Forosemide magnifies the exercise-induced elevation of plasma vasopressin concentration in horses

Research in Veterinary Science 1993, 55, 151-155

Furosemide magnifies the exercise-induced elevation of plasma vasopressin concentration in horses K. H. McKEEVER, K. W. H1NCHCLIFF, Department of Veterinary Clinical Sciences, J. L. COOLEY, D. R. LAMB, School of Health, Physical Education and Recreation, 7"he Ohio State University,

Columbus, Ohio 43210 USA

The purpose of this study was to test the hypothesis that furosemide administration before exercise would cause greater increases in plasma arginine vasopressin (AVe) concentration in exercising horses than exercise alone. Six adult, clinically normal, unfit mares underwent three randomly ordered 60 minute standard exercise tests on an equine treadmill to examine the effect of furosemide administration on plasma AVP concentration. In one trial, furosemide (1 mg kg -1) was infused four hours before exercise (FUR-4) and a placebo (10 ml saline) was infused two minutes before exercise; in another trial the placebo was infused four hours before exercise and drug was infused two minutes before exercise (FUR-2); in a third trial a placebo was infused four hours and two minutes before exercise (CON). During the treadmill test each mare ran up a fixed 4 ° incline for one hour at a speed previously determined to produce a heart rate of 65 per cent of each horse's maximum heart rate. Venous blood samples were collected at rest in the stall, immediately before exercise while standing on the treadmill, and at 15 minute intervals during the treadmill test. Plasma AVP concentration was measured by radioimmunoassay. In the CON trial, plasma AVP concentration increased 561 per cent (P<0-05) from 6.3 + 1.0 pg m1-1 (mean + SE) at rest to 38"8 + 12.8 pg m1-1 at the end of the 60 minute run. During the FUR-2 trial, AVP increased 1185 per cent (P<0.05) from 5-9 + 1.7 pg m1-1 to 75.8 _+ 17-7 pg m1-1. During the FUR4 trial, AVP increased 3624 per cent (P<0"05) from 3.3 + 0.5 pg m1-1 to 122.9 + 33.7 pg m1-1 at the end of the exercise. These data demonstrate that furosemide significantly enhances the AVP response to submaximal exercise in the horse.

ARGININE vasopressin (AVP) is a posterior pituitary hormone associated with the acute and chronic defence of blood pressure, plasma volume and fluid and electrolyte balance (Gauer and Henry 1976). The physiological actions of AVP include vasoconstriction, decreased free water clearance, and increased fibrinolysis (Gauer and Henry 1976, Gilmore 1983, Cowell 1986). It is well recognised that exercise causes an increase in plasma AVP concentration in humans, dogs and horses and that the increases in plasma AVP concentrations are correlated with exercise intensity (Wade and Freund 1990, McKeever et al 1991a, Zambraski 1990). The mechanism for the release of AVP appears to involve osmoreceptors in the hypothalamus and cardiopulmonary volume receptors (Gauer and Henry 1976, Gilmore 1983). It has been recently reported that AVP increases during steady-state submaximal exercise in horses, but that the increase does not become significant until between 20 and 40 minutes of exertion (McKeever et al 1991a). Vasopressin increases during submaximal exercise in horses without a change in free water clearance (McKeever et al 1991a). It has been suggested that AVP secreted during exercise acts on the vasculature to control blood pressure, whereas, sustained elevations in AVP following exercise may stimulate thirst and drinking and decrease free water clearance by the kidneys (McKeever et al 1991a, Zambraski 1990). Furosemide is a potent diuretic used to treat exercise-induced pulmonary haemorrhage in horses (Purohit et al 1979, Hinchcliff et al 1991). In exercising horses, furosemide appears to cause a decrease in plasma volume and right ventricular end-diastolic pressure (Hinchcliff et al 1991) and an increase in plasma osmolality (Hinchcliff 1990). It is likely that these furosemide-induced

151

152

K. H. McKeever, K. W. Hinchcliff , J. L. Cooley, D. R. Lamb

changes affect the AVP response to exercise; however, there have been no studies published on the effects of furosemide on the AVP response to exertion. Cowell (1986) has reported that exercise causes an increase in AVP, and that there is a high correlation between exercise intensity, plasma AVP concentration and fibrinolysis during exercise. Furosemide may increase AVP secretion during exercise, which could affect fibrinolysis in the horse. An increase in AVP may offset some of the effects of furosemide when it is used as a pre-race medication for the treatment of exercise-induced pulmonary haemorrhage. The purpose of the present study was to test the hypothesis that furosemide administration before exercise would cause greater increases in plasma AVP concentrations in exercising horses than exercise alone. Materials and methods Six normal mares ranging in age from four to 13 years and in bodyweight from 427 to 567 kg had their left carotid arteries relocated subcutaneously at least six months before the experiment. The horses were accustomed to running on a treadmill, but were otherwise untrained. This experiment was performed in accordance with the Guiding Principles in the Care and Use of Animals of the American Physiological Society and with the regulations of the Animal Care and Use Committee of the Ohio State University. The general experimental design included three randomly ordered standardised exercise tests separated by no less than seven days. In one trial, furosemide (1 mg kg -I) was infused four hours before exercise (Fua-4) and a placebo (10 ml saline) was infused two minutes before exercise; in another trial the placebo was infused four hours before exercise and the drug was infused two minutes before exercise (FUR-2); in a third trial a placebo was infused four hours and two minutes before exercise (CON). The above dose of furosemide was chosen because it has been shown (Hinchcliff et al 1991) to produce a pronounced diuresis. Furosemide was administered either four hours or two minutes before exertion in the present study so as to separate the effects due to fluid loss and a reduction in plasma volume (four hours) from the direct cardiovascular actions of the drug (two minutes). During the exercise test the mares ran on a treadmill (Sato, Uppsala, Sweden) at a fixed 4 ° incline for one hour at a speed previously deter-

mined during an incremental speed test (Rose and Evans 1987) to produce a heart rate of 65 per cent of each mare's maximum heart rate. The trials were conducted between 07.00 and 13.00 with an ambient temperature of 17-5 _+ 0.6°C (mean + sE) and humidity of 65.0 _+ 1.6 per cent. Access to feed and water was denied during the five hour experimental period. Before the study, an introducer catheter (Argon, Houston, Texas) was percutaneously placed in the right jugular vein of each horse. The catheters were inserted using a local anaesthetic (lignocaine) and sterile conditions. A polyethylene (PE 240, Becton Dickinson, Parsipany, New Jersey) tube was inserted through the introducer catheter and positioned within the right atrium. Catheter placement was verified before and after the trial by examination of the atrial pressure waveform displayed on a physiological recording system. Blood samples (20 ml) for measurement of plasma vasopressin (AVP) concentration were obtained from the catheter at rest and at 15 minute intervals during the treadmill test. The blood was placed into-pre-chilled EDTA-treated tubes (Vacutainer, Becton Dickinson, Parsipany, New Jersey) and centrifuged (1500 g) for 20 minutes at 4°C. Plasma samples were stored at -80°C for later analysis. Previously reported radioimmunoassay procedures were used to measure concentrations of AVP in plasma (McKeever et al 1991a). The assays were conducted in single runs to avoid interassay variation. Plasma for the measurement of AVP was extracted using octadecasilica columns (Instar, Stillwater, Minnesota) before radioimmunoassay. Standards and samples of horse plasma were run in duplicate. Additionally, a plasma sample from one of the horses was spiked with 40 pg of AVP standard as a control. A recovery of 89 per cent of this added AVP standard was achieved after extraction and radioimmunoassay. Parallelism data indicated excellent linearity for assays of serial diluted samples of plasma. Comparison of the reactivity of the vasopressin antibody used in the kit with several peptides indicated a cross-reactivity with AVP of 100 per cent; with lysine vasopressin of 600 per cent; with oxytocin, under 0.01 per cent; and with vasotocin, 0.14 per cent. The horse, like humans, camels, cats, dogs and rats produces AVP (McDonald 1977). Pigs, guinea pigs and armadillos produce lysine vasopressin (McDonald 1977). The assay was sensitive to 2.5 pg m1-1 (two

Furosemide and AVP during submaximal exercise

standard deviations above zero), and exhibited a within-assay coefficient of variation of 8 per cent. An analysis of variance for repeated measures was used to detect significant main effects due to treatment group or time, plus group by time interactions. One-way analysis of variance for repeated measures, Dunnett's test and a Student's Newman-Keul's test were used as appropriate to determine the significant differences between baseline and exercise means within each group and the differences between treatment means at each time period. The null hypothesis was rejected at P<0-05.

153

group between 45 and 60 minutes of exercise, whereas in the FUR-4 group, AVP continued to increase throughout the exercise trial. Compared to the resting 0 minutes control, one hour of exercise caused a 516 per cent increase in AVP in the CON trial, a 1185 per cent increase in the FUR-2 trial, and a 3624 per cent increase in the FUR-4 trial. Discussion

The resting and exercise AVP concentrations measured in the CON trial of the present experiment are similar to those previously reported for horses (McKeever et al 1987, 1991a, b) and Results humans (Wade and Claybaugh 1980, Convertino Mean _+ SE plasma AVP concentrations are et al 1981, Freund et al 1987). The present authors shown in Fig 1. There were no differences did not observe a significant increase in AVP until (P>0.05) in AVP for samples collected while the 30 minutes of exercise in the present experiment, horses were either standing in the stall or four which is consistent with previously reported studhours later while standing on the treadmill imme- ies showing that AVP did not increase during the diately before the treadmill test. There were first 15 to 40 minutes of submaximal steady-state increases in plasma AVP in all three groups of exercise in horses (McKeever et al 1991a, b). Both horses after 30 minutes of exercise (P<0.05), but the duration and the intensity of exercise can influthere were no significant differences between the ence the release of AVP (Wade et al 1989, Wade groups. After 45 minutes of exercise, the AVP con- and Freund 1990) and the timing of the AVP centration was significantly greater in the FUR-4 response observed in the present study may have and FUR-2 groups than in the CON group. been influenced by the relatively low intensity of Interestingly, AVP reached a plateau in the FUR-2 the treadmill test. This point is supported by the 1 75 observation that AVP is released sooner in humans exercising at work intensities above 70 per cent of ~: 150 V%MAX(Wade and Freund 1990). Interestingly, in A--A FUR-4 z # another study of horses, McKeever et al (1992) o 125 A reported that AVP increased significantly by the Z ° fourth, 1.5 minute stages of an incremental exercise test, that is, after six minutes. Plasma AVP z 4 100 t may have increased earlier in the previously uJ o. 7 5 . reported study than in the present study because ~t the intensity of that stage of the exercise was O o3 ~. 5o. about 85 per cent of that shown to produce maximal heart rate, well above the reported threshold 25. for AVP release (Wade and Freund 1990). Data from the present study demonstrate that 00, ~ furosemide administration magnified the plasma Stall 0 15 30 45 60 AVP response to steady-rate submaximal exercise. MINUTES Furthermore, furosemide given four hours before FIG 1: Mean + SE for plasma vasopressin concentrations while exercise had a greater effect on plasma AVP t h a n resting in the stall, four hours later while standing on the treadmill furosemide given just before the exercise. It is (0 minutes) and after 15, 30, 45 and 60 minutes of exertion. In the well recognised that furosemide acts directly on CON trial the horses received no drug; in the FUR-2 trial, the horses were given furosemide two minutes before exercise; and in the the cardiovascular system in addition to its FUR-4 trial the horses received furosemide four hours before the diuretic actions. Plasma volume was decreased standardised exercise test. Means with different superscripts (*, t, during the initial four hours following furosemide or #) were significantly (P<0.05) different

oz I 00__00 FUR-2CON °

+

]

154

K. H. McKeever, K. W. HinchcIiff, J. L. Cooley, D. R. Lamb

administration (Hinchcliff et al 1991). In the present study, furosemide was administered either four hours or two minutes before exertion so as to separate the effects due to fluid loss and a reduction in plasma volume from the direct cardiovascular actions of the drug. Along with the haemodynamic effects, it is likely that the AVP response observed in all three trials was also influenced in part by an increase in plasma osmolality during exercise. A 1 to 2 mOsm litre -1 increase in plasma osmolality is sufficient to cause an increase in plasma AVP (Wade et al 1989). Osmolality data from the horses used in the present study have been reported previously (Hinchcliff 1990). Osmolality increased slightly (from resting means + SE of 286 + 1, 285 +_ 1, 284 + 1 mOsm kg-1 up to 287 _+ 1,289 _+ 1,286 + 1 mOsm k g -1 at 60 minutes of exercise during the CON, FUR-2 and FUR-4 trials, respectively) in all treatment groups (Hinchcliff 1990), but the increase was significant only during the FUR-2 exercise trial. Thus, changes in plasma osmolality do not explain the greater increase in plasma AVP in the FUR-2 and FUR-4 trials. The changes in plasma AVP may be more closely related to furosemide-induced decreases in right ventricular end-diastolic pressure (RVEDP) than to elevated osmolality. A decrease in RVEDP would correspond to a decrease in both atrial filling pressure and atrial stretch (Holmgren 1956, Gauer and Henry 1976, Gilmore 1983). In turn this decreased filling would be sensed by the cardiopulmonary baroreceptors, which would stimulate the release of AVP via afferent vagal pathways to the hypothalamus (Gauer and Henry 1976). There was no significant RVEDP during the CON trial; however, RVEDP decreased significantly from an average of 21 + 2 mm Hg immediately before exercise to 7 +2 mm Hg after 60 minutes in the FUR-2 trial (Hinchcliff 1990, Hinchcliff et al 1990). The decrease in RVEDP was more substantial, from 18 +- 3 down to 2 + 1.5 mm Hg in the FUR-4 trial, and this decline was significantly greater than that measured in FUR-2 (Hinchcliff 1990, Hinchcliff et al 1991). In the FUR-2 trial furosemide did not have time to exert a diuretic effect, but may have caused vasodilation and a reduction in venous return and cardiac filling pressures, as evidenced by a reduction in RVEDP. This may have stimulated a greater release of AVP in the FUR-2 trim than in the CON trial. Additionally, the fact that furosemide given four hours before exercise caused a 14 per cent decrease in plasma volume

(Hinchcliff 1990) to decrease further venous return may explain the greater decreases in RVEDP in FUR-4 than in FUR-2. Thus, in the FUR-4 trial, furosemide may have elicited the greatest AVP response by the combined effects of a decrease in plasma volume and a direct effect on venous compliance. The authors are unaware of any published reports on the effects of furosemide on the AVP response to exercise in horses. Thus, the present study addressed a basic physiological question rather than a clinical question, that is, what is the effect of superimposed dehydration on the hormonal response to prolonged submaximal running. Therefore, the constant load submaximal exercise test was used in the present experiment to study the basic versus clinical effects of furosemide administration. In summary, furosemide administration before exercise caused greater increases in plasma AVP concentration during exercise; this effect may be a response to decreased cardiac filling pressures elicited by the haemodynamic and volume depletion effects of the drug. The effect of furosemide on AVP may be important to the exercising horse, because AVP stimulates tissue plasminogen activating factor and fibrinolysis during exercise (Cowell 1986). In fact, there appears to be a relationship between exercise intensity and AVP concentration (Cowell 1986, McKeever et al 1992). If AVP stimulates fibrinolysis and furosemide stimulates greater AVP release, it may be counterproductive to give furosemide as a treatment for exercise-induced pulmonary haemorrhage. However, it should be recognised that a low intensity exercise test was used, and therefore one cannot extrapolate these results to acute short term, high intensity exercise. It may be, though, that there is another more important mechanism behind the effect of furosemide in controlling bleeding. The increase of AVP late in exercise appears to be related to volume depletion. This endocrine response may be important in the acute control of blood flow and pressure during exercise and the stimulation of drinking and renal mechanisms for the replenishment of lost fluid volume after exercise (Wade and Freund 1990, Zambraski 1990).

Acknowledgements The authors thank Kevin Kirby and Shelly Zimmerman for their help in the collection and

Furosemide and AVP during submaximal exercise

processing of the data. This study was supported by the Ohio Thoroughbred Research Fund, the Ohio Standardbred Research Fund, and the Ohio Animal Health Trust. References CONVERTINO, V. A., KEIL, L. C. & BERNAUER, E. M. (1981) Plasma volume, osmolality, vasopressin, and renin activity during graded exercise in man. Journal of Applied Physiology 50, 123-128 COWELL, J. A. (1986) Effects of exercise on platelet function, coagulation, and fibrinolysis. Diabetes and Metabolism Review 1,501-512 FREUND, B. J., CLAYBAUGH, J. R., DICE, M. S. & HASHIRO, G. M. (1987) Hormonal and vascuIar fluid responses to maximal exercise in trained and untrained males. Journal of Applied Physiology 63,669-675 GAUER, O. H. & HENRY, J. P. (1976) Neuralhumoral control of plasma volume. In International Review of Physiology: Cardiovascular Physiology II. Ed. A. C. Guyton and A. W. Cowley. Vol 9. Baltimore, University Park Press. pp 145-190 GILMORE, J. P. (1983) Neural control of extracellular volume in the human and nonhuman primate. In Handbook of Physiology: The Cardiovascular System, Peripheral Circulation and Organ Blood Flow. Section 2, vol IlI, part 2, chapter 24. Bethesda, Maryland: American Physiological Society. pp 885-915 HINCHCLIFF, K. W., McKEEVER, K. H. & MUIR, Ill, W. W. (1991) Furosemide-induced changes in plasma and blood volume in horses. Journal of Veterinary Pharmacology and Therapeutics 14, 411-417 HINCHCLIFF, K. W., McKEEVER, K. H. & MUIR, III, W. W. (1990) Extra renal effects of furosemide during exercise. Proceedings, American Association of Equine Practitioners Annual Convention 36, • 193-197 HINCHCLIFF, K. W. (1990) Modification of the physiologic responses to sustained submaximal exertion in horses by phenylbutazone and furosemide. PhD dissertation, Ohio State University. pp 1-164 HOLMGREN, A. (1956) Circulatory changes during muscular work in man with special reference to arterial and central venous pressures in the systemic circulation. Scandinavian Journal of Clinical Laboratory Investigation 8, Suppl 24, 1-97 McDONALD, L. E. (1977) The pituitary gland. In Veterinary En docrinology and Reproduction, 2nd edn. Ed L. E. McDonald. Philadelphia. Lea & Febiger. pp 35-36

155

McKEEVER, K. H., HINCHCLIFF, K. W. & COOLEY, J. L. (1991b) Acute volume load during exercise in horses: atrial natriuretic peptide, vasopressin, and hemodynamics. Medicine and Science in Sports and Exercise 23, S 104 McKEEVER, K. H., HINCHCLIFF, K. W., SCHMALL, L. M. & MUIR, IlI, W. W. (1991a) Renal tubular function in horses during exercise. American Journal of Physiology (Regulatory, Integrative, Comparative Physiology) 261, 553-560 McKEEVER, K. H., HINCHCLIFF, K. W., SCHMALL, L. M., LAMB, D. R. & MUIR, III, W. W. (1992) Changes in plasma renin activity, aldosterone, and vasopressin, during incremental exercise in horses. American .Journal of Veterinary Research 53, 1290-1293 McKEEVER, K. H., SCHURG, W. A., JARRETT, S. H. & CONVERTINO, V. A. (1987) Exercise training-induced hypervolemia in the horse. Medicine and Science in Sports and Exercise 19, 21-27 PUROHIT, R. C., NACHREINER, R. F., HUMBURG, J. M., NORWOOD, G. L. & BECKETT, S. D. (1979) Effect of exercise, phenylbutazone, and furosemide on the plasma renin activity and angiotensin I in horses. American Journal of Veterinary Research 40, 986-990 ROSE, R. J. & EVANS, D. L. (1987) Cardiovascular and respiratory function in the athletic horse. In Equine Exercise PhysioIogy Ih Eds J. R. Gillespie and N. E. Robinson. Davis, ICEEP Publications, pp 1-24 WADE, C. E. & CLAYBAUGH, J. R. (1980) Plasma renin activity, vasopressin concentration, and urinary excretory responses to exercise in men. Journal of Applied Physiology 49, 930-936 WADE, C. E. & FREUND, B. J. (1990) Hormonal control of blood volume during and following exercise. In Perspectives in Exercise Science and Sports Medicine, vol 3: Fluid Homeotasis during Exercise. Eds C. V. Gisolfi and D. R. Lamb. CarmeI, Benchmark Press. pp 207-245 WADE, C. E., FREUND, B. J. & CLAYBAUGH, J. R. (1989) Fluid and electrolyte homeotasis during and following exercise. In Hormonal Fluid and Electrolytes. Eds J. R. Claybaugh and C. E. Wade. New York, Plenum. pp /-44 ZAMBRASKI, E. J. (1990) Renal regulation of fluid homeostasis during exercise. In Perspectives in Exercise Science and Sports Medicine voi 3: Fluid Homeotasis during Exercise. Eds C. V. Gisolfi and D. R. Lamb. Carmel, Benchmark Press. pp 247-280

Received April 3, 1992 Accepted February 2, 1993