The Veterinary Journal 189 (2011) 83–88

Contents lists available at ScienceDirect

The Veterinary Journal

journal homepage: www.elsevier .com/ locate/ tv j l

The effects of hypertonic dehydration changes on renal function and argininevasopressin in the horse during pulling exercises

Ana Muñoz a,b,*, Cristina Riber a,b, Pablo Trigo b, Francisco M. Castejón b, Raquel G. Lucas c, Jorge Palacio d

a Department of Animal Medicine and Surgery, School of Veterinary Medicine, University of Córdoba, Córdoba, Spainb Equine Sport Medicine Centre (CEMEDE), University of Córdoba, Spainc Department of Animal Medicine and Surgery, Alfonso X el Sabio University, Madrid, Spaind Department of Animal Medicine and Surgery, School of Veterinary Medicine, University of Zaragoza, Zaragoza, Spain

a r t i c l e i n f o

Article history:Accepted 30 June 2010

Keywords:DehydrationElectrolytesExerciseHorsesVasopressin

1090-0233/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.tvjl.2010.06.024

* Corresponding author at: Department of Animal Mof Veterinary Medicine, University of Córdoba, Córdob68; fax: +34 957 21 10 93.

E-mail address: pv1mujua@uco.es (A. Muñoz).

a b s t r a c t

This study investigated the effect of hypertonic dehydration on exercise in horses. Valencian draft horses(n = 43) with hypertonic dehydration following water and food deprivation and competing in pullingevents (COM), were compared to euhydrated horses (n = 11) during a pulling exercise test (PET). Bothgroups were divided into bodyweight (BW) categories A (6350 kg), B (351–450 kg), and C (P451 kg),and pulled 2, 2.25, and 2.5 times BW, respectively. Exercise duration (ED) was 1.3 ± 0.8 min. Heart rate,respiratory rate, rectal temperature and blood samples were taken at rest, after exercise, and during thefirst 30 min of passive recuperation.

The only difference between BW categories was the packed cell volume (PCV), which was higher afterexercise in COM horses in group A (52.7 ± 2.3%) than in B (49.3 ± 3.4%), so the data for the three BW cat-egories were combined. The COM horses had higher levels of resting plasma proteins, albumin, urea, cre-atinine, sodium and arginine vasopressin (AVP) than PET horses. Exercise induced significantly greater(P < 0.05) increases in heart rate and lactate in the COM horses. AVP was negatively correlated with EDand positively with PCV, plasma proteins, albumin, urea, and sodium. Peak AVP concentrations did notdiffer in the two groups.

� 2010 Elsevier Ltd. All rights reserved.

Introduction

Exercise induces hypovolaemia, the degree of which will de-pend on the duration and intensity of the exercise and the environ-mental conditions (Convertino, 1987; McKeever et al., 1993b;McKeever and Hinchcliff, 1995; McKeever, 2002). An initial de-crease in plasma volume (PV) will result from increased capillaryhydrostatic pressure, leading to extrusion of water and electrolytesfrom the vascular compartment (Convertino, 1987; McKeever,1998). Further decreases occur due to sweating, with significantlosses of water and electrolytes (Hodgson et al., 1994). The changesin blood pressure and the loss of body water and electrolytes stim-ulate endocrine responses to maintain internal homeostasis.

Arginine vasopressin (AVP) is a posterior pituitary hormonewith important physiological actions, such as vasoconstriction,limiting free water clearance, stimulating thirst and drinking, andenhancing the uptake of sodium and water from the large colon(Wade and Freund, 1990; McKeever et al., 1991). Changes in the

ll rights reserved.

edicine and Surgery, Schoola, Spain. Tel.: +34 957 21 10

activity of AVP have been reported during exercise in the horse.For example, McKeever et al. (1992) found that AVP increased from4.0 pg/mL at rest to 95 pg/mL by 10 m/s during an incrementaltreadmill exercise test. Similarly, Kokkonen et al. (2002) reportedthat plasma AVP increased from 2.9 pmol/mL at rest to29.1 pmol/mL when the horses had reached 7 m/s.

During steady-state sub-maximal exercise, AVP did not rise un-til after 20–40 min of exertion (McKeever et al., 1991). Muñoz et al.(2007) reported serum AVP concentrations in successful endurancehorses, with means of 3.6, 4.4, 3.0, and 3.1 pg/mL at rest, and at 27,55 and 72 km, respectively, without significant differences be-tween the distances covered. AVP was positively correlated withdistance, serum Na, Cl, Ca, and Mg levels, total serum proteins,and albumin concentrations in endurance horses (Riber et al.,2007).

Water deprivation will induce an increase in plasma AVP levels.Houpt et al. (1989) described increases from 1.53 to 4.32 pg/mL inhealthy ponies after a water deprivation of 24 h. Similarly, humanwrestlers had higher plasma AVP concentrations when they weredehydrated (Bartok et al., 2004).

In the present study, we have investigated plasma AVP concen-trations in horses with different bodyweights (BWs) during pullingexercises. Following water and food deprivation to reduce BW the

Table 2Results of the multivariate analysis (F values and probability, P) and the effects ofthree main factors, horse group, bodyweight (BW) and sampling time.

Horse groupa BW categoryb Sampling timec

HR F = 86.06* (P = 0.000) F = 1.096 (P = 0.323) F = 24.038* (P = 0.000)RR F = 8.997* (P = 0.003) F = 3.183* (P = 0.043) F = 14.94* (P = 0.000)RT F = 2.000 (P = 0.158) F = 3.000 (P = 0.066) F = 24.00 (P = 0.000)PCV F = 58.47* (P = 0.000) F = 3.730* (P = 0.025) F = 105.5* (P = 0.000)TPP F = 949.0* (P = 0.000) F = 5.520* (P = 0.004) F = 64.89* (P = 0.000)ALB F = 341.4* (P = 0.000) F = 2.340 (P = 0.098) F = 29.62* (P = 0.000)BUN F = 559.8* (P = 0.000) F = 0.050 (P = 0.955) F = 50.19* (P = 0.000)CREAT F = 97.57* (P = 0.000) F = 4.226 (P = 0.016) F = 22.97* (P = 0.000)LA F = 86.05*(P = 0.000) F = 0.062 (P = 0.940) F = 267.4* (P = 0.000)Na F = 327.7* (P = 0.000) F = 0.300 (P = 0.720) F = 67.80* (P = 0.000)K F = 37.73* (P = 0.000) F = 2.930 (P = 0.055) F = 49.50* (P = 0.000)Cl F = 73.00* (P = 0.000) F = 2.400 (P = 0.094) F = 35.80* (P = 0.000)AVP F = 656.2* (P = 0.000) F = 1.468 (P = 0.232) F = 52.12* (P = 0.000)

HR, heart rate; RR, respiratory rate; RT, rectal temperature; PCV, packed cell vol-ume; TPP, total plasma protein; ALB, albumin; BUN, blood urea nitrogen; CREAT,creatinine; LA, lactate; Na, sodium; K, potassium; Cl, chloride; AVP, argininevasopressin.

a COM, competition, and PET, pulling exercise test.b A (6350 kg), B (351–450 kg), and C (P451 kg).c R, before exercise; E, during the first minutes following exercise; 5REC, 10REC,

15REC, and 30 REC, passive recovery at 5, 10, 15 and 30 min, respectively.* Significant at P < 0.05.

84 A. Muñoz et al. / The Veterinary Journal 189 (2011) 83–88

animals were dehydrated at the beginning of the competition. Acontrol group of horses underwent similar exertion but wereeuhydrated.

The main objectives were (1) to describe changes in AVP in rela-tion to exercise intensity, hydration, and electrolyte markers andrenal function in horses with different BWs and pulling differentloads, and (2) to determine whether hydration status affected theAVP response to exercise. We hypothesised that AVP would in-crease more in horses pulling greater loads and that this increasewould be related to hydration status. We also hypothesised thathorses with hypertonic dehydration will have higher resting AVPlevels and would experience a greater rise of AVP during exercise.

Materials and methods

Horses

This research was approved by the Ethical Committee for Animal Experimenta-tion of the Cardenal Herrera-CEU University.

Fifty-four healthy male Valencian draft horses were divided into two groups:(1) 43 competed in pulling competitions (COM) and (2) 11 underwent a pullingexercise test (PET). According to their BWs, both groups were divided into three cat-egories: A (6350 kg), B (351–450 kg), and C (P451 kg). Table 1 lists the age, BW andload of the two groups. The BW was obtained before competition and dehydrationwas not clinically scored. All animals were sound and in active training.

Pulling competitions (COM)

COM consisted of horses being led by the owner over a 60 m track of hard beachsand. The animals pulled a carriage loaded with sand at 2, 2.25, and 2.5 times theirBW (groups A, B, and C, respectively). The track was divided into three phases of20 m each. After each phase, the horses had a compulsory stop to check the abilityof the animal to continue pulling the load. The total time to cover the track was re-corded, with the compulsory stop considered as part of this time.

During COM events, owners limit water and food supply before the event to re-duce BW. The mean time without food was 15.0 ± 4.2 h and without water was10.0 ± 3.3 h.

Pulling exercise test (PET)

The PET was carried out on the same type of track and the same rules of com-petition applied, although the horses were not deprived of water or food. The foodcomprised a commercial concentrate source which together with alfalfa hay wassupplied as a proportion of BW. PET was performed at least 5 h after feeding andwater was provided ad libitum.

Data acquisition

In both COM and PET horses, data were taken before exercise (R), during thefirst minutes following exercise (E), and at 5, 10, 15, and 30 min during a passiverecovery (5REC, 10REC, 15REC, and 30REC). Heart rate (HR, bpm) was obtained byauscultation, while respiratory rate (RR, rpm) and rectal temperature (RT, �C) wereobtained by conventional methods. Jugular venous blood samples were also col-lected at the same times. Data of mean ambient temperature (�C) and humidity(%) were recorded. The animals did not have access to water or food during the sam-pling period.

Analysis of blood samples

Packed cell volume (PCV, %) was determined in the field. The remaining ve-nous blood samples were poured into two tubes containing EDTA and heparin-

Table 1Mean bodyweight (BW), load and age of the 54 horses included in this study.

COM (n = 43)

A B C

N 22 12 9BW (kg) 289.9 ± 39.27 400.1 ± 25.91 512.0 ± 62Load (kg) 587.1 ± 128.5 870.6 ± 122.8 1075 ± 13Age (years) 8.667 ± 3.927* 8.412 ± 3.583 9.333 ± 3.6

COM, competition; PET, pulling exercise test. A–C refer to different BW categories (see* Significant differences between both groups at P < 0.05.

lithium. Both tubes were centrifuged within 5 min of withdrawal, and EDTA plas-ma and heparinised plasma were collected. No protease inhibitor was used. Thesamples were refrigerated during transport to the laboratory and frozen for lateranalysis.

Heparinised plasma was analysed for total plasma protein (TPP, g/dL), albumin(ALB, g/dL), blood urea nitrogen (BUN, mmol/L), creatinine (CREAT, lmol/L) and lac-tate (LA, mmol/L) by spectrophotometry. The concentrations of Na, K, and Cl (mmol/L) were measured using an analyzer with selective electrodes (Vetlyte).

The percentage of variation of PV was calculated from a modified Van Beau-mont’s approach (Van Beaumont et al., 1973), taking into account TPP changes, be-cause PCV is affected by splenic contraction.

%DPV ¼ ½ðTPPR=TPPEÞ � 1� � 100

where R indicated resting values and E exercise values.Although this formula assumed that TPP concentrations from the vascular com-

partment did not vary with exercise, it has been used previously in horses (Schottet al., 1997).

The concentrations of AVP (pg/mL) were measured in EDTA plasma using aradioimmunoassay (for vasopressin 100 T), which had been validated for horses.The assay was specific for AVP (cross-reactivity with AVP, 100%; lysine vasopressin<0.1%; oxytocin >0.1%; vasotocin <0.1%, and desmopressin <0.1%). The sensitivitywas 1.3 pg/mL, with between-assay coefficients of variation between 3.7% and10% and a within-assay coefficient of variation of 6.9%. The extraction recovery ofknown quantities of AVP was 100.1%.

Statistical analysis

Data are presented as means ± standard deviation (SD). Normality of the vari-ables was checked with a Shapiro–Wilk test and the homocedasticity of the vari-ance with a Levene test. The parameters that were not normally distributed orwere heterocedastic (HR, RR, BUN, CREAT, K, Cl, and AVP) were transformed loga-rithmically. A multivariate test was performed over the three factors: group(COM and PET), BW category (A, B, and C) and sampling time (R, E, 5REC, 10REC,15REC, and 30REC).

PET (n = 11)

A B C

3 3 5.65* 283.3 ± 34.0 396.7 ± 37.89 586.0 ± 67.340.9* 569.4 ± 119.0 814.2 ± 109.1 1219 ± 123.039* 12.33 ± 1.280 7.333 ± 0.485 6.200 ± 1.186

text for specific weights for each category).

A. Muñoz et al. / The Veterinary Journal 189 (2011) 83–88 85

The differences between BW categories within each group and the differ-ences between groups for each BW category were assessed using an ANOVAfor independent samples. The effect of sampling time was evaluated using anANOVA for repeated measurements and a post hoc test was done to assess be-tween which sampling times there existed differences (Tukey’s test). The corre-lations between the studied variables were assessed (Pearson product-momentcorrelation). Statistical significance was fixed at P < 0.05 (Statistica for Windows,v.6.0).

Fig. 1. Mean ± SD of heart rate (HR), respiratory rate (RR), and rectal temperature(RT) in euhydrated (PET) horses and in horses with hypertonic dehydration (COM)during pulling exercises; � significant differences between PET and COM horses foreach sampling time. (A-a): Indicates significant differences between rest andexercise; and (B-b): significant differences between exercise and recuperation.Lower case letters (a, b): significant differences in COM horses. Capital letters (A, B):significant differences in PET horses. P < 0.05.

Results

Multivariate analysis showed significant influences of the threefactors, BW, hydration and sampling time (Table 2).

Effects of BW and pull exercise in COM and PET horses

In COM, PCV was higher in horses in group A (52.7 ± 2.3%) thanin group B (49.3 ± 3.4%) at E. No significant differences betweenBW categories were found in other parameters and data were pro-cessed together for each sampling time. Exercise duration (ED) was1.2 ± 1.3 min (range: 0.4–5.3 min) in BW category A, 1.5 ± 0.9 min(range: 0.4–4.1 min) in B, and 1.2 ± 0.6 min (0.4–2.5 min) in C,without significant differences between them. The PV underwentsimilar decreases in the three BW categories at E (10.9%, 10.2%,and 10.9%, respectively).

The effects of exercise for COM and PET horses, and the differ-ences between them at each exercise level, are presented in Figs.1–4. Exercise in COM led to significant increases in all the variables.The R values were recovered at 15REC for ALB, Cl, and AVP and at30REC for PCV, TPP, BUN, CREAT, Na, K, and AVP. The HR, RR, RT,and LA remained significantly higher than R values at 30REC.

During PET, horses in category B had a lower PCV at E(44.8 ± 1.0% B, 46.27 ± 1.0% A, and 47.5 ± 1.1% C), while horses in Ahad lower CREAT at 5REC (41.4 ± 5.2 lmol/L A, 56.9 ± 7.8 lmol/L

Fig. 2. Mean ± SD of packed cell volume (PCV) and lactate (LA) in euhydrated (PET)horses and in horses with hypertonic dehydration (COM) during pulling exercises; �

significant differences between PET and COM horses for each sampling time. (A-a):Indicates significant differences between rest and exercise; and (B-b): significantdifferences between exercise and recuperation. Lower case letters (a, b): significantdifferences in COM horses. Capital letters (A, B): significant differences in PEThorses. P < 0.05.

Fig. 3. Mean ± SD of total plasma proteins (TPP), albumin (ALB), urea (BUN), andcreatinine (CREAT) concentrations in euhydrated horses (PET) and in horses withhypertonic dehydration (COM) during pulling exercises; � significant differencesbetween PET and COM horses for each sampling time. (A-a): Indicates significantdifferences between rest and exercise; and (B-b): significant differences betweenexercise and recuperation. Lower case letters (a, b): significant differences in COMhorses. Capital letters (A, B): significant differences in PET horses. P < 0.05.

Fig. 4. Mean ± SD of Na, Cl, K, and arginine vasopressin (AVP) concentrations ineuhydrated horses (PET) and in horses with hypertonic dehydration (COM) duringpulling exercises; � significant differences between PET and COM horses for eachsampling time. (A-a): Indicates significant differences between rest and exercise;and (B-b): significant differences between exercise and recuperation. Lower caseletters (a, b): significant differences in COM horses. Capital letters (A, B): significantdifferences in PET horses. P < 0.05.

86 A. Muñoz et al. / The Veterinary Journal 189 (2011) 83–88

B, and 59.5 ± 7.8 lmol/L C). Horses in group C had a higher Cl at30REC (106.8 ± 1.6 mmol/L C, 104.0 ± 2.6 mmol/L A, and 102.7 ±1.2 mmol/L B). Differences at R, 10REC, and 15REC were not ob-served between BW categories. The ED was similar for the three cat-egories: A, 1.2 ± 0.3 min, range: 0.5–3.2 min; B, 1.3 ± 1.0 min, range:0.5–4.2 min; and C, 1.2 ± 0.5, range: 0.3–2.3 min. The decrease in PVwas 14.7%, 14.5%, and 16.1% for categories A, B, and C, respectively.

Exercise induced increases in all the variables in PET. Statisti-cally similar values for R were found at 5REC for K; at 10REC for

RR, TPP, ALB, CREAT, Na, and Cl; at 15REC for HR and PCV; and at30REC for BUN and AVP. LA remained higher than R after 30REC(Figs. 1–4).

Differences between COM and PET

There were no significant differences in ED between COM andPET in the three BW categories. COM showed significantly higher

Table 3Results of the analysis of linear correlation between the studied variables in horses during the pulling exercises.

ED Load HR RR RT PCV TPP ALB BUN CREAT Na K Cl LA

Load �0.140HR �0.060 0.030RR 0.110 �0.110 0.390RT 0.190 �0.130 0.470 0.600*

PCV �0.210 0.020 0.630* 0.320 0.360TPP �0.530 0.210 0.530 0.380 0.280 0.600*

ALB �0.480 0.150 0.400 0.310 0.200 0.550 0.740*

BUN �0.590 0.210 0.500 0.290 0.210 0.600* 0.740* 0.650*

CREAT �0.070 0.160 0.510 0.550 0.510 0.450 0.530 0.400 0.450Na �0.510 0.230 0.590 0.340 0.250 0.640* 0.720* 0.620* 0.720* 0.570K �0.150 0.060 0.610* 0.380 0.230 0.560 0.630* 0.520 0.560 0.440 0.640*

Cl �0.270 0.100 0.560 0.450 0.350 0.620* 0.750* 0.620* 0.530 0.550 0.710* 0.690*

LA �0.120 0.060 0.730* 0.530 0.640* 0.580 0.560 0.400 0.540 0.550 0.630* 0.510 0.570AVP �0.750* 0.250 0.570 �0.680* 0.120 0.660* 0.730* 0.680* 0.670* 0.410 0.810* 0.550 0.570 0.680*

HR, heart rate; RR, respiratory rate; RT, rectal temperature; PCV, packed cell volume; TPP, total plasma protein; ALB, albumin; BUN, blood urea nitrogen; CREAT, creatinine;Na, sodium; K, potassium; Cl, chloride; LA, lactate; AVP, arginine vasopressin.* Significant at P < 0.05.

A. Muñoz et al. / The Veterinary Journal 189 (2011) 83–88 87

TPP, ALB, BUN, CREAT, Na, and AVP at all sampling times. HR andLA were higher at E and during the recuperation in COM, but notat R. The RR was higher in COM, without significant differences be-tween groups at 5REC. Both Cl and LA were higher in COM duringthe recuperation. No significant differences were found in RT be-tween PET and COM (Figs. 1–4).

The increase from R at E was higher in PET in RR (3.1-fold PET,1.7-fold COM), ALB (1.4-fold PET, 1.2-fold COM), BUN (1.5-fold PET,1.3-fold in COM), and CREAT (1.6-fold PET, 1.4-fold COM). In con-trast, the increases in HR (2.6-fold COM, 2.1-fold PET) and LA(6.1-fold COM, 6.5-fold PET) were more intense in COM. The in-crease in AVP was similar (1.6-fold COM, 1.5-fold PET).

The recovery time to achieve values statistically similar to R waslonger for HR, RR, PCV, and Na in COM and HR and RR values didnot recover (Figs. 1–4). Table 3 presents the correlation analysisbetween the variables at E.

Discussion

This study was the first to investigate the effects of hypertonicdehydration changes on renal function and AVP in the horse duringpulling exercises. Although the horses pulled different loads, thedifferences between BW categories were minor suggesting thatpulling strength was not linearly related to body size, with heavierhorses able to pull proportionally more intense loads with thesame relative exercise intensity.

The only difference between BW categories in COM was thehigher PCV at E in category A, compared to B. Because of the lackof differences in other parameters (PV, TPP, ALB, BUN, and CREAT),it was initially thought that the higher PCV resulted from a moreintense contraction of the spleen in response to stress. We havepreviously measured catecholamine concentrations under similarconditions in horses (Lucas, 2004) and found that the highest riseduring E was found in weight category B (increase of adrenalinein E from R: A 45.82%, B 933.4%, and C a decrease of 14.88%; in-crease of noradrenalin in E from R: A 218.9%, B 416%, and C215.5%). The higher PCV in category A was therefore not directlyrelated to the intensity of the stress exercise or, alternatively, cat-echolamines were released at the beginning of exercise and in-creased values were not found at the end of E.

The higher resting values of TPP, ALB, BUN, CREAT, Na, and AVPin COM were consistent with hypertonic dehydration. In particular,the resting AVP was 2.3-times higher in COM than in PET. The re-lease of AVP is known to be induced by an increase in extracellulartonicity, detected by osmoreceptors in the hypothalamus, and a

decrease in PV, detected by cardiopulmonary baroreceptors (Wadeand Freund, 1990; McKeever and Hinchcliff, 1995).

The higher HR at E in COM may have resulted from dehydration.To maintain cardiac output, HR will generally increase to compen-sate a decrease in stroke volume (Saltin, 1964; Wade and Clayb-augh, 1980). An unexpected finding in this study was the greaterdecrease in PV during PET, which was consistent with the moremarked increases in ALB, BUN, and CREAT in response to exercise.

The unexpectedly high increases in plasma LA may have beendue to the intensity of the exercise stress inducing either anaerobicmuscle stimulation or more intense muscle contractions duringCOM in comparison with PET. Horses could also be more motivatedto exert themselves during competition compared to experimentalconditions. The slow recovery of LA in both COM and PET horsesafter E could have resulted from the passive recovery, a limitationof the blood flow, or a reduced oxidative capacity of the muscles.

McKeever et al. (1993a) observed that furosemide administra-tion and subsequent volume depletion before exercise caused agreater increase in AVP. Their result did not agree with our findingswhere the increase in AVP in COM and PET was similar in responseto exercise.

The persistence of elevated AVP after the cessation of exercisein both groups has been described previously (McKeever et al.,1991). The decrease in AVP in the recovery phase following COMand PET occurred in spite of the fact that the horses did not havefree access to water. A possible explanation may lie in the influenceof compartmental fluid shifts, uptake of Na and water from the co-lon, changes in free water clearance, or the release of angiotensin IIor cortisol (Wade and Claybaugh, 1980; Zambraski, 1990).

An exponential increase in AVP with work intensity and dura-tion has been reported in humans and horses (Wade and Clayb-augh, 1980; Convertino et al., 1981; Freund et al., 1987;McKeever et al., 1992). In contrast, we found a negative correlationbetween ED and AVP. This could have been the result of the sub-maximal intensity and the short duration of exercise. McKeeveret al. (1991, 1993b) demonstrated that the increase in AVP duringsub-maximal exercise was delayed 20–40 min because of neuraland endocrine (atrial natriuretic peptide) cardiopulmonary sup-pression of AVP. The elevation of AVP did not therefore occur untildehydration had reduced venous return, cardiac filling, and pre-sumably atrial volume and pressure.

The lack of correlation between AVP and HR reflected a method-ology problem because HR was obtained by auscultation at the endof E. HR rapidly decreased immediately after exercise and HR wasnot a reliable index of cardiovascular demands in the currentstudy. Although it has been reported that increased temperature

88 A. Muñoz et al. / The Veterinary Journal 189 (2011) 83–88

will increase AVP in humans (Takamata et al., 1995), we did notfind a positive correlation between RT and AVP.

The physiological functions of AVP explained the positive corre-lations with PCV, TPP, ALB, and BUN. The lack of correlation withCREAT could be the result of the release of CREAT from muscleactivity, since this parameter did not accurately represent glomer-ular filtration rate. An increase in natraemia would be associatedwith a rise in osmolality, leading to AVP secretion (McKeever,2002), and hence the positive correlation between AVP and Na.

This study had some methodological limitations, which areinherent with studies conducted under field conditions. COM andPET occurred on different days and locations and with differentenvironmental conditions. Data were collected during the compet-itive season, between March and August, with environmentalmean temperatures of 13.3–23.2 �C and with relative humidity of49–80%. Because of the rules of the competitions, we were not per-mitted to weigh the horses after exercise. It would have been inter-esting to study the same horses under COM and PET conditions, butthe horses were transported home soon after competition. Itshould also be noted that COM horses were transported to thecompetition venue, while PET horses were exercised on their farm.

Conclusions

Minor differences were found in the response to exercise ofhorses with different BWs pulling different loads. Horses withhypertonic dehydration had higher concentrations of AVP andother markers of hydration and renal function at rest. However,the increase in these parameters during exercise was similar ineuhydrated horses, even though HR and LA showed higher in-creases in the dehydrated group. Finally, AVP concentration wassignificantly correlated with exercise duration, hydration, valuesfor kidney markers and natraemia.

Conflict of interest statement

None of the authors of this paper has a financial or personalrelationship with other people or organizations that could inappro-priately influence or bias the content of the paper.

Acknowledgments

The authors are indebted to the owners and trainers of thehorses included in this research who allowed us to study theirhorses despite the intensity and stress of the competitions. This re-search has been generously supported by the Cardenal Herrera-CEU University (Projects PRUCH 02/12 and PRUCH 03/17).

References

Bartok, C., Schoeller, D.A., Sullivan, J.C., Clark, R.R., Landry, G.L., 2004. Hydrationtesting in collegiate wrestlers undergoing hypertonic dehydration. Medicineand Science in Sports and Exercise 36, 510–517.

Convertino, V.A., 1987. Fluid shifts and hydration status: effects of long-termexercise. Canadian Journal of Sports Science 12, 136S–139S.

Convertino, V.A., Keil, L.C., Bernauer, E.M., 1981. Plasma volume, osmolality,vasopressin, and renin activity during graded exercise in man. Journal ofApplied Physiology 50, 123–128.

Freund, B.J., Claybaugh, J.R., Dice, M.S., Hashiro, G.M., 1987. Hormonal and vascularfluid responses to maximal exercise in trained and untrained males. Journal ofApplied Physiology 63, 669–675.

Hodgson, D.R., Davis, R.E., McConaghy, F.F., 1994. Thermoregulation in the horse inresponse to exercise. British Veterinary Journal 150, 219–234.

Houpt, K.A., Thornton, S.N., Allen, W.R., 1989. Vasopressin in dehydrated andrehydrated ponies. Physiology and Behavior 45, 659–661.

Kokkonen, U.M., Pösö, A.R., Hyyppä, S., Huttunen, P., Leppäluoto, J., 2002. Exercise-induced changes in atrial peptides in relation to neuroendocrine responses andfluid balance in the horse. Journal of Veterinary Medicine A 49, 144–150.

Lucas, R.G., 2004. Systemic Adaptations to Draught Exercise in Draft Horses. PhDThesis. Cardenal Herrera-CEU University, Valencia, Spain.

McKeever, K.H., 1998. Fluid balance and renal function in exercising horses.Veterinary Clinics of North America: Equine Practice 14, 23–44.

McKeever, K.H., 2002. The endocrine system and the challenge of exercise.Veterinary Clinics of North America: Equine Practice 18, 321–353.

McKeever, K.H., Hinchcliff, K.W., 1995. Neuroendocrine control of blood volume,blood pressure, and cardiovascular function in horses. Equine VeterinaryJournal 18, 77–81.

McKeever, K.H., Hinchcliff, K.W., Schmall, L.M., Muir, W.W., 1991. Renal tubularfunction in horses during submaximal exercise. American Journal of Physiology– Regulatory, Integrative and Comparative Physiology 30, 553–560.

McKeever, K.H., Hinchcliff, K.W., Schmall, L.M., Reed, S.M., Lamb, D.R., Muir, W.W.,1992. Plasma renin activity, aldosterone, and vasopressin during incrementalexercise in horses. American Journal of Veterinary Research 53, 1290–1293.

McKeever, K.H., Hinchcliff, K.W., Cooley, J.L., Lamb, D.R., 1993a. Furosemidemagnifies the exercise-induced elevation of plasma vasopressin concentrationin horses. Research in Veterinary Science 55, 151–155.

McKeever, K.H., Hinchcliff, K.W., Reed, S.M., Robertson, J.T., 1993b. Plasmaconstituents during incremental treadmill exercise in intact andsplenectomised horses. Equine Veterinary Journal 25, 233–236.

Muñoz, A., Riber, C., Trigo, P., Castejón, F.M., 2007. Serum concentrations of renin,angiotensin, aldosterone and arginine vasopressin in endurance horses. In:Proceedings of the XIX International Conferences of Equine VeterinaryMedicine, Buenos Aires, Argentina, p. 5.

Riber, C., Trigo, P., Bataller, A., Muñoz, A., 2007. Neuroendocrine control of hydrationand electrolyte balance in endurance horses. In: Proceedings of the XIXInternational Conferences of Equine Veterinary Medicine, Buenos Aires,Argentina, p. 6.

Saltin, B., 1964. Aerobic and anaerobic work capacity after dehydration. Journal ofApplied Physiology 19, 1114–1118.

Schott II, H.C., McGlade, K.S., Molander, H.A., Leroux, A.J., Hines, M.T., 1997. Bodyweight, fluid, electrolyte, and hormonal changes in horses competing in 50- and100-mile endurance rides. American Journal of Veterinary Research 58, 303–309.

Takamata, A., Mack, G.W., Stachenfeld, N.S., Nadel, E.R., 1995. Body temperaturemodification of osmotically induced vasopressin secretion and thirst in humans.American Journal of Physiology Regulation and Integrative ComparativePhysiology 269, 874–880.

Van Beaumont, W., Strand, J.C., Petrofsky, J.S., Hipskind, S.G., Greenleaf, J.E., 1973.Changes in total plasma content of electrolytes and proteins with maximalexercise. Journal of Applied Physiology 34, 102–106.

Wade, C.E., Claybaugh, J.R., 1980. Plasma renin activity, vasopressin concentrationand urinary excretory responses to exercise in men. Journal of AppliedPhysiology 49, 930–936.

Wade, C.E., Freund, B.J., 1990. Hormonal control of blood volume during andfollowing exercise. In: Gisolfi, C.V., Lamb, D.R. (Eds.), Perspectives in ExerciseScience and Sport Medicine. Fluid Homeostasis during Exercise. Bencharmarck,Carmel, IN, USA, pp. 207–245.

Zambraski, E.J., 1990. Renal regulation of fluid homeostasis during exercise. In:Gisolfi, C.V., Lamb, D.R. (Eds.), Perspectives in Exercise Science and SportMedicine. Fluid Homeostasis during Exercise. Bencharmarck, Carmel, IN, USA,pp. 245–280.