apnea in many patients (53).

Mechanism of TER.   Why is it necessary for drive to increase a finite amount (TER) before a net dilating effect is manifest, and why does the threshold differ so much among patients? Two broad mechanisms may be entertained.

  1. A finite increase in respiratory drive is required before dilator activity can begin increasing. The primary stimuli for activating pharyngeal dilators during sleep are chemical drive and negative pharyngeal pressure (15). Development of an obstructive event at eupneic levels of chemical drive (e.g., at sleep onset or at the onset of air dial downs here) indicates that the negative pharyngeal pressure associated with eupneic respiratory drive is insufficient to maintain airway patency. Since inspiratory efforts are the source of negative pharyngeal pressure, and the only means by which these efforts can be increased during sleep is an increase in chemical drive (51), chemical drive must increase if upper airway patency is to be restored without arousal, regardless of whether this is accomplished directly via the increase in drive per se or indirectly via a more negative pharyngeal pressure. When breathing is stimulated by CO2 in sleeping animals (13, 16, 55) and sleeping healthy humans (26, 34), diaphragm activity increases continuously with CO2 whereas dilator activity increases only above a threshold level of stimulation. Activity of pharyngeal dilators increases progressively during obstructive events in OSA patients (5, 7, 28, 29, 31, 38), thereby reflecting their responsiveness to increasing chemical drive, acting directly or indirectly via the negative pressure reflex. However, the increase in chemical drive, relative to eupnea, required for dilator activation in OSA patients is not known. It may be that differences exist among patients in the magnitude of increase in chemical stimulation, or in the magnitude of negative pharyngeal pressure, required for recruitment of dilators to be initiated.
  2. A finite increase in dilator activity needs to be reached before a net dilating influence can be generated. In the passive pharynx, reduction in pharyngeal pressure below values that result in flow limitation is commonly associated with a paradoxical reduction in maximum flow (negative effort dependence) (21, 42). Thus it is possible that the more negative pharyngeal pressure during stimulated breathing may, acting via negative effort dependence, offset the distending action of the activated dilators, resulting in no net dilating force until activity exceeds a threshold value. This threshold may vary among patients.

Recently, Patil et al. (32) compared OSA patients and normal subjects with respect to their ability to decrease critical closing pressure (PCRIT) during sustained reductions in airway pressure. OSA patients failed to decrease PCRIT whereas normal subjects did. This suggests that TER is higher in OSA patients than in normal subjects. Interestingly, the response of the genioglossus was not different, suggesting that the second mechanism mentioned above may be responsible for differences in TER. Further studies are needed to identify the reason(s) for TER variability.

It is not clear whether TER is a fixed property in individual patients. Given the different neural and mechanical mechanisms that may account for TER, it is possible that this threshold may vary with body or head and neck position, or as a result of changes in so-far-unidentified control mechanisms that affect dilator muscles recruitment. Additional studies are needed to address this important issue.

Relation Between Chemical Drive and Maximum Flow Beyond TER

In 4 of 12 patients in whom TER could be determined, maximum flow (FMAX) increased in a graded and orderly way as drive increased further (Fig. 2, A and B). In the remaining eight patients, a wide range of FMAX was observed over a narrow range of respiratory drive (Fig. 2C). This may reflect a very high slope between FMAX and drive beyond TER. Thus the level of drive during FMAX measurement (i.e., during dial down) was inferred from the last breath, which occurred 1–2 breaths earlier. Given the breath-by-breath variability in VE (Table 2), it may be expected that actual drive during dial down was different from the inferred value by up to 2–3 l/min. Such random measurement errors may still be consistent with an apparent orderly response when the slope is relatively low but would result in a wide range of responses over a small range of inferred drive levels if the slope were high. Alternatively, this type of relation (i.e., as in Fig. 2C) may result if TER were to vary from time to time in association with changes in head and neck position or in other unmonitored variables.

Arousal Threshold (TA)

Arousal threshold in OSA patients is typically reported as the inspiratory pressure generated just before arousal at the end of obstructive events (DPMAX) (2a). Reported DPMAX values vary considerably between patients (3, 4, 25, 30, 40, 46, 49, 56). As well, there is considerable within-patient variability in DPMAX (2), reflecting differences in depth of sleep during the night. The average DPMAX reported in a given patient is, therefore, the level associated with arousal nearly half the time and is comparable to the TA50 reported here. As with DPMAX, TA50 varied widely among patients (40 to >268% increase in drive over baseline) and displayed within-patient variability [width of the probability line (Fig. 2)]. The present study is, however, the first to report TA as the increase in respiratory output above eupnea. Even though arousal threshold has been reported to be higher in OSA patients than in normal subjects (2a), when TA is expressed as the percent increase in eupneic ventilatory drive that results in arousal, it becomes clear that most OSA patients arouse in the face of very modest, safe changes in blood gas tensions. Thus, in nearly half the patients (10/21), TA50 was <100%, and in 15/21 (71%) of patients, it was <150% (Fig. 5). As indicated earlier, such increases in drive result from very small changes in gas tensions. It therefore appears that in most OSA patients arousal mechanisms operate with an unnecessarily large safety margin.

Chemical Responses

To our knowledge, this is the first study in which ventilatory responses to CO2 and hypoxia were assessed during sleep with normalized upper airway resistance. It is also the first in which breath-by-breath changes in drive were quantified in the course of inspired gas changes that last several breaths, thereby mimicking the pattern of pulmonary gas tensions during obstructive events. The selection of gas mixtures also allowed us to estimate how much respiratory drive would increase breath-by-breath in each patient if alveolar ventilation were transiently eliminated, along with a modest degree of O2 saturation, as would happen during an apnea or moderate hypopnea (Table 3). The results showed marked interindividual differences (SDs, Table 3). This is not surprising since normal ventilatory responses to CO2 and hypoxia are highly variable when measured by standard techniques (36). However, on average, chemical responses were surprisingly fast and large. Thus, with the changes in blood gas tensions expected during an apnea or moderate hypopnea, respiratory output would reach an average 287% of the eupneic level by breath 5 (hypoxia in 6% CO2, Table 3). Because of circulatory delays ({approx}10 s), the increase in drive in breath 5 was in response to changes in pulmonary gas tensions by early breath 3 and is, therefore, unavoidable even if the airway opened at breath 3, the earliest it could do so during air breathing. Accordingly, these findings indicate that, given such chemical responses, a large ventilatory overshoot is unavoidable following even a three-breath event unless the patient's chemical response is well below average, or airway resistance remains very high after opening to substantially damp the ventilatory response.

Our patients had severe OSA. Whether the average response observed here is excessive relative to normal subjects or to patients with milder OSA is not known. The mild, altitude-related hypoxemia in Calgary (Table 2) may have contributed. Other lines of evidence, however, suggest that dynamic ventilatory responses may be higher in OSA patients than in normal subjects. Loop gain during sleep is higher in OSA patients than in normal subjects (48, 54). Likewise, Hudgel et al. (18) found that ventilatory responses to single breath CO2 challenges were higher in awake OSA patients than in normal subjects. Furthermore, in the current patients, PETCO2 increased an average 13.4 mmHg during inhalation of 6% CO2 (Table 3). Inhalation of 6% CO2 is comparable to rebreathing in that PETCO2 cannot rise above mixed venous PCO2 before recirculation. When measured by the rebreathing technique, mixed venous PCO2 is normally only 10 mmHg higher than PETCO2 (35), thereby suggesting that mixed venous-to-alveolar CO2 gradient was above normal in our patients, perhaps reflecting the higher metabolic rate due to obesity. A higher gradient increases the dynamic ventilatory response to CO2 (23, 50).

Interrelation of Chemical Responsiveness, TER, and TA

By expressing chemical responsiveness, TER, and TA in the same units ({Delta}VE, relative to eupnea), it is possible to examine the interaction between these three variables in generating OSA. Figure 7 shows the estimated average change in respiratory drive associated with {approx}4% desaturation along with an increase in PETCO2 to mixed venous level, such as would occur during a brief obstructive event (from Table 3). Average arousal threshold (TA50) is indicated. The solid circle illustrates a situation where arousal threshold at the moment is about average. In one case, TER is lower than TA, while in the other it is higher. This display helps explain the dominant features of OSA and illustrates how instability is determined by the interaction of several unrelated variables as opposed to being a function of a single abnormality.

  1. Given the average results illustrated in Fig. 7, and the impossibility of altering the time course of chemical drive until two to three breaths after opening, arousal is difficult to avoid even when the upper airway opens reflexly (i.e., TER < TA). Thus, to avoid arousal, TA must not only be higher than TER, but the difference must also exceed the obligatory increase in drive over the period of circulatory delay. According to the data of Fig. 7, this difference must be large. For example, if TER were 80% (e.g., TER2, Fig. 7), TA must be >180% (i.e., drive at breath 5), and if TER were 30%, TA must be >130% (drive at breath 4). Of course, the difference between TA and TER need not be so large in a patient in whom the rate of rise in drive is below average. However, this would require a combination of TER, TA, and chemical responsiveness that is rare in this type of patient (Figs. 5 and 6 and Table 4). This analysis explains why arousal occurs so frequently at the end of obstructive events. Furthermore, it shows why occurrence of arousal need not signify that arousal was responsible for airway opening.
  2. The above analysis also explains why some individuals, whose upper airway is susceptible to closure during sleep, do not develop OSA, or may only have it intermittently (52). Thus stable breathing may occur if 1) chemical response is slow enough to reduce the difference between TA and TER required to avoid arousals, and/or 2) the difference between TA and TER is large, because TER is low, TA is high, or both. Under these conditions arousal following an obstructive event may be avoided. It has been well documented that arousals further increase respiratory output and aggravate the postapneic overshoot (17, 24, 53). Therefore, avoidance of arousal promotes a stable response with the result that the patient may develop an occasional obstructive event as opposed to repetitive OSA. Because arousal threshold varies considerably across the night (2), the difference between TA and TER is also likely to vary from time to time, and this may account for periods of stable breathing, particularly during delta sleep, in patients who have intermittent OSA (52, 53). Further studies are needed to determine the differences in these indexes of ventilatory stability between the patients who have repetitive OSA throughout sleep and those with OSA that occurs intermittently or not at all despite a susceptible upper airway.
  3. The fast rate of increase in chemical drive coupled with the inevitable circulatory delay provide an explanation for the persistence of the characteristic discontinuous flow behavior at the end of obstructive events even in the absence of arousal (53). Thus, assume that TER is low, at 50%. The upper airway starts to open by breath 3 (see Fig. 7). However, chemical drive continues to rise rapidly through breath 4 and possibly breath 5, reaching 130% or 180% above baseline (230% or 280% of baseline) before it begins to decline. This large additional increase in chemical drive should result in a rapid and more complete opening of the airway, particularly when the relation between FMAX and chemical drive is steep beyond TER (Fig. 2).


Figure 7
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 7. Estimated average time course of respiratory drive following onset of an obstructive apnea or moderate hypopnea. TA50, average drive level at which arousal occurs half the time. Note that, on average, chemical drive at breath 5 is well above average TA50. Circles, a specific example used for illustration (see text) in which TA (solid circle) is near average while effective recruitment threshold is above (TER1) or below (TER2) arousal threshold.

 
The frequent occurrence of arousal along with a sudden increase in flow at the end of obstructive events have been the cornerstones of the thesis that arousal is responsible for airway opening (38). The present findings and analysis indicate that both phenomena are explainable, and in fact almost unavoidable, given the intensity of chemical responses, the prevailing arousal thresholds, and the relation between FMAX and respiratory drive in the absence of arousal in patients with OSA. Accordingly, these phenomena are no longer compelling reasons for advocating an essential role for arousal in terminating obstructive events.

Clinical Implications

Each of the three "control of breathing" variables studied here, chemical responsiveness, TA, and TER, may be amenable to modification through appropriate, but different, nonmechanical interventions. For example, TER may be effectively reduced or eliminated by breathing CO2-enriched gas, oxygen breathing may substantially dampen brisk chemical responses, and sedatives can increase arousal threshold. However, unless these therapies are targeted to the appropriate patients, they may not be effective or may exacerbate the problem. For example, where TER and/or PCLOSE is/are very high, it may not be possible to raise baseline respiratory drive by the required amount while maintaining sleep, and only mechanical interventions may succeed in such patients. Likewise, administering oxygen to someone in whom chemical responsiveness is not excessive may simply prolong the obstructive events, and use of sedatives where arousal threshold is already high would aggravate hypoxemia.

The values obtained for each of the three measured variables varied widely among patients (Table 4), and there was little or no correlation between them (Figs. 5 and 6). Although the extent to which the values observed on a given night/body position (as was done here) reflect fixed, or reasonably fixed, properties of individual patients remains to be determined, these findings suggest that the mechanism(s) of instability may vary considerably from patient to patient. This may explain why the use of O2 (20, 44, 45), CO2 (19, 20, 27), and sedatives (8, 14, 39) in previous studies on unselected patients produced inconsistent results. Accordingly, a reevaluation of these therapies, when used selectively, may be indicated.

The range of TER values required for reflex airway opening found here (Fig. 5) also explains why the use of pharmacological stimulants, such as acetazolamide, medroxyprogesterone, theophylline, and nicotine, proved ineffective (20, 44, 45). Thus the increase in drive produced by these agents (5- to 10-mmHg reduction in PaCO2, corresponding to 10–20% increase in drive) is trivial relative to what is required (Fig. 5).


   APPENDIX
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 
Estimation of Changes in Respiratory Drive During an Obstructive Apnea or Moderate Hypopnea

With the 6% CO2 mixture, inspired PCO2 ({approx}42 mmHg) was slightly above baseline PETCO2 (37.6 ± 4.4 mmHg, Table 2), thereby approximating the rebreathing situation during obstructive events in which tidal volume decreases below the anatomic dead space. Following a change to 6% CO2, PETCO2 rose rapidly to a relatively stable level (e.g., Fig. 1C), representing the mixed venous value. Thus the response to 6% CO2 reflects what would happen during an obstructive event if there were no associated hypoxemia. On the other hand, since PETCO2 changed minimally while breathing hypoxic gas in 1% CO2 (Table 3), the response to this gas mixture reflects what happens with a 4% desaturation with no associated hypercapnia. A simple summation of the two responses would underestimate what would actually happen with combined hypoxemia and hypercapnia in view of the well-established multiplicative relation between the two stimuli (11). An amplification factor of 1.3 was applied. This amplification is in fact modest. For example, the ventilatory response to CO2 in the presence of 4% desaturation [({Delta}VE at breath 5 while breathing 3% CO2 in hypoxia minus {Delta}VE at breath 5 while breathing 3% CO2 in air)/difference in PETCO2 between the two hypoxic mixtures at breath 5; (6.9 – 3.3)/4.0, or 0.90 l·min–1·mmHg–1, Table 3] was twice as much as the normoxic response to CO2 ({Delta}VE/{Delta}PETCO2 at breath 5 with the 6% CO2 mixture; 6.1/13.4, or 0.45 l·min–1·mmHg–1).


   GRANTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
ACKNOWLEDGMENTS
 REFERENCES
 
This research was supported by the Canadian Institutes of Health Research.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
REFERENCES
 
Present address of M. Younes: 405-435 Ellice Avenue, Winnipeg, Manitoba, Canada R3B 1Y6.


   FOOTNOTES
 

Address for reprint requests and other correspondence: M. Younes, 405-435 Ellice Ave., Winnipeg, Manitoba, Canada R3B 1Y6 (e-mail: mkyounes{at}shaw.ca)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX
 GRANTS
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. American Academy of Sleep Medicine. 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 22: 667–689, 1999.[Web of Science][Medline]
  2. American Sleep Disorders Association. EEG arousals: scoring rules and examples. A preliminary report from the Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. Sleep 15: 174–184, 1992.[Web of Science]
  3. Berry RB, Asyali MA, McNellis MI, Khoo MC. Within-night variation in respiratory effort preceding apnea termination and EEG delta power in sleep apnea. J Appl Physiol 85: 1434–1441, 1998.[Abstract/Free Full Text]
  4. Berry RB, Gleeson K. Respiratory arousal from sleep: mechanisms and significance. Sleep 20: 654–675, 1997.[Web of Science][Medline]
  5. Berry RB, Kouchi KG, Bower JL, Light RW. Effect of upper airway anesthesia on obstructive sleep apnea. Am J Respir Crit Care Med 151: 1857–1861, 1995.[Abstract]
  6. Berry RB, Kouchi K, Bower J, Prosise G, Light RW. Triazolam in patients with obstructive sleep apnea. Am J Respir Crit Care Med 151: 450–454, 1995.[Abstract]
  7. Berry RB, McNellis MI, Kouchi K, Light RW. Upper airway anesthesia reduces phasic genioglossus activity during sleep apnea. Am J Respir Crit Care Med 156: 127–132, 1997.[Abstract/Free Full Text]
  8. Bland JM, Altman DG. Survival probabilities (the Kaplan-Meier method). BMJ 317: 1572, 1998.[Free Full Text]
  9. Carlson DM, Onal E, Carley DW, Lopata M, Basner RC. Palatal muscle electromyogram activity in obstructive sleep apnea. Am J Respir Crit Care Med 152: 1022–1027, 1995.[Abstract]
  10. Cirignotta F, Mondini S, Zucconi M, Gerardi R, Farolfi A, Lugaresi E. Zolpidem-polysomnographic study of the effect of a new hypnotic drug in sleep apnea syndrome. Pharmacol Biochem Behav 29: 807–809, 1988.[CrossRef][Web of Science][Medline]
  11. Comroe JH. Physiology of Respiration. Chicago, IL: Year Book Medical, 1974.
  12. 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 150: 475–480, 1994.[Abstract]
  13. Cunningham DJC, Robbins PA, Wolff CB. Integration of respiratory responses to changes in alveolar partial pressures of CO2 and O2 and in arterial pH. In: Handbook of Physiology. The Respiratory System. Control of Breathing. Bethesda, MD: Am. Physiol Soc, 1986, sect. 3, vol. II, pt. 2, chapt. 15, p. 475–528.
  14. Haxhiu MA, van Lunteren E, Mitra J, Cherniack NS. Comparison of the response of diaphragm and upper airway dilating muscle activity in sleeping cats. Respir Physiol 70: 183–193, 1987.[Web of Science][Medline]
  15. Hedner J, Grunstein R, Eriksson B, Ejnell H. A double-blind, randomized trial of sabeluzole—a putative glutamate antagonist-in obstructive sleep apnea. Sleep 19: 287–289, 1996.[Web of Science][Medline]
  16. Horner RL. Motor control of the pharyngeal musculature and implications for the pathogenesis of obstructive sleep apnea. Sleep 1996: 827–53, 1996.
  17. Horner RL, Liu X, Gill H, Nolan P, Liu H, Sood S. Effects of sleep-wake state on the genioglossus vs. diaphragm muscle responses to CO2 in rats. J Appl Physiol 92: 878–887, 2002.[Abstract/Free Full Text]
  18. Horner RL, Rivera MP, Kozar LF, Phillipson EA. The ventilatory response to arousal from sleep is not fully explained by differences in CO2 levels between sleep and wakefulness. J Physiol 534: 881–890, 2001.[Abstract/Free Full Text]
  19. Hudgel DW, Gordon EA, Thanakitcharu S, Bruce EN. Instability of ventilatory control in patients with obstructive sleep apnea. Am J Respir Crit Care Med 158: 1142–1149, 1998.[Abstract/Free Full Text]
  20. Hudgel DW, Hendricks C, Dadley A. Alteration in obstructive apnea pattern induced by changes in oxygen- and carbon-dioxide-inspired concentrations. Am Rev Respir Dis 138: 16–19, 1988.[Web of Science][Medline]
  21. Hudgel DW, Thanakitcharu S. Pharmacologic treatment of sleep-disordered breathing. Am J Respir Crit Care Med 158: 691–699, 1998.[Abstract/Free Full Text]
  22. Isono S, Feroah TR, Hajduk EA, Brant R, Whitelaw WA, Remmers JE. Interaction of cross-sectional area, driving pressure, and airflow of passive velopharynx. J Appl Physiol 83: 851–859, 1997.[Abstract/Free Full Text]
  23. Issa F, Morrison GD, Hadjuk E, Iyer A, Feroah T, Remmers JE. Digital monitoring of sleep-disordered breathing using snoring sound and arterial oxygen saturation. Am Rev Respir Dis 148: 1023–1029, 1993.[Web of Science][Medline]
  24. Khoo MC, Kronauer RE, Strohl KP, Slutsky AS. Factors inducing periodic breathing in humans: a general model. J Appl Physiol 53: 644–659, 1982.[Abstract/Free Full Text]
  25. Khoo MC, Shin JJ, Asyali MH, Kim TS, Berry RB. Ventilatory dynamics of transient arousal in patients with obstructive sleep apnea. Respir Physiol 112: 291–303, 1998.[CrossRef][Web of Science][Medline]
  26. Kimoff RJ, Cheong TH, Olha AE, Charbonneau M, Levy RD, Cosio MG, Gottfried SB. Mechanisms of apnea termination in obstructive sleep apnea. Role of chemoreceptor and mechanoreceptor stimuli. Am J Respir Crit Care Med 149: 707–714, 1994.[Abstract]
  27. Lo YL, Jordan AS, Malhotra A, Wellman A, Heinzer RC, Schory K, Dover L, Fogel RB, White DP. Genioglossal muscle response to CO2 stimulation during NREM sleep. Sleep 29: 470–477, 2006.[Web of Science][Medline]
  28. Longobardo GS, Gothe B, Goldman MD, Cherniack NS. Sleep apnea considered as a control system instability. Respir Physiol 50: 311–333, 1982.[CrossRef][Web of Science][Medline]
  29. Okabe S, Chonan T, Hida W, Satoh M, Kikuchi Y, Takishima T. Role of chemical drive in recruiting upper airway and inspiratory intercostal muscles in patients with obstructive sleep apnea. Am Rev Respir Dis 147: 190–195, 1993.[Web of Science][Medline]
  30. Okabe S, Hida W, Kikuchi Y, Taguchi O, Takishima T, Shirato K. Upper airway muscle activity during REM and non-REM sleep of patients with obstructive apnea. Chest 106: 767–773, 1994.[CrossRef][Web of Science][Medline]
  31. Onal E, Leech JA, Lopata M. Dynamics of respiratory drive and pressure during NREM sleep in patients with occlusive apneas. J Appl Physiol 58: 1971–1974, 1985.[Abstract/Free Full Text]
  32. Onal E, Lopata M, O'Connor T. Pathogenesis of apneas in hypersomnia-sleep apnea syndrome. Am Rev Respir Dis 125: 167–174, 1982.[Web of Science][Medline]
  33. Patil SP, Schneider H, Marx JJ, Gladmon E, Schwartz AR, Smith PL. Neuromechanical control of upper airway patency during sleep. J Appl Physiol 102: 547–556, 2007.[Abstract/Free Full Text]
  34. Phillipson EA, Sullivan CE. Arousal: the forgotten response to respiratory stimuli. Am Rev Respir Dis 118: 807–809, 1978.[Web of Science][Medline]
  35. Pillar G, Malhotra A, Fogel RB, Beauregard J, Slamowitz DI, Shea SA, White DP. Upper airway muscle responsiveness to rising PCO2 during NREM sleep. J Appl Physiol 89: 1275–1282, 2000.[Abstract/Free Full Text]
  36. Powles AC, Campbell EJ. An improved rebreathing method for measuring mixed venous carbon dioxide tension and its clinical application. Can Med Assoc J 118: 501–504, 1978.[Abstract]
  37. Rebuck AS, Slutsky AS. Measurement of ventilatory responses to hypercapnia and hypoxia. In: Regulation of Breathing, edited by Hornbein TF. New York: Dekker, 1981, vol. 17, p. 745–772 (Lung Biol. Health Dis.)
  38. Rechtschaffen A, Kales A. A Manual of Standardized Terminology, Techniques, and Scoring System for Sleep Stages of Human Subjects. NIH Publication. No. 204. Washington, DC: U.S. Government Printing Office, 1968.
  39. Remmers JE, deGroot WJ, Sauerland EK, Anch AM. Pathogenesis of upper airway occlusion during sleep. J Appl Physiol 44: 931–938, 1978.[Free Full Text]
  40. Series F, Series I, Cormier Y. Effects of enhancing slow-wave sleep by gamma-hydroxybutyrate on obstructive sleep apnea. Am Rev Respir Dis 145: 1378–1383, 1992.[Web of Science][Medline]
  41. Sforza E, Krieger J, Petiau C. Arousal threshold to respiratory stimuli in OSA patients: evidence for a sleep-dependent temporal rhythm. Sleep 22: 69–75, 1999.[Web of Science][Medline]
  42. Tuck SA, Remmers JE. Mechanical properties of the passive pharynx in Vietnamese pot-bellied pigs. II. Dynamics. J Appl Physiol 92: 2236–2244, 2002.[Abstract/Free Full Text]
  43. Vazquez JC, Tsai WH, Flemons WW, Masuda A, Brant R, Hajduk E, Whitelaw WA, Remmers JE. Automated analysis of digital oximetry in the diagnosis of obstructive sleep apnoea. Thorax 55: 302–307, 2000.[Abstract/Free Full Text]
  44. Veasey SC. Pharmacotherapeutic trials for sleep-disordered breathing. In: Sleep Apnea, edited by Pack AI. New York: Dekker, 2002, vol. 166, p. 607–622 (Lung Biol. Health Dis.).
  45. Veasey SC, Guilleminault C, Strohl KP, Sanders MH, Ballard RD, Magalang UJ. Medical therapy for obstructive sleep apnea: a review by the Medical Therapy for Obstructive Sleep Apnea Task Force of the Standards of Practice Committee of the American Academy of Sleep Medicine. Sleep 29: 1036–1044, 2006.[Web of Science][Medline]
  46. Vincken W, Guilleminault C, Silvestri L, Cosio M, Grassino A. Inspiratory muscle activity as a trigger causing the airways to open in obstructive sleep apnea. Am Rev Respir Dis 135: 372–377, 1987.[Web of Science][Medline]
  47. Wasserman K, Whipp BJ, Casaburi R. Respiratory control during exercise. In: Handbook of Physiology. The Respiratory System. Control of Breathing. Bethesda, MD: Am. Physiol Soc., 1986, sect. 3, vol. II, pt. 2, chapt. 17, p. 595–620.
  48. Wellman A, Jordan AS, Malhotra A, Fogel RB, Katz ES, Schory K, Edwards JK, White DP. Ventilatory control and airway anatomy in obstructive sleep apnea. Am J Respir Crit Care Med 170: 1225–1232, 2004.[Abstract/Free Full Text]
  49. Wilcox PG, Pare PD, Road JD, Fleetham JA. Respiratory muscle function during obstructive sleep apnea. Am Rev Respir Dis 142: 533–539, 1990.[Web of Science][Medline]
  50. Younes M. The physiologic basis of central apnea and periodic breathing. Curr Pulmonol 10: 265–326, 1989.
  51. Younes M. Mechanisms of respiratory load compensation. In: Regulation of Breathing, edited by Dempsey JA, Pack AI. New York: Dekker, 1995, vol. 79, p. 867–922 (Lung Biol. Health Dis.).
  52. Younes M. Contributions of upper airway mechanics and control mechanisms to severity of obstructive apnea. Am J Respir Crit Care Med 168: 645–658, 2003.[Abstract/Free Full Text]
  53. Younes M. Role of arousals in the pathogenesis of obstructive sleep apnea. Am J Respir Crit Care Med 169: 623–633, 2004.[Abstract/Free Full Text]
  54. Younes M, Ostrowski M, Thompson W, Leslie C, Shewchuk W. Chemical control stability in patients with obstructive sleep apnea. Am J Respir Crit Care Med 163: 1181–1190, 2001.[Abstract/Free Full Text]
  55. Younes M, Park E, Horner RL. Pentobarbital sedation increases genioglossus respiratory activity in sleeping rats. Sleep 30: 478–488, 2007.[Web of Science][Medline]
  56. Zamagni M, Sforza E, Boudewijns A, Petiau C, Krieger J. Respiratory effort. A factor contributing to sleep propensity in patients with obstructive sleep apnea. Chest 109: 651–658, 1996.[CrossRef][Web of Science][Medline]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Regul. Integr. Comp. Physiol.Home page
P. N. Ainslie and J. Duffin
Integration of cerebrovascular CO2 reactivity and chemoreflex control of breathing: mechanisms of regulation, measurement, and interpretation
Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2009; 296(5): R1473 - R1495.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
M. Younes
Role of respiratory control mechanisms in the pathogenesis of obstructive sleep disorders
J Appl Physiol, November 1, 2008; 105(5): 1389 - 1405.
[Abstract] [Full Text] [PDF]

Home page
 PhysiologyHome page
Highlights From The Literature
Physiology, April 1, 2008; 23(2): 61 - 63.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
103/6/1929    most recent
00561.2007v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Younes, M.
Right arrow Articles by Hanly, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Younes, M.
Right arrow Articles by Hanly, P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2007 by the American Physiological Society.