• Open Access

Plasma and Pulmonary Fluid Endothelin in Horses with Seasonal Recurrent Airway Obstruction

Authors

  • L.R.R. Costa,

    1. From the Equine Health Studies Program, Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University
    Search for more papers by this author
  • S.C. Eades,

    1. From the Equine Health Studies Program, Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University
    Search for more papers by this author
  • C.S. Venugopal,

    1. From the Equine Health Studies Program, Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University
    Search for more papers by this author
  • R.M. Moore

    1. From the Equine Health Studies Program, Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University
    Search for more papers by this author

  • Dr Costa is presently affiliated with Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA. Dr Moore is presently affiliated with Department of Veterinary Clinical Sciences College of Veterinary Medicine, The Ohio State University, Columbus, OH. A portion of the data was presented in abstract form at the 2003 International Veterinary Emergency and Critical Care Symposium.

Corresponding author: Lais R.R. Costa, MV, MS, PhD, DACVIM, DABVP, 200 Westboro Road, North Grafton, MA 01536-1895; e-mail: lais.costa@tufts.edu.

Abstract

Background: Summer pasture-associated recurrent airway obstruction (SPA-RAO), a seasonal airway obstructive disease of horses, is characterized by clinical exacerbation after exposure to pasture during warm months of the year. Endothelin (ET)-1, potent bronchoconstrictor, mitogen, secretagogue, and proinflammatory mediator, has been implicated in the pathogenesis of asthma and equine heaves.

Hypothesis: Immunoreactive ET-1 concentrations increase during clinical exacerbation and return to basal values during periods of disease remission.

Animals: Twelve horses, 6 affected with SPA-RAO and 6 nonaffected.

Methods: Prospective, observational study. Bronchoalveolar lavage fluid (BALF), arterial and venous plasma samples, and clinical variables were obtained from affected horses during clinical exacerbation and remission. Samples and data of nonaffected horses were collected during the summer and winter on dates similar to affected horses. Immunoreactive ET-1 was determined using a commercial ELISA.

Results: The median and range ET-1 concentrations (pg/ml) in arterial (1.3, 0.7–1.8) and venous (1.3, 1.2–1.7) plasma and in BALF (0.3, 0.2–0.4), and pulmonary epithelial lining fluid (PELF) (25.5, 21–50) were greater in affected horses during clinical exacerbation compared with remission (P < .01). The concentrations of immunoreactive ET-1 were greater in affected horses during clinical exacerbation compared with nonaffected horses (P < .05).

Conclusions and Clinical Importance: During clinical exacerbation of SPA-RAO, ET-1 is increased in circulation and pulmonary secretions. Intervention with ET receptor antagonists should provide further information on the role of ET-1 in SPA-RAO.

Abbreviations:
BALF

bronchoalveolar lavage fluid

CSRE

clinical scores of respiratory effort

ΔPplmax

maximal pleural pressure difference

EDTA

ethylenediaminetetra acetic acid

ET

endothelin

ETr

endothelin receptor

MDI

metered dose inhaler

PELF

pulmonary epithelial lining fluid

RAO

recurrent airway obstruction

SPA-RAO

summer pasture-associated recurrent airway obstruction

A seasonal, naturally occurring obstructive pulmonary disease affecting horses residing on pasture in the southeastern United States was first described in the early 1970s.1 The disease is termed summer pasture-associated (SPA) obstructive pulmonary disease because the clinical signs are observed after exposure to pasture during the warm and humid months of the year, and remission of signs is observed during cooler months of year.1,2 The clinical exacerbation of this seasonal recurrent airway obstruction (RAO) is characterized by bronchoconstriction, goblet cell metaplasia, airway inflammation with accumulation of mucoid secretion, and neutrophils into the airway, resulting in airway obstruction.3

Endothelins (ETs) are potent bronchoconstrictors, mitogens, secretagogues, and proinflammatory mediators.4 There are 3 isoforms of endothelin, ET-1, ET-2, and ET-3, encoded by 3 different genes. The lung is the major site of ET-1 and ET-3 expression, as well as ET clearance.5 The most studied isoform, ET-1, has been implicated in the pathogenesis of airway inflammatory diseases such as human asthma and equine barn-associated RAO.6,7

The overall objective of this study was to evaluate the variations in circulating and pulmonary immunoreactive ET-1 concentrations during clinical exacerbation and remission of SPA-RAO. We first described the exacerbation-remission pattern of the disease. Then, we determined and compared the concentrations of ET-1 in pulmonary secretions, arterial and venous plasma samples obtained from nonaffected control horses and horses affected with SPA-RAO during different times of the year. We hypothesized that immunoreactive ET-1 concentrations would be increased during clinical exacerbation of the disease and return to basal values during periods of disease remission.

Methods and Materials

Animals

The experimental protocol of this study was approved by the Institutional Animal Care and Use Committee of Louisiana State University. Twelve horses acquired by donation, including 6 affected (1 castrated male and 5 intact females; 3 Quarter Horses, 1 Paint, 1 Quarter Horse cross, and 1 Thoroughbred; median age 15 years, ranging from 7 to 27) and 6 nonaffected control horses (4 castrated males and 2 intact females; 5 Thoroughbred and 1 Quarter Horse; median age 15 years, ranging from 14 to 17) were maintained on pastures at Louisiana State University in Baton Rouge, LA. All horses received regular deworming and core immunizations (tetanus toxoid, and vaccines against EEE, WEE, WNV, and rabies). Before entering the study, all horses were considered healthy based on a complete physical examination, serum biochemistry profile, CBC, and fibrinogen. The control horses had been part of the research herd for more than 1 year. The affected animals originated from south Louisiana and were donated because they had a history of recurrent signs of obstructive airway disease during the summer. At the beginning of the study, all 12 horses were evaluated clinically, and indirect measurement of maximal pleural pressure difference (ΔPplmax) and cytologic analysis of bronchoalveolar lavage fluid (BALF) were performed. All SPA-RAO-affected horses had signs of clinical exacerbation, including increased end-expiratory effort and wheezing, that responded to the administration of a parasympatholytic bronchodilator. All affected horses while showing signs of clinical exacerbation had ΔPplmax > 15 cmH2O and neutrophilic inflammation (>20%) on BALF cytology. All control horses had ΔPplmax < 10 cmH2O and <15% neutrophils on BALF cytology.

Experimental Design

The study extended from August 2001 to December 2002. Evaluation of the breathing pattern was performed daily by a single individual using a previously described clinical scoring system.3 The clinical score of respiratory effort (CSRE) was calculated with the formula: CSRE = (lateral aspect of nostrils + medial aspect of nostrils)/2 + abdominal lift. The degrees of nostril flaring and abdominal lift were subjectively scored by assigning values from 0 to 4.

All horses were fed a complete pelleted diet, with fiber content of 30%,a and maintained in pasture conditions, except after the times of sample and data collection. After each sampling, the horses were maintained in box-stalls with rubber mat flooring, without bedding, and they were fed the same complete pelleted diet at 1.5 pounds of feed per 100 pounds of body weight, divided in 2 feedings. Sampling times throughout the course of the study are depicted in Figure 1. Sample collection of affected horses during clinical remission of signs and control horses during summer and winter was performed with an interval of at least 4 weeks between samplings. Sample collections of affected horses during clinical exacerbation were performed when the affected horses showed marked signs of airway obstruction, ie, CSRE ≥ 5 and presence of expiratory wheezes. At times of clinical exacerbation, affected horses were kept in the stalls for a period of at least 4 weeks after clinical improvement (ie, CSRE ≤ 4), before returning them to pasture. While confined, the horses were hand walked daily on a paved area. Medical treatments were only administered to affected horses when deemed necessary and for as short a period as possible, and included aerosolized bronchodilators (albuterol metered dose inhaler [MDI] 2 μg/kg q6h, ipratropium bromide MDI 2 μg/kg q12h, or both) given through an Aeromaskb or oral preparation of bronchodilator (albuterol oral syrup 8 μg/kg PO q12h), and corticosteroid given systemically (dexamethasone at a single dose of 0.02 mg/kg, IV, or an initial dose of 0.04 mg/kg, IV, followed by tapering doses IV q24h, for 2 days). The treatments and the length of time until recovery from clinical exacerbation were recorded for each horse.

Figure 1.

 Experimental design. Sampling times for summer pasture-associated recurrent airway obstruction (SPA-RAO) during exacerbation and remission (bottom) and nonaffected horses during summer and winter (top).

Clinical Data and Sample Collection

The clinical evaluations (physical examination, CSRE, and pulmonary function testing determined by ΔPplmax with an esophageal balloon) were recorded and samples (arterial and venous blood and BALF) were obtained from each affected horse at least 3 times during the season of clinical exacerbation (midspring to midfall) and 3 times during clinical remission (winter). In addition, plasma samples and clinical evaluations of nonaffected control horses were collected twice during summer and twice during winter. Clinical evaluation included recording breathing pattern and CSRE, vital signs, cardiac auscultation, auscultation of the lung fields during tidal breathing and auscultation with a rebreathing bag (except in affected horses during clinical exacerbation). Mean (for parametric data) and median (for nonparametric data) pooled values were calculated for all variables for each individual animal for the seasons (exacerbation and remission for affected horses, and summer and winter for control horses).

Pleural Pressure Measurement

Pleural pressure was measured indirectly, using an esophageal balloon secured over the end of a catheter that was connected to a pressure transducerc interfaced with a physiographd as previously described.3 The changes in esophageal pressure measured during tidal breathing reflected the changes in pleural pressure, and were recorded as ΔPplmax as reported by Derksen and Robinson.8 A latex balloon,e 10 cm in length and 1-cm diameter, was placed over the end of a 2 m long, 2-mm internal diameter cannula.f The balloon was inserted through a lubricated short nasogastric tube that was passed into the rostral esophagus. Once the balloon was located between the heart and the diaphragm, the nasogastric tube was removed. The balloon was inflated with 1.5 mL of water and 5 measurements (peak-inspiratory pressure minus peak-expiratory pressure) were collected and averaged. The normal ΔPplmax during tidal breathing was considered to be <10 cmH2O.8

Blood Sample Collection

Venous blood samples were obtained from the jugular vein and arterial blood samples were obtained from the transverse facial artery and placed into polypropylene tubes containing 1.375 mg of ethylenediaminetetra acetic acid (EDTA) and 0.072 trypsin inhibition units of aprotinin per mL of blood. Blood samples collected from the jugular vein were also placed into tubes containing lithium heparin for blood urea nitrogen determination. Blood samples were centrifuged at 2,000 ×g for 10 minutes. Plasma samples were frozen at −70°C until assayed.

BALF Collection

Horses were sedated using a combination of xylazine hydrochloride (0.5 mg/kg, IV) and butorphanol tartarate (11 μg/kg, IV). After placement of a nose twitch, a 244-cm (11-mm outer diameter, 3-mm inner diameter) flexible silicone tubeg was passed through the nasal passage into the trachea and wedged in the distal airway. As the silicone tube was advanced through the trachea and carina, 30 mL of 1% lidocaine was injected through the tube, followed by 30 mL of air. The cuff was inflated with 4 mL of air, and 5 × 60-mL aliquots of warm sterile 0.9% saline were infused manually with 60 mL syringes.9 Immediately after infusion, BALF was collected and pooled into a sterile flask. The volume of fluid infused and retrieved was recorded for each horse. A portion of the pooled aspirate was placed in vacuum tubes containing EDTA for cytologic evaluation and heparin for determination of urea concentration, and a portion was placed into polypropylene tubes containing 1.375 mg of EDTA and 0.072 trypsin inhibition units of aprotinin per mL of sample. The BALF samples were centrifuged at 2,000 ×g for 10 minutes, and cell-free BALF samples were frozen at −70°C until assayed.

BALF Cytologic Evaluation

The concentrated smears of the BALF were prepared by cytocentrifugation.h Air-dried smears were stained with a modified Wright's solution and 200 cells were classified using light microscopy under high-field magnification (× 1,000) as neutrophils, lymphocytes, alveolar macrophages, mast cells, eosinophils, or epithelial cells, and results were expressed as percentages.

Quantification of Immunoreactive ET-1 in Plasma and BALF

The concentration of immunoreactive ET-1 in plasma and cell-free BALF samples was determined using a commercially available capture ELISA,i which has been validated for use in the horse.10,11 Validation of the assay included tests of inter- and intraassay variability and recovery. The reported inter- and intraassay coefficients of variability for equine plasma were 15.4 and 6.4%, respectively.10 The inter- and intraassay coefficients of variability for supernatants from lavage were 6.9 and 4.5%, respectively (F. Garza, unpublished data). The characteristics of the assay include reactivity of 100% to ET-1 (21 amino acids), 100% to ET-2 (21 amino acids), <5% to ET-3 (21 amino acids), <1% to big ET (38 amino acids), standard curve range from 0.625 to 10 fmol/mL, and the sensitivity of 0.5 fmol/mL or 0.125 pg/mL.10 Samples with ET-1 below the limit of detection were assigned a value of zero.

The assay was performed following the manufacturer's instructions. Briefly, the plasma samples were mixed with the precipitating agent at 1 : 1.5 ratio, cooled to 4°C and centrifuged at 3,000 ×g for 20 minutes. The supernatants were transferred into polypropylene tubes, desiccated with nitrogen gas, and dissolved in 500 μL of assay buffer. For purification and concentration of ET in BALF, the samples were subjected to solid-phase extraction using reverse phase sorbent C18 columns.j,12–16 The kit's endothelin stock solution (lyophilized human ET-1) was diluted serially (ranging from 0 to 10 fmol/mL) to serve as standards. The enzyme immunoassay was performed in duplicates using 200 μL of each serial dilution of 6 standards, positive and negative controls, processed plasma samples or extracted BALF samples, which were added to each well. Then, 50 μL of the detection antibody (mouse anti-ET monoclonal antibody) was added to all wells except the blank well, mixed and incubated for 16–24 hours at 20°C. The contents of the wells were discarded, and the wells were washed 5 times with washing buffer. Horseradish peroxidase-conjugated rabbit antimouse IgG antibody was added to each well and the plate was incubated for 3 hours at 37°C. The contents of the wells were discarded, the wells were washed again 5 times with washing buffer, and 200 μL the tetramethylbenzidine substrate was added to each well, incubated for 30 minutes at 20°C protected from light. The acid solution, called stop solution by the manufacturer, 50 μL, was added to each well and mixed. The optical densities at 405 and 620 nm were measured immediately with an ELISA plate reader. The concentration of immunoreactive ET in plasma or BALF was calculated as per the manufacturer's instructions and expressed as pg/mL. The ET quantification was performed by an individual unaware of the status of the horse from which the sample was obtained.

Determination of Urea in Plasma and BALF

For the pulmonary secretion, the dilution of the pulmonary epithelial lining fluid (PELF) resulting from the lavage procedure was estimated using urea as a marker.17–19 Blood urea nitrogen concentrations were determined enzymatically.k Urea concentrations in BALF samples were determined with a commercially available kitl following the protocol modifications described by Rennard et al.17

For calculation of volume of PELF, the urea concentrations in plasma and BALF were used, such that we accounted for the dilution factor associated with the lavage. The volume of PELF was determined by the following formula: VPELF (mL) = VBALF× ([Urea]BALF (mg/mL)/[Urea]Plasma (mg/mL)).19,20 The concentration of immunoreactive ET-1 in PELF was calculated based on the urea concentrations in plasma and BALF and expressed as pg/mL of PELF: [ET-1]PELF (pg/mL) = ([ET-1]BALF× [Urea]Plasma)/[Urea]BALF.20

Statistical Analyses

Data, including the vital signs, ΔPplmax, the percentage of inflammatory cells in BALF, and the concentrations of immunoreactive ET-1 in arterial and venous plasma, BALF, and PELF (ie, the immunoreactive ET-1 concentration corrected for the dilution of the lavage using urea as marker) were compared among groups. The 4 groups were the affected horses during exacerbation and during remission of disease, and the nonaffected horses during summer and during winter. For analysis, mean (for parametric analysis) and median (for nonparametric analysis) values for each individual horse for the seasons were then used; giving us a sample size (n) of 6 for each of the 4 groups. Vital signs and the percentage of inflammatory cells in BALF were analyzed using a mixed effect linear model, in which the effects of sampling time and disease status were considered fixed, and the effect of horse was considered random. Means were separated using a Tukey posthoc test. Percentage data were transformed using the angular transformation (arcsine square root) for analysis. The CSRE, ΔPplmax, immunoreactive ET-1 concentrations in arterial and venous plasma, in BALF samples and in PELF were analyzed using the Kruskal-Wallis statistic followed by Dunn's multiple comparison test. Significant association between groups or episodes was considered at P < .05. Nonnormal data are summarized by the median and quartiles. All analyses were performed with the use of SAS v 8.2.m

Results

The clinical variables reflecting the severity of signs (CRSE, ΔPplmax, and percentage of neutrophils in BALF) in horses affected with SPA-RAO during clinical exacerbation were significantly greater than during remission and greater than nonaffected horses in either summer or winter sampling periods (Figs 2–4, Table 1). Rectal temperatures, heart and respiratory rates were significantly greater in affected horses during clinical exacerbation compared with remission and to control horses (Table 1). As vital signs were obtained while the horses were in the stocks in the laboratory, anxiety associated with the restraint lead to the overall relatively high heart rates even in the control animals. Nonetheless, affected horses had significantly higher heart rate during clinical exacerbation.

Figure 2.

 Clinical score of respiratory effort (CSRE) of horses during exacerbation and remission of summer pasture-associated recurrent airway obstruction, and nonaffected horses during summer and winter. Box indicates 50% of the data, the whiskers represent the range, and the horizontal line within the box represents the median values. Different letters indicate statistical significance (P < .0001).

Figure 3.

 Pleural pressure difference (ΔPplmax, in cm of H2O) of horses during exacerbation and remission of summer pasture-associated recurrent airway obstruction, and nonaffected horses during summer and winter. Box indicates 50% of the data, the whiskers represent the range, and the horizontal line within the box represents the median values. Different letters indicate statistical significance (P < .0001).

Figure 4.

 Percentage of neutrophils in bronchoalveolar lavage fluid of horses during exacerbation and remission of summer pasture-associated recurrent airway obstruction, and nonaffected horses during summer and winter. Data points and mean are shown. Different letters indicate statistical significance (F = 79; 3, 20 df; P < .0001).

Table 1.   Summary of mean ± SD for vital signs and BALF cytologic findings.
GroupsAffected Horses
during Exacerbation
Affected Horses
during Remission
Nonaffected Horses
during Summer
Nonaffected Horses
during Winter
  • Letters indicate significant difference among groups:

  • a,b

    Different letters indicate statistical significance (F = 13.2; 3, 20 df; P < .0001).

  • c,d

    Different letters indicate statistical significance (F = 17.5; 3, 20 df; P < .0001).

  • e,f

    Different letters indicate statistical significance (F = 4.0; 3,20 df; P < .05).

  • g

    Differences among groups were not significantly different (F = 1.2; 3, 20 df; P > .3).

  • h,i,j

    Different letters indicate significant difference (F = 79; 3, 20 df; P < .0001).

  • k,l,m

    Different letters indicate significant difference (F = 64; 3, 20 df; P < .0001).

  • n,o

    Different letters indicate significant difference (F = 14; 3, 20 df; P < .0001).

  • q

    Differences among groups were not significant (F = 1.8; 3, 20 df; P > .1).

  • BALF, bronchoalveolar lavage fluid; HR, heart rate; RR, respiratory rate.

Rectal temperature (°F)101.3 ± 1.2a99.8 ± 0.7b100.1 ± 0.9b99.5 ± 0.9b
RR (breaths per minute)33 ± 8c20 ± 2d20 ± 4d15 ± 5d
HR (beats per minute)59 ± 6e51 ± 3f48 ± 4f50 ± 6f
BALF volume recovery55 ± 2%g50 ± 4%48 ± 7%48 ± 6%
Neutrophils in BALF63 ± 8%h14 ± 6%i8 ± 3%i4 ± 2%j
Macrophages in BALF10 ± 4%m36 ± 2%l34 ± 2%l50 ± 2%k
Lymphocytes in BALF19 ± 6%n43 ± 9%o52 ± 4%o39 ± 6%o
Mast cells in BALF1.7 ± 0.8%q4.0 ± 2.5%2.2 ± 0.6%1.8 ± 0.6%

During episodes of clinical exacerbation, thoracic auscultation of affected horses revealed wheezes, generally end-expiratory and in some cases inspiratory and expiratory; a rebreathing bag was not used. During clinical remission, which occurred in the winter, horses affected with SPA-RAO evaluated in this study had ΔPplmax < 15 cmH2O and CRSE ≤ 4, similar to values in the nonaffected horses, indicating reversibility of the airway obstruction. Moreover, during clinical remission, affected horses had normal breath sounds, and no wheezes or crackles even with use of a rebreathing bag. In all instances, nonaffected control horses had normal breath sounds upon auscultation of the lung with a rebreathing bag. The median CSRE for affected horses during exacerbation (median 5.5; range 5–6) was significantly (P < .001) greater than the CSRE during clinical remission (median 3; range 2–3.5), and the CSRE of nonaffected controls during summer (median 2.5; range 2–3) and during winter (median 2; range 1.5–2) (Fig 2). The ΔPplmax for affected horses during exacerbation (median 33.5 cmH2O, range 25–54) was significantly greater than during clinical remission (median 7 cmH2O, range 6–8) (P < .05), and nonaffected controls during summer (median 4 cmH2O, range 4–5) (P < .05) and during winter (median 4 cmH2O, range 3–5) (P < .001) (Fig 3). During clinical remission the signs of airway obstruction subsided, such that CSRE ≤ 4.0 and ΔPplmax < 10.0 cmH2O in most sampling times of affected horses during the winter. The percentage of neutrophils in BALF for affected horses during exacerbation was significantly greater than the percentage of neutrophils in BALF during clinical remission, and that of nonaffected controls during summer and winter (Table 1 and Fig 4). The percentages of macrophages and lymphocytes in BALF were correspondingly decreased during exacerbation.

The onset of clinical exacerbation varied among horses; the 1st clinical exacerbation episode of the season for each affected horse occurred in May, June, July, and August for 4 of the horses and September for 2 horses. Once the affected horses had recovered from their 1st episode of clinical exacerbation of the year, which meant 4 weeks of confinement in low-dust environment after the CSRE dropped to 4, the affected horses were placed back on pastures. Invariably, once the affected horses were again exposed to the pasture, the signs of clinical exacerbation returned within 2 days. The length of remission for each horse were 9 months for horse #1, inline image months for horses #2 and #3, inline image months for horse #4, 5 months for horse #5, and 6 months for horse #6. The number of episodes of clinical exacerbations and the number of episodes requiring treatments were 3 (none requiring treatment) for horse #1, 2 (none requiring treatment) for horse #2, 5 (2 requiring treatment) for horse #3, 3 (all requiring treatment) for horse #4, 5 (all requiring treatment) for horse #5, and 6 (all requiring treatment) for horse #6. The time to recover was 3–7 days for horse #1, 4–6 days for horse #2, 2–4 days for horse #3, 4–8 days for horses #4 and #5, 5–12 days for horse #6.

The severity of clinical signs during exacerbation episodes varied among affected horses. Three horses (horses #4–6) were considered more severely affected, because they had shorter length of remission, and after each exacerbation episode they required days of medical treatment in addition to the change to a dust-free environment, and 3 horses appeared moderately affected (horses #1–3), as they required no or minimal medical treatment, and their CSRE decreased within a few days of confinement in the dust-free environment. Comparison between clinical variables during clinical exacerbation of severely affected versus moderately affected horses revealed that the median rectal temperature (101.6 and 100.6°F, respectively), heart rate (60 and 54 beats per minute, respectively) and respiratory rate (40 and 24 breaths per minute, respectively), respiratory effort (CRSE of 6 and 5.25, respectively) tended to be higher in severely affected horses compared with moderately affected horses. Whereas their pulmonary function (ΔPplmax of 34 and 33 cmH2O) and percentage of neutrophils were as high as in the moderate affected as in the severely affected horses.

The immunoreactive ET-1 concentrations in venous plasma samples were significantly greater in affected horses during clinical exacerbation, compared with remission and nonaffected horses (Fig 5A). The immunoreactive ET-1 concentrations in arterial plasma samples were significantly greater in affected horses during clinical exacerbation, compared with those during clinical remission, but not compared with nonaffected controls (Fig 5B). The concentrations of immunoreactive ET-1 in BALF samples and PELF were significantly greater, in affected horses during clinical exacerbation compared with remission and those from nonaffected horses during summer (Fig 5C and D). There were no significant differences of ET-1 concentrations in BALF samples and PELF from affected horses during remission and those from nonaffected control horses during summer and winter. There was considerable variability in the concentrations of immunoreactive ET-1 in plasma samples (venous and arterial), as well as in pulmonary secretion (BALF and PELF) within each of the 4 groups.

Figure 5.

 Circulating and pulmonary fluid concentrations of immunoreactive ET-1. Endothelin-1 (ET) concentrations (pg/mL) in venous plasma (A), arterial plasma (B), bronchoalveolar lavage fluid (BALF) (C), and pulmonary epithelial lining fluid (PELF) (D). Box indicate 50% of the data, the whiskers represent the range, and the horizontal line within the box represents the median values. Different letters indicate statistical significance (P < .0001).

Comparison between ET-1 concentrations during clinical exacerbation of severely affected versus moderately affected horses revealed that the median immunoreactive ET-1 in BALF (0.32 and 0.02 pg/mL, respectively) and PELF (23.7 and 5.2 pg/mL, respectively) tended to be greater in the severely affected than in moderately affected horses, but the differences were not significant (P > .2 and P > .4, respectively). The ET-1 concentrations during clinical exacerbation of severely affected compared with moderately affected horses in venous plasma samples (1.23 and 1.08 pg/mL, respectively) and arterial plasma samples (1.25 and 1.20, respectively) did not differ.

Discussion

This study reports the clinical variables and concentrations of circulating and pulmonary ET during exacerbation and remission of SPA-RAO during the seasons of the year in comparison with nonaffected horses. In the present study, the circulating concentrations of ET-1 in affected horses during exacerbation were greater in venous and arterial plasma compared with remission, and only in venous plasma compared with nonaffected controls. Inconsistent and controversial systemic ET results have been reported in human asthmatic patients, where some studies report significant differences between asthmatics and nonasthmatics.6,21 In contrast, no difference between ET-1 concentrations in plasma samples from asthmatics and control subjects is reported by others.22 The circulating venous concentration of ET-1 was greater in barn-associated RAO-affected horses during crisis compared with remission and control samples.7

In the present study, the concentrations of ET-1 were significantly greater in BALF and PELF from affected horses during exacerbation, compared with remission, and nonaffected controls during the summer. Our results resemble the findings in horses affected with barn-associated RAO, whereby, despite the great variation of ET-1 concentration in affected horses during crisis, the controls have significantly lower ET concentrations.7 Similarly, our results also resemble those reported in human asthmatic patients where immunoreactive ET-1 concentrations are found to be greater in BALF samples of asthmatic patients compared with normal subjects.13,14,21 Similar to the present study, all of the aforementioned studies reported that the BALF samples were subjected to extraction using SE-Pak C18 columns, and that urea was utilized as a marker of dilution of the lavage. However, in contrast with the present study, the reported studies utilized a radioimmunoassay for determining ET-1 concentration, which has the lowest detectable limit (0.1 pg/mL), and it is considered to be slightly more sensitive than the ELISA (which has the lowest detectable limit as 0.125 pg/mL).

The severity of the clinical signs during exacerbation of SPA-RAO varied considerably among horses in this study, especially with respect to response to environmental management. Moreover, the concentrations of immunoreactive ET-1 in pulmonary secretion, BALF, and PELF, from affected horses during exacerbation had a wide variation between individual samples. It appeared that the more severely affected horses tended to have greater ET-1 in pulmonary secretions than the moderately affected horses. Although, individual variation is 1 of the limitations of studies performed in subjects affected with naturally occurring disease, they do provide invaluable information about the clinical spectrum of the disease and highlights the importance of further studies to evaluate the stratification of how changes in inflammatory mediators such as ET-1 relate to the severity of the disease. In addition, it is unclear if, despite the interval of at least 4 weeks after clinical recovery, the medical treatments had any long-lasting affect altering the ET-1 synthesis and release into the circulation or pulmonary secretions during the subsequent clinical exacerbation periods.

ET-1 is secreted by a number of cell types and interacts with the target receptors in an autocrine/ paracrine as well as endocrine modes of action. The increase in pulmonary as well as circulating immunoreactive ET-1 concentration during clinical exacerbation of SPA-RAO suggests paracrine and endocrine types of interaction and it resembles the findings in human asthma and barn-associated equine RAO.7,22 This increase in pulmonary and circulating ET-1 may result from increased ET-1 synthesis or release, delayed elimination, decreased degradation of ET-1, or a combination of these mechanisms. Interactions of ET-1 with its receptors are likely to be important with respect to physiologic effects of ET-1 in pulmonary tissue as well as clearance of ET-1 by the pulmonary system.

Endothelin receptor (ETr) B and, to a lesser extent, ETrA is overexpressed in bronchial smooth muscle of horses affected with SPA-RAO compared with sections from control horses.23 We speculate that even small increases in pulmonary and circulating ET-1 are likely to result in greater response in affected horses owing to the increase in receptor expression.

The overall increase in pulmonary and plasma ET-1 during clinical exacerbation of SPA-RAO suggests that ET-1 is possibly involved in the pathophysiology of this disease and could contribute to the bronchoconstriction, airway inflammation, and accumulation of neutrophilic mucoid secretion observed during clinical exacerbation of SPA-RAO. Intervention with ET receptor antagonists should provide further information on the role of ET-1in SPA-RAO. If ET-1 indeed plays a role in the pathogenesis of SPA-RAO, intervention in ET-1 production and metabolism, and antagonism of ETr may prove to be viable targets for pharmacological intervention and alleviation of signs associated with clinical exacerbation of the disease in affected horses.

Footnotes

aHorse Chow 200, Purina Mills, St Louis, MO

bTrudell Medical International, London, UK

cStatham Model P50 pressure transducer, Statham Instruments, Hato Ray, Puerto Rico

dModel 7D polygraph, Grass Medical Instruments, Quincy, MA

eLatex penrose tubing, Sherwood Medical Company, St Louis, MO

fCannula, PE 350 tubing, VWR Scientific Products, Willard, OH

gEquine broncho-alveolar lavage catheter, Bivona Inc, Gary, ID

hCytospin 3, Shandon Inc, Pittsburgh, PA

iBiomedica Gruppe, Austria, Distributed by American Research Products Inc, Belmont, MA

jSep-Pak C18 cartridge, Waters, Milipore Corporation, Temecula, CA

kOlympus AU600 Chemistry Analyzer, Melville, NY

lInfinityTM Urea Nitrogen, Thermo DMA, Louisville, CO

mSAS Institute, Cary, NC

Acknowledgments

This research project was conducted at the School of Veterinary Medicine, Louisiana State University, and supported by grants from the American Association of Equine Practitioners and the Equine Health Studies Program from Louisiana State University.

The authors thank Frank Garza for performing the daily CRSE and the ET quantification, Dr Giselle Hosgood for statistical advice, Dr Steve Gaunt for cytologic evaluation of BALF, Dr Kathy O'Reilly for the urea quantification, and Catherine Koch, Mike Keowen, Danielle Vallotton, Leslie Talley, Tarra Harden, Diane Savois, Elizabeth Dequeant, Jeffrey Cardinale, and Jenny Liford for their assistance in caring for the animals.

Ancillary