| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Sleep Disorders Unit, Repatriation General Hospital, Daw Park, 5041; and 2 School of Medicine, Flinders University, Bedford Park, South Australia 5042 Australia
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The sensation of increased
respiratory resistance or effort is likely to be important for the
initiation of alerting or arousal responses, particularly in sleep.
Hypoxia, through its central nervous system-depressant effects, may
decrease the perceived magnitude of respiratory loads. To examine this,
we measured the effect of isocapnic hypoxia on the ability of 10 normal, awake males (mean age = 24.0 ± 1.8 yr) to
magnitude-scale five externally applied inspiratory resistive loads
(mean values from 7.5 to 54.4 cmH2O · l
1 · s). Each subject scaled the loads during 37 min of isocapnic hypoxia (inspired O2 fraction = 0.09, arterial O2 saturation of ~80%) and during 37 min of
normoxia, using the method of open magnitude numerical scaling. Results
were normalized by modulus equalization to allow between-subject
comparisons. With the use of peak inspiratory pressure (PIP) as the
measure of load stimulus magnitude, the perception of load magnitude
(
) increased linearly with load and, averaged for all loaded
breaths, was significantly lower during hypoxia than during normoxia
(20.1 ± 0.9 and 23.9 ± 1.3 arbitrary units, respectively;
P = 0.048). declined with time during hypoxia
(P = 0.007) but not during normoxia (P = 0.361). Our result is remarkable because PIP was higher at all times
during hypoxia than during normoxia, and previous studies have shown that an elevation in PIP results in increased . We conclude that sustained isocapnic hypoxia causes a progressive suppression of the
perception of the magnitude of inspiratory resistive loads in normal
subjects and could, therefore, impair alerting or arousal responses to
respiratory loading.
obstructive sleep apnea; psychophysics; arousal
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
AROUSAL FROM SLEEP IS CONSIDERED one of the most important protective respiratory responses (24), and it has particular significance in disease states such as obstructive sleep apnea. In most instances, arousal terminates an obstructive sleep apnea with an abrupt shift in electroencephalograph (EEG) frequency (1, 23). Research in normal subjects indicates that within-subject arousal occurs at the same threshold of respiratory effort, regardless of the type of respiratory stimulus [i.e., hypoxia, hypercapnia, or resistive loading (13)]. It has been postulated (1) that subjects "sense" the increased work of breathing associated with these stimuli and respond by arousing from sleep. Sleep apnea patients require much greater respiratory effort to arouse during an obstructive event than normal subjects (1, 2). Increased sleep drive secondary to the persistent sleep fragmentation experienced by sleep apnea patients may in part explain this difference. Hypoxia may also contribute to increased arousal threshold. Acute isocapnic hypoxia [arterial O2 saturation (SaO2) = 80-90%] has been shown to depress central respiratory neural drive (9, 22) and produce decrements in neurocognitive function (3). Furthermore, sleep apnea patients with associated severe hypoxemia demonstrate a higher level of cognitive impairment than those with little hypoxemia (12).
In the present study, we examined the effects of isocapnic hypoxia on the ability of normal subjects to interpret the magnitude of inspiratory resistive loads while awake. If hypoxia reduces load perception during wakefulness, it may also increase the threshold of respiratory effort at which arousal occurs in sleep and thus may help explain the high arousal threshold observed in sleep apnea patients during airway occlusion. If hypoxia depresses load perception, there may be implications for other respiratory diseases in which perception of increased respiratory load is important in enabling corrective behaviors (16).
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Subjects
Informed consent was obtained from 10 normal male nonsmokers (mean age = 24.0 ± 1.8 yr) who participated in the study as paid volunteers. Subjects had no history of cardiac or respiratory problems and were free of medication at the time of the experiment. No subject had been born, or had spent any extended time, at altitude. Pulmonary function measurements were normal in every case (forced expiratory volume was 117 ± 4% of predicted; forced vital capacity was 107 ± 3% of predicted). Subjects were recruited from the general student population with the assistance of the Flinders University recruitment agency and by word of mouth from nonmedical or research personnel within the hospital. Of the 10 subjects used, three had participated in a previous experiment to evaluate the impact of inspired gas on threshold load detection. The remaining subjects had no experience of respiratory load magnitude estimation. Subjects were not informed of the hypothesis and remained blind to the inspired gas at all times.General Procedure
Subjects arrived in the laboratory after a light breakfast without caffeine. Data were collected on each subject in a single morning session. Each session involved two 17-min practice runs and two experimental periods (an isocapnic hypoxic and a normoxic period). Each experimental period had a 37-min duration. During each experimental period, suprathreshold inspiratory resistive loads were presented to and magnitude-scaled by the subjects. The apparatus and psychophysical techniques of open magnitude scaling of inspiratory loads used were similar to those described in studies by Burdon et al. (4) and Killian et al. (17). Finger tapping was conducted during the experimental periods in an attempt to independently measure the effects of hypoxia on cortical function (3).Measurements
All measurements were conducted with the subjects sitting comfortably on a bed with back support. EEG monitoring from two surface electrodes (C3-A2, Compumedics S-series, Abbotsford, Victoria, Australia) was used during data collection solely to confirm that wakefulness was maintained for the duration of the study. No analysis of the recorded EEG was undertaken, although the data were retained for future reference. Electrocardiograms (ECG) were continuously recorded (Hewlett Packard 78342, Andover, MI). A full face mask with unidirectional valves (8900 series, Hans-Rudolph, KS) was fitted to the face with a silicone seal (Ultimate Seal, Hans-Rudolph). The mask was held in place by an adjustable retaining cap and checked for leaks by having the subject exhale with moderate force against a closed expiratory port. In addition, a perforated tube encircling the perimeter of the mask and attached to a CO2 analyzer (Poet II, 602-3, Criticare Systems) served as a leak detector throughout each experiment. Subjects were asked to choose their preferred route of breathing (mouth, nose, or a combination) and then retain this for the duration of the study. SaO2 was measured with a pulse oximeter (Poet II, model 602-1, Criticare Systems). Inspiratory flow was determined using a pneumotachograph (Jaeger) and integrated electronically to provide breath-by-breath tidal volume. Mask pressure (DTX transducer, Spectramed) and inspiratory flow were used to calculate inspiratory resistance. Gas was continuously sampled from a port in the mask, and end-tidal PCO2 (PETCO2) and inspired O2 fraction were determined using CO2 and O2 (Medgraphics, MN) analyzers. The output from a peizo-electric transducer (Leg Movement Sensor, Compumedics) provided data on finger tapping frequency. All data were acquired on a 486 IBM laptop computer using an acquisition system (DATAQ Instruments) with a sampling rate of 150 Hz per channel.Resistive Loading Apparatus
The apparatus used to provide resistive loading was similar to that described previously (4, 17) for inspiratory resistive scaling. Briefly, two concentric tubes were constructed from PVC pipe. The inner tube [80 mm (ID)] had panels removed, and the cleared areas were replaced with filter paper (Whatman International). The position of an internal plunger determined the area of paper that the subjects inspired through and, therefore, the inspiratory resistance. Five plunger settings were chosen for the experiment to give a wide range of inspiratory resistances (see Fig. 1). The inspiratory resistive circuit was housed in an outer PVC tube connected to a Hans-Rudolph two-way tap. This allowed the resistive loading apparatus to be bypassed or included in the inspiratory line as required. Loads were introduced during expiration and remained in place throughout the inspiratory phase of the following breath (i.e., the loaded breath). A buzzer and light were activated two breaths before the introduction of a load to cue the subject. For the purposes of magnitude scaling, loads were presented every three to four breaths, with the load remaining in place for one breath only.
|
Calibration of resistance apparatus. The flow vs. pressure relationship for each setting of the resistance circuit was determined by using a suction device to produce a wide range of flows. Negative pressure and the corresponding flow were graphed, and the resistance was calculated at a flow of 0.5 l/s for background (resistive apparatus bypassed) and each of the five resistor settings (Fig. 1). The third resistor setting (R3 in Fig. 1) was used as the reference resistor for the duration of the experiment. The curves generated demonstrate that our circuit had similar characteristics to that used by Burdon et al. (4).
Protocol
The protocol is represented diagrammatically in Fig. 2, and an explanation of each of the components is detailed in this section.
|
Finger tapping practice run. Subjects were instructed to tap as rapidly as they could with the index finger of their dominant hand for 10-s periods until they felt comfortable with the technique and were instructed to retain this technique for all subsequent tests.
Magnitude scaling practice session. Before each experimental period (isocapnic hypoxia or normoxia), a random sequence of the five loads was presented while the subject breathed room air with the mask in place. Subjects were informed that these loads were typical of what they would likely experience and were then given the following verbal instructions, modified from a text used by Stevens (32).
"We are going to present a series of loads to your breathing, and we want you to put a number on the loads according to their size. Write the numbers down as you go. No two loads will be exactly the same; however, you may not be able to tell the difference between some of them. First, we will give you a reference load, and we want you to write down the number of your choice that represents how big you think this load is. We will then give you a series of loads, and we want you to put numbers on them, in proportion to their size. If you feel that the second load is twice as big as the reference, then put a number twice as big as your reference number. If you feel it is two and a bit times greater, then put a number two and a bit times bigger. You can use whole numbers, fractions, or decimals, just as long as you are comfortable with the numbers you use." Subjects were then presented with and were required to assign a numeric value to the reference load (~20 cmH2O · l1 · s). Thirty loads were then magnitude-scaled
by the subject. For this and every subsequent presentation of 30 loads,
the same five plunger settings were used on six occasions, using a
random sequence. A new random sequence of loads was generated for each series of 30 load presentations. The magnitudes assigned to these loads
during both practice runs were not used in subsequent analyses. With
the presentation of the reference load during the second practice
session, subjects were reminded of the quantity they had assigned to
that load during their first session and asked to write that same value
down. This enabled the subjects to reinforce their earlier
quantification of the reference.
Experimental periods. There were two experimental periods: normoxia (breathing from a reservoir bag filled from a compressed air source) and isocapnic hypoxia (FIO2 = 0.09, SaO2 ~80%). Before the introduction of the experimental gas, subjects spent 5 min quietly breathing room air without intervention but with all respiratory signals recorded. Isocapnia was maintained in both normoxia and hypoxia experiments by a manual bleed of CO2 into the inspiratory line. After the 5-min quiet breathing period, the reference load was presented again, and the subjects were reminded of the number they had previously assigned to this load and instructed to write that number down on the paper provided.
Subjects were then switched to the experimental gas (9% O2 in N2, or medical air), and a new, random 30-load sequence (5 loads, each presented 6 times) was given over ~15 min. Subjects recorded their estimation of magnitude after every load presentation. After the first series of loads, a finger tapping set (five periods: two 10-s periods followed by 10-s gaps, then 10 s of finger tapping and a 1-min break, followed by two additional 10-s periods with a 10-s break) was performed. This finger tapping sequence is similar to that described by Lezac (20). This was followed by another 30-load sequence and further finger tapping. Consequently, each experimental period of isocapnic hypoxia and normoxia consisted of 60 load presentations. The experimental periods varied from 37 to 43 min, but the duration of the normoxic and isocapnic hypoxic period did not differ by more than 3 min for each subject. The order of presentation of the experimental gases was alternated among subjects. Consequently, one-half of the subjects breathed a normoxic mixture during the first experimental period, and the remaining subjects breathed a hypoxic mixture during the first experimental period. Each experimental period was separated by 45 min of rest, with the mask and seal removed. After the 45-min rest period, the mask was refitted, and the entire sequence of events was repeated with the alternate gas. The rest period plus the time required to refit the mask, amounted to a minimum of 60 min between experimental periods, thereby allowing the subjects to fully recover from any depressant effects of hypoxia on ventilation (10).Analysis of Data
Individuals were free to place whatever value they thought appropriate on the reference load and scaled all subsequent loads relative to this reference. Values for perceived magnitude of the inspiratory resistive loads () varied considerably with the choice
of scales between subjects. To ensure that each subject's data
provided equal weighting to the group results, individual values were
transformed by the method of modulus equalization (31).
Briefly, the geometric mean of all perception values recorded throughout the study was calculated (grand modulus) along with the
geometric mean of all values recorded for each subject (individual modulus). Perception values for each subject were then divided by the
appropriate individual modulus and multiplied by the grand modulus,
thereby adjusting individual perception values to a common scale
(31, 34).
Peak inspiratory pressure (PIP), peak inspiratory flow (PIF), inspired
minute ventilation (
I), inspiratory time
(TI), expiratory time (TE), and total breath
time (Ttot) were calculated for every breath, and mean
values were derived for the 5-min rest period (i.e., baseline) and over
early (1-5 min), mid (5-13 min), and late (20-32 min)
periods of isocapnic hypoxia and normoxia. SaO2 and
PETCO2 were sampled continuously and
averaged for the same time periods. Inspiratory resistance (R) was
calculated as the slope of the linear regression of pressure vs. flow
for loaded breaths only.
Relationships between , R, and PIP were assessed according to
conventional methods described for the perception of inspiratory resistive loads (5, 17, 19).
Mean R, PIP, TI, and values for each resistance setting
were calculated for the early (1-5 min), mid (5-13 min), and
late (20-32 min) periods of isocapnic hypoxia and normoxia.
Values of for each individual were determined at the group mean
values of R and PIP (isocapnic hypoxia and normoxia data combined) for
each resistance setting. Group mean R and group mean PIP for
each load setting were calculated from the arithmetic mean R and mean
PIP for all presentations of the given load setting (i.e., all
subjects, normoxia and hypoxia, and repeated presentations in each
subject combined). This enabled statistical comparisons of for
isocapnic hypoxia and normoxia at identical values of R and PIP. The
equations and reasoning for using linear equations to express the
relationship between and PIP are given below.
Finger tapping data were analyzed by a Fourier analysis of the frequency and amplitude characteristics of the record. Preliminary analysis showed that finger tapping frequency by this method differed by <5% from manually scored finger tapping frequency. Mean finger tapping frequency for each subject was calculated for the isocapnic hypoxic and normoxic experimental periods.
Examination of equations defining the relationship between the
variables , R, PIP, and TI.
The natural logarithms of , R, and PIP were determined, and linear
regression was used to derive the exponents (n) and
constants (k; natural logarithm intercepts) in the equations
= kRn and = k(PIP)n. Power functions of this type
are generally accepted as the best descriptor of psychophysical
relationships (30, 33, 34). In
simple terms, the exponent in power functions represents the curvature
of the relationship between the increase in sensation and the magnitude
of the increase in the physical stimulus producing that sensation. A
feature of a power function is that, when the curvilinear relationship
is plotted in log-log coordinates, the graph becomes linear, with the
slope of the line being a direct measure of the exponent. However, in
our experiments, the graphical relationship between and PIP
appeared linear and was, therefore, also examined using the equation
= a(PIP) + b, with a and
b representing the slope and intercept values,
respectively. Coefficients of determination
(r2) and the residual sum of squares (SS) for
each equation were determined in linear coordinates to enable
comparisons between the models.
= k(PIP)nTIm.
To determine whether the inclusion of inspiratory time in the equation
improved the fit of the data, this expression was compared with the
equation = k(PIP)n by
r2 and SS values.
Statistical analyses.
Comparisons of SS between the two models relating and PIP
[ = k(PIP)n and = a(PIP) + b] were made using Student's
paired t-tests. Models relating , PIP, and TI
[ = k(PIP)nTIm
and = k(PIP)n] were also
compared using the t-test. ANOVA for repeated measures [including the Greenhouse-Geisser correction for multi-sample asphericity (21)] was used to examine differences between
isocapnic hypoxia and normoxia in 1) the ventilatory
parameters and 2) the components k, n,
a, and b. ANOVA for repeated measures was
also used to test for differences in (calculated at group mean
values of PIP and R) between gases and with respect to time. All
significant ANOVA effects were explored using Tukey's post hoc tests.
All data are reported as means ± SE. Differences were considered
significant when P < 0.05.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Ventilatory Parameters During Baseline Rest Periods, Isocapnic Hypoxia, and Normoxia
There were no significant differences in any of the ventilatory parameters in the 5-min baseline room air-breathing periods before the experimental runs (Table 1). During isocapnic hypoxia, SaO2 decreased significantly over the first 8 min to ~80% and remained at that level for the remainder of the experimental period (Fig. 3, Table 1). Isocapnia was maintained throughout the study (Fig. 3, Table 1). During isocapnic hypoxia there was an increase in PIP, PIF, tidal volume (VT), andI compared with resting values (Table 1). As
expected, the increase in I was mainly due to an
increase in VT, with a small, nonsignificant increase in
TI. There was also a significant increase in PIP compared
with baseline during normoxia, and I was elevated
during the late period of normoxia. PIP during isocapnic hypoxia was
significantly elevated at all times compared with during normoxia
(Table 1). PIF, VT, and I were also
elevated during hypoxia compared with during normoxia, but differences
only reached statistical significance at 1-5 and 5-13 min.
There were no differences in TI, TE, or
Ttot between hypoxia and normoxia.
|
|
Perceived Magnitude of Added Loads
The group mean values of R experienced by the subjects for each setting on the apparatus were calculated from the pressure/flow values for all presentations (isocapnic hypoxia and normoxia combined). For each of the five resistance settings used, the experienced group mean R was 7.53 ± 0.03, 10.47 ± 0.05, 19.66 ± 0.11, 38.20 ± 0.30, and 54.44 ± 0.47 cmH2O · l1 · s.
Relationship between and R.
The of added loads increased curvilinearly with R and was well
described by Stevens' power function (r2 = 0.942 ± 0.006, SS = 43.33 ± 9.11; see Table
2 and Fig.
4). Values for the exponent
n and the intercept k that best fit the equation
= kRn are shown in Table 2.
During 1-5 min of hypoxia, n was significantly lower
and k was significantly higher than during the same time in
normoxia (P < 0.05). After 20-32 min of hypoxia,
k was significantly lower compared with 1-5 min of
hypoxia. The overall mean value of , calculated at the group mean
values of R, was not significantly different between hypoxia and
normoxia (20.09 ± 1.02 vs. 19.78 ± 1.05, respectively).
During hypoxia, however, computed at the group mean values of R
declined with time (P = 0.007) and was significantly
lower after 20-32 min (17.71 ± 4.22) compared with 1-5
min (22.71 ± 4.51) of hypoxia. By contrast, did not change
across the early, mid, or late periods of normoxia.
|
|
Relationship between and PIP.
Although the relationship between and PIP (see Table 2 and Fig. 4)
was well described by Stevens' power function
(r2 = 0.937 ± 0.011, SS = 34.09 ± 8.013), the exponents were not different from unity
(n = 1.079 ± 0.068; P = 0.249).
Furthermore, a linear model provided a significantly better fit to the
data (SS = 21.37 ± 3.48) than Stevens' power function
(P = 0.040) and was therefore retained for the
remaining analyses.
= a(PIP) + b
are shown for both hypoxia and normoxia in Table 2. There were no
differences in a between hypoxia and normoxia. However,
b was higher during 1-5 min of hypoxia compared with
the same period in normoxia and declined after 20-32 min of
hypoxia. The overall group mean value of , calculated at the group
mean values of PIP, was significantly lower during hypoxia than during
normoxia (20.10 ± 0.91 vs. 23.88 ± 1.27; P = 0.048), and, furthermore, declined with time during hypoxia
(P = 0.007) but not during normoxia (P = 0.361).
and TI.
Although TI on the loaded breath increased with R (from
2.29 ± 0.11 s at 7.53 cmH2O · l1 · s to 2.99 ± 0.17 s at 54.44 cmH2O · l1 · s;
P = 0.003), there were no differences in TI
between hypoxia and normoxia (P = 0.376), no changes
with time (P = 0.538), and no time or resistance
interactions between hypoxia and normoxia. Furthermore, the
conventional expression = kPIPnTIm
(overall n = 1.004 ± 0.076, m = 0.451 ± 0.234) did not improve the fit over that of the linear
model = a(PIP) + b, despite an
additional curve parameter (curvilinear SS = 24.59 ± 7.15; linear SS = 21.37 ± 3.48; P = 0.576).
Finger Tapping
Fourier analysis of the finger tapping data demonstrated no significant differences in the dominant frequency and associated amplitude between the isocapnic hypoxia and normoxia periods or within either of these periods. The mean frequency of finger tapping during isocapnic hypoxia and normoxia was 5.97 ± 0.21 and 5.86 ± 0.23 taps/s, respectively (P = 0.177).| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The main finding of the present study was that, in normal,
healthy, male subjects, the of externally applied inspiratory resistive loads decreased progressively during sustained isocapnic hypoxia but remained unchanged during normoxic breathing. This finding
held when either R or PIP of the loaded breath was used to quantify the
magnitude of the stimulus provided by the externally applied resistive
loads. The result is all the more remarkable because PIP was
significantly elevated at all times during hypoxia compared with
normoxia. Previous studies (4) have shown that an increase
in respiratory drive (i.e., PIP) generally results in an increase in
the of inspiratory resistive loads.
Some differences existed in our results when measured R vs. measured
PIP was used to define the magnitude of the applied load. There
was no overall difference in between hypoxia and normoxia when R
was used as the measure of load, but was lower during hypoxia than
during normoxia when PIP was used to define respiratory stimulus
arising from load application. Killian et al. (17) demonstrated that PIP bears a closer relationship to what is perceived as load than R and probably better represents the stimulus provided by
external resistive loads. We consider that the reduction observed in
load estimation by subjects during hypoxia, using PIP as the independent variable, is strong evidence for an effect of hypoxia on
the mechanisms subserving load evaluation.
Previous researchers (4, 5, 17)
have investigated the impact of adding loads to breathing on using
a range of inspiratory loads extending from just above background to
>75 cmH2O · l1 · s. To our
knowledge, there have been no previous studies that have systematically
investigated the effects of hypoxia on the perception of resistive
respiratory loads. However, in the study of Burdon et al.
(4), one of the stimuli used to increase respiratory drive
was isocapnic hypoxia. The relationship between R and , derived from
their data using hypoxia to increase respiratory drive, shows
considerable scatter. This was not the case for the other two
respiratory stimulants used in that study, exercise and hypercapnia.
The authors observed that "Following the hypoxic experiment many
subjects complained of difficulty in maintaining cognitive ability"
(4) and offered this as a possible explanation for their
findings. Our data also imply that the sensory mechanisms used to
assess the magnitude of inspiratory resistive loads are impaired during
hypoxia. Whether this is due to a "cognitive" (i.e., cortical)
deficit or to an interruption of afferent information in sensory
pathways at another level remains unclear.
The time course of the ventilatory response to sustained isocapnic
hypoxia is bimodal: early stimulation of respiration is followed, in a
few minutes, by a decline in respiratory neural drive (9,
22, 26). Each of the neuroeffectors shown to be associated with hypoxic ventilatory depression (i.e., endogenous opioids, adenosine, and GABA) has been shown to be capable of blocking
sensory pathways involved in pain perception (6-8,
11, 15, 27, 28,
29). It has been suggested by Killian and Campbell
(18) that the model for the processing of pain is equally valid for the processing of the proprioceptive information necessary for inspiratory load evaluation. Thus the same neuroeffectors that
depress respiration during continuing hypoxia may also diminish the of inspiratory resistive loads.
Although we postulate that the reduction in observed during hypoxia
may involve the same neuroeffectors associated with hypoxic ventilatory
depression, our data did not show a statistically significant decline
in ventilation across the hypoxic period. However, ventilation was
elevated above quiet breathing during the normoxic period, with the
difference reaching statistical significance by 20-32 min. We
hypothesize that the changes in ventilation observed during normoxia
were a consequence of the tasks subjects had to perform. Because the
tasks were identical during the hypoxic period, they would probably
have induced similar increases in ventilation, and these increases may
have been sufficient to mask hypoxic ventilatory rolloff.
Another potential explanation for our finding of reduced during
sustained hypoxia is that hypoxia induced cerebral arterial dilatation
and increased cerebral blood flow, resulting in cerebral hypocapnia.
However, this is an unlikely explanation because we observed that the
of inspiratory resistive loads progressively decreased during 32 min of isocapnic hypoxia. In contrast, previous studies in humans
(35) show that sustained isocapnic hypoxia produces a
sudden reduction in internal jugular PCO2
(i.e., in the first minute) that remains unchanged over 15 min.
Previous animal studies (22) suggest that local metabolic
regulation of cerebral blood flow counteracts, within 30-120 s,
the effects of any cerebral overperfusion induced by hypoxia. Thus, in
the same way that these authors argue that cerebral hypocapnia is unlikely to explain the progressive ventilatory rolloff during sustained isocapnic hypoxia, we suggest it is also unlikely to explain
the progressive decline in sensory of respiratory loads during hypoxia.
Our results show that hypoxia altered perceptual interpretation of inspiratory stimuli but did not change finger tapping frequency. Finger tapping was originally used as a test of manual dexterity and was not designed to specifically assess the impact of hypoxia. However, it has been shown by one group (3) to decrease with acute hypoxia, and the relative simplicity of the test enabled us to incorporate it into the experimental protocol. Berry et al. (3), using linear trend analysis, demonstrated a reduction in tapping speed at levels of hypoxia similar to those used in our experiments. However, these trend differences were shown between rather than within subjects. Although considerable efforts were made to match experimental and control groups, intersubject differences may have contributed to their results. Our study, which failed to demonstrate differences in finger tapping frequency, used a more rigorous statistical approach. Finger tapping is a simple motor task and, as such, may be relatively immune to minor degrees of central nervous system depression in contrast to more complex tasks involving cognitive evaluation. Alternatively, the hypoxia in our experiments may have had relatively little effect on this motor function, whereas it may still have affected the sensory pathways involved in respiratory proprioception.
In our study, the equation describing the power relationship between
and PIP that included a TI component [ = k(PIP)nTIm]
did not improve the fit of the data provided by the simple linear equation relating and PIP [ = a(PIP) + b]. This should not be taken to imply that is
unaffected by changes in inspiratory duration. Previous researchers
have manipulated TI across a wide range of values and shown
that as TI increases, increases (17). The
reason why it did not appear to contribute to in our experiments is
probably due to the fact that TI varied little both during and between the experiments.
Possible relevance of results to respiratory arousals in sleep.
How respiratory loading is perceived to induce an arousal in sleep is
still poorly understood. We believe it is possible that the or
respiratory effort during wakefulness is related to the sensing of the
respiratory load during sleep. If we accept this premise, it is
possible to obtain some idea of how hypoxia might elevate the sensory
threshold required for arousal in sleep. Gleeson et al.
(13) found a threshold of respiratory effort for arousal in sleeping normal subjects equivalent to ~16.7 cmH2O of
peak-negative esophageal pressure equivalent, in our study, to a mask
PIP of ~15.3 cmH2O (transpulmonary pressure at PIP). We
calculated from our data that, to reach the level of associated
with a PIP of 15.3 cmH2O during normoxia, a 20-33%
increase in PIP would be required after 15-32 min of hypoxia.
These data imply that hypoxia during sleep could delay arousal until
the level of respiratory effort is increased by as much as one-third
above that necessary to produce arousal under normoxic conditions.
Clinical implications for sleep apnea, asthma, and chronic obstructive pulmonary disease. The impact of hypoxia on load evaluation is potentially significant for patients suffering from obstructive sleep apnea. This clinical population has been shown to arouse from sleep at much higher levels of respiratory effort than normal subjects (2). A possible contributor to this elevated arousal threshold is the depressant effect of hypoxia on the central nervous system. The perception of loaded breathing is reduced in asthmatic children (16) and adults (25) who suffer nearly fatal attacks and in chronic obstructive pulmonary disease patients (14). Our data suggest that such patients, apparently already at risk from a relative inability to perceive and react to an inspiratory challenge, may have perception distorted further during times of increased airway obstruction if hypoxia is present.
In summary, this study has shown that, in healthy, male subjects, sustained isocapnic hypoxia leads to a progressive decline in the of externally applied inspiratory resistive loads. This finding has
implications for defensive arousal responses from sleep (e.g., in
obstructive apnea) and for disorders such as asthma and chronic
obstructive pulmonary disease, in which the reduction in the perception
of inspiratory load, due to hypoxia, may cause delays in patients
accessing appropriate treatment.
| |
ACKNOWLEDGEMENTS |
|---|
Support for research was provided by The National Health and Medical Research Council (Australia) and the Asthma Foundation of South Australia.
| |
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: Address for reprint requests and other correspondence: R. D. McEvoy, Sleep Disorders Unit, Repatriation General Hospital, Daw Park SA 5041, Australia (E-mail: doug.mcevoy{at}rgh.sa.gov.au).
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. §1734 solely to indicate this fact.
Received 13 October 1998; accepted in final form 22 February 2000.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Berry, RB,
and
Gleeson K.
Respiratory arousal from sleep: mechanisms and significance.
Sleep
20:
654-675,
1997[ISI][Medline].
2.
Berry, RB,
Kouchi KG,
Der DE,
Dickel MJ,
and
Light RW.
Sleep apnea impairs the arousal responses to airway occlusion.
Chest
109:
1490-1496,
1996
3.
Berry, DT,
McConnell JW,
Phillips BA,
Carswell CM,
Lamb DG,
and
Prine BC.
Isocapnic hypoxemia and neuropsychological functioning.
J Clin Exp Neuropsychol
11:
241-251,
1989[ISI][Medline].
4.
Burdon, JG,
Killian KJ,
and
Campbell EJ.
Effect of ventilatory drive on the perceived magnitude of added loads to breathing.
J Appl Physiol
53:
901-907,
1982
5.
Burdon, JG,
Killian KJ,
Stubbing DG,
and
Campbell EJ.
Effect of background loads on the perception of added loads to breathing.
J Appl Physiol
54:
1222-1228,
1983
6.
Christie, MJ,
Chesher GB,
and
Bird KD.
The correlation between swim-stress induced antinociception and [3H] leu-enkaphalin to brain homogenates in mice.
Pharmacol Biochem Behav
15:
853-857,
1981[ISI][Medline].
7.
Christie, MJ,
Trisdikoon P,
and
Chesher GB.
Tolerance and cross tolerance with morphine resulting from physiological release of endogenous opiates.
Life Sci
31:
839-845,
1982[ISI][Medline].
8.
Chudler, EH,
and
Dong WK.
The role of the basal ganglia in nociception and pain.
Pain
60:
3-38,
1995[ISI][Medline].
9.
Easton, PA,
Slykerman LJ,
and
Anthonisen NR.
Ventilatory response to sustained hypoxia in normal adults.
J Appl Physiol
61:
906-911,
1986
10.
Easton, PA,
Slykerman LJ,
and
Anthonisen NR.
Recovery of the ventilatory response to hypoxia in normal adults.
J Appl Physiol
64:
521-528,
1988
11.
Farrell, P,
and
Gustafdon A.
Exercise stress and endogenous opiates.
In: Enkephalins and Endorphins, edited by Plotnikoff NP,
Faith RE,
Murgo AJ,
and Good RA.. New York: Plenum, 1986, p. 47-57.
12.
Findley, LJ,
Barth JT,
Powers DC,
Wihoit SC,
Boyd DG,
and
Suratt PM.
Cognitive impairment in patients with obstructive sleep apnea and associated hypoxemia.
Chest
90:
686-690,
1986
13.
Gleeson, K,
Zwillich CW,
and
White DP.
The influence of increasing ventilatory effort on arousal from sleep.
Am Rev Respir Dis
142:
295-300,
1990[ISI][Medline].
14.
Gottfried, SB,
Altose MD,
Kelson SG,
and
Cherniak NS.
Perception of changes in airflow resistance in obstructive pulmonary disorders.
Am Rev Respir Dis
124:
566-570,
1981[ISI][Medline].
15.
Haier, RJ,
Quaid K,
and
Mills JC.
Naloxone alters pain perception after jogging.
Psychiatry Res
5:
231-232,
1981[ISI][Medline].
16.
Kifle, Y,
Seng V,
and
Davenport PW.
Magnitude estimation of inspiratory resistive loads in children with life-threatening asthma.
Am J Respir Crit Care Med
156:
1530-1535,
1997
17.
Killian, KJ,
Bucens DD,
and
Campbell EJ.
Effect of breathing patterns on the perceived magnitude of added loads to breathing.
J Appl Physiol
52:
578-584,
1982
18.
Killian, KJ,
and
Campbell EJ.
Dyspnea in lung biology.
In: Health and Disease: The Thorax, edited by Roussos C,
and Macklem PT.. New York: Marcel Dekker, 1985, vol. 29, chapt. 26, p. 787-828.
19.
Killian, KJ,
Mahutte CK,
and
Campbell EJ.
Magnitude scaling of externally added loads to breathing.
Am Rev Respir Dis
123:
12-15,
1981[ISI][Medline].
20.
Lezac, MD.
Neuropsychological Assessment. New York: Oxford Univ. Press, 1995, p. 680.
21.
Ludbrook, J.
Repeated measurements and multiple comparisons in cardiovascular research.
Cardiovasc Res
28:
303-311,
1994
22.
Neubauer, JA,
Melton JE,
and
Edelman NH.
Modulation of respiration during brain hypoxia.
J Appl Physiol
68:
441-451,
1990
23.
Phillipson, EA,
Bowes G,
Sullivan CE,
and
Woolf GM.
The influence of sleep fragmentation on arousal and ventilatory responses to respiratory stimuli.
Sleep
3:
281-288,
1980[ISI][Medline].
24.
Phillipson, EA,
and
Sullivan CE.
Arousal: the forgotten response.
Am Rev Respir Dis
118:
807-809,
1978[ISI][Medline].
25.
Ruffin, RE,
Latimer KM,
and
Schembri DA.
Longitudinal study of near fatal asthma.
Chest
99:
77-83,
1991
26.
Sajkov, D,
Neill A,
Saunders NA,
and
McEvoy RD.
Comparison of effects of sustained isocapnic hypoxia on ventilation in men and women.
J Appl Physiol
83:
599-607,
1997
27.
Segerdahl, M,
Ekblom A,
and
Sollevi A.
The influence of adenosine, ketamine, and morphine on experimentally induced ischemic pain in healthy volunteers.
Anesth Analg
79:
787-791,
1994
28.
Sollevi, A.
Adenosine for pain control.
Acta Anesth Scand Suppl
110:
135-136,
1997.
29.
Solomon, GS,
Kay N,
and
Morley JE.
Endorphins a link between personality, stress, emotions, immunity and disease.
In: Enkephalins and Endorphins, edited by Plotnikoff NP,
Faith RE,
Murgo AJ,
and Good RA.. New York: Plenum, 1986, p. 129-144.
30.
Stevens, SS.
Neural events and the psychophysical law.
Science
170:
1043-1050,
1970
31.
Stevens, SS.
Issues in psychophysical measurement.
Psychol Rev
78:
426-450,
1971[ISI].
32.
Stevens, SS.
Psychophysics. New Brunswick, MA: Transaction Books, 1975, p. 30.
33.
Stevens, SS,
and
Gallanter EH.
Ratio scales and category scales for a dozen perceptual continua.
J Exp Psychol
54:
377-411,
1957[ISI][Medline].
34.
Stubbing, DG,
Ramsdale EH,
Killian KJ,
and
Campbell EJM
Psychophysics of inspiratory muscle force.
J Appl Physiol
54:
1216-1221,
1983
35.
Suzuki, A,
Nishimura M,
Yamamoto H,
Miyamoto K,
Kishi F,
and
Kawakami Y.
No effect of brain blood flow on ventilatory depression during sustained hypoxia.
J Appl Physiol
66:
1674-1678,
1989
This article has been cited by other articles:
|
|
 |
|
  D. J. Eckert, R. D. McEvoy, K. E. George, K. J. Thomson, and P. G. Catcheside Effects of hypoxia on genioglossus and scalene reflex responses to brief pulses of negative upper-airway pressure during wakefulness and sleep in healthy men J Appl Physiol, May 1, 2008; 104(5): 1426 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Hlavac, P. G. Catcheside, A. Adams, D. J. Eckert, and R. D. McEvoy The effects of hypoxia on load compensation during sustained incremental resistive loading in patients with obstructive sleep apnea J Appl Physiol, July 1, 2007; 103(1): 234 - 239. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. O'Donnell, R. B. Banzett, V. Carrieri-Kohlman, R. Casaburi, P. W. Davenport, S. C. Gandevia, A. F. Gelb, D. A. Mahler, and K. A. Webb Pathophysiology of Dyspnea in Chronic Obstructive Pulmonary Disease: A Roundtable Proceedings of the ATS, May 1, 2007; 4(2): 145 - 168. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Eckert, P. G. Catcheside, D. L. Stadler, R. McDonald, M. C. Hlavac, and R. D. McEvoy Acute Sustained Hypoxia Suppresses the Cough Reflex in Healthy Subjects Am. J. Respir. Crit. Care Med., March 1, 2006; 173(5): 506 - 511. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Eckert, P. G. Catcheside, R. McDonald, A. M. Adams, K. E. Webster, M. C. Hlavac, and R. D. McEvoy Sustained Hypoxia Depresses Sensory Processing of Respiratory Resistive Loads Am. J. Respir. Crit. Care Med., October 15, 2005; 172(8): 1047 - 1054. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Moosavi, R. B. Banzett, and J. P. Butler Time course of air hunger mirrors the biphasic ventilatory response to hypoxia J Appl Physiol, December 1, 2004; 97(6): 2098 - 2103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D.J. Eckert, P.G. Catcheside, and R.D. McEvoy Blunted sensation of dyspnoea and near fatal asthma Eur. Respir. J., August 1, 2004; 24(2): 197 - 199. [Full Text] [PDF]
|
|
|
|
|
|
D. J. Eckert, P. G. Catcheside, J. H. Smith, P. A. Frith, and R. D. McEvoy Hypoxia Suppresses Symptom Perception in Asthma Am. J. Respir. Crit. Care Med., June 1, 2004; 169(11): 1224 - 1230. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK |
