J. Physiol. (1974), 236, pp. 403-412 With 4 text-ftgure8 Printed in Great Britain
403
THE EFFECT OF ANTIDIURETIC HORMONE ON HUMAN SWEATING
By JUDITH A. ALLEN AND I. C. RODDIE From the Department of Physiology, The Queen's University of Belfast, Belfast BT9 7BL
(Received 29 June 1973) SUMMARY
1. Changes in insensible perspiration and sweating were followed in normal subjects by continuously monitoring total body weight loss in environmental temperatures of 18, 29 and 370 C. 2. Pharmacological doses of ADH had no effect on cutaneous water loss at 180 C. 3. Pharmacological doses of ADH are capable of increasing the rate of cutaneous water loss in human subjects who are close to or above the thermal sweating threshold. 4. Physiological doses of ADH had no effect on cutaneous water loss in either cool or hot environments. 5. At normal rates of secretion in the body, ADH probably does not influence human sweat secretion. INTRODUCTION
Antidiuretic hormone (ADH) promotes transport of water through toad and frog skin (Heller, 1945; Fuhrman & Ussing, 1951; Koefoed-Johnsen & Ussing, 1953). It also plays an important part in the water balance of man by its effect on the renal tubular epithelium. It was therefore of interest to investigate its effects on cutaneous water loss in man. ADH acts in the kidney by increasing the permeability to water of the walls of the distal convoluted tubules and collecting ducts. In its presence water is reabsorbed from the tubular lumen and conserved by the body. If no circulating ADH is present, water is not reabsorbed and diuresis ensues. If ADH has a similar action on the sweat gland tubules, it might be expected to decrease the rate of sweating by causing reabsorption of water from the bypotonic tubular fluid. If ADH increases the permeability of human skin to water, it might be expected to alter the rate of insensible perspiration. In previous work on the subject, sweating was measured from small
JUDITH A. ALLEN AND I. C. RODDIE 404 areas of skin or by single measurements of body weight. Some workers found that sweating was unaltered following treatment with ADH (Amatruda & Welt, 1953; Pearcy, Robinson, Miller, Thomas & de Brota, 19,56; Ladell & Whitcher, 1960; Ratner & Dobson, 1964; Senay & van Beaumont, 1969). Others reported that ADH decreased the rate of sweating (Hankiss, 1959; Mases, Falet, Joly & Houdas, 1962; Fasciolo, Totel & Johnson, 1969; Schlein, Spooner, Day, Pickering & Cade, 1971). Ladell (1948) found that in men working in moderate heat, ADH increased the rate of sweating of light sweaters and decreased the rate of sweating of heavy sweaters. In the present experiments, the effects of ADH administration on whole body sweating and insensible perspiration in man have been studied. The methods differ from those in earlier investigations in that continuous records of evaporative water loss from the entire body surface were made before, during and after ADH infusion or injection with the body in a relatively steady-state condition. This permitted the time course of the changes in evaporative water loss to be described with greater precision. METHODS
Changes in whole body sweating and insensible perspiration were followed by continuously monitoring total body weight loss (Allen, Grimley & Roddie, 1971). The subjects wore only shorts and lay supine on a bed in a heat chamber in which the temperature could be maintained at any desired level. At the start of an experiment, relative humidity in the chamber was in the range of 30-40 % and did not alter during an experiment. The subjects were not dehydrated nor water-loaded before the experiments. Mouth and forehead and hand skin temperatures were monitored using thermocouples. The subjects were all students or staff in the Department of Physiology. There were three types of experiment. Group 1. The effect of a large dose of ADH was studied in each of five subjects at environmental temperatures of 18, 29 and 370 C. Subjects were allowed to equilibrate at 370 C for 1 hr before any measurements were made. In each experiment, after a control period of 30 min, Pitressin (Parke, Davis & Co.) 10 u. was injected s.c., and recording was continued for a further 90 min. A further experiment was performed on one of these subjects in an environmental temperature of 340 C. After a control period of 30 min, 5 u. Pitressin diluted in 0-9 % saline was infused i.v. over a period of 30 min via a catheter in an antecubital vein. Group 2. Two-hour control experiments were carried out on the same five subjects under identical conditions at 29 and 37° C but with no injection of Pitressin. Group 3. The effects of physiological doses of ADH (Lauson, 1951; Barraclough & Jones, 1970) were studied in the same five subjects at environmental temperatures of 29 and 370 C. After a 30 min control period, 10 m-u. Pitressin was injected i.v. via an indwelling catheter in an antecubital vein. Observations were continued for a further 90 min. In additional experiments at 290 C on two of the same subjects, Pitressin waes infused i.v. for a 20 min period at a rate of 45 m-u./hr.
EFFECT OF ADH ON SWEATING
405
RESULTS
Fig. 1 shows records of weight loss from experiments at 18, 29 and 370 C. The gradient of the weight loss slopes represents the rate of total body weight loss of the subject, a steeper slope indicating a greater rate of weight loss. The fluctuations on the trace are due to respiratory and other movements. It can be seen that during the first 30 min, the rate of weight ADH
Weight loss (g)
t [
I0
Weight loss (g) 20
1 _
Weight loss (g) 20
>
A
h
\
K\Xf~s\Xf \ \\' ..................................................
I
............
66"
I
60 120 Min Fig. 1. The effect of 10 u. Pitressin injected s.c. on the rate of total body weight loss in environmental temperatures of 180 C (top trace); 290 C (middle 0
trace) and 37° C (bottom trace).
loss was low at 180 C, only slightly greater at 290 C, and much faster at 370 C. After the 30 min control period, 10 u. Pitressin was injected as indicated. At 180 C, the gradient of the weight loss slopes did not alter following the injection of Pitressin. At both 29 and 370 C, however, the rate of weight loss gradually increased after injection of pitressin so that 30 min after its administration the weight loss slopes were considerably steeper than during the control period. Values for weight loss were calculated by integrating the area under the weight loss slope every 30 sec and the rate of total body weight loss was then expressed in units of g. m-2. hr-1 using the formula of Du Bois & Du Bois (1916) to calculate surface area. This allowed direct comparison of results from subjects of different sizes.
406 JUDITH A. ALLEN AND I. C. RODDIE Fig. 2 shows the average results for the five subjects in the three environmental temperatures. At 180 C, the mean control rate of weight loss was 18g.m-2.hr-' (s.E..+2.4). For the three 30 min periods following the injection of Pitressin, the rates of weight loss were not significantly different being 18 ( ± 1 2), 15-6 ( ± 1.2) and 18 ( ± 1.2) g. m-2.hr-1. Mouth temperature was not significantly altered throughout, but hand skin temperature fell significantly following injection of Pitressin. Forehead skin temperature (0C)
290 C
180 C
370 C
38
34[ 38
Hand skin temperature
[
4
-OOOOOOOOOCO0
F
26
Mouth temperature (0C)
38
120 Weight loss
_004
36[I -
60
(g. M2. hr-')
____ _______ 0
t
ADH
_______
120
0 t
ADH
120
____
0
120
ADH
Fig. 2. The effect in five subjects of s.c. injection of IO u. Pitressin in environmental temperatures of 18, 29 and 370 C. + represents the mean value for the group with its s.E.
At 290 C (Fig. 2) the mean control rate of weight loss of the five subjects was 30 g.m-2. hr-1 ( ± 3 6). This was not significantly altered during the 30 min immediately following the injection being 33-6 g.m-2. hr-1 ( 3 6), but thereafter became significantly elevated to 46-8 g.m-2. hr-1 ( 4 8) (P < 0-02) and was still elevated at the end of the experiment being 45-6 g.m-2. hr-1 ( ± 3 6). Visible sweating was apparent at the end of the experiment in two subjects. In the control experiments on the same five subjects at 290 C there was no such increase in the rate of weight loss, mean values for the four 30 min periods being 34-8 ( ± 4 8), 34-8 ( ± 4- 8), 31-2 ( ± 7-2) and 28-8 ( + 2-4) g.m-2. hr-1. Mouth and forehead skin temperatures at 290 C were not significantly altered following injection of Pitressin
Min
407 EFFECT OF ADH ON SWEATING (Fig. 2). Hand skin temperature fell from 34-7 (± 0-9) to 33-8 (± 1.2)0 C, but the decrease was not significant. It was, however, absent in the control experiments. At 370 C (Fig. 2) the mean control rate of weight loss of the five subjects was 744 g. m-2. hr-1 (+ 6-0), rising to 88-8 (± 3.6), 96'0 (± 3.6) and 93'6 (± 2.4) g.m-2. hr-1 for the three 30 min periods following injection of Pitressin. Mouth, forehead and hand skin temperatures remained constant Pitressi n 5 U. I.V.
Mouth
38 temperature (0C) 36
Forehead skin 37 temperature (OC) 35 Hand skin 36 temperature (0C) 34
0-'
O
Iwo
96
72 Weight loss
(g. M2 .hr-')
48 24 _ 0
0
20
40
60
80
100
120
140
160
180
Min
Fig. 3. The effects in one subject of i.v. infusion of 5 u. Pitressin in an environmental temperature of 340 C.
throughout. In the control experiments at 370 C there was a gradual small increase in the rate of weight loss over the 2 hr period but this was not a significant rise. The mean values of weight loss during the four 30 min periods were 75-6 (±+ 72), 80-4 (± 4.8), 84-0 (± 48) and 85*2 (± 48) g .m-2. hr-1. In one subject, the i.v. infusion of 5 u. pitressin over 30 min produced a similar prolonged increase in the rate of weight loss from a control level of 528 g. m-2. hr-1 to a peak of 84-0 (Fig. 3). This became apparent within 5 min of starting the infusion and the control rate of weight loss was not regained until 90 min after the Pitressin infusion had ended. Mouth temperature remained constant at 36.60 C throughout. Forehead skin temperature fell from 36-9 to 35.80 C during the infusion and reached
JUDITH A. ALLEN AND I. C. RODDIE control levels 80 min after the end of the infusion. Hand skin temperature fell from 36-0 to 34.80 C during the infusion, and regained control levels 90 min after it had ended. 408
Effects of physiological doses of pitressin At an environmental temperature of 29° C, an isv. bolus of 10 m-u. pitressin produced no significant alteration in the rate of weight loss in five subjects (Fig. 4). The mean rate of weight loss was 21-6 g.m-2. hr-' (± 1.2) during the control period and was 24-0 (+ 1-2), 22-8 (± 1-2) and 24-0 (± 1.2) g.m-2. hr-1 for the three 5 min periods immediately following the injection. Mouth and forehead and hand skin temperatures were similarly unaffected Pitressin 10 M-u. Iv. 96-
E
72
370 C
I,480 -C
24
@OOO*029C 0
20
40
60
80
100
120
Min Fig. 4. The effect of 10 m-u. Pitressin injected i.v. on the rate of weight loss in environmental temperatures of 29 and 370 C. Each graph is the mean of five experiments.
At an environmental temperature of 370 C, 10 m-/t. pitressin again had no significant effect on the rate of weight loss of the five subjects or on the body and skin temperatures (Fig. 4). At 29° C, i.v. infusion of pitressin for 20 min periods at a rate of 45 m-u./ hr had no effect on the rate of weight loss or on the body and skin temperatures of two subjects. DISCUSSION
ADH in a dose of 10 u. s.c. at 180 C had no effect on the rate of cutaneous water loss. At this temperature, no active sweating is taking place as has been demonstrated by atropine blockade (Allen, Jenkinson & Roddie, 1973), so the results suggest that ADH has no effect on insensible perspiration in man and therefore no effect on the permeability of human skin to water.
409 EFFECT OF ADH ON SWEATING The same dose at 29 and 370 C significantly elevated the rate of weight loss by about 50 and 30 % respectively. Previous experiments using atropine blockade of active sweating (Allen et al. 1973) suggest that at 290 C the subjects were sweating slightly and at 370 C were sweating freely. A similar increase of sweating was not seen in the control experiments, therefore the results suggest that when one is close to or above the sweating threshold, pharmacological doses of ADH are capable of increasing the rate of cutaneous water loss. The level of sweating reached at an ambient temperature of 370 C was only moderate and the effects of ADH might well have been different if the heat stress had been greater. However, it was felt that steady-state conditions could be achieved more easily with moderate degrees of heat stress. The effects of ADH were slow in onset, probably due to the s.c. method of administration. They did not appear until approximately 30 min after its injection. They were prolonged, being still marked 90 min after injection. When administered i.v., there was a rapid increase in the rate of weight loss, but the effect was still prolonged, continuing for approximately 90 min after the end of the infusion. The mechanism underlying this increase in cutaneous water loss remains obscure. As previously suggested, if at all active on the sweat glands, ADH might be expected to decrease sweating. There are several possible explanations of why the reverse occurred. In the absence of good data on core and mean skin temperatures before and during the increase, the explanations are at best hypothetical. First, in these doses, ADH causes intense peripheral vasoconstriction as evidenced by the extreme pallor ofthe subjects following its administration, and the decreases in hand skin temperature at 18 and 290 C which presumably reflect decreases in hand blood flow. Thus it seems possible that as a result of cutaneous vasoconstriction, heat loss from the body is cut down with a consequent rise in deep body temperature. This would act as a direct stimulus to the sweat mechanisms leading to increased cutaneous water loss. Mouth temperature was monitored throughout and showed no evidence of a rise. However, it may be an insufficiently sensitive method of detecting increases in deep body temperature in which only a very small increase may be sufficient to activate the sweat apparatus. More information about cutaneous blood flow, mean skin temperature and true core temperature would be needed to argue this case fruitfully. A second possible explanation of the results is that ADH is acting on the myoepithelial cells which are known to be present around the sweat gland tubules (Ellis, 1965). In the mouse mammary gland, these cells have been shown to contract in response to small amounts of vasopressin (Linzel
410 JUDITH A. ALLEN AND I. C. RODDIE 1955). This action would only produce ejection of preformed sweat already present in the tubules and would therefore be expected to have only a transient effect on the rate of sweating and could not account for the prolonged increases seen in the present experiments. A third possibility is that the sweat produced was of emotional or mental origin or was similar to that observed by Randall (1946), in the presence of strong visceral sensations. Mild abdominal sensations of increased peristalsis were often associated with these large doses of ADH, and it is possible that they may have precipitated emotional sweating. If the increased sweating was induced by non-thermoregulatory pathways, a generalized skin vasoconstriction due to ADH may have compensated for the increased heat loss due to sweating and so explain the absence of change in mouth temperature. However, abdominal sensations were not experienced by all the subjects and their time course did not always correlate with that of the sweating. They were often present before increased sweating was apparent, and in the experiment in which pitressin was infused i.v., abdominal discomfort still persisted at the end of the experiment when sweating had returned to control levels. Also, emotional sweating is usually thought to be most marked on the palms of the hands and the soles of the feet (Kuno, 1956), and it was on the chest that visible sweating was observed in some of these experiments. A fourth possibility remains - that ADH has some direct stimulatory action on the cells of the sweat glands themselves. The significance of this increased sweating in response to these large doses of ADH is difficult to assess. This dose is probably in the pharmacological rather than the physiological range of ADH levels considering the normal circulating plasma levels and the amount needed to inhibit a diuresis (Lauson, 1951). However, it was used initially since it is within the dose range recommended by the manufacturers for the treatment of diabetes insipidus. The effects of ADH on the water balance of the body as observed in these experiments are directly opposite to its water conserving effects in the kidney, which is surprising. It is interesting to note that Weiner (1945) observed that in men performing work exposure to heat resulted in a reduction in urine flow even when water was drunk ad libitum. Urine flow was approximately 50 % less than during work in a cool environment. The greatest antidiuresis was often noted at the time of the most profuse sweating and release of ADH from the pituitary has been suggested as an explanation (Ladell, 1948). The effects of more physiological doses were then studied. Initially these were given as an i.v. bolus of 10 m-u. The normal plasma levels of ADH are 3-6 ,au./ml. (Lauson, 1967), which represents a total plasma ADH content of 10-5-21 m-u. Thus the injection was increasing circulating
EFFECT OF ADH ON SWEATING 411 ADH by 50-100 %. This had no effect on cutaneous water loss at 29 or 370 C. However, the half life of circulating ADH in man is very short due to rapid inactivation, so it was also given by i.v. infusion in two subjects at a rate of 45 m-u./hr which is more than sufficient to produce maximal antidiuresis (Lauson, 1951), and over the 20 min infusion period is within the range of plasma ADH levels reached after overnight dehydration (Lauson, 1967). These physiological doses did not induce any significant effect on the rate of weight loss, or on mouth or skin temperatures. The subjects experienced no symptoms and no cutaneous pallor was apparent. In conclusion, large doses of ADH, either 10 u. s.C. or 5 u. infused i.v. are capable of increasing the rate of cutaneous water loss in human subjects who are close to or above the sweating threshold, but do not affect cutaneous water loss in a cold environment. Physiological doses of ADH have no effect on cutaneous water loss in either hot or cold environments, and at normal rates of secretion in the body the hormone probably does not influence human sweat secretion. REFERENCES AWIN, J. A., GRIMLEY, J. F. & RODDIE, I. C. (1971). A body balance to measure sweat rates in man. Bio-med. Engng 6, 468-471. ALLEN, J. A., JENKINSON, D. J. & RODDIE, I. C. (1973). The effect of,/-adrenoceptor blockade on human sweating. Br. J. Pharmac. 47, 487-497. AMATRUDA, T. T. & WELT, L. G. (1953). Secretion of electrolytes in thermal sweat. J. apple. Physiol. 5, 759-772. BARRACLOUGH, M. A. & JoNEs, N. F. (1970). The effect of vasopressin on the reabsorption of sodium, potassium and urea by the renal tubules in man. Clin. Sci. 39, 517-527. Du Bois, D. & Du Bois, E. F. (1916). Clinical calorimetry. A formula to estimate the approximate surface area if height and weight be known. Arch8 intern. Med. 17, 863-871. ELLis, R. A. (1965). Fine structure of the myoepithelium of the eccrine sweat glands of man. J. cell Biol. 27, 551-563. FASCIOLO, J. C., TOTEL, G. L. & JOHNSON, R. E. (1969). Antidiuretic hormone and human eccrine sweating. J. apple. Physiol. 27, 303-307. FUHRMAN, F. & USSING, H. H. (1951). A characteristic response of the isolated frog skin potential to neurohypophysial principles and its relation to the transport of sodium and water. J. cell. comp. Physiol. 38, 109-130. HARNss, J. (1959). Effect of antidiuretic hormone on sweating as proof of its extrarenal action. Am. J. med. Sci. 238, 452-455. HELLER, H. (1945). The effect of neurohypophysial extracts on the water balance of lower vertebrates. Biol. Rev. 20, 147-158. KOEFOED-JOHNSEN, V. & USSING, H. H. (1953). Contributions of diffusion and flow to passage of D20 through living membranes. Acta physiol. scand. 28, 60-76. KuNo, Y. (1956). Human Perspiration. Springfield: Charles C. Thomas. LADELL, W. S. S. (1948). The effect of pituitrin upon performance in moderate heat. S. Afr. J. med. Sci. 13, 145-150.
412
JUDITH A. ALLEN AND 1. C. RODDIE
LADELL, W. S. S. & WmTciiER, H. W. (1960). The effect of pituitrin on sweating. J. Physiol. 154, 44-45 P. LAuSON, H. D. (1951). The problem of estimating the rate of secretion of antidiuretic hormone in man. Am. J. Med. 11, 135-156. LAUSON, H. D. (1 967). Metabolism of antidiuretic hormone. Am. J. Med. 42, 713-744. LINzELL, J. L. (1955). Some observations on the contractile tissue of the mammary glands. J. Physiol. 130, 257-267. MAsEs, P., FALET, R., JOLY, R. & HOUDAS, Y. (1962). Action de l'hormone antidiuretique sur les eliminations d'eau cutanees. C. r. Seanc. Soc. Biol. 156, 73-74. PEARcY, M., ROBINSON, S., MILLER, D. I., THoMAs, J. T. & DE BROTA, J. (1956). Effects of dehydration, salt depletion and pitressin on sweat rate and urine flow. J. apple. Physiol. 8, 621-626. RANDALL, W. C. (1946). Sweat gland activity and changing patterns of sweat secretion on the skin surface. Am. J. Physiol. 147, 391-398. RATNER, A. C. & DOBSON, R. L. (1964). The effect of antidiuretic hormone on sweating. J. invest. Derm. 43, 379-381. ScHLTIN, E. M., SPOONER, G. R., DAY, C., PICKERING, M. & CADE, R. (1971). Extrarenal water loss and antidiuretic hormone. J. apple. Physiol. 31, 569-572. SENAY, L. C. & VAN BEAUMONT, W. (1969). Antidiuretic hormone and evaporative weight loss during heat stress. Pfluigers Arch. ges. Physiol. 312, 82-90. WEINER, J. S. (1945). The diuretic response of men working in hot and humid conditions. J. Physiol. 103, 36P.
