| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Harvard-MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge 02139; 2Endocrinology-Hypertension Division, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston 02115; 4Department of Anesthesia and Critical Care, Massachusetts General Hospital, Boston, Massachusetts 02114; and 3Department of Cardio-Thoracic Surgery, McGill University, Montreal, Quebec, Canada H3G 1A4
Submitted 11 June 2003 ; accepted in final form 17 September 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
sympathetic; parasympathetic; syncope; bed rest; autonomic function
In this study, we applied the cardiovascular system identification (CSI) technique to quantify sympathetic and parasympathetic responsiveness with regard to HR regulation. The CSI technique (29, 30) evaluates the relation between second-to-second fluctuations in physiological signals, such as HR, arterial blood pressure (ABP), and instantaneous lung volume (ILV), to enable dynamic assessment of important physiological mechanisms without perturbing normal system operation. CSI assumes that the couplings of the small spontaneous fluctuations in these signals about their mean values may be represented by a linear, time-invariant model. This model will generally change when there is a change in physiological state. The linearity assumption has been tested and validated (10).
Figure 1 depicts the closed-loop model of HR and blood pressure regulation. The model consists of four couplings, circulatory mechanics, HR baroreflex, ILV 
HR, and ILV ABP; their transfer properties are represented below in terms of impulse response functions. Impulse response is defined by the response of the coupling mechanism to an input at time 0 with unit area and arbitrarily short duration.
|
Circulatory mechanics impulse response represents the ABP wavelet generated with each cardiac contraction. The input to circulatory mechanics, the pulsatile HR (PHR), is defined as a train of impulses occurring at the times of contraction of the ventricle. It may be constructed from the times of the occurrence of the QRS complexes in the ECG. Circulatory mechanics primarily reflect the mechanical properties of the heart, great vessels, and peripheral circulation. The amplitude of the circulatory mechanics impulse response is related to stroke volume, and the characteristic decay time of this impulse response is related to the product of peripheral resistance and arterial compliance. The HR baroreflex represents the change in HR in response to an impulse in ABP mediated via the HR baroreflex. The ILV HR impulse response reflects the change in HR in response to a very rapid inspiration/expiration; this impulse response is centrally mediated via the autonomic nervous system. ILV ABP represents the mechanical effects of respiration on ABP mediated in part by the effects of intrathoracic pressure on venous return.
In Fig. 1, there is a fifth predefined coupling, sinoatrial (SA) node, which relates HR tachogram (HRT) and PHR. HRT is defined to be a stepwise continuous process; its value corresponds to the reciprocal of the current interbeat interval for the time corresponding to the duration of that interval (3). The SA node is an "integrate-and-fire" device and is not identified from the experimental data, because its dynamics are predefined. NHR and NABP represent the fluctuations in HR and ABP, respectively, that are not attributable to the couplings discussed above (30). NHR may reflect, for example, perturbations to HR resulting from cerebral inputs, and NABP may represent perturbations to ABP resulting from autoregulation of local vascular resistance in different tissue beds. In Fig. 1 the power spectra of NHR and NABP are presented.
The CSI technique models the couplings between each of the physiological signals by two linear time-invariant autoregressive exogenous input (ARX) difference equations (24)
 | (1) |
 | (2) |
On the basis of the CSI technique, we further developed (see APPENDIX) and applied a new method to quantify separately parasympathetic and sympathetic responsiveness in the regulation of HR. As mentioned previously, widely used signal-processing techniques, such as HR spectral analysis, for autonomic function evaluation have generally not been able to effectively separate parasympathetic and sympathetic function (1). Our method was developed and validated by utilizing animal and human experimental data. Our quantification of autonomic responsiveness pertains only to autonomic tone in regulation of the SA node (HR), which may be different from that in regulation of other tissue beds. However, for simplicity, we refer to this type of autonomic responsiveness simply as parasympathetic and sympathetic responsiveness.
The objective of this work is to study the effect of simulated microgravity on cardiovascular autonomic control and its association with OI. Given the pivotal role of the autonomic function in maintaining blood pressure and normal hemodynamics, we tested the following hypotheses: 1) autonomic function is significantly altered after exposure to simulated microgravity and 2) autonomic responsiveness is associated with incidence of OI. In addition, we further evaluated the possibility of predicting microgravity-induced OI using only baseline measures. We utilized a prolonged head-down-tilt bed-rest protocol as a model of ground-based microgravity (6).
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
Bed-rest study protocol. Subjects were admitted to the hospital for a 3-day (subjects 1-4) or 5-day (subjects 5-29) pre-bed-rest ambulatory period, during which baseline testing was done and an isocaloric diet was maintained (200 meq sodium, 100 meq potassium, 2,500 ml fluid). Then subjects underwent 9 (subject 1), 14 (subjects 2-4), or 16 (subjects 5-29) days of 4° head-down-tilt bed rest with the same diet. Finally, 2 (subjects 1-4) or 3 (subjects 5-29) days of the post-bed-rest period were scheduled for the recovery, when the subjects were allowed ad libitum activity but continued the constant diet.
Throughout the inpatient course, the subjects maintained a constant light-dark cycle. Subjects 1-21 participated in a non-sleep-deprivation protocol (16 h light-8 h darkness), while subjects 22-29 participated in a sleep-deprivation protocol with a shorter period of darkness (17.9 h light-6.1 h darkness). In addition, routine vital signs were monitored every 8 h, and daily weights were recorded.
At the end of bed rest, midodrine was given to seven subjects in a randomized blinded fashion before the tilt-stand test (see Orthostatic tolerance testing) as part of a parallel study on countermeasures of OI (34). It was hypothesized that midodrine causes venous and arteriolar constriction, thus increasing venous return and ABP. Because administration of midodrine may modify the hemodynamic response to the tilt-stand testing, standing data from these subjects at the end of bed rest were excluded from the analysis.
Orthostatic tolerance testing. On the last day of the pre-bed-rest phase and the last day of the bed-rest phase (end-bed-rest day), a tilt-stand test was used to determine orthostatic tolerance. After baseline data in the supine position were recorded, the subjects were tilted upright on a tilt table to 30° for 
10 min. The angle of tilt was then increased to 60° for 10 min. Finally, the subjects were tilted to an upright posture. After each posture was attained and a hemodynamic steady state was reached, data for CSI analysis were collected (see Data collection for CSI). The test was immediately terminated, and the subject was returned to the supine position if there was evidence of a sudden drop in blood pressure and/or difficulty in appropriately responding to questions, i.e., mental status changes consistent with presyncopal symptoms. After the CSI testing in the upright posture, the subjects were allowed to move their legs or walk as needed while maintaining the upright posture for 120 min and being closely monitored for presyncopal symptoms.
Data collection for CSI. In the supine position before the tilt-stand test and in each tilt posture, 8 min of standard surface ECG, ABP, and ILV data were recorded continuously and noninvasively to facilitate offline CSI analysis. Continuous blood pressure was recorded from the middle finger of the right or left hand with a fingertip cuff transducer (Portapres, TNO, or Finapres, Ohmeda). ILV was measured with a two-belt chest-abdomen inductance plethysmograph (Respitrace, Ambulatory Monitoring Systems). To calibrate ILV, the subject alternately filled and emptied an 800-ml calibrated Spirobag. During data collection, the subjects were instructed to breathe in response to auditory cues spaced at random intervals ranging from 1 to 15 s with a mean of 5 s. This random breathing protocol broadens the spectral content of ILV, thereby facilitating system identification while preserving normal ventilation (4).
Data analysis. Approximately 6-min segments of ILV, ABP, PHR, and HRT (derived from ECG) were analyzed to determine the impulse response functions and noise spectra in the model in Fig. 1 (see Ref. 30 for details about solving Eqs. 1 and 2). The data were decimated from 100 to 1.25 Hz. Zero-mean ABP and HRT were normalized with respect to their corresponding time-averaged values and zero-mean ILV with respect to its standard deviation. In accordance with the previous CSI studies (29, 30), the impulse response curves were each characterized with two parameters: peak amplitude and characteristic time. These parameters are defined as
 |
 |
For the two autonomically mediated couplings, HR baroreflex and ILV HR, the peak amplitudes are determined mainly by parasympathetic responsiveness in HR regulation, and the characteristic time reflects the sympathetic-parasympathetic balance, because characteristic times are much longer for sympathetically than for parasympathetically mediated impulse response functions (30).
The perturbing noise sources are presented in terms of their power spectra. Three parameters were calculated to characterize the power spectra: total power (0-0.5 Hz), low-frequency (LF) power (0-0.15 Hz), and high-frequency (HF) power (0.15-0.5 Hz) (30). It is well accepted that the LF component in HR and ABP spectra is usually mediated by parasympathetic and sympathetic systems, whereas the HF component is mediated by the parasympathetic system (39).
In addition, the impulse response of ILV HR was analyzed further to obtain explicit, separate quantification of parasympathetic and sympathetic responsiveness in HR regulation. The APPENDIX introduces the quantification technique in detail.
Statistical analysis. To test the hypothesis stated in the introduction, univariate comparisons between the impulse response parameters from pre- and end-bed-rest data and between supine and standing posture for each subject were conducted by using the paired t-test (35). Unpaired t-tests were used to compare parameters of subjects who did and did not experience OI during tilt-stand testing. P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
Supine vs. standing. The impulse response parameters calculated from data obtained in supine and standing postures are shown in Table 1. The peak amplitudes of the HR baroreflex, ILV HR, and circulatory mechanics impulse response functions were significantly decreased during standing. The peak amplitude of the ILV ABP increased, but not at a statistically significant level (P < 0.1). The characteristic times of the HR baroreflex, ILV HR, and circulatory mechanics impulse response functions were significantly increased during standing, consistent with increased sympathetic-parasympathetic balance. The total power and LF power of NHR and NABP spectra increased during standing. A direct estimation of parasympathetic and sympathetic responsiveness was calculated from the ILV HR impulse response (see APPENDIX). Consistent with the well-accepted finding in the literature (19), our results showed that parasympathetic responsiveness decreased significantly on standing whereas sympathetic responsiveness increased significantly on standing (Table 2).
|
|
Before vs. after bed rest. The impulse response parameters obtained from data collected in the supine position on the pre- and end-bed-rest days were compared. Peak amplitudes of the HR baroreflex and ILV HR impulse response functions were significantly reduced, whereas the peak amplitude of the ILV ABP impulse response function was significantly increased, after bed rest (Table 1). The characteristic time of the ILV HR impulse response function increased significantly. The characteristic time of the HR baroreflex impulse response function increased, but not at a statistically significant level (P < 0.1). The HF power of the NHR spectra decreased significantly after bed rest (Table 1). The direct calculation of parasympathetic and sympathetic responsiveness indicates that prolonged head-down-tilt bed rest impaired sympathetic and parasympathetic responsiveness (Table 2).
Table 3 lists the number of subjects who tolerated the tilt-stand testing and those who had a presyncopal event before and after bed rest. Before bed rest, 66% of patients tolerated tilt-stand testing, and 36% of patients tolerated tilt-stand testing after bed rest. Fisher's exact categorical test (35) was performed, with hypothesis H0 being that the same proportion of subjects had a presyncopal episode before and after bed rest and hypothesis H1 being that a greater proportion of subjects had a presyncopal episode after than before bed rest. The P value of the test was 0.037, indicating that bed rest significantly decreases orthostatic tolerance. The seven subjects who received midodrine (see METHODS) were excluded from the statistical analysis of tilt tolerance after bed rest.
|
Toleration of tilt-stand testing. Although monitoring of presyncopal symptoms lasted >2 h for each subject, presyncope always occurred during the tilting or the early standing period. The time to presyncope was 25.5 ± 10.8 and 22.2 ± 6.9 (SD) min before and after bed rest, respectively. Therefore, we categorize our subjects as tilt-tolerant or tilt-intolerant without a more detailed differentiation according to their presyncope-free survival times.
Before bed rest, 10 of 29 subjects were tilt intolerant. Eight subjects displayed a vasovagal pattern of tilt intolerance with an abrupt change in ABP and HR, whereas two followed a dysautonomic pattern of tilt intolerance with a gradual drop in ABP and a gradual increase in HR. After bed rest, 14 of 22 subjects were tilt intolerant (excluding the 7 subjects who received midodrine at the end of bed rest). Twelve subjects had a vasovagal pattern and two had a dysautonomic pattern of tilt intolerance.
Table 4 compares the pre-bed-rest autonomic responsiveness between the subject group that tolerated the tilt-stand testing and the group that experienced presyncopal episodes before bed rest. Parasympathetic responsiveness was higher for the tilt-intolerant group, whereas sympathetic responsiveness was lower for this group.
|
In Table 5, data collected on the pre-bed-rest day are used to compare subjects who tolerated the tilt-stand testing with those who experienced a presyncopal episode on the last day of bed rest. Therefore, this is an attempt to associate tilt tolerance after bed rest with baseline measures. Those subjects who experienced a presyncopal event after bed rest had a higher parasympathetic responsiveness and a lower sympathetic responsiveness before bed rest (at baseline). Figure 2 displays the pre-bed-rest parasympathetic and sympathetic areas of individual pre-bed-rest tilt-tolerant subjects categorized by end-bed-rest tilt tolerance.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
Supine vs. standing. It is well accepted that the postural change from supine to standing results in a relative shift from parasympathetic-dominant to sympathetic-dominant cardiovascular control (19). We assess the validity of our technique by using this physiological finding. Table 1 demonstrates a decreased parasympathetic responsiveness and a shift to sympathetic control during standing (reflected by changes in peak amplitudes and characteristic times of the HR baroreflex and ILV HR impulse responses; see Data analysis for the relation of autonomic function to the parameters in Table 1). In Table 2, parasympathetic area decreased significantly on standing before or after bed rest, whereas sympathetic area increased significantly on standing after bed rest. These results demonstrate that the CSI approach to quantifying autonomic function can track changes due to postural shift.
In addition, Table 1 demonstrates changes in the mechanically mediated coupling: circulatory mechanics. The decrease in the peak amplitude and the increase in the characteristic time with standing suggest a decrease in stroke volume and an increase in peripheral resistance, respectively, as would be expected with a change in posture from supine to standing. The LF power in the NHR power spectrum was significantly increased in the standing posture compared with the supine posture. This suggests increased sympathetic modulation of HR in the standing position (1). The NABP power spectrum in the standing posture was increased relative to the supine posture in the LF and HF bands. This may be possibly due to increased lower extremity muscle activity on standing (1, 30).
Effect of bed rest. Results in Tables 1 and 2 demonstrate that parasympathetic and sympathetic responsiveness decreased significantly after bed rest, and there was a shift to the state of sympathetic-dominant control (reflected by elongation of characteristic times in the HR baroreflex impulse response and ILV HR impulse response). Using different techniques, such as analyzing R-R interval of HR or performing Valsalva's maneuver (11, 13, 17), many researchers have shown that parasympathetic activity is impaired after simulated or actual microgravity exposure. However, there remains a lack of consensus about the changes in sympathetic activity. For example, some (36) showed a reduction in muscle sympathetic nerve activity after 14 days of bed rest, whereas others reported an increase (22, 31) in muscle sympathetic nerve activity reflex control after bed rest. In another study (37), HR power spectrum and urinary catecholamine response were analyzed. However, the two techniques resulted in different conclusions with regard to changes in sympathetic activity after bed rest. Fritsch-Yelle and co-workers (18) found subnormal increases in plasma norepinephrine level on the assumption of upright posture in returning astronauts after 16 days of spaceflight. This indicates a functional change in the neurogenic feedback loop, which includes arterial baroreceptors, brain stem, spinal tracts, and sympathetic nerves. The different results may be accounted for by the various durations of bed rest or spaceflight, the small number of subjects in some of the studies, the number of subjects who had presyncopal episodes, and, most importantly, the different techniques employed to evaluate autonomic activity. One of the advantages of the CSI approach to quantifying autonomic responsiveness is that it estimates the modulation of the efferent autonomic activity on the SA node directly without involving a reflex mechanism, thereby simplifying the interpretation of the results.
In addition, there was a significant increase in the peak amplitude of the ILV ABP impulse response after bed rest (Table 1). This change might be related to a bed-rest-induced change in respiratory mechanics, a change in the compliance of the great vessels, a change in the diastolic filling properties of the heart, and/or a redistribution of blood within the vasculature resulting from bed rest. The NHR noise power spectra showed a reduction in HF power, perhaps reflecting the decrease in parasympathetic activity. No significant change in parameters of the circulatory mechanics coupling was found.
Tilt-stand test tolerance. Our results suggest that those who tolerated the tilt-stand test before bed rest had a lower parasympathetic responsiveness and a higher sympathetic responsiveness measured on the day of the test than those who did not tolerate the test (Table 4). The difference in parasympathetic responsiveness between the two groups is statistically significant in the supine posture, whereas that of sympathetic responsiveness is significant in the upright posture, perhaps because the supine posture is a parasympathetic-dominant state, whereas the upright posture is a sympathetic-dominant state. The result shows that autonomic responsiveness is associated with tilt-stand test tolerance. Furthermore, combining this result and the fact that more subjects failed the tilt-stand test (Table 3) after bed rest and the result that bed rest impaired parasympathetic and sympathetic responsiveness, one may speculate that autonomic control, sympathetic responsiveness in particular, may be mechanistically involved in determining tilt tolerance. The relation between OI and various factors directly or indirectly related to autonomic function has been studied (8, 11, 26, 38). However, a general consensus has not been reached. Our results are consistent with those of Fritsch-Yelle et al. (18) and Waters et al. (41), who also showed a greater sympathetic response on standing on the landing day in the tilt-tolerant than in the tilt-intolerant group.
Prediction of OI. Subjects with a lower parasympathetic responsiveness and a higher sympathetic responsiveness before bed rest tended to tolerate the tilt-stand test after bed rest (Table 5, Fig. 2). These data suggest that an altered autonomic nervous system state might be related mechanistically to a predisposition to end-bed-rest OI, and they also raise the possibility of predicting end-bed-rest OI by using pre-bed-rest measures. Few studies demonstrated similar statistical significance in this aspect. Fritsch-Yelle et al. (18) and Waters et al. (41) were the first to suggest the possibility of predicting which individuals would be susceptible to OI after spaceflight. They found that baseline peripheral vascular resistances and systolic and diastolic pressures before flight were significantly lower in the postflight presyncopal group than in the other group. Because autonomic function plays a major role in regulating these hemodynamic variables, one might infer from their results a preflight intergroup difference in autonomic responsiveness. Although various factors may be involved in the development of OI after microgravity exposure, our results suggest that autonomic function might specifically play a role in this process.
To test whether CSI measures obtained before bed rest might be used to predict orthostatic tolerance after bed rest, we retrospectively constructed the following algorithm: 1) tilt-stand test-tolerant before bed rest, 2) parasympathetic area (supine) 
0.020, and 3) sympathetic area (standing) > 0.
This algorithm (Fig. 2) successfully identified 8 of the 8 subjects (100%) who were tilt tolerant after bed rest and 12 of the 14 subjects (86%) who were tilt intolerant (P < 0.005). The two misidentified subjects had a vasovagal syncope pattern. This algorithm will need to be tested prospectively in future studies. If it is possible to identify individuals at greatest risk of OI, it may be possible to target these individuals for application of countermeasures.
Most of our tilt-intolerant subjects followed a vasovagal syncope pattern (34), whereas only two subjects showed a dysautonomic pattern. Both patterns have been documented after spaceflight (8, 28). In addition, the results may have relevance to the development of vasovagal syncope on adoption of an upright posture in Earth-bound patients. During vasovagal syncope, the efferent responses are increased vagal activity, especially to the heart, and decreased sympathetic activity. Our results suggest that patients with lower sympathetic responsiveness and higher parasympathetic responsiveness may be predisposed to vasovagal syncope.
Limitations. When subjects in this study became presyncopal before the CSI data collection was complete in the upright position, they were returned to the supine position for recovery; thus only data collected before the onset of presyncopal symptoms could be analyzed. Because a significant number of subjects became presyncopal during the tilt-stand test after bed rest, this limitation has a more serious effect on end-bed-rest than on pre-bed-rest results. This might be the reason that the comparison for sympathetic responsiveness between before and after bed rest in the upright posture (Table 2) is not statistically significant.
Conclusions. In summary, we studied cardiovascular autonomic function in healthy subjects before and after 16 days of head-down-tilt bed rest. We found that bed rest impairs parasympathetic and sympathetic responsiveness. Higher parasympathetic and lower sympathetic responsiveness before bed rest identified individuals who were more susceptible to OI before and after bed rest. We proposed an algorithm to predict OI after bed rest from pre-bed-rest measures. Our findings may have significance for studying Earth-bound orthostatic hypotension as well as for designing effective countermeasures to postflight OI.
|
APPENDIX |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
HR and HR baroreflex couplings are modulated by efferent autonomic activity, certain characteristics of the impulse responses should contain information about autonomic tone. However, the key question is whether there is a way to distinctly identify the parasympathetic component and the sympathetic component. Triedman and co-workers (39) reported that the positive wave and the negative wave in the ILV HR impulse response (Fig. 1) for humans were nearly obliterated after total cardiac autonomic blockade. Berger and co-workers (5) studied the dynamic behavior of the canine cardiac pacemaker. The impulse responses (see Fig. 4, G and N, in Ref. 5) of the SA node discharge rate under pure vagal stimulation and pure sympathetic stimulation mimic the initial upright wave and the delayed negative wave in the ILV HR impulse response, respectively (Fig. 1).
Given the experimental evidence and assuming linearity of the system, we can interpret and model the ILV HR impulse response as follows (Fig. 3). The initial upright wave represents the brief increase in HR mediated by parasympathetic withdrawal as a result of a positive impulse ILV input (inspiration). The increase begins at time < 0, indicating that HR rises in anticipation of the corresponding inspiration, which reflects the time delay between central initiation of cardiorespiratory activity and the physical onset of inspiration (5, 39). The delayed negative deflection is the consequence of sympathetic withdrawal, which is slower than the parasympathetic response. Therefore, the ILV HR impulse response consists of two separable components, each reflecting the modulation of autonomic efferent activity at the SA node. If we divide the impulse response into two components at the point where it first crosses zero after reaching the peak (Fig. 3), the areas covered by the two components quantify parasympathetic and 
-sympathetic efferent activity responsiveness, respectively. There is no significant overlap in time between the parasympathetic response and the sympathetic response (see Fig. 4, G and N, in Ref. 5). Our separation approach assumes that this overlap can be neglected. Analysis of data from supine and standing postures (Table 2) helps validate this technique for separating parasympathetic and sympathetic responsiveness.
|
The HR baroreflex impulse response (Fig. 1) can also be modeled as reflecting parasympathetic and sympathetic components. However, in response to an impulse change in ABP, stimulation of parasympathetic function and suppression of sympathetic function lead to a decrease in HR. Moreover, the responses in HR induced by the two mechanisms normally overlap each other in the HR baroreflex impulse response. Therefore, we employ the ILV HR impulse response to deduce autonomic responsiveness.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
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.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONAPPENDIXGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  X. Chen and R. Mukkamala Selective quantification of the cardiac sympathetic and parasympathetic nervous systems by multisignal analysis of cardiorespiratory variability Am J Physiol Heart Circ Physiol, January 1, 2008; 294(1): H362 - H371. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Grenon, S. Hurwitz, N. Sheynberg, X. Xiao, B. Judson, C. D. Ramsdell, C. Kim, R. J. Cohen, and G. H. Williams Sleep restriction does not affect orthostatic tolerance in the simulated microgravity environment J Appl Physiol, November 1, 2004; 97(5): 1660 - 1666. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Grenon, S. Hurwitz, N. Sheynberg, X. Xiao, C. D. Ramsdell, C. L. Mai, C. Kim, R. J. Cohen, and G. H. Williams Role of individual predisposition in orthostatic intolerance before and after simulated microgravity J Appl Physiol, May 1, 2004; 96(5): 1714 - 1722. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
