Journal of Applied Physiology

Selected Contribution: Sex differences in osmotic regulation of AVP and renal sodium handling

Nina S. Stachenfeld, Andres E. Splenser, Wendy L. Calzone, Matthew P. Taylor, David L. Keefe


To determine sex differences in osmoregulation of arginine vasopressin (AVP) and body water, we studied eight men (24 ± 1 yr) and eight women (29 ± 2 yr) during 3% NaCl infusion [hypertonic saline infusion (HSI); 120 min, 0.1 ml · kg body wt−1 · min−1]. Subjects then drank 15 ml/kg body wt over 30 min followed by 60 min of rest. Women were studied in the early follicular (F; 16.1 ± 2.8 pg/ml plasma 17β-estradiol and 0.6 ± 0.1 ng/ml plasma progesterone) and midluteal (L; 80.6 ± 11.4 pg/ml plasma 17β-estradiol and 12.7 ± 0.7 ng/ml plasma progesterone) menstrual phases. Basal plasma osmolality was higher in F (286 ± 1 mosmol/kgH2O) and in men (289 ± 1 mosmol/kgH2O) compared with L (280 ± 1 mosmol/kgH2O, P < 0.05). Neither menstrual phase nor gender affected basal plasma AVP concentration (P[AVP]; 1.7 ± 4, 1.9 ± 0.4, and 2.2 ± 0.5 pg/ml for F, L, and men, respectively). The plasma osmolality threshold for AVP release was lowest in L (x-intercept, 263 ± 3 mosmol/kgH2O, P < 0.05) compared with F (273 ± 2 mosmol/kgH2O) and men (270 ± 4 mosmol/kgH2O) during HSI. Men had greater P[AVP]-plasma osmolality slopes (i.e., sensitivity) compared with F and L (slopes = 0.14 ± 0.04, 0.09 ± 0.01, and 0.24 ± 0.07 for F, L, and men, respectively,P < 0.05). Despite similar Na+-regulating hormone responses, men excreted less Na+ during HSI (0.7 ± 0.1, 0.7 ± 0.1, and 0.5 ± 0.1 meq/kg body wt for F, L, and men, respectively, P < 0.05). Furthermore, men had greater systolic blood pressure (119 ± 5, 119 ± 5, and 132 ± 3 mmHg for F, L, and men, respectively,P < 0.05) than F and L. Our data indicate greater sensitivity in P[AVP] response to changes in plasma osmolality as the primary difference between men and women during HSI. In men, this greater sensitivity was associated with an increase in systolic blood pressure and pulse pressure during HSI, most likely due to a shift in the pressure-natriuresis curve.

  • estrogen
  • progesterone
  • testosterone
  • sodium regulation
  • arginine vasopressin
  • cortisol

cardiovascular disease-relatedmorbidity and mortality are lower in women than in men throughout middle age (10, 19). During and after menopause, this “risk advantage” can be maintained with estrogen administration (19). The mechanism for these gender differences is usually associated with estrogen; however, there are data to support a role for androgens in the differing disease development between men and women (22). One important estrogen-related mechanism involved in the lowering of cardiovascular disease risk among women is sodium and water regulation (7, 20). However, both estrogen and testosterone receptors are present in the hypothalamus of adult primates (16); therefore, both may alter control of the cardiovascular, fluid, and sodium regulation systems in the hypothalamus (24, 25, 29, 30).

Estrogen and testosterone modulate the synthesis of arginine vasopressin (AVP) in the paraventricular and supraoptice nuclei in rats (24, 25) as well as the osmotic regulation of AVP release by the pituitary in humans (26, 28). Previous studies have found greater 24-h urinary AVP (7) and greater resting and exercise plasma AVP concentration (P[AVP]) (27) in men compared with women. Unfortunately, it is difficult to determine whether these differences are due to estrogen or testosterone effects on osmotic regulation of AVP. Moreover, no human studies have specifically addressed gender differences or how estrogen, progesterone, or testosterone may interact to alter AVP release to impact the cardiovascular, water, and sodium regulation systems.

Human studies to determine the mechanisms for sex differences in body fluid regulation are crucial to developing effective treatments for conditions such as hypertension and cardiovascular and renal disease that are equally efficacious for men and women. The purpose of this study was to determine gender differences in the P[AVP]responses to hypertonic saline infusion (HSI) by comparing men with women in the early follicular menstrual phase (i.e., low estrogen-low progesterone) and then again in the midluteal menstrual phase (i.e., high estrogen-high progesterone). We hypothesized that greater estrogen and progesterone would enhance the AVP response to osmotic stimulation. Therefore, men would only differ from women in the midluteal phase, thereby implicating estrogen as the primary mechanism for changes in osmotic regulation of AVP. To determine the functional consequences of changes in plasma osmolality regulation of AVP, we also examined sex differences in body water and in the cardiovascular system.


Study Design

Subjects were healthy, nonsmoking women (n = 8, age 29 ± 2 yr, range 22–35 yr) and men (n = 10, age 24 ± 1 yr, range 20–33 yr). All subjects were interviewed about their medical history. The women participated in two HSI tests: one in the early-follicular phase (2–4 days after the beginning of menstrual bleeding) and one conducted in the midluteal phase (7–9 days after the luteinizing hormone peak). The first phase corresponded to low estrogen-low progesterone and the latter to high estrogen-high progesterone. The midluteal phase was determined individually by the use of ovulation prediction kits (OvuQuick, Quidel, San Diego, CA). To verify the phase of the menstrual cycle, plasma levels of estrogen and progesterone were assessed from a baseline blood sample. The male subjects were exposed to HSI only once. The subjects gave written, informed consent to participate in the study, which had prior approval by the Human Investigations Committee of Yale University School of Medicine.

Infusion Studies

For each experiment, the women and men arrived at the laboratory at ∼8:30 AM after eating a low-fat (∼300 kcal) breakfast at home and after refraining from caffeine for 12 h before the experiment. The subjects were also instructed to avoid high-salt foods and alcohol for 24 h before the experiment. After reporting to the laboratory, each subject voided her/his bladder, entered an environmental chamber (27°C, 30% relative humidity), was weighed to the nearest 10 g on a beam balance, and then allowed to rest seated for a 60-min control period. During this period, a 20-gauge Teflon catheter was placed in an antecubital or forearm vein in each arm with a heparin block (20 U/ml) to maintain catheter patency. A cuff was positioned for automatic readings by a sphygnomanometric device (Colin Medical Instruments, Komaki, Japan) to monitor changes in blood pressure. A three-lead electrocardiogram (Colin Medical Instruments) provided continuous heart rate monitoring. At the end of the 60-min control period, a baseline blood sample was taken, and a urine sample was collected.

After these control samples were obtained, hypertonic (3.0% NaCl) saline was infused at a rate of 0.1 ml · kg body wt−1 · min−1 for 120 min through one of the catheters. Blood was sampled at 10, 20, 30, 40, 50, 60, 75, 105, and 120 min during the infusion from the other catheter, and a urine sample was obtained at the end of the infusion. After a 30-min seated recovery period, the subject drank water (15 ml/kg body wt) during the next 30 min. Thirty minutes after the drinking period was completed, plasma volume was determined with the Evans blue dye procedure (seeBlood Volume below). Blood samples were obtained at 30, 60, and 120 min after infusion, and urine was collected at 60 and 120 min after infusion. The subjects were weighed at the end of the infusion period, after the drinking period, and again at the end of the protocol.

All blood samples were analyzed for hematocrit (Hct), hemoglobin concentration ([Hb]), plasma total protein concentration, plasma osmolality, plasma creatinine concentrations, P[AVP],plasma cortisol concentrations (P[Cort]), and serum concentrations of sodium (S[Na]) and potassium (S[K]). Blood samples at baseline, at the end of the infusion, and at 60 and 120 min after the infusion were also analyzed for the plasma concentrations of atrial natriuretic peptide (P[ANP]), aldosterone (P[Ald]), and plasma renin activity (PRA). Baseline blood samples from the women were also analyzed for 17β-estradiol and progesterone. Urine was analyzed for volume, osmolality, sodium, potassium, and creatinine concentrations.

Blood samples were separated immediately into aliquots. The first was analyzed for [Hb] and Hct. A second aliquot was transferred to a heparinized tube, and a third aliquot for the determination of S[Na] and S[K] was placed into a tube without anticoagulant. All other aliquots were placed in tubes containing EDTA. The tubes were centrifuged, and the plasma was taken off the heparinized sample and analyzed for sodium, potassium, osmolality, creatinine, and aldosterone. The EDTA samples were analyzed for concentrations of AVP, cortisol, ANP, and PRA.

Blood Analysis

Hemoglobin was measured in triplicate by the cyanomethemoglobin technique and Hct in triplicate by the microhematocrit method. Plasma and urine sodium and potassium were measured by flame photometry (Instrumentation Laboratory, model 943), plasma osmolality by freezing point depression (Advanced Instruments 3DII), plasma proteins by refractometry, and plasma creatinine concentration by colorimetric assay (Sigma Diagnostic Products). PRA, P[Ald], P[Cort], P[ANP], and plasma 17β-estradiol and progesterone concentrations were measured by radioimmunoassay. P[AVP] was determined by radioimmunoassay after extraction from plasma by the methods described by Freund et al. (13, 14), on octadecylsilane cartridges (SEP-PAK C18, Waters Associates, Needham, MA). Extracted samples were assayed by using a disequilibrium assay with the extracts incubated with the antiserum at 4°C for 72 h, followed by the addition of 125I-labeled AVP (New England Nuclear, Boston, MA). Bovine albumin-coated charcoal was used for separation of free and antibody-bound labeled AVP. This assay is highly specific for AVP with the antiserum prepared against a lysine vasopressin-thyroglobin conjugate and has a sensitivity of 0.6 pg/ml. The extraction recovery was 87%. Inter- and intra-assay coefficients of variation for the midrange standards were, respectively, as follows: 7.11 and 10.71% for 3.1 pg/ml P[AVP], 3.6 and 4.5% for 12.1 g/dl P[Cort] (Diagnostic Products, Los Angeles, CA), 2.8 and 5.0% for 4.8 ng · ml−1ANG · h−1 PRA (Immuno Biological Laboratories), 3.8 and 5.2% for 181 pg/ml P[Ald] (Diagnostic Products), 6.0 and 8.2% for 61.8 pg/ml P[ANP] (Diasorin, Stillwater, MN), 5.6 and 8.3% for 81 pg/ml plasma 17β-estradiol (Diagnostic Products), and 2.1 and 2.7% for 1.5 ng/ml plasma progesterone (Diagnostic Products).

Blood Volume

Absolute blood volume was measured by dilution of a known amount of Evans blue dye. An accurately determined volume of dye (by weight, because the specific density is 1.0) was injected into an arm vein, and a blood sample was taken at 10, 20, and 30 min for determination of dilution after complete mixing had occurred (10 min). Blood was also sampled at 20 and 30 min to ensure that complete mixing had occurred by the 10-min sample. If not, the 20-min sample was used for analysis. However, if complete mixing had not occurred by the 20-min sample, the blood volume measurement was not used. Plasma volume was determined from the product of the concentration and volume of dye injected divided by the concentration in the plasma after mixing, taking into account 1.5% lost from the circulation within the 10 min. Blood volume was calculated from plasma volume and Hct concentration, corrected for peripheral sampling.


Changes in plasma volume (ΔPV) were estimated from changes in [Hb] and Hct from the pre-HSI sample, according to the equation (17)%ΔPV=100×{([Hbb]/[Hba])×(1Hcta·102)/ (1Hctb·102)}100 where subscripts a and b denote measurements at time a and pre-HSI, respectively.

We used the pre-HSI [Hb] and Hct to estimate the baseline plasma volume from the Evans blue dye measurement at the end of the experiment.

Fractional excretions of water (FEH2 O) and sodium (FENa) were calculated from the following equationsFEH2O=(UV/GFR)·100 FENa=(UV·U[Na]/GFR·[Na+]f)·100 [Na+]f=the Donnan factor for cations(0.95)·S[Na] where the subscript f is glomerular filtrate, Uv is urine flow rate, GFR is the glomerular filtration rate, U[Na] is the urine excretion of Na concentration, and S[Na] is in protein-free solution (meq/kgH2O). Glomerular filtration rate was estimated from creatinine clearance.

Data Analysis

For each subject, osmotic regulation of AVP was determined by plotting P[AVP] as a function of plasma osmolality during HSI. The sensitivity of P[AVP] to changes in plasma osmolality provides the slope of this relationship, and the intercept provides the threshold for P[AVP] release. Body water handling was determined through the assessment of overall fluid balance and the renal clearance of free water.


The variables over time were analyzed by conditions (menstrual phase) and group (women vs. men) using ANOVA for repeated measures. We used separate ANOVA models to test differences between the men and women for each menstrual phase and to test the women within the menstrual phases. When significant differences were found, orthogonal contrasts tested differences between specific means related to the hypothesis of interest. Data are expressed as means ± SE. Differences were considered statistically significant at P < 0.05.

Sample size calculation.

Expected P[AVP] responses within and between groups are derived from data from our laboratory during HSI (26). In an earlier study, during HSI, P[AVP]-plasma osmolality threshold decreased by 6 mosmol/kgH2O with estrogen treatment. An estimate of the pooled standard deviation for the group was 3 mosmol/kgH2O.

The desired statistical test is two-sided at an α-level of 0.05 with 80% power to detect a difference. On the basis of our previous work, 80% power is sufficient to detect a significant alteration in plasma osmolality. For a two-sided test, Z (α) = 1.96; for 80% power, Z (β) = 0.84. The formula for calculating sample size for continuous response variables is (6)N=2{[Z(α)+Z(β)]2(s)2/(d)2} Substituting the values, the calculated sample size is eight subjects per group.


Baseline (pre-HSI)

The men were taller, weighed more, and were younger than the women (Table 1, P < 0.05). Basal Hct and [Hb] were higher in the men compared with the women in both menstrual phases (Table 2,P < 0.05). Plasma osmolality was lower in the women in the luteal phase compared with men; within both groups of women, plasma osmolality was lower in the midluteal phase compared with the follicular phase (Fig. 1,P < 0.05). Estimated plasma and blood volumes were greater in the men compared with women in both menstrual phases; within both groups of women, estimated plasma and blood volumes were greater in the follicular compared with the midluteal phase (Table 1,P < 0.05). There were no baseline differences in P[AVP] or P[Cort] between men and women in either menstrual phase (Figs. 1 and 2).

View this table:
Table 1.

Subject characteristics and baseline PV, BV, P[E2], P[P4], P[AVP], and POsm

View this table:
Table 2.

Blood responses at rest and during HSI and recovery in men and women

Fig. 1.

Plasma osmolality (POsm), plasma concentration of arginine vasopressin (P[AVP]), and percent change in plasma volume at rest, in response to hypertonic saline infusion and recovery in the follicular and midluteal menstrual phases in women, and in men. *Different from the follicular phase. §Different from the midluteal phase. #Different from the men. Differences were accepted as significant at P < 0.05. Data are means ± SE.

Fig. 2.

Plasma renin activity (PRA), plasma aldosterone (P[Ald]), plasma atrial natriuretic peptide (P[ANP]), and plasma cortisol (P[Cort]) concentrations at rest, in response to hypertonic saline infusion and recovery in the follicular and midluteal menstrual phases in women, and in men. *Different from the follicular phase. §Different from the midluteal phase. #Different from the men. Differences were accepted as significant at P < 0.05. Data are means ± SE.

Baseline PRA and P[Ald] were higher in the midluteal phase compared with the follicular phase within women and compared with the men (Fig. 2, P < 0.05). Urine flow, osmolality and sodium excretion (Fig.3), and free water and osmotic clearances were unaffected by menstrual phase or gender, indicating that baseline hydration levels were consistent between the menstrual phases and between the men and women (Table 3). Basal heart rate was unaffected by menstrual phase or gender, but systolic, diastolic, and mean arterial pressures were greater in the men than in the women in both menstrual phases (Table4, P < 0.05).

Fig. 3.

Renal Na+ and cumulative Na+excretion at rest corrected for body weight (BW) during hypertonic saline infusion and recovery in the follicular and midluteal phases in women and in men. *Different from the follicular phase. §Different from midluteal phase. Differences were accepted as significant atP < 0.05. Data are means ± SE.

View this table:
Table 3.

Renal osmoregulatory responses at rest and during HSI and recovery

View this table:
Table 4.

Cardiovascular responses at rest and during HSI and recovery

Hypertonic Saline Infusion

Among women, HSI increased plasma osmolality and plasma volume similarly during the follicular and midluteal phases. During the early part of the infusion, however, plasma osmolality during the midluteal phase remained significantly below that of the follicular phase and below that of the men (Fig. 1, P < 0.05). In contrast, P[AVP] in the men increased above that of the women in both menstrual phases (Fig. 1, P < 0.05). Linear regression analysis of the individual subjects' data during HSI indicated significant correlations between P[AVP] and plasma osmolality [mean r = 0.82 ± 0.02 for women (both phases) and mean r = 0.83 ± 0.04 for the men (P < 0.05)]. The slope of the P[AVP]-plasma osmolality relationship during HSI was unaffected by menstrual phase but was greater in the men compared with women in both menstrual phases (Table 1, Fig.4, P < 0.05). The mean abscissal intercept of the linear P[AVP]-plasma osmolality relationship was shifted downward in the women during the midluteal phase relative to the follicular phase but was unaffected by gender (Table 1, Fig. 4, P < 0.05).

Fig. 4.

Mean P[AVP] responses to increases in POsm during hypertonic saline infusion in the follicular and midluteal phases in women and in men. Data are means ± SE.

P[Cort] fell steadily during the infusion in the women in both menstrual phases but began to increase in the men at 75 min of infusion and continued to increase through the end of the infusion (Fig. 2, P < 0.05). PRA and P[Ald]decreased during HSI in the men and the women in both menstrual phases, although P[Ald] was greater in the midluteal phase than in the follicular phase within women (Fig. 2, P < 0.05). Although renal sodium excretion increased during HSI in the men and the women in both menstrual phases, the magnitude of increase was lower in the men compared with the women (Fig. 3 and Table 3,P < 0.05). Heart rate and blood pressure were unaffected by HSI in the women, but systolic and pulse pressures increased by the end of the infusion in the men (Table 4,P < 0.05).


Plasma osmolality and percent changes in plasma volume were similar between menstrual phases and between men and women during recovery from HSI (Fig. 1). Throughout recovery, P[AVP]remained greater in the men compared with the women (Fig. 1,P < 0.05), and P[Cort] remained greater in the men until 90 min into recovery (Fig. 2, P < 0.05). At the end of recovery, there were no menstrual phase or sex differences in PRA, P[Ald], or P[ANP].

During recovery, renal function, electrolyte excretion (Fig. 3 and Table 3), and overall fluid balance (i.e., fluid intake − urine output) were not affected by phase of the menstrual cycle or by gender. Due to the greater sodium retention during the infusion in the men, cumulative sodium excretion was lower in men than in women by the end of recovery (Fig. 3, P < 0.05). Throughout recovery, blood pressure remained greater in men compared with women, and systolic blood pressure and pulse pressure were greater during recovery in men compared with their own baseline values (Table 4, P < 0.05).


The greater P[AVP]-plasma osmolality slope observed in men indicated greater P[AVP] sensitivity to changes in plasma osmolality during HSI compared with women in both phases of the menstrual cycle. This greater sensitivity occurred without differences in free water clearance, thereby suggesting lower renal sensitivity to AVP in men. In contrast, within women, the osmotic threshold was shifted to a lower plasma osmolality in the midluteal phase relative to the follicular phase of the menstrual cycle, although without a change in the P[AVP]-plasma osmolality slope. As with the men, this shift in the osmotic threshold for AVP release occurred with no change in free water clearance or body fluid balance, suggesting a change in the osmotic operating point for fluid regulation over the menstrual cycle, most likely due to estrogen actions on AVP release (28). HSI increased P[Cort] in men but not in women in either menstrual phase. The mechanism for the P[Cort] response is not apparent from our data but may be due to an interaction between AVP and corticotropin-releasing hormone (CRH) in the hypothalamus that is unique to men or perhaps due to greater stress responses to the high levels of thirst induced by HSI. Finally, there were no changes in blood pressure or sodium excretion observed between the menstrual phases in women. However, HSI lowered sodium excretion and increased blood pressure in men, suggesting a hypertensive shift in the pressure natriuresis curve in men, presumably related to testosterone.

A primary difference between men and women was the greater P[AVP]-plasma osmolality slope in men, indicating greater sensitivity of systems regulating AVP synthesis and/or release in response to changes in plasma osmolality. There was no difference in osmotic AVP sensitivity between the two menstrual phases, suggesting that androgens are a primary factor in this shift. Our study does not conclusively demonstrate that testosterone is the primary cause for the sex differences in P[AVP]-plasma osmolality slope; however, the hypothalamus and pituitary have binding sites for testosterone (24) and testosterone has effects independent of estrogen on hypothalamic AVP stimulation in rats (8). In women, the P[AVP]-plasma osmolality slope during dehydrating exercise is unaffected by menstrual phase or by administration of oral contraceptives containing estrogen (30). In the present study, we also found a shift to a lower osmotic threshold for AVP release in the midluteal phase within women but no difference due to gender, suggesting estrogen is the primary cause of the change in osmotic threshold but has little effect on osmotic AVP sensitivity.

Our data do not explain why estrogen and testosterone would modulate osmotic AVP regulation differently, but both estrogen and testosterone are probably having their effects within the central nervous system. Estrogen and androgens cross the blood-brain barrier, and vasopressin-producing cells in the paraventricular and supraoptic nuclei have binding sites for both testosterone and estrogen that affect osmotic regulation (1, 8, 24, 25), thereby affecting synthesis and release of AVP. Estrogen and androgens may also modulate hypothalamic AVP release indirectly through catecholaminergic neurons that bind estrogen and testosterone and project to the paraventricular and supraoptic nuclei (18) and by sex steroid modulation of cholinergic and angiotensinergic innervation of vasopressinergic cells in the paraventricular and supraoptic nuclei (29).

Another mechanism involved in the sex differences in osmotic AVP regulation may be changes in plasma volume. The volume expansion produced by HSI should have increased central venous pressure by 2.2 mmHg (15), which is a sufficient stimulus to load cardiopulmonary baroreceptors. Loading cardiopulmonary baroreceptors in itself can provide a countersignal concerning fluid status to that of the elevated plasma osmolality. In this way, cardiopulmonary baroreceptor loading can modulate the AVP response to the increased plasma osmolality. Thus a lower cardiopulmonary baroreceptor sensitivity in men relative to women would make them less sensitive to the inhibitory stimulus of plasma volume expansion and is a possible mechanism for the greater P[AVP]-plasma osmolality slope in the men during HSI. Watenpaugh et al. (34), however, recently demonstrated that men and women have similar cardiovascular and renal responses to water immersion in which the primary stimulus is plasma volume expansion and cardiopulmonary baroreceptor loading.

Another peripheral mechanism that might have contributed to the greater P[AVP]-plasma osmolality slope in the men could be the increase in pulse pressure during HSI, which would reflect an increase in stroke volume or a reduction in arterial compliance. An increase in stroke volume seems an unlikely mechanism because greater stroke volume increases would have attenuated, rather than augmented, the P[AVP]-plasma osmolality slope in the men through countereffects on cardiopulmonary baroreceptors. However, a sodium load and/or vasopressin secretion can reduce arterial compliance in males. For example, in rats, the pressor response to vasopressin secretion is greater in males vs. females due to reduced total peripheral resistance in females (32), which is mediated by estrogen (31). Thus, in our study, greater arterial resistance during HSI may have augmented the pulse pressure responses in the men, which in turn increased the P[AVP]-plasma osmolality slope during HSI.

The increase in systolic blood pressure in the men during HSI may be due to an upward shift in the pressure-natriuresis curve (22). Sodium retention was greater in the men vs. the women, which is consistent with this upward shift, already reported in spontaneously hypertensive male vs. female rats (22) and in ovariectomized female rats given testosterone (5). Moreover, there is evidence that estrogen modifies the renin-angiotensin system (21), and recent evidence in transgenic hypertensive mice indicates that estrogen may act to reduce vasoconstriction and hypertension by shifting the vasoconstrictor-vasodilation balance in the kidney (3). Finally, androgens may have direct effects on proximal sodium reabsorption in the kidney, although evidence that androgen receptors exist in the proximal tubules is still preliminary (22).

Despite the greater P[AVP]-plasma osmolality slope in the men, the fall in free water clearance due to the osmotic load was unaffected by gender, suggesting greater renal sensitivity to AVP in women. This is consistent with data in postmenopausal women given estrogen, in whom water retention and renal concentrating response (urine osmolality/plasma osmolality) during HSI increased despite similar P[AVP], suggesting that estrogen enhances renal free water retention (26). On the other hand, more recent data suggest lower renal sensitivity to changes in P[AVP]in response to a water load in young women (4). Moreover, combined estrogen and progesterone administration to young women reduced the free water clearance response to increases in P[AVP] during exercise-induced dehydration (28).

Basal P[Cort] was similar between men and women, regardless of menstrual phase; however, P[Cort] increased only in men during HSI. Both estrogen (11) and androgens (11) have some control over the hypothalamic synthesis and release of CRH. CRH, released by the paraventricular nucleus, is the first component of the hypothalamic-pituitary-adrenal-axis system that ultimately leads to the release of cortisol. Modulation of AVP release by the posterior pituitary is a part of the negative feedback control of ACTH secretion (23) that is modified by testosterone (2) and estrogen (16). Thus the greater P[Cort] in men during HSI may be related to greater AVP release by the posterior pituitary.

In addition to the hypothalamic-pituitary-adrenal-axis system control, P[Cort] can be an indication of physiological or psychological stress, so the greater P[Cort] may be a stress response to HSI unique to the men. Men and women generally have similar basal and stress-induced blood cortisol response (9), although cortisol metabolism is lower in women (12, 33), supporting a greater P[Cort]response in the women. Thus a greater stress response in the men to HSI is an unlikely explanation for the differences in P[Cort]response to HSI.

Our data indicate a greater sensitivity in the P[AVP]response to changes in plasma osmolality as the primary difference between men and women during HSI. In men, this greater sensitivity was associated with an increase in systolic blood and pulse pressures and P[Cort] during HSI and thus may play a role in the greater cardiovascular morbidity in men compared with women. The greater blood pressure and sodium retention observed in men may also indicate a shift in the pressure-natriuresis curve, which may be due to differential sensitivity to the renin-angtiotensin system or related to direct effects on the kidney. Finally, this greater AVP sensitivity to osmotic stimulation may have also led to increases in cortisol and blood pressure during the infusion, although the functional significance of the greater adrenal stimulation is not apparent from our data.


Recent data from a number of medical and scientific disciplines have indicated that sex differences exist in all body systems. These differences are apparent at the cellular to the whole body level, with biochemical differences from both hormonal and genetic origins. These sex differences have important implications for the treatment of disease and the design of clinical trials and basic research studies. This particular study addresses the regulation of hormones involved in the control of body fluids and sodium, which has important implications for chronic diseases such as cardiovascular and renal disease and hypertension. In addition, these studies are also relevant to acute conditions, including hyponatremia and temperature regulation. The regulation of body temperature in humans is known to interact with systems that regulate volume and osmotic pressure of the extracellular fluid. Dehydration and heat exposure can result in heat exhaustion and, in extreme conditions, death. Thus understanding sex differences in the control of body fluids will play a role in the treatment of both chronic and acute medical conditions.


We gratefully acknowledge the technical support of Cheryl Weseman and David Blair, the clinical support of Dr. Celso Silva, and the cooperation of the volunteer subjects. We also thank Dr. Loretta DiPietro for contributions to the writing of this manuscript.


  • This work is supported by the U.S. Army Medical Research and Materiel Command under contract DAMD17-96-C-6093.

  • The views, opinions and/or findings contained in this report are those of the authors and should not be construed as an official Department of the Army position, policy or decision unless so designated by other documentation.

  • In conduct of research where humans are the subjects, the investigators adhered to the policies regarding the protection of human subjects as prescribed by 45 CFR 46 and 32 CFR 219 (Protection of Human Subjects).

  • Address for reprint requests and other correspondence: N. S. Stachenfeld, The John B. Pierce Laboratory, 290 Congress Ave., New Haven, CT 06519 (E-mail: nstach{at}

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract