INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

CHRONIC OBSTRUCTIVE pulmonary disease (COPD) is a common condition encountered in horses and is generally associated with chronic cough, bronchial hypersecretion, and dyspnea. It is a clinical entity specific to horses, somewhat similar to human asthma because of its allergic etiology but showing clinical similarities to human chronic airflow obstruction.

This respiratory disorder is most frequently observed in mature subjects, which are riding horses mainly used for jumping, endurance, eventing, and dressage, in other words horses working generally at submaximal intensities. Despite the fact that COPD horses rarely perform maximal or supramaximal exercise, reduced exercise tolerance and reluctance to work are often reported by the owners of these horses.

The reasons for this exercise intolerance remain unknown. Indeed, although several studies have been performed on resting horses to determine the functional consequences of chronic airflow obstruction, studies concerning the physiological adjustments during exercise in COPD horses are rare. Persson and Lindberg (24) have measured heart rate (HR), oxygen consumption (O₂), and expired minute ventilation (E) in exercising saddlebred and standardbred COPD horses and compared the results with healthy horses. They reported different adjustments according to the breed of the horses, i.e., increase in red blood cell volume in relation to body weight and restricted E in standardbreds and excessive HR response in saddlebreds. Unfortunately, they did not measure arterial blood-gas tensions during exercise, and an objective assessment of the pulmonary functional modifications of their COPD horses at rest was not reported. Another study by Kvart et al. (16) reported no significant differences in arterial blood-gas tensions in COPD horses exercising either after being kept on straw bedding and fed hay or after being fed silage and kept in a dust-free environment. In this study as in the previous one, no information about the clinical status of the horses at the time of the tests was available. In humans, a limited exercise capacity is also a major feature of chronic airflow obstruction. Many factors contribute to this limitation: some of them are now understood, and others are still the subject of controversy. It seems that human COPD subjects are limited during exercise by their ventilatory apparatus rather than by their cardiovascular system (25). Most of the work has led to the conclusion that the inadequacy of the ventilatory response of COPD patients (3), the mechanical constraints due to airflow obstruction (13), the excessive cost of breathing, i.e., >40% of the total exercise O₂ (17), and hence the fatigue of the inspiratory muscles (11), are probably the main determinants of reduction in their exercise tolerance. Moreover, in severely hypoxemic patients, O₂ availability apparently limits intramuscular oxidative metabolism because hypoxemia increases the net anaerobic metabolism for ATP production (19).

It is well known, however, that the respiratory adjustments to exercise of healthy humans and healthy horses are quite different. The same discrepancies may consequently be expected as regards the exercise-induced adaptations in subjects suffering from chronic airflow obstruction, and, therefore, it would be hazardous to extrapolate the observations made in humans to the horse.

The present study was conducted to understand better the mechanisms leading to the decrease in exercise capacity observed in COPD horses, namely, to assess whether an acute COPD crisis may impair the gas exchange and reduce the horse's O₂ and/or E during a submaximal exercise. Therefore, ventilatory and cardiorespiratory measurements as well as arterial and mixed venous blood-gas tensions were measured during a standardized treadmill exercise test in COPD horses while they were either in clinical remission (R) or in acute crisis (C).

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Horses

For 1 mo, they were housed in well-ventilated stables on wood shavings and received good-quality oats as concentrates and grassilage as forage. They did not receive any treatment during the 15 days preceding the experiments. During this period, they were trained daily on the treadmill and regularly accustomed to the laboratory procedures.

The training aimed mainly at having the horses calm and in confidence on the treadmill rather than to improve their fitness. Therefore, it consisted in 20 min of work per day, with 5 min of walking, 10 min of trotting with an increasing slope from 0 to 10%, 2 min of galloping (8 m/s at 0% slope), and 3 min of walking again on a flat treadmill.

Measurements

Mechanics of breathing. Before the tests, and to control whether they were in the appropriate clinical state, the horses underwent an evaluation of their mechanics of breathing. Esophageal pressure was measured by means of an esophageal balloon catheter made from a condom sealed over the end of a polyethylene catheter (4 mm ID, 6 mm OD, 220 cm long) positioned with its tip in the middle thoracic esophagus and connected to a pressure transducer (Bentley, Trantec M800, Medical Electronic Construction, Charleroi, Belgium). Respiratory airflow was simultaneously measured with a Fleish pneumotachograph no. 4 mounted on a face mask and coupled to a differential pressure transducer (Valydine M1 45, Validyne Engineering, Northridge, CA) with two identical catheters (4 mm ID, 6 mm OD, 220 cm long). The parameters of mechanics of breathing were immediately calculated by a computerized system (Hemodynamic Respiratory System, Medical Electronic Construction). Calibration of volume was performed with a 2-liter pump, of airflow with a rotameter, and of pressure with a water manometer. Technical details are reported elsewhere (2).

Blood measurements. The transverse facial artery was catheterized with a 20-gauge catheter (Baxter, Brussels, Belgium) secured to the skin with glue. A 110-cm plastic extension line was attached to the catheter to allow blood to be drawn at a distance from the horse. The catheter and the extension were flushed with heparinized saline between sampling to maintain patency. Immediately before the collection of each sample, 10 ml of blood and saline were drawn and discarded. Arterial blood samples were withdrawn during the last 5 s of each step of the test into 2-ml syringes, the dead space of which was filled with sodium heparin (10,000 IU/ml). The syringes were capped and stored in crushed ice until analysis, which was performed within a maximum of 15 min after collection.

O₂ and carbon dioxide output ( CO₂ ). A mass spectrometer (MGA 2000, Case, Biggin Hill, Kent, UK) was used to sample air in one flow tube. The sampling capillary was positioned 2 cm from the open end of the tube and continually measured O₂ and CO₂ concentrations in the inspired and expired respiratory gases on a breath-by-breath basis. Airflow signals from each nostril and respiratory O₂ and CO₂ concentration signals underwent analog-to-digital conversion and were recorded by using a data-acquisition system. Tidal volume (VT), respiratory frequency (f), E (BTPS), O₂ (STPD) and CO₂ (STPD) were instantaneously calculated by online and breath-by breath computer analysis (Case, Biggin Hill, Kent, UK). The concept of this method of determination of O₂ was described for the very first time by Beaver et al. (5). The accuracy of the method of flow measurements by ultrasonic pneumotachographs (7) as well as calculation of O₂ has been previously assessed (8).

Measurement of HR. HR values were recorded using a horse tester (Apparatus for Medical Guidance, Dinant, Belgium) throughout each investigation and recovery period: HR was calculated and recorded during consecutive periods of 5 s. After each experiment, the stored data were displayed on a microcomputer.

Experimental Protocol

1

1

Run up to fatigue. Before the first test and after the second one, a rapid incremental test was performed on the treadmill with a slope at 0% to determine maximal O₂ (O_{2 max}) in R and in C. A 5-min warm-up period at 3.0 m/s was followed by 1-min stages at 4, 8, 9, 10, and 11 m/s. The O₂ reached when the horses stopped running despite encouragement was considered as peak O₂ (O_{2 peak}) rather than as O_{2 max}, because no plateau was observed and HR was most probably not maximal. The numbers of steps completed during this test was expressed as the number of the complete steps plus a decimal equivalent on the basis of the number of seconds duration of the last step when the exercise was terminated before completion of a full step.

Standardized submaximal test (SST). The horses were catheterized and instrumented out of the laboratory. The tests were performed on a treadmill (Equispeed, Versailles, MI) located in a laboratory where the temperature and the relative humidity were kept constant (15°C and 55-60%, respectively). After an 8-min warm-up (5-min walk and 3-min trot on the level), the SST consisted of 7-min exercise of increasing intensity at 4.2 m/s, with a slope of 2, 4, 6, 8 (1 min each for each step), and 10% (3 min). The treadmill was then lowered to 0%, and the horses trotted for 2 min more before being stopped.

Calculations and Statistical Analysis

On the paper recording at 50 mm/s, Ppl_max, i.e., pleural pressure (Ppl) changes between the minimal and maximal Ppl; maximal airflow () changes, i.e., the flow changes between peak inspiratory and expiratory flows; the mechanical work of breathing (Wrm), i.e., the area enclosed in the Ppl/VT loop; and RL, i.e., the ratio of the changes to the Ppl changes at isovolume 50%, were calculated (1).

Alveolar ventilation (A) was estimated from the ratio of CO₂ to Pa_CO2 (expressed as a fractional gas concentration). In addition, the physiological dead space volume-to-tidal volume ratio (VD/VT) was calculated. Alveolar PO₂ (PA_O2) was calculated by the ideal alveolar equation.

was calculated with the Fick equation and stroke volume was estimated by dividing by HR.

An analysis of variance (repeated measures) was used to assess significant differences between the values obtained in R and in C.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Pulmonary Function Tests at Rest

Run Up to Fatigue

1

1

1

1

Standardized Submaximal Test

The A in C was maintained at a level almost similar to the level observed in R, whereas the VD/VT ratio and dead space ventilation (D) were significantly lower in C. Last, RL, Ppl_max, and Wrm per cycle, per liter, per minute, and per liter of O₂ consumed were significantly higher in C condition. When the horses were in C, RL decreased significantly from rest to exercise.

Table 3 gives the results of the blood analyses and particularly the PO₂ and PCO₂. The horses were significantly more hypoxemic and hypercapnic when exercised in C. Additionally, was lower and higher in C condition. The arterial-mixed venous differences in PCO₂ and the alveolar-arterial difference in PO₂ were the same in both conditions, while the arterial-mixed venous difference in PO₂ was lower in crisis (Fig. 1). Hb saturation was lower in C, but, because of a higher arterial concentration in Hb, the Ca_O2 was not significantly different.

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Alveolar (PAO2), arterial (PaO2), and mixed venous () PO2 values in horses with chronic obstructive pulmonary disease, in clinical remission (A) and in acute crisis (B), during a standardized treadmill exercise test of increasing intensity (horses were trotting at 4.2 m/s, and slope of treadmill was progressively increased). Values are means ± SE for 5 horses. Gray areas, PaO2- differences; hatched areas, PAO2-PaO2 alveolar-arterial PO2 differences. When horses were in crisis, at any time of the test, PAO2, PaO2, and were significantly lower (analysis of variance, P < 0.05) compared with remission. PAO2-PaO2 difference was the same in both conditions, whereas PaO2- difference was significantly lower in crisis in comparison with remission. 10a, 10b, 10c: 1st, 2nd, 3rd min at 10%, respectively.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

As suspected, most of the exercise-induced adjustments observed in COPD horses were totally different from those previously reported in humans. In exercising COPD men, two clinical syndromes have been determined according to whether pulmonary emphysema or chronic bronchiolitis is the predominant factor, i.e., subjects are either hyperoxic or "pink and puffing," or hypoxemic or "blue and bloated," respectively (13). In horses, diffuse pulmonary emphysema (which induces hyperoxia during exercise) is rarely observed: our horses were all hypoxemic. In COPD men, E at a given O₂ is increased (3). This increase in E is achieved by an increase in flow and f while VT is maintained constant (11). Conversely, in the present study, the COPD horses in C had a lower E, mainly because they had a lower f. In humans, ventilatory equivalents for O₂ and CO₂ as well as VD/VT are reported to increase (14), whereas the opposite was observed in our horses.

The plasma lactate concentration was slightly but significantly higher in the COPD horses in C than in R. Because lactic acid ultimately results in CO₂, this could partly explain why CO₂ was significantly higher during C. However, this difference in lactate is too low to be responsible for the entire increase in CO₂ in C. An increase in lactate has also been observed in exercising humans with COPD (18). In humans, an impairment of oxidative phosphorylation, an early activation of anaerobic glycolysis, and a decrease in activities of mitochondrial enzymes have been put forward to explain this observation (18), although a number of other factors, such as inappropriate O₂ delivery to the working muscles or reduction in lactate clearance, may also be involved in this increase in lactate production. Other authors have also suggested that this phenomenon could be the reflection of a general state of unfitness (20). This hypothesis can, however, be ruled out in the present work. With the same horses being compared at a 1-wk interval, their fitness should be unchanged from one test to the other.

Exercise-Induced Hypercapnia and Hypoxemia in COPD Horses in C

Diffusion limitation is one of the main reasons for hypoxemia during heavy exercise in healthy horses, being responsible for a large proportion, i.e., from 55% (22) to 70% (30), of the widening in PA_O2-Pa_O2 difference. Consequently, this factor could have been responsible for the more severe hypoxemia occurring in COPD horses in C. However, although the absolute values of Pa_O2 and PA_O2 were lower in C, the PA_O2-Pa_O2 difference was not different in C compared with R (Fig. 1). This suggests that diffusion limitation was not more severe in C than in R.

The lack of compensatory hyperventilation has been demonstrated to be an important factor contributing to the exercise-induced hypoxemia of healthy horses (4). In the present study, the E of the horses in C was significantly lower than that in R. Because in the same time, D and VD/VT were lower, A was not significantly decreased by the C. However, although maintained at the same level in C as in R, A was obviously too low to ensure an appropriate CO₂ removal when the horses were in C. The question of the origin of the source of this greater CO₂ while O₂ is the same in both conditions now requires addressing.

Last, hypoxemia may be related to / inadequacy. Actually, information about exercise-induced modifications of / ratio in horses has been rather conflicting. On the basis of the multiple inert-gas elimination technique, this ratio has been demonstrated to remain quite adequate during maximal exercise in healthy horses (31) or to be slightly impaired (27) and, last, to be significantly worsened, accounting for 41% of the exercise-induced hypoxemia (22). COPD horses in C at rest suffer from / inadequacy (29). Further studies are now necessary to assess whether this inadequacy is further exacerbated with exercise or whether the exercise-induced adjustments ultimately recruites unperfused capillaries and consequently decreases VD.

Compensatory Mechanisms in Submaximally Exercised COPD Horses

The fact that O₂ was the same in both C and R suggests that, in COPD horses in C, there were compensatory mechanisms that allowed these horses to maintain an adequate O₂ during submaximal exercise.

From a ventilatory point of view, although E was lower, ventilation was optimized by the lowering of D and VD/VT. The lowering of f, increasing the time period between two breaths, probably improved O₂ extraction from the inspired air as assessed by the lower E/O₂.

At the level of blood-gas transport, at least two mechanisms occurred in C to balance the impairment of gas exchange. The first was the significant increase in Hb concentration, which allowed the Ca_O2 to remain the same as in R despite the lower Sa_O2 and Pa_O2. The second was the increase in HR leading to a higher , which in turn counterbalanced the lower arterial-mixed venous O₂ content difference resulting in the same O₂.

Despite the fact that these compensatory mechanisms should reach a limit when the horse approaches its maximal exercise capacity, our horses' O_{2 peak} values seemed to be unaffected by the C. However, the O_{2 peak} in C was reached at a lower workload than in R. Further studies are neccessary to understand why O_{2 peak} was independent of the respiratory status of the horses.

Mechanical Cost of Breathing

It has already been suggested that, in healthy horses during strenuous exercise, the O₂ resp reaches a substantial percentage of O₂ and that, in this species as in humans, there is a "critical level of ventilation" above which any further increase of O₂ would be entirely consumed by the respiratory muscles (1). Because the O₂ resp is higher during a COPD crisis, the critical level is in turn probably lowered in these conditions and consequently reached at a lower exercise intensity than in normal conditions. This could explain the fact that despite 1) its more severe hypoxemia and hypercapnia, 2) its relative hypoventilation in regard to its CO₂, and 3) its potential ability to ventilate more, as assessed by the data obtained in R, the horse in C sets its ventilation at a lower level, mainly by decreasing its f. Dempsey (10) postulated that this special exercise adjustment could occur when the respiratory motor output or drive is not sufficiently augmented because of a strong feedback inhibition signal coming from the overstressed chest wall to the brain stem. Why this special adjustment occurs through a reduction of f rather than VT may be explained by the constraint of minimum average force of the respiratory muscles, which is an important determinant in the control of f (21). At a given E and with given characteristics of the respiratory system (RL and CL), a particular f would be least costly in terms of work of breathing. In the present work, RL was higher in C; therefore, f was lower, probably to reduce the flow-resistive work of breathing. The fact that there was no compensating increase in VT could be explained by the modification of the elastic properties of the lungs, as assessed by the reduction of CL measured before the test. In this condition an increase of VT would have increased disproportionately the elastic work of breathing.

When the horses were in C, their RL decreased significantly once they started to work (from 0.20 ± 0.05 kPa · l¹ · s at rest to 0.085 ± 0.005 kPa · l¹ · s during the last step). This observation has already been reported in asthmatic humans, in whom in the few first minutes of work there is bronchodilation manifested by a decrease in airflow resistance (20). This phase is thought to be due to catecholamine release in response to the stress of physical exercise. This physiological adjustment does not occur in healthy horses as reported in a previous study (1) or in the COPD horses when they were in remission. This is explained by the fact that in these animals there is no bronchomotor tone as assessed by previous studies that showed that bronchodilators have no influence on the mechanics of breathing in healthy horses and in COPD horses in R (6).

Exercise Intolerance in COPD

The first test has been performed to try to determine O_{2 max} to express the intensity of exercise of the second test in terms of relative workload. Unfortunately, for unknown reasons, the COPD horses were obviously unable to run fast enough to reach their maximal oxidative capacity. In both conditions, i.e., C and R, they stopped to run while reaching the same level of O₂, which has been termed O_{2 peak} in the present work. Nevertheless, O_{2 peak} was reached earlier when they were in C, suggesting a decrease in resistance to exercise. The same observations have been made in humans suffering from chronic airflow obstruction who stopped to work at submaximal HR and submaximal ventilation, making it difficult to isolate the real limiting factor. Killian et al. (15) suggested that exercise tolerance in this case was more related to individual behavior and motivation than to true physiological modifications.

On the other hand, all the horses were able to complete the submaximal tests, even when in C. However, the tests in C were obviously more difficult to perform and necessitated encouragement of the horses, especially at the end of the test. It was surprising that our horses in C experienced an obvious discomfort during exercise, although their O₂ was not different from R and their plasma lactate concentrations were within normal and even rather low ranges.

Recently, Killian et al. (15) and Gosselin et al. (12) demonstrated that the main symptom limiting exercise in humans suffering from COPD is often peripheral muscle fatigue rather than dyspnea, supporting the view that poor peripheral muscle function may contribute significantly to limit exercise in COPD. Moreover, in a recent state-of-the-art article, Wagner (30) pointed out that a reduction of muscle O₂ conductance, for example, by a reduction of the muscle capillary bed, could provide a limitation of O_{2 max} in patients suffering from COPD. Further investigations are thus necessary to examine these possibilities in horses.

Nevertheless, the present work led to the suggestions that exercise intolerance in heavily exercising COPD horses could occur because 1) the compensatory mechanisms are probably overwhelmed during high-intensity exercise, O_{2 max} is probably reduced and reached earlier, and time to fatigue shortened, leading to an earlier occurrence of exhaustion; 2) the mechanical work of breathing is higher during C and may induce respiratory muscle fatigue, then respiratory discomfort, and finally exercise intolerance; and 3) the higher level of blood lactate during exercise in C probably leads to an earlier occurrence of metabolic acidosis when the effort is further prolonged or intensified.

In conclusion, this work has shown that, when COPD horses in C perform a submaximal exercise, they are more hypoxemic and hypercapnic and they have a lower ventilatory level than when they are in R. Their net anaerobic metabolism is slightly higher. Their O₂ remains unchanged, which is probably due to the occurrence of compensatory mechanisms, such as a better ventilatory equivalent for O₂, a higher HR and , and a higher Hb concentration.