EQUINE EXERCISE PHYSIOLOGY 5
Equine vet. J., Suppl. 30 (1999) 107-113
Ventilation-perfusion relationships during exercise in
Standardbred trotters with red cell hypervolaemia
PIA FUNKQUIST*, P . D. WAGNER*, G. HEDENSTIERNAT, S. G. B. PERSSON and GoREL NYMAN
Department of Large Animal Clinical Sciences, Swedish University of Agricultural Sciences and tDepartment
of Clinical Physiology, Uppsala University Hospital, Uppsala, Sweden and #Department of Medicine-0623,
University of California, San Diego, La Jolla, California, USA.
Keywords: horse; red cell hypervolaemia; exercise; inert gases; pulmonary gas exchange
In order to evaluate the pulmonary gas exchange during
exercise in Standardbred
hypervolaemia (RCHV), 12 horses with RCHV were
compared with 9 normovolaemic (NV) horses. %2 and
were determined with an open bias flow system.
Cardiovascular and haemodynamic data were recorded
during exercise at 4 different speeds on a treadmill.
Pulmonary gas exchange was assessed by conventional
blood gas variables (arterial and mixed venous blood gas
teyions), and the ventilation-perfusion distribution
OA/Qwm estimated by the multiple inert gas elimination
technique. VA and AaDo2 were calculated. Dispersions of
perfusion and ventilation distribution (SDQ, SDV)
HR, RR, Q t , VO~, VA, log SDV, C(a-V)02 and lactate did
not differ between groups. The degree of hypoxaemia was
more pronounced in the RCHV than in the NV (Pa02 = 54
and 59 mmHg; AaDO2 = 41 and 34 mmHg in RCHV and NV,
respectively, at highest workload). Further, pH was lower in
the RCHV and Pacop and 0C02 was significantly higher in
the RCHV during the course of exercise (pH = 7.24 and 7.29;
Pac02 = 56 and 51 mmHg; VCO~
in RCHV and NV, respectively, at highest workload). The
Pa02 predicted from the VA/Q distribution was higher than
actually measured in blood during heavy exercise which may
suggest a certain diffusion limitation over the alveolar-
capillary membranes in both groups but there was no
difference between the 2 groups. The more pronounced
hypoxaemia observed in RCHV trotters was mainly caused
by increased VA/Q mismatch expressed as a significantly
increased log SDQ (0.78 and 0.45 in RCHV and NV,
respectively, at highest workload).
trotters with red cell
= 156 and 135 mvkg x min
Red cell hypervolaemia (RCHV) is commonly found in racing
Standardbred trotters examined for impaired racing performance
in Sweden (Persson 1967; 1968; 1983). The marked impairment
in performance in the RCHV trotters noticed by trainers and
'Author to whom correspondence should be addressed.
confmed by exercise tolerance testing has also been shown to
be accompanied by an evident drop in racing form (Persson and
Osterberg 1999). Further, the racing trotters with RCHV
frequently show exercise-induced pulmonary haemorrhage
(Persson 1982; Funkquist et al. 1994).
The exercise-induced hypoxaemia commonly observed in
healthy horses (Bayly et al. 1983; Thomton et al. 1983) has been
suggested to be a result of a considerable diffusion limitation
of oxygen, an increase in VA/Gmismatch and hypoventilation
(Wagner et al. 1989; Nyman et al. 1995; Seaman et al. 1995). It
has been reported that an increased circulating red cell volume,
caused by mobilisation of erythrocytes from the spleen during
exercise, may be a contributor to the exercise-induced
hypoxaemia (Persson et al. 1987). In the hypervolaemic horse,
the red cell volume is significantly greater than normal values for
racing trotters of the same age and sex (Persson 1967; Persson et
al. 1996). The effect of this increase on ventilation-perfusion
relationships and pulmonary gas exchange during exercise in
RCHV horses is not yet known.
The multiple inert gas elimination technique, developed by
Wagner et al. (1974b), for assessing the ventilation-perfusion
distribution (VA/@ has been used in pulmonary function studies
in both resting and exercising horses (Hedenstiema et al. 1987;
Wagner et al. 1989; Nyman et al. 1995). This method is also
applicable in this study.
The aim of this study was to evaluate the pulmonary gas
exchange during exercise in racing Standardbred t r o t t e r s with
RCHV and to compare these results with data from
normovolaemic (NV) horses.
Materials and methods
This study comprised horses with diagnosed RCHV and horses
with NV. Prior to the study, each horse underwent a clinical
physical examination, including routine haematological
examination and ECG at rest to exclude detectable diseases. The
upper respiratory airways were examined at rest with a flexible
fibre optic endoscope (CF-1OL)'. Lung biopsies were taken
through the tenth intercostal space about 20 cm above the ventral
lung border (Dahl et al. 1987) and no horse showed any
morphological sign of pulmonary disease.
' h o to 6 days before the haemodynamic study, the horses
108 Ventilation-perfusion relationships
TABLE 1: Blood volume data: the different blood volumes
determined at the standardised exercise test before the
study. Values presented are means i s.d.
% of pred
Red cell volume NV
72 f 7
98 * 11
138 f 18
Plasma volume NV
54 f 4
54 f 5
NV = normovolaemic horses (n = 9), RCHV = horses with red cell
hypervolaemia (n = 12).
performed a standardised submaximal exercise test to determine
the plasma volume, red cell volume, total blood volume and
working capacity (Persson 1967, 1983, 1997). Plasma, red cell
and total blood volumes were determined with Evans blue dye
dilution space and the post exercise packed cell volume (PCV),
as previously described (Persson 1967, 1986). The results from
the blood volume determinations are presented in Table 1.
Hypervolaemic (RCHV) group
Twelve racing Standardbred trotters with RCHV, 8 geldings
and 4 stallions, age 4-9 years (mean 6.4 years) and weighing
408-517 kg (mean 465 kg) were studied. These horses were
selected from the horses referred to the university clinic during a 2
year period for exercise tolerance testing. The horses were referred
to the clinic with a history of impaired performance, and in five
horses, the history also included recurrent epistaxis. All horses had
previously performed well in races. They all had different trainers,
and, consequently, their training schedules differed.
The selection was based on a measurement of the red cell
volume. Horses with a red cell volume deviating significantly
(>2 s.d.) from the reference values for normally performing
Standardbred racehorses of the corresponding sex and age
(Persson et al. 1996) were included. According to the
standardised exercise test, all horses had an impaired working
capacity in relation to their red cell volume (Persson and
Forssberg 1987; Persson 1997).
Normovolaemic (NV) group
Nine healthy, trained Standardbred trotters, 7 geldings and 2
stallions, age 3-12 years (mean 6.2 years) and weighing
395-535 kg (mean 484 kg) were used for reference. All horses
in the NV group were owned by the department at the time of the
study, but they had been in training and raced on the track until
they arrived at the department. The horses had been regularly
trained so that, at the time of the study, their physical fitness
corresponded to racing conditions. The training schedule
consisted of interval training 2 days a week and low-intensity
exercise one day a week.
The red cell volume was within the normal range for a horse
in training when age and sex were taken into account (Persson et
al. 1996). According to the standardised exercise test, all horses
had a normal working capacity in relation to their red cell
volume (Persson and Forssberg 1987; Persson 1997).
All catheterisations were performed under local anaesthesia with
the horses unsedated and standing. An arterial catheter (18 ga,
200 mm)2 was introduced percutaneously into the transverse
facial artery in 19 horses, and in the other 2 horses, a catheter
was inserted percutaneously with an introducer kit (8F
intr~ducer)~, into the previously raised carotid artery. The left
jugular vein was catheterised (8F introducer) for infusion of a
saline solution of dissolved inert gases. The right jugular vein
was used for 2 catheters, each inserted with an introducer kit (8F
introducer) and advanced to the main pulmonary artery. One of
these 2 catheters (8F, 1 .OO m)4, was used for sampling of mixed
venous blood. The second one (7F, 1.25 m)5 had a thermistor-tip
to allow measurement of pulmonary artery blood temperature.
Blood gas measurements and lactate analysis
Arterial and central venous blood, respectively, were obtained
anaerobically from the transverse facial or carotid artery and from
the pulmonary artery for measurements of blood gases and other
blood parameters. The samples were kept airtight on ice for 30
min until assayed with standard electrode technique (ABL 300)6.
Tonometry correction for horse blood was made. Measured
blood gases were corrected to the actual blood temperature
according to Kelman (1966, 1967). Arterial and mixed venous
pH, arterial oxygen and carbon dioxide tensions (Pa02, Paco2)
and mixed venous oxygen and carbon dioxide tensions (PVo2,
PVco2) were measured. Arterial and mixed venous oxygen
saturation and blood haemoglobin concentration (Hb) were
measured spectrophotometrically (Hemoxymeter OSM 3)6. The
alveolar-arterial oxygen difference (AaDo2) was calculated as
the difference between PA02 and Pa02. Oxygen content was
calculated as 1.39 x Hb (gfl) x saturation + 0.003 x Po2 (mmHg)
and then the arteriovenous O2 content difference (C(a-V )02) was
calculated. Packed cell volume (PCV) was measured by the
conventional technique. The
concentration (La) was determined with an enzymatic lactate
analyser (Analox GM-7)7.
arterial plasma lactate
Cardiac output (Qt) was computed through mass balance from
measured oxygen uptake and C(a-902 (Fick principle). Bipolar
ECG was recorded from surface electrodes to obtain HR.
Oxygen uptake and carbon dioxide output (VO~, VCO~) were
determined by collecting gas from the expired air in an open
flow system (flow-through system) without valves, as described
by Nyman et al. (1995). The gases were analysed for oxygen and
carbon dioxide content using a gas analyser8 integrated into
Oximeter 32009. The plastic tube between the face mask and a
gas-mixing steel drum was heated to prevent condensation of
water vapour, which would cause losses of the water soluble
gases. The gas samples for measuring %'02 and VCO~ were
drawn from the tube just before the fan.
Alveolar ventilation (\jT*) was calculated on basis of expired
C02 and arterial Pco2. The respiratory rate (RR) was determined
from end-tidal Co2 records.
The use of inert gas technique during exercise
The ventilation-perfusion distribution was estimated by the
multiple inert gas elimination technique (Wagner et al. 197413;
Evans and Wagner 1977). The 6 inert gases, sulphur
Pia Funkquist et al.
TABLE 2: General cardiopulmonary variables
Variable Group Walk Slow trot Moderate trot Fast trot Results ANOVA
C( a-B )02
0.79 f 0.12
0.80 f 0.13
74 f 20
78 f 9
37 f 3
42 f 4*
12.8 f 1.2
14.5 f 1.4*
0.3 f 0 . 1
0.5 f 0.2
7.9 f 0.9
8.5 f 1.4
173 f 53
216 f 54
63 i 19
63 f 6
205 f 66
263 f 36
93.8 i 5.3
89.0 f 5.5
2.0 f 7.3
7.8 f 6.5
43.5 f 1.1
44.7 f 2.5
7.44 f 0.02
7.43 f 0.02
99.0 f 0.7
98.4 f 0.4
57 f 7
60 f 5
47 f 8
48 f 7
0.92 f 0.28
0.83 f 0.14
142 f 14
134 f 9
44 f 3
49 f 3 '
15.5 f 0.9
16.9 f 1.2*
0 . 9 f 0.3
0 . 0 f 0.5
15.3 f 2 . 1
14.9 f 1.1
366 f 40
392 * 28
80 f 15
84 f 12
907 f 158
909 f 133
88.8 f 7.4
8 3 . 1 f 6.1
1 1 . 6 f 9.4
15.3 f 5 . 9
42.3 i 2 . 4
44.0 f 1.9
7.44 f 0.02
7.43 f 0.02
98.6 f 0.5
97.9 f 0.5
96 i 9
103 f 11
8 1 f 17
97 f 13'
0.92 f 0 . 3 1
0.94 f 0.13
180 f 9
177 f 11
49 f 3
55 f 3 '
17.0 f 1.0
19.1 f 1 . 0 '
2.6 f 0.9
2.5 f 1 .O
18.2 f 1.5
19.3 f 1 . 3
516 f 38
537 f 64
97 f 11
95 f 13
1463 f 267
1668 f 239
75.0 i 4.4
6 8 . 1 f 6.6
21.4 f 9.8
28.7 f 8.7
45.9 f 4.0
48.0 f 3.4
7.41 f 0.04
7.38 f 0.03
96.9 i 1 .O
95.2 f 1.9
132 i 18
129 f 19
135 f 25
156 f 1 7 '
1.07 f 0.16
2 1 1 f8
207 f 9
59 f 2 '
18.5 f 0.6
20.4 f 1 . O '
14.3 f 5.0
12.0 f 4.4
20.2 f 1 . 6
20.5 f 1 . 5
634 f 56
628 f 92
102 f 12
2246 i 534
5 9 . 1 f 4.5
54.0 f 6.6
34.4 f 8.4
41.4 f 7 . 1
51.4 i 5 . 3
56.2 f 5.7*
7.29 * 0.04
7.24 f 0.05*
89.2 f 4.0
84.4 f 5 . 9 '
P = 0.031t
P = 0.0048
P = 0.046
P = 0.048
P = 0.016
P = 0.022t
NV = normovolaemic horses (n = 9 ) , RCHV = horses with red cell hypervolaemia (n = 1 2 ) . Values presented are means i s.d. 30, =
oxygen consumption; Vco, = carbon dioxide production; R = respiratory exchange ratio; HR = heart rate; PCV = packed cell volume;
[Hb] haemoglobin; [La] = lactate; C(a-B)o, = arterio-venous oxygen content difference; Qt = cardiac output; RR = respiratory rate; VA =
alveolar ventilation; Pao, = arterial oxygen tension; AaDo, = alveolar-arterial oxygen difference; Paco, = arterial carbon dioxide tension;
SATa = arterial 0, saturation.
P values indicate the significance of differences between groups during the course of exercise. *Indicates a significant difference
(P4.05) between groups at respective workload according to post hoc analysis. tlndicates a significant interaction (P4.05), i.e. a
significant increase in the difference between groups with higher workload.
hexafluoride, ethane, cyclopropane, enflurane, diethyl ether and
acetone, were dissolved in isotonic sodium chloride solution
which was infused into the jugular vein. The infusion rate was
120 m l at walk and was gradually increased to 550 mUmin
during heavy exercise. After 10 min of infusion at the f i t level,
arterial and mixed venous blood samples were drawn, and mixed
expired gas was collected f r o m the heated mixing drum. This
infusion rate and time was used to increase the absolute gas
concentration of soluble gases in order to increase the signal-to-
noise ratio and to ensure a steady-state condition. Inert gas
samples during exercise were drawn during the last 45 s of each
exercise level, when a steady-state level of VO2 was attained, as
judged from expired gas data.
G a s concentrations in the blood samples and expirate were
110 Ventilation-perfusion relationships
TABLE 3: Inert gas data
Variable Group Walk Slow trot
2.49 * 0.50
3.06 f 0.40
Moderate trot Fast trot Results ANOVA
Mean V&Q NV
1.25 * 0.25
3.08 * 0.40
3.66 * 0.86
3.64 f 0.65
Mean vA/ Q,V NV
1.38 f 0.27
2.06 f 0.87
2.77 i 0.57
3.65 * 0.46
0.32 * 0.07
0.42 * 0.12
0.31 * 0.07
0.38 * 0.12
0.64 * 0.16
0.63 f 0.41
3.46 * 0.36
4.26 f 1.02
4.45 i 1.05
6.26 * 3.19
0.45 * 0.14
0.78 f 0 . 2 6 '
P = 0.0056
P = 0.0085
0.32 i 0.07
0.34 * 0.08
0.31 * 0.06
0.38 * 0 . 1 9
0.32 f 0.05
0.40 * 0.06
P = 0.0061
0.38 f 0.09
0.43 f 0.03
Log SDV NV
0.32 f 0.05
0.38 * 0.04
0.96 f 0.17
0.91 * 0.37
0.49 rt 0.11
0.47 f 0.14
0.52 f 0.50
21.3 * 8.6
22.8 i 7.3
Pred-meas Paoa NV
o * o
o * o
13.5 * 7.2
13.5 * 6.0
o * o
3.7 f 4.3
9.5 2 8.3
15.3 f 7.0
9.6 f 8.3
8 . 1 * 4.4
8.0 i 3.5
5.2 f 4 . 1
4.4 f 2.0
6.2 f 3.3
NR. = norrnovolaemic horses (n = 9), RCHV = horses with red cell hypervolaernia (n = 12). Values presented are means i s.d. Mean
$/Q,Q = mean of blood flow distribution; mean VA/ Q,V mean of ventilation distribution; Log SDQ = dispersion of blood flow distribution;
Log SDV = distribution of ventilation distribution; Pred-meas Pao, = predicted Pao2-rneasured Pao,; RSS = residual sum of squares.
P values indicate the significance of differences between groups during the course of exercise. *Indicates a significant difference
(P4.05) between groups at respective workload according to post hoc analysis. +Indicates a significant interaction (P4.05), i.e. a
significant increase in the difference between groups with higher workload.
measured by the method of Wagner et al. (1974a), using a gas
chromatograph (Hewlett Packard 5880A)'O. The arterialhixed
venous and mixed expiredmixed venous concentration ratios
(retention and excretion, respectively) were calculated for each
gas, and their solubility in blood was measured in each horse by
a 2-step procedure (Wagner et al. 1974b).
The exercise test was performed on a high-speed equine
treadmill" at 4 different workloads. Measurements during
walking (mean * s.d. 1.9 f 0.1 d s ) were made with the
treadmill horizontal, and at the 3 trotting speeds, slow, moderate
and fast trot (4.2 i 0.2, 6.8 f 0.3 and 8.9 * 0.3 d s ) , with the
treadmill at an inclination of 3.6".
The cardiorespiratory and inert gas samples were collected
over the last 45 s of each exercise level, when a steady state of
002 was attained, as judged from expired gas concentrations (at
a constant bias flow rate). Therefore, the average exercise times
at the different speeds were: walking, 11 min; slow trot, 5 min;
moderate trot, 4 min; and fast trot, 3 min.
Unless otherwise indicated, all data are presented as mean * s.d.
Analysis of variance (ANOVA) was used to examine differences
between RCHV and NV horses during the course of exercise.
Further analysis of interactions between the 2 groups of horses
and increasing speeds was carried out. When relevant according
to the ANOVA, post hoc analysis using the Tukey honest
significant difference (HSD) test was performed to compare the
two groups at a specific speed. Differences were considered
significant when P< 0.05.
Oxygen uptake (Table 2)
V O ~
higher in RCHV compared to NV and the difference increased
during the course of exercise. Further, no differences in 0 , were
calculated between groups.
was similar between the 2 groups. V C O ~
Oxygen transport variables (Table 2)
Both Hb and corresponding PCV values for RCHV horses were
significantly elevated over values for NV at all speed levels. No
differences in C(a-V)oz were found.
Cardiovascular and haemodynamic function (Table 2)
Cardiac output and HR did not differ between groups. Further,
pH was significantly lower during the course of exercise in
RCHV than in W.
Pulmonary gas exchange (Table 2, Fig 1)
Mean arterial Po2 corrected to body temperature was
significantly lower in RCHV, and PCO~
in RCHV than in NV (Fig 1). Arterial oxygen saturation was
significantly lower in RCHV compared to NV and the difference
increased during the course of exercise. The AaD02 increased
during the course of exercise in both groups and was
significantly greater in RCHV (Table 2).
was significantly higher
Ventilation-pelfusion relationships. (Table 3, Fig 2)
The residual sums of square of VA/Q distributions (RSS) were in
Pia Funkquist et al.
Log S W
o Pao, NV
0 Pao, RCHV
0 P a q NV
2 d s
.4 d s
7 d s
Fig I: Arterial oxygen tension (Pao,) and arterial carbon dioxide
tension (Paco,) during graded exercise. h V = normovolaemic horses
(n = 9). RCHV = horses with red cell hypervolaemia (n = 12 ). Mean
an acceptable range, as proposed by Wagner and West (1980),
and were similar over the entire exercise range and in
comparison with earlier studies in our department (Nyman et al.
1995). Therefore, the results from the multiple inert gas
elimination technique used for evaluating the ventilation-
perfusion relationships in the present study were technically
acceptable. The distributions of perfusion and ventilation,
presented on a log scale, were, in all cases, unimodal and narrow
up to 4 m/s. However, increased ventilation-perfusion mismatch,
reflected by an increase in log SDQ and log SDV, was seen in
both groups over the course of exercise (Fig 2). Mean log SDQ
was significantly higher in RCHV compared to NV and, further,
the difference increased during the course of exercise (Table 3).
In addition, mean o ' / Q
ratios of both blood flow and ventilation
were significantly higher in RCHV than in NV during the course
of exercise (Table 3). The intrapulmonary shunt was small in all
horses and did not differ between groups. There was
a higher Pa02 predicted from the VA/Q distribution than actually
measured in blood during moderate and heavy exercise. This
was seen in both groups and there was no difference between
RCHV and NV (Table 3).
Horses and protocol
To allow comparison of RCHV and NV horses, the horses were
exercised at equivalent speeds, and the agreement in HR, V02
and lactate levels between the 2 groups indicates that similar
work was performed. However, there might be some differences
in state of training between groups, and there also might be a
difference between horses in the RCHV group since the horses
had different trainers, each using varying training methods.
Earlier studies have shown that, at the chosen speeds, exercise
time and inclination of the treadmill, fit Standardbred trotters
exercise at approximately 10,40,70 and 95% of IQmax
et al. 1995).
Ventilation and oxygen uptake
Although si@icant increases in VCO~
over the values in NV, the VO~,
were observed in RCHV
VA and RR were similar. In the
Fig 2: Ventilation-perfusion (VdQ) inequaliv expressed by the second
moment of the perfusion distribution on a log scale (log SDQ) during
graded exercise. h V = nonnovolaemic horses (n = 9). RCHV = horses
with red cell hypervolaemia (n = 12). Mean f s.d.
present study, RCHV could maintain an adequate V O ~
graded submaximal exercise test, probably because of
compensatory mechanisms, such as a better oxygen delivery to
the tissue resulting from higher Hb concentrations. As expected,
the mean VA/Q ratio of both blood flow and ventilation shifted
to higher values with increasing workload. This greater increase
in ventilation than in cardiac output, resulting in a shift of mean
VA/Q ratios to higher values, is one of the most important
mechanisms for adjusting the pulmonary gas exchange during
exercise (Dantzker and D'Alonzo 1986). Surprisingly, both the
mean VA/Q ratio of blood flow and ventilation were higher in
RCHV than in NV.
The lower pH in RCHV was unexpected since measured
lactate levels and calculated bicarbonate concentrations were
similar between the groups during exercise. Interestingly, P a q
and Vc02 were higher in RCHV than in NV which could reflect
a different muscle metabolic response.
Pulmonary gas exchange
In accordance with other studies (Bayly et al. 1983, 1987;
Thomton et al. 1983; Persson et al. 1987; Nyman et al. 1995),
exercise-induced hypoxaemia was also seen in the horses in the
present study. Interestingly, a more severe hypoxaemia and
larger AaDo2 occurred in horses with RCHV. The Paco2
increased with speed in both groups and was also significantly
higher in the RCHV than in the NV (56 and 51 mmHg,
respectively, at highest speed). However, as earlier reported, the
fall in Pa02 was much greater than that associated with the
increase in Pac02; therefore, hypoventilation is responsible for
only a minor part of the exercise-induced hypoxaemia (Bayly et
al. 1989). The remaining cause of the arterial hypoxaemia may
be a result of mismatch of ventilation and perfusion in the lung,
right-to-left shunts or diffusion limitation (Bayly et al. 1989;
Wagner et al. 1989; Art et al. 1990; Nyman et al. 1995).
Shunt and G/Q inequality
Intrapulmonary shunting, computed from the multiple inert gas
elimination technique, was negligible under all conditions and
decreased with exercise. An important contributor to the
exercise-induced hypoxaemia was the increase in the dispersion
112 Ventilation-perfusion relationships
of the perfusion distribution. It has been shown that, for a given
level of perfusion, a wider dispersion of vA/Q ratios results in a
lower arterial Pao2 (West 1969). In agreement with an earlier
study on Standardbred trotters (Nyman et al. 1995), log SDQ
increased during the highest workload in both NV and RCHV.
Interestingly, log SDQ was significantly increased in RCHV
over that of NV during the course of exercise, and the difference
was most evident at the highest workload (log SDQ 0.78
and 0.45, respectively). At the highest workload, close to
V0zmax, vA/Q inequality computed from the multiple inert gas
elimination technique, as the difference between PA02 and
predicted Pao2, was responsible for 36% of the AaDo2 of
37 mmHg in NV and for 44% of the AaDo2 of 43 mmHg in
RCHV. The increase in vA/Q inequality during intense exercise
in both man and horse is suggested to be the result of a low-
grade interstitial oedema caused by increased fluid transudation
secondary to pulmonary hypertension (West et al. 1993; Birks et
al. 1997). Some authors have suggested that horses do not
develop pulmonary oedema during strenuous exercise (Bayly et
al. 1987; Lekeux and A r t 1994). However, in the work of West
et al. (1997), it was proposed that fluid is certainly moving out
of the capillaries and that the probable reason why oedema is not
clinically seen is due to the relatively short duration of a race
performed at maximal workload. As reported previously,
mean pulmonary arterial pressure was significantly higher in
the trotters with RCHV included in this study (range
94-152 mmHg) compared with NV (range 64-100 mmHg) and
exercise-induced pulmonary haemorrhage was detected in the
trachea in 11 horses, all belonging to the RCHV group
(Funkquist et al. 1999). West et al. (1993) have shown that stress
failure of pulmonary capillaries, including disruptions of the
capillary endothelial and alveolar epithelial layers, exists in
horses with exercise-induced pulmonary haemorrhage. In that
report, red blood cells were seen both in the alveolar space and
in the interstitium, and, in addition, interstitial oedema and fluid
in the alveolar space were observed.
In the absence of oxygen diffusion limitation, predicted Pa02
from inert gas data and the actual measured Pao2 values would
agree. As shown in earlier studies, no differences in predicted-
measured Pa02 were found during exercise below 70% of
v 0 2 ~ ~ ~ .
In line with previous reports from short-term intense
exercise (Wagner et al. 1989; Nyman et al. 1995; Seaman et al.
1995; Hopkins et al. 1998), diffusion limitation of oxygen was
the major contributor to the arterial hypoxaemia. During the
highest workload the diffusion limitation, calculated as the
difference between measured Pa02 and predicted Pao2, was
responsible for 58 and 53% of the AaDo2 in NV and RCHV
horses, respectively. The arterial oxygen saturation at the highest
workload was significantly lower in RCHV than in NV (84.4%
in RCHV and 89.2% in NV). It has been suggested that the mean
capillary transit time may be too short to allow a complete
diffusion equilibrium of oxygen during maximal exercise in
horses (Bayly et al. 1987; Persson et al. 1987; Constantinopol et
al. 1989). Further, the inverse relationship between Pa02 and
RCV has been suggested to be caused by a reduction in the
capillary transit time with higher RCV (Persson et al. 1987). To
estimate the mean capillary transit time during the highest
work load in the present study, the formula V(c)/Q t, where the
value for V(c) (total capillary blood volume) is based on
morphometric data from Standardbred trotters (Constantinopol
et al. 1989), was used. The calculated transit time, 386 ms (NV)
and 404 ms (RCHV), did not differ between the groups and
agreed with data from other reports (Bayly et al. 1987; Lekeux
and A r t 1994).
From the present study, it is concluded that, although
exercise-induced hypoxaemia and hypercapnia were found in
both normovolaemic and red cell hypervolaemic Standardbred
trotters, the changes were more pronounced in RCHV trotters. A
considerable amount of diffusion limitation contributed to the
hypoxaemia during short term intense exercise in both groups.
The more pronounced hypoxaemia observed in the RCHV
horses was caused mainly by an increase in VA/Q inequality,
evidenced as increased dispersion of perfusion distribution, log
SDQ. The significant mismatch in vA/Q together with the
presence of blood in trachea in 92% of the RCHV horses after
exercise support the theory that stress failure of the capillary
endothelial and alveolar epithelial layers results in exercise-induced
pulmonary haemorrhage and in a low-grade pulmonary oedema.
The authors gratefully acknowledge the expert technical
assistance of Eva-Maria Hedin, Hanieth Wagner, Ann-Marie
Lofgren, Karin Morgan, Karin Thulin, Kristina Karlstrom, Yvette
Eriksson, Marie Ederoth, Erik Granstrom and Bo Eriksson.
Further, we acknowledge the services of the patient technicians in
the large animal clinic. This work was supported by the AGA
Medical Research Fund and the C. August Carlsson Foundation.
'Olympus Optical Co., Japan.
*Viggo-Spectramed, Swindon, UK.
'Arrow Int. Inc., Reading, Pennsylvania, USA.
'Swan-Ganz, Edwards Lab., Santa Ana, California, USA
6Radiometer, Copenhagen, Denmark.
'Analox Instruments Ltd, London, UK.
'Servomex, Sussex, UK.
'Isler Bioengineering AG, Switzerland.
"Hewlett Packard, Palo Alto, California, USA.
"Sato, Uppsala, Sweden.
Art, T., Anderson, L., Woaks, A.J., Roberts, C., Butler, P., Snow, D.H. and Lekeux,
P. (1990) Mechanics of breathing during strenuous exercise in Thoroughbred
horses. Resp. Physiol. 82, 279-294.
Bayly, W.M., Grant, B.D., Breeze, R.G. and Kramer, J.W. (1983) The effects of
maximal exercise on acid-base balance and arterial blood gas tension in
Thoroughbred horses. In: Equine Exercise Physiology, Eds: D.H. Snow, S.G.B.
Penson and R.J. Rose, Granta Editions, Cambridge, pp 400-407.
Bayly, W.M., Schulz, D.A., Hodgson, D.R. and Gollnick, P.D. (1987) Ventilatory
response to exercise in horses with exercise-induced hypoxaemia. In: Equine
Exercise Physiology 2, Eds: J.R. Gillespie and N.E. Robinson, ICEEP
Publications, Davis, California. pp 172-182.
Bayly, W.M., Hodgson, D.R., Schulz, D.A., Dempsey, J.A. and Gollnick, P.D. (1989)
Exercise-induced hypercapnia in the horse. J. appl. Physiol. 67, 1958-1966.
Birks. E.K., Mathieu-Costello, O., Fu, Z., qler, W.S. and West, J.B. (1997) Very
high pressures are required to cause stress failure of pulmonary capillaries in
Thoroughbred racehorses. J. appl. Physiol. 82, 1584-1592.
Constantinopol, M., Jones, J.H., Weibel, E.R., Taylor, C.R., Lindholm, A. and
Karas, R.H. (1989) Oxygen transport during exercise in large mammals 11.
Oxygen uptake by the pulmonary gas exchanger. J. appl. Physiol. 67,871-878.
Dahl, L.G., Gillespie, J.R., Kallings, P., Persson, S.G.B. andThornton, J.R. (1987)
Effects of cold environment on exercise tolerance in the horse. In: Equine
Exercise Physzology 2, Eds: J.R. Gillespie and N.E. Robinson, ICEEP
Publications, Davis, California. pp 235-242.
Dantzker, D.R. and D'Alonzo, G.E. (1986) The effect of exercise on pulmonary
gas exchange in patients with severe chronic obstructive pulmonary disease.
Pia Funkquist et al.
Am. Rev. respi,: Dis. 134, 1135-1139.
Evans, J.W. and Wagner, P.D. (1977) Limits on VA/Q distributions from analysis
of experimental inert gas elimination. J. appl. Physiol. 42, 889-988.
Funkquist, P., Nyman, G. and Persson, S.G.B. (1994) Haemodynamic responses to
exercise in t r o t t e r s with red cell hypervolaemia and exercise induced
pulmonary haemorrhage. In: Proceedings of the 10th International Conference
of Racing Analysts and Veterinarians, R&W Publications, Newmarket, pp
Funkquist, P., Nyman, G. and Persson, S.G.B. (1999) Haemodynamic responses to
exercise in Standardbred trotters with red cell hypervolaemia. Equine vet. J. In
Hedenstierna, G., Nyman, G., Kvart, C. and Funkquist, B. (1987) Ventilation-
perfusion relationships in the standing horse: an inert gas elimination study.
Equine vet. J. 19, 514-519.
Hopkins. S.R., Bayly. W.M., Slocombe, R.F., Wagner, H. and Wagner, P.D. (1998)
Effect of prolonged heavy exercise on pulmonary gas exchange in horses. J.
appl. Physiol. 84, 1723-30.
Kelman, G.R. (1966) Digital computer subroutine for the conversion of oxygen
tension into saturation. J. appl. Physiol. 21, 1375.1376.
Kelman, G.R. (1967) Digital computer procedure for the conversion of Pco, into
blood CO, content. Respi,: Physiol. 3, 111-115.
Lekeux, P. and Art, T. (1994) The respiratory system: Anatomy, physiology and
training. In: The Athletic Horse, Eds: D.R. Hcdgson and R.J. Rose, W.B.
Saunders Co., Philadelphia, Pennsylvania. pp 79-127.
Nyman, G., Bjark, M.,
Funkquist, P., Persson, S.G.B. and Wagner, P.D. (1995)
Ventilation-perfusion relationships during graded exercise in the Standardbred
trotter. Equine vet. J., Suppl. 18, 63-69.
Persson, S.G.B. (1967) On blood volume and working capacity in horses. Acta Vet.
Scad., Suppl. 19, 1-189.
(1968) Blood volume, state of training and working capacity of
race horses. Equine vet. J. 1, 52-62.
Persson, S.G.B. (1982) Overtraning som orsak till nedsatt prestation hos travhastar
(Overtraining as a cause of impaired performance of trotters). Nord. Vet.
Congress, Copenhagen, pp 193-196 Abstract 14.
Persson, S.G.B. (1983) Evaluation of exercise tolerance and fimess in the
performance horse. In: Equine Exercise Physiology, Eds: D.H. Snow, S.G.B.
Persson and R.J. Rose, Granta Publications, Cambridge. pp 44-457.
Persson, S.G.B. (1986) Practical aspects of blood volume measurement: Procedure
for determination of total red cell volume (CV) in the horse. In: Proceedings
of the International Conference on Equine Sports Medicine. San Diego,
Calfornia. p 5 1,
Persson, S.G.B. (1997) Heart rate and blood lactate responses to submaximal
treadmill exercise in the normally performing Standardbred trotter - age and
sex variations and predictability from the total red blood cell volume. J. vet.
med. Ass. 44, 125-132.
Persson, S.G.B. and Forssberg, P. (1987) Exercise tolerance in Standardbred
trotters with T-wave abnormalities in the electrocardiogram. In: Equine
Exercise Physiology 2, Eds: J.R Gillespie and N.E. Robinson, ICEEP
Publications, Davis, California. pp 772-780.
Persson, S.G.B. and Osterbexg, I. (1999) Racing performance in red blood cell
hypervolaemic Standardbred trotters. Equine vet. J., Suppl. 30,617-620.
Persson, S.G.B., Kallings, P. and Ingvast-Larsson, C. (1987) Relationships between
arterial oxygen tensions and cardiocirculatory functions during submaximal
exercise in the horse. In: Equine Exercise Physiology 2. Eds: J.R Gillespie and
N.E. Robinson, ICEEP Publications, Davis, California. pp 161-171.
Persson, S.G.B., Funkquist, P. and Nyman G. (1996) Total blood volume in the
normally performing standardbred trotter: Age and sex variations. J. vet. med.
ASS. 43, 57-64.
Seaman, J., Erickson, B.K., Kubo, K., Hiraga, A., Kai, M., Yamaya, Y. and Wagner,
P.D. (1995) Exercise induced ventilatiodperfusion inequality in the horse.
Equine vet. J. 27, 104-109.
Thornton, J., Esdn-Gustavsson. B., Lindholm, A,, McMiken, D. and Persson. S.
(1983) Effects of training and detraining on oxygen uptake, cardiac output,
blood gas tensions, pH and lactate concentrations during and after exercise in
the horse. In: Equine Exercise Physiology, Eds: D.H. Snow, S.G.B. Persson
and R.J. Rose, Granta Editions, Cambridge, pp 470-486.
Wagner, P.D. and West, J.B. (1980) In: Ventilation-Perfusion Relationships i n
Pulmonary Gas Exchange, Ed: J.B West, Academic Press, New York.
Wagner, P.D., Laravuso, R.B., uhl, R.R. and West, J.B. (1974a) Continuous
distributions of ventilation-perfusion ratios: theory. J. appl. Physiol. 36,588-599.
Wagner, P.D., Naumann, P.F. and Laravuso, R.B. (1974b) Simultaneous
measurement of eight forein gastes in blood by gas chromatography. J. appl.
Wagner, P.D., Gillespie. J.R., Landgren, G.L., Fedde, M.R., Jones, B.W., DeBowes,
R.M., Pieschl, R.L. and Erickson, H.H. (1989) Mechanism of exercise-induced
hypoxaemia in horses. J. appl. Physiol. 66, 1227-1233.
West, J.B. (1969) Ventilation-perfusion inequality and over all gas exchange in
computer models of the lung. Respi,: Physiol. 7, 88-110.
West. J.B., Tyler, W.S., Birks, E.K. and Mathieu-Costello, 0. (1997) Exercised-
induced pulmonary haemorrhage. In: Proceedings of the 1997 Dubai
International Equine Symposium, Eds: N.W. Rantanen and M.L. Hauser,
Matthew R. Rantanen Design, California, USA. pp 353-368.
West, J.B., Mathieu-Costello, O., Jones, J.H., Birks, E.K., Logemann, R.B. Pascoe,
J.R. and Qler, W.S. (1993) Stress failure of pulmonary capillaries in
racehorses with exercise-induced pulmonary haemorrhage. J. appl. Physiol.