Heart and Circulatory Physiology

Mechanisms of postspaceflight orthostatic hypotension: low α1-adrenergic receptor responses before flight and central autonomic dysregulation postflight

Janice V. Meck, Wendy W. Waters, Michael G. Ziegler, Heidi F. deBlock, Paul J. Mills, David Robertson, Paul L. Huang


Although all astronauts experience symptoms of orthostatic intolerance after short-duration spaceflight, only ∼20% actually experience presyncope during upright posture on landing day. The presyncopal group is characterized by low vascular resistance before and after flight and low norepinephrine release during orthostatic stress on landing day. Our purpose was to determine the mechanisms of the differences between presyncopal and nonpresyncopal groups. We studied 23 astronauts 10 days before launch, on landing day, and 3 days after landing. We measured pressor responses to phenylephrine injections; norepinephrine release with tyramine injections; plasma volumes; resting plasma levels of chromogranin A (a marker of sympathetic nerve terminal release), endothelin, dihydroxyphenylglycol (DHPG, an intracellular metabolite of norepinephrine); and lymphocyte β2-adrenergic receptors. We then measured hemodynamic and neurohumoral responses to upright tilt. Astronauts were separated into two groups according to their ability to complete 10 min of upright tilt on landing day. Compared with astronauts who were not presyncopal on landing day, presyncopal astronauts had 1) significantly smaller pressor responses to phenylephrine both before and after flight; 2) significantly smaller baseline norepinephrine, but significantly greater DHPG levels, on landing day; 3) significantly greater norepinephrine release with tyramine on landing day; and 4) significantly smaller norepinephrine release, but significantly greater epinephrine and arginine vasopressin release, with upright tilt on landing day. These data suggest that the etiology of orthostatic hypotension and presyncope after spaceflight includes low α1-adrenergic receptor responsiveness before flight and a remodeling of the central nervous system during spaceflight such that sympathetic responses to baroreceptor input become impaired.

  • tyramine
  • phenylephrine
  • dihydroxyphenylglycol
  • tilt tests
  • arginine vasopressin

astronauts who suffer from orthostatic hypotension and presyncope after short-duration spaceflight have at least two characteristics that separate them from astronauts who do not suffer from orthostatic hypotension and presyncope. First, they have lower vascular resistance than nonsusceptible astronauts, not only on landing day, but before flight as well (12, 33). Before flight, the low resistance is not associated with low supine or upright plasma norepinephrine levels, and the subjects do not become presyncopal during tilt testing. This suggests that astronauts who will become presyncopal after spaceflight have vascular responsiveness that is inherently lower than that of nonpresyncopal astronauts.

The second known difference between presyncopal and nonpresyncopal astronauts is the microgravity-induced decreases in sympathetic responses to orthostatic stress. Astronauts who are not presyncopal after spaceflight have supine and upright plasma norepinephrine levels that are significantly greater than their preflight levels (12, 33). In contrast, astronauts who do become presyncopal have the same or even lower upright plasma norepinephrine levels than they had preflight (12, 33). This suggests a spaceflight-induced disruption of proper adrenergic function, but it has not been learned where in the baroreflex arc the “lesion” may be.

The purpose of this study was to pursue the mechanisms of low preflight vascular resistance and low landing day norepinephrine release in presyncopal astronauts by testing two hypotheses. First, we proposed that pressor responses to intravenous injections of the α1-adrenergic receptor agonist phenylephrine would be lower in presyncopal than in nonpresyncopal astronauts. Second, we hypothesized that the amount of norepinephrine released by intravenous tyramine, an indirect sympathomimetic that forces release of norepinephrine from sympathetic nerves, would be similar in presyncopal and nonpresyncopal astronauts both before and after flight, but that, on the anding day, norepinephrine release with upright posture would be smaller in the presyncopal than in the nonpresyncopal astronauts.


Twenty-three astronauts, 21 men and 2 women, aged 41 ± 1 yr, participated in this Johnson Space Center Institutional Review Board approved protocol, after signing written, informed consent. They flew on missions lasting from 5 to 18 days. Six months before flight, subjects visited the laboratory and were made familiar with all the equipment and procedures. Studies were performed 10 days before launch, 2–3 h after landing, and 3 days after landing. On each study day, subjects had abstained from taking any medications for the previous 12 h, had not eaten a heavy meal within 4 h, had consumed a light snack within the previous 2 h, and had not exercised maximally within 24 h. Subjects were weighed for calculation of drug doses, laid on the tilt table, and instrumented for ECG, arterial pressure (Dinamap, GE Medical Systems Information Technologies; Milwaukee, WI), and beat-to-beat wrist arterial pressure (CBM 7000, Colin Medical Instruments; San Antonio, TX). Two-dimensional echocardiography (Biosound Genesis II, Esoate; Indianapolis, IN) was used to determine the aortic cross-sectional area at the level of aortic cusp insertion. Ascending aortic flow was sampled with pulsed-Doppler (Biosound Genesis II, Esoate; Indianapolis, IN) for determination of stroke volume. Indwelling catheters were placed in the antecubital veins in both arms: one for withdrawal of blood and one for infusion of drugs. After a 20-min supine rest, a baseline blood sample was withdrawn for analyses of catecholamines, arginine vasopressin, chromogranin A (a marker of sympathetic nerve terminal release), endothelin, hematocrit, dihydroxyphenylglycol (DHPG, an intracellular metabolite of norepinephrine), and β2-adrenergic receptors on the lymphocytes. Plasma and red blood cell volumes were measured with carbon monoxide rebreathing (24). Data were collected on digital, audio, and videotape and paper recorders.

Pharmacological Studies

All pharmacological studies were conducted while the subjects were in the supine position. Tyramine and phenylephrine were the two drugs administered. The order of administration was randomized from subject to subject. Adequate recovery times were given between drug injections for baseline parameters to return to normal.

Phenylephrine. Three baseline measurements of arterial pressure, R-R interval, and aortic outflow were taken. A bolus injection of 0.13 mg/1.73 m2 body surface area (BSA) of phenylephrine was then administered, and measurements were continued until all parameters returned to baseline. The procedure was then repeated using 0.26 mg/1.73 m2 BSA of phenylephrine.

Tyramine. Three baseline measurements of arterial pressure, heart rate, and aortic outflow were made. A bolus injection of 2.0 mg/1.73 m2 BSA of tyramine was then administered, and continuous measurements were obtained for exactly 4 min, at which time a blood sample was drawn for a plasma norepinephrine level. All variables were monitored until arterial pressure and heart rate returned to preinjection levels. The digital tapes and videotapes from the ultrasound system were synchronized by means of an event marker of which the output was sent to both machines simultaneously. After recovery, the procedure was repeated by using 4.0 mg/1.73 m2 BSA of tyramine.

Tilt test. When all baseline measurements had returned to normal after the drug injections, all hemodynamic measurements were continued for one additional minute. The tilt table was then raised to the 80° upright position, and measurements were continued for 10 min or until presyncopal symptoms intervened. At the end of tilt, the final blood sample was taken for analysis of catecholamines, arginine vasopressin, endothelin, and hematocrit. In the case of presyncopal subjects, the blood was drawn immediately as the tilt table returned to the supine position.


Phenylephrine. Analog data were digitized using standard data-acquisition software. Beat-to-beat systolic and diastolic pressures and R-R intervals were extracted, synchronized, and entered into a spread-sheet. Arterial pressures and heart rates for the 10 beats preceding the injection were calculated. Stroke volumes were measured and cardiac outputs and total peripheral resistances were calculated. All variables are reported at their maximum within the first arterial pressure ramp. Delta responses from baseline were calculated. In addition, regression analyses were performed to calculate individual baroreflex slopes of R-R intervals versus systolic pressures during the postinjection pressure increase.

Tyramine. Hemodynamic measurements were made similarly to those during the phenylephrine injections.

Biochemistry. Lymphocytes were isolated, and β2-adrenergic receptor sensitivity was assayed according to previously published methods (27). Arginine vasopressin was measured with a double-antibody radioimmunoassay based on the method of Glick and Kagan (14). Endothelin (121) was measured by an ELISA kit from Biomedica Gruppe (Vienna, Austria). Plasma norepinephrine and epinephrine levels were measured by radioenzymatic assay (21). Chromogranin A was measured by the method of Stridsberg (32). DHPG was measured by HPLC with electrochemical detection (HPLC-EC) (16). The HPLC-EC assay also elutes a norepinephrine peak, and DHPG levels were standardized to the norepinephrine peak.


All results are presented as means ± SE. All data were tested for normalcy and equal variance using the Kolmogorov-Smirnov test and the Levene Median test, respectively. Astronauts were separated into a presyncopal or nonpresyncopal group based on the occurrence of presyncope (those who could not maintain 10 min of upright posture) on landing day. Differences were analyzed by using two-way repeated-measures analysis of variance. The effects of interest were group (presyncopal or nonpresyncopal) and day (preflight, landing day, and 3 days after landing). The Tukey test for multiple comparisons was performed to document differences when there were significant main effects. For all tests, significance was set at P ≤ 0.05.


Incidence of Presyncope

Ten of the 23 astronauts in this study, including both of the females, became presyncopal during the tilt test on landing day. For all analyses, subjects were separated into presyncopal and nonpresyncopal groups. Figure 1 presents a representative presyncopal subject (subject 1) during preflight (A) and landing day (B) tilt tests. Preflight, this subject maintained arterial pressure throughout the 10 min of upright tilt (horizontal arrow). The upright heart rate was maintained between about 70 and 80 beats/min. On landing day, this subject could remain upright for only about 6 min (horizontal bar). Systolic, diastolic, and pulse pressures fell steadily until it became necessary to terminate the test. Upright heart rate climbed to a high of ∼115 beats/min. This subject did not become nauseated but experienced lightheadedness at the end of the test. This arterial pressure pattern is similar to that of eight of the other presyncopal subjects. Figure 2 depicts preflight and landing day data from the 10th subject (subject 2), who was the only one to experience frank vasovagal syncope on landing day. It is the only such occurrence we have ever recorded in several hundred stand/tilt tests on landing day. In this subject, there was a sudden drop in heart rate to 20 beats/min, followed by a string of premature ventricular contractions and spontaneous recovery. Figure 2, C–E, presents the plasma norepinephrine, epinephrine, and arginine vasopressin levels in this subject. The epinephrine was the highest we have ever measured on landing day. The norepinephrine was at the level of the nonpresyncopal subjects, and the vasopressin was the highest measured in this study. These responses occurred within 30 s of the upright posture. Because the responses of this subject were so different from those of the other presyncopal subjects, his data were not grouped with the others.

Fig. 1.

Original record showing arterial pressure and heart rate [HR; in beats/min (bpm)] in a single subject. A: recorded before flight; B: recorded on landing day. Horizontal bars indicate the period of upright tilt. Before flight, arterial pressure was stable and the subject completed 10 min in the upright position, indicated by the arrow. On landing day arterial pressure was not maintained, and the test was terminated after 390 s upright. BP, blood pressure.

Fig. 2.

Responses to upright tilt in a different subject from the one represented in Fig. 1. A and B: arterial pressure and HR before flight (A) and on landing day (B). Horizontal bars indicate the period of upright tilt. Before flight, arterial pressure was stable and the subject completed 10 min in the upright position. On landing day, this subject experienced vasovagal syncope. CE: supine and upright norepinephrine (C), epinephrine (D) and arginine vasopressin (E) before flight (open circles), on landing day (solid circles), and 3 days after landing (shaded circles). tyr 1, 2.0 mg/1.73 m2 body surface area (BSA) of tyramine; tyr 2, 4.0 mg/1.73 m2 BSA of tyramine.

Table 1 presents hemodynamic data from the tilt tests before flight, on landing day, and 3 days after landing in presyncopal and nonpresyncopal subjects. There were no significant intergroup hemodynamic differences before flight, although upright total peripheral resistance and systolic pressure tended to be higher in the nonpresyncopal group (P = 0.07, P = 0.08, respectively). On landing day, presyncopal subjects had lower upright vascular resistance and systolic and diastolic pressures than the nonpresyncopal subjects. Within groups, supine systolic and diastolic pressures were significantly higher than preflight in the nonpresyncopal group, and their upright pressures tended to be higher than preflight, although the differences were not significant.

View this table:
Table 1.

Hemodynamic responses to tilt

Responses to Phenylephrine

Figure 3 depicts hemodynamic responses to intravenous phenylephrine injections. Before flight, increases in systolic pressure and total peripheral resistance were smaller in the astronauts who would become presyncopal on landing day than in those who would not (P = 0.02 for systolic, P = 0.03 for resistance). On landing day, as well as 3 days after landing, responses to phenylephrine were not different from those before flight in either group. Intergroup differences in responses showed the same trends as preflight but were not significant.

Fig. 3.

Maximum responses of systolic (A) and diastolic pressure (B) and peripheral resistance (C) in response to intravenous injections of phenylephrine (0.26 mg/1.73 m2 BSA) in presyncopal (hatched white bars) and nonpresyncopal (hatched shaded bars) subjects before flight, on landing day, and 3 days after landing. *P ≤ 0.05.

Cardiac baroreflex responses during phenylephrine injections were as follows. In presyncopal subjects, slopes preflight, on landing day, and 3 days after landing were 28 ± 9, 29 ± 8 (P = 0.93, from preflight), and 44 ± 7 ms/mmHg (P = 0.05, from preflight), respectively. Baroreflex slopes in nonpresyncopal subjects preflight, on landing day, and 3 days after landing were 43 ± 19, 26 ± 10 (P = 0.54, from preflight), and 21 ± 7 (P = 0.94, from preflight) ms/mmHg, respectively. There were no intergroup differences before flight or on landing day, but 3 days after landing, baroreflex slopes of the presyncopal subjects were larger than those of the nonpresyncopal subjects (P = 0.001).

Norepinephrine, DHPG, and Epinephrine

Figure 4A depicts plasma norepinephrine levels before flight, on landing day, and 3 days after landing in presyncopal (left) and nonpresyncopal (right) subjects. On landing day, nonpresyncopal subjects had significantly greater plasma norepinephrine levels at baseline, at both tyramine doses, and during tilt than they had either preflight or 3 days after landing. On landing day, presyncopal subjects did not have greater norepinephrine levels at baseline or during tilt but did have greater levels at the low dose of tyramine than they had had preflight. Nonpresyncopal subjects exhibited a significantly greater upright plasma norepinephrine than the presyncopal subjects.

Fig. 4.

Plasma norepinephrine (A) and dihydroxyphenylglycol (DHPG) (B) levels in 9 presyncopal and 13 nonpresyncopal subjects. Norepinephrine values are preflight (open circles) on landing day (solid circles), and 3 days after landing (shaded circles) at baseline, tyramine injections, and upright tilt. DHPG levels were only measured in the supine position, preflight (open bars), and on landing day (solid bars). tyr 1, 2.0 mg/1.73 m2 BSA of tyramine; tyr 2, 4.0 mg/1.73 m2 BSA of tyramine. *P ≤ 0.05, landing day vs. preflight; †P ≤ 0.05, landing day vs. 3 days after landing; ‡P ≤ 0.05, presyncopal vs. nonpresyncopal on landing day.

Figure 4B depicts resting DHPG levels. DHPG levels were significantly greater on landing day in the presyncopal but not in the nonpresyncopal group.

Figure 5 represents norepinephrine release with tyramine and tilt. Before flight, there were no differences in norepinephrine release between presyncopal and nonpresyncopal subjects at either dose of tyramine. After flight, presyncopal subjects, but not nonpresyncopal subjects, had significantly greater norepinephrine release than they had had preflight at the low (P ≤ 0.05) dose of tyramine and a similar trend at the higher dose. With tilt, presyncopal subjects had the same release of norepinephrine before and after flight. Nonpresyncopal subjects had a response on landing day that was greater than that before flight (P ≤ 0.05) and also greater than that of the presyncopal subjects on landing day (P ≤ 0.05).

Fig. 5.

Plasma norepinephrine responses to intravenous tyramine and upright tilt in presyncopal (A, n = 9) and nonpresyncopal (B, n = 13) subjects, preflight (open bars), on landing day (solid bars), and 3 days after landing (shaded bars). Baseline norepinephrine values used to calculate the changes for the tyramine injections as well as for the tilt are the supine norepinephrine values shown in Fig. 4. tyramine 1 = 2.0 mg/1.73 m2 BSA; tyramine 2 = 4.0 mg/1.73 m2 BSA. *P ≤ 0.05, within group; †P ≤ 0.05, between groups.

Figure 6A depicts supine and upright plasma epinephrine levels before flight, on landing day, and 3 days after landing. Before flight, there were no differences in responses between groups. On landing day, the presyncopal subjects had upright levels that were significantly higher than their own preflight as well as those of the nonpresyncopal subjects.

Fig. 6.

Plasma epinephrine (A) and arginine vasopressin (B) levels in presyncopal (n = 8) and nonpresyncopal (n = 11) astronauts preflight (open circles) on landing day (solid circles), and 3 days after landing (shaded circles). *P ≤ 0.05, landing day vs. preflight; †P ≤ 0.05, landing day vs. 3 days after landing; ‡P ≤ 0.05, presyncopal vs. nonpresyncopal on landing day.

Arginine Vasopressin

Figure 6B depicts supine and upright plasma arginine vasopressin levels before flight, on landing day, and 3 days after landing. Before flight, there was very little vasopressin release with tilt in either group, but on landing day the presyncopal group had upright levels that were significantly higher than their own preflight as well as those of the nonpresyncopal subjects.

Hematocrit, Blood Volume, Endothelin, Chromogranin A, and β2-Adrenoreceptors

Table 2 presents hematocrit, blood volume, endothelin, chromogranin A, and β2-adrenoreceptor data before flight, on landing day, and 3 days after landing. Upright hematocrits were greater than supine hematocrits in both groups, and on all occasions there were no group or day effects. Plasma volumes were significantly reduced on landing day in both groups, but there was no significant intergroup difference on any day. There were no group or day effects for red blood cell volume, endothelin, or chromogranin A. Lymphocyte β2-adrenoreceptors showed only a day effect; sensitivity was significantly reduced 3 days after landing compared with preflight.

View this table:
Table 2.

Hematocrit, blood volume, endothelin, chromogranin A, and β2-adrenergic receptor sensitivity preflight, on landing day, and 3 days after landing in presyncopal and nonpresyncopal astronauts


This study sought to determine the mechanisms of the low vascular resistance and the hypoadrenergic responses that have been reported in astronauts who become presyncopal on landing day. There are important new findings from this study. First, astronauts who are destined to become presyncopal on landing day can be identified before flight by their low pressor responses to phenylephrine. On the other hand, astronauts, both presyncopal and nonpresyncopal, do not have pressor responses to phenylephrine on landing day that are different from preflight. Second, norepinephrine release in response to tyramine injections is not reduced on landing day, suggesting that the absence of adequate norepinephrine release with tilt is not due to inadequate synthesis or availability in the sympathetic nerve terminals. Third, the high vasopressin and epinephrine responses to tilt in presyncopal astronauts suggest that the afferent limb of the baroreflex arc is not disrupted by spaceflight. Therefore, we suggest that central integration of baroreflex afferent input becomes dysfunctional as a result of spaceflight in a subset of astronauts. Taken together, these results offer the possibility of predicting susceptibility to postflight orthostatic hypotension before flight and also shed new light on the mechanisms of presyncope on landing day.

Patterns of Presyncope

In the course of performing several hundred stand and tilt tests within hours of shuttle landings, we have documented the same arterial pressure pattern shown in Fig. 1 in every presyncopal subject until now: that is, arterial pressure which declines from the time of upright posture, until it falls to the point at which the test must be terminated, at times ranging from 2 to 9 min (12, 33). The incident of frank vasovagal syncope depicted in Fig. 2 is the first such occurrence we have seen on landing day and is thus notable. During that particular tilt test, the epinephrine release was the highest we have ever reported on landing day, and the plasma norepinephrine level was more in the range of the nonpresyncopal subjects than of the other presyncopal subjects, even after only 30 s of tilt. In addition, the release of arginine vasopressin was higher than that of any other subject, either presyncopal or nonpresyncopal. This hyperadrenergic syncope is the rare occurrence after spaceflight. Most presyncopal astronauts had hypoadrenergic responses. Only this one astronaut had vasovagal bradycardia with hyperadrenergic activity.

α1-Adrenergic Receptor Responsiveness

These new data show that pressor responses to the α1-adrenergic agonist phenylephrine are not reduced by space-flight but are actually low before flight in astronauts who will become presyncopal on landing day. Because cardiac baroreflex responses in this study showed no group or day effect, the intergroup differences in the pressor responses could not be due to differences in baroreflex-mediated bradycardia, which might limit the pressure excursions. Therefore, we suggest that there is a primary difference between groups in α1-adrenergic receptor responsiveness. This finding is in direct accord with our previous reports that, before flight, presyncopal astronauts have low vascular resistance but do not have low plasma norepinephrine levels (12, 33).

Other investigators have suggested that α-adrenergic function and vascular responses may be impaired by spaceflight (6, 34) and thus contribute to orthostatic hypotension. Our data show that α1-adrenergic receptor responses to phenylephrine injections, in the supine position, are not affected by space-flight. However, they do not preclude the possibility that α-adrenergic receptor-mediated vasoconstriction may be compromised when vascular smooth muscle cells are contracting against a full load, as during the upright tilt (26). They also do not rule out the possibility that α2-receptor hyporesponsiveness or a change in α-receptor myogenic vasoconstrictor interactions (23) may contribute to low vascular resistance after spaceflight.

In the present study, the preflight trend toward lower vascular resistance in presyncopal subjects did not reach significance as it had in our prior reports (12, 33). However, there were only two women in this study. Because we (33) have shown previously that presyncopal men have higher vascular resistance than have presyncopal women, the failure to reach significance could be due to the small number of female subjects.

Cardiac-Baroreflex Responses

Baroreflex responses during phenylephrine injections were not different between groups on any test day and were not changed from preflight to landing day. This may seem incongruent with our prior reports that baroreflex function is reduced after spaceflight (10, 11). However, phenylephrine provides only a hypertensive stimulus. In the previous studies, both hypotensive and hypertensive stimuli were delivered in the form of positive and negative pressure applied to the carotid baroreceptors. In those reports, most of the loss of baroreflex function was from the hypotensive buffering portion of the response. The hypertensive buffering capacity remained intact. Thus the fact that R-R interval responses to phenylephrine-induced pressure increases were not different after spaceflight is not surprising. Differences between groups may have been elucidated if we had added a hypotensive stimulus, such as a nitroprusside infusion. It is not clear why presyncopal subjects had an increase in baroreflex sensitivity during phenylephrine injections 3 days after landing. The pressor responses to the injections were not greater on that day.

Norepinephrine Responses

The presyncopal astronauts in this study did not have adequate sympathetic responses on landing day. Not only did they not respond to upright posture, they also did not have the expected increase in baseline levels of norepinephrine that should result from their lower blood volume (Fig. 4). An important new finding from this study is that those low norepinephrine responses are not the result of low norepinephrine stores. The amount of norepinephrine released with the tyramine injections actually was significantly greater on landing day than preflight in presyncopal subjects (Fig. 4). Furthermore, the increased baseline levels of DHPG, the intraneuronal norepinephrine metabolite of monoamine oxidase, further suggest that, in presyncopal subjects, there is abundant norepinephrine production. Basal efflux of DHPG is taken to reflect norepinephrine that has translocated from the vesicles to the neuroplasm and been deaminated by monoamine oxidase (15). It may also reflect norepinephrine transporter function.

We believe that the most likely explanation for this finding is the following scenario. During weightlessness, the central nervous system receives no baroreceptor input to counter upright posture. This causes a loss of the normal efferent sympathetic activity associated with upright posture. In the presyncopal subset of astronauts, this loss of hypotensive afferent input results in a central remodeling, such that the capability for proper integration and initiation of appropriate sympathetic efferent responses to hypotensive input is lost. Probably not coincidentally, this subset of astronauts is the same group who does not maintain high levels of vascular resistance even before flight. During this inflight adaptation, the reduction in efferent sympathetic activity does not appear to reduce norepinephrine biosynthesis. We suggest that norepinephrine builds up in the neuronal vesicles, spills over into the axoplasm, and is quickly metabolized by monoamine oxidase to DHPG, thus preventing feedback inhibition. The lack of any preflight to postflight change in chromogranin A (which is coreleased with norepinephrine but has a long half life) (Table 2) suggests that norepinephrine efflux is not reduced or increased inflight, and absolute baseline norepinephrine levels are not reduced on landing day. Thus norepinephrine biosynthesis, norepinephrine stores, and baseline norepinephrine efflux are not reduced by spaceflight in these presyncopal astronauts. In fact, the markedly and significantly greater increase in tyramine-induced norepinephrine levels on landing day in presyncopal subjects (Fig. 5) suggests that spaceflight may have increased the tyramine releasable pool. However, norepinephrine responses that depend on intact central integration of baroreceptor input and intact efferent sympathetic electrical signals are severely affected by spaceflight in these individuals.

Recent data from hindlimb-suspended rats support this idea. After suspension, there is a change in central integration of baroreflex input (29). Those authors measured afferent aortic depressor nerve traffic versus efferent renal sympathetic nerve traffic during pharmacologically induced arterial pressure changes. They reported a significant attenuation of the gain of the efferent versus afferent traffic after hindlimb suspension and suggested dysfunction of central integration of baroreceptor input (29). In a follow-up study, they report enhanced GABA-mediated inhibition of efferent sympathetic traffic in the rostral ventrolateral medulla (RVLM) after hindlimb suspension (28). It is possible that similar changes may explain our findings in humans.

To summarize the sympathetic findings, returning shuttle astronauts divide into two groups during tilt testing. The largest group of predominantly male astronauts does not develop presyncope, has increased norepinephrine levels, and has increased peripheral vascular resistance. The second group has more women. They fail to increase norepinephrine appropriately on landing day and become presyncopal. One other study of six returning astronauts found that none developed presyncope during tilt testing on landing day and all had increased sympathetic nerve electrical activity (22). Those findings are not surprising because all of those six astronauts were male. They responded to tilt testing like the 13 nonpresyncopal male astronauts we tested. Because that study used microneurography as a measure of sympathetic activity, we can now say that nonpresyncopal astronauts have increased sympathetic nerve electrical activity that releases extra norepinephrine, enhances peripheral vascular resistance, and helps maintain arterial pressure during upright posture.

Vasopressin Responses

Arginine vasopressin responses to upright tilt were significantly greater in presyncopal than in nonpresyncopal subjects on landing day. This is not an unexpected finding but may be an important clue in dissecting out the mechanism of their hypotension.

Vasopressin has many sites and mechanisms of action. It is synthesized in the supraoptic and paraventricular nuclei (SON and PVN) of the hypothalamus and secreted by the posterior pituitary. Vasopressin responds to osmoreceptor input but also is released in response to decreases in cardiopulmonary or arterial baroreceptor input with decreases in blood volume or pressure (20). Cardiopulmonary and arterial baroreceptor afferents have their primary termination point in the nucleus of the solitary tract (NTS) in the medulla, and there are extensive ascending catecholaminergic tracts from the NTS to the SON and PVN (18, 20). This is the pathway by which vasopressin release is stimulated in response to baroreceptor input. Vasopressin is known to surge during severe orthostatic stress (7, 13, 31). This release in our presyncopal subjects offers insight into the mechanisms of their orthostatic hypotension. It suggests that the afferent portion of the baroreflex arc is intact.

There also are extensive descending vasopressinergic pathways from the SON and PVN back to the NTS, such that vasopressin acts a central neuromodulator. Vasopressin released into the circulation also acts as a central modulator. An important central effect of vasopressin is a marked inhibitory effect on sympathetic vasomotor centers and reduction of efferent sympathetic activity (17, 19, 30). Thus the increase in vasopressin in presyncopal subjects could actually contribute to their low upright norepinephrine levels.

In addition to the effects of vasopressin via the descending pathways from the SON and PVN, circulating vasopressin also has effects on the central integration of baroreflex function. The primary site of this action is the area postrema, a circum-ventricular organ in the brain stem, devoid of a blood-brain barrier, which has vasopressin receptors, and projections to the NTS, the dorsal motor nucleus of the vagus, the lateral parabrachial nucleus, and the ventrolateral medulla (18). These effects also act to reduce efferent sympathetic traffic. The projections to the ventrolateral medulla may have special significance for the rat data mentioned above. The increased GABA-ergic inhibition of efferent sympathetic activity in the RVLM in the suspended rats may be due at least in part to the increased vasopressin. To the extent that these findings relate to humans after spaceflight, the increased vasopressin may provoke GABA-ergic inhibition of efferent sympathetic activity in humans.

Although vasopressin is a potent vasoconstrictor, it does not raise arterial pressure at physiological concentrations due to its central inhibitory actions (9, 20). Plasma concentrations must reach much higher levels than those measured in this study, such as those reached during severe hemorrhage, to increase arterial pressure (2, 8). Thus the release of vasopressin into the circulation does not prevent or reverse orthostatic hypotension.

Our findings in these presyncopal subjects (high vasopressin, inadequate norepinephrine with upright tilt) are similar to those shown in diabetic patients with autonomic neuropathy (6). The authors of that study interpreted their data in the following way. The high vasopressin indicated that the baroreceptor afferent neurons, as well as ascending tracts to the SON and PVN in the hypothalamus, were intact, whereas the low norepinephrine suggested that efferent neurons were impaired (6). Our data suggest a central efferent defect.

These findings can be summarized in the following way. The fact that norepinephrine release in response to tyramine injections is not lower on landing day than before flight indicates that, in presyncopal subjects, synthesis and storage of norepinephrine in the sympathetic nerve terminals is intact. The high vasopressin release with upright posture suggests that the afferent limb of the baroreflex, all the way up to the SON and PVN is intact. Thus, the low upright norepinephrine release must indicate that the “lesion” is either central or efferent and is possibly aggravated by the high vasopressin.

Epinephrine Responses

In addition to enhanced vasopressin release, the presyncopal subjects in this study also had epinephrine release on landing day that was significantly higher than both that of nonpresyncopal subjects and that of themselves before flight (Fig. 6). Release of catecholamines from the adrenal medulla can provide a safety factor by substituting for low norepinephrine release from the sympathetic nerves. However, epinephrine has less of an effect on raising resistance and more of an effect on cardiac stimulation than does norepinephrine. In fact, epinephrine may even have a vasodilatory effect on the vasculature (31). Thus the response seems to have little effect in protecting blood pressure.

Changes in Hematocrit with Tilt

Hematocrit increased with tilt on all occasions in both groups, and tilted values were never different between groups. This indicates that there is no difference in transcapillary filtration during upright tilt after spaceflight.

β-Adrenergic Receptors

It is interesting that the lymphocyte β2-adrenoreceptors are significantly desensitized (irrespective of group) 3 days after landing but not on landing day. We suggest that catecholamine levels, even in the nonpresyncopal subjects, were not increased until the stress of landing occurred. Perhaps the catecholamine exposure had not been long enough by the time our measurements were taken on landing day to effect a significant change (3). However, catecholamine levels probably are high in the days after landing. Presyncopal subjects recover rapidly, and by postflight day 3 most are able to complete the tilt test without incident with adequate norepinephrine levels. Therefore, we suggest that, in both groups, the low blood volume and hectic activity following landing generate higher levels of sympathetic activity than preflight and cause the downregulation of these receptors. This may not be reflected in our laboratory measurements of norepinephrine, because of the 20-min quiet rest preceding the blood draw. The postflight desensitization of the receptors was not reflected in heart rates 3 days after landing (Table 1). Although β2-adrenoreceptors on lymphocytes parallel those in the heart (4) and are known to have chronotropic effects (1, 5), the status of the β1-receptors, which also have important chronotropic effects, are not known in this study. In addition, it is likely that vagal activity is somewhat withdrawn during this time, contributing to the slightly higher heart rates. These data, in conjunction with the unchanged responses to phenylephrine, indicate that changes in peripheral adrenergic receptor function probably are not a significant factor in defining the mechanisms of postspaceflight orthostatic hypotension and presyncope. This supports our hypothesis that the “lesion” is in the ability of the central nervous system to respond appropriately to incoming information from the baroreceptors.

It is clear that postflight orthostatic hypotension is severe in some astronauts and virtually nonexistent in others. The lingering question is, why. Although the more susceptible group does start out at somewhat of a disadvantage with the low vascular resistance and low vascular responsiveness, that does not explain why spaceflight seems to affect them so adversely. We (33) reported earlier that women are more susceptible to postflight orthostatic hypotension than men and offered possible mechanisms, but in that study, and in the present study, there are a large number of males who also become presyncopal. There still is no complete explanation for their susceptibility. However, after months of weightlessness, 80% of male astronauts become presyncopal on landing day (25), suggesting that after prolonged exposure to weightlessness, most astronauts' brains will “forget” how to cope with gravity.


These findings have obvious importance for National Aeronautics and Space Administration. If postflight orthostatic hypotension can be predicted preflight, it offers the possibility of identification of appropriate, individually specific countermeasure development. This laboratory is currently evaluating the efficacy of the peripheral α-agonist midodrine as a countermeasure for postflight orthostatic hypotension. If susceptible crew members have been identified preflight, the countermeasure can be given before reentry to provide protection in anticipation of the first onset of gravitational forces during landing.

In addition to the flight-related significance, this study also offers important insight into the subtle differences in arterial pressure control strategies among healthy, normotensive people. We found, in our course of studies of the effects of spaceflight, that those who operate with low resistance are more susceptible to loss of arterial pressure control after spaceflight. Whereas those with higher resistance are protected from presyncope on landing day, it is unknown if, in the long term, they may be more susceptible to the development of hypertension or other cardiovascular diseases.

On a more basic level, our observations offer a unique perspective on the physiology of blood pressure control and stress responses. The sympathetic nervous system operates via the baroreflex loop to maintain blood pressure in response to changes in posture and blood volume. It also participates in the “fight or flight” response to stress. In a weightless environment there is decreased need for baroreflex-mediated sympathetic responses. One might expect decreased activity of the baroreflex components of the sympathetic nervous system to occur as a means of conserving resources. It is an expensive system to maintain when there is no need. We suggest this may occur in some subjects, who are then, on return to Earth, unable to reactivate the system for several days. On the other hand, during a spaceflight mission, stress responses with release of epinephrine from the adrenal medulla are most certainly called into play and seem to be maintained. The postural baroreflex response of the sympathetic nervous system is such an unvarying human response that it seems “hardwired” into the brain circuitry. However, exposure to weightlessness reveals that the circuit has plasticity that can decrease responses of brain efferent pathways that stimulate the sympathetic nerves.


We are indebted to the astronauts who participated in this study. We also thank Dr. Dominick D'Aunno for invaluable assistance in the preparation and administration of the pharmacological agents during all testing.

Present address for W. W. Waters: Wyle Laboratories, 1290 Hercules, Houston, TX 77058.


This research was funded by National Aeronautics and Space Agency Grant NASA Research Announcement 96-OLMSA-01-051 and partially supported by National Institutes of Health Grant M01RR-00827.


  • 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