Assessment of body fluids and blood parameters associated with rapid weight change in heavy horses

Abstract

Horses competing in pulling competitions often undergo rapid weight change to enter lower weight categories. The aim of this study was to assess weight change and the associated changes of body water compartments in pulling horses. Weight change was attributed primarily to body water losses; however, losses from other sources were also indicated.

The Heavy Horse Pull is a competition that showcases horses’ strength and anaerobic power. Teams of 2 draft horses pull a weighted sled over 4.3 m at increasing load until a maximum weight is reached, often up to 2.5 times their body weight (1). Since this is a strength event, horses are divided into weight categories to ensure equal and fair competition. Similar to some human athletes who attempt to “make weight,” owners may restrict the food and water intake of horses prior to weigh-in (commonly referred to as “shrinkage”). Though not well-documented in scientific literature, some pulling competitions explicitly state that excessive shrinking is grounds for disqualification, indicating the commonality of this method. Previously, we found that between arrival and pre-competition (day of competition, before the pulling event), body weight can increase up to 8.6% in the middleweight division horses, which is indicative of the total weight lost prior to weigh-in on arrival (1). However, it is still unclear if this weight shrinkage is mainly achieved through the restriction of water or feed.

The objectives of this study were therefore to further quantify the weight change, the source of weight change, and the associated body water changes in horses competing in the Heavy Horse Pull by monitoring weight, blood chemistry, and body compartments fluid status.

A total of 52 horses competed in the 2016 Calgary Stampede Heavy Horse Pull [43 Belgians, 5 Percherons, 3 Shires, and 1 Clydesdale; median age: 8 y (IQR: 7 to 11); 12 light, 22 medium, 18 heavyweight].

A mandatory weight monitoring and physical examination (Calgary Stampede official “Fitness to Compete”) was conducted on all 52 horses upon arrival. This included a hydration status assessment (mucous membranes, jugular and capillary refill, sunken eyes), cardiac auscultation, strength of pulse, and lameness examination.

The divisions of the competition are based on team (2 horses) entry weight (lightweight: < 1451 kg, middleweight: 1452 to 1587 kg, heavyweight: > 1588 kg). All horses (n = 52) were weighed on a digital scale between 6 am and 7 am upon arrival (day 1: Arrival) and assigned to a division. All horses were again weighed on the morning before competition (days 2, 3, and 4 for lightweight, middleweight, and heavyweight, respectively: Competition day) for the purpose of the study, although this had no effect on division designation. Out of the 52 horses competing, 16 (31%) were voluntarily enrolled in the study for blood and data collection for which owners completed a consent form. This study was approved by the University of Calgary Veterinary Sciences Animal Care Committee.

Of the 16 horses for which consent was obtained for blood collection, 15 were sampled on arrival and 15 were sampled at pre-competition due to difficulties sampling 2 different horses. Venous blood samples were drawn in vacutainer tubes at 2 time points: on arrival and on the morning of the competition. A small portion (< 2 mL) of this blood was immediately tested for whole blood lactate (using a handheld analyzer, Lactate Scout+; EKF Diagnostics, Penarth, Wales), total protein, and packed cell volume (PCV). The remaining sample was left to clot, then centrifuged (10 min at 1500 × g) and frozen onsite at −20°C until undergoing chemistry analysis later that day.

Bioelectrical impedance spectroscopy (BIS) is used to determine the body composition of humans and has been validated in horses (2–7). This technique measures the resistivity of the body to a current, which is then used to assess total body water (TBW), extracellular water (ECW), and intracellular water (ICW) (2). This is achieved by passing a low frequency current between sets of electrodes, based on the understanding that bone and fat will have high resistivity (poor conductivity), while water will have low resistivity (good conductivity) (7). Bioelectrical impedance spectroscopy was conducted on 14 horses whose owners had consented to additional data collection (14 horses tested on arrival; 10 horses tested pre-competition due to owner compliance or difficulties sampling). A body proportion factor (Kb) of 1.42 was used as previously described, based on body measurements from draft horses (8). Resistivity coefficients previously determined by Ward et al (2) were applied. Four hypodermic needles acted as electrodes and were placed at anatomical locations, as previously described (2,8,9). The distance between both sets of electrodes was measured to ensure consistent placement on both days of measurements. Bioelectrical impedance spectroscopy readings were taken on 2 days (arrival and competition days), each of which consisted of 5 scans using a spectrometer (ImpediVet SFB7; ImpediMed, Pinkenba, Australia). Horses were not sedated, but were restrained in stalls. Analysis was conducted using the manufacturer’s software.

Body weight difference between arrival and pre-competition was tested using a paired t-test for all horses (n = 52). A Wilcoxon matched-pairs signed rank test was used for within category (light, middle, and heavy) weight comparisons, to assess differences in blood chemistry parameters between arrival and pre-competition, and to assess body water compartments. Associations between blood chemistry parameters from arrival samples were examined using a Pearson Correlation (data from all sampled horses together). Pearson Correlations were also used to analyze the associations between body weight and body water compartments. All data are presented as median and interquartile range. A level of P ≤ 0.05 was considered to be significant.

No horses had to be excluded from the study or competition based on the veterinary examination. The clinical assessment of the hydration status revealed that 90% of horses were < 5% dehydrated, with 1 at 5% to 8% dehydrated, and 4 at 8% to 10% dehydrated (10). Body weight for all horses increased overall by 5.4% between arrival [median: 778.8 kg (IQR: 753.2 to 841.4 kg)] and pre-competition [median: 818.3 kg (IQR: 783.8 to 869.2 kg)] (P < 0.0001). Body weight increased in the light and middleweight categories by 7.4% (P = 0.002) and 5.8% (P < 0.0001), respectively, but did not change in the heavyweight division. Had the pre-competition weights been used instead of the arrival weights, 4 of the 6 lightweight teams would have qualified as middleweight teams. Similarly, 8 of the 11 middleweight teams would have qualified for the heavyweight division.

Significant differences in plasma parameters between arrival and pre-competition are presented in Table 1. Correlations between weight change and plasma parameters were not significant. However, correlations between plasma parameters on arrival were found. Hematocrit was correlated with creatinine (r = −0.65, P = 0.01), total protein (r = 0.60, P = 0.02), sodium (r = −0.62, P = 0.01), and lactate (r = 0.63, P = 0.02). Sodium was correlated with chloride (r = 0.78, P = 0.001), and total protein (r = −0.65, P = 0.008). Potassium was correlated with total protein (r = 0.64, P = 0.01), glucose (r = 0.65, P = 0.009), and aspartate aminotransferase (AST) (r = 0.57, P = 0.03). Creatine kinase (CK) and AST were also correlated (r = 0.59, P = 0.02).

Table 1.

Results of plasma analyses from the 2016 Heavy Horse Pull (n = 15).

	Arrival	Pre-Competition
Packed cell volume (%)	36.0 (34.4 to 42.0)	34.0 (32.0 to 34.0)a P = 0.001
Total protein (g/L)	78.0 (77.0 to 82.0)	75.0 (73.0 to 77.0)a P = 0.009
Albumin (g/L)	34.0 (30.5 to 35.0)	31.0 (30.0 to 33.0)
Creatinine (mmol/L)	105.0 (101.5 to 115.0)	106.0 (100.5 to 123.0)
Urea (mmol/L)	5.6 (5.3 to 6.1)	4.8 (4.4 to 5.6)
Glucose (mmol/L)	6.5 (6.2 to 6.9)	5.4 (5.2 to 5.9)a P < 0.0001
AST (IU/L)	355.0 (330.0 to 486.0)	349.0 (329.0 to 416.5)a P = 0.002
Lactate (mmol/L)	0.8 (0.6 to 0.9)	0.5 (0.4 to 0.6)a P = 0.02
Sodium (mmol/L)	135.0 (133.5 to 136.0)	133.0 (131.5 to 134.0)
Potassium (mmol/L)	3.4 (3.0 to 4.2)	4.0 (3.7 to 4.3)
Chloride (mmol/L)	97.0 (95.5 to 99.5)	98.0 (96.0 to 99.0)
Calculated osmolarity (mmol/kg)	271.0 (268.0 to 272.0)	265.0 (262.5 to 267.5)a P = 0.03

Median and interquartile ranges shown.

^aIndicates significantly different from arrival.

AST — aspartate aminotransferase.

No differences were found in water compartments between arrival and pre-competition for horses overall (all 3 weight categories). However, TBW increased between arrival and pre-competition for the combined lightweight and middleweight horses (n = 8) [TBW_Arrival = 402.0 L (range: 343.1 to 433.7 L), TBW_Competition = 429.9 L (range: 381.5 to 452.8 L)], P = 0.008. Additionally, ECW showed a trend to increase following arrival [ECW_Arrival = 154.6 L (range: 134.4 to 158.0 L), ECW_Competition = 167.1 L (range: 147.4 to 184.9 L)], P = 0.08. ICW did not change [ICW_Arrival: 261.9 L (range: 234.5 to 278.1 L), ICW_Competition: 252.6 L (range: 231.4 to 275.0 L)]. Strong positive correlations were found between body weight and TBW (r = 0.65, P = 0.007) (Figure 1), and body weight and ECW (r = 0.74, P = 0.0009) (Figure 2). A moderate positive correlation was found between body weight and ICW (r = 0.51, P = 0.04) (Figure 3).

Figure 1.

Correlations between body weight and total body water for horses competing in the Heavy Horse Pull.

Figure 2.

Correlations between body weight and extracellular fluid for horses competing in the Heavy Horse Pull.

Figure 3.

Correlations between body weight and intracellular fluid for horses competing in the Heavy Horse Pull.

This study reports the physiological effects associated with the “shrinkage” or rapid weight loss in heavy horses competing in pulling events. We have previously examined this practice during competition (1). Although the change in weight was quantified, the changes in body water compartments and the source of those losses (body water or gut filling) could not be determined at that time. Due to the international nature of the event, standardized recording of the horses’ weight on-farm was not feasible. However, the weight gained during the ~34 to 72 h between arrival and competition is assumed to be indicative of the weight that was originally lost before check-in. This change in weight between arrival and competition day was similar to what we reported for a previous Heavy Horse Pull competition (1). This indicates the commonality and consistency of the practice of “shrinkage.” Based on the physical indicators of dehydration, 10% of the horses (5 individuals) were found to be dehydrated between 5% to 8% (10). Similarly, body weight increased between arrival and competition day in the light- and middle-weight categories by 7.4% and 5.8%, respectively. We analyzed blood samples and used BIS as an attempt to document the body water shifts to differentiate between gut and water losses as a form of rapid weight change in this population of heavy horses.

Blood sample analysis was used as an indicator of hemoconcentration. While mild to moderate decreases in hematocrit (5.7% change), total protein (3.9% change), and plasma lactate (46.2%) were observed between arrival and pre-competition, there were no significant changes in electrolytes, creatinine, or urea (Table 1). Overall this suggests some moderate compensated hemoconcentration on arrival. Similarly, the lack of correlations between weight change and plasma parameters suggest that the changes in weight were not solely due to changes in body water (dehydration and hypovolemia). Of note, the plasma lactate levels stayed within normal physiological ranges but decreased significantly (P = 0.04) between arrival and pre-competition for all horses. Unlike data we reported for a previous Heavy Horse Pull competition (1), no severe hemoconcentration was observed. As expected, creatinine being an approximate marker of renal glomerular filtration, was correlated with other plasma parameters (total protein, sodium, lactate) that change with hydration and intravascular volume status. Hypokalemia was observed in 4 horses upon arrival, possibly indicating that the horses had been kept off feed or that they received furosemide before arrival (11). The hyperglycemia noted on arrival was attributed to the stress of transportation and handling. The correlation between CK and AST was expected, as both are markers of muscle damage.

Bioelectrical impedance spectroscopy was used to monitor body water shift between compartments between arrival and pre-competition. Unlike radioactive or dye dilution techniques, BIS is not a reference method to assess TBW and ECW (12,13), but has been validated in horses (2). Additionally, a Kb value specific to draft horses has been suggested (8). Bioelectrical impedance spectroscopy has previously been compared against deuterium dilution methods, and was found to have limits of agreement of ±11.6% for TBW (2). A benefit of the BIS method is that it provides an estimate of the ECW and ICW in addition to TBW without being invasive. The BIS technique is practical for field studies where dilution techniques cannot be used on client-owned horses. The TBW and ICW calculated in the present study were respectively ~41% and ~32% greater than values previously described in a group of mixed breed adult horses, using the same device and methods (2). This result matched well with the 31.0% to 42.0% heavier body weight of the draft horses studied here. Tracer dilution has found TBW to range from 55.7% to 67.7% of total body weight (6) in euhydrated horses (Standardbred, Thoroughbreds, Percheron) and ponies, which is slightly greater than the median TBW at 53.5% of body weight in the present population of heavy horses on arrival. This confirms the moderate dehydration status of the horses on arrival and suggests that the BIS measurements are accurate for TBW. The increases found in TBW, along with the correlations between water compartments and body weight, indicate the role that water plays in attaining rapid weight change. However, there were a number of horses that underwent large weight changes without signs of dehydration or hemoconcentration, indicating losses from sources other than body water. This is potentially due to a change in intestinal content or “gut filling.” Additionally, it is common practice in racing horses to reduce the percentage of roughage while increasing the percentage of concentrate (grain), resulting in water losses from the gastrointestinal tract. This can result in weight loss by up to 10% (14). It is possible that some of the weight loss was achieved by methods similar to these, however that cannot be confirmed by this study.

In conclusion, the horses arrived in a mild to moderate state of dehydration. The data suggest that the rapid weight change can be attributed primarily to body water losses (based on blood and BIS analysis); however, other sources were also indicated.