Vol. 93, Issue 1, 195-200, July 2002
Modeling the acute- and late-phase responses to peripheral
airway cooling and desiccation
Michael S.
Davis1,
Chris
M.
Royer1,
Mark
Payton2, and
Brian
Buttress1
Departments of 1 Physiological Sciences and
2 Statistics, Oklahoma State University, Stillwater,
Oklahoma 74078
 |
ABSTRACT |
Acute bronchoconstriction
after isocapnic hyperpnea can be produced in most asthmatic
individuals. However, the existence of a late-phase response is
less certain. We used a canine model of isocapnic hyperpnea to test the
hypothesis that this discrepancy is due to differences in the challenge
threshold for the responses. Acute-phase and late-phase
bronchoconstriction was measured in nine dogs after peripheral airway
exposure to unconditioned air. Additionally, bronchoalveolar lavage
fluid (BALF) was obtained during the late-phase response. The
acute-phase response was a polynomial function with a decreasing slope
at higher challenges, whereas the late-phase response suggested that a
minimum threshold of challenge severity was needed to produce
late-phase bronchoconstriction. BALF leukocyte and eicosanoid
concentrations had linear relationships with challenge severity. Our
data support the hypothesis that acute- and late-phase posthyperpnea
responses have different dose-response relationships, a fact that may
explain the frequent lack of a late-phase response. However, our data
suggest that mild inflammation can be induced with relatively lower
challenge severity.
exercise-induced asthma; airway inflammation; bronchoconstriction
| |
INTRODUCTION |
HYPERPNEA INCREASES THE
REQUIREMENT for the respiratory mucosa to warm and humidify the
inspired air. Large increases in the minute ventilation, particularly
when the inspired air is substantially below body temperature, can
exceed the capacity of the upper airways to complete the conditioning
process, resulting in peripheral airway cooling and desiccation. The
acute-phase response to hyperpnea has been extensively characterized in
both humans and numerous laboratory animals (20). However,
whether this stimulus results in a second phase of airway obstruction
is the subject of considerable debate (23). Some
investigators argue that a late-phase response is a reproducible
feature of hyperpnea (2, 4, 15, 18, 25), whereas others
maintain that a late-phase response is not a feature per se of this
stimulus but rather an epiphenomenon attributable to other factors
(3, 5, 14, 17, 30, 31). The difficulty in clearly
demonstrating a late-phase response in humans stands in contrast to
data derived from a canine model of peripheral airway cooling and
desiccation, in which a late-phase response can be reliably measured
(9, 11). Given the consistency with which this animal
model reproduces the comparable acute-phase response in humans
(10), it seemed unlikely that the late-phase response
could be unique to dogs. Rather, we hypothesized that the apparent
differences between the canine model and the human data were the result
of experimental design, specifically that the late-phase response is a
feature of near-maximal challenge of the peripheral airways with
unconditioned air and that submaximal challenges that produce an
acute-phase response might not exceed the threshold necessary to
produce a late-phase response. Thus we sought to model the
dose-response effects of peripheral airway cooling and desiccation on
the expression of the acute-phase and late-phase responses.
| |
MATERIALS AND METHODS |
Study Design
Male mongrel dogs (~20 kg, >6 mo old) were each used in two
protocols. In each protocol, dogs were anesthetized as
previously described (7), a bronchoscope was advanced in
to the lower airways until wedged in each of five separate sublobar
airways, and baseline peripheral airway resistance (Rp) was
measured. Each airway was then challenged once with one of five
challenges: 200, 500, 1,000, 1,500, and 2,000 ml/min insufflation of
room-temperature dry air with 5% CO2 added for 5 min. The
acute-phase response after the challenge was then recorded for 10 min,
and the bronchoscope was removed. Approximately 4.5 h after
completion of the challenges, the dogs were reanesthetized and the
challenged airways were relocated with the bronchoscope by using airway
branching maps constructed at the beginning of the experiment. In the
first protocol, only Rp was measured during the late-phase response. In
the second protocol, bronchoalveolar lavage was performed in the
challenged airways, and the recovered fluid (BALF) was analyzed for
cellular and eicosanoid concentrations. All dogs were used in both
protocols, and the same airways received the same challenges in both protocols.
Methods
All animals were handled and maintained in accordance with the
Policy and Procedures Manual published by the Oklahoma State University
Institutional Animal Care and Use Committee. For the experiments, dogs
were anesthetized with an intravenous bolus of thiopental (25 mg/kg)
and fentanyl (1 µg/kg) and maintained under anesthesia with
additional intravenous boluses of fentanyl (1 µg/kg) every 15 min.
Depth of anesthesia was assessed by heart rate, blood pressure, canthal
reflex, and the presence of spontaneous movements and breathing. Dogs
were intubated and mechanically ventilated (17 ml/kg) with room air.
End-tidal CO2 was monitored with a CO2 analyzer
and maintained at ~4.5% by adjusting ventilator frequency.
Measurement of Rp.
A bronchoscope (5 mm OD) was inserted through an airtight portal of the
endotracheal tube and gently wedged into a sublobar segmental bronchus.
The suction port (1.2 mm) of the bronchoscope was connected to a
pressure transducer used to measure airway pressure in the subtended
lung segment (Pb). Compressed, dry, room-temperature 5%
CO2 in air was delivered at a rate of 200 ml/min through
the suction port of the bronchoscope into the wedged sublobar segment.
Rp was measured by stopping the ventilator during exhalation such that
the unobstructed areas of the lung equilibrated with atmospheric
pressure at functional residual capacity. Under these conditions, Pb
decays to a plateau at a pressure greater than the alveolar pressure
(atmospheric) in the surrounding unobstructed lung so that Rp = Pb · 200 ml
1 · min1.
BALF recovery and analysis.
Three 20-ml aliquots of warmed Hanks' buffered saline solution
were instilled through the bronchoscope into the wedged sublobar segment and then recovered by gentle aspiration. The recovered BALF was
pooled and the volume determined. Nucleated cells were counted by using
a hemacytometer. Differential counts of cytocentrifuged cells treated
with a modified Wright-Giemsa stain were done. The balance of the
sample was centrifuged, and the supernatant was filtered through a
C18 Sep-Pak exchange cartridge (Waters, Milford, MA) and
eluted with 4 ml of methanol. Aliquots of the eluted sample were dried
under vacuum, reconstituted with buffered saline, and analyzed with
commercially available ELISA kits for thromboxane B2,
prostaglandin E2 (Neogen, Lexington, KY), prostaglandin
F2
, prostaglandin D2, leukotriene
B4, and sulfidopeptide leukotrienes C4-E4 (Cayman Chemical, Ann Arbor, MI).
Analysis
Analysis of variance procedures were utilized to assess the
effect of challenge. Dog was included in the model to account for
dog-to-dog variability. PROC MIXED in PC SAS Version 8.2 was used.
Because challenge had meaningfully numeric levels, one degree of
freedom contrasts were calculated to account for the linear and
quadratic effects of challenge to response. Tests were performed using
the type III sums of squares associated with these contrasts to test
the significance of the quadratic effect given the linear effect in the
model. If the quadratic effect was not deemed significant, it was
removed and the linear-only effect model was fit. All contrasts and
effects were judged significant if their respective P value was <0.05.
| |
RESULTS |
The acute-phase response data were best modeled as a binomial
function expressing the response as a percentage of the prechallenge resistance and the challenge flow rate as the independent variable. The
constant for the independent variable "challenge2" was
negative, causing the slope of the model to decrease with increasing
challenge magnitude (Fig. 1). In
addition, all challenges produced significant acute-phase airway
obstruction compared with the control flow rate (200 ml/min).
The late-phase response data were also modeled as a binomial function
expressing the response as a percentage of the prechallenge resistance,
but in this model, the constant for the independent variable
challenge2 was positive, resulting in a model whose slope
becomes steeper with increasing challenge magnitude. Furthermore, only
the highest challenge (2,000 ml/min) produced late-phase obstruction
that was significantly different from control.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 1.
Dose-response effects of peripheral airway hyperpnea on
acute-phase and late-phase airway resistance. A: acute-phase
response is equal to  1.6 × 105(x2) + 0.11(x) + 97, r2 = 0.37, P = 0.0576; in which x = challenge flow
rate in ml/min. B: late-phase response is equal to 9.4 × 105(x2) 0.1(x) + 80, r2 = 0.40, P = 0.02. * Significantly different from control
(200 ml/min), P < 0.05.
|
|
All BALF cell and eicosanoid concentrations had significant linear
relationships to challenge severity (Figs.
2 and 3).
In the case of macrophages, lymphocytes, neutrophils, and
sulfidopeptide leukotrienes, the slope of the relationship was
sufficiently steep (relative to the variability of the values within
each category) to result in significant differences between BALF from
control lobes and that recovered from the two highest challenges. For eosinophils and the rest of the eicosanoids measured, only the BALF
from the airways with the highest challenge had significantly greater
eosinophil concentrations compared with control airways.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 2.
Dose-response effects of peripheral airway hyperpnea on late-phase
bronchoalveolar lavage fluid (BALF) nucleated cells. A:
macrophage regression slope = 0.037, r2 = 0.51, P = 0.0004. B: lymphocyte regression slope = 0.011, r2 = 0.59, P = 0.0057. C: neutrophil regression slope = 0.011, r2 = 0.46, P = 0.001. D: eosinophil regression slope = 0.0036, r2 = 0.42, P = 0.0009. * Significantly different from control (200 ml/min),
P < 0.05.
|
|
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 3.
Dose-response effects of peripheral airway hyperpnea on late-phase
BALF eicosanoids. A: leukotriene (LT) B4
regression slope = 1.8 × 103,
r2 = 0.45, P = 0.0239. B: LTC4 regression slope = 4.6 × 103, r2 = 0.46, P = 0.0010. C: thromboxane B2
(TxB2) regression slope = 1.1 × 102, r2 = 0.36, P = 0.0031. D: PGD2 regression
slope = 1.5 × 102,
r2 = 0.33, P = 0.0183. E: PGF2 regression slope = 2.7 × 102, r2 = 0.44, P = 0.0010. F: PGE2 regression
slope = 7.6 × 105,
r2 = 0.51, P = 0.0472. * Significantly different from control (200 ml/min),
P < 0.05.
|
|
| |
DISCUSSION |
The widespread acceptance of the late-phase response to peripheral
airway cooling and desiccation has been elusive because of the number
of studies in humans that have failed to demonstrate that a second
phase of airway obstruction occurred after exercise or isocapnic
hyperpnea or that such a phenomenon occurring was directly associated
with that challenge (3, 5, 14, 17, 30). The existence of a
dose-response relationship between challenge severity and the
acute-phase response, and the resulting need for standardization of
exercise tests, has been previously described (1, 24). In
this study, we have produced data to suggest that the late-phase
response has a distinctly different dose-response relationship from the
acute-phase response. These data help to reconcile the conflicting
findings of those former studies and the studies that have demonstrated
late-phase airway obstruction after exercise.
The mean airway wall temperature at any location in the respiratory
tract represents a specific net heat loss over time. During normal
ventilation, heat loss and recovery in the airway wall oscillate with
the respiratory cycle: heat is lost when the relatively cooler air
moves distally during inhalation, and some (but not all) of that heat
is recovered when the relatively warmer air moves proximally during
exhalation (21). The difference between the heat lost
during inhalation and the heat recovered during exhalation is the net
heat loss, and this net heat loss directly corresponds to the magnitude
of airway cooling. The amount of heat lost during ventilation is
correlated with the magnitude and duration of ventilation and is
inversely correlated with the temperature of the inspired air
(19). Thus equivalent respiratory heat loss can be
generated by short periods of hyperpnea with cold air and prolonged
periods of hyperpnea with warm air. Similarly, large increases in
ventilation with room temperature air may approximate the same
magnitude of respiratory heat loss as smaller increases in ventilation
with extremely cold air. From a practical standpoint, it is not
necessary to use cold air to produce peripheral airway cooling if the
magnitude of either of the other two variables is sufficiently great.
However, the magnitude of airway cooling is maximized with the use of
extremely cold air, large increases in minute ventilation, and long
duration of hyperpnea. The canine wedged bronchoscope model used in
this study results in less heat loss per volume of air during
"inhalation," but by eliminating heat recovery during exhalation,
it produces the same net heat loss, the same mean airway wall
temperature, and therefore the same initial stimulus as a comparable
oscillatory pattern of air movement (12). The validity of
this comparison is further supported by the fact that the canine
wedged-bronchoscope model consistently reproduces virtually every
aspect of the acute-phase response to hyperpnea in humans
(10).
Our study demonstrates that the relationship between hyperpnea
and the acute-phase response is not linear (Fig. 1). Rather, the slope
of the relationship tends to be steeper at lower values of challenge
severity, thus producing an easily detectable response at relatively
lower levels of challenge. The relationship between challenge severity
and the late-phase response is also not linear (Fig. 1), but in
contrast to the acute phase, the slope of this relationship tends to be
low at the lower challenge levels and only becomes steep under severe
challenge conditions. The implications of these relationships for human
studies of hyperpnea and exercise-induced asthma are that challenges
that produce acute-phase obstruction may not be sufficiently severe to
produce an episode of late-phase airway obstruction. Many human studies
of the hyperpnea-induced late-phase bronchoconstriction have titrated
the initial challenge to a percentage of the maximal heart rate for the
test subject (3, 5, 14, 17) or to a specific minimum
decrease in forced expiratory volume in 1 s during the acute-phase
response (31), whereas others have allowed the subjects to
inspire room-temperature air (3, 14). Our data suggest
that this approach may preclude the development of a detectable late
phase of airway obstruction because of insufficient respiratory heat
loss. To reliably demonstrate late-phase airway obstruction,
investigators need to select a study population that is capable of
achieving very high minute ventilations and/or use frigid inspired air
to produce a challenge that results in a maximal acute-phase response.
Thus the phenomenon of late-phase airway obstruction after exercise or
isocapnic hyperpnea may be a feature of only the most severe challenges.
So how severe must a challenge be to elicit late-phase airway
obstruction? It is widely accepted that a valid indicator of the
magnitude of the challenge is the magnitude of airway cooling and that
this is in turn related to the velocity of air and the temperature of
the air. Freed et al. (13) measured the airway wall
temperatures produced by varying the unidirectional flow of
room-temperature air through a bronchoscope wedged in a canine sublobar
bronchus and found that 500, 1,000, 1,500, and 2,000 ml/min produced
decreases in airway wall temperature of 0.7, 2.3, 4.9, and 7.0°C,
respectively. Based on similar modeling of the effects of inspired air
conditions and minute ventilation on airway temperature in humans by
McFadden et al. (22), these same decreases in airway wall
temperatures would be produced by increasing minute ventilation by
~10, 33, 70, and 100 l/min over resting minute ventilation when the
inspired air temperature is 18.6°C. Thus the highest challenge is
quite close to the maximal minute ventilation, and only strenuous
activity while breathing subfreezing air could be reasonably expected
to produce sufficient thermal challenge to elicit both an acute-phase
and late-phase bronchoconstriction response.
Despite the apparent requirement for a near-maximal challenge to
produce late-phase airway obstruction, airway inflammation appears to
be more readily elicited by submaximal challenges (Figs. 2 and 3). A
change in macrophage concentration was most readily induced, followed
by a change in neutrophil concentration. Induction of an influx of
eosinophils was most subtle, and thus is only reliably detectable with
the most severe challenges. However, it is important to note that, at
least within this scale of challenge severity, there was not an
apparent threshold of activation for the pathways leading to leukocyte
recruitment or mediator production. Rather, production of eicosanoids
and recruitment of inflammatory cells may be initiated at even the
milder challenges. This finding has considerable significance with
regard to winter athletes and other individuals who routinely exercise
in cold climates. A large number of epidemiologic studies have
suggested that repeated strenuous exercise in cold weather can lead to
airway inflammation and hyperreactivity similar to asthma, a syndrome
that has been referred to as ski asthma (16, 26-29).
This hypothesis is supported by recent studies demonstrating that
repeated challenge of canine peripheral airways with the highest
challenge used in this study (2,000 ml/min) can produce many of the
features of "ski asthma," including airway injury and inflammation
(6-8). However, the results of the present study suggest that repeated maximal challenges might not be necessary to produce airway inflammation. Rather, winter athletes may suffer airway injury and inflammation even during less strenuous activity.
The results of this study help explain the apparent discrepancies
between human and animal studies regarding the existence of a
late-phase response to exercise or isocapnic hyperpnea. Based on our
data, it appears that although an acute-phase response requires a
relatively mild stimulus, the development of late-phase airway
obstruction requires a considerably stronger stimulus and thus may not
be achievable in many human subjects. However, airway inflammation
secondary to exercise or isocapnic hyperpnea may be more readily
induced, a fact that is supported by the relatively high prevalence of
chronic airway inflammation in winter athletes.
| |
ACKNOWLEDGEMENTS |
This study was funded by Oklahoma Center for the Advancement of
Science and Technology Health Research Program Grant 99-001.
| |
FOOTNOTES |
Address for reprint requests and other correspondence:
M. S. Davis, 264 McElroy Hall, Stillwater, OK 74078 (E-mail:
msdavis{at}okstate.edu).
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published March 22, 2002;10.1152/japplphysiol.00074.2002
Received 29 January 2002; accepted in final form 20 March 2002.
| |
REFERENCES |
1.
Assoufi, BK,
Dally MB,
Newman-Taylor AJ,
and
Denison DM.
Cold air test: a simplified standard method for airway reactivity.
Bull Eur Physiopath Respir
22:
349-357,
1986[ISI][Medline].
2.
Bierman, CW,
Spiro SG,
and
Petheram I.
Characterization of the late response in exercise-induced asthma.
J Allergy Clin Immunol
74:
701-706,
1984[ISI][Medline].
3.
Boner, AL,
Vallone G,
Chiesa M,
Spezia E,
Fambri L,
and
Sette L.
Reproducibility of late phase pulmonary response to exercise and its relationship to bronchial hyperreactivity in children with chronic asthma.
Pediatr Pulmonol
14:
156-159,
1992[ISI][Medline].
4.
Chhabra, SK,
and
Ojha UC.
Late asthmatic response in exercise-induced asthma.
Ann Allergy Asthma Immunol
80:
323-327,
1998[ISI][Medline].
5.
Crimi, E,
Balbo A,
Milanese M,
Miadonna A,
Rossi GA,
and
Brusasco V.
Airway inflammation and occurrence of delayed bronchoconstriction in exercise-induced asthma.
Am Rev Respir Dis
146:
507-512,
1992[ISI][Medline].
6.
Davis, MS,
and
Freed AN.
Repeated dry air hyperventilation causes airway inflammation, edema, and mucosal squamous metaplasia (Abstract).
Am J Respir Crit Care Med
157:
A672,
1998.
7.
Davis, MS,
and
Freed AN.
Repetitive hyperventilation causes peripheral airway obstruction and eosinophilia.
Eur Respir J
14:
57-62,
1999[Abstract].
8.
Davis, MS,
and
Freed AN.
Repeated hyperventilation causes peripheral airways inflammation, hyperreactivity, and impaired bronchodilation in dogs.
Am J Respir Crit Care Med
164:
785-789,
2001[Abstract/Free Full Text].
9.
Davis, MS,
McCulloch S,
Myers T,
and
Freed AN.
Eicosanoids modulate hyperpnea-induced late phase airway obstruction and hyperreactivity in dogs.
Respir Physiol
129:
357-365,
2002[ISI][Medline].
10.
Freed, AN.
Experimental models of exercise-induced asthma.
In: Exercise-Induced Asthma, edited by McFadden ER.. New York: Dekker, 1999, p. 319-349. (Lung Biol. Health Dis. Ser.)
11.
Freed, AN,
and
Adkinson NF, Jr.
Dry air-induced late phase responses in the canine lung periphery.
Eur Respir J
3:
434-440,
1990[Abstract].
12.
Freed, AN,
Kelly LJ,
and
Menkes HA.
Airflow-induced bronchospasm: imbalance between airway cooling and airway drying.
Am Rev Respir Dis
136:
595-599,
1987[ISI][Medline].
13.
Freed, AN,
Wang D,
and
Menkes HA.
Dry-air induced constriction: effects of pharmacological intervention and temperature.
J Appl Physiol
62:
1794-1800,
1987[Abstract/Free Full Text].
14.
Hofstra, WB,
Sterk PJ,
Neijens HJ,
Kouwenberg JM,
Mulder PGH,
and
Duiverman EJ.
Occurrence of a late response to exercise in asthmatic children: multiple regression approach using time-matched baseline and histamine control days.
Eur Respir J
9:
1348-1355,
1996[Abstract].
15.
Iikura, Y,
Inui H,
Nagakura T,
and
Lee TH.
Factors predisposing to exercise-induced late asthmatic responses.
J Allergy Clin Immunol
75:
289-295,
1985.
16.
Karjalainen, EM,
Laitinen A,
Sue-Chu M,
Altraja A,
Larsson L,
Bjermer L,
and
Laitinen LA.
Airway inflammation and bronchial hyperresponsiveness in young elite cross-country skiers (Abstract).
Am J Respir Crit Care Med
155:
A962,
1997.
17.
Karjalainen, J.
Exercise response in 404 young men with asthma: no evidence for a late asthmatic reaction.
Thorax
46:
100-104,
1991[Abstract].
18.
Lee, TH,
Nagakura T,
Papageorgiou N,
Iikura Y,
and
Kay AB.
Exercise-induced late asthmatic reactions with neutrophil chemotactic activity.
N Engl J Med
308:
1502-1505,
1983[Abstract].
19.
McFadden, ER.
Respiratory heat exchange and asthma.
In: Exercise-Induced Asthma, edited by McFadden ER.. New York: Dekker, 1999, p. 47-76. (Lung Biol. Health Dis. Ser.)
20.
McFadden, ER, Jr
(Editor).
Exercise-Induced Asthma. New York: Dekker, 1999. (Lung Biol. Health Dis. Ser.)
21.
McFadden, ER, Jr,
and
Pichurko BM.
Intraairway thermal profiles during exercise and hyperventilation in normal man.
J Clin Invest
76:
1007-1010,
1985[ISI][Medline].
22.
McFadden, ER, Jr,
Pichurko BM,
Bowman HF,
Ingenito E,
Burns S,
Dowling N,
and
Solway J.
Thermal mapping of the airways in humans.
J Appl Physiol
58:
564-570,
1985[Abstract/Free Full Text].
23.
Peroni, DG,
and
Boner A.
Exercise-induced asthma: is there space for late-phase reactions.
Eur Respir J
9:
1335-1338,
1996[ISI][Medline].
24.
Silverman, M,
and
Anderson SD.
Standardization of exercise tests in asthmatic children.
Arch Dis Child
47:
882-889,
1972[ISI][Medline].
25.
Speelberg, B,
Panis EAH,
Bijl D,
van Herwaarden CLA,
and
Bruijnzeel PLB
Late asthmatic responses after exercise challenge are reproducible.
J Allergy Clin Immunol
87:
1128-1137,
1991[ISI][Medline].
26.
Sue-Chu, M,
Henricksen AH,
and
Bjermer L.
Non-invasive evaluation of lower airway inflammation in hyper-responsive elite cross-country skiers and asthmatics.
Respir Med
93:
719-725,
1999[ISI][Medline].
27.
Sue-Chu, M,
Karjalainen EM,
Altraja A,
Laitinen A,
Laitinen LA,
Noess AB,
Larsson L,
and
Bjermer L.
Lymphoid aggregates in endobronchial biopsies from young elite cross-country skiers.
Am J Respir Crit Care Med
158:
597-601,
1998[Abstract/Free Full Text].
28.
Sue-Chu, M,
Larsson L,
and
Bjermer L.
Prevalence of asthma in young cross-country skiers in central Scandinavia: differences between Norway and Sweden.
Respir Med
90:
99-105,
1996[ISI][Medline].
29.
Sue-Chu, M,
Larsson L,
Moen T,
Rennard SI,
and
Bjermer L.
Bronchscopy and bronchoalveolar lavage findings in cross-country skiers with and without "ski asthma."
Eur Respir J
13:
626-632,
1999[Abstract].
30.
Varga, EM,
Eber E,
and
Zach MS.
Cold air challenge for measuring airway reactivity in children: lack of a late asthmatic reaction.
Lung
168:
267-272,
1990[ISI][Medline].
31.
Zawadski, DK,
Lenner KA,
and
McFadden ER, Jr.
Re-examination of the late asthmatic response to exercise.
Am Rev Respir Dis
137:
837-841,
1988[ISI][Medline].
J APPL PHYSIOL 93(1):195-200
8750-7587/02 $5.00
Copyright © 2002 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
