159
EQUINE VETERINARY JOURNAL Equine vet. J . (1979), 11(3), 159-166
Chronic Obstructive Pulmonary Disease (COPD) in Horses: Aetiological Studies: Responses to Intradermal and Inhalation Antigenic Challenge E. A. McPHERSON, G. H. K. LAWSON, JILL R. MURPHY and JANET M. NICHOLSON University o f Edinburgh Veterinary Field Station, Easter Bush, Roslin, Midlothian EH25 9RG R. G. BREEZE* and H. M. PlRlE University o f Glasgow Veterinary School, Department of Veterinary Pathology, Bearsden Road, Bearsden, Glasgow G6 I I QH SUMMARY Micropolyspora faeni and Aspergillus fumigatus were identified as common causes of respiratory hypersensitivity in horses affected with chronic obstructive pulmonary disease (COPD). Rye grass pollen and an Actinomycete evoked respiratory allergy in a few horses. Not infrequently, individual horses were found to have respiratory hypersensitivity to two or more antigens. The methods used to examine for allergy were intradermal testing and inhalation challenge with environmental antigens. An intradermal test using an M faeni extract was demonstrated to be suitable for diagnostic use in horses previously accurately diagnosed as suffering from COPD. In contrast, the A fumigatus antigen used proved unsatisfactory for such a purpose. Skin reaction to M faeni and A fumigatus extracts by horses affected with COPD indicated that the hypersensitivity was a dual one-a weak response shortly after injection followed by an Arthus-like response 4 to 8 hours later. As a parameter for monitoring responses to inhalation challenge, maximum intrathoracic pressure change (max A Ppl) proved satisfactory, whereas changes in partial pressure of arterial oxygen (PaOJ did not.
INTRODUCTION McPHERSON et a1 (1 978) described the identification of horses affected with COPD by measurement of the partial pressure of arterial oxygen (PaOJ and the maximum change in intrathoracic pressure (max ppl), the former being subnormal and the latter elevated in affected animals. From 100 horses referred to us principally because of poor work performance or chronic coughing, 38 were found to be affected (Group A), 34 were not affected (Group C) and 28 were in a doubtful category (Group B). These groups of horses were intradermally tested with environmental allergens believed to cause hypersensitive disease in other species (Austwick, 1966; Pepys, 1969a) and subsequently challenged by the inhalation of the same agents or extracts of them. The results are recorded and discussed in this paper.
a
* Present address : Washington State University, Department of Veterinary Microbiology and Pathology, Pullman, Washington 99164, USA.
MATERIALS AND METHODS ~
~ i ~ ~ l ~ The loo horses were assigned to Groups A, and and kept under the conditions described by McPherson et (1978)-
*
Techniques The max A Ppl, PaO, and haematological examinations were carried out as described by McPherson ef a1 (1978). Partial pressure of arterial carbon dioxide (PaCO,) was determined on a Corning pH/blood gas 161 analyser (Corning Medical, Halstead, Essex). Intradermal tests Antigens Bencard intradermal skin testing solutions (Bencard, Brentford, England) were used for grass pollens (0.02 per cent, B2), tree pollens (0.2 per cent, B3), mixed moulds (0.5 per cent, A13), mixed moulds (1 per cent, MS), Cladosporium herbarum (1 per cent, M3) and A fumigatus
EQUINE VETERINARY JOURNAL
160
(0.5 per cent, 2000). Bencard control solution was also used in conjunction with these antigens. Antigens from M faeni were prepared as described by Lawson et a1 (1979). The acetone precipitated M faeni antigen was diluted with 0.25 per cent phenol saline to give a solution containing 1.4 mg dry matter antigenlml. The control was a nutrient broth (Oxoid CMI) diluted volume for volume with 0.5 per cent phenol saline. Extracts were also prepared from an uncharacterised Actinomycete (705/69) isolated from the nasopharynx of a horse. This organism was grown for 7 days at 37°C in glucose broth; after sonication for 30 min at 8 microns amplitude on ice, the extract was concentrated approximately 20 times with carbowax at 5°C and finally dialysed against phosphate buffered saline (0.01 M pH 7.2). The isolate was lost before it was identified. The solution used for skin testing contained approximately 25 mg/ml of dry matter in 0.25 per cent phenol saline. Techniques 0.05 ml of the test and control solutions were injected intradermally into the closely clipped or shaven neck of the horse. The diameter of the wheal was measured (mm) horizontally at 4, 1, 2, 4,6, 8 and, if necessary, 10 hours after injection.
Grass pollens. Rye grass pollen collected during harvesting of a hay crop, cut while the grass was still flowering and kept under dry conditions in the laboratory, was used at a dose of approximately 1 gm. This was insufflated into a Cox’s face mask (Arnolds, London) over a period of about 10 min. Horses were exposed when the history suggested pollen involvement or when no other challenge had produced a positive response. Control inhalations. Where circumstances permitted, each horse was made to inhale an antigen to which it had not reacted in the skin test, as well as those to which it was likely to respond, thus providing a control challenge. At the outset, horses were exposed to an aerosol of 5 ml of 0.25 per cent phenol saline to which none reacted. Statistical methods Chi-square tests and analysis of variance, followed when appropriate by Duncan’s multiple range test, were employed as described by McPherson et al(1978). Student’s “ t ” test was employed in assessing response to challenge and McNemar’s xztest corrected for continuity was employed in comparing the results of skin and intradermal tests.
RESULTS Intradermal tests Horses frequently responded by producing swellings Inhalation challenge to the control injections. This confused the interpretIn the case of fungal extracts, a final volume of 5 ml ation of skin tests employing specific allergens. Only was administered as an aerosol over a period of 20 min. reactions to an antigen which exceeded the mean reaction A Wright’s nebuliser (Aerosol Products, London) of horses to the control plus two standard deviations converted the solution into an aerosol which was led by (X 2 SD) were considered positive. The control a plastic tube, 10 mm internal diameter, into a face mask means were calculated from the reactions of 75 horses constructed from a rigid plastic bucket and a flexible to the Bencard control solution and 90 horses to the plastic sleeve. An aperture 9 cm in diameter was nutrient broth solution. These means, and the decision cut in the bottom of the bucket which was suspended by a levels derived from them (> X 2 SD) for evaluating strap over the horse’s poll region. The plastic sleeve response to antigens, are shown diagramatically in embraced the bucket tightly and was sealed to the face Figs 1 and 3. with plastic foam held in position by a self-adhesive Using these criteria for interpretations, the responses nylon strip. The tube from the nebuliser passed through of all horses tested with a variety of antigens are set the face seal into the bucket. The attendant wore a out in Table I and the statistical significance of the mask and the administration was carried out in a well- differences in response given by Group A (affected) and ventilated room with a wide open door. Group B C (doubtful or not affected) was determined. The resting max A Ppl and PaO, were determined on Significantly more of the horses affected with COPD the morning of the test. After inhalations were com- showed skin sensitivity to A fumigatus and A4 faeni than pleted, clinical observations were made hourly and did horses in the other two classes. monitoring of the foregoing parameters was repeated 4 to 5 hours after inhalation. When no changes had taken place, observation continued and, where necessary, A+n=25 further recordings were made up to 10 hours after B+n= 8 completion of the inhalations. ~ + n =6 M ,faeni, A fumigatus and Actinomycete 705169. The 3 h B control first 2 antigens were prepared as described by Lawson .-YC ii+2sd 15 et a1 (1979) and the last as described above. The 0 8-n=17 challenge dose of M faeni and A fumigatus was approx- I A- n = 10 imately 12 mg of antigen extract (estimated as dry matter), ;10 C-n=14 whereas that of the Actinomycete was 50 mg (estimated as .-Qm dry matter), suspended in all cases in 5 ml normal saline. 5 5 NB control n = 92 Mouldy hay. This material was used in the earlier 5 part of the work and occasionally later, when horses 0 0.5 1 2 4 6 8 failed to respond to the more specific antigens. Bales of poor quality fodder (mouldy or heated straw or hay Hours + = i for horses responding positively as available) were shaken up in a loosebox in which the zforgroups A, B and c = e , X & A - = T i for horses responding negatively animal on test was confined for 1 hour before returning n = number of horses to its own loosebox for observation and monitoring as Fig I . Skin response of horses to M faeni antigen (0.05 nil described. intradermally).
+
+
+
~1
L
L
EQUINE VETERINARY JOURNAL
161
TABLE I INTRADERMAL SKIN TEST RESPONSE OF HORSES TO ENVIRONMENTAL ANTIGENS Group A
t
Group B + C
-
t
-
Chi-square value
Probability
0.931 0.647 0.515 0.499 0.076 0.266 11.926 12.808
> 0.05 > 0.05
(PI
~~
2 3 25 13
*Grass pollen *Tree pollen *Mixed mould (A13) *Mixed mould (A5) * Cladosporium herbarum Actinomycete spp * Aspergillus fumigatus Mieropolyspora faeni
15
22 17 25
31 30 9 21 19 15 17 10
34 33 14 22 20 19 34 31
5
6 27 19 18 22 5 14
> 0.05 > 0.05 > 0.05 > 0.05 < 0.001 < 0.001
* Bencard antigen solutions Descriptions of skin reactions Each of the antigens employed failed to evoke a response in some horses. When horses responded, the wheals tended to be rounded and relatively circumscribed while developing and were dome-shaped, tense and painful to the touch. Later, the swellings softened and diffused into the surrounding tissue. The oedema then tended to move ventrally and, in severe reactions, the superficial lymphatics were visible. In most cases, the reactions were flattened and no longer clearly defined 4 to 6 hours after reaching their maximum but, in a few severe reactions, 12 to 24 hours passed before the swellings completely disappeared. No skin necrosis occurred. As descriptions are subjective, reliance was placed on measurement alone. Response to A4 faeni The mean diameter of the wheals at each time of measurement for horses in Groups A, B and C which gave positive (A+, B+, C + ) and negative (A-, B-, C-) response to M.faeni antigen were plotted in Fig 1. The top of the hatched area is the calculated mean response (X 2 SD) for all animals to control nutrient
+
POSITIVE SKIN
TESTS
M f a e n i (25 horses)
POSITIVE S K I N TESTS A f u m i g a t u s (17 horses)
n
broth solution. There was, by 6 and 8 hours, a clear separation of the responses of horses which showed skin sensitivity to A4 faeni from those which did not. In horses belonging to Groups A and B this had occurred by 4 hours. In the reacting horses in Group C the response was late, at 6 and 8 hours, hence the mean wheal diameter for this group at 4 hours is within the hatched area. If interpretation of the test had been made only at 4 hours 12 per cent of positive reactions would have been missed; at 6 or at 8 hours, 10 per cent. If, however, readings were made at both 6 and 8 hours only 4 per cent would have been missed. The times taken to reach a positive reaction and the maximum size of wheal are shown in Fig 2. The timing of the maximum response to the M fueni skin test indicates that this response could involve an Arthus-like phenomenon. When those horses which clearly reached a positive response between 4 and 8 hours were then compared with animals failing to respond in such a manner in respect of their skin reactions early after injection, it was found that they differed. The mean response 1 hour post-injection of the A+ was significantly greater than that of the A- horses (Fig 1) and that of all horses to the control nutrient broth solution ( F = 4.89; P
EQUINE VETERINARY JOURNAL Equine vet. J . (1979), 11(3), 159-166
Chronic Obstructive Pulmonary Disease (COPD) in Horses: Aetiological Studies: Responses to Intradermal and Inhalation Antigenic Challenge E. A. McPHERSON, G. H. K. LAWSON, JILL R. MURPHY and JANET M. NICHOLSON University o f Edinburgh Veterinary Field Station, Easter Bush, Roslin, Midlothian EH25 9RG R. G. BREEZE* and H. M. PlRlE University o f Glasgow Veterinary School, Department of Veterinary Pathology, Bearsden Road, Bearsden, Glasgow G6 I I QH SUMMARY Micropolyspora faeni and Aspergillus fumigatus were identified as common causes of respiratory hypersensitivity in horses affected with chronic obstructive pulmonary disease (COPD). Rye grass pollen and an Actinomycete evoked respiratory allergy in a few horses. Not infrequently, individual horses were found to have respiratory hypersensitivity to two or more antigens. The methods used to examine for allergy were intradermal testing and inhalation challenge with environmental antigens. An intradermal test using an M faeni extract was demonstrated to be suitable for diagnostic use in horses previously accurately diagnosed as suffering from COPD. In contrast, the A fumigatus antigen used proved unsatisfactory for such a purpose. Skin reaction to M faeni and A fumigatus extracts by horses affected with COPD indicated that the hypersensitivity was a dual one-a weak response shortly after injection followed by an Arthus-like response 4 to 8 hours later. As a parameter for monitoring responses to inhalation challenge, maximum intrathoracic pressure change (max A Ppl) proved satisfactory, whereas changes in partial pressure of arterial oxygen (PaOJ did not.
INTRODUCTION McPHERSON et a1 (1 978) described the identification of horses affected with COPD by measurement of the partial pressure of arterial oxygen (PaOJ and the maximum change in intrathoracic pressure (max ppl), the former being subnormal and the latter elevated in affected animals. From 100 horses referred to us principally because of poor work performance or chronic coughing, 38 were found to be affected (Group A), 34 were not affected (Group C) and 28 were in a doubtful category (Group B). These groups of horses were intradermally tested with environmental allergens believed to cause hypersensitive disease in other species (Austwick, 1966; Pepys, 1969a) and subsequently challenged by the inhalation of the same agents or extracts of them. The results are recorded and discussed in this paper.
a
* Present address : Washington State University, Department of Veterinary Microbiology and Pathology, Pullman, Washington 99164, USA.
MATERIALS AND METHODS ~
~ i ~ ~ l ~ The loo horses were assigned to Groups A, and and kept under the conditions described by McPherson et (1978)-
*
Techniques The max A Ppl, PaO, and haematological examinations were carried out as described by McPherson ef a1 (1978). Partial pressure of arterial carbon dioxide (PaCO,) was determined on a Corning pH/blood gas 161 analyser (Corning Medical, Halstead, Essex). Intradermal tests Antigens Bencard intradermal skin testing solutions (Bencard, Brentford, England) were used for grass pollens (0.02 per cent, B2), tree pollens (0.2 per cent, B3), mixed moulds (0.5 per cent, A13), mixed moulds (1 per cent, MS), Cladosporium herbarum (1 per cent, M3) and A fumigatus
EQUINE VETERINARY JOURNAL
160
(0.5 per cent, 2000). Bencard control solution was also used in conjunction with these antigens. Antigens from M faeni were prepared as described by Lawson et a1 (1979). The acetone precipitated M faeni antigen was diluted with 0.25 per cent phenol saline to give a solution containing 1.4 mg dry matter antigenlml. The control was a nutrient broth (Oxoid CMI) diluted volume for volume with 0.5 per cent phenol saline. Extracts were also prepared from an uncharacterised Actinomycete (705/69) isolated from the nasopharynx of a horse. This organism was grown for 7 days at 37°C in glucose broth; after sonication for 30 min at 8 microns amplitude on ice, the extract was concentrated approximately 20 times with carbowax at 5°C and finally dialysed against phosphate buffered saline (0.01 M pH 7.2). The isolate was lost before it was identified. The solution used for skin testing contained approximately 25 mg/ml of dry matter in 0.25 per cent phenol saline. Techniques 0.05 ml of the test and control solutions were injected intradermally into the closely clipped or shaven neck of the horse. The diameter of the wheal was measured (mm) horizontally at 4, 1, 2, 4,6, 8 and, if necessary, 10 hours after injection.
Grass pollens. Rye grass pollen collected during harvesting of a hay crop, cut while the grass was still flowering and kept under dry conditions in the laboratory, was used at a dose of approximately 1 gm. This was insufflated into a Cox’s face mask (Arnolds, London) over a period of about 10 min. Horses were exposed when the history suggested pollen involvement or when no other challenge had produced a positive response. Control inhalations. Where circumstances permitted, each horse was made to inhale an antigen to which it had not reacted in the skin test, as well as those to which it was likely to respond, thus providing a control challenge. At the outset, horses were exposed to an aerosol of 5 ml of 0.25 per cent phenol saline to which none reacted. Statistical methods Chi-square tests and analysis of variance, followed when appropriate by Duncan’s multiple range test, were employed as described by McPherson et al(1978). Student’s “ t ” test was employed in assessing response to challenge and McNemar’s xztest corrected for continuity was employed in comparing the results of skin and intradermal tests.
RESULTS Intradermal tests Horses frequently responded by producing swellings Inhalation challenge to the control injections. This confused the interpretIn the case of fungal extracts, a final volume of 5 ml ation of skin tests employing specific allergens. Only was administered as an aerosol over a period of 20 min. reactions to an antigen which exceeded the mean reaction A Wright’s nebuliser (Aerosol Products, London) of horses to the control plus two standard deviations converted the solution into an aerosol which was led by (X 2 SD) were considered positive. The control a plastic tube, 10 mm internal diameter, into a face mask means were calculated from the reactions of 75 horses constructed from a rigid plastic bucket and a flexible to the Bencard control solution and 90 horses to the plastic sleeve. An aperture 9 cm in diameter was nutrient broth solution. These means, and the decision cut in the bottom of the bucket which was suspended by a levels derived from them (> X 2 SD) for evaluating strap over the horse’s poll region. The plastic sleeve response to antigens, are shown diagramatically in embraced the bucket tightly and was sealed to the face Figs 1 and 3. with plastic foam held in position by a self-adhesive Using these criteria for interpretations, the responses nylon strip. The tube from the nebuliser passed through of all horses tested with a variety of antigens are set the face seal into the bucket. The attendant wore a out in Table I and the statistical significance of the mask and the administration was carried out in a well- differences in response given by Group A (affected) and ventilated room with a wide open door. Group B C (doubtful or not affected) was determined. The resting max A Ppl and PaO, were determined on Significantly more of the horses affected with COPD the morning of the test. After inhalations were com- showed skin sensitivity to A fumigatus and A4 faeni than pleted, clinical observations were made hourly and did horses in the other two classes. monitoring of the foregoing parameters was repeated 4 to 5 hours after inhalation. When no changes had taken place, observation continued and, where necessary, A+n=25 further recordings were made up to 10 hours after B+n= 8 completion of the inhalations. ~ + n =6 M ,faeni, A fumigatus and Actinomycete 705169. The 3 h B control first 2 antigens were prepared as described by Lawson .-YC ii+2sd 15 et a1 (1979) and the last as described above. The 0 8-n=17 challenge dose of M faeni and A fumigatus was approx- I A- n = 10 imately 12 mg of antigen extract (estimated as dry matter), ;10 C-n=14 whereas that of the Actinomycete was 50 mg (estimated as .-Qm dry matter), suspended in all cases in 5 ml normal saline. 5 5 NB control n = 92 Mouldy hay. This material was used in the earlier 5 part of the work and occasionally later, when horses 0 0.5 1 2 4 6 8 failed to respond to the more specific antigens. Bales of poor quality fodder (mouldy or heated straw or hay Hours + = i for horses responding positively as available) were shaken up in a loosebox in which the zforgroups A, B and c = e , X & A - = T i for horses responding negatively animal on test was confined for 1 hour before returning n = number of horses to its own loosebox for observation and monitoring as Fig I . Skin response of horses to M faeni antigen (0.05 nil described. intradermally).
+
+
+
~1
L
L
EQUINE VETERINARY JOURNAL
161
TABLE I INTRADERMAL SKIN TEST RESPONSE OF HORSES TO ENVIRONMENTAL ANTIGENS Group A
t
Group B + C
-
t
-
Chi-square value
Probability
0.931 0.647 0.515 0.499 0.076 0.266 11.926 12.808
> 0.05 > 0.05
(PI
~~
2 3 25 13
*Grass pollen *Tree pollen *Mixed mould (A13) *Mixed mould (A5) * Cladosporium herbarum Actinomycete spp * Aspergillus fumigatus Mieropolyspora faeni
15
22 17 25
31 30 9 21 19 15 17 10
34 33 14 22 20 19 34 31
5
6 27 19 18 22 5 14
> 0.05 > 0.05 > 0.05 > 0.05 < 0.001 < 0.001
* Bencard antigen solutions Descriptions of skin reactions Each of the antigens employed failed to evoke a response in some horses. When horses responded, the wheals tended to be rounded and relatively circumscribed while developing and were dome-shaped, tense and painful to the touch. Later, the swellings softened and diffused into the surrounding tissue. The oedema then tended to move ventrally and, in severe reactions, the superficial lymphatics were visible. In most cases, the reactions were flattened and no longer clearly defined 4 to 6 hours after reaching their maximum but, in a few severe reactions, 12 to 24 hours passed before the swellings completely disappeared. No skin necrosis occurred. As descriptions are subjective, reliance was placed on measurement alone. Response to A4 faeni The mean diameter of the wheals at each time of measurement for horses in Groups A, B and C which gave positive (A+, B+, C + ) and negative (A-, B-, C-) response to M.faeni antigen were plotted in Fig 1. The top of the hatched area is the calculated mean response (X 2 SD) for all animals to control nutrient
+
POSITIVE SKIN
TESTS
M f a e n i (25 horses)
POSITIVE S K I N TESTS A f u m i g a t u s (17 horses)
n
broth solution. There was, by 6 and 8 hours, a clear separation of the responses of horses which showed skin sensitivity to A4 faeni from those which did not. In horses belonging to Groups A and B this had occurred by 4 hours. In the reacting horses in Group C the response was late, at 6 and 8 hours, hence the mean wheal diameter for this group at 4 hours is within the hatched area. If interpretation of the test had been made only at 4 hours 12 per cent of positive reactions would have been missed; at 6 or at 8 hours, 10 per cent. If, however, readings were made at both 6 and 8 hours only 4 per cent would have been missed. The times taken to reach a positive reaction and the maximum size of wheal are shown in Fig 2. The timing of the maximum response to the M fueni skin test indicates that this response could involve an Arthus-like phenomenon. When those horses which clearly reached a positive response between 4 and 8 hours were then compared with animals failing to respond in such a manner in respect of their skin reactions early after injection, it was found that they differed. The mean response 1 hour post-injection of the A+ was significantly greater than that of the A- horses (Fig 1) and that of all horses to the control nutrient broth solution ( F = 4.89; P
