Sex differences in blood pressure control during 6° head-down tilt bed rest

Abstract

Spaceflight-induced orthostatic intolerance has been studied for decades. Although ∼22% of the astronaut corps are women, most mechanistic studies use mostly male subjects, despite known sex differences in autonomic control and postflight orthostatic intolerance. We studied adrenergic, baroreflex, and autonomic indexes during continuous infusions of vasoactive drugs in men and women during a 60-day head-down bed rest. Volunteers were tested before bed rest (20 men and 10 women) and around day 30 (20 men and 10 women) and day 60 (16 men and 8 women) of bed rest. Three increasing doses of phenylephrine (PE) and sodium nitroprusside were infused for 10 min after an infusion of normal saline. A 20-min rest period separated the phenylephrine and sodium nitroprusside infusions. Autonomic activity was approximated by spectral indexes of heart rate and blood pressure variability, and baroreflex sensitivity was measured by the spontaneous baroreflex slope. Parasympathetic modulation and baroreflex sensitivity decreased with bed rest, with women experiencing a larger decrease in baroreflex sensitivity by day 30 than men. The sympathetic activation of men and parasympathetic responsiveness of women in blood pressure control during physiological stress were preserved throughout bed rest. During PE infusions, women experienced saturation of the R-R interval at high frequency, whereas men did not, revealing a sex difference in the parabolic relationship between high-frequency R-R interval, a measurement of respiratory sinus arrhythmia, and R-R interval. These sex differences in blood pressure control during simulated microgravity reveal the need to study sex differences in long-duration spaceflight to ensure the health and safety of the entire astronaut corps.

orthostatic intolerance after spaceflight and spaceflight analogs has been well documented over the past decades (8, 10, 15, 30, 53). The etiology of postflight orthostatic intolerance has yet to be clearly elucidated, but autonomic and baroreflex dysfunctions are among the contributory mechanisms (8, 15, 21). Most bed rest and spaceflight studies have shown decreased vagal activity (2, 15, 16, 22) and baroreflex sensitivity (15, 16, 22, 29, 39). Changes in sympathetic activity during and after spaceflight are more controversial; both an increase (11, 22, 26, 30) and no change (45, 62) in sympathetic activity have been observed when measured by heart rate (HR) or blood pressure (BP) variability (30, 62), catecholamine analysis (11, 26, 30), or muscle sympathetic nerve activity (45). However, most spaceflight and bed rest studies seeking to understand the mechanisms contributing to postflight orthostatic intolerance have included few, if any, women (2, 15, 29). Inclusion of female data is necessary for comprehensive insights into postflight orthostatic intolerance, since women comprise ∼22% of the astronaut corps (35).

Female astronauts are more susceptible to postflight orthostatic intolerance and presyncope than male astronauts (8, 9, 35, 64), which may be in part due to their different mechanisms of BP regulation and the effect of spaceflight on autonomic control. Women have more parasympathetically dominated autonomic control than men during physiological stress (3, 27), where orthostatic BP regulation occurs via parasympathetic withdrawal more than sympathetic activation (4, 14), such that women experience a larger increase in HR than men during cardiovascular stress (35). The opposite is true in men (3, 27), who typically have a greater sympathetic activation than women during physiological stress (37, 56), such that they respond to orthostatic challenges with greater increases in vascular resistance (35). Most women, as well as a small subset of men, develop autonomic dysfunction during spaceflight (64) and have compromised sympathetic responsiveness to orthostatic stress after flight. These sex differences in autonomic control and the prevalence of autonomic dysfunction in women after spaceflight suggest there could exist a higher degree of autonomic dysfunction in women after bed rest.

Although Custaud et al. (18) did not find sex differences in the cardiac baroreflex during bed rest, this study only considered a 7-day bed rest. Most studies (1, 7, 14, 44) agree that women have lower baroreflex sensitivity than men; however, sex differences in baroreflex function after long-duration bed rest have yet to be examined. Long-duration bed rest studies are essential, particularly with the end of the Shuttle program, since future missions to asteroids or Mars will require months to years of space travel. In this study, we examined autonomic and arterial baroreflex function over a range of pharmacologically altered arterial BPs in men and women before and during a 60-day bed rest. Continuous stepwise infusions of vasoactive drugs have been previously implemented to detect baroreflex and autonomic function abnormalities in diabetic patients (23). This methodology has also been used to study the baroreflex in men and women since adequate baroreflex function is necessary to compensate for the acute effects of the vasoactive drugs (59). To our knowledge, this methodology has never been implemented to examine sex differences throughout long-duration bed rest. The purpose of this study was to assess catecholamine, HR variability, and baroreflex sensitivity responses to continuous infusions of vasoactive drugs to determine sex differences in BP control before and during long-duration bed rest. We hypothesized that the methodology used in this study would allow us to explore the steady-state properties of autonomic control throughout a large range of BPs, such that we could detect three main differences or changes: 1) a sex difference in BP regulation before bed rest, 2) the effects of bed rest leading to autonomic dysfunction, and 3) an interaction between sex and bed rest to explain contributory factors to the higher incidence of postflight orthostatic hypotension in women.

METHODS

Subjects.

Twenty men (age: 32.5 ± 1.3 yr, weight: 82.2 ± 2.6 kg, height: 175.8 ± 1.8 cm) and ten women (age: 37.2 ± 2.8 yr, weight: 60.6 ± 3.7 kg, height: 160.2 ± 1.7 cm) participated at least 60 days in 6° head-down bed rest conducted at the General Clinical Research Center (GCRC) Satellite Flight Analogs Research Unit at the University of Texas Medical Branch (UTMB) in Galveston, TX. Study protocols and procedures were reviewed and approved by the National Aeronautics and Space Administration Johnson Space Center Committee for the Protection of Human Subjects, UTMB Institutional Review Board, and UTMB GCRC Advisory Committee. Subjects received written and verbal explanations of the study objectives and protocols and provided signed informed consent before participation.

Subjects were healthy, normotensive, nonsmokers, not obese, and taking no medications. Women were premenopausal, with regular menses lasting 28 ± 2 days. Bed rest standard conditions were used in accordance with the National Aeronautics and Space Administration Flight Analogs Project (49). Briefly, all subjects adhered to a strict 6° head-down tilt bed rest, 24 h/day, for up to 90 days. Diet was composed of food similar to those found in the Shuttle food system, with a carbohydrate, fat, and protein ratio of 55:30:15. Baseline caloric intake was 35.7 kcal/kg body wt, which was adjusted as necessary to maintain a weight within 3% of a subject's weight on day 3 of bed rest. All subjects were required to drink a minimum of 28.5 ml/kg of fluid each day, comprised mainly of distilled water. Subjects were not allowed nicotine, alcohol, caffeine, cocoa, chocolate, tea, or other herbal beverages. Subjects were awakened each day at 06:00 and required to begin sleep at 22:00 each day, with no napping allowed during the day. Before all test days for this study, subjects consumed a light snack of complex carbohydrates within 2 h before testing. Subjects were not allowed medication or maximal exercise within the preceding 24 h.

Protocol.

To assess baroreflex responses during bed rest, we used two infusion protocols designed to assess BP control over a wide range of pressures. Stepwise venous infusions of phenylephrine (PE), an α₁-adrenergic receptor agonist, were used to induce vasoconstriction, whereas sodium nitroprusside (SNP), a nitric oxide donor, was used to induce vasodilation. Testing occurred 1 day before bed rest (20 men and 10 women) and near bed rest day 30 (20 men and 10 women) and day 60 (16 men and 8 women). In women, the protocol was always performed during menses to control for hormonal levels (66).

Twenty-gauge intravenous catheters (BD Insyte Autoguard, Becton Dickinson, Franklin Lakes, NJ) were inserted into suitable veins for drug infusions and blood draws on contralateral arms. After a 20-min rest period, normal saline was infused for 10 min using an infusion pump (Baxter FLO-GARD 6201, Round Lake, IL). At the end of the infusion, a 3-ml blood sample was drawn to measure baseline catecholamines. Three 10-min infusions of increasing dosages of PE (0.4, 0.8, and 1.6 μg·kg⁻¹·min⁻¹) were administered. A 20-min rest followed the PE infusions to allow the subject's HR and BP to return to baseline. After the rest period, normal saline was infused again for 10 min followed by three 10-min infusions of SNP (0.4, 0.8, and 1.2 μg·kg⁻¹·min⁻¹), similar to the above procedure.

Not all subjects completed all three dosages of both drugs. Infusions were stopped if diastolic BP (DBP) changed by 15 mmHg, systolic BP (SBP) rose by 50 mmHg or fell by 25 mmHg, absolute SBP went below 70 mmHg, HR went below 40 beats/min, cardiac rhythm changed, or the subject requested termination.

ECG (Escort II, Medical Data Electronics, Arleta, CA), respiration, and finger arterial BP (Finometer, Finapres Medical Systems) were digitized and recorded at 250 Hz (Notocord, France). Offline analyses were performed with MATLAB (MathWorks, Natick, MA). Norepinephrine (NE) assays were conducted at the Yale University GCRC Core laboratory using HPLC with electrochemical detection (ESA, Chelmsford, MA), where sensitivity was from 2 to 5 pg/ml with an intra-assay variation of 1.9–2.5%.

HR variability and BP variability.

Artifacts, ectopic beats, and occasional deep breaths, as determined from the respiration signal, were manually removed from the R-R interval and BP data series. Finometer calibration steps were also removed from the BP signal for the accurate detection of SBP and DBP. R-R intervals and SBP were resampled at 4 Hz using a cubic spline interpolation, and the mean and linear trend were removed before spectral power calculation. Power spectral densities were calculated using Welch's method of modified periodograms (65) with segments overlapping by 50% and a 1,024-point fast Fourier transform. Power spectral densities were integrated using the trapezoidal method over the specified frequency range to determine low-frequency (LF; 0.04–0.15 Hz) and high-frequency (HF; 0.15–0.40 Hz) power (60). Autonomic activity was approximated by LF SBP, an index of sympathetic activity (48, 51, 55), and the HF R-R interval, an index of parasympathetic modulation (51, 54, 60).

Baroreflex sensitivity.

The spontaneous baroreflex slope, a measure of arterial baroreceptor sensitivity, was determined from beat-to-beat changes in SBP and R-R interval (6, 58). Sequences of three or more beats where SBP increased by at least 1 mmHg and the following R-R interval lengthened by at least 10 ms, as well as sequences where a decrease in SBP was followed by a shortening of the R-R interval, were identified. Least-squared linear regression was used to determine the slope and correlation coefficient of each ramp. Baroreflex gain was defined as the average slope of the ramps during each infusion where the squared correlation coefficient was >0.85.

Plasma volume.

Plasma volume was measured in all of the subjects using the carbon monoxide rebreathing technique (46, 63) before bed rest and around bed rest days 28 and 57 as part of a separate protocol (53). The contribution of the plasma volume index, which corrects plasma volume by body surface area (53), to baroreflex sensitivity during bed rest was examined to determine if decreases in baroreflex sensitivity were due to bed rest or hypovolemia.

Statistical analysis.

All statistical analyses were performed using Stata, IC software (version 11.2, StataCorp LP, College Station, TX) and setting the two-tailed α to reject the null hypothesis at 0.05. To meet the assumptions required of the statistical analyses, all of the spectral measures, NE, and spontaneous baroreflex slope required a log transformation. The NE outcome also included a zero offset before log transformation to improve the distribution of residuals and thereby meet the statistical assumptions.

As described above, our longitudinal experimental design involved repeated measures of our dependent variables before, during, and after bed rest. Not all subjects completed the full 60 days of bed rest due to hurricane evacuations (4 men and 2 women); however, the statistical techniques that we used enabled analysis of all available data (i.e., no subject in this data set was dismissed due to deviation from the standard bed rest conditions). These individuals supplied pre-bed rest and in-bed rest observations but no post-bed rest data due to their evacuation. Thus, our analysis included complete data from n = 24 subjects plus partial data from those 6 subjects who were unable to complete the protocol but were nevertheless able to contribute some data for analysis.

Separate mixed-effects linear regression models were used to evaluate the effects of sex, bed rest day, and dose on our continuously scaled dependent variables. Our statistical model included fixed-effects dummy-coded β-coefficients comparing men with women, each in-bed rest and post-bed rest time period versus baseline, and each of the three doses (0.4, 0.8, and 1.6 μg·kg⁻¹·min⁻¹ for PE and 0.4, 0.8, and 1.2 μg·kg⁻¹·min⁻¹ for SNP) versus normal saline baseline. We also included all possible two-way and three-way interaction terms in our model enabling us to test our hypotheses that sex and dose have interactive effects on changes over time. As typical with mixed-effects modeling, we included random intercept terms to accommodate the longitudinal design of the experiment, allowing each participant to have their own offset. Random slopes were not indicated in our evaluation of model residuals and did not provide better fit to the data. Overly influential outliers (those with standardized residuals greater than 3 or less than −3) were removed to meet statistical assumptions. Less than 2% of the measurements were found to be overly influential. All results are reported as means ± SE.

To evaluate the effect of plasma volume on baroreflex sensitivity, mixed-modeling analysis was performed on the baseline spontaneous baroreflex slope using both bed rest day and plasma volume index as factors and on the baseline spontaneous baroreflex slope using just bed rest day as an independent variable. The two models were compared to determine the contribution of plasma volume to changes in the spontaneous baroreflex slope.

RESULTS

Effect of bed rest.

Shifts in R-R interval, autonomic function, and baroreflex sensitivity illustrate the cardiovascular deconditioning that accompanies bed rest. R-R interval decreased with bed rest (P < 0.001). The sex-by-day interaction (P < 0.05) for the PE baseline and infusions indicated a larger decrease in R-R interval with bed rest in men. Main effects of bed rest day showed a decrease in SBP (P < 0.005) and DBP (P < 0.03), yet the sex-by-day interactions (P < 0.03) indicated the effect was due to men, such that SBP and DBP did not decrease with bed rest in women. Arterial baroreflex sensitivity decreased with bed rest (P < 0.001) during the entire protocol, and this decrease was greater in women than in men after 30 days of bed rest, as evidenced by the sex-by-day interaction during the SNP baseline and infusions (P < 0.02). Statistical analysis to determine the effect of plasma volume on the decrease in the spontaneous baroreflex slope yielded a main effect of bed rest day (P < 0.01 for pre-PE saline and P < 0.02 for pre-SNP saline) but no significant contribution from plasma volume index in the mixed model. Additionally, the model with both bed rest day and plasma volume index as factors was not significantly different from the model with just plasma volume as a factor, indicating the decrease in spontaneous baroreflex slope was due to bed rest day and not hypovolemia.

Bed rest additionally altered autonomic function, as measured by catecholamine response and spectral HR variability. Catecholamine analysis yielded an enhanced drug response during bed rest; the main effects of bed rest were attenuated plasma NE levels during the PE infusions (P < 0.001) and elevated plasma NE levels during the SNP infusions (P < 0.01), with an enhanced response at the lowest SNP dose during bed rest (day-by-dose interaction, P < 0.03). HR variability analysis further detected alterations in autonomic balance. Bed rest was characterized by diminished parasympathetic modulation (HF R-R interval, P < 0.04) and increased sympathetic activity (LF SBP, P < 0.005), both of which were significant during the entire protocol.

Effect of PE.

Group means ± SE of the measured and calculated parameters during each of the PE infusions are shown in Table 1. Mean R-R interval, SBP, and DBP increased with PE infusion (P < 0.001). A typical response to the PE infusions is shown in Fig. 1. Women had lower SBP and DBP than men (P < 0.005), experienced a larger increase in DBP from baseline than men with higher PE doses (P < 0.05), and tended to experience a larger increase in SBP from baseline than men with higher PE doses (P = 0.058; Fig. 2). Plasma NE levels decreased (P < 0.001). Significant sex differences in the response to the highest dose of PE were found in autonomic activity and baroreflex sensitivity analyses. Baroreflex gain increased with PE (P < 0.001), but women did not experience as large an increase in baroreflex gain as men during the 1.6 μg·kg⁻¹·min⁻¹ PE infusion (sex-by-dose interaction, P < 0.05). Spectral analysis indicated that PE infusions were characterized by increased parasympathetic modulation (HF R-R interval, P < 0.001) and decreased sympathetic activity (LF SBP, P < 0.001). As with baroreflex sensitivity analysis, HF R-R interval at the highest PE dose differed between sexes (sex-by-dose interaction, P < 0.02). Men experienced increased HF R-R interval with every PE dose, whereas the women's response was saturated such that HF R-R interval during the 1.6 μg·kg⁻¹·min⁻¹ PE infusion was not different from that during the 0.8 μg·kg⁻¹·min⁻¹ PE infusion.

Table 1.

Heart rate variability, blood pressure variability, and catecholamine results during PE infusions

	Men				Women
	Normal saline	0.4 μg·kg−1·min−1 PE	0.8 μg·kg−1·min−1 PE	1.6 μg·kg−1·min−1 PE	Normal saline	0.4 μg·kg−1·min−1 PE	0.8 μg·kg−1·min−1 PE	1.6 μg·kg−1·min−1 PE
R-R interval, msb,c,d
Pre-bed rest	932 ± 31	1,008 ± 37	1,051 ± 38	1,189 ± 42	852 ± 30	934 ± 37	1,012 ± 45	1,127 ± 50
Bed rest day 30	871 ± 27	947 ± 33	1,032 ± 38	1,178 ± 40	822 ± 44	931 ± 39	994 ± 40	1,104 ± 56
Bed rest day 60	826 ± 28	920 ± 36	982 ± 42	1,125 ± 43	823 ± 37	907 ± 44	980 ± 51	1,098 ± 51
SBP, mmHga,b,c,d
Pre-bed rest	127 ± 3	132 ± 3	135 ± 3	146 ± 3	108 ± 3	116 ± 2	124 ± 3	131 ± 4
Bed rest day 30	116 ± 2	123 ± 3	129 ± 3	141 ± 3	106 ± 3	113 ± 2	118 ± 3	137 ± 4
Bed rest day 60	119 ± 4	125 ± 4	131 ± 4	142 ± 4	106 ± 3	112 ± 2	122 ± 3	139 ± 4
DBP, mmHga,b,c,d,e
Pre-bed rest	68 ± 3	71 ± 3	73 ± 3	79 ± 3	54 ± 4	60 ± 3	65 ± 4	67 ± 3
Bed rest day 30	61 ± 2	65 ± 2	68 ± 2	73 ± 2	54 ± 3	58 ± 2	60 ± 2	68 ± 3
Bed rest day 60	63 ± 3	67 ± 3	71 ± 3	76 ± 3	56 ± 3	61 ± 3	64 ± 3	69 ± 4
Norepinephrine, pg/mlb,c
Pre-bed rest	220 ± 24	124 ± 15	98 ± 12	91 ± 11	144 ± 21	99 ± 16	86 ± 16	81 ± 14
Bed rest day 30	200 ± 21	120 ± 18	88 ± 15	83 ± 12	155 ± 22	105 ± 17	71 ± 12	63 ± 11
Bed rest day 60	205 ± 27	107 ± 17	83 ± 15	67 ± 10	177 ± 28	88 ± 16	74 ± 9	60 ± 8
Baroreflex gain, ms/mmHgb,c,e
Pre-bed rest	16.0 ± 1.5	18.4 ± 1.4	22.8 ± 2.0	28.9 ± 2.7	16.0 ± 1.6	19.1 ± 2.3	25.9 ± 2.6	25.5 ± 3.2
Bed rest day 30	12.6 ± 0.9	15.6 ± 1.3	21.0 ± 2.3	25.4 ± 1.5	14.5 ± 2.2	19.8 ± 3.0	20.1 ± 2.3	21.1 ± 3.0
Bed rest day 60	12.7 ± 0.9	16.0 ± 1.5	19.5 ± 2.2	21.9 ± 1.8	14.6 ± 3.0	19.9 ± 4.8	22.5 ± 4.7	23.6 ± 3.8
LF R-R interval, ms2c,e
Pre-bed rest	998 ± 178	1,183 ± 230	1,666 ± 469	1,996 ± 466	545 ± 102	1,343 ± 473	1,971 ± 563	1,962 ± 616
Bed rest day 30	501 ± 47	950 ± 146	1,334 ± 271	1,679 ± 264	664 ± 158	1,256 ± 411	2,552 ± 1092	1,789 ± 364
Bed rest day 60	531 ± 81	763 ± 108	1,022 ± 149	1,505 ± 332	514 ± 148	1,074 ± 392	1,568 ± 464	2,235 ± 900
HF R-R interval, ms2b,c,e
Pre-bed rest	619 ± 108	884 ± 150	1,348 ± 239	1,687 ± 274	610 ± 171	1,179 ± 341	2,255 ± 661	2,088 ± 505
Bed rest day 30	334 ± 63	772 ± 173	1,490 ± 385	1,583 ± 241	533 ± 238	1,278 ± 455	2,172 ± 646	2,039 ± 480
Bed rest day 60	279 ± 63	602 ± 113	881 ± 214	1,403 ± 183	856 ± 559	1,649 ± 834	2,314 ± 784	2,243 ± 746
LF SBP, mmHg2b,c
Pre-bed rest	7.59 ± 0.97	7.17 ± 0.76	6.20 ± 0.83	4.66 ± 0.69	8.38 ± 2.04	7.41 ± 1.79	6.21 ± 1.24	7.41 ± 2.29
Bed rest day 30	8.14 ± 1.29	8.51 ± 1.20	7.80 ± 0.82	6.52 ± 1.01	8.66 ± 2.15	7.21 ± 1.54	10.88 ± 3.52	7.65 ± 1.66
Bed rest day 60	7.68 ± 0.92	7.19 ± 0.68	6.24 ± 0.63	7.36 ± 1.47	7.33 ± 1.75	6.81 ± 1.26	8.41 ± 1.54	9.00 ± 4.06

Values are means ± SE. PE, phenylephrine; SBP, systolic blood pressure; DBP, diastolic blood pressure; LF, low frequency; HF, high frequency.

^aMain effect of sex;

^bmain effect of day;

^cmain effect of drug dose;

^dsex-by-day interaction;

^esex-by-dose interaction.

Fig. 1.

Typical systolic blood pressure (SBP; top) and R-R interval (bottom) responses during phenylephrine (PE) infusions. Data were plotted for one male subject during bed rest.

Fig. 2.

Model-predicted means and SE of the response to increasing doses of PE for each sex before and after 60 days of bed rest. HF, high frequency; LF, low frequency.

Effect of SNP.

Group means ± SE of the measured and calculated parameters during each of the SNP infusions are shown in Table 2. A typical R-R interval and SBP response to the SNP infusions is shown in Fig. 3. Mean R-R interval decreased with SNP (P < 0.001). Mean SBP was not affected by the drug in men and increased in women with SNP (sex-by-dose interaction, P < 0.001). Mean DBP decreased with SNP (P < 0.03), but this decrease was dictated by the men's response (sex-by-dose interaction, P < 0.04). Baroreflex gain decreased with SNP (P < 0.001). The significant sex-by-dose interaction (P < 0.03) in baroreflex gain suggests that women experienced a larger decrease in baroreflex gain than men at the lower SNP doses. As expected, NE levels increased during the SNP infusions (P < 0.001). Autonomic activity, as indicated by spectral analysis, was also altered by the SNP infusions. Parasympathetic modulation (HF R-R interval) decreased and sympathetic activity (LF SBP) increased with SNP (P < 0.001). The parasympathetic (HF R-R interval) withdrawal caused by the SNP infusions was larger in women than in men (sex-by-dose interaction, P < 0.001) and was magnified during bed rest for both sexes (day-by-dose interaction, P < 0.005). These sex differences in autonomic regulation during SNP are shown in Fig. 4.

Table 2.

Heart rate variability, blood pressure variability, and catecholamine results during SNP infusions

	Men				Women
Normal saline	0.4 μg·kg−1·min−1 SNP	0.8 μg·kg−1·min−1 SNP	1.2 μg·kg−1·min−1 SNP	Normal saline	0.4 μg·kg−1·min−1 SNP	0.8 μg·kg−1·min−1 SNP	1.2 μg·kg−1·min−1 SNP
R-R interval, msa,b
Pre-bed rest	958 ± 32	837 ± 26	771 ± 26	745 ± 24	903 ± 40	787 ± 38	738 ± 36	728 ± 38
Bed rest day 30	862 ± 29	733 ± 25	669 ± 20	634 ± 19	848 ± 39	684 ± 30	628 ± 32	588 ± 34
Bed rest day 60	849 ± 34	713 ± 25	654 ± 20	636 ± 22	843 ± 41	678 ± 31	620 ± 28	583 ± 29
SBP, mmHga,c,d
Pre-bed rest	121 ± 3	120 ± 3	119 ± 3	120 ± 3	109 ± 2	114 ± 3	114 ± 3	114 ± 5
Bed rest day 30	114 ± 3	116 ± 3	112 ± 3	111 ± 3	109 ± 3	113 ± 4	113 ± 4	114 ± 4
Bed rest day 60	118 ± 4	119 ± 3	116 ± 3	114 ± 4	106 ± 2	110 ± 4	113 ± 4	112 ± 4
DBP, mmHga,c
Pre-bed rest	64 ± 2	61 ± 2	59 ± 2	60 ± 2	57 ± 3	58 ± 3	57 ± 3	54 ± 4
Bed rest day 30	60 ± 3	60 ± 2	59 ± 2	59 ± 2	61 ± 3	62 ± 3	61 ± 3	63 ± 3
Bed rest day 60	64 ± 3	63 ± 3	61 ± 2	60 ± 3	59 ± 3	61 ± 3	62 ± 3	60 ± 3
Norepinephrine, pg/mla,b,e
Pre-bed rest	133 ± 17	216 ± 21	294 ± 23	326 ± 22	115 ± 19	191 ± 29	282 ± 43	330 ± 47
Bed rest day 30	129 ± 15	261 ± 24	309 ± 24	340 ± 27	107 ± 16	233 ± 31	339 ± 47	408 ± 46
Bed rest day 60	137 ± 17	279 ± 24	337 ± 26	386 ± 26	119 ± 23	262 ± 52	333 ± 50	331 ± 47
Baroreflex gain, ms/mmHga,b,c,d
Pre-bed rest	14.1 ± 1.0	12.3 ± 0.9	11.1 ± 0.9	10.6 ± 0.8	17.7 ± 1.8	13.5 ± 1.7	12.6 ± 1.0	13.3 ± 1.2
Bed rest day 30	11.6 ± 0.9	9.5 ± 0.8	8.6 ± 0.6	7.6 ± 0.5	14.6 ± 2.8	9.9 ± 1.2	8.6 ± 1.3	8.7 ± 1.5
Bed rest day 60	11.9 ± 1.0	9.0 ± 0.8	8.0 ± 0.4	7.6 ± 0.5	15.3 ± 3.3	9.9 ± 1.4	8.7 ± 0.9	10.4 ± 1.3
LF R-R interval, ms2a,b,d
Pre-bed rest	747 ± 128	723 ± 79	876 ± 161	986 ± 249	1c147 ± 318	664 ± 174	588 ± 153	660 ± 187
Bed rest day 30	541 ± 78	527 ± 81	611 ± 135	493 ± 131	714 ± 305	354 ± 105	313 ± 85	314 ± 123
Bed rest day 60	505 ± 76	450 ± 58	458 ± 80	480 ± 138	656 ± 231	268 ± 61	201 ± 41	212 ± 53
HF R-R interval, ms2a,b,d,e
Pre-bed rest	469 ± 83	355 ± 84	315 ± 80	300 ± 78	740 ± 147	376 ± 104	322 ± 94	336 ± 74
Bed rest day 30	287 ± 82	201 ± 63	164 ± 47	117 ± 42	541 ± 268	135 ± 40	135 ± 58	132 ± 73
Bed rest day 60	277 ± 73	170 ± 62	84 ± 24	85 ± 28	600 ± 270	132 ± 57	101 ± 49	111 ± 75
LF SBP, mmHg2a,b
Pre-bed rest	6.76 ± 0.92	12.54 ± 3.33	18.42 ± 5.10	16.98 ± 3.16	11.76 ± 4.34	9.00 ± 1.78	8.38 ± 1.36	7.25 ± 1.26
Bed rest day 30	7.58 ± 0.74	14.40 ± 4.17	21.57 ± 5.94	29.25 ± 12.02	7.17 ± 1.59	9.70 ± 1.81	12.85 ± 2.66	15.10 ± 2.86
Bed rest day 60	7.47 ± 0.81	14.22 ± 2.74	20.60 ± 4.56	21.32 ± 4.12	5.93 ± 1.12	7.43 ± 1.29	8.09 ± 1.01	10.22 ± 1.47

Values are means ± SE. SNP, sodium nitroprusside.

^aMain effect of day;

^bmain effect of drug dose;

^csex-by-day interaction;

^dsex-by-dose interaction;

^eday-by-dose interaction.

Fig. 3.

Typical SBP (top) and R-R interval (bottom) responses during the SNP infusions. Data were plotted for one male subject during bed rest.

Fig. 4.

Model-predicted means and SE of the response to increasing doses of SNP for each sex before and after 60 days of bed rest.

DISCUSSION

Overall findings.

Bed rest caused a decrease in R-R interval, baroreflex sensitivity, and spectral indexes of parasympathetic modulation and an increase in spectral indexes of sympathetic activity. Catecholamine levels during the infusion of vasoactive drugs and the parasympathetic withdrawal with SNP infusions were also enhanced with bed rest. Autonomic response to infusion was as expected, with PE causing an increase in R-R interval, SBP, baroreflex sensitivity, and parasympathetic modulation. SNP had opposite effects on autonomic function, with decreases in R-R interval, baroreflex sensitivity, and parasympathetic modulation and an increase in sympathetic activity. The differences in sympathetic activity during infusions of both drugs were significant in both catecholamine levels and spectral analysis indexes. However, the more interesting findings of this study are the observed sex differences during the drug infusions. During the pre-SNP saline and SNP infusions, men tended to have higher sympathetic modulation than women, whereas women suppressed parasympathetic modulation more than men, and men tended to increase sympathetic activity more than women during the SNP infusions. This parasympathetic suppression in women extended to the baroreflex sensitivity results, where women experienced more decreased baroreflex sensitivity than men at lower SNP doses and a quicker decrease in baroreflex sensitivity with bed rest than men.

BP responses to vasoactive drugs.

As expected, SBP increased during PE; however, we did not observe the expected decrease in mean SBP during SNP (Fig. 4) and even found an increase in mean SBP during SNP in women. BP responses were similar to those observed by Korner et al. (43), who examined the steady-state properties of the baroreflex by injecting PE and glyceryl trinitrate in small doses to the right atrium over 15–20 s. These investigators showed a vasodilator injection followed by a drop in BP, a plateau, and a subsequent recovery in BP and a PE injection followed by increase in BP, where a recovery did not occur within the plotted area. The difference in the recovery of the BP responses to the PE and SNP infusions could be due to the differences in the catecholamine response. Previous studies (23, 34) have found that the increase in plasma NE during hypotension is larger than the decrease seen during hypertension. We observed similar results, where the magnitude of the increase in NE during the SNP infusions (175 ± 7 pg/ml) was significantly larger (t-test, P < 0.001) than the magnitude of the decrease in NE during the PE infusions (101 ± 5 pg/ml). Thus, the lack of change in SBP during SNP seems to be indicative of a larger compensatory response to the vasodilator than the vasoconstrictor in women.

Sex differences in autonomic control.

Determination of baroreflex curves for men and women was precluded by the lack of decrease in SBP in women during the SNP infusions. Tank et al. (59) were able to draw such curves; however, they did not control for the menstrual cycle and included women taking oral contraceptives, whereas we only tested women during the early follicular phase and none of the women were taking hormones. Previous studies (38, 50) have shown that sympathetic activity, as measured by plasma NE and muscle sympathetic nerve activity, as well as the sympathetic and cardiovagal baroreflex, change between the phases of the menstrual cycle. Although we were not able to draw the baroreflex curves for the sexes, we showed sex differences in baroreflex response to vasoactive drugs. Specifically, the lesser spontaneous baroreflex slope response in women for the higher doses of PE suggests that women reach the saturation region of the baroreflex at a lower BP than men. The larger decrease in spontaneous baroreflex slope in women for the lower dose of SNP could indicate a less effective baroreflex than men, particularly during low-level stress on the cardiovascular system. This result, together with Christou et al.'s finding (12) of less effective baroreflex buffering in women, suggest a contribution of the baroreflex to sex differences in orthostatic tolerance.

Our results agree with previous studies (3, 4, 27, 56) showing that women are characterized by enhanced parasympathetic responsiveness and reduced sympathetic activity in the autonomic control of BP compared with men. However, we have shown this sex difference during a range of constant infusions of vasoactive drugs instead of during orthostatic maneuvers or autonomic blockades. We have also shown that long-duration bed rest does not affect sex differences in the autonomic control of BP. The differences in autonomic regulation were clearly observed during SNP infusions, where women suppressed parasympathetic modulation (HF R-R interval) more than men. During PE infusions, both sexes increased parasympathetic modulation, yet HF R-R interval increased more in women for the lower doses of PE and was saturated by the 0.8 μg·kg⁻¹·min⁻¹ infusion. The saturation of HF R-R interval does not indicate poor parasympathetic control or a parasympathetic ceiling in women, but likely a further increase in parasympathetic stimulation. The HF R-R interval quantifies the respiratory sinus arrhythmia (5, 51, 54), which is primarily parasympathetically mediated. At very high levels of vagal stimulation, the respiratory sinus arrhythmia is diminished and dissociated from parasympathetic activity (5, 32, 47), causing a saturation, or even a decrease, of respiratory measures of parasympathetic activity (24, 33, 47). Goldberger et al. (33) showed that a quadratic relationship outperformed a linear relationship between log-transformed HF R-R interval and R-R interval, accounting for a decrease in HF R-R interval at very high levels of vagal activity, as indicated by large R-R intervals caused by PE infusions. Although they identified an age effect for predicting the quadratic curves, they did not find a sex effect. Our results indicate a saturation in the women's HF R-R interval and no sex difference in R-R intervals (Fig. 2), suggesting that such a quadratic relationship might be horizontally shifted to account for sex, where women reach the HF R-R interval ceiling at a lower R-R interval than men.

Autonomic dysfunction with bed rest.

The differences in autonomic activity between men and women were preserved throughout 60 days of bed rest. However, bed rest caused autonomic dysfunction, as evidenced by the decrease in parasympathetic modulation (HF R-R interval) and baroreflex sensitivity. We found an increase in sympathetic activity when measured by LF SBP, no main effect of bed rest on catecholamines, and no change in the response of either measurement to the drug doses with bed rest. Baseline sympathetic activity or the response to an orthostatic stimulus have been reported to decrease (11, 20, 57), not change (15, 17, 52, 57), or increase with bed rest (40, 52). The differences in these findings may be due to different study lengths, stimulus during the measurement, or methodology, where the particular method could be measuring muscle nerve activity, sympathetic tone, or cardiac sympathetic activity. Although the effect of bed rest on sympathetic activity has been controversial, our results agree with the consensus that parasympathetic modulation (17, 40, 42, 52) and baroreflex sensitivity (13, 21, 28, 39, 57) decrease with bed rest.

We not only found that parasympathetic modulation decreased with bed rest but also that the compensatory parasympathetic withdrawal during SNP infusions was magnified by bed rest. Furthermore, although plasma volume has been previously found to be both a main contributor (41, 42) and not significant (13, 15, 61) in reductions of baroreflex sensitivity, our results showed that the decrease in baroreflex sensitivity was not due to decreased plasma volume. In addition to showing that the known effects of bed rest on autonomic control are evident during continuous infusions of vasoactive drugs and that the reduction in baroreflex sensitivity is a function of bed rest day rather than reduction in plasma volume, we found that women experienced a larger decrease in baroreflex sensitivity than men by day 30 of bed rest. This sex difference in the time course of baroreflex dysfunction could potentially contribute to sex differences in orthostatic tolerance after bed rest.

Comparison with spaceflight.

Multiple studies have examined autonomic function or baroreflex sensitivity during and after spaceflight; however, these studies consisted of entirely or mostly male subject groups, precluding the possibility of examining sex differences (2, 11, 16, 19, 22, 25, 29–31, 62). Although some studies have found that baroreflex sensitivity in flight is decreased (22) or similar to that in the supine position preflight (19, 62), most studies have agreed that baroreflex sensitivity is decreased after spaceflight (16, 29, 30). The reduction in baroreflex sensitivity after spaceflight is consistent with our findings during bed rest and those of other bed rest studies (13, 15, 28, 39, 57). Spaceflight has also been shown to cause changes in autonomic function similar to those we found in bed rest, such as diminished vagal activity (2, 16, 22, 62), increased sympathetic activity (11, 22, 26, 30, 31), and increased sympathovagal balance (2, 30). Future studies should specifically address whether the sex differences in autonomic function and baroreflex sensitivity observed in our bed rest study are also present during and after spaceflight.

Limitations.

Due to uncontrollable circumstances, equipment availability, and preset maximum drug doses, certain limitations could have impacted our analyses. The aim of the study centered on sex differences, yet subject recruitment and availability limited the number of women to half of that of men. Missing data points due to hurricane evacuations and early termination of drug infusions further limited the statistical analyses that could be performed. Limitations of the hardware included an analog ECG output update rate of 125 Hz. This is below the recommended sampling rate for HR variability analysis, and although the digitizing of the signal at 250 Hz fell within guidelines, the accuracy of the HR variability measures could have been improved with a higher output rate. Autonomic activity was approximated by spectral analyses instead of invasive direct methods. Although the contribution of muscle sympathetic nerve activity to BP regulation during physiological stressors is unknown in women (36), our study could have benefited from other measures of sympathetic activity. The dosage of the drug infusions and termination criteria prevented an examination of both saturation regions of the baroreflex curve in most subjects, such that the lower saturation region of the baroreflex could not be examined in women. Finally, the rest period was chosen to be long enough to eliminate the effect of the PE infusions and for HR and BP to visually return to baseline. However, paired t-tests showned that plasma NE (P < 0.001, change: −66 ± 6 pg/ml) was lower and mean R-R interval (P = 0.003, change: 19 ± 6 ms) was higher during the pre-SNP saline infusion than the pre-PE saline infusion.

Conclusions.

Long-duration bed rest suppresses parasympathetic modulation and baroreflex function in men and women, with women experiencing a larger decrease in baroreflex sensitivity than men after 30 days of bed rest. Despite this bed rest-induced autonomic dysfunction, the parasympathetic dominance in women and the sympathetic dominance in men are preserved throughout bed rest. Our results during PE infusions suggest that women have a different parabolic relationship between HF R-R interval and R-R interval than men, thus reaching a respiratory sinus arrhythmia ceiling at a lower R-R interval than men. This study reveals the need for studying sex differences in BP control after long-duration spaceflight, since autonomic control and baroreflex sensitivity are thought to contribute to postflight orthostatic intolerance.

GRANTS

This work was funded by National Aeronautics and Space Administration (NASA) Grant BRC-2003-0000-0543 and sponsored by the NASA Flight Analogs Project. This work was conducted at the National Center for Research Resources (NCRR)-funded (NCRR Grant M01-RR-0073) General Clinical Research Center at the University of Texas Medical Branch (Galveston, TX).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: N.M.A., M.B.S., and S.H.P. performed experiments; N.M.A., M.B.S., and R.P.-S. analyzed data; N.M.A., M.B.S., S.M.L., and R.P.-S. interpreted results of experiments; N.M.A. prepared figures; N.M.A. drafted manuscript; N.M.A., M.B.S., S.M.L., and R.P.-S. edited and revised manuscript; N.M.A., M.B.S., S.M.L., R.P.-S., and S.H.P. approved final version of manuscript; S.H.P. conception and design of research.