Inhaled corticosteroids are now recommended as first-line therapy for asthma in patients of all ages because these drugs avoid some of the undesirable effects of systemic corticosteroid administration.1 One of the most widely used drugs is fluticasone propionate (FP), which also is recommended for the management of recurrent airway obstruction (RAO) and inflammatory airway disease in horses. In RAO-affected horses, FP (2 mg twice daily for 3 weeks) improves lung function, and a similar dose administered for 2 weeks enhances the improvement in lung function caused by environmental management alone.2,3 There are no published papers either on the shorter-term effects of FP in RAO-affected horses or on the effectiveness of FP for prevention of exacerbations of RAO. In the current study, therefore, we compared the effects of FP (3 and 6 mg administered twice daily for 3 days) and dexamethasone (DEX) (0.1 mg/kg IV q24h for 3 days) in horses with RAO. We measured serum cortisol (COR) as an indicator of potential systemic effect and monitored the feet for signs of laminitis. In a 2nd trial, we compared 6 mg FP with DEX for prevention of the development of airway obstruction when RAO horses were moved from pasture to the stable. Previous investigations of aerosol FP in horses have used the Equine Aeromask.2,3 In the present study, we used the less cumbersome EquineHaler,a which has been demonstrated to deliver Tc99m-labeled FP effectively into equine lungs.4
Background: Efficacy of inhaled fluticasone propionate (FP) for management of recurrent airway obstruction (RAO) has only been evaluated after several weeks' treatment.
Objectives: To compare efficacy of (1) 3-day treatments with FP to dexamethasone (DEX) for management of RAO; and (2) FP and DEX to no treatment in prevention of acute RAO exacerbations.
Animals: Nine RAO affected horses.
Methods: Crossover studies in RAO-affected horses compared (a) 3-day treatment of RAO exacerbation with FP (3 and 6 mg q12h) and DEX (0.1 mg/kg q24h) and (b) FP (6 mg q12h) and DEX (0.1 mg/kg q24h) to no treatment for prevention of acute exacerbations of RAO. Treatment efficacy and unwanted effects were judged from maximal change in pleural pressure (ΔPplmax), serum cortisol (COR), bronchoalveolar lavage (BAL) cytology, and subjective scores for respiratory distress and lameness.
Results: In treatment trial, DEX and FP (6 mg) significantly decreased ΔPplmax by 48 and 72 hours, respectively; FP (3 mg) had no significant effect. DEX decreased COR more than did FP. In prevention trial, both DEX and FP (6 mg) prevented the increase in ΔPplmax that occurred in untreated horses. Both treatments decreased COR to the same degree. FP and DEX had no effects on bronchoalveolar lavage fluid (BALF) cytology and there was no evidence of laminitis.
Conclusions and Clinical Importance: FP (6 mg q12h) is as effective as DEX for prevention of acute exacerbations of RAO and lower doses should be evaluated. High-dose FP is not as effective as DEX for treatment of RAO exacerbations.
bronchoalveolar lavage fluid
maximal change in pleural pressure
recurrent airway obstruction
Materials and Methods
The All-University Committee for Animal Use and Care of Michigan State University approved the protocols. Nine horses (5 mares and 4 geldings), 14–33 years old (mean ± SD, 24.7 ± 6.0 years), weighing 502 ± 73 kg, with a previous diagnosis of RAO, were used for the investigations. When housed and fed hay, all horses had demonstrated atropine-reversible airway obstruction. Animals were maintained on pasture and supplemented with a pelleted diet except during periods of stabling and exposure to hay and straw bedding, which were used to induce acute exacerbations of RAO.5
Clinical Respiratory Score and Maximal Change in Pleural Pressure (ΔPplmax)
Clinical signs of airway obstruction were numerically scored by summing the degree of nasal flaring and abdominal effort (each scored 1–4).6,7 To measure ΔPplmax, horses were intubated with an esophageal balloonb sealed over the end of a 240-cm-long fluorinated ethylene propylene teflon catheterc (2.4 mm internal diameter, 4 mm external diameter) with lateral holes drilled in the portion covered by the balloon. The catheter was inserted so that the balloon was placed caudal to the heart base but cranial to the diaphragm. This location provides a good estimate of the pressure within the pleural cavity at that location.8 Pressure changes within the balloon were detected by means of a pressure transducerd and recorded on a portable physiograph.e The equipment was calibrated daily against a water manometer. Fluctuations in esophageal pressure were recorded for 20 breaths. The differences between peak inspiratory and peak expiratory pressures were calculated to derive ΔPplmax.
Bronchoalveolar Lavage (BAL) Cytology
A 3-m endoscope or Bivona BAL tube was wedged in a bronchus. Five (Protocol 1) or 3 (Protocol 2) 100-mL aliquots of phosphate-buffered saline were infused and aspirated and the samples were pooled. Leukocyte density per microliter BAL fluid was determined with a hemocytometer. The percentage of each type of white blood cell was determined by counting 200 cells on a Wright-Giemsa-stained cytocentrifuged preparation.
To evaluate the possible presence of laminitis during corticosteroid treatment, we scored hoof temperature from 1 (cool) to 3 (hot), digital pulse from 1 (barely or not palpable) to 3 (bounding full pulse), and lameness severity according to Obel laminitis grade 0–4.9 The horse's posture was evaluated at rest in its stall, the feet were palpated to determine hoof temperature, digital arteries were palpated at the fetlock, an attempt was made to pick up the front feet, and the horse was led around the stall by means of a halter and lead shank and its gait observed.
Measurement of COR
COR was measured by a commercially available radioimmunoassay10 at the Diagnostic Center for Population and Animal Health at Michigan State University.
Drugs and Administration
FP (125 μg per actuation) in HFA propellant was provided in metered-dose canisters. For drug administration, 2 actuations from the metered-dose canister were delivered into the spacer of the EquineHaler, which then was placed over the nostril for 3 breaths. The procedure was repeated until the entire dose was delivered.
The investigation consisted of 2 protocols conducted in sequence: comparison of the efficacy of FP (2 doses) and DEX for treatment of RAO, and comparison of the FP and DEX with no treatment for the prevention of clinical exacerbations of RAO. Both protocols had 3 treatments, the sequence of which differed among horses.
Comparison of FP and DEX for Treatment of RAO Exacerbation. In 8 horses, we used a nonblinded crossover protocol to compare the efficacy of FP (3 and 6 mg q12h; 0900 and 1900 hours) and DEX (0.1 mg/kg q24h; 0900 hours) for treatment of RAO. The experiment was planned so that the sequence of treatments differed among horses. Horses were brought to the barn, fed hay, bedded on straw, and observed daily until clinical score was at least 5.0, at which point ΔPplmax was measured. The horse qualified for inclusion in the trial if ΔPplmax was >10 cmH2O at that time, 24 hours later, and 15 minutes beyond 24 hours. Scores for clinical signs of respiratory distress, lameness, hoof temperature, and digital pulse scores were recorded, a blood sample was taken for measurement of COR, BAL was conducted, and the 3-day treatment begun (always in the morning). All scores were recorded and ΔPplmax was measured 24, 48, and 72 hours after treatment initiation; blood samples for COR measurement were taken at 36 and 72 hours. There was a minimum of 21 days at pasture between each arm of the protocol.
Comparison of FP and DEX to No Treatment for Prevention of RAO Exacerbation. In 6 horses, we used a Williams modification of a Latin square design to compare the effects of FP (6 mg q12h, 0900 and 1900 hours) and DEX (0.1 mg/kg q24h, 0900 hours) to no treatment on the development of an exacerbation of heaves during stabling. The protocol began with recording of the clinical respiratory score when horses had been at pasture for a minimum of 21 days. If clinical score was 4 out of 8 or less, ΔPplmax was measured and if it was <10 cmH2O, the measurement was repeated 24 hours later. If ΔPplmax was still <10 cmH2O on 2 consecutive measurements 15 minutes apart, the horse qualified for the protocol. Clinical score, lameness, hoof warmth, and digital pulse scores were recorded, BAL was performed, and a blood sample for measurement of COR was taken. Treatment began immediately and continued for a total of 7 days. For the 1st 3 days, horses remained on pasture. On the morning of the 4th day, horses were brought to the laboratory and kept in an outdoor paddock for 60 minutes before measurement of ΔPplmax and recording of clinical score, lameness, hoof warmth, and digital pulse scores, and taking a blood sample for measurement of COR. Horses then were placed in stalls, fed hay, and bedded on straw, where they continued to receive treatment for the remainder of the protocol. On Days 5–8, ΔPplmax was measured and scores for clinical respiratory signs, lameness, hoof warmth, and digital pulses were recorded daily. On Day 8, BAL was performed and a blood sample for measurement of COR also was taken. The horses then were returned to pasture for a minimum of 21 days before the protocol was repeated with another treatment.
We first tested for differences between means of the qualifying periods. No differences were detected and we averaged the 2 qualifying measurements that were 15 minutes apart to obtain a baseline against which to compare other time periods. The data from all protocols were analyzed by 2-way repeated measures analysis of variance, with treatment and time as the fixed effect and horse as a random effect. When data were not normally distributed, appropriate transformations were made. When significant main effects or treatment-time interactions occurred, means were compared by Tukey's highly significant difference test. Significance was set at P < .05. Statistical analysis was performed on commercial software.f
Comparison of FP and DEX for Treatment of RAO Exacerbation
Of the 8 horses in the trial, 1 did not respond to any treatment, but all others improved with DEX. This horse, which had previously exhibited bronchodilatation in response to atropine, therefore was classified as steroid unresponsive, and the data on the remaining 7 horses was analyzed. Mean baseline values for clinical score, ΔPplmax, bronchoalveolar lavage fluid (BALF) cytology, COR concentration, and scores for hoof temperature and lameness did not differ among treatments. When compared with baseline, DEX treatment significantly decreased clinical score at 24 (P < .001), 48 (P < .001), and 72 hours (P < .001), whereas FP (6 mg) did not decrease score until 72 hours (P= .007) (Fig 1A). By contrast, FP (3 mg) had no effect on clinical score. Fluticasone (3 mg) failed to decrease ΔPplmax, whereas FP (6 mg) significantly decreased ΔPplmax at 72 hours (P= .042) and DEX decreased ΔPplmax at 48 (P= .004) and 72 hours (P < .001; Fig 1B). In the case of both clinical score and ΔPplmax, there were no significant differences between DEX and FP (6 mg), or between the 2 FP doses. However, the improvements were significantly greater with DEX than with 3 mg FP (P= .008). Individual values at baseline and 72 hours are shown in Figure 2. Treatment with 3 mg FP decreased ΔPplmax in only 4 of 7 horses and the mean change was a 6.6 ± 58.7% increase. By contrast, treatment with 6 mg FP decreased ΔPplmax in 5 horses (mean ± SD decrease was 37.7 ± 45.7%) and DEX decreased ΔPplmax in all 7 horses (mean ± SD decrease was 70.4 ± 19.2%).
Baseline BALF cytology did not differ among treatment groups and there was no effect of treatment (Table 1). At baseline, COR concentration did not differ among treatments. There were highly significant effects of treatment, time and treatment-time interactions on COR concentration (Fig 3). With all 3 treatments, COR had decreased significantly by 36 hours and did not decrease significantly more by 72 hours (P= .009 for FP 3 mg; and P < .001 for FP 6 mg and DEX). At 36 hours, COR concentration after DEX was significantly lower than after the FP 3 mg (P < .001) and FP 6 mg (P= .002) treatments. The difference continued to 72 hours when COR after DEX was significantly less than after both FP 3 mg (P < .001) and FP 6 mg (P= .003). At no time did COR concentrations differ between the 2 FP doses.
|Treatment||Time||Total Cells/μL||Inflammatory Cell Percentage|
|FP3||Base||290 ± 140||46 ± 19||15 ± 7||39 ± 21||0 ± 0||0 ± 0|
|72 hours||404 ± 371||54 ± 19||12 ± 6||35 ± 18||0 ± 0||0 ± 0|
|FP6||Base||454 ± 282||41 ± 35||13 ± 9||39 ± 29||0 ± 0||1 ± 2|
|72 hours||238 ± 121||39 ± 29||17 ± 7||44 ± 26||0 ± 0||0 ± 1|
|DEX||Base||331 ± 221||37 ± 26||17 ± 10||44 ± 19||0 ± 0||1 ± 2|
|72 hours||596 ± 358||47 ± 33||22 ± 19||31 ± 21||0 ± 0||0 ± 0|
There was no significant change in hoof warmth, lameness score, or digital pulse during the treatments (Table 2). Although horses occasionally received a lameness score of 2, this was sometimes at baseline and was not associated with any particular treatment or treatment duration.
|Lameness||Base||0 (0–2)||0 (0–2)||0 (0–1)|
|24||0 (0–2)||0 (0–2)||0 (0–1)|
|48||0 (0–2)||0 (0–1)||0 (0–1)|
|72||0 (0–2)||0 (0–1)||0 (0–1)|
|Hoof temperature||Base||1 (1–2)||2 (1–2)||1 (1–2)|
|24||1 (1–2)||1 (1–2)||1 (1–2)|
|48||2 (1–2)||1 (1–2)||2 (1–2)|
|72||1 (1–2)||1 (1–2)||1 (1–2)|
|Digital pulse||Base||1 (1–1)||1 (1–2)||1 (1–2)|
|24||1 (1–2)||1 (1–2)||1 (1–2)|
|48||1 (1–2)||1 (1–2)||1 (1–2)|
|72||1 (1–2)||1 (1–2)||1 (1–2)|
Comparison of FP and DEX to No Treatment for Prevention of RAO Exacerbation. In the absence of corticosteroid treatment, only 3 of the 6 horses developed obvious clinical signs of RAO. For the group as a whole, mean clinical score began to increase once horses were placed in the stable and was significantly increased on Days 5 through 8 of the protocol (ie, 1–5 days after the start of stabling [P < .001]). On Days 6 and 7, clinical score was significantly less during DEX (Day 6, P= .035; Day 7, P= .003) and FP (Day 6, P= .003; Day 7, P= .003) treatments than in the absence of treatment, but the 3 treatments did not differ significantly on Day 8. At no time was there any significant difference between DEX and FP treatments (Fig 4A). Although mean ΔPplmax increased when horses received no treatment, at no time was it significantly increased above the values on Days 1 and 4. On Day 6, ΔPplmax was significantly less in the presence of DEX (P= .025) and FP (P= .011) treatments than without treatment (Fig 4B). On Day 7 FP continued its protective effect (P= .011) but DEX did not (P= .065). On Day 8 both DEX (P= .001) and FP (P < .001) maintained ΔPplmax below the value with no treatment.
In the 3 horses that developed obvious signs of severe airway obstruction in the absence of treatment, ΔPplmax was 21.5, 39.3, and 32.8 cmH2O on Day 8. By contrast, with FP and DEX treatment, ΔPplmax was 6.2, 5.0, and 3.2 and 6.6, 4.8, and 13.7 cmH2O, respectively.
Baseline values of BALF cytology did not differ significantly and there was no effect of any treatment (Table 3). COR concentration was significantly decreased and to a similar degree by both DEX and FP treatment (Fig 5). The decrease was evident by Day 4 (P < .001 for both treatments) and continued to Day 8 (P < .001 for DEX; P= .002 for FP). In the absence of corticosteroid treatment, there was a slight increase in COR at Day 4 (immediately before stabling) and then a significant decrease (P= .015) by Day 8 as the animals developed airway obstruction. Even after this decrease, however, the COR concentration was still within the reference range and significantly greater than after corticosteroid administration.
|Treatment||Time||Total Cells/μL||Inflammatory Cell Percentage|
|None||Day 1||237 ± 49||28 ± 16||32 ± 15||41 ± 23||0 ± 0||0 ± 0|
|Day 8||258 ± 105||34 ± 25||20 ± 13||45 ± 23||0 ± 0||1 ± 1|
|FP6||Day 1||268 ± 180||19 ± 32||24 ± 10||57 ± 25||0 ± 0||0 ± 0|
|Day 8||197 ± 135||42 ± 28||20 ± 19||37 ± 22||0 ± 0||1 ± 1|
|DEX||Day 1||210 ± 159||30 ± 29||23 ± 12||47 ± 21||0 ± 0||1 ± 1|
|Day 8||283 ± 88||32 ± 23||17 ± 13||51 ± 24||0 ± 0||1 ± 1|
None of the treatments had any significant effect on lameness, hoof temperature, or digital pulse scores. Median values and ranges (data not shown) were similar to those in the treatment protocol.
This investigation demonstrated that aerosol FP at a high dose (6 mg q12h), delivered with the EquineHaler, has beneficial effects in the initial treatment of RAO and can be used for prevention of exacerbations of airway obstruction in RAO-affected animals that are in remission. Furthermore, this high dose of FP is as effective as DEX (0.1 mg/kg IV) for prevention of exacerbations but less consistently effective than DEX for short-term treatment.
One shortcoming of the protocols was the lack of blinding of individuals who judged the clinical respiratory, lameness, hoof temperature, and digital pulse scores. Lack of blinding was in part because of budgetary constraints and also because we used the objective measure ΔPplmax to assess the severity of lung disease. A decrease in ΔPplmax is a simple field indicator of improved lung function because it can be a consequence of a decrease in pulmonary resistance to airflow, an increase in lung compliance, dilatation of peripheral airways, or a decrease in respiratory rate as a consequence of improved gas exchange and reduction of airway inflammation. Although the clinical respiratory score we used is strongly associated with ΔPplmax, it is most strongly correlated with the peak expiratory flow (PEF).11 In horses with RAO, PEF is highest during exacerbations of disease and decreases during remission.12 Clinical score thus provides an indicator of lung function that is complementary to ΔPplmax. The other subjective scores were used to assess the possible onset of laminitis. Because of our concerns about this very painful condition, we were more likely to err on the high side of scoring and bias our scores against the drugs.
To determine the efficacy of FP for treatment of RAO, we compared 2 FP doses (3 and 6 mg given q12h) with DEX at a dose (0.1 mg/kg IV q24h) that is known to improve lung function as measured by ΔPplmax.7,13 Because a significant decrease in ΔPplmax occurs as little as 90 minutes after IV administration of DEX and the effect peaks at 6 hours,14 it was not surprising therefore that DEX significantly decreased both ΔPplmax and clinical score in all horses within 24–48 hours of initiation of treatment. Although FP (3 mg q12h) decreased ΔPplmax slightly in 4 of 7 horses, overall it was ineffective in decreasing either ΔPplmax or clinical score. However, the higher dose (6 mg q12h) had statistically significant beneficial effects on both ΔPplmax and clinical score, but these did not become significant until 24–48 hours later than with DEX treatment. Furthermore, the mean magnitude of improvement in ΔPplmax with FP (6 mg) was only 37.7% compared with 70.4% during DEX treatment.
Comparing the results of the present study with others is hampered in part by differences in aerosol delivery devices. Bearing that in mind, in 1 of the other investigations of the efficacy of FP for treatment of RAO, 4 horses received 2 mg twice daily for 3 weeks.3 At the end of this period, ΔPplmax had decreased significantly and was no longer different from the remission value. Our data show that FP at a higher dose can be effective in a shorter period of time. Interestingly, at the end of the 3-week treatment with FP,3 the BALF neutrophil percentage was decreased. We saw no such decrease in our short-term trial, a result similar to another investigation in which FP (2 mg q12h) improved lung function within 2 weeks but neutrophil numbers in BALF were unchanged.2 The same variable results have been reported in trials with DEX, in which longer-term drug administration has been associated with a decrease in BALF neutrophil percentage,13,15–17 but short-term administration18 or administration of a low dose (0.04 mg/kg) caused no change.19
Based on the results of the treatment protocol, we selected the dose of FP (6 mg) to be used in the prevention protocol, in which we compared it to DEX and to no treatment. Horses were only allowed to enter the protocol after at least 21 days on pasture and if ΔPplmax was ≤10 cmH2O, a value indicating remission in RAO-affected horses. Because we wanted to mimic the conditions under which FP might be administered in practice, the severity of airway inflammation as demonstrated by BALF cytology was not a criterion for entry into the study. Horses were treated for 3 days before being stabled and continued for 4 days thereafter. In the presence of corticosteroid treatment, ΔPplmax, and clinical score were significantly less than in the absence of treatment (ie, an acute exacerbation did not occur). In contrast to the results of the treatment protocol, no statistically significant difference in efficacy of the 2 corticosteroids for prevention of acute exacerbation was observed. During FP treatment, ΔPplmax remained <10 cmH2O in every horse, and with DEX treatment, ΔPplmax only exceeded 10 cmH2O in 1 horse, in which it reached only 13.7 cmH2O.
Comparing the results of our 2 protocols indicates that FP (6 mg q12h) has greater efficacy as a preventative than as a therapeutic agent. One possible explanation for this is more efficacious delivery of the drug into the lung when horses are in remission than during acute exacerbations of RAO. During exacerbations, diffuse airway obstruction by mucoid secretions and bronchospasm limits delivery of aerosols into the peripheral airways of RAO-affected horses.20 For this reason, when FP is being used for therapy during exacerbations of RAO, administration of a short-acting bronchodilator such as albuterol a few minutes before administration of FP may increase efficacy.
Negative feedback effects of corticosteroids on adrenal function resulting in decreased COR concentrations previously have been reported during treatment of RAO with both DEX21 and inhaled beclomethasone dipropionate.22 In the treatment protocol of the present investigation, all 3 corticosteroid treatments decreased COR concentration, but the decrease was less with FP than with DEX. Neither of the FP doses decreased COR concentration as much as DEX did. By contrast, in the prevention trial, FP (6 mg q12h) decreased COR to the same extent as DEX. This difference in the results may have been because of differences in treatment duration or to the absence and presence of airway obstruction in the prevention and treatment protocols, respectively. The increased ΔPplmax characteristic of acute RAO exacerbations is caused in part by mucoid accumulations and bronchospasm that limit delivery of FP into the lung periphery.20 When ΔPplmax is <10 cmH2O as it was at the start of the prevention trial, the airways are much less obstructed and FP likely is delivered into the lung more effectively. Absorption from the lung into the systemic circulation may have been responsible for the negative feedback effects on the adrenal cortex and the resultant decrease in COR. In humans, the adrenal effects of FP absorbed from the gut or lung are limited by extensive 1st-pass liver metabolism but, despite this, suppression of COR production does occur and long-term FP administration suppresses the response to ACTH stimulation.1 The extent of hepatic metabolism of FP in horses is unknown.
Although previous investigations of the efficacy of corticosteroids for treatment of RAO have not mentioned laminitis as a complication, the fear persists, and for this reason, we were careful to check the feet of our horses daily for lameness, heat, and digital pulse. No signs of acute laminitis were observed in either the treatment or prevention protocols, and the scores for lameness, hoof temperature, and digital pulse did not increase during corticosteroid treatment. In its responses to vasoactive mediators, the hoof circulation behaves like a dermal vascular bed. The observation that DEX decreases skin blood flow and temperature by increasing the sensitivity of blood vessels to catecholamines makes it unlikely that hoof temperature will increase during DEX administration.23
In summary, inhaled FP administered at a high dose (6 mg) has beneficial effects for the treatment and, even more so, for the prevention of exacerbations of RAO. Future therapeutic trials of FP should incorporate pretreatment with a bronchodilator such as R-albuterol a few minutes before FP administration. Under such conditions, lower doses of FP may prove effective. However, the almost consistently beneficial effect of DEX in the treatment of RAO makes it the drug of choice for management of the RAO-affected horse with severe airway obstruction. Once the horse shows improvement, the horse can be maintained with FP. The dose of FP for prevention of RAO exacerbations needs further investigation. Although 6 mg twice daily is highly effective, it is likely that a lower dose can be used.
aEquineHaler, Stumpedyssevej 26, 2970 Hørsholm, Denmark
bTrojan condom, Carter-Wallace Inc, New York, NY
cTeflon catheter, A. Diagger Scientific Division, Wheeling, IL
dPressure transducer, Model DP/45-28, Validyne, Northridge, CA
ePortable physiograph, Dash model II and model 18, Astro-Med, West Warwick, RI
fSigma-Stat, Version 2.0, SysStat Software, San Jose, CA
Conflicts of interest: Other than the funding for this research project that was provided to Michigan State University, none of the authors have received any reimbursement from Stirling Products.