Usefulness of phrenic nerve stimulation to measure upper airway collapsibility in normal awake subjects

Respiratory Physiology & Neurobiology 130 (2002) 57 – 67 www.elsevier.com/locate/resphysiol

Usefulness of phrenic nerve stimulation to measure upper airway collapsibility in normal awake subjects C. Sanfac¸on a, E´. Ve´rin b, I. Marc a, F. Se´rie`s a,* a

Centre de Recherche, Hoˆpital La6al, Institut Uni6ersitaire de Cardiologie et de Pneumologie de l’Uni6ersite´ La6al, 2725 Chemin Sainte-Foy, Sainte-Foy, Quebec, Canada G1V 4G5 b Ser6ice de Physiologie, Centre Hospitalier Uni6ersitaire de Rouen, Rouen, France Accepted 30 October 2001

Abstract Upper airway (UA) collapsibility can be characterized during sleep by looking at the changes in inspiratory flow limitation (IFL) with changing nasal pressure. IFL can be induced during wakefulness using phrenic nerve stimulation (PNS) applied during exclusive nasal breathing. The aim of the study was to evaluate the possibility of measuring UA critical pressure (Pcrit) in normal awaked subjects using electrical PNS (EPNS) or bilateral anterior magnetic phrenic stimulation (BAMPS). Instantaneous flow, esophageal (Peso) and mask pressures (Pmask), and genioglossal (GG) end-expiratory EMG activity were recorded in 13 normal subjects (4F, 9M) with randomly changing Pmask (0 to −20 cmH2O). For each trial, we examined the relationship between maximal inspiratory flow (V: I max) of IFL twitches and the corresponding Pmask. Pcrit could be determined in 12 subjects (mean − 33.59 16.3 cmH2O). No difference in Pcrit values was found between the EPNS and BAMPS methods but the strength of the V: I max/Pmask relationship was higher with BAMPS. GG end-expiratory EMG activity increased with decreasing Pmask but no significant relationship was found between the slope of the GG end-expiratory EMG activity/Pmask relationship and Pcrit. We conclude that: (1) Pcrit can be measured during wakefulness in normal using PNS; (2) Pcrit measurements may be easier and more reliable with BAMPS than EPNS; and (3) Pcrit does not seem to be influenced by the pressure-related changes in GG end-expiratory EMG. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Airways, upper, collapsibility; Flow, limitation, inspiratory; Mammals, humans; Muscles, upper airways; Nerve, phrenic, stimulation

1. Introduction

* Corresponding author. Tel.: + 1-418-656-4747; fax: +1418-656-4762. E-mail address: [email protected] (F. Se´rie`s).

Obstructive sleep apnea (OSA) is a condition in which the upper airways (UA) collapse during sleep and results in a decrease or cessation of airflow despite continuing inspiratory efforts.

1569-9048/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 4 - 5 6 8 7 ( 0 1 ) 0 0 3 3 9 - 5

58

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

OSA is highly prevalent in the general population (Young et al., 1993) and is associated with a high rate of mortality and morbidity (Barbe´ et al., 1998; Peppard et al., 2000). The pressure at which UA close (critical pressure or Pcrit) represents a good indicator of UA stability. During sleep, Pcrit is subatmospheric in normal, but it is less negative or even positive in snorers and apneic patients and is correlated with the frequency of sleep-related obstructive breathing disorders (Gleadhill et al., 1991; Issa and Sullivan 1984). The effectiveness of different therapeutic procedures such as mandibular advancement (Isono et al., 1995), uvulopatalopharyngoplasty (Schwartz et al., 1992) and weight loss (Schwartz et al., 1989) in improving sleep-related obstructive breathing disorders parallels the decrease in critical pressure. Several techniques are available to measure Pcrit during sleep, including the analysis of the relationship between UA pressure and the velopharyngeal cross-sectional area following abrupt changes in the UA pressure (Isono et al., 1993) or the relationship between maximal inspiratory airflow (V: I max) and UA pressure for limited respiratory cycles (IFL) at varying levels of airway pressure (negative and positive) (Schwartz et al., 1988). These techniques may be quite easily achieved in OSAS while asleep since these patients fall rapidly asleep with continuous positive pressure even with complex instrumentation due to chronic partial sleep deprivation. However, in normal and in non-apneic snorers this technique is much more complicated since it requires the application of negative UA pressure and even sometimes sleep deprivation that has been shown to influence the UA stability (Se´ rie`s et al., 1994). It would therefore be of considerable interest to be able to accurately measure UA collapsibility during wakefulness. This has been attempted by Suratt et al. (1984) by applying a sudden decrease in UA pressure (− 10 to −60 cmH2O). However, Pcrit measurements could be achieved in only 67% of subjects and an important overlap was found between values obtained in normal and apneics subjects. We applied the method described by Schwartz et al. (1988) in awoken subjects using low level of negative pressure (between −30 and −40 cmH2O), but this technique requires a

marked collaboration of patients who often fight against the collapsing force by voluntarily contracting their UA musculature to maintain UA patency when submitted to such negative pressure. It is now well accepted that the pharynx behaves like a collapsible tube (Starling resistor) (Schwartz et al., 1989), the fine-tuning in the timing and magnitude of UA and respiratory muscle activity playing a key role in the maintenance of UA patency. UA closure will occur when dilating forces produced by the contraction of UA dilator do not counterbalance the collapsing forces of the transpharyngeal negative pressure gradient and tissue weight. Physiologically, UA dilator muscle activation proceeds that of the inspiratory muscle (Strohl et al., 1980) during wakefulness in such a way that the peak EMG activity is reached before that of the diaphragm. In apneics the loss of this pre-activation pattern during sleep parallels the occurrence of obstructive breathing events (Hudgel and Harasick, 1990). This loss of coordination between UA musculature and inspiratory muscle activity observed during sleep plays an important role in the pathophysiology of obstructive sleep apnea. We have previously found that this dissociation between the activation of the UA dilators and the diaphragm can be induced during wakefulness by the mean of electrical phrenic nerve stimulation (EPNS), and that this procedure results in a dramatic increase in UA resistance and IFL in the absence of previous activation of UA musculature in normal (Se´ rie`s et al., 1999) and apneic subjects using bilateral anterior magnetic phrenic stimulation (BAMPS) (Se´ rie`s et al., 2000). We reasoned that phrenic nerve stimulation (EPNS and BAMPS) could be used at different mask pressure (Pmask) levels to determine Pcrit in awake normal by looking at the V: I max/Pmask relationship of IFL stimulated breaths. 2. Material and methods

2.1. Subjects Thirteen non-snoring subjects were recruited for this study. A screening history in each subject

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

discloses no medical illness or anatomic abnormalities that could cause UA occlusion. None of the subjects complained of any symptoms suggestive of obstructive sleep disorder. A conventional in-lab polysomnographic study was completed in the five subjects who were studied with BAMPS. They were on no medication except for one female who took oestroprogestatives. The internal review board of our institution approved this protocol and informed consent was obtained from each subject.

2.2. Protocol Surface recording of the right and left costal diaphragmatic EMG activities were obtained by bipotential skin electrodes or by silver cup electrodes (for four subjects of the EPNS group) placed on the axillary line in the 6– 8 right and left intercostal spaces and connected to a electromyograph (Biopac system/Biopac, Santa Barbara, CA). We proceeded to this change because silver cup electrodes may offer a wider skin contact surface, therefore potentially increasing the quality of M-waves signal. However, no difference was identified with both types of electrode. After local anesthesia (1 ml of viscous xylocaine 2%) an esophageal balloon was inserted through one nare into the lower third of the esophagus as assessed by the occlusion technique (Baydur et al., 1982). In the eight subjects investigated with EPNS, a pressure tipped catheter (Gaeltec, model CT/S X1058, Hackensack, NJ) passed through the other nostril in the nasopharynx at 8 cm from the nares recorded pharyngeal pressure. The two catheters were securely taped to the nose. A plastic nasal stent (Petruson, 1990) was also placed in the anterior nares to prevent inspiratory collapse of the nares. A tightly fitting nasal continuous positive pressure mask was then placed over the nose, its airtightness being assessed by occluding its opening during maximal inspiratory efforts. Another catheter was passed through another opening of the mask to measure the pressure inside the mask (Pmask). Esophageal and mask catheters were connected to differential pressure

59

transducers (Validyne MP 459 100 cmH2O). Esophageal pressure was referenced to mask pressure. The mask was then connected to a pneumotachograph (Hans Rudolph, model 1124673850A, Kansas City, MO), fixed to a non-rebreathing valve (Whisper Swivel, Respironics, Murrysville, PA) and to flexible tubing. The distal end of the tube was attached to a three-way valve: one opening being connected through a 180-L capacitance opened to a vacuum source and the other to a variable resistance. The subjects were studied supine with the head supported by a pre-molded firm pillow to ascertain that the head and neck position did not change during the experiment. The electrical activity of the genioglossus (GG) was recorded in subjects undergoing the EPNS procedure. EMG electrodes were mounted on a mouthpiece made from dental impression (Doble et al., 1985) with the two Teflon-coated stainless wires sewn through the mouthpiece and in contact with the mouth floor. The EMG signals were amplified (Grass CP122, Quincy, MA), filtered (10 Hz – 3 KHz), rectified and integrated with a moving time averager with a time constant of 100 msec (MA 1000, CWE, Ardmore, PEN). To assure good position of electrodes on the mouthpiece, we asked the subjects to perform reproducible quantifiable maneuvers previously described by Mezzanotte et al. (1992). These maneuvers included swallowing, maximal protrusion of the tongue against the maxillary alveolar ridge, and a Mu¨ eller maneuver.

2.3. Study design After the different sensors have been installed, the supine subjects were instructed to breathe exclusively through the nose without having any respiratory or behavioral training prior to the experiment. For EPNS, twitch pulses were delivered from a Grass stimulator (S88, Quincy, MA) through a stimulus isolation unit (Grass SIU 5A, Quincy, MA). The phrenic nerve was stimulated at the anterior border of the sternocleidomastoid muscle bilaterally, at the level of the cricoid carti-

60

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

lage, using two bipolar electrodes with saline soaked felt tips 5 mm in diameter with 2 cm between electrodes. The stimulator was set to deliver a square wave pulse of 0.1 msec duration. Using a low intensity EPNS, train stimulation at 1 Hz was used to assess the phrenic nerve location. A recruitment procedure was realized to determine the supramaximal level of stimulus intensity that was associated with a plateau in the amplitude of the diaphragmatic M-waves (motor evoked potential). EPNS was then further increased by 10–20% to ascertain supramaximality. With BAMPS, twiches were delivered by two coupled Magstim 200 stimulator units (Magstim Ltd, Whitland, Dyfed, UK) equipped with a 43 mm figure-of-eight coil as described earlier by Mills et al. (1996). Various levels of negative pressures were applied through the CPAP mask in a randomized order (0 cmH2O to − 10 for BAMPS and − 20 cmH2O for EPNS) in 2 cmH2O steps. At each pressure setting, a 1 min stabilization period was observed before initiating the stimulation procedures. Ten stimulations were then obtained at supramaximal stimulus intensity at each negative pressure level. All ten stimulations had to be completed before Pmask was modified. Twitches were applied at the end of expiration as assessed by on-line instantaneous flow tracing.

measured. V: I max values of flow-limited breaths were plotted against the corresponding Pmask using a least-square regression analysis to establish the V: I max/Pmask relationship. Pcrit was given as the zero-flow intercept obtained from the regression analysis between V: I max and Pmask. At each pressure level tonic genioglossal EMG activity was estimated as GG end-expiratory EMG activity expressed by the percentage of baseline of tonic EMG values obtained at atmospheric pressure. The influence of negative airway pressure on tonic GG EMG activity was assessed by the individual relationship between breath-bybreath values of tonic EMG activity and the Pmask. The slope of this relationship was used to quantify the increase in GG EMG activity induced by negative airway pressure. In order to assess the influence of this EMG response on UA stability measured by EPNS, we analyzed the correlation between the slope of the GG end-expiratory EMG activity/Pmask relationship and two different indices of UA stability, Pcrit and the slope of the V: I max/Pmask relationship, respectively. Nasopharyngeal resistance was measured by the ratio of pharyngeal pressure/V: I max obtained at V: I max. Statistical significance was set at PB 0.05.

2.4. Data and statistical analysis

3.1. Anthropometric data

Flows and all pressure tracings were recorded on a microcomputer. Pmask, GG EMG and flows were also collected on a paper recorder. We excluded stimulated breaths where EPNS was performed at expiratory flow values higher than 150 ml sec − 1 or at inspiratory flow higher than 50 ml sec − 1 and/or during unstable GG EMG condition (swallow) and those demonstrating a nonflow limited pattern. Breathing cycles were identified as flow limited when the inspiratory flow plateaued or decreased while the twitch inspiratory efforts (esophageal pressure) increased. Esophageal pressure before stimulation should lie between + 1 and − 1 cmH2O. At each mask pressure, maximal twitch flow (V: I max) of each twitch flow-limited breathing was

This study was performed in 13 subjects, four men and four women. The mean age was 35.39 11.4 year (mean9 S.D.), with a 22–54 year range. Among men, the mean age was 40.99 8.9 year, body mass index was 24.99 2.6 kg m − 2, and neck circumference 37.592.4 cm with no significant difference between the EPNS and BAMPS groups. In females, the mean age was 2391 year, body mass index was 22.69 1.6 kg m − 2, and neck circumference 34.39 0.5 cm.

3. Results

3.2. Flow-limited breaths induced by PNS No IFL was observed during spontaneous breathing in any of our subjects. PNS induced partial UA closure as demonstrated by the pres-

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

ence of twitch flow-limited breaths. A representative example of twitch flow limitation induced by EPNS is shown in Fig. 1, as illustrated by the dissociation between the inspiratory effort (Peso) and twitch flow. It should be emphasized that no rise in phasic EMG genioglossal activity was observed before EPNS onset. The number of stimulated breaths retained for analysis varied from one subject to the other due to the application of our strict inclusion and exclusion criteria. Considering only the twitches that fulfilled these criteria, 87.29 12.4% of them presented the characteristic features of flow limitation with EPNS and 100% with BAMPS and were then considered for analysis.

3.3. Maximal inspiratory airflow V: I max/Pmask relationship Fig. 2 illustrates the effect of decreasing subatmospheric mask pressure on flow characteristics in a representative subject with EPNS (No. 1). In panel A, PNS performed at atmospheric pressure induced IFL. This flow-limited pattern worsened as the mask pressure was reduced with a progressive decrease in V: I max (panel B– D).

61

Figs. 3 and 4 represent the individual (V: I max)/ Pmask relationships for each IFL twitch with EPNS and BAMPS, respectively. A significant relationship between V: I max and Pmask was obtained in 12 of the 13 subjects (mean slope 30.59 14.6 ml sec − 1 cmH2O, PB 0.005). In these subjects the zero-flow intercept of the simple regression equation was calculated to determine Pcrit. The mean Pcrit was −33.59 16.3 cmH2O with a − 19.7 to −75.9 cmH2O range. No difference was found between the Pcrit values measured with the EPNS method (− 36.59 18.6 cmH2O) and the BAMPS technique (−29.49 13.2 cmH2O). Excluding the outlier value (subject No. 8, Fig. 3), Pcrit was − 29.699.8 cmH2O.

3.4. Changes in end-expiratory EMG acti6ity during EPNS In order to assess the influence of changes in end-expiratory EMG activity on UA stability, we first looked at the individual relationships between end-expiratory EMG activity of the genioglossus and the corresponding Pmask (Fig. 5). In seven out of the eight subjects end-expiratory EMG activity increased with decreasing Pmask. In subject No. 4, end-expiratory EMG activity remained unchanged with decreasing Pmask. Since increase in end-expiratory EMG activity may influence V: I max values, we looked at the relationship between the individual values of the slope of end-expiratory EMG activity/Pmask and the characteristics of flow dynamics (Pcrit, slope of the V: I max/Pmaskrelationship) (Fig. 6A and B). No relationship was found between the negative pressureinduced changes in tonic GG EMG activity and Pcrit (P=0.74, R= 0.15), nor with the slope of the V: I max/Pmask relationship (P =0.27, R=0.48).

3.5. Nasopharyngeal resistance Fig. 1. Example of airflow limitation induced by EPS at a Pmask during continuous negative airway pressure. When EPS is performed at end-expiration, no previous phasic activation of the genioglossus (GG) is observed. At a certain level of oesophageal pressure (Peso), inspiratory flow falls (V: I max) (*) despite increasing twitch oesophageal pressure. This flow pattern signs the presence of a partial obstruction in the UA. The narrow indicates the beginning of the EPNS.

Baseline nasopharyngeal resistance was 1.49 0.9 cmH2O L − 1 sec − 1. No significant relationship was found between the baseline nasal resistance values and Pcrit values (P=0.3). Since the upstream resistance is an important determinant of flow limitation characteristics, we looked at the

62

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

Fig. 2. Examples of flow-limited breaths obtained at different Pmask with EPNS: A, at 0 cmH2O; B, at − 6 cmH2O; C, at −12 cmH2O; and D, at − 18 cmH2O. V: I max (identified by *) decreased significantly with decreasing Pmask.

influence of the changes in nasal resistance with decreasing mask pressure on flow-limited stimulated breaths. In all subjects, a significant negative relationship was found between V: I max of flow-limited twitches and upstream resistance (R range 0.32 –0.81, P50.03).

4. Discussion The major findings in this study are that with a progressive decrease in UA pressure, Pcrit can be calculated during wakefulness using electrical or magnetic phrenic nerve stimulation techniques in the majority of normal subjects by analyzing the relationship between V: I max and Pmask. Pcrit values obtained with EPNS did not differ from those obtained with BAMPS even if this the range of negative pressure was less with this last procedure (down to −20 and −10 cmH2O, respectively). Further, the significance of the V: I max/ Pmask relationship and the corresponding correlation coefficients were higher with the BAMPS technique even when the analysis of the EPNS data was restricted to the 0 to − 10

cmH2O pressure range. We believe that this is the direct consequence of the higher number of flowlimited twitches obtained with BAMPS to that with EPNS (Se´ rie`s et al., 1999). From a practical point of view, these results indicate that the determination of Pcrit according to the presently described protocol is easier with BAMPS than with EPNS, and that this practical advantage is not at the expense of the results’ accuracy. Pcrit is a good determinant of UA stability and plays an important role in the pathophysiology of sleep-related disorders as illustrated by the differences in Pcrit obtained between normal, snorers, and apneic subjects (Isono et al., 1993, 1995). We are aware that any direct comparison between Pcrit measured during wakefulness using PNS and with modifying pharyngeal pressure during sleep cannot be made from this study due to the absence of measurements during sleep in the present study and to differences in subjects’ characteristics as compared with the previous reports (Brooks and Strohl, 1992; Schwartz et al., 1988). However, present results seem to indicate that Pcrit values obtained with EPNS tend to be more negative than those obtained previously in normal sleeping

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

subjects (−29.997.3 cmH2O) (Philip-Joe¨ t et al., 1996). It could be anticipated that differences in tonic EMG between wakefulness and sleep could account for a such difference. Animal studies clearly illustrated that the sleep-related loss in neuro-muscular activity is associated with a decrease in UA volume both during inspiratory and end-expiratory phases (Goh et al., 1986). The influence of tonic EMG activity on Pcrit values has been demonstrated in normal subjects during sleep by a positive relationship between end-expiratory GG response to negative UA pressure

63

and Pcrit values (Philip-Joe¨ t et al., 1996). However, now it will be of particular interest to investigate if the differences in Pcrit that exist between sleeping normal, snorers and SAHS can be reproduced during wakefulness with this technique. Pcrit measurement is influenced by the value of the maximal twitch inspiratory flow induced by PNS at atmospheric pressure and by the slope of the V: I max/Pmask relationship. These two factors can also be influenced by the changes in UA dilator muscles activity. It is well documented that negative pressure reflexly increases phasic and

Fig. 3. Individual relationships between V: I max and Pmask for IFL twitches with EPNS. A significant relationship was observed between these variables in seven of the eight subjects. The intercepts of these least-square regressions represent Pcrit. It could not be measured in subject No. 7 due to the insignificant relationship that was found between those variables.

64

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

Fig. 4. Same relationship as in Fig. 3 with the BAMPS technique. Pcrit measurements could be obtained in every subject.

tonic inspiratory EMG in many UA dilator muscles both in humans (Aronson et al., 1991; Leiter and Daubenspeck, 1990) and in animals (Mathew 1984). This increase in EMG activity is proportional to the level of negative pressure applied. We found that the tonic GG EMG, estimated by end-expiratory EMG activity, increased with a decreasing Pmask in the majority of our subjects (seven out of the eight). In order to assess the influence of negative pressure-induced increase in tonic EMG activity on Pcrit values, we looked at the relationship between the slope of the GG end-expiratory EMG activity/Pmask relationship and the individual values of Pcrit and the slope of the V: I max/Pmask relationships. The absence of correlation that was found between EMG response and Pcrit or with the slope of V: I max/Pmask relationship differs from the previous results obtained during sleep (Philip-Joe¨ t et al., 1996). This may be

accounted for by the differences in the GG EMG response to negative pressure during wakefulness and sleep as demonstrated by the slope of the relationship between GG end-expiratory EMG activity/Pmask during wakefulness (− 0.359 0.21% max EMG activity/cmH2O) and sleep (− 1.579 1.42% max EMG activity/cmH2O) (Philip-Joe¨ t et al., 1996). Given the fact that the negative pressure-induced increase in tonic and phasic UA EMG activities vanish over short time periods (14), and that our measurements were made after a stabilization period of 1 min at each pressure level; it is likely that the acute EMG response had dropped at the time when data collection was made in the present and previously published study. Therefore, we believe that the steeper increase in EMG activity with a decreasing airway pressure observed during sleep can result from the increase in CO2 consecutive to the increase in

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

UAR. Such an effect of ventilatory drive could be dramatically less during wakefulness where hypercapnic ventilatory drive is higher and behavioral factors significantly contribute to ventilatory control. Theoretically, some subjects may have developed voluntary behavioral EMG changes to fight against the collapsing force induced by negative pressure. In fact, if subjects had increased their UA dilating force, they should have modified their end-expiratory genioglossal EMG activity. Since no significant relationship was found between Pcrit and the slope of the end-expiratory EMG activity, we can

65

assume that if these behavioral changes were present, they were minimal and did not significantly influence Pcrit in these conditions. Since EMG responses of the different velopharyngeal dilator muscles are similar to those of the genioglossus (Mathew et al., 1982; Van Der Touw et al., 1994), it is reasonable to assume that the change in end-expiratory EMG pattern was the same in the other UA dilator muscles. It is therefore unlikely that Pcrit measurements obtained by PNS were influenced by the changes in tonic dilator muscles response to decreasing airway pressure during wakefulness.

Fig. 5. Individual relationship between GG end-expiratory EMG activity and Pmask for flow-limited twitch. Tonic genioglossal activity increased in seven subjects with decreasing Pmask. In subject No. 4, GG end-expiratory EMG activity did not change in response to reduction in Pmask.

66

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67

feasibility of this technique to determine Pcrit in normal awake subjects. Further studies are needed to explore if PNS could also be used to determine Pcrit in subjects with obstructive sleeprelated disorders since it mimics during wakefulness the delayed phasic activation of UA dilators that is present during obstructed breaths in sleep (Hudgel and Harasick, 1990). As discussed above, Pcrit values during wakefulness may be somewhat different from those obtained during sleep (e.g. slightly negative values compared to atmospheric or positive values during sleep). However, we speculate that these Pcrit values should be significantly less negative in SAHS than in normal subjects due to differences in UA mechanical properties such as differences in UA shape and dimension (Rodenstein et al., 1990). We conclude that EPNS offers a new practical way to determine UA collapsibility during wakefulness in normal subjects. The question of the applicability of this technique in identifying patients running the risk of having sleep-related obstructive breathing disturbances will be important to address in further studies. Fig. 6. Relationships between Pcrit and the slope of the GG end-expiratory EMG/Pmask relationship (Fig. 6A) and between the slopes of V: I max/Pmask and GG end-expiratory EMG activity/Pmask relationships (Fig. 6B). No significant relationship was found between these parameters.

Nasopharyngeal resistance represents the upstream resistance to the UA collapsing site and therefore may largely influence the dynamics of passive UA measured with EPNS. Thoracopulmonary deflation can modify nasal resistance (Se´ rie`s et al., 1990); therefore, negative UA pressure could also theoretically influence V: I max values. This is supported by the systematic negative correlation that we found between nasopharyngeal resistance and V: I max of flow-limited twitches. However, it is important to mention that no relationship was found between upstream resistance response and individual Pcrit values. This is not in discrepancy with the influence of nasal resistance on UA collapsibility since it must be reminded that measurements were obtained with a nasal stent. This study was conducted to determine the

Acknowledgements This work was supported by Medical Research Council of Canada MT 13768. The authors want to thank all the volunteers who participated in this study and S. Simard for his help in the completion of the statistical analysis.

References Aronson, R.M., Carley, D.W., O8 nal, E., Wilborn, J., Lopata, M., 1991. Upper airway activity and the thoracic volume dependence of upper airway resistance. J. Appl. Physiol. 70, 430 – 438. Barbe´ , F., Pericas, J., Munoz, A., Findley, L., Anto, J.M., Agusti, A.G.N., 1998. Automobile accidents in patients with sleep apnea. Am. J. Respir. Crit. Care Med. 158, 18 – 22. Baydur, A., Berhrakis, P., Zin, W.A., Jaeger, M., Milic-Emili, J., 1982. A simple method for assessing the validity of oesophageal balloon technique. Am. J. Respir. Crit. Care Med. 126, 788 – 791.

C. Sanfac¸ on et al. / Respiratory Physiology & Neurobiology 130 (2002) 57–67 Brooks, L.J., Strohl, K.P., 1992. Size and mechanical properties of the pharynx in healthy men and women. Am. J. Respir. Crit. Care Med. 145, 1394 –1397. Doble, E.A., Knuth, S.L., Daubenspeck, J.A., Bartlett, D., 1985. A non-invasive intraoral electromyographic electrode for genioglossus muscle. J. Appl. Physiol. 58, 1378 –1382. Gleadhill, I.C., Schwartz, A.R., Schubert, N., Wise, R.A., Permutt, S., Smith, P.L., 1991. Upper airway collapsibility in snorers and in patients with obstructive hypopnea and apnea. Am. J. Respir. Crit. Care Med. 143, 1300 –1303. Goh, A.S., Issa, F.G., Sullivan, C.E., 1986. Upper airway dilating forces during wakefulness and sleep in dogs. J. Appl. Physiol. 61, 2148 –2155. Hudgel, D.W., Harasick, T., 1990. Fluctuation in timing of upper airway and chest inspiratory muscle activity in obstructive sleep apnea. J. Appl. Physiol. 69, 443 – 450. Isono, S., Morrison, D.L., Launois, S.H., Feroah, T.R., Whitelaw, W.A., Remmers, J.E., 1993. Static mechanics of the velopharynx of patients with obstructive sleep apnea. J. Appl. Physiol. 75, 148 – 154. Isono, A., Tanaka, A., Sho, Y., Konno, A., Nishono, T., 1995. Advancement of the mandible improves velopharyngeal airway patency. J. Appl. Physiol. 79, 2132 – 2138. Issa, F.G., Sullivan, C.E., 1984. Upper airway closing pressure in snorers. J. Appl. Physiol. 57, 528 –535. Leiter, J.C., Daubenspeck, J.A., 1990. Selective activation of the genioglossus in humans. J. Appl. Physiol. 68, 2181 – 2187. Mathew, O.P., Abu-Osba, Y., Thach, T., 1982. Influence of upper airway pressure changes on genioglossus muscle respiratory activity. J. Appl. Physiol. 52, 438 –444. Mathew, O.P., 1984. Upper airway negative-pressure effects on respiratory activity of upper airway muscles. J. Appl. Physiol. 56, 500 – 505. Mezzanotte, W.S., Tangel, D.J., White, D.P., 1992. Waking genioglossal electromyogram in sleep apnea patients versus normals controls (a neuromuscular compensatory mechanism). J. Clin. Invest. 89, 1571 –1579. Mills, G.H., Kyroussis, D., Hamnegard, C.H., Polkey, M.I., Green, M., Moxham, J., 1996. Bilateral anterior magnetic stimulation of the phrenic nerves from an anterolateral approach. Am. J. Respir. Crit. Care Med. 154, 1099 –1105. Peppard, P.E., Young, G.T., Palta, M., Skatrud, J., 2000. Prospective study of the association between sleep-disordered breathing and hypertension. N. Engl. J. Med. 342, 1378 – 1384. Petruson, B., 1990. Snoring can be reduced when nasal airflow is increased by the nasal dilator Nozovent. Arch. Otolaryngol. Head Neck Surg. 116, 462 – 464.

67

Philip-Joe¨ t, F., Marc, I., Se´ rie`s, F., 1996. Effects of genioglossal response to negative airway pressure on upper airway collapsibility during sleep. J. Appl. Physiol. 80, 1466 – 1474. Rodenstein, D.O., Dooms, G., Thomas, Y., Liistro, G., Stanescu, D.C., Cule´ e, C., Aubert-Tilkens, G., 1990. Pharyngeal shape and dimensions in healthy subjects, snorers, and patients with obstructive sleep apnea. Thorax 45, 722 – 727. Schwartz, A.R., Smith, P.L., Wise, R.A., Gold, A.R., Permutt, S., 1988. Induction of upper airway occlusion in sleeping individuals with subatmospheric nasal pressure. J. Appl. Physiol. 64, 535 – 542. Schwartz, A.R., Smith, P.L., Wise, R.A., Bankman, I., Permutt, S., 1989. Effect of positive nasal pressure on upper airway pressure – flow relationships. J. Appl. Physiol. 66, 1626 – 1634. Schwartz, A.R., Schubert, N., Rothman, W., Godley, F., Marsh, B., Eisele, D., Nadeau, J., Permutt, S., Gleadhill, I., Smith, P.L., 1992. Effect of uvulopalatopharyngoplasty on upper airway collapsibility in obstructive sleep apnea. Am. Rev. Respir. Dis. 145, 527 – 532. Se´ rie`s, F., Cormier, Y., Desmeules, M., 1990. Influence of passive changes of lung volume on upper airways. J. Appl. Physiol. 68, 2159 – 2162. Se´ rie`s, F., Roy, N., Marc, I., 1994. Effects of sleep deprivation and sleep fragmentation on upper airway collapsibility in normal subjects. Am. J. Respir. Crit. Care. Med. 150, 481 – 485. Se´ rie`s, F., Demoule, A., Marc, I., Sanfac¸ on, C., Derenne, J.Ph., Similowsi, T., 1999. Inspiratory flow dynamics during phrenic nerve stimulation in awake normals. Am. J. Respir. Care Med. 160, 614 – 620. Se´ rie`s, F., Straus, C., Demoule, A., Attali, V., Arnulf, I., Derenne, J.P., Similowski, T., 2000. Assessment of upper airway dynamics in awaked sleep apnea patients with phrenic nerve stimulation. Am. J. Respi. Crit. Care Med. 162, 795 – 800. Strohl, K.P., Hensley, M.J., Hallett, M., Saunders, N.A., Ingram, R.H., 1980. Activation of upper airway muscles before onset of inspiration in normal humans. J. Appl. Physiol. 49, 638 – 642. Suratt, P.M., Wilhoit, S.C., Cooper, K., 1984. Induction of airway collapse with subatmospheric pressure in awake patients with sleep apnea. J. Appl. Physiol. 57, 140 – 146. Van Der Touw, T., O’Neil, N., Brancastisano, A., Amis, T., Wheatley, J., Engel, L.A., 1994. Respiratory-related activity of soft palate muscles: augmentation by negative upper airway pressure. J. Appl. Physiol. 76, 424 – 432. Young, T., Palta, M., Dempsey, J., Weber, S., Badr, S., 1993. The occurrence of sleep-disordered breathing among middle-aged adults. N. Engl. J. Med. 328, 1230 – 1235.