INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE UPPER AIRWAY is a complicated structure exhibiting a range of behaviors during breathing. Normally, the upper airway is widely patent throughout inspiration. During certain physiological and pathophysiological conditions, however, the upper airway collapses during inspiration. Collapse of the upper airway may completely or partially obstruct inspiratory airflow (I). Obstructive sleep apnea represents complete I obstruction at the nasopharynx, oropharynx, or hypopharynx (19, 23, 25, 36). Sniffing (2, 3, 14, 15, 24), snoring (4, 12), and stridor (6, 11, 20) represent partial I obstruction at the nares, pharynx, larynx, and extrathoracic trachea. Whenever partial I obstruction occurs, collapse of a flow-limiting site (FLS) limits airflow to a maximal level (I_max) (13, 14). Under these circumstances, I does not exceed I_max as inspiratory effort increases progressively. Thus, although the upper airway FLS may vary in location, the phenomena of collapse and I limitation are similar throughout the upper airway.

The airflow dynamics of the upper airway are affected markedly by the activity of its associated muscles. It is now recognized that upper airway muscles act in at least two ways. First, their contraction may affect airway caliber by dilating or constricting the airway (16, 17, 21, 31, 37, 38, 40, 41). Second, their contraction may reduce collapsibility by stiffening the airway (16, 17, 21, 34). Hypothetically, either action may affect the level of I_max during upper airway collapse and airflow limitation. However, it is unclear how upper airway muscle activity actually modulates I_max (1, 27).

Methods have been developed to determine the mechanism for alterations in I_max (14, 34). It has been established that the level of maximal flow through biological tubes is determined by two physiological parameters. The first is the transmural pressure (Ptm) at which the FLS collapses [the critical Ptm (Ptm')], which is an index of airway collapsibility. The second is the airway resistance upstream to the FLS (Rus), which reflects the caliber of the upstream airway segment. Thus, study of airflow dynamics during I limitation can help us to determine whether changes in I_max are related to changes in airway caliber or collapsibility.

In this study, we examined the modulation of I_max by muscle activity under conditions of I limitation. The study of maximal I dynamics requires the simultaneous measurement of muscle activity at the onset of I limitation, I_max, and the Ptm' of the FLS. Such a study is most easily performed in the nasal airway because it is easy to instrument, demonstrates I limitation, and contains the dilator naris portion of the nasalis muscle (the alae nasi) that is under voluntary control. In a recent study, we used a nasal catheter to locate the FLS of the nasal airway and to characterize nasal I dynamics under conditions of flow limitation (14). To examine how upper airway muscles influence I_max, we used the catheter method to study the effect of varying alae nasi activity on I_max and its determinants (Ptm' and Rus).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects. For this study, we selected only subjects who were able to voluntarily flare their nasal alae. Our six subjects (five men and one woman) were all Caucasian health care workers. One subject had a history of allergic rhinitis, but none of the subjects had a disorder of the lower respiratory tract. Each subject was free of upper respiratory tract symptoms on the day of study. The methods used were approved by the Research and Development Committee of the DVA Medical Center-Northport, and informed consent was obtained from the subjects.

Experimental apparatus. Our methods for studying maximal nasal airflow dynamics with the use of a nasal catheter have been previously presented in detail (14). Briefly, a nasal mask was connected in series with a pneumotachograph (model 3813, Hans Rudolph, Kansas City, MO) measuring I (Fig. 1). Mask pressure (Pmask) was measured with a pressure transducer (model 23ID, Spectramed, Oxnard CA) from a pressure port in the mask. Nasopharyngeal pressure (Pnp), a pressure downstream to the collapsible segment of the nasal airway, was measured by using anterior rhinometry through the left nostril.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Apparatus used to study maximal inspiratory airflow (I) dynamics of right nostril. A pneumotachograph in series with a nasal mask measures I, and a pressure transducer measures mask pressure (Pmask) from a port in the mask. Left nostril is occluded, and nasopharyngeal pressure (Pnp) is measured by anterior rhinometry. Surface EMG electrodes on the left ala measure activity of alae nasi. To measure pressure at flow-limiting site (PFLS), 2 fused catheters with side ports are place within the right nasal opening.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Schematic demonstration of our method for detecting a shift of the flow-limiting site (FLS). A: FLS located at nasal opening. Top: I tracing demonstrates onset of I limitation at maximal inspiratory airflow (Imax). Middle panels: Pmask is upstream to FLS and reaches its nadir at Imax; pressure immediately downstream to FLS (Pproximal) continues to decrease beyond Imax. Bottom: pressure 2 mm downstream to Pproximal (Pdistal) also continues to decrease beyond Imax. B (bottom panels): FLS has shifted downstream to Pproximal but remains upstream to Pdistal (i.e., the FLS has shifted downstream by <2 mm). Pproximal is now upstream to FLS and reaches its nadir at Imax. Pdistal is located immediately downstream to FLS and could be used to approximate PFLS. If FLS were to shift >2 mm downstream, both Pproximal and Pdistal would be upstream to FLS and would reach their nadirs at Imax.

Experimental protocol. Each of our subjects performed a series of sniffs while varying the level of alae nasi activity by voluntarily flaring the nasal alae. Subjects controlled the level of alar flaring by viewing the EMGan on the polygraph tracing (Fig. 3). We randomized the order of EMG activity levels and asked subjects to replicate each level of activation for three consecutive sniffs. Each subject was allowed to coordinate alae nasi activation with inspiratory effort comfortably. Reproducibility of the maximal nasal I dynamics was facilitated by spraying subjects' nostrils with a long-acting nasal decongestant (0.05% oxymetazoline hydrochloride).

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Tracing of alae nasi raw EMG activity and moving average alae nasi EMG (EMGan; downward deflection in this tracing) during 3 series of practice alar flares. First 3 flares represent maximal attempts. Subsequent 2 series of 3 flares represent 2 lesser levels of effort. Tracing is analog signal recorded at paper speed of 5 mm/s.

Data analysis. For each sniff, we determined the three parameters characterizing maximal nasal I: I_max, Ptm', and Rus. To accomplish this, we identified the moment I_max was attained and obtained the corresponding values of EMGan (EMGan'), PFLS (PFLS'), and Pmask (Pmask'; Fig. 4). We then fitted the values into the following equations

(1)

(2)

As in a previous investigation of alae nasi activity (9), EMGan activity was normalized by dividing the EMGan', in arbitrary units, by the peak value obtained during maximal voluntary flaring of the nasal alae. Using this method, we expressed the EMGan' as a percentage of each subject's maximum.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Polygraph recording of 2 sniffs performed at different levels of alae nasi activity. A: alae nasi activity is 0% of maximum that subject can develop. Imax (vertical line) is 530 ml/s. Critical transmural pressure (Ptm') of FLS (0.5 cmH2O) is calculated by subtracting critical Pmask from critical PFLS (PFLS'  Pmask'). B: subject has voluntarily flared the nostrils, and EMGan activity is 89% of maximum at Imax. Imax has increased to 1,020 ml/s. Ptm' of FLS has decreased to 2.7 cmH2O. In this example, alae nasi activation has increased Imax and decreased nasal Ptm'. Pnp' = Pnp at Imax.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

All six subjects demonstrated a FLS at the nasal opening. Voluntary alar flaring produced values of EMGan' that ranged from 1.3 ± 1.6 (range, 0-3.8%) to 72.3 ± 11.0% (range, 55.2-81.6%) of maximal EMG activity. Although all subjects were able to match their maximal EMG activity throughout the study, they were unable to synchronize I_max with values of EMGan' above 82% of maximal activity. Alae nasi activation did not cause downstream migration of the FLS in any subject.

The effects of alae nasi activity on I_max, Ptm' and Rus are illustrated in Fig. 5 and Table 1. Alae nasi activity significantly increased I_max and decreased Ptm' (Table 1). All subjects demonstrated a significant correlation between EMGan' and both I_max and Ptm' (Fig. 5).

View larger version (27K): [in this window] [in a new window]   View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effect of progressive alae nasi activation on right nostril's Imax and its determinants Ptm' and upstream resistance (Rus). A: subjects 1-3; B: subjects 4-6. Alae nasi activation increased Imax and decreased Ptm' in all subjects, and both were significantly correlated with EMGan activity. Effect of alae nasi activation on Rus was variable. Rus increased in 2 subjects, decreased in 2 subjects, and was unchanged in 2 subjects.

View this table: [in this window] [in a new window]   Table 1.   Effect of alae nasi EMG activity on Imax, Ptm', and Rus

Alae nasi activity did not consistently affect Rus (Table 1). Figure 5 demonstrates a negative correlation between EMGan' and Rus for two subjects (subjects 3 and 6), a positive correlation for two subjects (subjects 1 and 2), and no correlation for two subjects (subjects 4 and 5).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In this study, we examined how an upper airway muscle affects I_max. We accomplished this by studying the effect of varying alae nasi activity on nasal I_max. Our methods allowed us to attribute any increase in I_max to a decrease in either Ptm' or Rus. Patients sniffed through one nostril while flaring the ala with varying levels of effort. Airflow dynamics were studied with a nasal catheter at the onset of I limitation (I_max). In all subjects, high levels of alae nasi activity were associated with an increased I_max and a decreased Ptm' of the nasal FLS. We also found that a decrease in Rus in two subjects contributed to observed increases in I_max. Moreover, the nasal FLS remained at the nasal opening over the range of EMG activity observed. We conclude from these findings that alae nasi activity increased I_max in one of two ways (Eq. 2). First, decreases in the Ptm' of the nasal FLS in each subject caused an increase in I_max. Second, decreases in Rus in some subjects also increased I_max.

Our method for assessing the response of flow dynamics to ala nasi activity depends on establishing the precise location of the nasal FLS. We accomplished this by locating the point in the airway at which the pressure in the nasal catheter continued to decrease beyond I_max. This methodology required a high level of precision. As demonstrated in Fig. 4A (PFLS), even a decrease of 0.2 cmH₂O in catheter pressure lasting 50 ms after I_max is enough to establish a downstream location for the nasal catheter. Because we used a computer to record both the catheter pressure and the I every 2 ms, our methods had the necessary precision to identify the nasal FLS. In previous use of this methodology (14) and in the present study, we have demonstrated a FLS at the nasal opening in 10 of 11 unique subjects (1 subject participated in both studies). One subject had a FLS located at a constriction at the junction of the greater alar and lateral cartilages (the limen nasi). None of our subjects had a FLS at the nasal valve. In contrast to our experience, previous investigators (2, 3) have suggested that the nasal FLS is usually deeper within the nasal airway at the nasal valve. We believe, however, that the nasal FLS is usually located at the nasal opening and that the methods previously employed (2, 3) lacked the precision to accurately localize the nasal FLS. Therefore, because of the precision of the catheter method, we have been able to establish the exact location of the FLS, and have determined it to be at the nasal opening for most of our subjects.

The catheter method allowed us to determine the effect of alae nasi activity on airflow dynamics in the nose. We recognize, however, that our methodology may have influenced our measurements of nasal function. First, we placed two catheters in the nasal opening; this may have narrowed the lumen and increased Rus. The combined area of the catheters was 0.06 cm², which represented no more than 6% of the average area of the nasal opening (>1.0 cm²; 18), suggesting that the effect of these catheters on Rus was minimal. Second, we recognize that the lower margin of the nasal mask funneled air into the nasal opening (Fig. 1). This may have produced marked pressure losses caused by convective acceleration and consequent increases in Rus. Whatever the mechanism, increases in Rus would cause decreases in I_max at any level of Ptm' (Eq. 2). Regardless of the magnitude of artifacts in our measurements of I_max and Rus, however, we would not expect any relationship between such artifacts and alae nasi activity. Thus, we do not believe that the catheter method affected the relationship between alae nasi activity and our measurements of nasal airway function.

Once the FLS was located, we were able to characterize the function of the FLS and upstream segment and to examine how their mechanical properties modulated I_max. Specifically, we could apportion changes in I_max to alterations in either Ptm' or Rus. In our subjects, increases in I_max with alae nasi activation were associated with decreases in Ptm'. In contrast, the response of Rus to alae nasi activation was variable. These findings suggest that increases in I_max were caused by decreases in Ptm' that reflected diminished collapsibility of the FLS.

Although the Ptm' of the nasal FLS decreased consistently in all subjects, the response of Rus to alae nasi activation was more variable. This variable response occurred despite the known dilatory effect of alae nasi activation on the nasal airway under conditions of submaximal flow. The variability in the response of Rus can be explained in two ways. First, it is likely that airflow turbulence occurred at I_max. When airflow is turbulent, airway resistance increases with increasing airflow. Consequently, Rus might have increased in individuals who experienced a large increase in I_max with alae nasi activation. Conversely, Rus might have remained the same or decreased in individuals in whom I_max did not increase greatly. Alternatively, the variable response of Rus can be explained by the effect of alae nasi activation on Ptm'. When the Ptm' of the FLS decreased, the segment upstream to the FLS was exposed to more subatmospheric intraluminal pressures during a vigorous sniff. This decrease in airway pressure may have narrowed the upstream segment and increased its resistance. The greater the decrease in Ptm' during alae nasi activation, the greater would be the tendency for Rus to increase. Thus, an interaction between the dilating effect of the alae nasi and changes in I_max and Ptm' could explain the variable response of Rus to alae nasi activation. Regardless of the mechanism, our data indicate that alterations in alae nasi activity have no consistent effect on Rus.

Previous investigators have differed on the mechanism by which alae nasi activity increases nasal airflow (1, 27). Strohl and associates (38) have demonstrated that, in humans breathing at submaximal flows, maximal alar flaring decreases nasal resistance by 29% compared with inspiration without alar flaring. This finding suggests that the major effect of the alae nasi is to dilate the nasal airway and decrease resistance under non-flow-limited conditions (at submaximal levels of airflow). Others (5, 8, 15) believe that the chief effect of the alae nasi is to stiffen the airway (decrease collapsibility), thereby preventing flow-limited conditions from developing.

Our findings suggest that the alae nasi may have increased airflow in both of these ways. First, the alae nasi decreases resistance by dilating the nasal airway (38), an effect that occurs consistently at submaximal flows and variably at I_max. Second, our findings demonstrate that alae nasi activation decreased the Ptm' of the nasal FLS. When the Ptm' falls, airway pressure must become more negative for the FLS to collapse and limit flow, and greater levels of nasal I can be accommodated (Eq. 2) as inspiratory effort rises with exertion. Once nasal I limitation occurs, however, further increases in inspiratory effort no longer generate any increase in nasal airflow, which can be supplemented only by mouth breathing (26). Thus, nasal I can increase substantially when the alae nasi activate, because they prevent the development of flow limitation with increasing ventilatory effort and because they decrease nasal resistance over the entire range of submaximal airflow.

In this study, we found the nasal valve to be located downstream to the nasal FLS. Its downstream location should determine its role in modulating airflow dynamics as follows. The nasal valve is the narrowest site within the nasal airway and would be expected to contribute significantly to the resistance of the downstream segment of the airway (15). As the downstream resistance rises (e.g., during congestion of the inferior turbinate), greater inspiratory effort is required to reach I_max. The level of nasal I_max, however, should not be affected by changes in downstream resistance. Nasal I_max can be affected only by the collapsibility of the FLS and by the nasal Rus rather than the downstream resistance. Thus, increasing the resistance at the nasal valve should decrease submaximal flows at any level of inspiratory effort without affecting the nasal I_max.

A model of the nasal airway in which changing the resistance at the nasal valve does not affect I_max appears to conflict with previously reported observations. Several studies (22, 28, 29) have demonstrated an increased peak nasal I_max after decongestion or surgical widening of the downstream nasal airway segment. Because the studies did not monitor the alae nasi, however, we do not know the contribution of changes in alae nasi activity to their findings. Decreasing nasal resistance (increasing submaximal airflow) could increase alae nasi activity and increase nasal I_max. An increase in I_max delivered to the upper or lower airway has been demonstrated to increase respiratory drive in humans (7, 10, 30). Such a mechanism can be elucidated if alae nasi activity is examined at various levels of downstream resistance.

Our findings in the nasal airway may also apply to the phenomena of collapse and flow limitation in the pharynx and the larynx (4, 6, 11, 12, 19, 20, 25, 35, 36). Regardless of the site in the upper airway, the principles of flow through a collapsible tube (the Starling resistor model) have accounted for airflow dynamics under flow-limited conditions. The Starling resistor is a passive model, with a FLS, an upstream segment, and a downstream segment with fixed mechanical properties. The upper airway, however, has a complex musculature with activity that changes dynamically with respiration. It is well recognized that the state of neuromuscular activity can dramatically alter the flow dynamics across several potential sites of I limitation throughout the upper airway. In our study, we have examined how flow dynamics in one such site are influenced by voluntary activation of a single muscle group. Our methods have allowed us to determine the effect of this muscle group on I_max and to discern differences in its effect on the FLS and the upstream segment. Using similar methods in the pharyngeal airway of an animal model, investigators have demonstrated that pharyngeal dilator muscle activity increases I_max by decreasing the collapsibility of the pharyngeal airway (32, 33). Moreover, like the function of the alae nasi in the nasal airway, the pharyngeal dilator muscles can increase I_max even as Rus increases (33). Thus, under conditions of flow limitation, dilator muscles from different FLS in the upper airway increase airflow by decreasing airway collapsibility regardless of their effect on Rus. This mechanism of action can only be examined after carefully locating the FLS and characterizing its function and that of the upstream segment.