Horses are frequently transported across continents for participation in racing, other types of competition, or breeding programs. Transportation of horses by air is becoming more common, and horses that are transported in this manner are typically extremely valuable. Stress associated with transportation may play a major role in the development of transportation-related respiratory tract disease (shipping fever).1–12 As a result, there is a need (from a clinical standpoint) for better understanding of the mechanisms that lead to transport-associated stress in horses. The effects of transportation on the physiologic responses of horses to stressors have been studied; transportation causes a variety of physiologic responses, such as decreased body mass, elevated HR and respiratory rate, increased circulating concentrations of cortisol and ACTH, and changes in various other indices.1–14
It has been suggested that power spectral analysis of HRV is a noninvasive index of autonomic nervous activity.15–17 The HRV in the HF power spectrum is primarily attributable to parasympathetic nervous activity, and sympathetic and parasympathetic nervous activity have been shown to contribute to the LF power spectrum.15–18 Therefore, it has been suggested that the LF:HF ratio is an index of cardiac sympathovagal balance.15–18
The purpose of the study reported here was to evaluate (by use of power spectral analysis) HRV indices as quantitative measures of the autonomic nervous activity and response to stressors in horses during transportation. We hypothesized that HRV indices are more sensitive indicators of the effects of transportation on autonomic nervous activity than is the more commonly used index (HR) and that transportation by air and transportation by road similarly affect responses of the autonomic nervous system in horses.
Materials and Methods
Horses—Six healthy warmbloods that had been purchased in Belgium from private owners by the Japan Racing Association and were being shipped to Japan were enrolled in the study. The horses included 1 Dutch Warmblood and 5 Belgian Warmbloods. There were 5 geldings and 1 female, and the mean ± SD age was 9.2 ± 2.1 years. Protocols for the study were reviewed and approved by the Animal Welfare and Ethics Committee of the Japan Racing Association Equine Research Institute.
Transportation—Horses were transported from Brugge, Belgium, to Narita International Airport, Japan (near Tokyo), by a combination of transportation by road and transportation by air. Seven to 9 days prior to the journey, the horses were transported a short distance by use of a horse van and housed in a quarantine barn (quarantine conditions) near Brugge until they were shipped to Japan. Horses were housed at the quarantine facility in individual stalls that had shaved wood bedding, which was similar to their home environments. They were fed a concentrate feed twice each day and had ad libitum access to hay and water.
During transportation to Narita International Airport, the horses were transported by road (road-transport conditions; approx 4.5 hours) from the quarantine barn near Brugge to Schiphol Airport in Amsterdam by use of a horse van, which was fully loaded with 8 horses; the other 2 horses in the van were owned by the Japan Racing Association but were not part of the study. Horses were transferred from the van to commercially available 3-horse air-transport stalls and were housed indoors at the airport before being loaded into an airplanea (waiting conditions; 5.5 hours). Each air-transport stall was loaded with 3 horses. Fifteen horses (housed in 5 air-transport stalls) were transported in the airplane (air-transport conditions). A total of 10 of the horses being transported were owned by the Japan Racing Association; the other 5 horses (not part of the study) were privately owned. During the flight, indices of cabin air quality (dust, CO2, and NH3 concentrations; humidity; and temperature) were monitored. The horses remained in the aircraft until they arrived at Narita International Airport, Japan. The flight from Schiphol Airport to Narita International Airport was 11 hours in duration, and the total duration of transport from the quarantine barn to Narita International Airport (road-transport, waiting, and air-transport conditions) was 21 hours. During the flight, horses were offered silage hay and water every 2 hours. Horses were moved to a quarantine stable near the airport within 15 minutes after the airplane landed at Narita International Airport; this period was not included as an experimental condition, and data were not obtained after air-transport conditions other than rectal temperature, which was monitored every 12 hours.
ECG recordings—Electrocardiograms were recorded continuously during quarantine conditions for a period of 24 hours from each horse 2 to 4 days after they had arrived at the quarantine barn near Brugge (ie, 5 days before transportation from Brugge to Schiphol Airport). The ECGs were recorded continuously during road-transport, waiting, and air-transport conditions. The ECG recordings were obtained by use of a modified base-apex lead19 and a Holter-type monitor.b Horses wore an elastic surcingle to which the Holter monitor was attached. Horses wore the same instrumentation during recording of ECGs during quarantine, road-transport, waiting, and air-transport conditions. Horses were allowed to move freely in their stalls without restriction during recording of quarantine-period ECGs and during quarantine after the flight at Narita International Airport. During transportation by road and air, the horses' heads were loosely cross tied.
Data analysis—The ECGs were analyzed by use of an ECG processor analyzing systemc as previously described.18 The software identified R waves and calculated the R-R interval tachogram (raw HRV data) in sequential order. Noise that the software identified as R waves was eliminated manually via visual inspection and examination of values that were outside the interval of 75% to 125% of the mean value. A spline curve was fit to the tachogram from which data sets of 512 points were resampled at 200-millisecond intervals. The values were selected as a compromise to balance the need for a large time series for accuracy versus the need for reasonably short recording periods.16 Each data set was applied to a Hamming window, and a fast-Fourier transformation was performed to obtain the power spectrum density of the fluctuation. Frequency of the LF power was set at 0.01 to 0.07 Hz, and frequency of the HF power was set at 0.07 to 0.6 Hz.18 The HR, HRV indices (LF power and HF power), and LF:HF ratio were determined from each recording. The LF power and HF power were each calculated as the area under the curve within the respective frequency range for each hour of the experiment. The HRV indices were calculated for each horse (by use of mean values for each hour of the experiment) and for each experimental condition (quarantine, road-transport, waiting, and air-transport conditions; mean values for each hour and mean values for each experimental condition were calculated by use of the mean values for each horse during those periods).
Statistical analysis—Values were expressed as mean ± SD or as mean ± SE (for some data to improve clarity); within-horse and between-horse variability were expressed for some data. Comparisons among data that were obtained during quarantine, road-transport, waiting, and air-transport conditions were performed by use of a 1-way repeated-measures ANOVA and a Holm-Sidak post hoc pairwise multiple comparisons procedure. For nonnormally distributed data that could not be transformed (LF power), analysis was performed by use of a Friedman repeated-measures ANOVA on ranks. Pairwise comparisons were made by use of the Student-Newman-Keuls method. Nonlinear regression analysis was performed to evaluate the fit of the quarantine conditions data to a diurnal rhythm. To remove the potential influence of diurnal rhythm on comparisons of HRV indices, a 2-way repeated-measures ANOVA was performed to compare values obtained during the 3 transportation-related experimental conditions (road-transport, waiting, and air-transport conditions) with values obtained during the corresponding times of day during quarantine conditions. For these comparisons, experimental condition and time were used as factors. Least squares linear regression and forward stepwise multiple regression analyses were performed to evaluate the influence of HRV indices on HR. Values of P ≤ 0.05 were considered significant.
Results
During road-transport conditions, horses were exposed to a 4.5-hour van ride on a major highway that had a smooth road surface, and there were few delays (ie, the van ride was uneventful). During waiting conditions, horses remained in their air-transport stalls and were kept inside a building until they were loaded into the airplane. The horses remained in the airplane from the time they were loaded until after the airplane landed. There was little air turbulence during the flight. Indices of the quality of air indicated that the cabin of the airplane was well ventilated, the air quality was good (low concentrations of dust, CO2, and NH3), there was moderate humidity, and the environment was thermoneutral for horses during the flight. All horses remained calm during transportation by road and by air. Three of 6 horses had pyrexia (rectal temperature > 38.6°C) upon arrival at Narita International Airport in Japan; however, the rectal temperature of these horses was within the reference interval by the following day.
The HRV indices during quarantine conditions all had significant (P < 0.001 to 0.020) diurnal rhythms (Figure 1) as determined by their fit to the following nonlinear regression equation: Y = a (sin [m (X – p)]) + q, where Y is the value of the dependent variable, X is the number of hours into the protocol, and a, m, p, and q (amplitude, frequency, temporal offset, and magnitude offset, respectively) are fitted variables that determine the parameters of the response curve. The values of the variables a, m, p, and q and the coefficient of determination were determined for the HRV indices as follows: HR (a, 1.30; m, −25.4; p, −7.45; q, 38.9; R2 = 0.454), LF power (a, 471, m, −25.6, p, −8.00, q, 3,437; R2 = 0.349), HF power (a, 167, m, −25.0, p, 71.1, q, 970; R2 = 0.280), and the LF:HF ratio (a, 0.877, m, −25.6, p, 10.2, q, 4.83; R2 = 0.606).
Hourly mean ± SE HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in 6 warmbloods during 24 hours of quarantine (stall rest in quarantine stalls 5 days before transport [white circles]) conditions, 4.5 hours of road-transport (transportation by road in a horse van [black triangles]) conditions, 5.5 hours of waiting (waiting in air-transport stalls prior to being loaded into an airplane [black squares]) conditions, and 11 hours of air-transport (transportation by air while housed in air-transport stalls [black circles]) conditions. Time of day is represented in military format. Sinusoidal waveforms are nonlinear regressions fit to diurnal variation of the quarantine condition variables.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Hourly mean ± SE HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in 6 warmbloods during 24 hours of quarantine (stall rest in quarantine stalls 5 days before transport [white circles]) conditions, 4.5 hours of road-transport (transportation by road in a horse van [black triangles]) conditions, 5.5 hours of waiting (waiting in air-transport stalls prior to being loaded into an airplane [black squares]) conditions, and 11 hours of air-transport (transportation by air while housed in air-transport stalls [black circles]) conditions. Time of day is represented in military format. Sinusoidal waveforms are nonlinear regressions fit to diurnal variation of the quarantine condition variables.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Hourly mean ± SE HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in 6 warmbloods during 24 hours of quarantine (stall rest in quarantine stalls 5 days before transport [white circles]) conditions, 4.5 hours of road-transport (transportation by road in a horse van [black triangles]) conditions, 5.5 hours of waiting (waiting in air-transport stalls prior to being loaded into an airplane [black squares]) conditions, and 11 hours of air-transport (transportation by air while housed in air-transport stalls [black circles]) conditions. Time of day is represented in military format. Sinusoidal waveforms are nonlinear regressions fit to diurnal variation of the quarantine condition variables.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
The mean ± SD HRs during quarantine, road-transport, waiting, and air-transport conditions were 38.9 ± 1.5 beats/min, 41.7 ± 5.6 beats/min, 41.5 ± 4.3 beats/min, and 48.8 ± 5.6 beats/min, respectively (Figures 1 and 2). Heart rate was significantly (P = 0.013) higher during air-transport conditions than it was during quarantine conditions. The LF power was significantly higher during quarantine (3,454 ± 1,087 milliseconds2) and air-transport (3,101 ± 567 milliseconds2) conditions versus road-transport (1,824 ± 432 milliseconds2) and waiting (2,072 ± 616 milliseconds2) conditions. The HF powers (range, 509 to 927 milliseconds2) and the LF:HF ratios (range, 4.1 to 6.2) during each experimental condition were not significantly different.
Mean ± SE HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in the 6 horses in Figure 1 during quarantine, road-transport, waiting, and air-transport conditions. Bars with asterisks represent significant (P ≤ 0.05) differences between data. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Mean ± SE HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in the 6 horses in Figure 1 during quarantine, road-transport, waiting, and air-transport conditions. Bars with asterisks represent significant (P ≤ 0.05) differences between data. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Mean ± SE HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in the 6 horses in Figure 1 during quarantine, road-transport, waiting, and air-transport conditions. Bars with asterisks represent significant (P ≤ 0.05) differences between data. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
The HR was highly correlated with LF power during road-transport conditions (R2 = 0.979; P < 0.001; Figure 3), and they were related by the following equation: LF power = (–4,639 + [155 × HR {beats/min}]). The HR and HF powers during road-transport conditions were significantly (P = 0.01) correlated (R2 = 0.920) and were related by the following equation: HF power = (–119 + [15.0 × HR {beats/min}]). However, results of forward stepwise regression indicated that there was no improvement in fit if HF power were added as a variable (in addition to LF power). The HR during air-transport conditions was significantly (P = 0.019) correlated (R2 = 0.477) with the LF:HF ratio. No other HRV indices were significantly related to HR, although the other regressions all had low power (range, 0.05 to 0.26).
Relationship between the HR and the HF power (A), LF power (B), and LF:HF ratio (C) in the 6 horses in Figure 1 during quarantine, road-transport, waiting, and air-transport conditions. Nonsignificant correlations and linear regressions (dashed lines) and significant linear regressions (solid lines) are shown. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Relationship between the HR and the HF power (A), LF power (B), and LF:HF ratio (C) in the 6 horses in Figure 1 during quarantine, road-transport, waiting, and air-transport conditions. Nonsignificant correlations and linear regressions (dashed lines) and significant linear regressions (solid lines) are shown. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Relationship between the HR and the HF power (A), LF power (B), and LF:HF ratio (C) in the 6 horses in Figure 1 during quarantine, road-transport, waiting, and air-transport conditions. Nonsignificant correlations and linear regressions (dashed lines) and significant linear regressions (solid lines) are shown. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Comparisons among data obtained during the 3 transportation-related periods (road-transport, waiting, and air-transport conditions) and the diurnally equivalent time periods during quarantine conditions revealed significant differences. For HR and HRV indices, significant differences were detected between the following experimental conditions: for HR, air-transport conditions versus quarantine conditions (P = 0.010) and air-transport conditions versus time (P = 0.035); for LF power, road transport conditions versus quarantine conditions (P = 0.010), road-transport conditions versus time (P = 0.019), waiting conditions versus quarantine conditions (P = 0.027), waiting conditions versus time (P = 0.018), and air-transport conditions versus time (P = 0.041); and for the LF:HF ratio, air-transport conditions versus time (P = 0.033).
Comparison of the mean between-horse variability in HR and HRV indices among all experimental conditions in the present study was greater than the within-horse variability in HR, LF power, HF power, and the LF:HF ratio by 41.2%, 42.5%, 108%, and 134%, respectively. These values were calculated as the ratio of the between-horse to within-horse variances for each experimental treatment. Data from the present study and 2 other studies14,20 that compared HR and HRV indices in horses during various types of stress (24 hours of stall rest vs 21 hours of transportation by road and ferry in five 2-year-old Thoroughbreds14 and 24 hours of stall rest vs 24 hours of withholding of food in 5 Thoroughbreds [mean ± SE age, 7.8 ± 1.5 years]20) were summarized (Figure 4).
Mean values for HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in the present study and in 2 other studies14,20 that compared HR and HRV indices in horses during various types of stress (24 hours of stall rest [SR] vs 21 hours of transportation by road and ferry [R] in five 2-year-old Thoroughbreds14; 24 hours of stall rest [C] vs 24 hours of withholding of food [F] in 5 Thoroughbreds [mean ± SE age, 7. 8 ± 1. 5 years]20; and quarantine [Q], road-transport [RT], waiting [W], and air-transport [AT] conditions in 6 warmbloods [mean ± SD age, 9.2 ± 2.1 years] in the present study). Positive error bars represent 1 SD of between-horse variability, and negative error bars represent 1 SD of within-horse variability. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Mean values for HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in the present study and in 2 other studies14,20 that compared HR and HRV indices in horses during various types of stress (24 hours of stall rest [SR] vs 21 hours of transportation by road and ferry [R] in five 2-year-old Thoroughbreds14; 24 hours of stall rest [C] vs 24 hours of withholding of food [F] in 5 Thoroughbreds [mean ± SE age, 7. 8 ± 1. 5 years]20; and quarantine [Q], road-transport [RT], waiting [W], and air-transport [AT] conditions in 6 warmbloods [mean ± SD age, 9.2 ± 2.1 years] in the present study). Positive error bars represent 1 SD of between-horse variability, and negative error bars represent 1 SD of within-horse variability. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Mean values for HR (A) and HRV indices (HF power [B], LF power [C], and LF:HF ratio [D]) in the present study and in 2 other studies14,20 that compared HR and HRV indices in horses during various types of stress (24 hours of stall rest [SR] vs 21 hours of transportation by road and ferry [R] in five 2-year-old Thoroughbreds14; 24 hours of stall rest [C] vs 24 hours of withholding of food [F] in 5 Thoroughbreds [mean ± SE age, 7. 8 ± 1. 5 years]20; and quarantine [Q], road-transport [RT], waiting [W], and air-transport [AT] conditions in 6 warmbloods [mean ± SD age, 9.2 ± 2.1 years] in the present study). Positive error bars represent 1 SD of between-horse variability, and negative error bars represent 1 SD of within-horse variability. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 73, 4; 10.2460/ajvr.73.4.515
Discussion
The present study was conducted to determine whether the response of horses to stressors during air-transport conditions caused changes in HRV indices and whether the changes were similar to those that developed during road-transport conditions. The experimental protocol included confinement of horses to stalls and periods of transportation by road and by air; this protocol was chosen so that results of the present study could be compared with results of other studies14,20 in which HR and HRV indices were determined in horses that were exposed to stressors. Stall rest conditions in the other studies14,20 were similar to quarantine conditions in the present study, although horses in the present study were not housed in their usual stalls and had only been in the quarantine stalls for 2 to 4 days before ECGs were recorded. As a result, these horses might have had a moderate degree of stress during quarantine conditions because of the unfamiliar environment, and the values of the HRV indices (obtained during quarantine conditions) might reflect a mild stress response (compared with values obtained in horses during resting conditions in their usual stalls). This possibility should be kept in mind when comparing values obtained during quarantine conditions in the present study (rest in unfamiliar stalls and environment) with values obtained during control conditions (rest in familiar stalls and environment) in other studies.
Analysis of the HR and HRV indices that were determined during quarantine conditions in the present study revealed evidence of diurnal rhythm, as has been identified in another study,14 although the pattern of diurnal rhythm in the present study was not as pronounced as it was in the other study.14 The coefficients of determination for the nonlinear regressions of HR, LF power, and HF power were all lower in the present study than those in the other study,14 although the fit for the LF:HF ratio was better (and was significant) in the present study. It is unclear whether the more apparent diurnal rhythm that was detected in the other study14 was attributable to differences in breeds (warm-bloods in the present study vs Thoroughbreds in the other study14) or to the possibility that horses in the present study during quarantine conditions (ie, in an unfamiliar environment) were not as relaxed as were horses in other studies14,20 during control conditions (ie, in a familiar environment). The high correlation of HRV indices with diurnal rhythm in the present study did not seem to be exclusively a function of age because the horses in another study20 that had diurnal variation in these indices were older than horses in the present study.
The comparisons among HRV indices during quarantine conditions and during transportation-related conditions (road-transport, waiting, and air-transport conditions) that were performed by use of 2-way repeated-measures ANOVA revealed more differences among data obtained during each experimental condition than did comparisons that were performed by use of 1-way ANOVA (in which mean values for each experimental condition were compared without regard for diurnal rhythm). This finding suggests that diurnal rhythm during quarantine conditions might mask differences between data that were obtained during quarantine versus transportation-related conditions. Because data for each of the transportation-related conditions were obtained during short periods that did not overlap, comparisons of data for diurnal rhythm could be made only with data obtained during quarantine conditions in the present study.
During road-transport conditions in the present study, HR and LF power were extremely high during the first hour, then decreased markedly, which was similar to the pattern that was detected in another study14; this finding suggests that horses had the most excitement during the initial period of transportation by road and that they acclimated during the first hour. This pattern was detected in young horses (that had modest experience of short durations of road-transport conditions) in the other study14 and in old horses (that had moderate experience of regional transportation by road) in the present study. The HR peaked during the third hour of waiting conditions in the present study. We were unable to observe the horses directly during that period; however, because they were loaded into the airplane during the waiting period, it is likely that the spike in HR that was detected during that period was associated with loading of the air-transport stalls into the airplane. During air-transport conditions, HR (48.8 beats/min) was higher than during any other study interval and was significantly (P = 0.010) higher (by 9.9 beats/min) during this period than during quarantine conditions (38.9 beats/min).
Many commercial aircraft are pressurized so that cabin pressure is greater than the barometric pressure outside the cabin during flight; however, that pressure is less than the atmospheric barometric pressure at sea level.21–24 It has been reported that when the mean ± SD cabin pressure in an airplane is decreased to 585 ± 15 mm Hg, oxyhemoglobin saturation significantly decreases to 92 ± 2% in humans; however, HR during the flight does not differ from that before the flight.21 Given that the ideal alveolar gas equation indicates that alveolar oxygen tension is > 60 mm Hg at this cabin pressure, it seems unlikely that the increased HR detected in horses during air-transport conditions in the present study was attributable to hypoxic stimulation; rather, the increased HR was more likely associated with excitement or stress. Other authors25 that investigated transportation of horses by air detected a deterioration of cabin air quality during periods when the airplane used in that report was stationary; however, cabin ventilation and air quality were good in the airplane that was used in the present study. Differences in design or function of the ventilation systems of the airplane used in the other study25 versus the airplane used in the present study presumably were responsible for this difference in air quality, particularly because human passengers shared the cabin environment with the cargo space in the airplane in the other study.25
The mean values of the HF power during transportation-related (road-transport, waiting, and air-transport) conditions were 35% to 45% lower than the respective value during quarantine conditions in the present study, and all hourly mean values of the HF power were lower during transportation-related conditions than they were during quarantine conditions. However, the high variability of values obtained during quarantine conditions prevented any of the differences from being significant. This pattern of reduced HF power during transportation-related conditions versus quarantine conditions was similar to the pattern that was detected in another study14 in which researchers investigated the effects of transportation in horses.
The LF power was significantly lower during road-transport and waiting conditions (47% and 40%, respectively) than during quarantine conditions in the present study. The LF power was also significantly lower during road-transport and waiting conditions than during air-transport conditions. In another study,14 investigators detected a lower LF power during transportation than during stall rest. There were no significant differences in the LF:HF ratio among any of the transport conditions in the present study.
The patterns of association that were detected among HRV indices and HR in the present study suggest that these variables were related. For nearly all HRs, HF power was lower during stressful (road-transport, waiting, and air-transport) conditions than it was during quarantine conditions. Similarly, nearly all values of LF power during road-transport and waiting conditions were lower than values during quarantine conditions. However, values of LF power during air-transport conditions were similar to values during quarantine conditions. During air-transport conditions, there were few HR values recorded that were within 4 beats/min of any HR value recorded during quarantine conditions. For data that were collected during stressful conditions, there were few overlaps among values of LF power that were associated with a given HR; a similar relationship between data obtained during transportation and data obtained during stall rest was found in another study.14 These findings (in the present study and in the other study14) suggest that analysis of HRV indices and HR may reveal stress responses in horses.
During both road-transport and air-transport conditions in the present study, changes in HR were associated with changes in HRV indices. The HR during road-transport conditions was highly correlated with both LF power (R2 = 0.979) and HF power (R2 = 0.920); however, the addition of HF power to a multiple regression analysis did not significantly improve the fit when LF was already included. The slope of the regression equation relating LF power and HR (slope, 155; R2 = 0.979; P < 0.001) during road-transport conditions in the present study was twice the value of the slope relating LF power and HR (slope, 78.2; R2 = 0.802; P < 0.001) reported in another study14 during similar conditions in horses. The HF power during transportation in the other study14 was also significantly correlated with HR (R2 = 0.725; P < 0.001) and had a similar slope (11.8) to that observed during road-transport conditions in horses in the present study (slope, 15.0; R2 = 0.920; P = 0.002). The LF:HF ratio was moderately correlated with HR during air-transport conditions in horses in the present study (R2 = 0.477; P = 0.019) and during transportation in the other study14 (R2 = 0.570; P < 0.001), although the slopes had opposite signs (slope, −0.249 vs 0.176, respectively; data not shown).
The present study and 2 other studies14,20 used similar methods to evaluate relationships between HR and HRV indices in horses. These studies have investigated the responses of horses to quarantine, road-transport, waiting, and air-transport conditions; 24 hours of stall rest and 21 hours of transportation conditions; and 24 hours of stall rest and 24 hours of food-withholding conditions, respectively. The finding of greater between-horse than within-horse variability in HR and HRV indices is notable. In the present study and in one of the other studies,14 the between-horse variability and the within-horse variability in HR increased 2- to 10-fold, and variability in HRV indices decreased to a lesser extent during transport conditions than during resting (quarantine and stall rest, respectively) conditions. Variability of data obtained from horses during withholding of food versus during control conditions in one of the other studies20 had the opposite pattern (lower variability in HR and higher variability in HRV indices during food-withholding vs stall rest conditions).
The patterns of change in HR and HRV indices that were found in the present study and in the other studies14,20 reveal similarities and differences among the responses of horses to different stressors. The responses of horses to road-transport conditions in the present study were similar to the responses of horses to transport (by road and ferry) detected in one of the other studies14 (tachycardia was initially detected followed by decreasing HR during transport, and there was a higher mean HR in horses during transport conditions than during stall rest14 or quarantine [present study] conditions), although the duration of road-transport conditions in the present study was only 25% as long as the combined duration of transportation by road and ferry in that other study.14 Associated with this response were decreases in HF power and LF power. An opposite pattern was detected during food-withholding conditions in one of the other studies20 (HR decreased and LF power and HF power increased during withholding of food vs control conditions). Indeed, on the basis of the values of the HRV indices, the results of that study20 suggest that the excitement associated with feeding might be more stressful than withholding of food in horses and that withholding of food may cause a hypometabolic state in which there is an increased para-sympathetic influence on HRV indices. The responses of HRV indices to air-transport conditions in the present study contrasted with the responses detected in the other studies14,20 in that there was a marked increase in LF power associated with an increased HR during transportation by air, which was consistent with intense sympathetic stimulation.
The responses of horses to the stress of food-withholding conditions in another study20 might have been expected to be different than the responses of horses to transportation in the present study, given that withholding of food involves caloric deprivation but does not include stimuli (motion, vibration, and sound) that trigger prolonged periods of excitement. The overall pattern of responses to withholding of food, including increased frequency of second-degree atrioventricular block,20 suggests a strong influence of parasympathetic autonomic activity on HR in horses. Conversely, the increased HR (associated with decreased LF power and HF power) in horses during transportation that was found in another study14 suggests that there was less parasympathetic influence and greater sympathetic influence on HR or that LF power was decreased more because of parasympathetic withdrawal (as indicated by reduced HF power) than by increased sympathetic autonomic activity.
Insufficient data were available to determine the influence of age and breed of horses on HR and HRV indices in the present study. Young Thoroughbreds that were included in another study14 might have been more excitable than the older horses that were included in the present study and in another study20; however, the highest resting (during control conditions) HRs were detected in older Thoroughbred horses in one of the other studies,20 and those horses were nearly as old as the horses in the present study. Between-horse variability in HR was approximately twice as high in the young horses during stall rest14 as it was in the old horses during quarantine conditions in the present study. However, the responses of the young horses in that other study were sufficiently uniform that the diurnal pattern in HR and HRV indices was detectable and significant in a small (n = 5) group of horses.14
In the present study, HR and HRV indices in horses were sensitive to changes in the relationships among those variables, and their values were influenced by the type of stress to which horses were exposed. The patterns of these relationships among horses appeared to be consistent within a given type of stress (eg, road-transport conditions) but were different during other conditions (eg, transportation by air). These differences may have been caused by distinct autonomic responses to particular stimuli. Further research is needed to determine whether the relationships between HR and HRV indices can be used to quantify stress responses in horses.
ABBREVIATIONS
| HF | High frequency |
| HR | Heart rate |
| HRV | Heart rate variability |
| LF | Low frequency |
747–400 Combi, Boeing, Chicago, Ill.
SM-60, Fukuda Denshi Co Ltd, Tokyo, Japan.
ECG processor analyzing system, Softron Co Ltd, Tokyo, Japan.
References
- 1.
Oikawa M, Jones JH. Studies of the causes and effects of transport-associated stress and shipping fever in athletic horses. In: Kohn CW, ed. Guidelines for horse transport by road and air. New York: American Horse Shows Association, 2000;35–62, 117–123.
- 2.
Leadon DP & Watkins K. Body weight, rectal temperature, haematology and blood biochemistry prior to transport and for three days after arrival, in racehorses transported by air to international races in Hong Kong. In: Kohn CW, ed. Guidelines for horse transport by road and air. New York: American Horse Shows Association, 2000;71–81.
- 3.
Baucus KL, Ralston SL, Nockels CF, et al. Effects of transportation on early embryonic death in mares. J Anim Sci 1990; 68:345–351.
- 4.
Baucus KL, Squires EL, Ralston SL, et al. Effect of transportation on the estrous cycle and concentrations of hormones in mares. J Anim Sci 1990; 68:419–426.
- 5.
Clark DK, Friend TH & Dellmeier G. The effect of orientation during trailer transport on heart rate, cortisol and balance in horses. Appl Anim Behav Sci 1993; 38:179–189.
- 6.
Waran NK. The behaviour of horses during and after transport by road. Equine Vet Educ 1993; 5:129–132.
- 7.
Smith BL, Jones JH, Carlson GP, et al. Effect of body direction on heart rate in trailered horses. Am J Vet Res 1994; 55:1007–1011.
- 8.
Smith BL, Jones JH, Carlson GP, et al. Body position and direction preferences in horses during road transport. Equine Vet J 1994; 26:374–377.
- 9.
Hobo S, Kuwano A & Oikawa M. Respiratory changes in horses during automobile transportation. J Equine Sci 1995; 6:135–139.
- 10.
Waran NK & Cuddeford D. Effects of loading and transport on the heart rate and behaviour of horses. Appl Anim Behav Sci 1995; 43:71–81.
- 11.
Nambo Y, Yoshihara T, Kuwano A, et al. Effect of transport stress on concentrations of LH and FSH in plasma of mares: a preliminary study. J Equine Sci 1996; 7:1–5.
- 12.
Waran NK, Singh N, Robertson V, et al. Effects of transport on behaviour and heart rates of Thoroughbred horses. Appl Anim Behav Sci 1993; 38:76–77.
- 13.
Waran NK, Robertson V, Cuddeford D, et al. Effects of transporting horses facing either forwards or backwards on their behaviour and heart rate. Vet Rec 1996; 139:7–11.
- 14.↑
Ohmura H, Hiraga A, Aida H, et al. Changes in heart rate and heart rate variability in Thoroughbreds during prolonged road transportation. Am J Vet Res 2006; 67:455–462.
- 15.
Akselrod S, Gordon D, Ubel FA, et al. Power spectrum analysis of heart rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control. Science 1981; 213:220–222.
- 16.↑
Pagani M, Lombardi F, Guzzetti S, et al. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympatho-vagal interaction in man and conscious dog. Circ Res 1986; 59:178–193.
- 17.
Pomeranz B, Macaulay RJ, Caudill MA, et al. Assessment of autonomic function in humans by heart rate spectral analysis. Am J Physiol 1985; 248:H151–H153.
- 18.↑
Kuwahara M, Hashimoto S, Ishii K, et al. Assessment of autonomic nervous function by power spectral analysis of heart rate variability in the horse. J Auton Nerv Syst 1996; 60:43–48.
- 19.↑
Ohmura H, Hiraga A, Matsui A, et al. Physiological responses of young Thoroughbreds during their first year of race training. Equine Vet J Suppl 2002;(34):140–146.
- 20.↑
Ohmura H, Boscan PL, Solano AM, et al. Changes in heart rate, heart rate variability, and atrioventricular block during withholding of food in Thoroughbreds. Am J Vet Res 2012; 73:508–514.
- 21.↑
Kelly PT, Swanney MP, Frampton C, et al. Normobaric hypoxia inhalation test vs. response to airline flight in healthy passengers. Aviat Space Environ Med 2006; 77:1143–1147.
- 22.
Humphreys S, Deyermond R, Bali I, et al. The effect of high altitude commercial air travel on oxygen saturation. Anaesthesia 2005; 60:458–460.
- 23.
Cottrell JJ. Altitude exposures during aircraft flight: flying higher. Chest 1988; 93:81–84.
- 24.
Cottrell JJ, Lebovitz BL, Fennell RG, et al. Inflight arterial saturation: continuous monitoring by pulse oximetry. Aviat Space Environ Med 1995; 66:126–130.
- 25.↑
Leadon DP, Daykin J, Backhouse W, et al. Environmental, haematological and blood biochemical changes in equine transit stress. In: Kohn CW, ed. Guidelines for horse transport by road and air. New York: American Horse Shows Association, 2000;63–70.
