Clinical Physiology (1992) 12,505-526
REVIEW ARTICLE
Gravitational stress and volume regulation P. Norsk Danish Aerospace Medical Centre of Research, Rigshospitalet,Copenhagen, Denmark
(Received 16 September 1991; accepted 31 October 1991)
Summary. During the past 3 decades, groundbased experiments have been performed in order to investigate the effects of increased and decreased gravitational stress, respectively, on the renal response in humans. Experiments that simulate an increase in gravitational load (+G,)to the subjects (centrifugation, passive head-up titlt [HUTI or lower body negative pressure [LBNP] have clearly demonstrated a decrease in renal sodium and water excretion. Simultaneously, increases in plasma levels of arginine vasopressin (AW), renin activity (PRA), aldosterone (PA), norepinephrine (NE) and decreases in ANP have been observed. Additionally, experiments that have utilized immersion of seated subjects to simulate a decreased gravitational stress (-0 Gz) have demonstrated that renal water and sodium excretion increases by 100-400% and that plasma AVP, PRA, PA, and NE concentrations are reduced and ANP levels increased. Alternative experimental models conducted to simulate the effects of weightlessness in humans such as head-down tilt (HDT) and lower body positive pressure (LBPP) have yielded less consistent results than those of water immersion (WI) with respect to renal function. However, compared to a seated control HDT clearly induces an increased rate of renal fluid and sodium excretion. The demonstration that central volume expansion during WI is accompanied by an increase in renal fluid and electrolyte excretion and that central hypovolaemia during centrifugation, HUT, and LBNP is accompanied by the opposite effects indicate that changes in central blood volume is an important determinant of the renal functional changes. Results of experiments in humans during weightlessness in space are inconsistent and difficult to interpret. However, they have indicated that a cephalad redistribution of blood and fluid occurs and that this is accompanied by a decrease in total body fluid. Experimental models that, respectively, increase and decrease the gravitational stress in humans constitute promising tools in the investigation of the physiology and pathophysiology of volume regulation. Correspondence: Peter Norsk, M.D., DAMEC Research, Rigshospitalet7522, STagensvej, DK-2200 Copenhagen, Denmark.
505
506
P. Norsk Introduction
All bodies on the surface of the Earth are constantly subjected to gravitational stress induced by the reactional force of the ground. Since the human body has an elongated shape, the position of the body in relation to the direction of the gravitational force is important in determining the effects on distribution of fluid and blood in the body (Wood et al., 1981; Balldin, 1986; Glaister, 1988). Thus, the gravitational stress exerted on the cardiovascular system is more pronounced when the body is in the upright than in the supine position. Furthermore, changes in body positions have pronounced effects on fluid and electrolyte homeostasis. For example in supine subjects more fluid and electrolytes are excreted than when the subjects are upright. These features are of interest in understanding the physiological and patophysiological mechanisms of volume regulation in humans. With the initiation of manned space flight programmes in the US and the former USSR, more efforts have been devoted to the physiological effects of increased and decreased gravitational stress (Nicogossian & Garshnek, 1989). The observations that weightlessness (the perfect environment of no gravitational stress) induces a substantial loss in body mass and that this loss is most probably due to an augmented renal fluid excretion, led investigators to further explore the mechanisms of the relationship between gravitational stress and volume regulation (Norsk & Epstein, 1991). Procedures, which aim at decreasing (water immersion, head-down tilted bedrest) or increasing (head-up tilt, lower body negative pressure) gravitational stress have been widely used in recent years as tools in cardiovascular and renal research (Epstein, 1978; Blomqvist & Stone, 1983). Furthermore, some of these models have been employed in investigations of deranged fluid and electrolyte metabolism during patophysiological states such as cirrhosis and cardiovascular diseases (Epstein et ul., 1977; Bichet et al., 1983; Epstein, 1988). The present review aims at describing the renal and hormonal alterationsin humans during procedures that increase or decrease gravitational stress or stimulate these effects and their impact in altering volume homeostasis. Gravitationalstrese and G-forces: concepts and definitions
On the surface of the Earth, gravitational stress is usually a reactional, mechanical force of the ground exerted upon a body due to gravitation. In aviation and space medicine (Wood et ul., 1981; Glaister 1988; Guy & Prisk, 1989), the term G is a measure of gravitational stress or stress exerted on a body by mechanical forces during changes in velocity. It is defined as the ratio of an applied acceleration, a, to the gravitational constant, g (9.81 m:s2). Thus, G =a:g. The human body is affected by a gravitational stress of 1G when passively standing, sitting, or lying on the surface of the Earth. During acceleration in a vehicle or ritation in a centrifuge, the stress of mechanical forces acting on the body may attain several Gs. In this way, an increase in gravitational stress can be simulated.
Gravitational stress
507
n
Fig. 1. The standard aeromedical terminology regarding direction of G-forces on a human body. A headward acceleration (e.g. created by the force of the ground due to gravity acting on an upright body) induces by convention a G-force (stress) in the opposite direction which is designated +G.. A back to chest acceleration (e.g. created by the force of the ground acting on a supine subject lying on the back) induces a G-force defined as +G.. A left to right shoulder acceleration (e.g. created by the force of the ground on a supine subject lying on his left lateral side) induces a G-force of +G,. Minus indicates a direction of G-force in the opposite direction. The term 0 G is used for weightlessness.
The direction of a G-force (stress) is defined according to the anatomy of the human body as indicated in Fig. 1 (Glaister, 1988; Guy & Prisk, 1989). By convention, it is opposite to the direction of the reactional, mechanical force which induces the stress. Therefore, the direction of the G-vector is the same as the direction of displacement of body compartments. The following definitions apply: Z indicates a foot to head (-G,) or head to foot (+G,) direction; x a back to chest (-G,) or chest to back (+G,) direction; and y a left to right shoulder (-Gy) or right to left shoulder (+Gy) direction. Thus, a seated or standing human on the surface of the Earth is subjected to + 1 G, and a supine human lying on the back is subjected to +1 G,. Weightlessness is designated as 0 G. Increased gravitational stress CENTRIFUGATION
An increase in gravitational stress can be simulated in humans by centrifugation. When an upright seated human subject is centrifuged, the Gz-stress is increased. The cardiovascular system is primarily affected since the hydrostatic pressure gradieas are augmented (Wood et al., 1981; Balldin, 1986). Thus, venous and arterial pressures decrease in the cephalad parts of the body and increase in the caudad portions. Blood is subsequently redistributed resulting in a decreased intrathoracic blood volume (Lindberg et al., 1960; Blomqvist & Stone, 1983; Guy & Prisk, 1989). The initial experiments investigating the renal response to a simulated increase in gravitational stress was reported by Stauffer & Errebo-Knudsen in 1950. Well
508
P. Norsk
hydrated human subjects underwent +3 Gz centrifugation for 1 min in the seated position. During the 15-min period following centrifugation, urine flow rate decreased by 50% whereas no decreases occurred during a corresponding seated control. Piemme and co-workers (Piemme et al., 1966) extended these observations by subjecting human subjects to +2 and +3 Gz for 90 min. Compared to +1 Gz,urine output decreased significantly from 6.2 to 1.9 ml min-' and solute-free water clearance from 3.0 to -0.5 ml min-' at +3 Gz.Since occurrence of nausea might induce an antidiuresis, care was taken to ensure that the subjects did not experience vestibular symptoms. Since Rogge and co-workers (Rogge et al., 1%7), have demonstrated that peripheral venous plasma concentration of antidiuretic hormone ( = ADH, when measured by bioassay and arginine vasopressin [AVP]when measured by radioimmunoassay) increases during 30 min of +2 Gz,it is likely that stimulation of ADH may have constituted one of the mechanisms mediating this observed decrease in renal water excretion. Epstein and co-workers (Epstein ef al., 1974) have investigated the effects of exposure to +2 Gz for 20-30 min on renal sodium handling. The subjects were investigated in either the sodium depleted (intake 10 mEq day-') or sodium replete (intake 200 mEq day-') state. Both groups of subjects demonstrated a significant decrease in renal sodium excretion during centrifugation from 7.8 2 2.6 to 2.5 & 0.6 pEq min-' in the sodium depleted subjects and from 121 2 13 to 602 13 pEq min-' in the sodium replete subjects. Since creatinine clearance also decreased during centrifugation indicating that glomerular filtration rate (GFR)fell, it was concluded that the decrease in renal sodium excretion was primarily attributable to decrements in the filtered load of sodium. Changes in hormonal determinants of renal sodium handling might also have contributed to the anti-natriurssissince these authors also documented a consistent albeit heterogenous increase in plasma renin activity (PRA) immediately following centrifugation. Rogge and co-workers (Rogge ef al., 1967) also reported that peripheral venous PRA levels increased during exposure to +2 Gz. Chimoskey (1970) investigated renal haemodynamics in dogs during accelerations up to +6 Gz for 20-30 s durations. The renal arterial-venous pressure gradient was unchanged up to +3 Gz and thereafter declined to 50% of the 1Gzvalue. Renal flow velocity decreased in proportion to +Gz.Therefore, intrarenal vasoconstriction might constitute a mechanism of decrease in renal fluid and sodium excretion during an increase in +Gz. Thus, during well controlled experiments in humans a consistent tinding is an antidiuresis and anti-natriuresis during increased Gz levels. Simultaneously, AVP and PRA increase. The mechanisms of the renal responses may entail interactions of several complex mechanisms involving a decrease in central blood volume and renal perfusion. However, this suggestion is still speculative since no studies have been conducted that involve simultaneous measurements of cardiovascular and renal variables during centrifugation in humans.
+
Gravitational stress
509
HEAD-UP TILT
The renal response to passive posture changes of the human body, e.g. changing the G-stress from +1 Gx to +1 Gz, have only been investigated to a limited extent. In order to induce these changes, passive head-up tilt (HUT) on a tilt table has been the model of choice. Brun and co-workers (Brun et al., 1946) were one of the first groups to investigate the renal response to HUT. These investigators observed that HUT of 60" induced a marked decrease in urinary output that was further decreased when the subject experienced circulatory collapse. It was concluded that these changes were attributable, primarily, to a release of ADH because a transfusion of 300-500 C.C. of blood from the orthostatic collapsed subjects to non-collapsed water loaded, supine subjects produced a comparable oliguria in the recipients. Only a few studies have directly investigated the renal responses to HUT in humans. Molzahn and co-workers (Molzahn et al., 1972) and Hesse and co-workers (Hesse et al., 1978) observed that urinary sodium excretion decreased by 45 and 14%, respectively. This decrease was not caused by changes in glomerular filtration fractions. Thus, an interaction of endocrine, circulatory, and intrarenal mechanisms probably participate in concert in the HUT-induced anti-diuresis and anti-natriuresis. In contrast to the limited data on renal responses to HUT in humans, endocrine variables of importance for renal function have been measured extensively during HUT procedures. The suggestion of Brun and co-workers (Brun et al., 1946) that increases in ADH mediate the HUT-induced anti-diuresishas been confirmed by direct measurements of ADH or A W (Davies et al., 1976; Sander-Jensen et al., 1986). The increase in plasma A W seems to follow a biphasic pattern: a two-fold increase during the initial 30 min of an 85" HUT followed by a 4-5 fold increase during the subsequent 30 min (Davies et al., 1976). It has been suggested that the decrease in central blood volume per se through inhibition of low-pressure baroreceptors account for the orthostatic stimulation of A W release (Share, 1988). However, results of studies performed by Norsk (1989), Bie and co-workers (Bie et al., 1986), Hammer and co-workers (Hammer et al., 1988), and recent results from our laboratory (not yet published) indicate that inhibition of arterial baroreflexes is a more important stimulus for AVP release during HUT than deloading of low-pressure baroreceptors. Hesse and co-workers (Hesse et d.,1978) investigated the effect of 40 min of 60" HUT in humans on sympathetic nervous activity (SNA) and PRA.During HUT, both variables increased. However, combining HUT with inflation of a G-suit to counteract the decrease in central blood volume abolished the increases in SNA and PRA.Thus, it is likely that inhibition of low-pressure baroreceptors in the intrathoracic region account for the increase in these variables during orthostatic manoeuvers. Schutten and co-workers (Schutten et al., 1987) investigated the effects of various The authors concluded that body positions on plasma atrial natriuretic peptide (A"). the upright body position and HUT of 30" induced a lower level of A N P than when the supine position was attained.
510
P. Norsk
In summary, HUT induces a decrease in renal excretion of fluid and sodium. Simultaneously, PRA, SNA, and plasma A W increase and ANP decrease which might all participate in the renal responses. Experimental evidence indicates that inhibition of low-pressure baroreceptors might account for the increases in PRA and SNA during HUT whereas the increased release of A W might primarily be due to deloading of arterial baroreceptors (personal observations). LOWER BODY NEGATIVE PRESSURE
By applying negative pressure around the caudad portions of the body from the waist and down, blood can be redistributed from the intrathoracic region to the abdomen and lower extremities (Wolthuis ef al., 1974; Norsk, 1989). Low levels of lower body negative pressure (LBNP) between 0 and -20 mmHg induce decreases in central venous pressure without affecting arterial variables (Mark & Mancia, 1983). Thus, in contrast to HUT, LBNP can simulate an increase in gravitational stress that is more selectively aimed at inhibiting low-pressure baroreceptors. In a recent study, Miller and co-workers (Miller ef al., 1991) investigated the effects of selective cardiopulmonary baroreceptor deactivation in humans during prolonged LBNP of -15 mmHg for 90 min. Urine was collected during voiding prior to and immediately after LBNP. There was a marked decrease in renal water and sodium excretion during LBNP compared to the supine position without application of this procedure. Simultaneously, central venous pressure decreased by 4.6 mmHg without affecting arterial variables and heart rate. Release of ANP decreased by 6-10 pg ml-' and SNA increased indicated by an increased level of plasma norepinephrine (NE)by 0.6 nmol 1-'. PRA, plasma aldosterone (PA) and plasma A W remained unchanged. It was concluded that selective stimulation of low-pressure baroreceptors induced a decrease in renal excretion of sodium and fluid and that these renal responses might be initiated by a decrease in ANP and an increase in SNA and plasma NE. Results of several studies (Goldsmith ef af., 1982; Leimbach ef al., 1984; Agewall ef al., 1990) are in compliance with the results of Miller and co-workers (Miller ef al., 1991) in regard to the endocrine responses of LBNP. During increased levels of LBNP from -20 to -80 mmHg, exceeding those that Miller and co-workers used, arterial pulse pressure decreases which is subsequentlyfollowed by a decrease in mean arterial pressure. During these circumstances with combined unloading of low- and highpressure baroreceptors, plasma A W and PRA increase (Mark ef af., 1978; Leimbach ef af., 1984; Nabel ef al., 1987; Sander-Jensen et al., 1988). Thus, the renal responses to LBNP mimic those of HUT. However, in contrast to the HUT model the relative importance of low- and high-pressure baroreceptors, respectively, on endocrine and renal responses caq be evaluated by using the LBNP model. Recent results from our laboratory (not yet published) and those of other groups (Goldsmith ef al., 1982; Miller ef af., 1991) indicate that selective inhibition of low-pressure baroreceptors induces a decrease in renal fluid and sodium excretion
Gravitational stress
511
through modulations of SNA, NE, and ANP release. Furthermore, deloading of arterial baroreceptors by further increasing the level of LBNP augments the renal response to a simulated gravitational stress by also inducing increases in PRA and AVP release. SUMMARY OF RENAL RESPONSES TO INCREASED GRAVITATIONAL STRESS
Increasing the G,-stress in humans induces a decrease in renal fluid and sodium excretion. The mechanisms of the renal response to a gravitational stress may entail increases in the release of AVP, PRA, PA, SNA, and a decrease in ANP release. By utilizing the HUT and LBNP models to induce or simulate an increase in G,-stress, the relative importance of low- and high-pressure baroreceptors, respectively, on endocrine responses can be investigated. Decreased gravitational stress WATER IMMERSION
In contrast to the models that simulate or induce an increase in +&-levels, the cardiovascular, endocrine, and renal adaptaton to head-out water immersion (WI) in humans have been extensively investigated. For a detailed description of the model, the reader is referred to several comprehensive reviews (Epstein, 1978; Norsk & Epstein, 1988; Epstein et al., 1989b; Krasney et al., 1989). It has been suggested that WI simulates aspects of weightlessness since intravascular hydrostatic gradients are counteracted by the water pressure exerted upon the body (Gauer et al. , 1970; Norsk & Epstein, 1988). The data on volume regulation in this model are usually based upon 3-8 h of investigation and upon comparing the data during WI with those from a similar seated control study. In brief, by immersing seated humans in thennoneutral water (34.5"C) to the level of the neck, blood is redistributed from the caudad portions of the body into the intrathoracic circulation. This induces increases in central blood volume of -700 ml (Arboreliuset al., 1972),heart volume of 180-250 ml (Lange et al. , 1974; Risch et al. , 1978), central venous pressure (CVP) of 7-18 mmHg (Echt et al. , 1W4; Gauer & Henry, 1976;Norsk et al. ,1985,1986), transmural CVP of 8-13 mmHg (Echt et al., 1974; Gauer & Henry, 1976), cardiac output (CO) of 17-36% (Begin et al., 1976; Norsk et al., l W ) , stroke volume (SV)of 3567% (Gauer & Henry, 1976; Norsk et al. , 1990), and arterial pulse pressure (PP) of 25% (Norsk et al., 1986). The increases in CVP, SV, and PP persist throughout prolonged WI of several hours duration (3-6 h) (Norsk et al. , 1985, 1986). Concomitantly, immersion induces a natriuresis and diuresis (Epstein, 1978). The magnitude and temporal profile of the diuresis of WI is dependent upon the hydration state of the subject (Behn et al. ,1%9). Thus, in fluid-restricted humans, the diuresis is primarily characterized by an increase in solute excretion and attain peak values after
5 12
P. Norsk
4-5 h of WI. In hydrated subjects, the diuresis is characterized by an increase in both
solute and free-water clearance and peak values are attained after 1-3 h of WI (Epstein, 1978). The magnitude and temporal profile of the natriuresis of WI is dependent upon the sodium balance of the subjects. In sodium replete subjects, the natriuresis peaks earlier than in depleted states and the absolute increase is greater. Furthermore, Epstein and co-workers (Epstein et al., 1973) have demonstrated that the increase in urinary flow rate and sodium excretion during WI exceed those of a supine control. The renal response to WI is mediated by several hormonal effectors. Suppression of PRA, PA, plasma AVP, plasma NE, and increases in plasma ANP act in concert to mediate the diuresis and natriuresis of immersion (Epstein, 1978; Norsk & Epstein, 1988; Epstein et al., 1989b). Furthermore, changes in intrarenal prostaglandine and dopamine activity and in intrarenal blood flow distribution might contribute to the renal responses (Epstein et al., 1979; Coruzzi et al., 1986). The quantitative importance of all these hormonal mediators is not known at present. In a more prolonged WI experiment conducted in our laboratory (Stadeager et al., personal communication), the effects of 12 h of WI-induced central hypervolaemia on renal fluid and sodium excretion were investigated. Renal sodium and water excretions were attenuated following the usually observed increase. However, renal sodium excretion was still elevated after 12 h of WI whereas renal water excretion attained control levels already after 7 h. Furthermore, the endocrine responses indicated that the initial increase in renal sodium excretion might be initiated by simultaneous suppressions of PRA, PA, and plasma NE and an increase in ANP release. However, the attentuation of the renal sodium excretion could have been induced by the observed attentuation of ANP release over the 12 h. These results illustrate the complexity of the mechanisms of the renal responses to WI. Usually, the increase in central blood volume during WI is considered to be the initiating factor of the renal response through stimulation of low-pressure baroreceptors (Gauer & Henry, 1976;Epstein, 1978). However, since results of several investigations from our laboratory have demonstrated that arterial pulse pressure increases (Norsk et al., 1985, 1986), stimulation of arterial baroreceptors must also have occurred. Furthermore, only little attention has been put on observed increases in blood and plasma volume which occur due to filtration of fluid from the interstitium to the intravascular space (Greenleaf, 1984). Theoretically, an increase in total blood volume and a decrease in plasma colloid osmotic pressure might also participate in the renal responses to WI. Of considerable interest is the recent discovery by Schulz-Knappe and co-workers (Schultz-Knappe et al., 1988) of an ANP related peptide in urine. The 32-amino-acid peptide, known as urodilatin, is composed of the entire sequence of a-human ANP plus a four-amino-acid N-terminal extension. Preliminary results obtained by Norsk and co-workers and Drummer and co-workers (personal communications) indicate that 12 h of WI induces an increased rate of urinary urodilatin excretion which
Gravitational stress
513
basically follows a temporal pattern similar to the increases in urine flow and sodium excretion. However, the importance of these discoveries has yet to emerge. Russian investigators have carried out prolonged immersions of several days duration. Gogolev and co-workers (Gogolev et al., 1980) carried out a study utilizing the so called 'dry' immersion technique for 7 days. The subjects were surrounded by a thin plastic covering so that the skin was protected from the water and they were recumbent during immersion. During control, the subjects were ambulatory. These authors reported that a negative water balance (difference between intake and renal output) of -700 ml was attained during the initial 24 h of immersion compared to a positive balance of 400 ml during control. During the subsequent days, the water balance approximated zero. The results of this study are difficult to evaluate since no specific details were given of the experimental design and since the subjects were recumbent during immersion. In summary, WI in humans clearly induces an increased rate of renal fluid and sodium excretion. The stimuli for the renal responses probably constitute stimulations of low- and high-pressure baroreceptors since central blood volume, CVP, and PP increase. Furthermore, an increase in plasma volume due to filtration of fluid from the interstitial to the intravascular space also might constitute a stimulus for the renal responses. Several endocrine elements such as AVP, PRA, PA, NE, and ANP probably participate as effectors in the natriuresis and diuresis of WI.The discovery of urodilatin, which is an ANP-related peptide produced by the kidney, has recently attracted attention as a possible candidate for the diuresis and natriuresis of central volume expansion, including WI (Goetz, 1991). HEAD DOWN TILTED BEDREST
By tilting human subjects head down 3-10", the gravitational stress exerted upon the human body can be minimized. This model was initially introduced by Russian investigators (Kakurin et al., 1976) as a more effective way than horizontal bedrest of simulating the effects of weightlessness on the cardiovascular system. This model has since been widely accepted by investigators in NASA and within the European space community. It should be noted that results of head-down tilt (HDT) are usually compared to results obtained with the subjects in the supine position in the hours prior to and following tilting (Nixon et al., 1979;Gaffney et al., 1985)or to measurements performed on the subjects while ambulatory (Noskov, 1982;Hargens et al., 1983). Thus, except for a few studies (Cathcard & Williams, 1955;Gharib et al., 1988) an appropriate time control with the subjects seated or supine have not been conducted. Compared to the supine position in the hours prior to tilting the immediate cardiovascular effects of HDT of 5-6" can be summarized as follows: CVP increases by 2-4 mmHg, CO by 15%, SV by -lo%, and left ventricular end diastolic diameter (LVED) by -20% (Nixon et al., 1979;Gaffney et al., 1985). During prolonged HDT experiments of 20-24 h duration, these changes usually occur within the first 0-5-2h.
5 14
P. Norsk
Thereafter, the increases in CVP, SV and CO gradually subside to attain pre-study levels after 20-24 h. Upon resuming the supine position after 20 h of HDT, CVP is decreased by 14%, SV by 19%, and CO by 17%, compared to supine, pre-tilt levels. Simultaneously, body weight decreases by 1.3 kg over 24 h of HDT. Since HDT experiments primarily have been performed in order to examine the effects upon the cardiovascular system, the renal response to HDT has not been well characterized. Cathcart & Williams (1955) compared the renal responses during - 12" HDT with that of supine rest for a comparable length of time. These authors observed no effect on renal fluid and electrolyte excretion of HDT. Other investigators have concluded that a negative sodium and fluid balance develops during HDT (Volicer et al., 1976) and that renal sodium and fluid excretions are augmented (Nixon et al., 1979; Hargens et al., 1983; Gaffney et al., 1985). However, since these investigations did not comprise time-control experiments and controlled intakes of food and fluid, the results of these studies are difficult to interpret. A prolonged HDT study of -3" for 12 h has been conducted in our laboratory (Norsk et al., data not published). The results concerning changes in CVP and renal sodium excretion are shown in Fig. 2, where they are also compared to a seated time control experiment and to seated WI for the same length of time in the same subjects. It is clear that changing the body position from seated (+1G,) to HDT (slightly negative GJ induces an augmented excretion rate of sodium. Concomitantly, CVP is increased by approximately 8 mmHg. Furthermore, the data reveals that seated WI induces an even larger increase in renal sodium excretion accompanied by a 12 mmHg increase in CVP. In summary, the HDT model has been primarily used to simulate the effects of weightlessness on the cardiovascular system. HDT induces increases in central blood volume as evidenced by increases in CVP,CO, SV, and LVED compared to both the supine and seated positions, respectively. None of the studies to date have convincingly demonstrated that these increases in central cardiovascular variables are accompanied by a diuresis and natriuresis when compared to the supine position (Cathcart & Williams, 1955; Nixon et al., 1979; Gaffney et al., 1985). Furthermore, compared to the supine position, HDT does not induce a statistically significant suppression of plasma levels of AVP, PRA, A, and NE (Nixon et al., 1979; Gaffney et al., 1985). However, compared to a seated control (+1 Gz) HDT clearly induces increases in renal fluid and sodium excretion and suppresses PRA, PA, and plasma NE (Gharib et al., 1988; personal observations in Fig. 2). Gogolev and co-workers (Gogolev et al., 1980)carried out an HDT study for 7 days. During control, the subjects were ambulatory. These authors reported that the water balance was zero during the initial 24 h of HDT cornpared to a positive balance of 400 ml during control. Over the subsequent days, the water balance increased to control levels. The results of this study are difficult to evaluate since no specific details were given of the experimental design.
515
Gravitational stress HDT or WI
"T
m w1
-5 1
Control
0
2
4
6
8
1
0
1
2
Time (h)
Flg. 2. Central venous pressure (CVP)and renal sodium excretion (UN.V) before, during and after 12 h of water immersion (WI)to the neck (triangle), head-down tilted bedrest (HDT) of -3" (filled circle), and a seated control (blank circle). Values are means 2 SE of 6 male subjects. *indicates significant changes (P
REVIEW ARTICLE
Gravitational stress and volume regulation P. Norsk Danish Aerospace Medical Centre of Research, Rigshospitalet,Copenhagen, Denmark
(Received 16 September 1991; accepted 31 October 1991)
Summary. During the past 3 decades, groundbased experiments have been performed in order to investigate the effects of increased and decreased gravitational stress, respectively, on the renal response in humans. Experiments that simulate an increase in gravitational load (+G,)to the subjects (centrifugation, passive head-up titlt [HUTI or lower body negative pressure [LBNP] have clearly demonstrated a decrease in renal sodium and water excretion. Simultaneously, increases in plasma levels of arginine vasopressin (AW), renin activity (PRA), aldosterone (PA), norepinephrine (NE) and decreases in ANP have been observed. Additionally, experiments that have utilized immersion of seated subjects to simulate a decreased gravitational stress (-0 Gz) have demonstrated that renal water and sodium excretion increases by 100-400% and that plasma AVP, PRA, PA, and NE concentrations are reduced and ANP levels increased. Alternative experimental models conducted to simulate the effects of weightlessness in humans such as head-down tilt (HDT) and lower body positive pressure (LBPP) have yielded less consistent results than those of water immersion (WI) with respect to renal function. However, compared to a seated control HDT clearly induces an increased rate of renal fluid and sodium excretion. The demonstration that central volume expansion during WI is accompanied by an increase in renal fluid and electrolyte excretion and that central hypovolaemia during centrifugation, HUT, and LBNP is accompanied by the opposite effects indicate that changes in central blood volume is an important determinant of the renal functional changes. Results of experiments in humans during weightlessness in space are inconsistent and difficult to interpret. However, they have indicated that a cephalad redistribution of blood and fluid occurs and that this is accompanied by a decrease in total body fluid. Experimental models that, respectively, increase and decrease the gravitational stress in humans constitute promising tools in the investigation of the physiology and pathophysiology of volume regulation. Correspondence: Peter Norsk, M.D., DAMEC Research, Rigshospitalet7522, STagensvej, DK-2200 Copenhagen, Denmark.
505
506
P. Norsk Introduction
All bodies on the surface of the Earth are constantly subjected to gravitational stress induced by the reactional force of the ground. Since the human body has an elongated shape, the position of the body in relation to the direction of the gravitational force is important in determining the effects on distribution of fluid and blood in the body (Wood et al., 1981; Balldin, 1986; Glaister, 1988). Thus, the gravitational stress exerted on the cardiovascular system is more pronounced when the body is in the upright than in the supine position. Furthermore, changes in body positions have pronounced effects on fluid and electrolyte homeostasis. For example in supine subjects more fluid and electrolytes are excreted than when the subjects are upright. These features are of interest in understanding the physiological and patophysiological mechanisms of volume regulation in humans. With the initiation of manned space flight programmes in the US and the former USSR, more efforts have been devoted to the physiological effects of increased and decreased gravitational stress (Nicogossian & Garshnek, 1989). The observations that weightlessness (the perfect environment of no gravitational stress) induces a substantial loss in body mass and that this loss is most probably due to an augmented renal fluid excretion, led investigators to further explore the mechanisms of the relationship between gravitational stress and volume regulation (Norsk & Epstein, 1991). Procedures, which aim at decreasing (water immersion, head-down tilted bedrest) or increasing (head-up tilt, lower body negative pressure) gravitational stress have been widely used in recent years as tools in cardiovascular and renal research (Epstein, 1978; Blomqvist & Stone, 1983). Furthermore, some of these models have been employed in investigations of deranged fluid and electrolyte metabolism during patophysiological states such as cirrhosis and cardiovascular diseases (Epstein et ul., 1977; Bichet et al., 1983; Epstein, 1988). The present review aims at describing the renal and hormonal alterationsin humans during procedures that increase or decrease gravitational stress or stimulate these effects and their impact in altering volume homeostasis. Gravitationalstrese and G-forces: concepts and definitions
On the surface of the Earth, gravitational stress is usually a reactional, mechanical force of the ground exerted upon a body due to gravitation. In aviation and space medicine (Wood et ul., 1981; Glaister 1988; Guy & Prisk, 1989), the term G is a measure of gravitational stress or stress exerted on a body by mechanical forces during changes in velocity. It is defined as the ratio of an applied acceleration, a, to the gravitational constant, g (9.81 m:s2). Thus, G =a:g. The human body is affected by a gravitational stress of 1G when passively standing, sitting, or lying on the surface of the Earth. During acceleration in a vehicle or ritation in a centrifuge, the stress of mechanical forces acting on the body may attain several Gs. In this way, an increase in gravitational stress can be simulated.
Gravitational stress
507
n
Fig. 1. The standard aeromedical terminology regarding direction of G-forces on a human body. A headward acceleration (e.g. created by the force of the ground due to gravity acting on an upright body) induces by convention a G-force (stress) in the opposite direction which is designated +G.. A back to chest acceleration (e.g. created by the force of the ground acting on a supine subject lying on the back) induces a G-force defined as +G.. A left to right shoulder acceleration (e.g. created by the force of the ground on a supine subject lying on his left lateral side) induces a G-force of +G,. Minus indicates a direction of G-force in the opposite direction. The term 0 G is used for weightlessness.
The direction of a G-force (stress) is defined according to the anatomy of the human body as indicated in Fig. 1 (Glaister, 1988; Guy & Prisk, 1989). By convention, it is opposite to the direction of the reactional, mechanical force which induces the stress. Therefore, the direction of the G-vector is the same as the direction of displacement of body compartments. The following definitions apply: Z indicates a foot to head (-G,) or head to foot (+G,) direction; x a back to chest (-G,) or chest to back (+G,) direction; and y a left to right shoulder (-Gy) or right to left shoulder (+Gy) direction. Thus, a seated or standing human on the surface of the Earth is subjected to + 1 G, and a supine human lying on the back is subjected to +1 G,. Weightlessness is designated as 0 G. Increased gravitational stress CENTRIFUGATION
An increase in gravitational stress can be simulated in humans by centrifugation. When an upright seated human subject is centrifuged, the Gz-stress is increased. The cardiovascular system is primarily affected since the hydrostatic pressure gradieas are augmented (Wood et al., 1981; Balldin, 1986). Thus, venous and arterial pressures decrease in the cephalad parts of the body and increase in the caudad portions. Blood is subsequently redistributed resulting in a decreased intrathoracic blood volume (Lindberg et al., 1960; Blomqvist & Stone, 1983; Guy & Prisk, 1989). The initial experiments investigating the renal response to a simulated increase in gravitational stress was reported by Stauffer & Errebo-Knudsen in 1950. Well
508
P. Norsk
hydrated human subjects underwent +3 Gz centrifugation for 1 min in the seated position. During the 15-min period following centrifugation, urine flow rate decreased by 50% whereas no decreases occurred during a corresponding seated control. Piemme and co-workers (Piemme et al., 1966) extended these observations by subjecting human subjects to +2 and +3 Gz for 90 min. Compared to +1 Gz,urine output decreased significantly from 6.2 to 1.9 ml min-' and solute-free water clearance from 3.0 to -0.5 ml min-' at +3 Gz.Since occurrence of nausea might induce an antidiuresis, care was taken to ensure that the subjects did not experience vestibular symptoms. Since Rogge and co-workers (Rogge et al., 1%7), have demonstrated that peripheral venous plasma concentration of antidiuretic hormone ( = ADH, when measured by bioassay and arginine vasopressin [AVP]when measured by radioimmunoassay) increases during 30 min of +2 Gz,it is likely that stimulation of ADH may have constituted one of the mechanisms mediating this observed decrease in renal water excretion. Epstein and co-workers (Epstein ef al., 1974) have investigated the effects of exposure to +2 Gz for 20-30 min on renal sodium handling. The subjects were investigated in either the sodium depleted (intake 10 mEq day-') or sodium replete (intake 200 mEq day-') state. Both groups of subjects demonstrated a significant decrease in renal sodium excretion during centrifugation from 7.8 2 2.6 to 2.5 & 0.6 pEq min-' in the sodium depleted subjects and from 121 2 13 to 602 13 pEq min-' in the sodium replete subjects. Since creatinine clearance also decreased during centrifugation indicating that glomerular filtration rate (GFR)fell, it was concluded that the decrease in renal sodium excretion was primarily attributable to decrements in the filtered load of sodium. Changes in hormonal determinants of renal sodium handling might also have contributed to the anti-natriurssissince these authors also documented a consistent albeit heterogenous increase in plasma renin activity (PRA) immediately following centrifugation. Rogge and co-workers (Rogge ef al., 1967) also reported that peripheral venous PRA levels increased during exposure to +2 Gz. Chimoskey (1970) investigated renal haemodynamics in dogs during accelerations up to +6 Gz for 20-30 s durations. The renal arterial-venous pressure gradient was unchanged up to +3 Gz and thereafter declined to 50% of the 1Gzvalue. Renal flow velocity decreased in proportion to +Gz.Therefore, intrarenal vasoconstriction might constitute a mechanism of decrease in renal fluid and sodium excretion during an increase in +Gz. Thus, during well controlled experiments in humans a consistent tinding is an antidiuresis and anti-natriuresis during increased Gz levels. Simultaneously, AVP and PRA increase. The mechanisms of the renal responses may entail interactions of several complex mechanisms involving a decrease in central blood volume and renal perfusion. However, this suggestion is still speculative since no studies have been conducted that involve simultaneous measurements of cardiovascular and renal variables during centrifugation in humans.
+
Gravitational stress
509
HEAD-UP TILT
The renal response to passive posture changes of the human body, e.g. changing the G-stress from +1 Gx to +1 Gz, have only been investigated to a limited extent. In order to induce these changes, passive head-up tilt (HUT) on a tilt table has been the model of choice. Brun and co-workers (Brun et al., 1946) were one of the first groups to investigate the renal response to HUT. These investigators observed that HUT of 60" induced a marked decrease in urinary output that was further decreased when the subject experienced circulatory collapse. It was concluded that these changes were attributable, primarily, to a release of ADH because a transfusion of 300-500 C.C. of blood from the orthostatic collapsed subjects to non-collapsed water loaded, supine subjects produced a comparable oliguria in the recipients. Only a few studies have directly investigated the renal responses to HUT in humans. Molzahn and co-workers (Molzahn et al., 1972) and Hesse and co-workers (Hesse et al., 1978) observed that urinary sodium excretion decreased by 45 and 14%, respectively. This decrease was not caused by changes in glomerular filtration fractions. Thus, an interaction of endocrine, circulatory, and intrarenal mechanisms probably participate in concert in the HUT-induced anti-diuresis and anti-natriuresis. In contrast to the limited data on renal responses to HUT in humans, endocrine variables of importance for renal function have been measured extensively during HUT procedures. The suggestion of Brun and co-workers (Brun et al., 1946) that increases in ADH mediate the HUT-induced anti-diuresishas been confirmed by direct measurements of ADH or A W (Davies et al., 1976; Sander-Jensen et al., 1986). The increase in plasma A W seems to follow a biphasic pattern: a two-fold increase during the initial 30 min of an 85" HUT followed by a 4-5 fold increase during the subsequent 30 min (Davies et al., 1976). It has been suggested that the decrease in central blood volume per se through inhibition of low-pressure baroreceptors account for the orthostatic stimulation of A W release (Share, 1988). However, results of studies performed by Norsk (1989), Bie and co-workers (Bie et al., 1986), Hammer and co-workers (Hammer et al., 1988), and recent results from our laboratory (not yet published) indicate that inhibition of arterial baroreflexes is a more important stimulus for AVP release during HUT than deloading of low-pressure baroreceptors. Hesse and co-workers (Hesse et d.,1978) investigated the effect of 40 min of 60" HUT in humans on sympathetic nervous activity (SNA) and PRA.During HUT, both variables increased. However, combining HUT with inflation of a G-suit to counteract the decrease in central blood volume abolished the increases in SNA and PRA.Thus, it is likely that inhibition of low-pressure baroreceptors in the intrathoracic region account for the increase in these variables during orthostatic manoeuvers. Schutten and co-workers (Schutten et al., 1987) investigated the effects of various The authors concluded that body positions on plasma atrial natriuretic peptide (A"). the upright body position and HUT of 30" induced a lower level of A N P than when the supine position was attained.
510
P. Norsk
In summary, HUT induces a decrease in renal excretion of fluid and sodium. Simultaneously, PRA, SNA, and plasma A W increase and ANP decrease which might all participate in the renal responses. Experimental evidence indicates that inhibition of low-pressure baroreceptors might account for the increases in PRA and SNA during HUT whereas the increased release of A W might primarily be due to deloading of arterial baroreceptors (personal observations). LOWER BODY NEGATIVE PRESSURE
By applying negative pressure around the caudad portions of the body from the waist and down, blood can be redistributed from the intrathoracic region to the abdomen and lower extremities (Wolthuis ef al., 1974; Norsk, 1989). Low levels of lower body negative pressure (LBNP) between 0 and -20 mmHg induce decreases in central venous pressure without affecting arterial variables (Mark & Mancia, 1983). Thus, in contrast to HUT, LBNP can simulate an increase in gravitational stress that is more selectively aimed at inhibiting low-pressure baroreceptors. In a recent study, Miller and co-workers (Miller ef al., 1991) investigated the effects of selective cardiopulmonary baroreceptor deactivation in humans during prolonged LBNP of -15 mmHg for 90 min. Urine was collected during voiding prior to and immediately after LBNP. There was a marked decrease in renal water and sodium excretion during LBNP compared to the supine position without application of this procedure. Simultaneously, central venous pressure decreased by 4.6 mmHg without affecting arterial variables and heart rate. Release of ANP decreased by 6-10 pg ml-' and SNA increased indicated by an increased level of plasma norepinephrine (NE)by 0.6 nmol 1-'. PRA, plasma aldosterone (PA) and plasma A W remained unchanged. It was concluded that selective stimulation of low-pressure baroreceptors induced a decrease in renal excretion of sodium and fluid and that these renal responses might be initiated by a decrease in ANP and an increase in SNA and plasma NE. Results of several studies (Goldsmith ef af., 1982; Leimbach ef al., 1984; Agewall ef al., 1990) are in compliance with the results of Miller and co-workers (Miller ef al., 1991) in regard to the endocrine responses of LBNP. During increased levels of LBNP from -20 to -80 mmHg, exceeding those that Miller and co-workers used, arterial pulse pressure decreases which is subsequentlyfollowed by a decrease in mean arterial pressure. During these circumstances with combined unloading of low- and highpressure baroreceptors, plasma A W and PRA increase (Mark ef af., 1978; Leimbach ef af., 1984; Nabel ef al., 1987; Sander-Jensen et al., 1988). Thus, the renal responses to LBNP mimic those of HUT. However, in contrast to the HUT model the relative importance of low- and high-pressure baroreceptors, respectively, on endocrine and renal responses caq be evaluated by using the LBNP model. Recent results from our laboratory (not yet published) and those of other groups (Goldsmith ef al., 1982; Miller ef af., 1991) indicate that selective inhibition of low-pressure baroreceptors induces a decrease in renal fluid and sodium excretion
Gravitational stress
511
through modulations of SNA, NE, and ANP release. Furthermore, deloading of arterial baroreceptors by further increasing the level of LBNP augments the renal response to a simulated gravitational stress by also inducing increases in PRA and AVP release. SUMMARY OF RENAL RESPONSES TO INCREASED GRAVITATIONAL STRESS
Increasing the G,-stress in humans induces a decrease in renal fluid and sodium excretion. The mechanisms of the renal response to a gravitational stress may entail increases in the release of AVP, PRA, PA, SNA, and a decrease in ANP release. By utilizing the HUT and LBNP models to induce or simulate an increase in G,-stress, the relative importance of low- and high-pressure baroreceptors, respectively, on endocrine responses can be investigated. Decreased gravitational stress WATER IMMERSION
In contrast to the models that simulate or induce an increase in +&-levels, the cardiovascular, endocrine, and renal adaptaton to head-out water immersion (WI) in humans have been extensively investigated. For a detailed description of the model, the reader is referred to several comprehensive reviews (Epstein, 1978; Norsk & Epstein, 1988; Epstein et al., 1989b; Krasney et al., 1989). It has been suggested that WI simulates aspects of weightlessness since intravascular hydrostatic gradients are counteracted by the water pressure exerted upon the body (Gauer et al. , 1970; Norsk & Epstein, 1988). The data on volume regulation in this model are usually based upon 3-8 h of investigation and upon comparing the data during WI with those from a similar seated control study. In brief, by immersing seated humans in thennoneutral water (34.5"C) to the level of the neck, blood is redistributed from the caudad portions of the body into the intrathoracic circulation. This induces increases in central blood volume of -700 ml (Arboreliuset al., 1972),heart volume of 180-250 ml (Lange et al. , 1974; Risch et al. , 1978), central venous pressure (CVP) of 7-18 mmHg (Echt et al. , 1W4; Gauer & Henry, 1976;Norsk et al. ,1985,1986), transmural CVP of 8-13 mmHg (Echt et al., 1974; Gauer & Henry, 1976), cardiac output (CO) of 17-36% (Begin et al., 1976; Norsk et al., l W ) , stroke volume (SV)of 3567% (Gauer & Henry, 1976; Norsk et al. , 1990), and arterial pulse pressure (PP) of 25% (Norsk et al., 1986). The increases in CVP, SV, and PP persist throughout prolonged WI of several hours duration (3-6 h) (Norsk et al. , 1985, 1986). Concomitantly, immersion induces a natriuresis and diuresis (Epstein, 1978). The magnitude and temporal profile of the diuresis of WI is dependent upon the hydration state of the subject (Behn et al. ,1%9). Thus, in fluid-restricted humans, the diuresis is primarily characterized by an increase in solute excretion and attain peak values after
5 12
P. Norsk
4-5 h of WI. In hydrated subjects, the diuresis is characterized by an increase in both
solute and free-water clearance and peak values are attained after 1-3 h of WI (Epstein, 1978). The magnitude and temporal profile of the natriuresis of WI is dependent upon the sodium balance of the subjects. In sodium replete subjects, the natriuresis peaks earlier than in depleted states and the absolute increase is greater. Furthermore, Epstein and co-workers (Epstein et al., 1973) have demonstrated that the increase in urinary flow rate and sodium excretion during WI exceed those of a supine control. The renal response to WI is mediated by several hormonal effectors. Suppression of PRA, PA, plasma AVP, plasma NE, and increases in plasma ANP act in concert to mediate the diuresis and natriuresis of immersion (Epstein, 1978; Norsk & Epstein, 1988; Epstein et al., 1989b). Furthermore, changes in intrarenal prostaglandine and dopamine activity and in intrarenal blood flow distribution might contribute to the renal responses (Epstein et al., 1979; Coruzzi et al., 1986). The quantitative importance of all these hormonal mediators is not known at present. In a more prolonged WI experiment conducted in our laboratory (Stadeager et al., personal communication), the effects of 12 h of WI-induced central hypervolaemia on renal fluid and sodium excretion were investigated. Renal sodium and water excretions were attenuated following the usually observed increase. However, renal sodium excretion was still elevated after 12 h of WI whereas renal water excretion attained control levels already after 7 h. Furthermore, the endocrine responses indicated that the initial increase in renal sodium excretion might be initiated by simultaneous suppressions of PRA, PA, and plasma NE and an increase in ANP release. However, the attentuation of the renal sodium excretion could have been induced by the observed attentuation of ANP release over the 12 h. These results illustrate the complexity of the mechanisms of the renal responses to WI. Usually, the increase in central blood volume during WI is considered to be the initiating factor of the renal response through stimulation of low-pressure baroreceptors (Gauer & Henry, 1976;Epstein, 1978). However, since results of several investigations from our laboratory have demonstrated that arterial pulse pressure increases (Norsk et al., 1985, 1986), stimulation of arterial baroreceptors must also have occurred. Furthermore, only little attention has been put on observed increases in blood and plasma volume which occur due to filtration of fluid from the interstitium to the intravascular space (Greenleaf, 1984). Theoretically, an increase in total blood volume and a decrease in plasma colloid osmotic pressure might also participate in the renal responses to WI. Of considerable interest is the recent discovery by Schulz-Knappe and co-workers (Schultz-Knappe et al., 1988) of an ANP related peptide in urine. The 32-amino-acid peptide, known as urodilatin, is composed of the entire sequence of a-human ANP plus a four-amino-acid N-terminal extension. Preliminary results obtained by Norsk and co-workers and Drummer and co-workers (personal communications) indicate that 12 h of WI induces an increased rate of urinary urodilatin excretion which
Gravitational stress
513
basically follows a temporal pattern similar to the increases in urine flow and sodium excretion. However, the importance of these discoveries has yet to emerge. Russian investigators have carried out prolonged immersions of several days duration. Gogolev and co-workers (Gogolev et al., 1980) carried out a study utilizing the so called 'dry' immersion technique for 7 days. The subjects were surrounded by a thin plastic covering so that the skin was protected from the water and they were recumbent during immersion. During control, the subjects were ambulatory. These authors reported that a negative water balance (difference between intake and renal output) of -700 ml was attained during the initial 24 h of immersion compared to a positive balance of 400 ml during control. During the subsequent days, the water balance approximated zero. The results of this study are difficult to evaluate since no specific details were given of the experimental design and since the subjects were recumbent during immersion. In summary, WI in humans clearly induces an increased rate of renal fluid and sodium excretion. The stimuli for the renal responses probably constitute stimulations of low- and high-pressure baroreceptors since central blood volume, CVP, and PP increase. Furthermore, an increase in plasma volume due to filtration of fluid from the interstitial to the intravascular space also might constitute a stimulus for the renal responses. Several endocrine elements such as AVP, PRA, PA, NE, and ANP probably participate as effectors in the natriuresis and diuresis of WI.The discovery of urodilatin, which is an ANP-related peptide produced by the kidney, has recently attracted attention as a possible candidate for the diuresis and natriuresis of central volume expansion, including WI (Goetz, 1991). HEAD DOWN TILTED BEDREST
By tilting human subjects head down 3-10", the gravitational stress exerted upon the human body can be minimized. This model was initially introduced by Russian investigators (Kakurin et al., 1976) as a more effective way than horizontal bedrest of simulating the effects of weightlessness on the cardiovascular system. This model has since been widely accepted by investigators in NASA and within the European space community. It should be noted that results of head-down tilt (HDT) are usually compared to results obtained with the subjects in the supine position in the hours prior to and following tilting (Nixon et al., 1979;Gaffney et al., 1985)or to measurements performed on the subjects while ambulatory (Noskov, 1982;Hargens et al., 1983). Thus, except for a few studies (Cathcard & Williams, 1955;Gharib et al., 1988) an appropriate time control with the subjects seated or supine have not been conducted. Compared to the supine position in the hours prior to tilting the immediate cardiovascular effects of HDT of 5-6" can be summarized as follows: CVP increases by 2-4 mmHg, CO by 15%, SV by -lo%, and left ventricular end diastolic diameter (LVED) by -20% (Nixon et al., 1979;Gaffney et al., 1985). During prolonged HDT experiments of 20-24 h duration, these changes usually occur within the first 0-5-2h.
5 14
P. Norsk
Thereafter, the increases in CVP, SV and CO gradually subside to attain pre-study levels after 20-24 h. Upon resuming the supine position after 20 h of HDT, CVP is decreased by 14%, SV by 19%, and CO by 17%, compared to supine, pre-tilt levels. Simultaneously, body weight decreases by 1.3 kg over 24 h of HDT. Since HDT experiments primarily have been performed in order to examine the effects upon the cardiovascular system, the renal response to HDT has not been well characterized. Cathcart & Williams (1955) compared the renal responses during - 12" HDT with that of supine rest for a comparable length of time. These authors observed no effect on renal fluid and electrolyte excretion of HDT. Other investigators have concluded that a negative sodium and fluid balance develops during HDT (Volicer et al., 1976) and that renal sodium and fluid excretions are augmented (Nixon et al., 1979; Hargens et al., 1983; Gaffney et al., 1985). However, since these investigations did not comprise time-control experiments and controlled intakes of food and fluid, the results of these studies are difficult to interpret. A prolonged HDT study of -3" for 12 h has been conducted in our laboratory (Norsk et al., data not published). The results concerning changes in CVP and renal sodium excretion are shown in Fig. 2, where they are also compared to a seated time control experiment and to seated WI for the same length of time in the same subjects. It is clear that changing the body position from seated (+1G,) to HDT (slightly negative GJ induces an augmented excretion rate of sodium. Concomitantly, CVP is increased by approximately 8 mmHg. Furthermore, the data reveals that seated WI induces an even larger increase in renal sodium excretion accompanied by a 12 mmHg increase in CVP. In summary, the HDT model has been primarily used to simulate the effects of weightlessness on the cardiovascular system. HDT induces increases in central blood volume as evidenced by increases in CVP,CO, SV, and LVED compared to both the supine and seated positions, respectively. None of the studies to date have convincingly demonstrated that these increases in central cardiovascular variables are accompanied by a diuresis and natriuresis when compared to the supine position (Cathcart & Williams, 1955; Nixon et al., 1979; Gaffney et al., 1985). Furthermore, compared to the supine position, HDT does not induce a statistically significant suppression of plasma levels of AVP, PRA, A, and NE (Nixon et al., 1979; Gaffney et al., 1985). However, compared to a seated control (+1 Gz) HDT clearly induces increases in renal fluid and sodium excretion and suppresses PRA, PA, and plasma NE (Gharib et al., 1988; personal observations in Fig. 2). Gogolev and co-workers (Gogolev et al., 1980)carried out an HDT study for 7 days. During control, the subjects were ambulatory. These authors reported that the water balance was zero during the initial 24 h of HDT cornpared to a positive balance of 400 ml during control. Over the subsequent days, the water balance increased to control levels. The results of this study are difficult to evaluate since no specific details were given of the experimental design.
515
Gravitational stress HDT or WI
"T
m w1
-5 1
Control
0
2
4
6
8
1
0
1
2
Time (h)
Flg. 2. Central venous pressure (CVP)and renal sodium excretion (UN.V) before, during and after 12 h of water immersion (WI)to the neck (triangle), head-down tilted bedrest (HDT) of -3" (filled circle), and a seated control (blank circle). Values are means 2 SE of 6 male subjects. *indicates significant changes (P
