Abstract

Materials and Methods

This case–control study used serum stored at −80°C from horses recruited in a previous study on the relationship between bronchoalveolar cells and cytokine gene expression.1 Serum from thirty‐one client‐owned Standardbred racehorses were included in the study, 29 of them were part of Lavoie et al (2011) (some horses were excluded in one or the other study because serum or cells were not available). Briefly, and as described previously, horses presented for decreased performance and the clinical investigation included a physical examination, thoracic auscultation, lameness examination, complete blood count (available for 27 horses), plasma aspartate aminotransferase (AST) and creatine kinase (CK), upper airway endoscopy, and bronchoalveolar lavage. Trainers were requested to train at race speed the day before presentation and to refrain from administering medication 2 weeks prior to presentation. There were 13 mares, 12 geldings, and 6 males. Age ranged from 2 to 10 years in the IAD group (median: 4) and from 2 to 9 years (median: 3) in the non‐IAD group. Horses were considered as having IAD if, in addition to exercise intolerance, they had increased neutrophils (>5%), eosinophils (≥1%), or mast cells (≥2%), in their bronchoalveolar lavage (Exercise intolerant IAD group) while horses without lower airway inflammation formed the Exercise intolerant non‐IAD group. Horses with lameness or a suspected cardiac condition were excluded from the study but in both groups more than one potential cause of poor performance was identified (exercise‐induced pulmonary hemorrhage, suspected dorsal displacement of the soft palate, mildly elevated muscle enzymes). Experimental procedures were performed in accordance with the Canadian Council of Animal Care and approved by the Université de Montréal Animal Care Committee.

Results

There were no significant differences between groups (Exercise intolerant IAD versus Exercise intolerant non‐IAD), sex or age for any of the 3 proteins analyzed (Figs 1A, ​A,2A,2A, ​A,3A3A and ​and4),4), nor was there an association between the type of inflammation found in bronchoalveolar lavage and acute phase proteins (Figs 1B and ​and3B),3B), except for CRP concentrations in horses with and without increased neutrophils in their bronchoalveolar lavage fluid (P = .047) (Fig 2B). More specifically, median (range) for Exercise intolerant non‐IAD versus Exercise intolerant IAD were 3.47 (0.06–34.94) versus 6.33 (0.06–80) μg/mL (P = .49) for SAA, 10.87 (2.05–29.03) versus 4.63 (0.02–31.81) μg/mL (P = .23) for CRP and 900.36 (607.99–2018.84) versus 749.54 (530.81–1076.95) μg/mL (P = .09) for haptoglobin. Mean ± SD were 6.52 ± 10.18 versus 15.66 ± 23.50 μg/mL with a difference between the means of 9.14 (95% confidence interval (CI): −3.24 to 21.52) for SAA, 12.50 ± 9.63 versus 9.23 ± 10.99 μg/mL with a difference between the means of 3.27 (95% CI: −4.66 to 11.19) for CRP, and 968.49 ± 394.08 versus 766.32 ± 152.34 μg/mL with a difference between the means of 202.17 (95% CI: −61.59 to 465.93). Neutrophils and other cells percentages in the IAD and the non‐IAD groups are summarized in Table 1. Seven horses had SAA levels considered above normal values (> 20 μg/mL³), 6 of which were in the IAD group (no significant association, P > .2, Fisher exact test). Two horses with IAD had concentration above the detection limit and 5 horses had undetectable levels of SAA, 4 of them in the IAD group. Four horses had undetectable level of CRP, all of them in the IAD group. There was no correlation between concentrations of SAA, CRP, and haptoglobin, or between these parameters and peripheral leukocytes, neutrophils, fibrinogen, or bronchoalveolar lavage cytology (P values >.05). Muscle enzymes CK and AST were not correlated with acute phase proteins. There was a weak positive correlation between SAA and age (r = .36; P = .047).

Open in a separate window

Figure 1

(A) Serum amyloid A protein (SAA) in horses with inflammatory airway disease (IAD, closed circles, n = 21) and exercise intolerance from other cause (open circles, n = 10). (B) SAA in bronchoalveolar cytology with (closed circles) or without (open circles) elevated neutrophils, mast cells, and eosinophils (all horses). One bronchoalveolar lavage fluid can be classified in more than one category. Horizontal bar represents the median.

Open in a separate window

Figure 2

(A) C‐reactive protein (CRP) in horses with inflammatory airway disease (IAD, closed circles, n = 21) and exercise intolerance from other cause (open circles, n = 10). (B) CRP in bronchoalveolar cytology with (closed circles) or without (open circles) elevated neutrophils, mast cells, and eosinophils (all horses). One bronchoalveolar lavage fluid can be classified in more than one category. Horizontal bar represents the median.

Open in a separate window

Figure 3

(A) Haptoglobin in horses with inflammatory airway disease (IAD, closed circles, n = 21) and exercise intolerance from other cause (open circles, n = 10). (B) Haptoglobin in bronchoalveolar cytology with (closed circles) or without (open circles) elevated neutrophils, mast cells, and eosinophils (all horses). One bronchoalveolar lavage fluid can be classified in more than one category. Horizontal bar represents the median.

Open in a separate window

Figure 4

Protein concentrations of serum amyloid A protein (SAA), C‐reactive protein (CRP), and haptoglobin by sex (A) and age group (in years) (B) (all horses). M: male, G: gelding, F: mare. Horizontal bar represents the median.

Only one horse with IAD (of 27 with complete blood count available) had peripheral leukocytes and neutrophils above the reference range (leukocytes: 13 × 10⁹/L (5.5–12.5 × 10⁹/L), neutrophils 9 × 10⁹/L (2.7–6.7 × 10⁹/L)). This horse had low concentrations of SAA, CRP, and haptoglobin. All horses had normal fibrinogen concentration (≤4 g/L, n = 27). There was no difference in age and proportion of mares between Exercise intolerant IAD and Exercise intolerant non‐IAD horses (Fig 4).

Discussion

The main finding of this study was that in this population, SAA, CRP, and haptoglobin concentrations were not detectably different between exercise intolerant horses with and without IAD, or between horses with different types of cell population in their bronchoalveolar lavage, except for slightly lower concentration of CRP in horses with airway neutrophilia. It is therefore unlikely that these 3 acute phase proteins would be useful as a diagnostic marker of lower airway inflammation in horses presented for exercise intolerance.

Acute phase proteins can increase in blood with noninfectious and infectious respiratory conditions. High SAA concentration has been documented in horses with influenza virus, Streptococcus zooepidemicus, and Rhodococcus equi infections,11, 12, 13 and CRP is elevated in horses with pneumonia.14 Among noninfectious conditions, both SAA and haptoglobin are increased in horses with heaves compared to controls kept in the same environment.8 The lack of differences between IAD and non‐IAD horses studied here could be because of the lesser degree of inflammation in IAD compared to heaves,9 or because of a relatively high baseline level of inflammation in this population of poor performers. The higher levels and frequency of detection of SAA in this study compared to levels and frequency previously reported in heaves supports the latter hypothesis (90 and 81% of non‐IAD and IAD horses had detectable SAA in this population, while only in 17% of controls and 67% of heaves‐affected horses had detectable SAA levels8). Nevertheless, the lack of difference in SAA, CRP, and haptoglobin between horses with and without lower airway inflammation in the present study is in agreement with, and further extends, the previous findings that SAA and fibrinogen concentrations are not different between poorly performing coughing and noncoughing thoroughbred racehorses in training.15 Training alone should not result in high concentrations of acute phase proteins but they might increase in a population of poor performers maintained in training. Specifically, SAA do not increase with acute strenuous exercise alone16 or following training of experienced horses.17 However, it increases following prolonged strenuous endurance exercise and in inexperienced endurance horses.17, 18 Similarly, haptoglobin and fibrinogen increased only at the end of a 2.5‐month training program, suggesting that fatigue or subclinical disease developed over time.19 This suggests that training in adverse conditions could lead to an increase in acute phase protein concentrations and could account for the apparent high baseline SAA in our population of horses. Interestingly, haptoglobin was found to decrease 2 and 7 days following acute strenuous exercise in Standardbred racehorses.16 Since trainers were asked to train horses at race pace before presentation, it could have created an unexpected confounding factor for haptoglobin in our population. Other potential confounding factors that could have modified acute phase protein blood levels are the administration of anti‐inflammatory drugs and the presence of exercise‐induced pulmonary hemorrhage. Even if trainers were requested to refrain from administering medication 2 weeks prior to presentation, horses were not tested for anti‐inflammatory drugs (or other medication) before entering the study. Evidence of exercise‐induced pulmonary hemorrhage was very common in this population of racehorses in training, independently of their group (8/10 and 19/21 horses in the non‐IAD and IAD group, respectively). The lack of difference between groups, whether it is because the pulmonary inflammation causes little systemic inflammation or because our non‐IAD horses were also exercise intolerant would be interesting to explore but nevertheless, in a clinical setting where one would try to differentiate IAD from non‐IAD in poorly performing horses, the acute phase proteins studied here are not promising. More specific markers of lung inflammation such as serum surfactant protein D might be more interesting in that context.20

There was a small positive correlation between age and SAA concentration in this population and a trend for a difference between horses of 2–5 years of age and 6–10 years of age. Variation of SAA with age in healthy human adults is not considered clinically relevant,21 but in one study, horses over 8 years of age had higher level of SAA than horses of other age groups (except for neonates).22 The lower CRP concentrations in horses with bronchoalveolar lavage neutrophilia (Fig 2B), when compared to other horses from the study was unexpected, especially since neutrophilic asthma is associated with systemic inflammation and increased CRP in humans.23 The difference observed here was small and of unknown biological significance. Finally, from the data available on record, it was not possible to identify the origin of increased acute phase protein in outlier individuals, not even in the one horse who had a 2.2–3.1 times increase compared to control's mean in all three acute phase proteins measured. Of note, all horses had normal fibrinogen concentrations and this could support the use of other acute phase proteins in a population of horses with undetected inflammatory conditions without overt signs of infections, and in which no other cause of counter performance can be identified.

One potential pitfall of this study is the small number of samples available, which could have prevented the identification of a significant difference between groups (type II error). A posteriori power calculation shows that with this sample size and the variation observed, a 50% change would have been unlikely to be detected for SAA and CRP (effective power: 8% and 31%) and likely to be detected for haptoglobin (effective power: 99%). Therefore, further studies with larger sample sizes could find differences of this magnitude between groups. While such potential findings could be helpful in understanding the pathophysiology of IAD, it remains that the large overlap observed between groups make these proteins unlikely diagnostic tools on an individual case level.

In conclusion, in this population of horses, there were no difference in the acute phase proteins measured between horses with inflammatory airway disease and other horses with exercise intolerance. Serum amyloid A, CRP, and haptoglobin are unlikely to be useful markers of IAD in that context.

Acknowledgments

Notes

The work presented here was performed at the Université de Montréal and supported by the Natural Sciences and Engineering Research Council of Canada (grant # 415448) and by an unrestricted grant from the Pfizer Clinical Research Fund.

The majority (29 of 31) of the samples analyzed in his study came from horses that were enrolled in a previous report on the relationship between bronchoalveolar lavage cytology and cytokine gene expression.1

Footnotes

References