J Appl Physiol
90: 989–996, 2001.
Basophils and exercise-induced hypoxemia
in extreme athletes
PATRICK MUCCI,1 FABIENNE DURAND,2 BERNARD LEBEL,3
JEAN BOUSQUET,3 AND CHRISTIAN PRÉFAUT2
1
Laboratoire d’Analyse Multidisciplinaire des Pratiques Sportives, Unité de Formation
et de Recherche en Sciences et Techniques des Activités Physiques et Sportives de Liévin,
Université d’Artois, 62800 Lievin; 2Laboratoire de Physiologie des Interactions, Unité Propre
à la Recherche et à l’Enseignement Supérieur EA701, and 3Unité Institut National de la Santé
et de la Recherche Médicale 454, Hôpital Arnaud de Villeneuve, 34295 Montpellier, France
Received 5 November 1998; accepted in final form 20 September 2000
may induce hypoxemia
in highly trained endurance athletes (5, 7, 26, 28, 29),
who are referred to as “extreme athletes” (2, 29). The
mechanisms involved in the development of this exercise-induced hypoxemia (EIH) have long been debated,
and two major explanations have been proposed: 1) a
lack of compensatory hyperpnea (7, 16, 26) and/or 2) a
gas exchange alteration that may result from functionally based mechanisms during exercise (16, 35). This
latter may involve ventilation-perfusion (19, 33) and
diffusion alterations induced by an incomplete O2 equilibrium between alveolar gas and pulmonary capillary
blood as a result of a rapid red blood cell pulmonary
transit time (7) and/or a pulmonary interstitial and
peribronchial vascular edema during exhaustive exercise (9, 15, 16, 19, 29, 33, 35).
A previous study (2) showed that the drop in arterial
partial pressure of O2 (PaO2) during exhaustive exercise was associated with a concomitant increase in
histamine release (%H) in extreme athletes, whereas
there was no change in either PaO2 or histamine level in
the untrained control group. Histamine is widely acknowledged to be an inflammatory mediator that
causes increased microvascular permeability to macromolecules (38) and therefore increased transcapillary
fluid movement. It has therefore been suggested that
this increase in histamine release could be related to a
change in pulmonary fluid movement and thus to a gas
exchange alteration such as EIH (2, 30). The finding
that EIH can be partly inhibited by prior inhalation of
nedocromil sodium, a drug thought to act by inhibiting
the release of histamine (30), provided evidence that
this increase in histamine release during exercise may
be involved in the development of EIH. More recently,
it was reported that this increase in %H can be suppressed in association with an apparent change in
ventilation-perfusion distribution (9). An interesting
question concerns the origin of this increase in histamine release. Histamine is a well-known inflammatory
mediator and potential contributor to mild hypoxemia
(30, 38). The increase in histamine release associated
with EIH may thus be the consequence of an inflammatory process in the lung and/or in the peripheral
muscles (38) during incremental exercise. This increase may also be the consequence of an elevated
basophil number and/or an elevated histamine content
of these cells (1, 17, 24) in extreme athletes.
Histamine is essentially contained in mastocytes
and basophils, and most of the histamine in whole
Address for reprint requests and other correspondence: P. Mucci,
UFR STAPS de Liévin-Université d’Artois, Laboratoire d’Analyse
Multidisciplinaire des Pratiques Sportives, 62800 Liévin, France
(E-mail: [email protected]).
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.
healthy men; incremental exhaustive exercise; cation concentration
IT IS WELL KNOWN THAT EXERCISE
http://www.jap.org
8750-7587/01 $5.00 Copyright © 2001 the American Physiological Society
989
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
Mucci, Patrick, Fabienne Durand, Bernard Lebel,
Jean Bousquet, and Christian Préfaut. Basophils and
exercise-induced hypoxemia in extreme athletes. J Appl
Physiol 90: 989–996, 2001.—This study examined whether
the increase in histamine release (%H, i.e., plasma histamine
expressed as a percentage of whole blood histamine) associated with exercise-induced hypoxemia (EIH) is related to
high training-induced changes in basophil and osmolarity
factors in arterial blood. All parameters were measured in 20
endurance athletes, 11 of whom presented an EIH (HThyp)
and 9 of whom were nonhypoxemic (HTnor), and in 10 untrained control subjects (UT). Measurements were made at
rest, at the maximal workload of an incremental exhaustive
exercise test, and at the fifth minute of recovery. %H increased during exercise in HThyp (P ⬍ 0.01) but did not
increase significantly in HTnor and UT controls. The results
indicated that 1) osmolarity and Na⫹ and K⫹ concentrations
did not differ between the two trained groups and 2) the
basophil count and basophil histamine content did not differ
among groups. We concluded that the %H increase associated with EIH was not due to a training effect on these
parameters. The relatively low increase in histamine content
during exercise in HThyp in comparison to HTnor (P ⬍ 0.05)
and UT (P ⬍ 0.01) and the low recovery vs. resting basophil
count only in HThyp (P ⬍ 0.01) suggested an accentuated
exercise-induced basophil degranulation in the hypoxemic
athletes.
990
BASOPHILS AND HIGH TRAINING
Subjects
Exercise Testing
Thirty young healthy men participated in this study. The
anthropometric characteristics and training regimens are
presented in Table 1. None of the subjects reported respiratory or cardiac disease or hypertension or were known to be
suffering from any chronic disease. None was on regular
medication. None of the subjects had a history of asthma,
exercise-induced asthma, or atopic disease or showed signs of
allergic disease as detected with the Phadiatop test, a serologic test with a sensitivity of 90% and specificity of 98% (37).
All were nonsmoking and presented normal spirometric values. Before admittance to the study, all subjects were evaluated for cardiovascular health. Subjects having an abnormal
12-lead electrocardiogram (ECG) tracing or a supine blood
pressure greater than 160/100 Torr at rest were excluded
from the study. A preliminary maximal exercise test on a
cycle ergometer was then performed. Subjects were excluded
An incremental exhaustive test was performed on a calibrated cycle ergometer (Monark 860, Varberg, Sweden). The
initial power setting was 30 W for UT and 60 W for HT for 3
min, with successive increases of 30 W every minute except
at the end of the test, when the increase was smaller to be as
close as possible to V̇O2 max. Minute ventilation (V̇E), O2
uptake (V̇O2), and CO2 output (V̇CO2) were measured continuously by use of a breath-by-breath automatic exercise metabolic system (CPX, Medical Graphics, St. Paul, MN). The
data were averaged during the last 20 s of each load. To
ensure that V̇O2 max was attained, at least three of the following criteria had to be met: 1) a plateau of V̇O2 with the last
increase in work rate (“leveling-off ” criterion); 2) attainment
of age-predicted maximal heart rate [210 ⫺ (0.65 ⫻ age) ⫾
10%]; 3) a respiratory exchange ratio ⬎ 1.1; and 4) an inability to maintain the required pedaling frequency (60 rpm)
Table 1. Physical and physiological characteristics of subjects
HThyp
HTnor
UT
Age, yr
Height, cm
Weight, kg
FEV1, %
V̇O2 max,
ml 䡠 min⫺1 䡠 kg⫺1
Pmax, W
HRmax, beats/min
23.4 ⫾ 1.6
22.8 ⫾ 0.9
26.0 ⫾ 1.2
182.3 ⫾ 1.9
180.5 ⫾ 1.6
176.5 ⫾ 2.5
72.9 ⫾ 2.2
71.0 ⫾ 3.1
73.8 ⫾ 3.0
110.5 ⫾ 2.6*
110.0 ⫾ 3.3*
100.6 ⫾ 2.6
65.6 ⫾ 1.3†
65.8 ⫾ 1.9†
44.2 ⫾ 1.6
365.0 ⫾ 10.0†
365.0 ⫾ 8.7†
267.0 ⫾ 8.3
187.7 ⫾ 3.2
188.5 ⫾ 2.7
188.6 ⫾ 1.6
Values are means ⫾ SE. FEV1, forced expiratory volume in 1 s expressed as a percentage of predicted value; V̇O2 max, maximal oxygen
uptake; Pmax, maximal load achieved during incremental exhaustive exercise; HRmax, maximal heart rate achieve during incremental
exhaustive exercise; HThyp, highly trained hypoxemic group; HTnor, highly trained nonhypoxemic group; UT, untrained control group.
Significantly different in trained groups from untrained group: * P ⬍ 0.05; † P ⬍ 0.001.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
METHODS
from the study if they had ST segment depression in ECG or
significant arrhythmias. The study was approved by the
institutional ethics committee, and all subjects gave written
consent to participate after the design and risks of the study
had been described to them.
Athletes. Twenty male endurance-trained athletes (HT)
were studied. The criteria for selection were as follows: age
19–30 yr and a maximal O2 uptake (V̇O2 max) ⬎60
ml 䡠 min⫺1 䡠 kg⫺1. They were assigned to one of two groups.
The first group comprised nine extreme athletes (HThyp),
i.e., highly trained athletes who develop EIH. This was defined as a drop in PaO2 of at least 8 Torr from resting values,
corrected for temperature, that lasts for at least the last
three steps of an incremental exercise test (2, 9, 29, 30).
Three of the HThyp were triathletes, and the six others were
cyclists. They participated in regional or national competition, and all had been training regularly for an average of
4.6 ⫾ 0.5 yr. They trained 16.3 ⫾ 1.1 h/wk. The second
trained group was composed of nine control athletes (HTnor),
i.e., highly trained athletes who do not exhibit EIH. The maximal decrease in PaO2 in this group was 3 Torr. Three of the
HTnor were triathletes, and the six others were cyclists. They
had participated in regional or national competition for an
average of 3.6 ⫾ 0.8 yr. All trained regularly for an average
of 13.7 ⫾ 1.1 h/wk. The two remaining athletes exhibited
PaO2 decreases of 5 and 6 Torr, respectively, i.e., above the
definition of resting hypoxemia (31) and under the threshold
of 8 Torr. We did not include these subjects in either of the
athlete groups to maintain distinct group difference. Nevertheless, these subjects were used in correlational analysis to
determine relationships between dependent variables.
Control subjects. Ten untrained men (UT), aged 20–30 yr
(26.0 ⫾ 1.2 yr), composed the untrained control group. None
trained in endurance sports, although they had active lifestyles with an average of 2.3 ⫾ 0.4 h/wk of physical activity.
blood is derived from basophil polymorphonuclear leukocytes (basophils) (38). It has been reported that 1) a
positive relationship exists between histamine release
and basophil histamine content in both normal and
nonmedicated asthmatic subjects (1) and 2) in both
normal and asthmatic subjects, there is a close association between peripheral whole blood histamine concentration after exercise and circulating basophil
counts (17, 24).
Exercise and training produce a series of complex
physiological changes that may alter circulating leukocytes (8, 12, 14, 20). Given the findings to date, we
hypothesized that high endurance training in athletes
would induce an increase in the number of basophils
and/or the histamine content of these cells, which could
explain the increased histamine release observed during exercise. However, there are conflicting data about
the effect of exercise on basophil count, with some
authors reporting an increase (14, 17) and others reporting no significant change (12, 20, 24) between preand postexercise. To test our hypothesis, we measured
basophil count and basophil histamine content at rest
and during incremental exhaustive exercise in highly
trained athletes who have shown EIH, as well as in
highly trained athletes who do not present EIH and in
untrained control subjects. In addition, we examined
blood osmolarity and the concentrations of Na⫹ and
K⫹, noninflammatory parameters known to increase
with exercise (11) and to influence histamine release
(4, 10).
BASOPHILS AND HIGH TRAINING
despite maximal effort and verbal encouragement. A threelead ECG (DII, V2, V5) (Quinton Q 3000, Seattle, WA) was
monitored continuously during testing and recorded at the
end of every minute of the entire test to determine heart rate.
It was possible to monitor and record the nine other leads at
any time.
Blood Analysis
histamine (%H). The specific release of histamine was determined as %H ⫽ 100 ⫻ PH/TH, where PH is the plasma
histamine for a given testing level in each subject and TH is
the corresponding total histamine.
Basophil count. Basophil counts were made in anticoagulated blood samples (EDTA) taken at the same time as those
for histamine assay. The blood was stained with Alcian blue,
which is a highly specific, accurate, and reproducible method
for basophil counting based on the staining of basophil granules as described by Gilbert and Ornstein (13). Counts were
made in a Fuchs-Rosenthal counting chamber.
Histamine content. Because most histamine is derived
from basophils (38), the histamine content of the basophils
was determined as HC ⫽ (TH ⫺ PH)/BC, where BC is the
basophil count corresponding at the testing level of the
plasma histamine and the total histamine assays (1).
Protocol
The subjects performed three exercise tests at intervals of
a few weeks, all in the morning between 9 and 11 AM to avoid
diurnal variations. On the first day, clinical interviews and
examinations were conducted, and height, body mass, blood
pressure, resting ECG, and resting lung spirometry were
recorded. The first exercise test was then performed to allow
the subject to adapt to the ergosystem and laboratory environment, to detect any exercise-related cardiac anomaly and
to determine V̇O2 max. During the second exercise test, blood
samples were drawn from the brachial artery for blood gas,
basophil, and histamine measurements. Arterial samples
were drawn during the last 20 s of the third and fifth minutes
of rest, of the last load corresponding to V̇O2 max, and of the
fifth minute of recovery. During the third exercise test, tympanic temperature was measured at the times that corresponded to the blood sampling; i.e., we used the mean value
in temperature obtained during the last 20 s of the third and
fifth minutes of rest, of the last load corresponding to V̇O2 max,
and of the fifth minute of recovery. Similar physiological
maximal values were achieved during the last two exercise
tests in each subject. During every exercise test, subjects
were monitored for ventilatory variables and three-lead ECG
for the 5 min preceding the exercise test, from the beginning
to the end of testing, and for the 10 min of recovery. Ventilatory variables (V̇O2, V̇CO2, and V̇E) were averaged over an
integral number of breaths during the last 20 s of each
minute of rest, exercise, and recovery. This allowed the
calculation of Ai-aDO2 using the standard formula PAiO2 ⫽
PiO2 ⫺ PACO2 [FIO2 ⫹ (1 ⫺ FIO2/R)], where Ai-aDO2 is ideal alveolararterial difference in PO2, PAiO2 is ideal alveolar PO2, PiO2 is inspired PO2, FIO2 is the inspired O2 fraction, R is respiratory
exchange ratio, PACO2 is alveolar PCO2, and PACO2 ⫽ PaCO2 (3).
Statistical Analysis
The values are means ⫾ SE. Data concerning subject
characteristics were compared for homogeneity among UT,
HTnor, and HThyp by using a one-way ANOVA after verification of a normal distribution. The means of difference (⌬)
between each testing level for a given histamine parameter
were calculated and then compared between the three groups
with this test. A two-way ANOVA with repeated measurements was used to compare mean values across testing levels
between the three subject groups, and a one-way ANOVA
with repeated measurements was used to test for withingroup changes. Linear regression analysis was used to define
the relationship between two variables, and statistical significance was tested by using correlation coefficients. Statistical significance for all tests was set at P ⬍ 0.05, two-tailed.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
Blood samples were drawn from the brachial artery of the
nondominant arm with a 1-mm-diameter catheter (Seldicath;
Plastimed, Paris, France).
Blood gases. Arterial blood gases were immediately analyzed for PaO2, arterial partial pressure of CO2 (PaCO2), and
pH at 37°C using the appropriate electrode (IL Meter 1306,
Milan, Italy). Because of the rise in central temperature
during exercise testing, there was a risk of overestimating
the PaO2 decrease. Blood gases therefore needed to be corrected by core temperature estimated with a tympanic probe
(YSI Probe 43TA, Yellow Springs Instruments, Yellow
Springs, OH) at each testing level. We observed an average
change of 0.7 ⫾ 0.1°C during the incremental test with a
maximal value at the fifth minute of recovery. Because we
corrected blood gases by the appropriate temperature increase, we used the following rationale to define EIH: given
the variability in PaO2 measurement at rest, hypoxemia in
the resting condition should be defined as a significant reduction of 5 Torr compared with normal individual values
(31). Because the tympanic probe is accurate to estimate
increases in central blood temperature, i.e., ⬃0.1 or 0.2°C,
and can thus be used to correct values in blood gases (32), and
also because we needed to detect mild hypoxemia, we added
3 Torr to the 5-Torr resting drop of PaO2 for a final value of 8
Torr to ensure a rigorous definition of EIH. Therefore, a drop
in PaO2 of at least 8 Torr that lasts for at least the last three
steps of an incremental exercise test (9) is necessary to
establish the presence of hypoxemia during exercise.
Osmolarity analysis. Blood osmolarity was immediately
measured with a cryometric technique using an advanced
micro-osmometer (3Mo⫹, Advanced Instruments, Needham
Heights, MA). K⫹ and Na⫹ blood concentrations were determined by use of an appropriate automatic analyzer (Vitros
700 XRC, Johnson and Johnson Clinical Diagnostic).
Plasma histamine assay. Blood samples for plasma histamine determination were obtained in Vacutainer tubes (Becton Dickinson, Rutherford, NJ) containing EDTA and were
immediately placed on ice. Plasma was obtained via centrifugation at 900 g for 10 min at 4°C in a model PR-J refrigerated centrifuge (Beckman Instruments, Irvine, CA). Aliquots
were separated 0.5–1 cm above the cells to avoid picking up
any leukocytes, especially basophils, and were frozen at
⫺80°C until assay. Quantification of histamine was performed by use of an enzyme immunoassay kit (Immunotech,
Marseille, France).
Whole blood histamine. Blood samples for whole blood
histamine were obtained in heparinized Vacutainer tubes
exclusively and were placed on ice immediately after being
drawn. Fifty microliters of whole blood were added to 950 ␮l
of distilled water, and aliquots were frozen at ⫺80°C until
histamine measurement. For whole blood histamine quantification, the diluted blood was then frozen and thawed twice
for cell lysis. After final thawing and the removal of membranes by centrifugation at 900 g for 10 min at 4°C, quantification of histamine was performed in supernatant as described above.
Histamine release. As previously described (2, 9, 23, 30),
histamine release was expressed as a percentage of the total
991
992
BASOPHILS AND HIGH TRAINING
Table 2. PaO2 and D(Ai-a)O2 during incremental exhaustive exercise testing
PaO2, Torr
HThyp
HTnor
UT
D(Ai-a)O2, Torr
Rest
Max
Rec
Rest
Max
Rec
99.2 ⫾ 2.0
97.7 ⫾ 3.0
99.7 ⫾ 1.6
88.8 ⫾ 1.4†§
102.7 ⫾ 4.9
113.0 ⫾ 2.9
105.8 ⫾ 2.2‡
114.5 ⫾ 2.5
109.4 ⫾ 1.9
4.9 ⫾ 1.5
11.1 ⫾ 2.0
8.0 ⫾ 1.5
29.3 ⫾ 1.6†§
19.4 ⫾ 3.4*
8.8 ⫾ 2.5
5.1 ⫾ 0.9
9.0 ⫾ 2.3
4.5 ⫾ 1.0
Values are means ⫾ SE. PaO2, arterial partial pressure of O2; D(Ai-a)O2, ideal alveolar-arterial difference in partial pressure of O2; Max,
maximum of the incremental exercise; Rec, 5th min of recovery. Significantly different in trained groups from untrained group: * P ⬍ 0.01;
† P ⬍ 0.001. Significantly different in HThyp from HTnor: ‡ P ⬍ 0.05, § P ⬍ 0.01.
RESULTS
Physical and Maximal Data
Blood Gases
PaO2 and Ai-aDO2 are reported in Table 2. At rest,
there were no significant differences in the mean PaO2
and Ai-aDO2 values between the groups. However,
during exercise testing, HThyp showed lower values
of PaO2 and higher values of Ai-aDO2 than UT (P ⬍
0.001) and HTnor (P ⬍ 0.01) at the maximal workload.
During recovery, the data for PaO2 were significantly
different between HThyp and HTnor (P ⬍ 0.05).
Histamine
There were no significant differences among the
three groups in plasma histamine and whole blood
histamine at any testing level. All showed a significant
increase in plasma histamine between rest and maximal workload (P ⬍ 0.01 and P ⬍ 0.05, respectively) but
no difference between rest and recovery. Whole blood
histamine was significantly higher at maximal workload and recovery in HThyp (P ⬍ 0.05), HTnor (P ⬍ 0.01)
and UT (P ⬍ 0.001) than at rest (Table 3). Baseline %H
values did not differ significantly among the three
groups. However, at maximal exercise, there was a
significant rise in HThyp (P ⬍ 0.01) but not in the nonhypoxemic groups (Fig. 1A). During recovery, the %H
values did not differ from the resting values in any
group. There were significant correlations in HThyp
1) between the increase in %H and the drop in PaO2
between rest and maximal exercise (n ⫽ 9, r ⫽ 0.74,
P ⬍ 0.05) (Fig. 2) but not in HTnor (n ⫽ 9, r ⫽ 0.09, P ⫽
Basophils
The basophil counts did not differ significantly between the groups at any testing level. The HT groups
and UT showed a significant decrease during exercise
(P ⬍ 0.01), but during recovery only UT and HTnor
returned to resting values, and not HThyp (P ⬍ 0.05)
(Fig. 1B).
Histamine Content
Basophil histamine content was not significantly different among the three groups at any testing level.
There was a significant increase in histamine content
in HThyp, HTnor, and UT between rest and maximal
exercise (P ⬍ 0.05, P ⬍ 0.01, and P ⬍ 0.001, respectively) and between rest and postexercise recovery (P ⬍
0.05, P ⬍ 0.05, and P ⬍ 0.001, respectively) (Fig. 1C).
However, the increase in basophil histamine content
between rest and maximal exercise was significantly
lower in HThyp than in HTnor or UT (P ⬍ 0.05) (Fig. 3).
Osmolarity
There were increases in blood osmolarity (P ⬍
0.001), K⫹ (P ⬍ 0.001), and Na⫹ (P ⬍ 0.001) between
rest and maximal exercise in all groups. The values of
all these parameters returned to resting level at the
fifth minute postexercise, except for the Na⫹ concentration in UT (Table 4). The resting values were not
different in the three groups; however, the maximal
values were lower in HThyp than in UT for osmolarity
(P ⬍ 0.01), K⫹ (P ⬍ 0.05), and Na⫹ (P ⬍ 0.005). There
were no significant difference between HThyp and
HTnor. There were no significant correlations between
the changes in these parameters and the change in %H.
Table 3. Changes in plasma and whole blood histamine during incremental exhaustive exercise testing
Plasma Histamine, nM
HThyp
HTnor
UT
Whole Blood Histamine, nM
Rest
Max
Rec
Rest
Max
Rec
1.57 ⫾ 0.13
1.70 ⫾ 0.10
1.66 ⫾ 0.19
2.50 ⫾ 0.27†
2.39 ⫾ 0.20†
2.50 ⫾ 0.39*
1.55 ⫾ 0.20
1.91 ⫾ 0.39
2.26 ⫾ 0.30
779.0 ⫾ 67.0
732.0 ⫾ 78.5
707.5 ⫾ 90.8
840.0 ⫾ 60.6*
978.5 ⫾ 87.3†
963.9 ⫾ 87.4‡
948.0 ⫾ 66.4*
994.3 ⫾ 74.6†
1,012.2 ⫾ 121.8‡
Values are means ⫾ SE. Significantly different from resting values: * P ⬍ 0.05; † P ⬍ 0.01; ‡ P ⬍ 0.001.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
No significant differences were found among the
three groups for age, height, body mass, or maximal
heart rate. The highly trained groups showed significantly higher values than the untrained subjects for
forced expiratory volume in 1 s (P ⬍ 0.05), V̇O2 max (P ⬍
0.001), and maximal load (P ⬍ 0.001) (Table 1).
0.82) or UT (n ⫽ 10, r ⫽ 0.16, P ⫽ 0.65), and 2) between
the changes in %H and the increase in Ai-aDO2 between
rest and maximal exercise (n ⫽ 9, r ⫽ 0.68, P ⬍ 0.05)
but not in HTnor (n ⫽ 9, r ⫽ 0.16, P ⫽ 0.73) or UT (n ⫽
10, r ⫽ 0.25, P ⫽ 0.49).
BASOPHILS AND HIGH TRAINING
993
Fig. 2. Relationship between increase in histamine release (⌬%H ⫽
%H at maximal exercise ⫺ %H at rest) and drop in PaO2 (⌬PaO2 ⫽
PaO2 at maximal exercise ⫺ PaO2 at rest) during exercise in HThyp
(n ⫽ 9 subjects).
Because Anselme et al. (2) and Préfaut et al. (30)
found a strong correlation between %H and the drop in
PaO2 in extreme athletes, we used %H as the histamine
release index, rather than plasma histamine or whole
blood histamine. Nevertheless, the %H index takes
into account variations in both plasmatic and total
histamine (2, 20, 27).
We found an increase in %H during incremental
exhaustive exercise in extreme athletes but not in
untrained subjects, as did Anselme et al. (2). However,
in our extreme athletes the mean increase in %H was
below the 0.1–0.2% found by these authors; e.g., one of
our subjects showed a decrease in %H and two showed
an increase of ⬍0.02%. Also, we did not find significant differences in %H at maximal exercise between
HThyp and the nonhypoxemic groups. Despite this, we
found the same significant correlation between changes
Fig. 1. Histamine release (%H; A), basophil count (basophils; B), and
histamine content (HC; C) in untrained control subjects (UT; open
bars), in nonhypoxemic highly trained subjects (HTnor; hatched
bars), and in highly trained subjects with exercise-induced hypoxemia (HThyp; solid bars) at rest (Rest), at maximal exercise (Max),
and at the 5th min of recovery (Rec). Significantly higher than
resting values within groups: *P ⬍ 0.05, **P ⬍ 0.01, ***P ⬍ 0.001.
DISCUSSION
The results of the present investigation showed that
the increase in histamine release during exercise was
not associated with elevated basophil count, basophil
histamine content, or blood osmolarity in extreme athletes. However, we found a lower increase in histamine
content in these athletes during exercise than in control athletes and untrained subjects and a decrease in
basophil count in each group at maximal exercise but
only in extreme athletes at 5 min postexercise.
Fig. 3. Comparison among UT, HTnor, and HThyp of changes in
basophil histamine content (⌬HC) between maximal and resting
levels (Max-Rest) and between recovery and resting level (Rec-Rest).
Significantly higher than HThyp group, *P ⬍ 0.05.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
Methodology
994
BASOPHILS AND HIGH TRAINING
Table 4. Changes in blood osmolarity and K⫹ and Na⫹ blood concentrations during incremental exhaustive
exercise testing
K⫹ concentration, g/l
Osmolarity, osmol/kgH2O
HThyp
HTnor
UT
Na⫹ concentration, mmol/l
Rest
Max
Rec
Rest
Max
Rec
Rest
Max
Rec
287.2 ⫾ 1.0
288.1 ⫾ 1.6
286.2 ⫾ 0.8
303.0 ⫾ 2.9*
303.6 ⫾ 2.3*
311.3 ⫾ 1.5†
287.4 ⫾ 1.6
288.5 ⫾ 2.6
286.2 ⫾ 1.4
4.42 ⫾ 0.06
4.23 ⫾ 0.08
4.29 ⫾ 0.06
6.18 ⫾ 0.19†
6.40 ⫾ 0.30†
6.78 ⫾ 0.17†
4.22 ⫾ 0.05
3.98 ⫾ 0.11
4.12 ⫾ 0.06
138.5 ⫾ 0.7
140.3 ⫾ 1.1
139.1 ⫾ 0.5
144.7 ⫾ 0.8†
146.9 ⫾ 0.9†
148.8 ⫾ 0.9†
138.6 ⫾ 0.4
141.9 ⫾ 1.3
140.4 ⫾ 0.6*
Values are means ⫾ SE. Significantly different from resting values: * P ⬍ 0.01; † P ⬍ 0.001.
in %H and the drop in PaO2 during exercise (2, 30) in
HThyp.
Exercise-Induced Hypoxemia and Histamine Release
Histamine and Incremental Exhaustive Exercise
We noted that the incremental exhaustive exercise
induced the same relative increase in total and plasma
histamine in the nonhypoxemic subjects, as has been
previously described after exercise in normal subjects
(8, 17, 24). These subjects thus showed no change in
mean %H. In contrast, %H increased in the extreme
athletes because of an accentuated augmentation in
plasma histamine (2, 23); i.e., during exercise, the
plasma and the total histamine levels increased in
about the same proportion in the nonhypoxemic subjects, whereas the increase in plasma histamine was
relatively more accentuated (mean ⬃50%) than the
increase in total histamine in the hypoxemic athletes.
Histamine release may be related to a peripheral inflammatory process (38) during exercise (14) in highly
Basophil Count
This study demonstrated no significant difference in
basophil count in the highly trained endurance groups
compared with the untrained group. Nieman et al. (27)
originally reported this result for athletes in general at
rest, and we have enlarged upon this result with similar data under exercise conditions and in a group of
highly trained athletes exhibiting EIH. We may thus
conclude that high training does not induce a high
basophil count and that therefore basophil number
cannot explain the increase in %H associated with
EIH. Other factors must therefore intervene.
We describe, for the first time to our knowledge, a
decrease in basophil count during an incremental exhaustive exercise in healthy subjects. This is the first
study that has specifically investigated changes in basophil count during incremental exercise, with the sole
exception of a 40-yr-old study using a much longer
exercise (8). Our data suggest that incremental exhaustive testing was accompanied by a “basopenia” in
all groups. This exercise-induced basopenia may have
been due to the counting method, which is based on
specific staining of basophil granules and the detection
of the stained cells. It also may have been due to a
partial degranulation of the basophils during exercise
and the consequent undercounting that occurs in the
presence of lightly stained basophils that are difficult
to distinguish from nonbasophil leukocytes with nuclear staining. These indeterminate cells are normally
omitted from the basophil count because of this uncertainty (13). This hypothesis is supported by the fact
that this basopenia disappeared during the recovery
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
The development of EIH in our highly trained athletes is consistent with reports in the literature. Seven
athletes maintained PaO2 within 8–15 Torr of the resting values, another two showed 15- to 20-Torr reductions in PaO2. Moreover, this drop in PaO2 was accompanied by a greater maximal Ai-aDO2 in HThyp than in
UT or HTnor. This last result for extreme athletes has
previously been suggested, with a significant correlation between the changes in PaO2 and the rise in AiaDO2 (16, 35). In the present study, we studied for the
first time %H during exercise in nonhypoxemic highly
trained athletes and show that HTnor did not present a
significant change in %H during incremental exercise,
as was the case for the untrained subjects. This result
strengthens the hypothesis of a relationship between
EIH and %H. The lack of increased %H in HTnor was
associated with a maximal Ai-aDO2 that was lower than
in HThyp. This is consistent with the hypothesis of the
contribution of histamine release to the impaired gas
exchange in EIH. Nonetheless, our results cannot determine whether %H increase is an effect or a cause of
EIH. The finding of Préfaut et al. (30) (a partial inhibition in EIH with antihistamine inhalation) provided
evidence that this increase in histamine release during
exercise may be involved in the development of EIH,
but it is probable that histamine alone cannot induce
gas exchange alteration and that another phenomenon
must occur (30).
trained athletes. However, HTnor did not present an
increase in %H, although they reached the same high
absolute level of exercise as HThyp. Furthermore, the
gas exchange impairment in these extreme athletes
was significantly correlated with %H. This suggests
that an alternative explanation for the increased histamine release during exercise may be that part of the
histamine released comes from the activation of basophils or mastocytes located in the lung after lung
inflammation (38). This hypothesis is supported by two
facts: 1) we drew blood from the artery and not the vein
(17) and 2) recent research with bronchoalveolar lavages has suggested that intense exercise may impair
the integrity of the pulmonary blood-gas barrier in elite
athletes and induce the release of inflammatory mediators (18).
BASOPHILS AND HIGH TRAINING
Histamine Content
The results of this study showed no difference in
histamine content between nonhypoxemic subjects and
extreme athletes. We thus cannot explain the increase
in %H in the extreme athletes by an elevated histamine content induced by high training. No other research, to our knowledge, has been conducted on the
cellular content in histamine during exercise since the
investigation of Duner and Pernow in 1958 (8). We
found, as did they, an increase in cellular histamine in
the untrained control subjects; we also found this increase in the highly trained athletes. This increase in
both groups could be due to an increase in histamine
synthesis during exercise. The literature is consistent
with this hypothesis: 1) exercise leads to elevated
plasma IL-1 activity (6, 34); and 2) IL-1 can induce an
increase in histamine synthesis from basophils (36).
The increase in basophil histamine content observed
in the extreme athletes during exercise was lower than
in the untrained or nonhypoxemic highly trained subjects. One explanation of a low basophil histamine
content associated with a decreased basophil count and
an increased %H in highly trained athletes during
exercise is an increased degranulation of the basophil
cells. As we showed above, the augmented basophil %H
in extreme athletes cannot be due to an effect of high
training on basophils or noninflammatory hyperosmolar stimulation. It may, however, be explained by a
systemic (14) and/or pulmonary (18) inflammatory process induced by the high workload achieved by extreme
athletes.
In conclusion, we cannot explain the increase in
histamine release associated with EIH by elevated
basophil number or histamine content in extreme athletes. This suggests that another cell type may intervene: mast cells. However, the observation of a low
recovery basophil count and a relatively low increase in
basophil content at maximal exercise associated with
the increase in %H suggested increased basophil degranulation during exercise in extreme athletes. This
degranulation was not related to a hyperosmolar stimulation during exercise in these athletes, but it could
be related to an inflammatory process. We hypothesize
that the increase in histamine release in extreme athletes is at least partly due to an elevated degranulation
of circulating basophils in response to stimulation by
inflammatory factors such as inflammatory cytokines
during exercise. This speculation obviously requires
further investigation.
We thank M. Rongier, P. Kippelen, and C. Caillaud for technical
expertise and N. Blondel (Laboratoire d’Analyses Multidisciplinaires
des Pratiques Sportives, Université d’Artois, France) and C. Fabre
and S. Berthoin (Laboratoire d’Études de la Motricité Humaine,
Université Lille II, France) for technical assistance.
This work was supported by a grant from the Conseil Régional du
Languedoc-Roussillon.
REFERENCES
1. Akagi K and Townley RG. Spontaneous histamine release and
histamine content in normal subjects and subjects with asthma.
J Allergy Clin Immunol 83: 742–749, 1989.
2. Anselme F, Caillaud C, Couret I, Rossi M, and Préfaut C.
Histamine and exercise-induced hypoxemia in highly trained
athletes. J Appl Physiol 76: 127–132, 1994.
3. Anthonisen NR and Fleetham JA. Ventilation: total, alveolar,
and dead space. In: Handbook of Physiology. The Respiratory
System. Gas Exchange. Bethesda, MD: Am. Physiol. Soc., 1987,
sect. 3, vol. IV, chapt. 7, p. 113–129.
4. Beauvais F, Shimara T, Inoue I, Hieblot C, Burtin C, and
Benveniste J. Regulation of human basophil activation. II.
Histamine release is potentiated by K⫹ efflux and inhibited by
Na⫹ influx. J Immunol 148: 149–154, 1992.
5. Caillaud C, Anselme F, and Préfaut C. Effects of two successive maximal exercise tests on pulmonary gas exchange in athletes. Eur J Appl Physiol 74: 141–147, 1996.
6. Cannon JG, Evans WJ, Hughes VA, Meredith CN, and
Dinarello CA. Physiological mechanisms contributing to in-
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
period in the nonhypoxemic men. Indeed, as with most
investigations (12, 20, 22, 24), we report no change in
basophil number between pre- and postexercise basophil counts in normal subjects. Therefore, we prefer the
term pseudobasopenia. The drop in basophil count
likely reflects a basophil degranulation, which suggests an inflammatory process induced by incremental
exercise. The basophil count in the extreme athletes,
however, had not returned to resting level at the fifth
minute of recovery. This result may be related to the
fact that this population presents an increase in %H
during exercise, and we hypothesized that this low
basophil count after exercise may have been due to a
complete degranulation of basophils. Again, this result
suggests an inflammatory response induced by incremental exercise that is enhanced in extreme athletes.
This high basophil degranulation may also be caused
by an increased stimulation due to several factors, such
as hyperosmolarity (10). In agreement with the literature (11), we found an increase in osmolarity and
cations during exercise. Nevertheless, the extreme athletes did not present a higher mean maximal value in
osmolarity than the untrained or nonhypoxemic
trained subjects. This observation is not consistent
with the hypothesis of a hyperosmolar activation of
histamine release in these athletes. Moreover, the untrained controls showed a higher mean maximal value
than the extreme athletes in Na⫹ blood concentration,
which supports the hypothesis of an inhibitory effect of
external Na⫹ on histamine release (4). We did not,
however, find a significant difference between hypoxemic and nonhypoxemic groups nor a significant relationship between change in %H and Na⫹ concentration. Our data thus do not support the hypothesis of an
influence of noninflammatory factors such as hyperosmolarity or blood concentration in cations on the increase in histamine release in extreme athletes. Other
factors could induce an elevated histamine release,
such as cytokines. For example, interleukin-1 (IL-1) is
known to be an inflammatory mediator and a histamine-releasing factor (21) and to increase with exercise
(6, 34). Recently, our laboratory showed that histamine
releasability was elevated in highly trained athletes
compared with untrained subjects (25). Obviously, further studies are needed to elucidate the mechanisms
involved to confirm or eliminate inflammatory and
cytokine hypotheses.
995
996
7.
8.
9.
10.
11.
12.
13.
15.
16.
17.
18.
19.
20.
21.
22.
creased interleukin-1 secretion. J Appl Physiol 61: 1869–1874,
1986.
Dempsey JA, Hanson PG, and Henderson KS. Exerciseinduced arterial hypoxemia in healthy human subjects at sea
level. J Physiol (Lond) 355: 161–175, 1984.
Duner H and Pernow B. Histamine and leukocytes in blood
during muscular work in man. Scand J Clin Lab Invest 10:
394–396, 1958.
Durand F, Mucci P, Safont L, and Préfaut C. Effects of nitric
oxide inhalation on pulmonary gas exchange during exercise in
highly trained athletes. Acta Physiol Scand 165: 169–176, 1999.
Findlay SR, Dvorak AM, Sobotka AK, and Lichtenstein
LM. Hyperosmolar triggering of histamine release from human
basophils. J Clin Invest 67: 1604–1613, 1981.
Freund BJ, Claybaugh JR, Dice MS, and Hashiro GM.
Hormonal and vascular fluid responses to maximal exercise in
trained and untrained males. J Appl Physiol 63: 669–675, 1987.
Fry RW, Morton AR, Crawford GPM, and Keast D. Cell
numbers and in vitro responses of leucocyte and lymphocyte
subpopulations following maximal exercise and interval training
sessions of different intensities. Eur J Appl Physiol 64: 218–227,
1992.
Gilbert HS and Ornstein L. Basophil counting with a new
staining method using alcian blue. Blood 46: 279–286, 1975.
Gleeson M, Almey J, Brooks S, Cave R, Lewis A, and
Griffiths H. Haematological and acute-phase responses associated with delayed-onset muscle soreness in humans. Eur J Appl
Physiol 71: 137–142, 1995.
Hammond MD, Gale GE, Kapitan KS, Ries A, and Wagner
PD. Pulmonary gas exchange in humans during exercise at sea
level. J Appl Physiol 60: 1590–1598, 1986.
Harms CA, McClaran SR, Nickele GA, Pegelow DF, Nelson
WB, and Dempsey JA. Exercise-induced arterial hypoxaemia
in healthy young women. J Physiol (Lond) 507: 619–628, 1998.
Harries MG, Burge PS, O’Brien I, Cromwell O, and Pepys
J. Blood histamine level after exercise testing. Clin Allergy 9:
437–444, 1979.
Hopkins SR, Schoene RB, Henderson WR, Spragg RG,
Martin TR, and West JB. Intense exercise impairs the integrity of the pulmonary blood-gas barrier in elite athletes. Am J
Respir Crit Care Med 155: 1090–1094, 1997.
Hopkins SR, Gavin TP, Siafakas NM, Haseler LJ, Olfert
JM, Wagner H, and Wagner PD. Effect of prolonged heavy
exercise on pulmonary gas exchange in athletes. J Appl Physiol
85: 1523–1531, 1998.
Janssen GME, van Wersch JWJ, Kaiser V, and Does JMM.
White cell system changes associated with a training period of
18–20 months: a transverse and a longitudinal approach. Int
J Sports Med 10: S176–S180, 1989.
Kaplan A, Baeza M, Reddigari S, and Kuna P. Histaminereleasing factors. Int Arch Allergy Appl Immunol 94: 148–153,
1991.
Keen P, McCarthy DA, Passfield L, Shaker HAA, and Wade
AJ. Leucocyte and erythrocyte counts during a multi-stage cycling race (‘the Milk Race’). Br J Sports Med 29: 61–65, 1995.
23. Lichtenstein LM and Osler A. Studies of the mechanisms of
hypersensitivity phenomena. J Exp Med 120: 507–529, 1964.
24. Morgan DJR, Moodley I, Phillips MJ, and Davies RJ.
Plasma histamine in asthmatic and control subjects following
exercise: influence of circulating basophils and different assay
techniques. Thorax 38: 771–777, 1983.
25. Mucci P, Anselme F, Caillaud C, Couret I, Rossi M, and
Préfaut C. Basophil releasability during exercise in young
highly trained and older athletes. Med Sci Sports Exerc 31:
507–513, 1999.
26. Mucci P, Prioux J, Hayot M, Ramonatxo M, and Préfaut C.
Ventilation response to CO2 and exercise-induced hypoxemia in
master athletes. Eur J Appl Physiol 77: 343–351, 1998.
27. Nieman DC, Brendle D, Henson DA, Suttles J, Cook VD,
Warren BJ, Butterworth DE, Fagoaga OR, and NehlsenCannarella SL. Immune function in athletes versus nonathletes. Int J Sports Med 16: 329–333, 1995.
28. Powers S, Dodd S, Lawler J, Landry G, Kirtley M, McKnight T, and Grinton S. Incidence of exercise hypoxemia in
elite endurance athletes at sea level. Eur J Appl Physiol 58:
298–302, 1988.
29. Préfaut C, Anselme F, Caillaud C, and Masse-Biron J.
Exercise-induced hypoxemia in older athletes. J Appl Physiol 76:
120–126, 1994.
30. Préfaut C, Anselme-Poujol F, and Caillaud C. Inhibition of
histamine release by nedocromil sodium reduces exercise-induced hypoxemia in master athletes. Med Sci Sports Exerc 29:
10–16, 1997.
31. Préfaut C, Bourgouin Karaouni D, Ramonatxo M, Michel
FB, and Macabies J. A one-year double-blind follow-up of blood
gas tensions and haemodynamics in almitrine bismesylate therapy. Eur Respir J 1: 41–50, 1988.
32. Sato KT, Kane N, Soos G, Gisolfi C, Kondo N, Sato K.
Reexamination of tympanic membrane temperature as a core
temperature. J Appl Physiol 80: 1233–1239, 1996.
33. Schaffartzik W, Poole DC, Derion T, Tsukimoto K, Hogan
MC, Arcos JP, Debout DE, and Wagner PD. V̇A/Q̇ distribution during heavy exercise and recovery in humans: implications
for pulmonary edema. J Appl Physiol 72: 1657–1667, 1992.
34. Shephard RJ, Rhind S, and Shek PN. Exercise and the
immune system: natural killer cells, interleukins and related
responses. Sports Med 18: 340–369, 1994.
35. St Croix CM, Harms CA, McClaran SR, Nickele GA, Pegelow PF, Nelsen WB, and Dempsey JA. Effect of prior exercise
on exercise-induced arterial hypoxemia in young women. J Appl
Physiol 85: 1556–1563, 1998.
36. Subramanian N and Bray MA. Interleukin 1 releases histamine in human basophils and mast cells in vitro. J Immunol 138:
271–275, 1987.
37. Vervloet D, Haddi E, Tafforeau M, Lanteaume A, Kulling
G, and Charpin D. Reliability of respiratory symptoms to
diagnose atopy. Clin Exp Allergy 21: 733–737, 1991.
38. West GB. Histamine, mast cells and basophils. Chem Immunol
49: 121–142, 1990.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
14.
BASOPHILS AND HIGH TRAINING