Endurance riding is a strenuous equestrian discipline; hence, much effort is spent on monitoring horses to ensure that their welfare is not compromised. The metabolic consequences of an endurance ride have been identified in several studies1–9 of exercise-induced changes in hematologic and biochemical variables in horses. Some of these studies1,5,7 also involved examination of the association between heart rate and biochemical variables and found an increased heart rate to be indicative of blood volume depletion and skeletal muscle fatigue. Clinical variables such as heart rate and cardiac recovery index are therefore commonly used to assess the metabolic status of the horses and their capability to continue the ride.10,11 However, whereas previous studies primarily focused on heart rate during the ride, to the authors' knowledge no investigations have been conducted into changes in heart rate during the recovery period.
A study12 involving human endurance athletes revealed an increase in heart rate and decrease in HRV subsequent to a long-distance run. Endurance horses are likewise anecdotally believed to have an increase in heart rate for hours after an endurance ride, which could suggest an altered sympathetic-parasympathetic balance resulting in a decrease in HRV and increase in susceptibility to arrhythmias. Frequent occurrence of supraventricular and ventricular premature beats has been identified in trotting horses during the immediate postrace recovery phase and is most likely related to vagal reactivation during deceleration of the heart rate.13,14 Nevertheless, despite the known systemic effects of endurance riding, no studies have been conducted to assess HRV or frequency of arrhythmias in endurance horses during recovery after an endurance ride.
Studies15,16 involving human endurance athletes have revealed increases in circulating concentrations and activities of cardiac biomarkers following exercise, leading to the hypothesis that prolonged exercise in humans affects the myocardium. A correlation has even been identified between magnitude of cardiac biomarker release and heart rate and HRV during recovery.12 Increases in plasma cTnI concentration have also been reported for horses after endurance rides, but in the associated study17 an association with heart rate was not investigated. The objective of the study reported here was to investigate changes in heart rate, HRV, and arrhythmia frequency in horses during the initial 12 hours of recovery after completing a 120- or 160-km endurance ride. A second objective was to evaluate changes from preride values in serum cTnI concentration and CK-MB activity after the ride and their potential correlation with heart rate.
Materials and Methods
Horses and competitions
Data were collected at 3 endurance competition venues from April to June 2012 in Glimåkra, Sweden, and Gartow and Nörten-Hardenberg, Germany. The weather and terrain were comparable at the 3 venues. Twenty-eight Arabian horses (4 stallions, 11 mares, and 13 geldings) with a mean ± SD age of 11 ± 3 years and body weight of 419 ± 32 kg (922 ± 70 lb) were enrolled from among teams competing in 120- or 160-km endurance rides (Concours de Raid d'Endurance International 2-star or 3-star). Written informed consent was obtained from horse owners. Ethical approval and permission to perform the study were attained from national governmental authorities when required (license No. M 115-12/Dnr 31-3234-12).
All horses passed a veterinary inspection on arrival at the venues the day before the ride. Echocardiographic examinations were performed to exclude horses with cardiac abnormalities; however, horses with mild valvular regurgitation were included.18 Echocardiography was repeated after the ride (data not reported). Competition results (eg, timekeeping data or speed) were obtained from the official records. The time when horses entered their final veterinary inspection was used to represent the finishing time of their ride.
ECG recording
Electrocardiographic recordings were obtained with a Holter recording systema with 2 channels and bipolar leads. The ground electrode and the combined negative electrode of leads I and II were placed dorsally over the region of the right scapula. The positive electrode of lead I was placed on the left side approximately 20 cm dorsal to the olecranon, and the positive electrode of lead II was placed to the left of the ventral midline and caudomedial to the pectoral muscles. Thus, lead I was optimized for deflections of atrial depolarization,19 whereas lead II was optimized for ventricular depolarization. The area was clipped of hair, and electrodes were secured with adhesive foam patches.b Preride recordings of approximately 4 hours' duration were obtained the evening before the ride. Recording of recovery ECGs commenced approximately 1 hour after horses either were eliminated from or finished the ride and continued for 12 hours after the ride or until the next morning (when the horses left the venue), whichever came first. During the recordings, horses were primarily confined to their box stalls where they ate and rested; however, owners were allowed to groom and walk their horses in accordance with their usual competition routines.
ECG analyses
The ECG recordings were divided into 30-minute periods. For every period, HRmean was calculated as the mean of normal (nonarrhythmic) beats, whereas HRV was determined by the SDNN.c Additionally, each period was subdivided into 5-minute sequences to assess the HRmin and HRmax of every 30-minute period. The 30-minute period between 8:00 pm and 9:00 pm in the preride recordings (when horses were at rest in the stables) with the most regular heart rate as assessed by RR tachograms was used as a reference value when recovery values were analyzed. For study purposes, the recovery period started when horses finished the ride and continued for 12 hours afterward.
Frequency of arrhythmias was assessed by semiautomatic ECG analysis and manual classification of SVPCs and VPCs as described elseswhere.13 Prematurity of beats was defined in accordance with the underlying heart rate. At heart rates < 60 beats/min, a shortening of the R-R interval by ≥ 16% was considered arrhythmic; at heart rates from 60 to 100 beats/min, the threshold was 10%; and at heart rates > 100 beats/min, the threshold was 4%. Supraventricular and ventricular tachycardia were defined as ≥ 4 consecutive SVPCs or VPCs, respectively.20
Blood samples
Blood samples were obtained from a jugular vein via 20-gauge needles at 3 points: the day before the ride, approximately 45 minutes after horses either finished the ride or were eliminated, and the morning after the ride. Because horses finished the ride at different times, intervals between the 3 sample collection points varied among horses. Samples were collected in 10-mL EDTA tubes, 8.5-mL serum separator tubes, and 4.5-mL sodium citrate tubes.d Blood samples in serum separator tubes were allowed to coagulate before centrifugation at 1,800 × g for 10 minutes. Half of each harvested serum sample was transferred into 2-mL cryotubese and frozen on dry ice until storage at −80°C. The remaining serum, EDTA-stabilized blood, and citrated plasma samples were kept in a cooler box until arrival at the laboratory. The EDTA-anticoagulated blood samples were used for a CBC,f fresh serum samples were used for biochemical analysis,g and citrated plasma samples were used for fibrinogen concentration measurement.h
Seven to 9 months after collection, stored serum samples were used for measurement of cTnli concentration and CK-MB activity.j Measurements of these cardiac biomarkers were performed twice, and the mean value was used in subsequent analyses. When the mean serum cTnI concentration was lower than the detection limit (0.003 ng/mL), a concentration of 0.003 ng/mL was assigned.21 Seventeen to 19 months after collection, stored serum samples were also used for measurement of haptoglobin concentration.k
Statistical analysis
Data for horses competing in 120- and 160-km endurance rides were processed together. Recovery values for HRmean, HRmin, HRmax, SDNN, and hematologic tests of horses that completed the ride were compared with preride values by means of repeatedmeasures mixed-model ANOVA with a Dunnett test for post hoc comparisons.l Repeated measures were accounted for by including time (hour or day) with an autoregressive covariance type 1 structure. Individual horse was included as a random effect. Model residuals were evaluated for independence and normal distribution. For variables that were not normally distributed, logarithmic transformation of the data was performed. Blood parameters were analyzed both with and without correction for hydration status as assessed by the percentage change in serum protein concentration.8 Associations between cardiac biomarkers and distance, velocity, and HRmean (during the initial 3 hours of the recovery period) were tested by means of linear regression. For horses completing the ride, numbers of premature beats were measured in the preride and recovery ECG recordings and were normalized per hour to correct for different recording durations. The hourly frequency of premature beats deviated from a normal distribution, and analyses were therefore performed with the Wilcoxon signed rank test. Values of P < 0.05 were considered significant.
Results
Horses
Twenty-eight horses were included in the study. Thirteen competed in a 120-km endurance ride, and 15 competed in a 160-km endurance ride (Figure 1). Nine of 13 horses completed the 120-km ride at a mean ± SD speed of 14.2 ± 2.5 km/h; 10 of 15 horses completed the 160-km ride at a mean speed of 14.7 ± 2.1 km/h. The remaining 4 horses in the 120-km ride and 5 horses in the 160-km ride completed a mean of 78 ± 24 km (range, 30 to 100 km) before they were eliminated. Causes of elimination included lameness (n = 6), metabolic concerns (2), and withdrawal by the rider (1). One of the horses eliminated for metabolic reasons received fluid therapy immediately after the ride. None of the horses received medical treatment overnight.
Flow diagram of Arabian horses included in a study involving assessment of cardiac variables in horses competing in a 120- or 160-km endurance ride.a Because of practical limitations, the use of ECG recorders was prioritized for horses in the 160-km ride. Therefore, not all horses in the 120-km ride had ECG readings.b Horses eliminated during the early stages of the ride were allowed to leave the venue directly, and a recovery ECG could therefore not be obtained.c The preride ECG of one horse and the recovery ECG of another horse were not available because of technical issues.
Citation: Journal of the American Veterinary Medical Association 248, 9; 10.2460/javma.248.9.1034
Echocardiographic examination revealed unremarkable left ventricular dimensions in all 28 horses. A few horses had mild insufficiencies of the pulmonary valve (n = 4), aortic valve (4), mitral valve (1), or tricuspid valve (1). There were no significant differences in echocardiographic findings between horses that comheart pleted the ride and those that were eliminated, and no horses were excluded from the study on the basis of echocardiographic findings.
ECG analyses
For analyses of ECG data, including heart rate, HRV and arrhythmias, only data regarding the 19 horses that completed the ride were included; data for the 9 eliminated horses were excluded. Recording of recovery ECGs commenced a mean of 58 ± 31 minutes after horses finished the ride. Plots of the instantaneous rate against time revealed heart rate fluctuations with different characteristics before and after the ride. Preride recordings generally included momentary peaks of large heart rate increases, whereas recovery recordings were dominated by less prominent but longer-lasting heart rate increases (Figure 2). Comparisons of preride and recovery values for heart rate and SDNN revealed that HRmean, HRmin, and HRmax were significantly increased and that SDNN was significantly decreased for approximately 10 hours during the recovery period (Figure 3).
Plots of instantaneous heart rate over time during preride (A) and postride (recovery [B]) periods for a representative Arabian horse that completed a 160-km endurance ride. Notice the transient peaks in heart rate in the preride recording, compared with the longer-lasting but less prominent heart rate increases during the recovery period. An overall decrease in heart rate is evident during recovery.
Citation: Journal of the American Veterinary Medical Association 248, 9; 10.2460/javma.248.9.1034
Mean HRmean (A), HRmin (B), HRmax (C), and HRV assessed by SDNN (D) during the initial 12 hours of recovery for 12 Arabian horses that completed a 120- or 160-km endurance ride. Each mark on the graphs refers to the value of the preceding 30-minute period. Because of varying starting times and durations of the recovery ECG recordings, the mean values of the initial 2 and final 2 hours of the recovery period were calculated on the basis of the smaller number of horses with ECG recordings available at the time. Error bars represent SD. *Value differs significantly (P < 0.05) from preride value.
Citation: Journal of the American Veterinary Medical Association 248, 9; 10.2460/javma.248.9.1034
None of the 12 horses that completed the ride and had ECG recordings made had VPCs in the preride recordings, whereas 6 had VPCs during recovery (Table 1). Supraventricular premature complexes were identified in 5 of 12 preride recordings and 11 of 12 recovery recordings. The normalized frequency of VPCs (premature beats per hour) was higher, albeit not significantly (P = 0.062), during the recovery period, whereas the frequency of SVPCs was not significantly different between preride and recovery recordings. The number of premature beats in individual horses was generally low; however, 1 horse had multiple arrhythmias during recovery. This horse completed a 160-km ride and had no premature beats in the preride recording but had 200 SVPCs (including 7 pairs, 1 triplet, and 1 episode of supraventricular tachycardia) and 27 VPCs (including 8 interpolated VPCs) during recovery (Figures 4 and 5). Overall, 6 horses had VPCs and 2 of those horses had multifocal ectopy (Figures 6 and 7).
Distribution of SVPCs (gray bars) and VPCs (white bars) during the recovery period of the horse in Figure 3 that had the highest number of premature beats during that time. The second hour comprises only 43 minutes of recording.
Citation: Journal of the American Veterinary Medical Association 248, 9; 10.2460/javma.248.9.1034
Lead II ECG recording of the same horse as in Figure 4 obtained during the sixth hour of the recovery period. Notice the VPC interpolated between 2 sinus beats (first arrowhead) followed by another VPC (second arrowhead). Numbers = Instantaneous heart rates. Paper speed = 25 mm/s. Gain = 10 mm/mV.
Citation: Journal of the American Veterinary Medical Association 248, 9; 10.2460/javma.248.9.1034
Representative lead I, II, and III ECG recordings of the same horse as in Figure 4, showing different VPC morphologies, suggesting multifocal ectopy. A—This morphology was observed for 25 VPCs, of which 8 occurred as interpolated. B—This morphology was observed for 2 VPCs. Paper speed = 50 mm/s. Gain = 10 mm/mV
Citation: Journal of the American Veterinary Medical Association 248, 9; 10.2460/javma.248.9.1034
Lead I, II, and III ECG recordings showing 2 morphologies (A and B) of the 3 VPCs observed in another horse in Figure 3 that had multifocal ectopy. Paper speed = 50 mm/s. Gain = 10 mm/mV.
Citation: Journal of the American Veterinary Medical Association 248, 9; 10.2460/javma.248.9.1034
Number of premature beats in ECG recordings obtained from 12 Arabian horses before (preride) and for up to 12 hours after (recovery) completion of a 120- or 160-km endurance ride.
| Preride | Recovery | ||||||
|---|---|---|---|---|---|---|---|
| Variable | Mean | Median | Range | Mean | Median | Range | P value |
| Duration (h) | 3.8 | 3.8 | 2.3–5.2 | 10.1 | 10.6 | 6.6–11.5 | — |
| SVPCs | |||||||
| Total No. | 1 | 0 | 0–10 | 19 | 2 | 0–200 | — |
| No./h | 0.3 | 0.0 | 0.0–2.6 | 1.8 | 0.3 | 0.0–18.7 | 0.63 |
| VPCs | |||||||
| Total No. | 0 | 0 | 0–0 | 3 | 1 | 0–27 | — |
| No./h | 0.0 | 0.0 | 0.0–0.0 | 0.3 | 0.0 | 0.0–2.5 | 0.06 |
Blood sample analysis
The 19 horses that completed the ride contributed 56 blood samples for analysis (3 samples/horse; 1 sample was missing the morning after the ride). The first set of recovery blood samples was obtained a mean of 36 ± 26 minutes after horses completed the ride, and the second set was obtained the following morning a mean of 12.7 ± 2.6 hours after the ride.
Serum cTnl concentration was significantly (P < 0.001) greater after the ride and CK-MB activity was significantly greater both after the ride (P < 0.001) and the next morning (P < 0.001), compared with preride values (Table 2). No significant linear correlations were identified between values of these cardiac biomarkers and velocity, distance, or HRmean.
Mean (95% confidence interval) values of clinicopathologic variables at various points for Arabian horses competing in a 120- or 160-km endurance ride.
| Variable | Preride (n = 19) | Ride completion (n = 19) | Morning after (n = 18) |
|---|---|---|---|
| Cardiac biomarkers | |||
| cTnl (ng/mL)* | 0.004 (0.003–0.005) | 0.010 (0.006–0.016)† | 0.005 (0.003–0.007) |
| CK-MB (ng/mL) | 3.18 (2.61–3.75) | 6.53 (5.06–8.01)† | 5.50 (4.88–6.12)† |
| CBC analytes | |||
| Total leukocyte count (× 109 cells/L) | 8.91 (8.30–9.53) | 13.30 (12.14–14.47)‡ | 10.98 (9.92–12.04)† |
| Neutrophil count (× 109 cells/L) | 6.06 (5.40–6.71) | 11.92 (10.63–12.78)† | 8.91 (7.81–10.01)† |
| Lymphocyte count (× 109 cells/L) | 2.14 (1.90–2.37) | 1.08 (0.97–1.19)† | 1.41 (1.21–1.60)† |
| Eosinophil count (× 109 cells/L) | 0.25 (0.19–0.32) | 0.04 (0.02–0.06)† | 0.08 (0.04–0.12)† |
| Thrombocyte count (× 1012 platelets/L) | 143.3 (118.1–168.4) | 163.9 (140.7–187.2)‡ | 158.1 (134.5–181.7) |
| Erythrocyte count (× 1012 RBCs/L) | 8.55 (8.17–8.92) | 9.13 (8.59–9.66)‡ | 8.70 (8.22–9.19) |
| Hemoglobin (mmol/L) | 8.70 (8.29–9.11) | 9.31 (8.76–9.86)‡ | 8.78 (8.33–9.23) |
| Hct (%) | 41.14 (39.26–42.02) | 43.94 (41.29–46.60)‡ | 41.07 (39.02–43.12) |
| Serum analytes | |||
| Protein (g/L) | 70.21 (67.88–72.54) | 70.75 (68.10–73.41) | 66.77 (64.19–69.34)† |
| Albumin (g/L) | 38.45 (37.49–39.41) | 39.25 (37.81–40.69) | 36.86 (35.73–37.98)† |
| Bilirubin μmol/L) | 23.76 (20.68–26.84) | 67.89 (58.00–77.78)† | 49.15 (41.63–56.67)† |
| γ-Glutamyltransferase (U/L) | 18.89 (14.14–23.65) | 21.68 (17.08–26.29)† | 20.78 (15.92–25.64)‡ |
| Serum amyloid A (mg/L)* | 0.22 (0.12–0.33) | 5.42 (1.55–15.20)† | 21.72 (5.80–74.94)† |
| Fibrinogen (g/L) | 3.54 (3.33–3.76) | 3.63 (3.45–3.82) | 4.21 (3.97–4.46)† |
| Haptoglobin (mg/L) | 1,646 (1,449–1,844) | 1,000 (796–1,203)† | 1,198 (978–1,417)† |
| Iron μmol/L) | 29.66 (27.25–32.08) | 12.30 (10.29–14.30)† | 14.36 (11.43–17.30)† |
| Creatine kinase (U/L)* | 242 (219–268) | 3,535 (2,482–5,033)† | 850 (629–1,149)† |
| Aspartate transaminase (U/L)* | 347 (319–376) | 595 (515–687)† | 610 (529–704)† |
| Creatinine μmol/L)* | 84.87 (78.65–91.58) | 112.41 (100.95–125.16)† | 90.68 (83.69–98.26) |
| Urea (mmol/L)* | 5.79 (5.30–6.32) | 9.97 (9.06–10.96)† | 9.21 (8.09–10.49)† |
| Sodium (mmol/L) | 140.0 (138.8–141.2) | 140.8 (139.8–141.9) | 140.0 (139.3–140.8) |
| Potassium (mmol/L) | 4.54 (4.08–5.01) | 2.96 (2.75–3.17)† | 3.44 (3.13–3.75)† |
| Chloride (mmol/L)* | 102.5 (101.5–103.6) | 98.1 (95.3–101.0)† | 99.0 (96.4–101.8)§ |
| Calcium (mmol/L) | 3.08 (3.04–3.13) | 3.01 (2.93–3.11)‖ | 2.87 (2.83–2.91)† |
| Magnesium (mmol/L)* | 0.88 (0.86–0.91)‖ | 0.84 (0.78–0.91) | 0.85 (0.80–0.90) |
Values were logarithmically transformed for comparisons among assessment points.
Value differs significantly (P < 0.001) from corresponding preride value.
Value differs significantly (P < 0.01) from corresponding preride value.
Value differs significantly (P < 0.05) from corresponding preride value.
One outlier was excluded from each of these calculations.
Significant changes from preride values were detected in all standard CBC and serum biochemical variables (except serum sodium and magnesium concentration) after the ride or the next morning (Table 2). All blood variables were reanalyzed (recovery vs preride values) after correction for hydration status. However, this correction had only a minor influence on the results, with thrombocyte and erythrocyte counts as well as hemoglobin concentration, Hct, and serum creatinine and sodium concentrations significantly increased the morning after the ride. Two obvious outliers in serum calcium and magnesium values were excluded from analysis because measurement errors were suspected.
Discussion
To the authors' knowledge, the present study was the first in which changes in heart rate and HRV were evaluated in horses during the recovery period after an endurance ride. The study revealed that horses had a higher heart rate and lower HRV for many hours after completing a 120- or 160-km ride, compared with preride values. These changes were further associated with a change in the pattern of heart rate fluctuations (Figure 2).
In the evening before the ride, some horses had behavioral signs of stress attributable to the unfamiliar surroundings, and this was most likely the cause of the large momentary increases in instantaneous heart rate detected in preride ECG recordings. In contrast, all horses were observed to be quiet, resting, or eating during the recovery period. The less prominent but longer-lasting heart rate increases during recovery were therefore most likely caused by internal rather than external stimuli. During the recovery period, the horses were primarily resting in their stalls, although owners were allowed to groom the horses and take them out of the box stalls. These activities may have biased comparisons between the resting period before the race and the recovery period.
Analyses of blood samples collected before and after the ride revealed metabolic changes, including exercise-induced leukocytosis, increases in acute-phase reactant concentrations, hemoconcentration, volume depletion, increases in muscle enzyme activities, and changes in electrolyte concentrations comparable to previously reported results for endurance horses after competition.1–9 The changes could have explained the increase in heart rate during the recovery period in the present study, through mechanisms such as increases in oxygen demand for cellular recuperation and sympathetic activity mediated by the baroreceptor reflex. Postride increases in serum concentrations and activities of the cardiac biomarkers cTnI and CK-MB were also identified, consistent with findings in another study.17 However, it was unclear whether myocardial damage was as a cause of the increase in heart rate during the recovery period in the present study.
In human athletes, a postexercise decrease in left ventricular function (known as exercise-induced cardiac fatigue) has been suggested to increase sympathetic activity to maintain cardiac output and compensate for reduced cardiac performance after prolonged exercise.22 Exercise-induced cardiac fatigue subsequent to prolonged exercise has been suggested to occur in horses.23 However, this suggestion requires further investigation, and therefore, it is unknown whether cardiac fatigue contributed to the increased heart rate during the recovery period in the horses of the present study.
The study reported here was also the first in which cardiac rhythm was evaluated in horses for multiple hours after prolonged exercise. The frequency of VPCs (normalized per hour) increased from the preride frequency, albeit not significantly, during the recovery period, possibly as a consequence of an increase in sympathetic activity, electrolyte changes, or myocardial damage. Although frequencies of premature beats were generally low, 1 horse had considerable arrhythmic activity, particularly during the initial hours of recovery (Figures 4 and 5). Previous studies have revealed a high prevalence of premature beats within the initial 4 to 60 minutes after exercise in trotting13,14 and warmblood18,24 horses. The ECG recordings could not be started immediately after the ride in the present study because horses were required to go through their final veterinary inspection before the ECG equipment could be applied. It would have been interesting to evaluate the frequency of premature beats within the earliest part of the recovery period given that vagal reactivation and potential sympathetic-parasympathetic imbalances could be presumed to be more pronounced at that time. We therefore suspect that the observed arrhythmia frequencies would have been even greater if this initial period had been included in the recovery ECGs. This suspicion, along with the observed increases in the prevalence and frequency of ventricular arrhythmias, supports the need for additional research into the effects of prolonged exercise on cardiac function in horses. Furthermore, considering that the horse with the highest number of premature beats during the recovery period completed a 160-km ride and passed the final veterinary inspection, we speculate that some endurance horses might be expected to have increased arrhythmic activity after an endurance ride without being noticed.
The similarity of the amino acid sequences of human, canine, feline, and equine cTnI enables the use of commercial human assays for veterinary species.25 The target sequences of the high-sensitivity assayi used in the study reported here are identical in human, canine, and feline cTnI and differ from equine cTnI by only a single amino acid.21,26 This assay has been previously used for measurement of serum cTnI concentration in horses,27 revealing mean concentrations of 0.003 ± 0.003 ng/mL (range, 0.0 to 0.011 ng/mL) in those with no evidence of cardiac disease and concentrations of 0.025 to 7.44 ng/mL in diseased horses with conditions ranging from mild leukocytosis to severe colic.
Serum CK-MB activity is a less specific cardiac biomarker in horses but was included in the present study to support the cTnI concentration results. Interpretation of CK-MB activity is complicated by crossreactivity with creatine kinase after skeletal muscle damage.28 Serum CK-MB activity in the present study had no correlation with serum creatine kinase activity; however, considering that CK-MB values remained significantly increased the morning after the endurance ride (in contrast to creatine kinase values), the observed increase was presumed to be of not only skeletal muscle but also myocardial origin.
Studies15,16 of human endurance athletes have revealed increases in both cTnI concentration and CK-MB activity after competition. The release of troponins with pathological myocardial damage in humans results from ischemic cardiac injury or direct myocardial trauma, myocarditis, or drug-induced toxic effects. With sudden myocardial injury, troponins are released from the cytosol and later from the structurally bound troponins linked to actin and tropomyosin in cardiomyocytes.16 Troponin release in humans with myocardial injury is therefore typically biphasic, resulting in a small initial release of troponins followed by a larger sustained release, with a peak in serum concentration up to 24 hours after injury.16 For athletes, it is uncertain whether the increase in circulating cTnI concentration is caused by actual myocardial damage. The relatively small increases in circulating troponin concentrations and rapid clearance following exercise15 have led to the hypothesis that the troponin release might be considered a physiologic response to exercise when no other clinical signs of cardiac dysfunction or pathological changes exist.16
The authors are unaware of any studies involving measurement of serum CK-MB activity in endurance horses, although plasma cTnI concentration was shown to increase in relation to endurance riding in a previous study.17 Investigators in that study found significant increases in plasma cTnI concentration after both 80- and 160-km rides, with no significant differences between horses that completed the ride and those that were eliminated and no relationship between speed and cTnI concentration in horses that completed the ride. A different assay was used in that study,17 which might account for the higher cTnI concentrations than those identified in the present study, both before and after the rides.
Similar to findings in the other study,17 increases identified in serum cTnI concentration in the present study were mild, and although significant when compared with preride values, it is uncertain whether the increase was clinically consequential. Circulating cTnI concentration can be high in various systemic conditions.27 Consequently, we could not conclude whether the increase observed in relation to endurance riding in our study was caused by myocardial damage or was merely a physiologic response to prolonged exercise. Nine horses had a serum cTnI concentration greater than the reported upper reference limit,27 with a maximum value of 0.084 ng/mL. However, these horses did not appear to have greater hematologic or biochemical changes, compared with the remaining horses. Therefore, their systemic condition did not appear to underlie the increase in serum cTnI concentration. The increases in cardiac biomarker values and increase in VPC frequency during recovery, albeit nonsignificant, coupled with the significant changes in heart rate and HRV identified in the present study could therefore have indicated that prolonged exercise causes some degree of adverse effects to equine heart tissue. However, considering the small sample size in the present study, this possibility remains speculative.
The primary limitation of the present study was the small number of horses used. Because only 9 of the original 28 enrolled horses were eliminated from the rides, the decision was made to exclude their data from the analyses. Had more data been available, we could have compared variables between horses that completed the ride and those that were eliminated. It would also have been interesting to have performed a follow-up examination several days after the ride. Echocardiographic evaluation of myocardial function at various points would also have been beneficial.
The present study revealed that horses competing in an endurance ride had an increase in heart rate and decrease in HRV attributable to increased sympathetic activity during the recovery period following the ride. The frequency of ventricular arrhythmias also increased during recovery, albeit nonsignificantly, although no significant changes from preride values in the frequency of supraventricular premature beats was detected. Cardiac biomarker values were greater than preride values following the ride, but the clinical relevance of that finding remains uncertain. In combination, these changes are hypothesized to have been related to an exerciseinduced cardiovascular response. Additional studies are needed to elucidate the nature of the relationship between prolonged exercise and these cardiac effects and their impact on horse welfare in endurance riding.
Acknowledgments
Funded by the University of Copenhagen.
None of the authors have a financial or personal relationship with other individuals or organizations that could inappropriately influence or bias the content of this report.
The authors thank Dr. Rebecca Langhorn for assistance with analysis of cardiac biomarkers, Dr. Søren Saxmose Nielsen for assistance with statistical analysis, and Tina Roust and Camilla Malec for handling of blood samples.
ABBREVIATIONS
| CK-MB | Creatine kinase isoenzyme MB |
| cTnl | Cardiac troponin I |
| HRmax | Maximum mean heart rate calculated over a |
| HRmean | 5-minute sequence within a 30-minute period Mean heart rate calculated over a 30-minute period |
| HRmin | Minimum mean heart rate calculated over a 5-minute sequence within a 30-minute period |
| HRV | Heart rate variability |
| SDNN | SD of normal R-R intervals calculated over a 30-minute period |
| SVPC | Supraventricular premature complex |
| VPC | Ventricular premature complex |
Footnotes
Televet 100, Engel Engineering Services GmbH, Heusenstamm, Germany.
Animal Polster, Snøgg AS, Kristiansand, Norway.
PROC MEANS, SAS Enterprise Guide, version 6.1, SAS Institute Inc, Cary, NC.
BD Vacutainer, Becton, Dickinson & Co, Franklin Lakes, NJ.
CryoPure, Sarstedt AG & Co, Nümbrecht, Germany.
ADVIA 2120 hematology system, Siemens Healthcare Diagnostics Inc, Tarrytown, NY.
ADVIA 1800 chemistry system, Siemens Healthcare Diagnostics Inc, Tarrytown, NY.
ACL TOP coagulation analyzer, Instrumentation Laboratory, Bedford, Mass.
ADVIA Centaur CP TnI-ultra, Siemens Healthcare Diagnostics Inc, Tarrytown, NY.
ADVIA Centaur CP CK-MB, Siemens Healthcare Diagnostics Inc, Tarrytown, NY.
PHASE haptoglobin assay, Tridelta Development Ltd, Maynooth, Kildare, Ireland.
PROC MIXED, SAS Enterprise Guide, version 6.1, SAS Institute Inc, Cary, NC.
References
1. Rose RJ, Purdue RA, Hensley W. Plasma biochemistry alterations in horses during an endurance ride. Equine Vet J 1977; 9:122–126.
2. Rose RJ, Hodgson DR, Sampson D, et al. Changes in plasma biochemistry in horses competing in a 160 km endurance ride. Aust Vet J 1983; 60:101–105.
3. Rose RJ, Arnold KS, Church S, et al. Plasma and sweat electrolyte concentrations in the horse during long distance exercise. Equine Vet J 1980; 12:19–22.
4. Rose RJ. Haematological changes associated with endurance exercise. Vet Rec 1982; 110:175–177.
5. Sloet van Oldruitenborgh-Oosterbaan MM, Wensing T, Barneveld A, et al. Heart rate, blood biochemistry and performance of horses competing in a 100 km endurance ride. Vet Rec 1991; 128:175–179.
6. Lindinger MI, Ecker GL. Ion and water losses from body fluids during a 163 km endurance ride. Equine Vet J 1995; 27:314–322.
7. Schott HC II, Marlin DJ, Geor RJ, et al. Changes in selected physiological and laboratory measurements in elite horses competing in a 160 km endurance ride. Equine Vet J Suppl 2006; (36):37–42.
8. Cywińska A, Szarska E, Górecka R, et al. Acute phase protein concentrations after limited distance and long distance endurance rides in horses. Res Vet Sci 2012; 93:1402–1406.
9. Boucher JH, Ferguson EW, Wilhelmsen CL, et al. Erythrocyte alterations endurance exercise in horses. J Appl Physiol 1981; 51:131–134.
10. Fielding CL, Meier CA, Balch OK, et al. Risk factors for the elimination of endurance horses from competition. J Am Vet Med Assoc 2011; 239:493–498.
11. Robert C, Benamou-Smith A, Leclerc J-L. Use of the recovery check in long-distance endurance rides. Equine Vet J Suppl 2002;(34):106–111.
12. Aagaard P, Sahlén A, Bergfeldt L, et al. Heart rate and its variability in response to running-associations with troponin. Med Sci Sports Exerc 2014; 46:1624–1630.
13. Buhl R, Petersen EE, Lindholm M, et al. Cardiac arrhythmias in Standardbreds during and after racing—possible association between heart size, valvular regurgitations, and arrhythmias. J Equine Vet Sci 2013; 33:590–596.
14. Physick-Sheard PW, McGurrin MKJ. Ventricular arrhythmias during race recovery in Standardbred racehorses and associations with autonomic activity. J Vet Intern Med 2010; 24:1158–1166.
15. Neumayr G, Gaenzer H, Pfister R, et al. Plasma levels of cardiac troponin I after prolonged strenuous endurance exercise. Am J Cardiol 2001; 87:369–371.
16. Shave R, Baggish A, George K, et al. Exercise-induced cardiac troponin elevation: evidence, mechanisms, and implications. J Am Coll Cardiol 2010; 56:169–176.
17. Holbrook TC, Birks EK, Sleeper MM, et al. Endurance exercise is associated with increased plasma cardiac troponin I in horses. Equine Vet J Suppl 2006;(36):27–31.
18. Buhl R, Meldgaard C, Barbesgaard L. Cardiac arrhythmias in clinically healthy showjumping horses. Equine Vet J Suppl 2010;(38):196–201.
19. Verheyen T, Decloedt A, De Clercq D, et al. Oesophageal electrocardiography in healthy horses. Equine Vet J 2012; 44:640–645.
20. Verheyen T, Decloedt A, De Clercq D, et al. Electrocardiography in horses—part 2: how to read the equine ECG. Vlaams Diergeneeskd Tijdschr 2010; 79:337–344.
21. Langhorn R, Willesen JL, Tarnow I, et al. Evaluation of a highsensitivity assay for measurement of canine and feline serum cardiac troponin I. Vet Clin Pathol 2013; 42:490–498.
22. Hautala A, Tulppo MP, Makikallio TH, et al. Changes in cardiac autonomic regulation after prolonged maximal exercise. Clin Physiol 2001; 21:238–245.
23. Amory H, Votion DM, Fraipont A, et al. Altered systolic left ventricular function in horses completing a long distance endurance race. Equine Vet J Suppl 2010;(38):216–219.
24. Barbesgaard L, Buhl R, Meldgaard C. Prevalence of exerciseassociated arrhythmias in normal performing dressage horses. Equine Vet J Suppl 2010;(38):202–207.
25. Wells SM, Sleeper M. Cardiac troponins. J Vet Emerg Crit Care 2008; 18:235–245.
26. Rishniw M, Simpson KW. Cloning and sequencing of equine cardiac troponin I and confirmation of its usefulness as a target analyte for commercial troponin I analyzers. J Vet Diagn Invest 2005; 17:582–584.
27. Serra M, Papakonstantinou S, Adamcova M, et al. Veterinary and toxicological applications for the detection of cardiac injury using cardiac troponin. Vet J 2010; 185:50–57.
28. Argiroudis SA, Kent JE, Blackmore DJ. Observations on the isoenzymes of creatine kinase in equine serum and tissues. Equine Vet J 1982; 14:317–321.
