Linköping University Medical Dissertation No. 1087

Cardiovascular responses to hypovolemic circulatory stress in women

With special reference to venous compliance and capacitance

Marcus Lindenberger

Division of Cardiovascular Medicine Department of Medical and Health Sciences Linköping University, Sweden

Linköping 2008

©Marcus Lindenberger, 2008

Cover picture: Left: Pooling of venous blood followed by filtration of fluid in the lower limbs in response to lower body negative pressure (LBNP), creating experimental hypovolemic circulatory stress. Right: Mobilization of venous blood and fluid from muscle and skin to the central circulation in response to LBNP in a woman.

Published articles have been reprinted and used with the permission of the copyright holder (Article I, Am J Physiol Regul Integr Comp Physiol; Articles II and III, Am J Physiol Heart Circ Physiol).

Printed in Sweden by LiU‐Tryck, Linköping, Sweden, 2008

ISBN 978‐91‐7393‐756‐6 ISSN 0345‐0082

To my family

You’ll Never Walk Alone Liverpool FC anthem and way of life

Contents

CONTENTS

ABSTRACT...... 3

LIST OF PAPERS ...... 5

ABBREVIATIONS...... 6

INTRODUCTION...... 7

AIMS ...... 13

MATERIALS...... 14 Ethics...... 14 Healthy volunteers (paper I‐IV)...... 14 Women prone to vaso‐vagal reaction (additional study) ...... 15

METHODS ...... 16 Lower body negative pressure (LBNP)...... 16 Cardiovascular responses...... 17 Changes in calf volume ...... 19 Changes in upper arm volume during LBNP...... 22 Blood samples...... 23 Data recordings...... 23 Statistics ...... 24

RESULTS ...... 25 Calf volumetric responses (paper I‐IV) ...... 25 The effect of capillary fluid filtration on compliance calculations ...... 29 Total hypovolemic response (paper I‐IV) ...... 30 Cardiovascular responses to acute hypovolemia (paper III‐IV)...... 30 Women prone to vaso‐vagal reaction during LBNP (additional study).. 37

Cardiovascular responses to hypovolemic circulatory stress in women

DISCUSSION ...... 40 Venous capacitance and compliance...... 40 Capillary fluid filtration in the calf and capillary filtration coefficient. 42 Cardiovascular responses to hypovolemic circulatory stress...... 45 Cardiovascular responses in women prone to vaso‐vagal reaction ...... 49 Methodological considerations and limitations...... 51

CONCLUSIONS ...... 53

POPULÄRVETENSKAPLIG SAMMANFATTNING ...... 54

ACKNOWLEDGEMENTS...... 58

REFERENCES ...... 60

Abstract

ABSTRACT

Acute haemorrhage is a leading cause of death in trauma. Young women (YW) seem more susceptible to hypovolemic stress than young men (YM), but the underlying mechanisms are not clear. Elderly subjects are more vulnerable to haemorrhage, with a decreased defence of central blood volume in elderly men, but the defence has not been evaluated in elderly women (EW). The aims were to assess differences in cardiovascular responses to hypovolemic circulatory stress, emphasizing compensatory mechanisms to maintain central blood volume in YW, EW and women prone to vaso‐vagal reaction (VW). Lower body negative pressure (LBNP) was used as a model for haemorrhage and to create acute hypovolemic stress. Volumetric techniques were used to assess venous compliance, capacitance and capillary fluid exchange both caused by LBNP in the calf and the response to maintain central blood volume. LBNP induced a comparable hypovolemic stimulus in YW and YM, with lower calf venous compliance and capacitance but higher net capillary fluid filtration in YW. YW responded with smaller vasoconstriction without association between P‐NE and peripheral vascular resistance in contrast to YM. Venous capacitance response was decreased with time in YW. Further, net capillary fluid absorption from peripheral tissues to central circulation was decreased in YW in response to hypovolemic stress. All in all, this indicates less efficiency to defend central blood volume in young women. Calf venous compliance and capacitance was maintained in EW compared to YW but capillary filtration was decreased, implying reduced capillary function with age. With increasing transmural pressures however, filtration and capillary filtration coefficient (CFC) increased indicating increased capillary susceptibility to transmural pressure load in dependent regions with age. Heart rate increase was attenuated in EW while peripheral vascular conductance was maintained suggesting reduced cardiovagal baroreceptor function in response to hypovolemia with age. Venous capacitance response and fluid absorption from peripheral tissues to central circulation were decreased with age, indicating less efficiency to defend central blood volume. LBNP induced a slower hypovolemic stimulus in VW compared with non‐ vagal women. Further, the cardiopulmonary baroreflex was less efficient, and the venous capacitance response from peripheral tissues to central circulation was decreased, which may explain their susceptibility to orthostatic challenge.

3 Cardiovascular responses to hypovolemic circulatory stress in women

4 List of papers

LIST OF PAPERS

The presented thesis is mainly based on the papers listed below, which will be referred to in the text by their Roman numbers. The results of some additional experiments will also be presented.

I. Sex‐related effects on venous compliance and capillary filtration in the lower limb. Am J Physiol Regul Integr Comp Physiol. 292: R852‐ R859, 2007.

II. Decreased capillary filtration but maintained venous compliance in the lower limb of aging women. Am J Physiol Heart Circ Physiol. 293: H3568‐H3574, 2007.

III. Lower capacitance response and capillary fluid absorption in women to defend central blood volume in response to acute hypovolemic circulatory stress. Am J Physiol Heart Circ Physiol. 295: H867‐H873, 2008.

IV. Reduced defense of central blood volume during acute hypovolemic circulatory stress in aging women. Submitted.

5 Abbreviations

ABBREVIATIONS

AUC Area under the curve BMI Body mass index BP Blood pressure BRS Baroreceptor sensitivity Β0, β1, β2 Characteristics of venous compliance and capacitance C Compliance Cap 50 Time to 50% of the calf venous capacitance response CFC Capillary filtration coefficient CO Cardiac output CV Coefficient of variation CVP Central venous pressure DBP Diastolic blood pressure ΔV Change in volume ΔP Change in pressure EW Elderly women FBF Forearm blood flow FVC Forearm vascular conductance FVR Forearm vascular resistance HR Heart rate HUT Head‐up tilt LBNP Lower body negative pressure MAP Mean arterial pressure NE Norepinephrine P‐NE Plasma Norepinephrine PP Pulse pressure SBP Systolic blood pressure SV Stroke volume YM Young men YW Young women VW Women responding with vaso‐vagal reaction V0 Unstressed venous volume

6 Introduction

INTRODUCTION

Cardiovascular responses to hypovolemic circulatory stress

Haemorrhage and hypovolemic circulatory stress is a leading cause of death in trauma (Becker et al. 2002, Sauaia et al. 1995, Wohltmann et al. 2001). Acute hypovolemia induces a cascade of physiological responses to maintain blood pressure homeostasis, including activation of the baroreflex axis. Cardiopulmonary (low‐pressure) receptors located in the pulmonary artery and right atrium, are unloaded during early hypovolemia, while the arterial baroreceptors in the aortic arch and carotid sinuses are unloaded when arterial pressure decreases during more pronounced hypovolemia (Mancia and Mark 1983, Mark and Mancia 1983). Afferent baroreceptor nerves inhibit the central sympathetic nerve centre, leading to increased sympathetic activation (Seller 1991), mediated by efferent sympathetic nerve endings releasing norepinephrine (NE) in the arterial wall as well as by release of NE and epinephrine from the adrenal medulla into the systemic circulation. The early cardiovascular response to central hypovolemia (within sec) includes cardiac excitation (also by withdrawal of parasympathetic nervous system) and arterial vasoconstriction (Chien 1967). This increases blood pressure and redirects the blood flow to vital organs such as the brain and heart. Furthermore, the vasoconstrictor response reduces blood flow to the venous section, decreasing peripheral venous pressure. Capacitance vein blood volume then decreases substantially, leading to mobilization of blood to the central venous circulation (Rothe 1983). These responses act momentarily to maintain homeostasis as a first line of defence against hypovolemia. Effective circulating blood volume is further increased through a slower but continuous net capillary absorption of extra‐vascular fluid which has a major impact on early plasma volume restitution (Lanne and Lundvall 1992). If the subject survives the first critical phase, extra‐ and intravascular volume is restored by increased thirst and water intake, by endocrine and reflex regulation of renal water and salt excretion as well as through erythropoesis (Fitzsimons 1998, Miller et al. 1991).

7 Cardiovascular responses to hypovolemic circulatory stress in women

The venous section of the cardiovascular system can be looked upon as a voluminous blood reservoir containing 70% of total blood volume, with another 15% in the heart and lungs (Rowell 1993), and due to its great compliance well designed to preserve a proper inflow of blood into the heart during various cardiovascular adjustments (Rothe 1979). Thus, central venous pressure and filling of the heart may be maintained at a fairly stable level despite variations in venous blood volume. Although neuropeptide Y has been shown to constrict superficial veins (Linder et al. 1996), no evidence exists that active venoconstriction of capacitance vessels in skeletal muscle provides an important mechanism translocating blood towards the central circulation (Stewart et al. 2001). Skeletal muscle and skin is a primary target for the baroreflex arterial vasoconstriction during hypovolemia and due to its large total mass in the human body, the functional importance of the compensatory capacitance response and net capillary fluid absorption seems vital (Lesh and Rothe 1969, Lundvall et al. 1993, Olsen et al. 2000, Rothe 1983, Skelton 1927). A potent net capillary fluid absorption is dependent on both high hydrodynamic conductivity as well as a decline in capillary pressure caused by reflex autonomic adjustments of both α‐ and β‐ adrenergic receptors affecting the pre‐ to post‐capillary resistance ratio, creating a net driving force over the capillary wall (Lundvall and Hillman 1978, Maspers et al. 1990).

Lower body negative pressure (LBNP) is an excellent model for acute haemorrhage and hypovolemic circulatory stress, by inducing central hypovolemia and unloading of baroreceptors (Convertino et al. 2008, Cooke et al. 2004). When evaluating differences in cardiovascular responses to LBNP it is of importance to consider the created hypovolemic stimulus since LBNP induces the pooling of blood in the lower limbs, and increased limb venous compliance and capacitance have been shown to induce greater orthostatic intolerance (Morikawa et al. 2001, Olsen et al. 2000, Tsutsui et al. 2002). Net capillary fluid filtration in the lower limb further increases the hypovolemic stimulus over time and its importance has been demonstrated in patients suffering from postural orthostatic tachycardia syndrome (POTS) (Lanne and Olsen 1997, Lundvall et al. 1993, Stewart 2003).

8 Introduction

Sex-related differences in response to acute hypovolemia

Young women are more susceptible to experimental hypovolemic circulatory stress than men (Convertino 1998, Franke et al. 2003, Fu et al. 2004, Gotshall 2000, White et al. 1996) but the mechanisms underlying the susceptibility are not clear and probably multi‐factorial (Fu et al. 2004). In accordance with some findings in the arterial tree (Sonesson et al. 1993) it may be hypothesized that women have greater venous compliance with greater blood pooling in the lower limbs and greater hypovolemia. This seems to be refuted however by two recent studies, presenting greater calf venous compliance in young men (Meendering et al. 2005a, Monahan and Ray 2004). However, the technique used to calculate calf venous compliance does not exclude the considerable contribution of fluid filtration to calf volume increase, possibly confounding the conclusions. No available data on sex differences in calf capillary fluid filtration and capillary fluid coefficient (CFC) exists, even though sex hormones seem to influence capillary permeability and body fluid homeostasis (Pechere‐Bertschi et al. 2000, Stachenfeld et al. 2001, Stachenfeld and Taylor 2005, Tollan et al. 1992). In line with these findings, a recent large animal study detected increased micro‐vessel permeability in females compared to male pigs (Huxley et al. 2005).

A decrease in baroreceptor sensibility (BRS) in young women compared with young men has been postulated by several authors (Franke et al. 2003, Laitinen et al. 1998, Shoemaker et al. 2001), and women seem to respond with diminished arterial vasoconstriction to the infusion of α‐receptor agonists, e.g. NE (Bowyer et al. 2001, Freedman et al. 1987, Kneale et al. 2000). Furthermore, a more pronounced decrease in stroke volume (SV) and in cardiac output (CO) has been proposed as the main mechanism for the susceptibility of hypovolemic circulatory stress (Convertino 1998, Fu et al. 2004, Fu et al. 2005), due to smaller and functionally stiffer hearts impeding cardiac filling in young women (Fu et al. 2004, Fu et al. 2005), as well as lower relative circulating blood volume in women (Convertino 1998, Fu et al. 2005, White et al. 1996). The central hypovolemia that occurs during hypovolemic circulatory stress is compensated for by the mobilization of blood from peripheral capacitance vessels towards the central circulation, as well as by net capillary fluid absorption from tissue to blood in order to defend central blood volume and increase venous return to the heart (Ablad and Mellander 1963, Convertino et al. 2004, Lundvall and Lanne 1989a, Mellander 1960, Olsen et al. 2000). A less

9 Cardiovascular responses to hypovolemic circulatory stress in women efficient defence of central blood volume could hypothetically explain the decreased SV and CO detected in women, but has not been assessed in women.

Age-related differences in response to acute hypovolemia

Mortality to trauma increases with increasing age in both women and men, even after adjustment for higher prevalence of pre‐existing diseases (George et al. 2003, OʹKeefe et al. 2001, Taylor et al. 2002). On the other hand, orthostatic tolerance does not seem to decrease with age (Laitinen et al. 2004). In men, this can be attributed to decreased venous compliance in the lower limbs with increasing age, protecting against large volume displacements to the legs and central hypovolemia (Monahan et al. 2001, Olsen and Lanne 1998, Tsutsui et al. 2002). Calf venous compliance in elderly women is unknown however, but there are reasons to suspect differences with age between women and men, since young men have greater calf venous compliance than young women (Meendering et al. 2005a, Monahan and Ray 2004). Further, women demonstrate a slower decrease in arterial compliance with age compared to men, an effect probably attributed to estrogen (Debasso et al. 2004, Tomiyama et al. 2003). Estrogens have been shown to affect cellular transcription of elastin and collagen, and estrogen‐receptors are known to exist in smooth muscle cells in veins and arteries (Kappert et al. 2006, Knaapen et al. 2005, Mendelsohn 2002). Aging men have been shown to have an intact calf capillary filtration and CFC (Lanne and Olsen 1997). The marked drop in estrogen level during menopause in women might however affect capillary fluid filtration and CFC, but this has not been evaluated in aging women.

Impaired cardiovagal BRS as well as decreased α‐ and β‐adrenergic receptor responses has been found with age in mixed groups of women and men, potentially affecting HR and vasoconstriction during hypovolemia (Brown et al. 2003, Dinenno et al. 2002, Jones et al. 2003, Laitinen et al. 1998, Schutzer and Mader 2003, van Brummelen et al. 1981). Further, Olsen et al. (2000) found a ~50% reduction in venous capacitance response with age in healthy men during experimental hypovolemic stress, which could seriously impede survival during severe haemorrhage (Olsen et al. 2000). Compensatory mechanisms to defend central blood volume and blood pressure during hypovolemia have not yet been evaluated in elderly women.

10 Introduction

Vaso-vagal syncope

Syncope is a common clinical problem that affects up to 3.5% of the general population (Savage et al. 1985) and occurs most frequently during upright posture (Mosqueda‐Garcia et al. 2000). Increased calf venous pooling has been reported in subjects prone to syncope during head‐up tilt (HUT) (Hargreaves and Muir 1992), although others have found similar decrease in central hypovolemia in subjects experiencing syncope vs. controls (Epstein et al. 1968, Mosqueda‐Garcia et al. 1997). Other mechanisms behind orthostatic and vaso‐ vagal syncope in women may be defect cardiopulmonary baroreflex (Thomson et al. 1997, Wasmund et al. 2003, Wijeysundera et al. 2001) or differences in the venous section with reduced venous return to the heart (Fuca et al. 2006).

This thesis focuses on compensatory cardiovascular responses in the early phase of acute hypovolemic circulatory stress with emphasis on venous capacitance function and transcapillary fluid absorption from skeletal muscle and skin to blood.

11 Cardiovascular responses to hypovolemic circulatory stress in women

12 Aims

AIMS

• To evaluate sex‐related differences as well as age‐related changes in calf venous compliance and capacitance in women.

• To examine sex‐related differences as well as age‐related changes in calf net capillary fluid filtration.

• To study sex‐related differences in the cardiovascular response to experimental hypovolemic circulatory stress in young, emphasizing compensatory mechanisms to maintain central blood volume.

• To assess age‐related changes in the cardiovascular response during experimental hypovolemic circulatory stress in women, emphasizing compensatory mechanisms to maintain central blood volume.

• To evaluate differences in the cardiovascular response during experimental hypovolemic circulatory stress in women prone to vaso‐ vagal reactions compared with non vaso‐vagal responding women.

13 Cardiovascular responses to hypovolemic circulatory stress in women

MATERIALS

Ethics

The studies were approved by the Ethics Committee at Linköping University. All volunteers gave informed consent according to the Declaration of Helsinki.

Healthy volunteers (paper I-IV)

Paper I‐IV comprised of a total of 50 volunteers (range 20‐75 yrs, mean 33.5±2.7 yrs), divided into three groups; 22 young women (YW, range 20‐27 yrs, mean 23.1±0.4 yrs), 16 young men (YM, range 20‐26 yrs, mean 23.2±0.5 yrs), 12 elderly women (EW, range 61‐75 yrs, mean 66.4±1.4 yrs). All volunteers were healthy with no history of hypertension and physical examination showed an absence of varicose veins, diabetes (normal HbA1c) or other systemic diseases. All were non‐smokers and based on an interview regarding earlier and current training activities, they were found to be of average physical fitness, excluding sedentary individuals and well‐trained athletes. In paper I, a total of 28 volunteers divided into 12 YW and 16 YM were studied. In paper II and IV, a total of 34 women divided into 22 YW and 12 EW were studied. In paper III, a total of 38 volunteers divided into 22 YW and 16 YM were studied. Table 1 shows the demographic values in YW, YM and EW at rest.

14 Materials

Women prone to vaso-vagal reaction during LBNP (additional study)

Five young women responded with a vaso‐vagal reaction during LBNP of 44 mmHg (VW, range 21‐25 yrs, mean 22.6±0.6 yrs) and their cardiovascular responses were compared with the group of YW that tolerated LBNP well (n=22). Three VW were detected as a part of article II‐IV, one during a pilot study and the last subject was first excluded due to bad data quality, but developed a vaso‐vagal reaction in the end. The five VW were later re‐ examined to complete their registrations (except for Plasma Norepinephrine, P‐NE). Table 1 shows the demographic values in VW at rest.

Table 1. Demographic resting values (mean±SE). YW YM EW VW n 22 16 12 5 Age (yrs) 23.1±0.4 23.2±0.5 66.4±1.4††† 22.6±0.6 Height (cm) 169±1 181±1*** 166±1 166±2 Weight (kg) 62±2 72±1*** 64±2 65±3 BMI (kg/m2) 21.7±0.4 21.8±0.4 23.4±0.5† 23.5±0.9 HR (beats/min) 60±2 56±2 63±2 67±4 SBP (mmHg) 106±1 116±2*** 138±5††† 103±3 DBP (mmHg) 63±1 60±1 82±1††† 62±4 MAP (mmHg) 77±1 78±1 101±2††† 79±4 PP (mmHg) 43±2 56±2*** 57±4††† 37±2 FBF (ml/100ml min‐1) 2.2±0.2 3.0±0.2** 1.9±0.2 1.6±0.2 FVR (FVR units) 37±3 28±2* 66±10†† 51±8# FVC (FVC units, 10E‐3) 30±2 37±3* 18±2†† 21±3 P‐ NE (pmol/l) 1.2±0.1 1.6±0.2 2.2±0.3†† ‐‐‐ ‐‐‐ YW, young women; YM, young men; EW, elderly women; VW, vagal women. BMI, body mass index; HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure; PP, pulse pressure; FBF, forearm blood flow; FVR, forearm vascular resistance; FVC, forearm vascular conductance; P‐NE,NE, Plasma Norepinephrine. * YM. vs.vs YW; † EW vs. YW; # VW vs. YW.

15 Cardiovascular responses to hypovolemic circulatory stress in women

METHODS

Lower body negative pressure (LBNP)

All measurements were carried out with the subjects in the supine position. The LBNP technique was used to create defined transmural pressure changes over the vessel walls as well as to simulate acute hypovolemic circulatory stress and orthostatism. LBNP is an excellent model for acute haemorrhage and hypovolemic circulatory stress by inducing central hypovolemia and unloading of baroreceptors and sympathetic activation (Convertino et al. 2008, Cooke et al. 2004). The advantages with LBNP compared to head‐up tilt (HUT) is that the subject remains in the supine position, which facilitates physiological measurements and minimizes the risk of confounding skeletal activity and change in transmural pressure is easier to define. During LBNP, 80% of the negative pressure is transmitted to the underlying muscle tissue of the leg irrespective of muscle depth, time and magnitude, leading to a defined increase in transmural pressure over the vessel wall, with concomitant vessel dilatation and blood pooling (Olsen and Lanne 1998). LBNP of 40 to 50 mmHg results in a similar shift in thoracic blood volume as passive HUT to 70° (Taneja et al. 2007), but regional blood pooling differs between LBNP and HUT. While both techniques lead to increased blood pooling in lower limbs and the pelvic region, LBNP induces a decrease in splanchnic blood volume, similar to in haemorrhage, while HUT brings on splanchnic filling (Taneja et al. 2007). Figure 1 illustrates the cardiovascular response in young men to LBNP of 11, 22 and 44 mmHg (% of resting values). Heart rate (HR) is unchanged during LBNP of 11 mmHg, but increases with increasing LBNP thereafter. Systolic blood pressure (SBP) is stepwise reduced from LBNP of 22 mmHg, while diastolic blood pressure (DBP) is stable or slightly raised, leading to progressively decreased pulse pressure (PP) with increasing LBNP, but maintained mean arterial pressure (MAP). Forearm vascular resistance (FVR) increases rapidly and the maximal response is seen within the first min after initiation of LBNP. After termination of LBNP, the cardiovascular parameters return to resting values within the first min.

16 Methods

Experiments were performed at a stable room temperature of 23‐25°C and started 1 h after a regular meal. The subjects were instructed to abstain from caffeine beverages on the day of the investigation. The experiments were performed at two separate occasions, each lasting 2‐3 h. The subjects were placed in the supine position with the legs and pelvis enclosed in an airtight box up to the level of the iliac crest with a seal fitted hermetically around the waist. The pressure in the LBNP chamber was measured continuously by a manometer (DT‐XX disposable transducer, Viggo spectramed, Helsingborg, Sweden) and held constant by a rheostat. The LBNP chamber was connected to a vacuum source permitting stable negative pressure to be produced within 5 sec and maintained at constant pressure for 8 min LBNP of 11, 22 and 44 mmHg were used, with at least 30 min rest between each experiment to assure a return to the basal state.

No subjects were taking any regular medication. YW were scheduled between day 7 and 21 after start of menstruation, not excluding contraceptive use (10 out of the 22 YW). The type of contraceptives was not registered. The time in the menstrual cycle was chosen in accordance with two large studies on gender and LBNP tolerance (Gotshall 2000, White et al. 1996). Cardiovascular responses to LBNP seems to be unaffected by menstrual phase (Claydon et al. 2006, Frey et al. 1986, Meendering et al. 2005b), and furthermore, venous compliance and capacitance do not change over the course of the menstrual cycle or by oral contraceptive use (Meendering et al. 2005a). EW were all postmenopausal and not on hormone replacement therapy.

Cardiovascular responses

Blood pressure (BP) and HR were measured non‐invasively in the left upper arm by oscillometric technique (Dinamap Pro 200, Critikon) directly prior to LBNP and 1, 3, 6 and 8 min after LBNP initiation, and further 1, 2 and 4 min after LBNP termination. Forearm blood flow (FBF) was measured in the right forearm by standard venous occlusion double‐looped mercury‐in‐silicone strain‐gauge plethysmography (Hokanson EC‐6, D.E. Hokanson, Bellevue, WA), with the forearm at heart level and the strain‐gauge at the maximal circumference. Occlusion of hand blood flow was accomplished by a wrist cuff inflated 100 mmHg above SBP at least one min before measurement of FBF. FBF was measured six times at baseline directly prior to LBNP and 30 sec, 1, 3,

17 Cardiovascular responses to hypovolemic circulatory stress in women

6 and 8 min after institution of LBNP, as well as 1, 2 and 4 min after LBNP termination. Simultaneously, BP was measured in the contra‐lateral arm and FVR as well as forearm vascular conductance (FVC) was calculated:

FVR = MAP/FBF

FVC = FBF/MAP

140 11 22 44

HR 120 (%) 100 105 SBP 100 (%) 95 90 110 DBP 105 (%) 100 95

100 PP 90 (%) 80

105

MAP100 (%) 95

100 FBF 75 (%) 50 300

FVR 200 (%) 100 8 min Figure 1. Hemodynamic responses to hypovolemic circulatory stress (% of resting values) caused by 8 min LBNP of 11, 22 and 44 mmHg in young men (n=16, mean±SE). HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; PP, pulse pressure; MAP, mean arterial pressure; FBF, forearm blood flow; FVR, forearm vascular resistance.

18 Methods

Changes in calf volume

Mercury‐in‐silicone strain‐gauge plethysmography was used to measure changes in calf volume (ml/100ml). The strain gauge was applied at the maximal circumference of the right calf with the subject in the supine position. In all subjects care was taken to place the calf 5cm below heart level and to avoid any confounding external pressure, the lowest part of the calf was at least 2cm above the floor of the LBNP chamber. Further, the subjects rested in the supine position for at least 30 min to ensure stable calf volume and arterial inflow prior to initiation of LBNP.

LBNP

1 ml/100ml 8 min Figure 2. Original tracing illustrating tissue volume changes in the calf in response to LBNP of 44 mmHg in a 23‐year‐old woman. Initial rapid increase in volume reflects capacitance response of 2.34 ml/100ml. The subsequently much slower but continuous increase reflects a net capillary filtration of plasma fluid into extra‐vascular tissue of 0.163 ml/100ml min‐1.

Calf capacitance response and capillary fluid filtration

At onset of LBNP there is an initial rapid increase of calf volume followed by a slower but continuous increase, and at cessation of LBNP there is a rapid decrease in calf volume (Fig. 2). These different phases reflect (1) capacitance response (blood pooling); (2) net capillary filtration of intravascular fluid to extravascular space; (3) rapid return of pooled blood after termination of LBNP corresponding to (1). This interpretation is aided by the fact that the

19 Cardiovascular responses to hypovolemic circulatory stress in women capacitance response is terminated within approximately 3 min (Lundvall et al. 1993, Schnizer et al. 1978). Thus, the calf capacitance response (ml/100ml) was calculated from the increase in calf volume at onset of LBNP to the line defined from the filtration slope between 3 and 8 min (Fig. 2). Since the compliance of the arterial bed is only ≈3% of that of the venous bed, almost exclusively venous blood is pooled (Rothe 1979), and the capacitance response is denoted venous capacitance response (Fig. 3 A). Net capillary fluid filtration (ml/100ml min‐1) was calculated assessing the increase in calf volume per min after min 3 (Fig 3 B). The total net capillary fluid filtration (ml/100ml) was calculated as rate of filtration times the time of the LBNP stimulus (8 min). The total calf volume increase (ml/100ml) was calculated as venous capacitance response + total net capillary fluid filtration. Further, the time from onset of LBNP to 50 % of the venous capacitance response (Cap 50) was defined. The coefficient of variation (CV) for measurements of calf capacitance response on two separate days was 10.4% in YM and 8.4% in EW while CV for measurements of calf capillary filtration was 9.4 % in YM and 9.6% in EW (LBNP of 44 mmHg).

A B 3 0.20 calf calf

) 1 ‐ 0.15 2 min

filtration, 0.10

(ml/100ml) 1

(ml/100ml 0.05 Venous capacitance,

0 Capillary fluid 0 11 22 44 11 22 44 LBNP (mmHg) LBNP (mmHg) Figure 3. Venous capacitance response (blood pooling) (A) and net capillary fluid filtration (B) in the calf in young men (mean±SE, n=16) in response to LBNP of 11, 22 and 44 mmHg. Both capacitance response and net fluid filtration increased with increasing LBNP levels (P < 0.0001).

20 Methods

Venous compliance

Venous compliance was measured in the calf. Compliance (C, ml/100ml mmHg‐1) is generally ascribed as a change in volume caused by a change in pressure:

C = ΔV/ΔP

In the calf ΔV reflects venous capacitance response (ml/100ml) and ΔP the increase in transmural pressure (80% of the applied negative pressure, mmHg). Olsen and Länne (1998) used negative pressure and applied a linear regression model when studying calf venous compliance at higher transmural pressure gradients (18‐51 mmHg) (Olsen and Lanne 1998). Halliwill et al. (1999) on the other hand developed the thigh cuff occlusion technique with a non‐linear regression model to calculate calf venous compliance also at slightly lower transmural pressure levels (10‐60 mmHg) (Halliwill et al. 1999). We studied calf venous compliance at transmural pressures of 9 to 36 mmHg, and combined these two previous techniques (negative pressure and non‐ linear regression, see the Results section).

Capillary filtration coefficient (CFC)

The capillary filtration coefficient (CFC, ml/100ml min‐1 mmHg‐1) in the calf was calculated as:

CFC = ΔV/ (ΔP x t)

ΔV denotes the total net capillary fluid filtration during LBNP (ml/100ml), ΔP denotes the LBNP induced change in transmural capillary pressure (mmHg), and t denotes time (min).

21 Cardiovascular responses to hypovolemic circulatory stress in women

Changes in upper arm volume during LBNP

Venous capacitance response and net capillary fluid absorption

Changes in upper arm volume were measured by air plethysmography to assess venous capacitance response and net transcapillary fluid absorption in response to LBNP of 11, 22 and 44 mmHg (Olsen et al. 2000). The air pletysmographs were cylindrical, 8cm long and made of transparent plastic. They had openings of different sizes to fit the subject’s upper arm, which was placed at heart level. To avoid venous stasis the size of the openings were chosen to be slightly larger than the arm circumference. The air slits between the skin and pletysmograph were then sealed with a soft latex compound that did not cause any additional pressure, venous stasis or irritation to the skin. The enclosed arm volume was calculated and changes in tissue volume measured with a piston recorder connected to the pletysmographs. Recordings

LBNP 1 ml/100ml

(4) (1) (5)

(3)

(2)

Figure 4. Original tracing illustrating tissue volume changes in the upper arm of a 22‐year‐old man during hypovolemic circulatory stress caused by 8 min of LBNP 44 mmHg. The initial rapid volume decrease reflects mobilization of regional blood from peripheral to central circulation (initial capacitance response, (1)), while the much slower, but continuous decline reflects net transcapillary fluid absorption (2). After cessation of LBNP, there is a rapid return of regional blood volume (final capacitance response (3)), with total volume of absorbed fluid after LBNP termination depicted in (4) and followed by a gradual, slow net capillary re‐filtration of fluid (5).

22 Methods ensured that the enclosed arm segment volume was stable for at least five min before each LBNP initiation. Application of LBNP leads to a rapid decrease in arm volume, followed by a much slower, but continuous decline during LBNP. At termination of LBNP, there is a rapid increase in tissue volume, followed by a slower increase. These different phases reflect (Fig. 4) (1) an initial mobilization of regional blood towards the central circulation (initial arm capacitance response); (2) net capillary absorption of extravascular fluid to intravascular space; (3) rapid recovery of regional blood after termination of LBNP (final capacitance response); (4) total net capillary fluid absorption during LBNP; (5) transcapillary filtration of fluid from the intra‐ to the extravascular space. This interpretation of tissue volume changes during acute hypovolemic circulatory stress has been validated with the aid of simultaneously measured blood and tissue volume changes in both animals (Ablad and Mellander 1963, Mellander 1960), and in humans using technetium marked erythrocytes simultaneously with plethysmographic recordings (Lundvall et al. 1993, Lundvall and Lanne 1989a). The arm capacitance response is fully developed within the first 2 min after institution of LBNP (Lundvall and Lanne 1989a). The net transcapillary fluid absorption (ml/100ml min‐1) was measured as the difference in arm volume before LBNP and one min after termination of LBNP (Fig. 4). Further, the rate of development of initial arm capacitance response was assessed by determining the change in arm volume 10, 20 and 30 sec after LBNP initiation.

Blood samples

Plasma levels of Norepinephrine (P‐NE) was measured both at rest and after four min of LBNP 44 mmHg, since by this time the increase in P‐NE has almost completely developed (Edfeldt 1993). The blood samples were kept on ice, centrifuged within 20 min, stored in a ‐70°C freezer and later analyzed with HPLC technique. The level of HbA1c was examined in EW, to exclude latent diabetes mellitus.

Data recordings

Volume recordings from the arm and calf as well as the pressure in the LBNP chamber were collected and amplified (MP 100A‐CE, Biopac Systems Inc, Santa Barbara, CA) for later analysis with PC software (AcqKnowledge v 3.7.0,

23 Cardiovascular responses to hypovolemic circulatory stress in women

Biopac Systems Inc, Santa Barbara, CA) Blood flow data were collected (Hokanson EC‐6, D.E. Hokanson, Bellevue, WA) for later analysis with PC software (NIVP ver. 5.29B, D.E. Hokanson, Bellevue, WA).

Statistics

Statistical evaluation of the collected data was performed in PC software (Statview ver 5.0.1, SAS Institute and SPSS ver 14, SPSS Inc).

All data are given with reference to soft tissue weight, excluding bone taken as 10% in the calf and upper arm and 13% in the forearm (Cooper et al. 1955, Hafferl and Thiel 1969). Values are expressed as mean ± SE. The significance of difference between and within the groups was principally tested by unpaired Student’s t‐test and paired Student’s t‐test, respectively. Area under the curve (AUC) for blood pressure parameters, FVR and FVC was calculated from institution to termination of LBNP (paper III‐IV + additional study). Repeated measures ANOVA was applied both between and within groups to assess the correlation between CFC and transmural pressure, arm volume recordings and LBNP level, speed of initial arm capacitance response as well as to assess changes in various cardiovascular parameters during LBNP and followed by Tukey’s simultaneous post hoc test when appropriate (paper I‐IV + additional study). When calculating compliance, a regression equation was adjusted to each subject’s own volume‐pressure data and β0, β1 and β2 stored for group comparison with unpaired Student’s t‐test (paper I‐III). Coefficient of variation, CV (%), was calculated on two different days. In paper II and IV, FVC rather than FVR was applied in between group measurements, since FVC is a better marker of vascular response when MAP differs between the two groups (DʹAlmeida and Lautt 1992). Simple regression analysis was applied to assess the association between FVR/FVC and P‐NE, both at rest and during LBNP of 44 mmHg (paper III‐IV). Multiple regression analysis was used to compare the slope of the regression lines between groups (paper III). Statistical significance was set to P < 0.05.

* † # denotes P < 0.05; ** †† ## denotes P < 0.01; *** ††† ### denotes P < 0.001; **** denotes P < 0.0001.

24 Results

RESULTS

Calf volumetric responses (paper I-IV)

Calculation of calf venous compliance (paper I-III)

Calf venous compliance (C, ml/100ml mmHg‐1), was calculated using a combined method of previously utilized techniques (Halliwill et al. 1999, Olsen and Lanne 1998). In paper I‐IV the studied pressure interval was relatively low, and the resulting capacitance‐pressure curve to different levels of lower body negative pressure (LBNP) was clearly non‐linear (e.g. Fig. 5 A), with larger volume changes (greater compliance) at lower transmural pressures as described by a quadratic regression equation:

Δ Calf volume = β0 + β1 x (transmural pressure) + β2 x (transmural pressure) 2

In this equation, β0 is the y‐intercept, and β1 and β2 are characteristics of the volume‐pressure curve. Since compliance is dependent on prevailing pressure, no single value can characterize the slope of this relation. To simplify data presentation, the first derivative of the volume‐pressure curve was calculated, creating a linear compliance‐pressure curve (e.g. Fig. 5 B):

C = β1 + 2 x β2 x transmural pressure

The slope of this curve equals the derivative of the compliance‐pressure curve:

Slope = 2 x β2

C, the two components β1 and β2 as well as the slope were used to determine differences in calf venous compliance.

25 Cardiovascular responses to hypovolemic circulatory stress in women

Sex-related differences in calf venous compliance and capacitance (paper I and III)

Calf venous capacitance increased with increasing LBNP in both young women (YW) and young men (YM) (P < 0.0001), but was smaller in YW than in YM at all LBNP levels (P < 0.01 overall, Fig. 5 A). The rate of capacitance response development (Cap 50) was equal in YW and YM at all LBNP levels. Calf venous compliance was greater at small transmural pressures and decreased with increasing pressures in both YW and YM (P < 0.001, Fig. 5 B) and compliance measured as β1 as well as β2 and slope was smaller in YW than in YM, (each P < 0.05, Fig. 5 B).

A calf

3

** 2

(ml/100ml) 1

Capacitance response, 0 B 0.12 ) calf 1

‐ * 0.08

0.04 (ml/100ml mmHg

Venous compliance, 0 001020304010 20 30 40 Transmural pressure (mmHg) Figure 5. (A) Venous capacitance response in the calf in relation to transmural pressure changes induced by LBNP in young women (YW, white dots, dashed lines) and young men (YM, black boxes, solid lines), mean±SE. Venous capacitance response was lower in YW than in YM (P < 0.01). (B) Corresponding compliance‐pressure curves in YW and YM. It is obvious that compliance decreases with increasing transmural pressure in both groups (P < 0.001). Venous compliance was greater in YM at lower transmural pressures (P < 0.05). * denotes sex‐related differences.

26 Results

Age-related effects on calf venous compliance and capacitance (paper II)

Calf capacitance response increased with increasing LBNP also in elderly women (EW) (P < 0.0001), and was similar to calf capacitance response in YW at all LBNP level and overall (Fig. 6 A). The rate of capacitance response development (Cap 50) was also equal at all LBNP levels in EW and YW. Calf venous compliance was greater at small transmural pressures also in the EW and decreased with increasing pressure (P < 0.001). No change in calf venous compliance was seen with increasing age (Fig. 6 B).

A calf

3

2

(ml/100ml) 1

Capacitance response, 0 B 0.10 ) calf 1

‐

mmHg 0.06

0.02 (ml/100ml

Venous compliance, 0 001020304010 20 30 40 Transmural pressure (mmHg)

Figure 6. (A) Venous capacitance response in the calf in relation to transmural pressure changes evoked by LBNP in elderly women (EW, black boxes, solid lines) and young women (YW, white dots, dashed lines), mean±SE. No age‐related difference in venous capacitance response was seen. (B) The corresponding venous compliance‐pressure curves in EW and YW. No age‐related change in compliance was detected.

27 Cardiovascular responses to hypovolemic circulatory stress in women

Sex-related differences in capillary fluid filtration and capillary filtration coefficient (CFC) (paper I and III)

The net capillary fluid filtration in the calf increased with increasing LBNP in both YW and YM (P < 0.0001), and was greater in YW at LBNP of 11 and 22 as well as overall (P < 0.05, Fig. 7). CFC was unaffected by increasing transmural pressure in both YW and ,YM and was 0.0043±0.0002 in YW and 0.0036±0.0002 (ml/100ml min‐1 mmHg‐1) in YM, being greater in YW (P = 0.02).

0.20 * calf

) 1 ‐ 0.15 min

filtration,

0.10 *

*

(ml/100ml 0.05

Capillary fluid 0 11 22 44 LBNP (mmHg) Figure 7. Net capillary fluid filtration in the calf in response to LBNP in young women (YW, white bars) and young men (YM, black bars), mean±SE. Net fluid filtration was larger in YW at LBNP 11 and 22 mmHg as well as overall (P < 0.05). * denotes sex‐related differences.

Age-related differences in capillary fluid filtration and CFC (paper II)

The net capillary fluid filtration in the calf increased with increasing LBNP also in EW (P < 0.0001). Net fluid filtration was smaller in EW than YW at LBNP of 11 and 22 mmHg (each P < 0.001), but similar at LBNP of 44 mmHg (P = 0.93, Fig. 8 A). In agreement, CFC was also reduced at LBNP of 11 and 22 mmHg (each P < 0.001), but equal at higher LBNP (P = 0.93). CFC was dependent on the prevailing transmural pressure in EW (P < 0.001) and increased by approximately 1/3 at LBNP of 44 mmHg compared with lower LBNP levels (P < 0.01), while CFC was pressure independent in YW (Fig. 8 B).

28 Results

A B 0.20 160 calf

) 1 ‐ 0.15 140 min

filtration, calf calf

0.10 120 (%) *** *** (%) ††† CFC, CFC,

(ml/100ml 0.05 100

Capillary fluid 0 80 11 22 44 02010 30 40 LBNP (mmHg) Transmural pressure (mmHg) Figure 8. (A) Net capillary fluid filtration in the calf in response to LBNP in elderly women (EW, black bars) and young women (YW, white bars), mean±SE. Net fluid filtration was reduced with age during LBNP 11 and 22 mmHg (P < 0.001), but equal at LBNP 44 mmHg (P = 0.93). (B) Change in CFC (%) with increasing transmural pressure in EW (black boxes, solid lines) and YW (white dots, dashed lines). A positive association between CFC and transmural pressure was found in EW (P < 0.001), and CFC increased 1/3 at high transmural pressure in EW (P < 0.01). * denotes age‐related differences, † denotes change in CFC with increasing transmural pressure.

The effect of capillary fluid filtration on compliance calculations (paper I and II)

The impact of capillary fluid filtration on calf compliance calculations was studied using either calf venous capacitance volume (net capillary filtration excluded) or total calf volume (not excluding net capillary filtration) in the volume‐pressure relationship. Either β1 or β2 were significantly affected by net capillary filtration in all groups studied (YW, YM and EW, Table 2).

Table 2. The effects of net capillary fluid filtration on compliance calculations.

Parameter β1 β2 Group Ven cap Total vol Diff (%) P Ven cap Total vol Diff (%) P YW 0.100 0.124 ‐24 0.0001 ‐0.0009 ‐0.0007 22 0.08 YM 0.130 0.149 ‐15 0.02 ‐0.0015 ‐0.0012 21 0.08 EW 0.092 0.094 ‐2 0.73 ‐0.0008 ‐0.0001 87 0.004

YW, young women; YM, young men; EW, elderly women. β1 and β2 are parameters of the compliance equation calculated using either calf venous capacitance (Ven cap, filtration excluded) or total calf volume increase (Total vol, filtration not excluded). Diff,

difference between β1 and β2 using Ven cap and Total vol in the compliance equation.

29 Cardiovascular responses to hypovolemic circulatory stress in women

Total hypovolemic response (paper I-IV)

Total calf volume increase (i.e. venous capacitance response + total filtration) during LBNP was equivalent between YW and YM as well as YW and EW at all LBNP levels (Table 3).

Table 3. Total calf volume increase Total calf volume increase Group 11 mmHg 22 mmHg 44 mmHg YW 1.00±0.05 1.94±0.08 3.47±0.11 YM 1.16±1.16±0.090.09 2.22±0.14 3.77±0.20 EW 0.90±0.08 1.72±0.10 3.30±0.14 Total calf volume increase (venous capacitance response + total net fluid filtration) in response to LBNP in young women (YW), young men (YM) and elderly women (EW). No sex‐ or age‐related differences were seen at any LBNP level.

Cardiovascular responses to acute hypovolemia

Assessment of arm capacitance response (paper III-IV)

The initial capacitance response, final capacitance response as well as net capillary fluid absorption were measured in the upper arm segment in response to LBNP (fig. 9). However, in many registrations there was a gradual reduction of the capacitance response with time during LBNP (Fig. 9 B). To be able to separate the capacitance response and fluid absorption and define the initial capacitance also in these curves, the following assumptions were applied in all registrations: first, the maximal arm volume reduction during the first two min of LBNP was identified. Then the total net capillary fluid absorption (4) was added, defining the total volume reduction at termination of LBNP (X, Fig. 9 B). From X a tangent was drawn adjoining the lowest part of the volume curve during the initial two min of LBNP, defining the initial capacitance response (Fig. 9 B).

30 Results

A LBNP B LBNP 1 ml/100ml 0.56 1.15; (4) 1.44; (1) (5) 1.57

1.14

1.43; (3) x (2)

Figure 9. (A) Original tracing illustrating tissue volume changes in the upper arm of a 22‐year‐old man during hypovolemic circulatory stress caused by 8 min of LBNP 44 mmHg (see fig. 4 for details). (B) Original tracing illustrating tissue volume changes in the upper arm during 8 min of LBNP 44 mmHg of a 21‐year‐ old woman. After the initial capacitance response (mobilization of blood to the central circulation), there is a gradual reduction of the capacitance during LBNP, resulting in a decreased final capacitance response. See text for further details.

Sex-related cardiovascular responses (paper III)

The maximal cardiac and peripheral responses to 8 min of LBNP 44 mmHg can be seen in Table 4. Changes in heart rate (HR) and blood pressure parameters were equal, while YW responded with smaller increase in forearm vascular resistance (FVR), but with increased Plasma Norepinephrine (P‐NE) (P < 0.05).

Table 4. Maximal cardiovascular responses evoked by LBNP of 44 mmHg. Group HR (%) SBP (%) DBP (%) MAP (%) PP (%) FVR P(%) P‐NE (%) YW 137±4 94±1 104±1 100±1 77±3 180±11* 189±16* YM 135±4 94±1 109±2 102±1 74±3 255±39 133±9 Percentage of resting values, mean ± SE. YW, young women; YM, young men. HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure; PP, pulse pressure; FVR, forearm vascular resistance; P‐NE, plasma Norepinephrine. * denotes sex‐related didifferences,fferences, P < 0.05.

31 Cardiovascular responses to hypovolemic circulatory stress in women

Figure 10 A shows the change in FVR (%) during LBNP of 44 mmHg in YW and YM. YW responded with reduced FVR during the first 3 min of LBNP (P < 0.05) as well as overall (P < 0.01) compared with YM. Figure 10 B shows the association between the increase in FVR (AUC) and P‐NE (% from resting value) in YW and YM during LBNP of 44 mmHg. No association was found in YW (R2 = 0.01, NS) in contrast to in YM (R2 = 0.59, P < 0.05). Further, the slope of the regression line was less steep in YW than in YM (P = 0.002), i.e. a similar increase in P‐NE in both sexes seemed to generate a much greater increase in FVR in YM than in YW. Also at rest P‐NE was associated with FVR in YM (R2 = 0.67, P < 0.05), but not in YW (R2 = 0.03, NS).

A ** B † 300 * * 1200 mmHg mmHg

800 44 44

units) 200 * resting values)

LBNP

LBNP 400

(FVR

of

(% FVR, 100 FVR, 0

0 4 8 0 100 200 Time (min) P‐NE increase, LBNP (%) Figure 10. (A) Increase in forearm vascular resistance (FVR, %) during hypovolemic circulatory stress caused by 8 min LBNP of 44 mmHg in young women (YW, white dots, dashed lines), and young men (YM, black boxes, solid lines), mean±SE. YW had smaller increases in FVR during the first three min of LBNP (P < 0.05) as well as overall (AUC, P < 0.01). (B) FVR increase (AUC) in relation to Plasma Norepinephrine (P‐NE) increase (%) during hypovolemic circulatory stress caused by 8 min LBNP of 44 mmHg. No association was seen between FVR and P‐NE in YW in contrast to in YM (R2 = 0.59, P < 0.05). * denotes sex‐related differences, † denotes a positive association between FVR and P‐NE.

32 Results

Sex-related defence of central blood volume (paper III)

Figure 11 A shows the initial arm capacitance response (mobilization of venous capacitance blood to central circulation) during LBNP in YW and YM. The initial response was well maintained in YW, being greater at LBNP 11 mmHg (P < 0.001), and subsequently increased similarly when increasing the hypovolemic stimulus (P < 0.0001). Furthermore, YW responded with a faster increase in arm capacitance response than YM at LBNP of 44 mmHg (P < 0.01, not shown). However, the capacitance response was less maintained with time in YW than in YM at all LBNP levels (Fig. 11 B).

A B 1.8 0 arm 10 (%) 1.2 *** capacitance,

(ml/100ml) ** 30 0.6 arm capacitance response, ** during LNBP Initial 0 Decrease in 50 * 11 22 44 11 22 44 LBNP (mmHg) LBNP (mmHg) Figure 11. (A) Initial arm capacitance response (mobilization of peripheral blood to the central circulation) during hypovolemic circulatory stress caused by LBNP in young women (YW, white bars) and young men (YM, black bars), mean±SE. The initial capacitance response was greater in YW during LBNP of 11 mmHg (P < 0.001). (B) The decrease in capacitance response (%) with time during 8 min of LBNP. The decrease was greater in YW at all LBNP levels (P < 0.05 and P < 0.01). * denotes sex‐related differences.

Figure 12 shows the net capillary fluid absorption during LBNP in YW and YM, which increased with increasing LBNP in both groups (P < 0.0001). Capillary fluid absorption was lower in YW at LBNP of 44 mmHg as well as overall (P < 0.05). Furthermore, YM responded with a more prominent increase in net capillary fluid absorption with increasing hypovolemia than YW (P < 0.05).

33 Cardiovascular responses to hypovolemic circulatory stress in women

* ) 1

‐ 0.12 min

absorption, *

0.08 (ml/100ml 0.04 arm capillary fluid

Net 0 10 20 30 40 50 LBNP (mmHg) Figure 12. Net capillary fluid absorption during hypovolemic circulatory stress caused by LBNP in young women (YW, white dots, dashed lines) and young men (YM, black boxes, solid lines), mean±SE. The fluid absorption was lower in YW during LBNP 44 mmHg as well as overall (P < 0.05). Further, YW responded with a less pronounced increase in fluid absorption with increasing hypovolemia than YM (P < 0.05). * denotes sex‐related differences.

Age-related cardiovascular responses (paper IV)

The maximal cardiac and peripheral responses to 8 min of LBNP of 44 mmHg can be seen in Table 5. HR increased less in EW (P < 0.01), while the increase in P‐NE and forearm vascular conductance (FVC) was similar to in YW. Further, a greater increase in MAP and DBP was seen in EW (P < 0.05).

Table 5. Maximal cardiovascular responses evoked by LBNP of 44 mmHg. Group HR (%) SBP (%) DBP (%) MAP (%) PP (%) FVC (%) P‐NE (%) EW 120±2** 95±2 109±2* 103±1* 78±3 59±7 189±20 YW 137±4 94±1 105±1 100±1 77±3 60±3 189±16 Percentage of resting values, mean ± SE. EW, elderly women; YW, young women. HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure; PP, pulse pressure; FVR, forearm vascular resistance; P‐NE, plasma Norepinephrine. * denotes age‐related differences.

Figure 13 shows the maximal change (%) in HR (A) and FVC (B) in EW and YW in response to LBNP of 11, 22 and 44 mmHg. HR increased with increasing LBNP in both groups (P < 0.0001), but EW responded with a less prominent augmentation in HR with increasing hypovolemia (P < 0.01), as well as a smaller HR increase at LBNP of 22 and 44 mmHg (both P < 0.01) than

34 Results

YW. FVC decreased with increasing LBNP levels (P < 0.001), but similarly in EW and YW (P > 0.58) and no FVC difference was seen at any LBNP level.

A ** B 100 (%)

140 (%)

FVC HR

** 80 120 change in change in

** 60 100 Maximal Maximal 10 20 30 40 50 10 20 30 40 50 LBNP (mmHg) LBNP (mmHg) Figure 13. (A) Maximal change in heart rate (HR, %) and (B) forearm vascular conductance (FVC, %) during hypovolemic circulatory stress caused by LBNP in elderly women (EW, black boxes, solid lines) and young women (YW, white dots, dashed lines), mean±SE. HR increased less with increasing hypovolemia in EW and was lower during LBNP 22 and 44 mmHg (P < 0.01). On the contrary, FVC decreased similarly with increasing LBNP level and no age‐related differences were seen. * denotes age‐related differences.

Age-related defence of central blood volume (paper IV)

Figure 14 A shows the initial arm capacitance response during LBNP in EW and YW. Both groups increased their initial capacitance response with increasing hypovolemic stimuli (EW, P < 0.01; YW, P < 0.0001), but the response was much smaller in EW with increasing hypovolemia compared to YW (P < 0.001). Further, initial capacitance response was markedly decreased in EW (P < 0.0001), but EW maintained the capacitance response during LBNP of 44 mmHg, while the YW decreased in capacitance response (EW vs. YW, P = 0.001, not shown). Figure 14 B shows the rate of capacitance response development during the first 30 sec of LBNP of 44 mmHg (% in relation to fully developed capacitance response). EW responded with a much slower increase in capacitance response, especially during the first 10 sec (P = 0.0001).

Figure 15 shows net capillary fluid absorption in the upper arm in EW and YW during LBNP. This increased with increasing hypovolemia in both groups

35 Cardiovascular responses to hypovolemic circulatory stress in women

(EW, P < 0.05; YW, P < 0.0001), but the increase was smaller in EW (LBNP of 0‐44 mmHg, P < 0.05). The fluid absorption was lower at LBNP of 44 mmHg (P < 0.01), as well as overall (P = 0.01) in EW.

A B 1.8 100 **** arm **** 75 1.2 **** **** **** mmHg (%) (ml/100ml)

50

44

capacitance respons, arm 0.6 25 **** LBNP Initial Capacitance response, 0 0 11 22 44 0102030 LBNP (mmHg) Time (sec) Figure 14. (A) Initial arm capacitance response (mobilization of peripheral blood to the central circulation) during hypovolemic circulatory stress caused by LBNP in elderly women (EW, black bars) and young women (YW, white bars), mean±SE. The arm capacitance response was reduced in EW (P < 0.0001). (B) Rate of capacitance response development during the first 30 sec of LBNP 44 mmHg in relation to fully developed capacitance response (%), mean±SE. The response was slower in EW (P < 0.0001). * denotes age‐related differences.

** 0.10 ) 1 ‐ min

absorption,

**

0.05 fluid

(ml/100ml

arm capillary

Net 0 10 20 30 40 50 LBNP (mmHg) Figure 15. Net capillary fluid absorption during hypovolemic circulatory stress caused by LBNP in elderly women (EW, black boxes, solid lines) and young women (YW, white dots, dashed lines), mean±SE. Fluid absorption increased less with increasing hypovolemia in EW (P < 0.05) and demonstrated decreased absorption at LBNP 44 mmHg (P < 0.01), as well as overall (P < 0.01). * denotes age‐related differences.

36 Results

Women prone to vaso-vagal reaction during LBNP (additional study)

All young women examined (n=27) tolerated LBNP of 11 and 22 mmHg well, but five young women (VW) developed signs of a vaso‐vagal reaction during the last five min of LBNP 44 mmHg. At this LBNP level only data from the first three min can be presented in VW, and final capacitance response and net capillary absorption could not be determined.

Calf volume responses in VW

Calf venous capacitance was similar in VW at all LBNP levels compared to YW (Table 6). However, the rate of development of calf capacitance response (Cap 50) was slower in VW (LBNP 22 mmHg, P < 0.05; LBNP 44 mmHg, P < 0.001; overall, P = 0.002, Table 6), with increasing difference with increasing LBNP (P < 0.001).

Calf net capillary fluid filtration as well as CFC was smaller in VW during LBNP of 11 mmHg (P, < 0.05) but similar in VW and YW at higher LBNP as well as overall (Table 6). Total calf volume increase was slightly smaller in VW at LBNP of 11 mmHg (P < 0.05), but similar at higher LBNP levels.

Table 6. Calf responses evoked by LBNP of 11, 22 and 44 mmHg. LBNP 11 mmHg 22 mmHg 44 mmHg Calf responses: VW YW VW YW VW YW Ven Cap (ml/100ml) 0.53±0.08 0.68±0.04 1.31±0.10 1.35±0.07 2.09±0.12 2.26±0.09

Cap 50 (sec) 13±2 10±1 26±1 * 19±1 44±7*** 26±2 Net filtr (ml/100ml min‐1) 0.025±0.005* 0.039±0.002 0.081±0.011 0.073±0.03 0.151±0.017 0.151±0.08 Total vol (ml/100ml) 0.73±0.09* 1.00±0.05 1.96±0.18 1.94±0.08 3.29±0.22 3.47±0.11 Values are mean±SE. VW, women prone to vaso‐vagal reaction; YW, non vaso‐vagal responding women. Ven Cap, Calf venous capacitance response; Cap 50, time to 50 % of the calf venous capacitance response; Net filtr, net capillary fluid filtration; Total vol, total calf volume increase (calf venous capacitance response + total net capillary fluid filtration during 8 min of LBNP). Venous capacitance developed slower in VW seen as greater Cap 50 (LBNP 22 mmHg, P < 0.05; LBNP 44 mmHg, P < 0.001), with the difference increasing with increasing LBNP (P < 0.001). Fluid filtration and total calf volume increase were smaller in VW at LBNP 11 mmHg (P < 0.05) but similar at higher LBNP. * denotes differences between VW and YW.

37 Cardiovascular responses to hypovolemic circulatory stress in women

Cardiovascular responses in VW

The cardiovascular response to LBNP in VW and YW can be seen in Figure 16. At LBNP of 11 mmHg, HR (%) and SBP (%) were unchanged in both VW and YW, and one min after initiation FBF, FVC and FVR were also unchanged in VW (FBF and FVR, P > 0.92, FVC > 0.72), while a substantial decrease in FBF and FVC with concomitant increase in FVR was seen in YW (P = 0.0006 and P = 0.002, respectively). Further, FVC was less reduced in VW compared to YW during the first 3 min (P < 0.05, LBNP 11 mmHg). With increasing hypovolemia, VW decreased faster in SBP and PP (%, both P < 0.05). Further, the decrease in SBP (%, P < 0.01) and PP (%, P < 0.05) was greater in VW at LBNP of 22 mmHg and 44 mmHg, respectively.

11 22 44 Figure 16. Hemodynamic responses to 140 hypovolemic circulatory stress (% of resting HR 120 values) caused by 8 min LBNP of 11, 22 and (%) 100 44 mmHg in women prone to vaso‐vagal 105 reaction (VW) and young women (YW), SBP 100 mean±SE. HR, heart rate; SBP, systolic (%) 95 blood pressure; DBP, diastolic blood ** pressure; PP, pulse pressure; MAP, mean 90 110 arterial pressure; FBF, forearm blood flow; FVR, forearm vascular resistance. No DBP 105 change in FBF and FVR was seen after (%) 100 initiation of LBNP 11 mmHg in VW in 95 contrast to the substantial change seen in 100 YW (P < 0.0006 and P < 0.002, respectively). PP SBP and PP decreased more in VW at LBNP (%) 80 * 22 mmHg (P < 0.01) and 44 mmHg 60 * 105 (P < 0.05), respectively, and both SBP and PP decreased faster with increasing MAP 100 hypovolemia (P < 0.05) in VW. * denotes (%) differences between VW and YW. † denotes 95 120 changes from resting values in YW.

100 FBF (%) 80 ††† 60 200

FVR 150 †† (%)

100 8 min

38 Results

Defence of central blood volume in VW

Figure 17 A shows the initial arm capacitance response during LBNP in VW and YW. VW responded with decreased initial capacitance response at LBNP of 44 mmHg as well as overall (P < 0.05). Figure 17 B shows the net capillary fluid absorption in the upper arm in VW and YW. VW presented a trend towards lower net absorption than YW at LBNP of 22 mmHg (P = 0.13).

A * B

1.8 * 0.12 ) 1 ‐ min

1.2 absorption, 0.09 (ml/100ml)

0.6 (ml/100ml 0.03 arm arm capacitanceInitial response, 0 Net capillary fluid 0 11 22 44 11 22 44 LBNP (mmHg) LBNP (mmHg) Figure 17. Initial arm capacitance response (mobilization of peripheral blood to the central circulation) during hypovolemic circulatory stress caused by LBNP in women prone to vaso‐vagal reaction (VW, black bars) and young women (YW, white bars), mean±SE. The arm capacitance response was reduced in VW at LBNP of 44 mmHg as well as overall (P < 0.05). (B) Net capillary fluid absorption during LBNP in VW and YW. Net fluid filtration seemed to be reduced in VW, but this did not reach significance. * denotes differences between VW and YW.

39 Cardiovascular responses to hypovolemic circulatory stress in women

DISCUSSION

Venous capacitance and compliance

The venous section of the cardiovascular system can be regarded as a capacious blood reservoir containing ~70% of the total blood volume with another 15% in the heart and lungs. The systemic venous section has ~30 times greater compliance than arteries, designed to preserve a proper inflow of blood to the heart and arterial section during various cardiovascular adjustments (Rothe 1979, Rowell 1993). Thus, central venous pressure (CVP) and filling of the heart may be maintained at a fairly stable level, despite variations in venous blood volume. The great venous compliance is the main mechanism for the pooling of blood in dependent regions, e.g. during standing, head‐up tilt (HUT) or lower body negative pressure (LBNP), and an increased calf venous compliance is associated with increased venous capacitance response, concomitant decrease in central blood volume and increased intolerance to LBNP (Morikawa et al. 2001, Olsen et al. 2000, Tsutsui et al. 2002).

Venous compliance is described as the relationship between change in venous volume and distending (transmural) pressure. At low pressure levels, the slope of the volume‐pressure curve is steep, and the compliance in the vein is high, meaning that a small change in pressure accompanies a large change in volume. At higher pressure the slope is less steep and compliance is lower (e.g. Fig. 5) (Wesly et al. 1975). This is because the early expansion of the veins involves no actual stretch of the elastic walls, but rather acts through a change in the geometry of the veins (Oberg 1967, Rajagopalan et al. 1979). Once the veins assume a circular cross‐section, subsequent increases in their transmural pressure are opposed by the development of increased tension in the walls. The volume‐pressure curve of a whole limb at rest represents the distributed properties of all veins (venules to large veins). In the present investigation, we studied a transmural pressure interval of 9‐36 mmHg where the volume‐ pressure relationship was clearly non‐linear (Figs. 5 A and 6 A). Accordingly,

40 Discussion we used a non‐linear regression equation model characterizing the volume‐ pressure curve to calculate calf venous compliance (Figs. 5 B and 6 B).

Women are known to be more susceptible to orthostatic stress than men (Convertino 1998, Franke et al. 2003, Fu et al. 2004, Gotshall 2000, White. et al 1996), and one possible mechanism might be increased blood pooling in the lower part of the body, leading to greater central hypovolemia. As expected venous compliance was high at low transmural pressure gradients, but gradually decreased with increasing pressure. However, calf venous compliance and venous capacitance were decreased in women at lower transmural pressures, in agreement with earlier studies (Meendering et al. 2005a, Monahan and Ray 2004). At higher transmural pressures relevant to standing and HUT on the other hand, women may in fact have greater venous compliance and blood pooling (Fig. 5 B). Differences in unstressed venous volume (V0) could influence the measured capacitance response, but the adopted protocol makes this unlikely. Resting blood flow to the lower limbs (Holland et al. 1998, Lewis et al. 1986) as well as LBNP induced increase in total peripheral resistance during seems to be similar in women and men (Fu et al. 2004, White et al. 1996) and concomitantly no sex difference was found in β0. However, we found reduced forearm vascular resistance (FVR) in women during LBNP (Fig. 10 A) (Frey and Hoffler 1988). A similar response also in the lower limbs could have lead to an underestimation of the differences in capacitance response. The underlying sex differences in venous wall architecture are at present unknown but may be linked to sex‐related hormonal influence on collagen/elastin ratio as well as wall thickness.

Calf venous compliance is reduced by ~45% in aging men (Olsen and Lanne 1998). There are reasons however to suspect a sex‐related difference in venous compliance with increasing age, due to the increased compliance in young men compared to women (Fig. 5 B) (Meendering et al. 2005a, Monahan and Ray 2004), and arterial compliance demonstrates a slower decrease in aging women compared to men (Debasso et al. 2004, Tomiyama et al. 2003). Venous compliance and capacitance was unchanged with age in women (Fig. 6 A and B). This could be a result of a slightly larger body mass index (BMI) in the elderly with differences in soft tissue‐to‐bone ratio, overestimating capacitance response in the elderly women, but the significance of this confounder seems minor since the difference in BMI was small (Table 1). Training increases calf venous compliance (Hernandez and Franke 2005, Monahan et al. 2001, Morikawa et al. 2001), but a possibly more sedentary lifestyle in elderly

41 Cardiovascular responses to hypovolemic circulatory stress in women women would have decrease their venous compliance. Again, the adopted protocol makes differences in V0 unlikely, and no difference in β0 between young and elderly women was found. The conclusion that venous compliance and capacitance is maintained in aging women seems credible and contrasts earlier findings in elderly men, where calf venous compliance is decreased (Monahan et al. 2001, Olsen and Lanne 1998, Tsutsui et al. 2002), possibly reflecting the effect of estrogen and its impact on cellular transcription of elastin and collagen in vascular smooth muscle cells (Kappert et al. 2006, Knaapen et al. 2005, Mendelsohn 2002).

Capillary fluid filtration in the calf and capillary filtration coefficient (CFC)

Apart from blood pooling, net capillary fluid filtration substantially increases the hypovolemic stimulus during LBNP (Fig. 2, Table 6) (Lanne and Olsen 1997, Lundvall et al. 1993). The hemodynamic importance is indicated in patients affected by the postural orthostatic tachycardia syndrome (POTS) where fluid filtration in lower limbs is increased (Stewart 2003). Net capillary fluid filtration is a resultant of hydrodynamic conductivity, measured as CFC, as well as increased transmural pressure gradients over the micro‐vessel walls, creating a transcapillary driving force. The pressure gradients are dependent on α‐ and β‐adrenergic adjustment of the pre‐ to post‐capillary resistance ratio (Lundvall and Hillman 1978, Maspers et al. 1990). The common view is that CFC is influenced by variations in the number of perfused capillaries due to local myogenic as well as axon reflex responses reacting on transmural pressure changes in the microcirculation (Folkow and Mellander 1970, Henriksen and Sejrsen 1977, Lundvall and Lanne 1989b, Sexton et al. 1994). CFC in men was of comparable magnitude as earlier described with similar techniques (Lanne and Olsen 1997, Mellander et al. 1964), but seemed unaffected by increasing transmural pressures from 9 to 36 mmHg as well as by changes in peripheral resistance, in line with the findings of Bentzer et al. (2001), who found CFC to be unaffected of the number of perfused capillaries (Bentzer et al. 2001).

Net capillary fluid filtration as well as CFC in the calf was ~20% higher in young women than men (Fig. 7). This is in accordance with findings by Huxley et al. (2005), who studied coronary micro‐vessel permeability in a large animal model, and found an increased permeability to proteins in micro‐

42 Discussion vessels in female pigs after the administration of aldosterone (Huxley et al. 2005). Possible mechanisms behind the sex difference could be a decreased α‐ adrenergic function in young women seen as their smaller increase in FVR compared to men during LBNP (Fig. 10 A). Furthermore, estrogen affects capillary fluid transfer. Tollan et al. (1992) proposed a direct effect of estrogen on capillary protein permeability, increasing filtration capacity (Tollan et al. 1992) and atrial natriuretic peptide (ANP) affects capillary filtration by increasing CFC and/or protein permeability, further augmented by estrogen (Groban et al. 1990, Stachenfeld et al. 2001, Wijeyaratne and Moult 1993).

Net capillary fluid filtration and CFC in the calf was reduced by 25‐30% in elderly compared to young women at a transmural pressure increase up to 18 mmHg (Fig. 8 A). The change in plasma Norepinephrine (P‐NE), forearm blood flow (FBF) and forearm vascular conductance (FVC) during LBNP was similar in both groups reflecting a maintained α‐adrenergic function with age (Table 5). A decreased visco‐elasticity of the calf skeletal muscle with age as shown in men might delay the capacitance response beyond the applied cut‐ off point between capacitance and filtration (3 min after initiation of LBNP) due to a slower pressure change in the tissue (Olsen and Lanne 1998). This seems unlikely since the time to 50% of the calf venous capacitance response (Cap 50) was equal in elderly and young women. Also, if a decrease in visco‐ elasticity would still be present, this would result in an overestimation of fluid filtration in the elderly women, indicating a larger difference than presented in this study. A defective transmission of negative pressure into the tissue would result in a reduction of fluid filtration. However, transmission of negative pressure into the calf is not affected by age (Olsen and Lanne 1998).

All of the elderly women were in a postmenopausal, estrogen deficient state (Berne et al. 2005). Because of the vasodilatory effects of estrogens as well as its effects on capillary protein permeability it seems reasonable to assume that this is one of the main mechanisms for the age‐related reduction in fluid filtration as well as CFC (Gorodeski et al. 1995, Groban et al. 1990, Huang et al. 1997, Pechere‐Bertschi et al. 2000, Stachenfeld et al. 2001, Sudhir et al. 1997, Tollan et al. 1992, Wijeyaratne and Moult 1993). This is corroborated by the fact that women on hormone‐replacement therapy (HRT) improve their endothelial function (Saitta et al. 2001). The reactivity of the arterioles seems to be impaired due to changed cellular mechanisms with endothelial and smooth muscle dysfunction, which might induce heterogeneity in flow between different capillaries (Muller‐Delp 2006). Further, capillary density in skeletal

43 Cardiovascular responses to hypovolemic circulatory stress in women muscle is also reduced with age and capillary basement membranes thicken in dependent regions because of long‐lasting increases in capillary pressure, indicating a loss of capillary function (Clough 1987, Croley et al. 2005, Harris 2005).

An interesting observation was that CFC was augmented by approximately 1/3 when increasing the applied transmural pressure from 18 to 36 mmHg in elderly women (Fig. 8 B). Increased fluid permeability at high micro‐vascular transmural pressures has been reported by several authors working on vascular bed preparations in isolated limbs as well as on single micro‐vessels (Neal and Michel 1996, Rippe et al. 1985, Wolf et al. 1989). These observations have been interpreted as the “stretched pore phenomenon”, caused by passive stretching of the micro‐vascular membrane with the formation of gaps in or between endothelial cells, preferentially in the venules where the inter‐ endothelial junctions appear to be less tight than in the true capillary section (Frokjaer‐Jensen 1982). Another possibility is that the imposed pressure distension more efficiently opens all micro‐vessels to flow (Rippe et al. 1985). The “stretched pore phenomenon” has been shown to be reversible with time after reduction of high micro‐vascular pressure (Neal and Michel 1996, Rippe et al. 1985, Wolf et al. 1989). Most studies have shown capillary walls to tolerate pressure elevations far beyond their physiological range without an increase in permeability and it is at present unknown whether capillary walls become more fragile, or if cellular junctions in the micro‐vessels become less tight with age. The increasing CFC with transmural pressure in the lower limbs of elderly women might give additional insight into the preponderance of leg oedema besides earlier known factors such as cardiac and venous insufficiency.

When calculating calf venous compliance it is of fundamental importance to exclude net capillary fluid filtration from total calf volume increase and define calf venous capacitance, since total filtration increases calf volume by another 50% during such a short time as 8 min in young women (Table 6). The venous congestion technique, developed by Halliwill et al. (1999) measures calf compliance based on calf volume decrease during one min of linear cuff deflation from 60 to 10 mmHg after 4 to 8 min of venous stasis, without separating capillary filtration from the capacitance response (Halliwill et al. 1999). We found significant changes in the calf compliance calculations if net capillary filtration was not excluded from total calf volume increase (Table 2).

44 Discussion

Cardiovascular responses to hypovolemic circulatory stress

Haemorrhage is a leading cause of death in trauma (Becker et al. 2002, Sauaia et al. 1995) and LBNP is an excellent model to simulate acute haemorrhage and hypovolemic circulatory stress in humans, by unloading the baroreceptors due to central hypovolemia (Convertino et al. 2008, Cooke et al. 2004). Young womene ar more susceptible to hypovolemic circulatory stress than young men (Convertino 1998, Franke et al. 2003, Fu et al. 2004, Gotshall 2000, White et al. 1996). The underlying mechanisms are not clear however, and are probably multi‐factorial (Fu et al. 2004). Total calf volume increase was similar in young women and men during 8 min of LBNP, creating comparable central hypovolemic stress, with the increased calf net capillary fluid filtration in young women compensating for the decreased capacitance response (Table 3).

P‐NE is a good marker of overall sympathetic response (Goldstein et al. 1983, Hjemdahl et al. 1989) and young women increased more in P‐NE than men during LBNP. Despite this, they increased less in FVR during hypovolemic stress (Table 4, Fig. 10 A) and no association between P‐NE and FVR was seen in contrast to in men. Furthermore, an equal increase in P‐NE seems to generate a smaller FVR increase in young women than in young men (Fig. 10 B). A less pronounced increase in arterial vasoconstriction to an infusion of α‐ receptor agonists (e.g. norepinephrine, NE) has earlier been reported in women, possibly a result of decreased sensitivity or density of peripheral vascular adrenergic receptors in women (Bowyer et al. 2001, Freedman et al. 1987, Kneale et al. 2000). Kneale et al. (2000) extended these observations by detecting increased vasodilatation in women to infusion of β‐receptor agonists and when NE was co‐infused with a β‐receptor antagonist, the sex difference in vasoconstriction disappeared. Thus, P‐NE in women might have a relatively higher affinity to β2‐ than α1‐receptors, impeding vasoconstriction (Kneale et al. 2000). Estrogen may also attenuate forearm vasoconstrictor responses to infusion of NE in women (Sudhir et al. 1997).

Initial arm capacitance response was well initiated in young women, but it was not maintained over time to the same extent as in young men (Fig. 11 A and B). Despite an increased CFC, net capillary fluid absorption from tissue to blood during hypovolemic stress was lower in young women (Fig. 12). A possible explanation is the smaller α1‐receptor mediated response in young

45 Cardiovascular responses to hypovolemic circulatory stress in women women (Fig. 10 A) (Bowyer et al. 2001, Freedman et al. 1987, Kneale et al. 2000), leading to diminished capillary pressure reduction with impaired transcapillary driving force (Maspers et al. 1990), also indicated by the relatively slower increase in net fluid absorption in women with increasing hypovolemia (Fig. 12). Different contractile responses to neurally released NE have been shown in vessels of various sizes. In small muscular arteries and arterioles, the distance between sympathetic nerve endings and smooth muscle cells is short, and responses to NE fast and initially great, but not maintained over time. In larger arteries, the distance between synapse and smooth muscle cell is greater resulting in a slower and smaller response, but maintained over time (Bevan 1979). It may be hypothesized that women primarily to a larger extent rely on arterial vasoconstriction of smaller arteries, i.e. fast but not maintained, while men rely on vasoconstriction in larger arteries with slower but maintained response over time (Bevan 1979, Ludwig et al. 2001). The faster increase in FVR at LBNP of 11 mmHg in women (not shown), the initially faster but less maintained arm capacitance response as well as the decreased net capillary fluid absorption in women supports this theory (Figs. 11 B and 12).

The earlier onset of hypotension in women during experimental hypovolemic circulatory stress has been associated with a more prominent decrease in stroke volume (SV) and cardiac output (CO) (Convertino 1998, Fu et al. 2004), even more evident during hypovolemic conditions (Custaud et al. 2002, Fu et al. 2005). A hypothesized mechanism for the faster decrease in SV has been a smaller and less distensible heart in women (Fu et al. 2004, Fu et al. 2005). We suggest an alternative explanation; the functional importance of the capacitance response and net capillary fluid absorption during hypovolemic stress might at first glance seem unimpressive, but since the total mass of skeletal muscle and skin in the human body is large (Skelton 1927), the combined effect of the capacitance response and transcapillary fluid absorption in response to hypovolemic stress corresponding to LBNP of 44 mmHg during a limited period of 8 min, could add another 500‐1000 ml to the effective circulating blood volume. A well maintained capacitance response and net capillary fluid absorption seem to constitute very important compensatory responses for the restitution of plasma volume during acute haemorrhage and hypovolemia (Lanne and Lundvall 1992, Lundvall et al. 1993). The inability for young women to uphold SV and CO might thus be due to the more prominent reduction in capacitance response and decreased net capillary fluid absorption, leading to decreased venous return.

46 Discussion

It seems that human evolution would benefit from greater tolerance against haemorrhage in young women since they give birth during which the danger of haemorrhage is prominent, and at first glance the presented data seem to be in contrast with this hypothesis. In pregnancy and especially during the last trimester however, other mechanisms develop that favour tolerance against haemorrhage. Intravascular volume is increased by ~50%, estrogen and progesterone substantially elevated and venous distensibility is increased compared to non‐pregnant women (Barwin and Roddie 1976, Berne et al. 2005, Fawer et al. 1978, Kuczkowski 2004). The beneficial effect of estrogen and progesterone on capillary permeability (Stachenfeld et al. 2001, Tollan et al. 1992) combined with the increased venous blood volume and venous distensibility gives the pregnant woman an efficient defence against a reduction in central blood volume if experiencing substantial blood loss.

Mortality in trauma increases with age, even after correctinge for pr ‐existing medical conditions (George et al. 2003, OʹKeefe et al. 2001, Taylor et al. 2002). A decreased cardiovagal baroreceptor sensitivity (BRS) but preserved sympathetic control of blood vessels has been demonstrated with increasing age in mixed groups of subjects (Brown et al. 2003, Laitinen et al. 1998, Ng et al. 1994, OʹMahony et al. 2000). The elderly women responded with smaller increases in heart rate (HR) compared to young women, even more evident with increasing hypovolemic circulatory stress, while the decrease in FVC was equal, indicating decreased cardiovagal BRS but maintained sympathetically mediated BRS with age in women (Fig. 13 A and B).

A ~50% reduction of the arm capacitance response was found in the elderly women further reduced with increasing hypovolemia compared to young women, in line with earlier findings in elderly men (Fig. 14 A) (Olsen et al. 2000). The slightly larger BMI in the elderly women might lead to differences in soft tissue‐to‐bone ratio between groups. However, this would have lead to an overestimation of the arm capacitance response in the elderly. Body fat (%) increases with age (Bruce et al. 1980), and elderly women might have had smaller V0 due to increased fat‐to‐muscle ratio, potentially underestimating their arm capacitance response. However, the fact that calf capacitance response was maintained with age implies that the importance of this confounder was minor. The discrepancy between the maintained calf venous capacitance response and the reduced compensatory venous capacitance response in mthe ar with age deserves attention (Figs. 6 A and 14 A). In the

47 Cardiovascular responses to hypovolemic circulatory stress in women calf, the applied transmural pressures over the venous walls were 9 to 36 mmHg, and the veins had most likely assumed circular shape. This means that venous compliance as well as the capacitance response to a large extent depends on the elastic materials (preferentially elastin and collagen) in the venous walls. In the arm however, the established venous transmural pressures are much lower during the induced hypovolemic stress (~1‐2 mmHg) (Olsen and Lanne 2000), and venous compliance as well as capacitance might rather reflect changes in geometry of the walls than the elastin‐to‐collagen ratio. Greater inertia due to thickened venous walls and lower visco‐elasticity with age might affect the change of venous geometry during changes in venous pressure (Bouissou et al. 1991). Furthermore, an age‐ related increase in collagen engaged in the tethering of veins could also lead to increased wall inertia with a delay in change of venous geometry, also explaining the much slower development of the arm capacitance response during the first 30 sec of LBNP in elderly women (Fig. 14 B) (Huisman et al. 1999, Olsen et al. 2000).

Net capillary fluid absorption in response to hypovolemic circulatory stress was also decreased in elderly women (Fig. 15). An effective net capillary fluid absorption is dependent on high hydrodynamic conductivity, and CFC was found to be reduced with age. Furthermore, sympathetic activation establishes a reflex reduction in capillary pressure due to adjustments of pre‐ to post‐ capillary resistance ratio caused by adrenergic α‐ and β‐receptors (Lundvall and Hillman 1978, Maspers et al. 1990). An intact α‐receptor function with age was indicated by the maintained FVC response (Fig. 13 B, Table 5). A decreased β‐receptor function has been proposed with age however (Schutzer and Mader 2003, van Brummelen et al. 1981), and this might reduce net capillary fluid absorption even further in elderly women (Fig. 15). The combination of a decrease in mobilization of peripheral blood to central circulation and reduced net capillary fluid absorption in response to hypovolemic circulatory stress could be an important contributing factor to the increased mortality seen in haemorrhage and burn trauma with increasing age (George et al. 2003, OʹKeefe et al. 2001, Taylor et al. 2002). In contrast to elderly women, men seem to respond with maintained net capillary fluid absorption to hypovolemic stress with age (Olsen et al. 2000). This supports recent clinical observations indicating that elderly women may be more susceptible than age‐matched men to hypovolemic stress during major surgery as well as haemorrhage and burn trauma (George et al. 2003, OʹKeefe et al. 2001, Shevde et al. 2002, Toraman et al. 2004).

48 Discussion

Cardiovascular responses to hypovolemia in women prone to vaso-vagal reaction

Syncope is a common clinical problem that affects up to 3.5% of the general population (Savage et al. 1985) and reflects a heterogeneous patient population (Shen et al. 2000). Syncope due to vaso‐vagal reactions typically occurs during standing (Mosqueda‐Garcia et al. 2000). Increased calf venous pooling has been suggestede as on potential underlying mechanism (Hargreaves and Muir 1992), but on the other hand similar decrease in central venous pressure (CVP) has been reported during HUT (Epstein et al. 1968, Mosqueda‐Garcia et al. 1997). Our data indicate that the hypovolemic stimulus was not greater in women prone to vaso‐ vagal reactions (VW) but rather slightly smaller, at least during LBNP of 11 mmHg (Table 6), suggesting other underlying mechanisms responsible for the increased susceptibility to vaso‐vagal reaction. However, not only the amount of venous blood pooled and fluid filtrated from blood to tissue affects the compensatory responses but also the speed by which the hypovolemic stimulus is instituted, and a faster induced hypovolemic stimulus has been shown to elicit a greater circulatory response (Lanne and Lundvall 1992). Despite a similar rate in instituting LBNP, the rate of the calf capacitance response initiation (Cap 50) was much slower in the VW, especially at higher LBNP levels (Table 6). The filling of venous capacitance vessels is dependent on arterial inflow of blood, pressure reduction in calf tissue caused by negative pressure as well as distending properties of the veins (venous walls and tethering of veins). Arterial inflow to the calf was not measured but similar arm and calf blood flow has been reported in women prone to syncope and controls (Stewart et al. 2004), and resting FBF as well as the changes in FBF and FVR at higher LNBP levels were equal in the two groups (Tabled 1 an Fig. 16). A decreased pressure transmission into the calf tissue seems unlikely (Olsen and Lanne 1998), but a reduced viscoelasticity in the calf tissue as well as differences in venous wall properties and tethering of veins might hypothetically play a role for the reduced rate of capacitance initiation. Further studies are needed to clarify these factors.

The decrease in forearm vascular conductance (FVC) was smaller in VW compared with YW during LBNP of 11mmHg (a LBNP level unloading mainly cardiopulmonary receptors) (Mancia and Mark 1983). Further, FBF, FVR as well as FVC were unchanged in VW compared to resting values during the early phase of LBNP of 11 mmHg in contrast to in YW (Fig. 16).

49 Cardiovascular responses to hypovolemic circulatory stress in women

This is in analogy with earlier findings of impaired forearm vasoconstriction and MSNA response during low LBNP levels in subjects prone to syncope, signalling impaired cardiopulmonary baroreflex function (Thomson et al. 1997, Wasmund et al. 2003, Wijeysundera et al. 2001). A reduced venous return to the heart has been suggested as the key factor behind the fall in blood pressure during a vaso‐vagal reaction (Fuca et al. 2008, Fuca et al. 2006). The decreased arm capacitance response as well as the tendency to decreased net capillary fluid absorption found in VW favours this hypothesis (Fig. 17 A and B), further supported by (Mizumaki et al. 1995, Yamanouchi et al. 1996). In daily life, the benefits in increasing venous return by raising the legs in vaso‐vagal subjects has since long been used as effective treatment, restoring consciousness. It is suggested that the slower onset of hypovolemic stimulus as shown in the reduced rate of calf capacitance response initiation might be one factor contributing to the diminished cardiovascular response in subjects prone to vaso‐vagal reactions (Figs. 16 and 17, Table 6) (Lanne and Lundvall 1992).

50 Discussion

Methodological considerations and limitations

When evaluating differences in cardiovascular responses to LBNP it is of importance to consider the hypovolemic stimulus (Olsen et al. 2000), since differences in calf blood pooling and net fluid filtration has been seen with both sex and age (Figs. 5 A, 7 and 8 A) (Meendering et al. 2005a, Monahand an Ray 2004, Olsen and Lanne 1998, Tsutsui et al. 2002). We measured calf volume increase as an indicative of hypovolemic load during LBNP, although blood pooling and net fluid filtration takes place in the whole part of the body enclosed in the LBNP chamber, including the pelvic region. Middle‐aged women seem to have increased pelvic blood pooling during LBNP than age‐ matched men (White and Montgomery 1996). Further, pelvic basal blood volume might be smaller in elderly women secondary to decreased size of the reproductive organs after menopause (Well et al. 2007), possibly causing decreased blood pooling in elderly during LBNP. However, the venous compartments in the legs rather than the pelvic or abdominal region seem to be of hemodynamic importance during LBNP (Halliwill et al. 1998). Thus, changes in calf volume seem to reflect central blood volume well.

The women were scheduled between day 7 and 21 after start of menstruation and estrogen levels were not measured. However, cardiovascular responses to LBNP seem to be unaffected by menstrual phase (Claydon et al. 2006, Frey et al. 1986, Meendering et al. 2005b). Furthermore, venous compliance and capacitance do not change over the course of the menstrual cycle (Meendering et al. 2005a). Ten of the women were using oral contraceptives and the type was not registered. Oral contraceptives can influence the cardiovascular system in many ways, potentially influencing our findings. However, calf venous compliance and capacitance seem to be unaffected by oral contraceptive use (Meendering et al. 2005a) and no differences were found in calf venous compliance, capacitance or in any of the cardiovascular responses measured comparing normally menstruating women (n=12) and women using oral contraceptives (n=10). Further, all elderly women were in an estrogen‐ deplete state. Our study design cannot differentiate between the effects of sex‐ or age‐related structural differences on measured cardiovascular responses from the effect of the diverse estrogen levels in young women. The presented data are likely to represent a general response to hypovolemic circulatory stress in young women compared to men as well as to elderly women.

51 Cardiovascular responses to hypovolemic circulatory stress in women

The group of women prone to vaso‐vagal reaction during LBNP was small (n=5) and as such data must be interpreted with caution, especially the negative findings. Nevertheless, the general findings on the cardiovascular responses fit well into earlier presented findings on vaso‐vagal syncope and we add new data indicating slower development of capacitance response in the calf as well as reduced arm venous capacitance response in women prone to vaso‐vagal reaction.

This thesis focuses on early compensatory responses to experimental haemorrhage (first minutes) while clinical studies on trauma routinely exclude early, pre‐hospital mortalities due to obvious limitations in study design (Gannon et al. 2002, Magnotti et al. 2008, OʹKeefe et al. 2001, Wohltmann et al. 2001). Clinical studies on sex differences in survival after trauma show a disperse pattern, e.g. a two‐fold increase in mortality in premenopausal women after burn trauma (McGwin et al. 2002, OʹKeefe et al. 2001), while others show either no sex advantage (Gannon et al. 2002, Magnotti et al. 2008) or a decreased risk in young women following trauma (Wohltmann et al. 2001). However, in the elderly population trauma uniformly decrease survival, and this may be accentuated in elderly women (refx2). Although it is difficult to compare the presented findings with clinical studies, our data indicate that women might be more susceptible to haemorrhage, especially at older age. This hypothesis is also in line with other experimental studies in humans (Convertino 1998, Franke et al. 2003, Fu et al. 2004, Gotshall 2000, White et al. 1996).

52 Conclusions

CONCLUSIONS

• Young women have lower calf venous compliance and capacitance response to LBNP than young men, at least at low transmural pressures. Calf venous compliance and capacitance are maintained in aging women. • Calf net capillary fluid filtration as well as CFC is higher in young women than in men. Elderly women have decreased filtration and CFC at low transmural pressure, implying reduced capillary function. At higher pressures, however, CFC is increased in elderly women, indicating increased capillary susceptibility to transmural pressure load in dependent regions. • LBNP induces a comparable hypovolemic stimulus in young women and men. Young women demonstrate however smaller vasoconstriction without an association between P‐NE and peripheral vascular resistance in contrast to in men. Further, the larger decrease in arm venous capacitance response with time as well as lower net capillary fluid absorption from tissue to blood indicates less efficiency to defend central blood volume in young women. • LBNP induces a comparable hypovolemic stimulus in elderly and young women. With increasing age however, heart rate increases less but peripheral vascular conductance is maintained in response to hypovolemic stress. Further, the decreased arm venous capacitance response and reduced net capillary fluid absorption from peripheral tissues to central circulation indicates less efficiency to defend central blood volume in elderly women. • LBNP induces a hypovolemic stimulus that is of similar magnitude but the rate of capacitance response in the calf is more slowly instituted in women prone to vaso‐vagal reaction compared with women not experiencing vaso‐vagal reactions. The cardiopulmonary baroreflex seems to be defect, and arm venous capacitance response is reduced in the vaso‐vagal women, pointing towards a reduced defence of central blood volume.

53 Cardiovascular responses to hypovolemic circulatory stress in women

POPULÄRVETENSKAPLIG SAMMANFATTNING

(Summary in Swedish)

Akut blodförlust och nedsatt blodvolym i kärlsystemet (hypovolemi) är en allvarlig komplikation vid olyckor, t ex trafikolyckor och större brännskador samt vid kirurgi. Tidigare experimentella studier har slagit fast att yngre kvinnor är känsligare för akut hypovolemisk stress än yngre män och även svimmar lättare, men den bakomliggande orsaken är inte klarlagd. Med ökande ålder så ökar känslighet för trauma och akut blodförlust hos både män och kvinnor, även efter att man tagit hänsyn till eventuell annan sjuklighet hos de äldre.

Akut blodförlust och hypovolemi leder till sjunkande blodtryck i centrala artärer (ex. aorta) som i sin tur avlastar baroreceptorer och aktiverar det sympatiska nervsystemet. Det initierar direkt en serie kardiovaskulära försvarsmekanismer som verkar i olika tidsfaser för att upprätthålla stabilt blodtryck och omfördela blodflödet till centrala organ. De viktiga mekanismerna i det akuta skedet består av en omedelbar ökad hjärtfrekvens och en generell kärlsammandragning (vasokonstriktion) i artärsystemet, som ger ett minskat blodflöde i perifera vävnader (armar och ben). Det leder vidare till en mycket snabb (sekunder) mobilisering av venblod från den perifera till den centrala cirkulationen, ett ökat blodtryck, samt omfördelning av blodvolymen till vitala områden som hjärna och hjärta. Under de första minuterna så förflyttas vätska från vävnader, framförallt muskulatur och hud, till blodbanan via kapillärerna för att på så sätt öka den cirkulerande blodvolymen. Totalt kan dessa två system på 5‐10 minuter öka den effektivt cirkulerande blodvolymen med minst en liter vid en stor blodförlust.

Den venösa sidan av det kardiovaskulära systemet består av en blodreservoar som innehåller ac: 70 % av den totala blodvolymen, och är designad för att kunna reglera inflödet av blod till hjärtat och artärsystemet under olika kardiovaskulära påfrestningar. En avgörande faktor för att blod effektivt ska kunna mobiliseras till den centrala cirkulationen är att venerna har en hög

54 Summary in Swedish eftergivlighet (hög compliance). Tillsammans med den stora kapaciteten som reservoaren har medför att även små tryckförändringar i perifera vener leder till en väsentlig ökning av den centrala blodvolymen.

I denna avhandling så har en minskning av central blodvolym (i bröstkorgen) åstadkommits experimentellt med hjälp av undertryck applicerat kring underkroppen (lower body negative pressure, LBNP) vilket ger definierade tryckgradienter (skillnad i tryck) över venväggarna. Det leder till en vidgning av venerna och ansamling av blod i nedre delen av kroppen (kapacitanssvar). Dessutom kommer vätska filtreras ut från blod till vävnad över kapillärväggarna (kapillär vätskefiltration). Bägge processerna leder till central hypovolemi dsom vi en akut blodförlust med aktivering av kardiovaskulära försvars‐mekanismer. Ett liknande svar fås också vid stående.

Delarbete I studerar blodansamlingen (kapacitanssvar), venös compliance samt kapillär vätskefiltration i nedre extremiteten på unga friska kvinnor och män (n=28). En större blodansamling eller ökat vätskeutträde hos kvinnor skulle leda till mer uttalad central hypovolemi och kunna förklara kvinnors ökade känslighet för akut experimentell blodförlust samt svimning. Kapacitanssvaret och venös compliance var dock snarare lägre hos kvinnor vid låga transmurala tryck, medan vid högre tryckgradienter minskade skillnaden och försvann. Den kapillära vätskefiltrationen och kapillära filtrationskoefficienten (CFC) var däremot större hos kvinnor än hos män. Sammantaget var den totala hypovolemin jämförlig mellan könen. Om inte den kapillära filtrationen subtraherades från kapacitanssvaret, vilket inte är brukligt vid studium av venös compliance, så kan det leda till feltolkningar av data.

Delarbete II visar på hur normalt åldrande påverkar venös compliance och kapillär vätskefiltration hos friska kvinnor och studerades i vaden på äldre (66 år) och unga kvinnor (23 år) (n=35). Ingen förändring av compliance med ökande ålder noterades till skillnad från tidigare studier på män, som uppvisar klart sänkt compliance vid naturligt åldrande. Kapillära filtrationen samt CFC var lägre hos de äldre kvinnorna indikerande defekt kapillärfunktion, till skillnad från tidigare studier på äldre män. Vid stegrade transmurala tryckgradienter (ökade tryck över mikrokärlen) så ökade CFC hos de äldre kvinnorna, talande för en ökad kapillär känslighet för höga transmurala tryck. Fynden tyder på könsspecifika förändringar i både venös och mikro‐cirkulatorisk funktion vid ökande ålder.

55 Cardiovascular responses to hypovolemic circulatory stress in women

Delarbete III studerar kompensatoriska försvarsmekanismer för att skydda och upprätthålla den centrala blodvolymen vid experimentell hypovolemi hos friska unga kvinnor (23 år, n=22) och män (23 år, n=16). LBNP användes för att skapa hypovolemisk cirkulatorisk stress. Kärlsammandragningen (perifera resistensen, PR) var mindre hos kvinnor och man fann ingen association mellan hPR oc noradrenalin till skillnad från hos män. Kvinnor hade ett välutvecklat initialt kapacitanssvar (mobilisering av venblod från perifera vävnader mot centrala cirkulationen) vilket dock minskade successivt med tiden till skillnad från hos män. Vidare så var absorptionen av vätska från vävnad till blod (kapillär vätskeabsorption) mindre hos kvinnor. Fynden indikerar att kvinnor inte har lika effektiva kompensationsmekanismer som män som skydd för akut hypovolemisk stress.

Delarbete IV belyser kompensatoriska försvarsmekanismer för att skydda och upprätthålla den centrala blodvolymen vid experimentell hypovolemi hos äldre kvinnor (66 år, n = 12) och unga kvinnor (23 år, n=22). LBNP användes för att skapa hypovolemisk cirkulatorisk stress. Hjärtfrekvensen ökade mindre hos äldre kvinnor medan de svarade med samma respons i perifer resistens och konduktans (fortledning). Äldre kvinnor hade ett kraftigt reducerat kapacitanssvar (mobilisering av venblod från perifera vävnader mot centrala cirkulationen) jämfört med yngre kvinnor. Vidare var absorptionen av vätska från vävnad till blod (kapillära vätskeabsorption) lägre hos äldre kvinnor. Fynden indikerar att äldre kvinnor har mindre effektiva kompensatoriska mekanismer än yngre som skydd för akut hypovolemisk stress.

Den sista och kompletterande studien belyser det övergripande kardiovaskulära svaret vid experimentell hypovolemi skapad av LBNP hos kvinnor som uppvisade en vaso‐vagal reaktion (förstadium till svimning, 23 år, n=5) jämfört med icke vaso‐vagala kvinnor (23 år, n=22). Kapacitanssvaret samt kapillärfiltrationen av vätska i benen var likvärdig mellan grupperna. Däremot så tog det mycket längre tid för det hypovolema svaret att utvecklas hos gruppen med kvinnor som uppvisade en vaso‐vagal reaktion. Vidare så uppvisade den vaso‐vagala gruppen sänkt känslighet hos kardiopulmonella baroreceptorer samt ett sänkt kapacitanssvar (mobilisering av venblod från perifera vävnader mot centrala cirkulationen) jämfört med icke vaso‐vagala kvinnor. Sammantaget kan fynden till viss del förklara den ökade känsligheten för akut hypovolemi hos vaso‐vagala kvinnor.

56 Summary in Swedish

57 Cardiovascular responses to hypovolemic circulatory stress in women

ACKNOWLEDGEMENTS

I would like to express my sincere gratitude to everyone who has made this stimulating, but at times demanding journey possible. I would especially like to mention the following:

Professor Toste Länne, my tutor, for sharing your invaluable knowledge in cardiovascular physiology, but more importantly for your immense encouragement and enthusiasm in my work and for discussing and supporting new ideas. I would also like to express my deepest gratitude for your support in my life outside of science, not forgetting your family, Marie and Ruben, for inviting me into your home on numerous occasions.

Dr Henrik Olsen, Helsingborg, co‐author, for helping me at the start of the project with practical details and for fruitful discussions throughout my work.

Dr Staffan Schön, head of the Department of Internal Medicine, Jönköping, for your understanding attitude towards my research, and to my fellow colleagues at the Department of Internal Medicine for all your support.

Professor Bo‐Eric Malmvall, head of Futurum, the Academy of Health Care for your personal support in helping me to obtain time for my research also in practice.

The staff at the Department of Clinical Physiology/Linköping University, especially Elisabeth Forsström, Inger Ekman and the late Professor Bengt Wranne for taking me on as an amanuensis, raising my interest in physiology. Also thank you for good company and for free coffee throughout these years; a necessity during long working hours and when producing creative thoughts.

Olle Eriksson, statistician at Linköping University, for invaluable help scrutinizing my data and your guidance in finding the appropriate statistical methods. Also, thank you for quick support when deadline draws closer, despite your busy schedule.

58 Acknowledgements

Fellow PhD students for constructive criticism and for sharing happy times together.

Christina Andersson and Elin Wistrand at the Department of Medical and Health Sciences (IMH) in Linköping and Anneli Ohlson, Marie‐Louise Arvidsson, Ulrica Axelsson‐Francke and Agneta Alvarsson in Jönköping for support with matters of practical importance, not my strongest subject.

Nora Östrup, FoU‐secretary, for skilful linguistic revision of this thesis and published articles.

All volunteers, women and men, young and old, my sincere respect and thanks.

My close friends for supporting me in tough times, helping me back on my feet again. You should know how grateful I am.

Last but not least,

My loving parents, Ulf and Irene, for supporting me throughout my life. Thank you for your understanding and generosity. To David, for being the best brother one can have. It saddens me knowing you cannot live your life to your full potential and follow in my footsteps as you always used to do before your tragic accident. My late sister, Mi, for your warmth and high spirit, being generous with yourself. You are in my heart, always. My nephew and godson Felix, for being the star of my life.

Karolina, my partner, for your big heart, patience and love, being the best companion imaginable. Also to Astrid and Smilla, our two lively cats, for keeping me company during sometimes tedious writing days.

The studies, upon which this thesis is based, were supported by the invaluable grants from Futurum, the Academy of Health Care, Jönköping County Council and also from funds at Linköping University, Medical Research Council Grant 12661, and the Swedish Heart‐Lung Foundation.

59 Cardiovascular responses to hypovolemic circulatory stress in women

REFERENCES

Ablad B, Mellander S. Comparative Effects of Hydralazine, Sodium Nitrite and Acetylcholine on Resistance and Capacitance Blood Vessels and Capillary Filtration in Skeletal Muscle in the Cat. Acta Physiol Scand 58: 319‐329, 1963. Barwin BN, Roddie IC. Venous distensibility during pregnancy determined by graded venous congestion. Am J Obstet Gynecol 125(7): 921‐923, 1976. Becker LB, Weisfeldt ML, Weil MH, Budinger T, Carrico J, Kern K, Nichol G, Shechter I, Traystman R, Webb C, Wiedemann H, Wise R, Sopko G. The PULSE initiative: scientific priorities and strategic planning for resuscitation research and life saving therapies. Circulation 105(21): 2562‐ 2570, 2002. Bentzer P, Kongstad L, Grande PO. Capillary filtration coefficient is independent of number of perfused capillaries in cat skeletal muscle. Am J Physiol Heart Circ Physiol 280(6): H2697‐2706, 2001. Berne RM, Levy MN, Koeppen BM, Stanton BA. Physiology, 5th Edition: Mosby Year Book Medical Publishers, 2005. Bevan JA. Some bases of differences in vascular response to sympathetic activity. Circ Res 45(2): 161‐171, 1979. Bouissou H, Julian M, Pieraggi MT, Maurel E, Thiers JC, Louge L. Structure of healthy and varicose veins. In: Return Circulation and Norepinephrine: An Update, edited by Vanhoutte P. Paris: Libbey Eurotext, 1991. Bowyer L, Brown MA, Jones M. Vascular reactivity in men and women of reproductive age. Am J Obstet Gynecol 185(1): 88‐96, 2001. Brown CM, Hecht MJ, Weih A, Neundorfer B, Hilz MJ. Effects of age on the cardiac and vascular limbs of the arterial baroreflex. Eur J Clin Invest 33(1): 10‐16, 2003. Bruce A, Andersson M, Arvidsson B, Isaksson B. Body composition. Prediction of normal body potassium, body water and body fat in adults on the basis of body height, body weight and age. Scand J Clin Lab Invest 40(5): 461‐ 473, 1980. Chien S. Role of the sympathetic nervous system in hemorrhage. Physiol Rev 47(2): 214‐288, 1967. Claydon VE, Younis NR, Hainsworth R. Phase of the menstrual cycle does not affect orthostatic tolerance in healthy women. Clin Auton Res 16(2): 98‐ 104, 2006.

60 References

Clough G. A quantitative study of the exchange microvasculature of muscles from the human foot and hand. Int J Microcirc Clin Exp 6(3): 237‐243, 1987. Convertino VA. Gender differences in autonomic functions associated with blood pressure regulation. Am J Physiol 275(6 Pt 2): R1909‐1920, 1998. Convertino VA, Ludwig DA, Cooke WH. Stroke volume and sympathetic responses to lower‐body negative pressure reveal new insight into circulatory shock in humans. Auton Neurosci 111(2): 127‐134, 2004. Convertino VA, Ryan KL, Rickards CA, Salinas J, McManus JG, Cooke WH, Holcomb JB. Physiological and medical monitoring for en route care of combat casualties. J Trauma 64(4 Suppl): S342‐353, 2008. Cooke WH, Ryan KL, Convertino VA. Lower body negative pressure as a model to study progression to acute hemorrhagic shock in humans. J Appl Physiol 96(4): 1249‐1261, 2004. Cooper KE, Edholm OG, Mottram RF. The blood flow in skin and muscle of the human forearm. J Physiol 128(2): 258‐267, 1955. Croley AN, Zwetsloot KA, Westerkamp LM, Ryan NA, Pendergast AM, Hickner RC, Pofahl WE, Gavin TP. Lower capillarization, VEGF protein, and VEGF mRNA response to acute exercise in the vastus lateralis muscle of aged vs. young women. J Appl Physiol 99(5): 1872‐1879, 2005. Custaud MA, de Souza Neto EP, Abry P, Flandrin P, Millet C, Duvareille M, Fortrat JO, Gharib C. Orthostatic tolerance and spontaneous baroreflex sensitivity in men versus women after 7 days of head‐down bed rest. Auton Neurosci 100(1‐2): 66‐76, 2002. DʹAlmeida MS, Lautt WW. Expression of vascular escape: conductance or resistance? Am J Physiol 262(4 Pt 2): H1196‐1196, 1992. Debasso R, Astrand H, Bjarnegard N, Ryden Ahlgren A, Sandgren T, Lanne T. The popliteal artery, an unusual muscular artery with wall properties similar to the aorta: implications for susceptibility to aneurysm formation? J Vasc Surg 39(4): 836‐842, 2004. Dinenno FA, Dietz NM, Joyner MJ. Aging and forearm postjunctional alpha‐ adrenergic vasoconstriction in healthy men. Circulation 106(11): 1349‐1354, 2002. Edfeldt H. Sympathetic baroreflex control of vascular resistance in skeletal muscle and skin of man. Malmö, Sweden, 1993. Epstein SE, Stampfer M, Beiser GD. Role of the capacitance and resistance vessels in vasovagal syncope. Circulation 37(4): 524‐533, 1968. Fawer R, Dettling A, Weihs D, Welti H, Schelling JL. Effect of the menstrual cycle, oral contraception and pregnancy on forearm blood flow, venous distensibility and clotting factors. Eur J Clin Pharmacol 13(4): 251‐257, 1978.

61 Cardiovascular responses to hypovolemic circulatory stress in women

Fitzsimons JT. Angiotensin, thirst, and sodium appetite. Physiol Rev 78(3): 583‐ 686, 1998. Folkow B, Mellander S. Measurements of capillary filtration coefficient and its use in studies of the control of capillary exchange. In: Capillary Permeability: The Transfer of Molecules and Ions between Capillary Blood and Tissue, edited by Crone C, Lassen N. Copenhagen: Munksgaard, 1970, p. 614‐623. Franke WD, Johnson CP, Steinkamp JA, Wang R, Halliwill JR. Cardiovascular and autonomic responses to lower body negative pressure: do not explain gender differences in orthostatic tolerance. Clin Auton Res 13(1): 36‐44, 2003. Freedman RR, Sabharwal SC, Desai N. Sex differences in peripheral vascular adrenergic receptors. Circ Res 61(4): 581‐585, 1987. Frey MA, Hoffler GW. Association of sex and age with responses to lower‐ body negative pressure. J Appl Physiol 65(4): 1752‐1756, 1988. Frey MA, Mathes KL, Hoffler GW. Cardiovascular responses of women to lower body negative pressure. Aviat Space Environ Med 57(6): 531‐538, 1986. Frokjaer‐Jensen J. Permeability of single muscle capillaries to potassium ions. Microvasc Res 24(2): 168‐183, 1982. Fu Q, Arbab‐Zadeh A, Perhonen MA, Zhang R, Zuckerman JH, Levine BD. Hemodynamics of orthostatic intolerance: implications for gender differences. Am J Physiol Heart Circ Physiol 286(1): H449‐457, 2004. Fu Q, Witkowski S, Okazaki K, Levine BD. Effects of gender and hypovolemia on sympathetic neural responses to orthostatic stress. Am J Physiol Regul Integr Comp Physiol 289(1): R109‐116, 2005. Fuca G, Dinelli M, Gianfranchi L, Bressan S, Lamborghini C, Alboni P. Do subjects with vasovagal syncope have subtle haemodynamic alterations during orthostatic stress? Europace 10(6): 751‐759, 2008. Fuca G, Dinelli M, Suzzani P, Scarfo S, Tassinari F, Alboni P. The venous system is the main determinant of hypotension in patients with vasovagal syncope. Europace 8(10): 839‐845, 2006. Gannon CJ, Napolitano LM, Pasquale M, Tracy JK, McCarter RJ. A statewide population‐based study of gender differences in trauma: validation of a prior single‐institution study. J Am Coll Surg 195(1): 11‐18, 2002. George RL, McGwin G, Jr., Windham ST, Melton SM, Metzger J, Chaudry IH, Rue LW, 3rd. Age‐related gender differential in outcome after blunt or penetrating trauma. Shock 19(1): 28‐32, 2003.

62 References

Goldstein DS, McCarty R, Polinsky RJ, Kopin IJ. Relationship between plasma norepinephrine and sympathetic neural activity. Hypertension 5(4): 552‐ 559, 1983. Gorodeski GI, Yang T, Levy MN, Goldfarb J, Utian WH. Effects of estrogen in vivo on coronary vascular resistance in perfused rabbit hearts. Am J Physiol 269(6 Pt 2): R1333‐1338, 1995. Gotshall RW. Gender differences in tolerance to lower body negative pressure. Aviat Space Environ Med 71(11): 1104‐1110, 2000. Groban L, Cowley AW, Jr., Ebert TJ. Atrial natriuretic peptide augments forearm capillary filtration in humans. Am J Physiol 259(1 Pt 2): H258‐263, 1990. Hafferl A, Thiel W. Lehrbuch der topographischen Anatomie. Berlin: Springer, 1969, p. 971. Halliwill JR, Lawler LA, Eickhoff TJ, Joyner MJ, Mulvagh SL. Reflex responses to regional venous pooling during lower body negative pressure in humans. J Appl Physiol 84(2): 454‐458, 1998. Halliwill JR, Minson CT, Joyner MJ. Measurement of limb venous compliance in humans: technical considerations and physiological findings. J Appl Physiol 87(4): 1555‐1563, 1999. Hargreaves AD, Muir AL. Lack of variation in venous tone potentiates vasovagal syncope. Br Heart J 67(6): 486‐490, 1992. Harris BA. The influence of endurance and resistance exercise on muscle capillarization in the elderly: a review. Acta Physiol Scand 185(2): 89‐97, 2005. Henriksen O, Sejrsen P. Local reflex in microcirculation in human skeletal muscle. Acta Physiol Scand 99(1): 19‐26, 1977. Hernandez JP, Franke WD. Effects of a 6‐mo endurance‐training program on venous compliance and maximal lower body negative pressure in older men and women. J Appl Physiol 99(3): 1070‐1077, 2005. Hjemdahl P, Fagius J, Freyschuss U, Wallin BG, Daleskog M, Bohlin G, Perski A. Muscle sympathetic activity and norepinephrine release during mental challenge in humans. Am J Physiol 257(5 Pt 1): E654‐664, 1989. Holland CK, Brown JM, Scoutt LM, Taylor KJ. Lower extremity volumetric arterial blood flow in normal subjects. Ultrasound Med Biol 24(8): 1079‐ 1086, 1998. Huang A, Sun D, Koller A, Kaley G. Gender difference in myogenic tone of rat arterioles is due to estrogen‐induced, enhanced release of NO. Am J Physiol 272(4 Pt 2): H1804‐1809, 1997.

63 Cardiovascular responses to hypovolemic circulatory stress in women

Huisman HW, Pretorius PJ, Van Rooyen JM, Malan NT, Eloff FC, Laubscher PJ, Steyn HS. Haemodynamic changes in the cardiovascular system during the early phases of orthostasis. Acta Physiol Scand 166(2): 145‐149, 1999. Huxley VH, Wang J, Whitt SP. Sexual dimorphism in the permeability response of coronary microvessels to adenosine. Am J Physiol Heart Circ Physiol 288(4): H2006‐2013, 2005. Jones PP, Christou DD, Jordan J, Seals DR. Baroreflex buffering is reduced with age in healthy men. Circulation 107(13): 1770‐1774, 2003. Kappert K, Caglayan E, Huntgeburth M, Baumer AT, Sparwel J, Uebel M, Rosenkranz S. 17Beta‐estradiol attenuates PDGF signaling in vascular smooth muscle cells at the postreceptor level. Am J Physiol Heart Circ Physiol 290(2): H538‐546, 2006. Knaapen MW, Somers P, Bortier H, De Meyer GR, Kockx MM. Smooth muscle cell hypertrophy in varicose veins is associated with expression of estrogen receptor‐beta. J Vasc Res 42(1): 8‐12, 2005. Kneale BJ, Chowienczyk PJ, Brett SE, Coltart DJ, Ritter JM. Gender differences in sensitivity to adrenergic agonists of forearm resistance vasculature. J Am Coll Cardiol 36(4): 1233‐1238, 2000. Kuczkowski KM. Labor analgesia for the parturient with cardiac disease: what does an obstetrician need to know? Acta Obstet Gynecol Scand 83(3): 223‐ 233, 2004. Laitinen T, Hartikainen J, Vanninen E, Niskanen L, Geelen G, Lansimies E. Age and gender dependency of baroreflex sensitivity in healthy subjects. J Appl Physiol 84(2): 576‐583, 1998. Laitinen T, Niskanen L, Geelen G, Lansimies E, Hartikainen J. Age dependency of cardiovascular autonomic responses to head‐up tilt in healthy subjects. J Appl Physiol 96(6): 2333‐2340, 2004. Lanne T, Lundvall J. Mechanisms in man for rapid refill of the circulatory system in hypovolaemia. Acta Physiol Scand 146(3): 299‐306, 1992. Lanne T, Olsen H. Decreased capacitance response with age in lower limbs of humans‐‐a potential error in the study of cardiovascular reflexes in ageing. Acta Physiol Scand 161(4): 503‐507, 1997. Lesh TA, Rothe CF. Sympathetic and hemodynamic effects on capacitance vessels in dog skeletal muscle. Am J Physiol 217(3): 819‐827, 1969. Lewis P, Psaila JV, Davies WT, McCarty K, Woodcock JP. Measurement of volume flow in the human common femoral artery using a duplex ultrasound system. Ultrasound Med Biol 12(10): 777‐784, 1986.

64 References

Linder L, Lautenschlager BM, Haefeli WE. Subconstrictor doses of neuropeptide Y potentiate alpha 1‐adrenergic venoconstriction in vivo. Hypertension 28(3): 483‐487, 1996. Ludwig DA, Vernikos J, Wade CE, Convertino VA. Blood pressure changes during orthostatic stress: evidence of gender differences in neuroeffector distribution. Aviat Space Environ Med 72(10): 892‐898, 2001. Lundvall J, Bjerkhoel P, Edfeldt H, Ivarsson C, Lanne T. Dynamics of transcapillary fluid transfer and plasma volume during lower body negative pressure. Acta Physiol Scand 147(2): 163‐172, 1993. Lundvall J, Hillman J. Fluid transfer from skeletal muscle to blood during hemorrhage. Importance of beta adrenergic vascular mechanisms. Acta Physiol Scand 102(4): 450‐458, 1978. Lundvall J, Lanne T. Large capacity in man for effective plasma volume control in hypovolaemia via fluid transfer from tissue to blood. Acta Physiol Scand 137(4): 513‐520, 1989a. Lundvall J, Lanne T. Much larger transcapillary hydrodynamic conductivity in skeletal muscle and skin of man than previously believed. Acta Physiol Scand 136(1): 7‐16, 1989b. Magnotti LJ, Fischer PE, Zarzaur BL, Fabian TC, Croce MA. Impact of gender on outcomes after blunt injury: a definitive analysis of more than 36,000 trauma patients. J Am Coll Surg 206(5): 984‐991; discussion 991‐982, 2008. Mancia G, Mark A. Cardiopulmonary baroreflexes in humans. In: Handbook of Physiology The Cardiovasular System, edited by Shepherd J. Washington DC: Bethesda, Md: Am Physiol Soc, 1983, p. 795‐813. Mark A, Mancia G. Arterial baroreflexes in humans. In: Handbook of Physiology The Cardiovasular System, edited by Shepherd J. Washington DC: Bethesda, Md: Am Physiol Soc, 1983, p. 755‐793. Maspers M, Bjornberg J, Grande PO, Mellander S. Sympathetic alpha‐ adrenergic control of large‐bore arterial vessels, arterioles and veins, and of capillary pressure and fluid exchange in whole‐organ cat skeletal muscle. Acta Physiol Scand 138(4): 509‐521, 1990. McGwin G, Jr., George RL, Cross JM, Reiff DA, Chaudry IH, Rue LW, 3rd. Gender differences in mortality following burn injury. Shock 18(4): 311‐ 315, 2002. Meendering JR, Torgrimson BN, Houghton BL, Halliwill JR, Minson CT. Effects of menstrual cycle and oral contraceptive use on calf venous compliance. Am J Physiol Heart Circ Physiol 288(1): H103‐110, 2005a. Meendering JR, Torgrimson BN, Houghton BL, Halliwill JR, Minson CT. Menstrual cycle and sex affect hemodynamic responses to combined

65 Cardiovascular responses to hypovolemic circulatory stress in women

orthostatic and heat stress. Am J Physiol Heart Circ Physiol 289(2): H631‐ 642, 2005b. Mellander S. Comparative studies on the adrenergic neuro‐hormonal control of resistance and capacitance blood vessels in the cat. Acta Physiol Scand Suppl 50(176): 1‐86, 1960. Mellander S, Oberg B, Odelram H. Vascular Adjustments to Increased Transmural Pressure in Cat and Man with Special Reference to Shifts in Capillary Fluid Transfer. Acta Physiol Scand 61: 34‐48, 1964. Mendelsohn ME. Genomic and nongenomic effects of estrogen in the vasculature. Am J Cardiol 90(1A): 3F‐6F, 2002. Miller JA, Floras JS, Skorecki KL, Blendis LM, Logan AG. Renal and humoral responses to sustained cardiopulmonary baroreceptor deactivation in humans. Am J Physiol 260(3 Pt 2): R642‐648, 1991. Mizumaki K, Fujiki A, Tani M, Shimono M, Hayashi H, Inoue H. Left ventricular dimensions and autonomic balance during head‐up tilt differ between patients with isoproterenol‐dependent and isoproterenol‐ independent neurally mediated syncope. J Am Coll Cardiol 26(1): 164‐173, 1995. Monahan KD, Dinenno FA, Seals DR, Halliwill JR. Smaller age‐associated reductions in leg venous compliance in endurance exercise‐trained men. Am J Physiol Heart Circ Physiol 281(3): H1267‐1273, 2001. Monahan KD, Ray CA. Gender affects calf venous compliance at rest and during baroreceptor unloading in humans. Am J Physiol Heart Circ Physiol 286(3): H895‐901, 2004. Morikawa T, Sagawa S, Torii R, Endo Y, Yamazaki F, Shiraki K. Hypovolemic intolerance to lower body negative pressure in female runners. Med Sci Sports Exerc 33(12): 2058‐2064, 2001. Mosqueda ‐Garcia R, Furlan R, Fernandez‐Violante R, Desai T, Snell M, Jarai Z, Ananthram V, Robertson RM, Robertson D. Sympathetic and baroreceptor reflex function in neurally mediated syncope evoked by tilt. J Clin Invest 99(11): 2736‐2744, 1997. Mosqueda‐Garcia R, Furlan R, Tank J, Fernandez‐Violante R. The elusive pathophysiology of neurally mediated syncope. Circulation 102(23): 2898‐ 2906, 2000. Muller‐Delp JM. Aging‐induced adaptations of microvascular reactivity. Microcirculation 13(4): 301‐314, 2006. Neal CR, Michel CC. Openings in frog microvascular endothelium induced by high intravascular pressures. J Physiol 492 ( Pt 1): 39‐52, 1996.

66 References

Ng AV, Callister R, Johnson DG, Seals DR. Sympathetic neural reactivity to stress does not increase with age in healthy humans. Am J Physiol 267(1 Pt 2): H344‐353, 1994. OʹKeefe GE, Hunt JL, Purdue GF. An evaluation of risk factors for mortality after burn trauma and the identification of gender‐dependent differences in outcomes. J Am Coll Surg 192(2): 153‐160, 2001. OʹMahony D, Bennett C, Green A, Sinclair AJ. Reduced baroreflex sensitivity in elderly humans is not due to efferent autonomic dysfunction. Clin Sci (Lond) 98(1): 103‐110, 2000. Oberg B. The relationship between active constrictiond an passive recoil of the veins at various distending pressures. Acta Physiol Scand 71(2): 233‐247, 1967. Olsen H, Lanne T. Reduced venous compliance in lower limbs of aging humans and its importance for capacitance function. Am J Physiol 275(3 Pt 2): H878‐886, 1998. Olsen H, Lanne T. Reduced capillary hydraulic conductivity in skeletal muscle and skin in Type I diabetes: a possible cause for reduced transcapillary fluid absorption during hypovolaemia. Diabetologia 43(9): 1178‐1184, 2000. Olsen H, Vernersson E, Lanne T. Cardiovascular response to acute hypovolemia in relation to age. Implications for orthostasis and hemorrhage. Am J Physiol Heart Circ Physiol 278(1): H222‐232, 2000. Pechere‐Bertschi A, Maillard M, Stalder H, Brunner HR, Burnier M. Blood pressure and renal haemodynamic response to salt during the normal menstrual cycle. Clin Sci (Lond) 98(6): 697‐702, 2000. Rajagopalan B, Bertram CD, Stallard T, Lee GD. Blood flow in pulmonary veins: III. Simultaneous measurements of their dimensions, intravascular pressure and flow. Cardiovasc Res 13(12): 684‐692, 1979. Rippe B, Haraldsson B, Folkow B. Evaluation of the ʹstretched pore phenomenonʹ in isolated rat hindquarters. Acta Physiol Scand 125(3): 453‐ 459, 1985. Rothe CF. Reflex control of the veins in cardiovascular function. Physiologist 22(2): 28‐35, 1979. Rothe CF. Venous system: Physiology of the capactiance vessels. In: Handbook of Physiology The Cardiovasular System, Peripheral circulation, edited by Shepherd J. Washington, D.C: Bethesda, Md: Am. Physiol. Soc, 1983, p. 397‐452. Rowell L. Human cardiovascular control. New York: Oxford University Press, 1993.

67 Cardiovascular responses to hypovolemic circulatory stress in women

Saitta A, Altavilla D, Cucinotta D, Morabito N, Frisina N, Corrado F, DʹAnna R, Lasco A, Squadrito G, Gaudio A, Cancellieri F, Arcoraci V, Squadrito F. Randomized, double‐blind, placebo‐controlled study on effects of raloxifene and hormone replacement therapy on plasma no concentrations, endothelin‐1 levels, and endothelium ‐dependent vasodilation in postmenopausal women. Arterioscler Thromb Vasc Biol 21(9): 1512‐1519, 2001. Sauaia A, Moore FA, Moore EE, Moser KS, Brennan R, Read RA, Pons PT. Epidemiology of trauma deaths: a reassessment. J Trauma 38(2): 185‐193, 1995. Savage DD, Corwin L, McGee DL, Kannel WB, Wolf PA. Epidemiologic features of isolated syncope: the Framingham Study. Stroke 16(4): 626‐629, 1985. Schnizer W, Klatt J, Baeker H, Rieckert H. [Comparison of scintigraphic and plethysmographic measurements for determination of capillary filtration coefficient in human limbs (authorʹs transl)]. Basic Res Cardiol 73(1): 77‐84, 1978. Schutzer WE, Mader SL. Age‐related changes in vascular adrenergic signaling: clinical and mechanistic implications. Ageing Res Rev 2(2): 169‐190, 2003. Seller H. Central baroreceptor reflex pathways. In: Baroreceptor Reflexes: Integrative Functions and Clinical Aspects, edited by Persson PB, H.R. K. Heidelberg: Springer‐Verlag, 1991, p. 45–74. Sexton WL, Poole DC, Mathieu‐Costello O. Microcirculatory structure‐ function relationships in skeletal muscle of diabetic rats. Am J Physiol 266(4 Pt 2): H1502‐1511, 1994. Shen WK, Low PA, Rea RF, Lohse CM, Hodge DO, Hammill SC. Distinct hemodynamic profiles in patients with vasovagal syncope: a heterogeneous population. J Am Coll Cardiol 35(6): 1470‐1477, 2000. Shevde K, Pagala M, Tyagaraj C, Udeh C, Punjala M, Arora S, Elfaham A. Preoperative blood volume deficit influences blood transfusion requirements in females and males undergoing coronary bypass graft surgery. J Clin Anesth 14(7): 512‐517, 2002. Shoemaker JK, Hogeman CS, Khan M, Kimmerly DS, Sinoway LI. Gender affects sympathetic and hemodynamic response to postural stress. Am J Physiol Heart Circ Physiol 281(5): H2028‐2035, 2001. Skelton H. The storage of water by various tissues of the body. Arch intern med 40: 140‐152, 1927.

68 References

Sonesson B, Hansen F, Stale H, Lanne T. Compliance and diameter in the human abdominal aorta‐‐the influence of age and sex. Eur J Vasc Surg 7(6): 690‐697, 1993. Stachenfeld NS, Keefe DL, Palter SF. Estrogen and progesterone effects on transcapillary fluid dynamics. Am J Physiol Regul Integr Comp Physiol 281(4): R1319‐1329, 2001. Stachenfeld NS, Taylor HS. Progesterone increases plasma volume independent of estradiol. J Appl Physiol 98(6): 1991‐1997, 2005. Stewart JM. Microvascular filtration is increased in postural tachycardia syndrome. Circulation 107(22): 2816‐2822, 2003. Stewart JM, Lavin J, Weldon A. Orthostasis fails to produce active limb venoconstriction in adolescents. J Appl Physiol 91(4): 1723‐1729, 2001. Stewart JM, McLeod KJ, Sanyal S, Herzberg G, Montgomery LD. Relation of postural vasovagal syncope to splanchnic hypervolemia in adolescents. Circulation 110(17): 2575‐2581, 2004. Sudhir K, Elser MD, Jennings GL, Komesaroff PA. Estrogen supplementation decreases norepinephrine‐induced vasoconstriction and ytotal bod norepinephrine spillover in perimenopausal women. Hypertension 30(6): 1538‐1543, 1997. Taneja I, Moran C, Medow MS, Glover JL, Montgomery LD, Stewart JM. Differential effects of lower body negative pressure and upright tilt on splanchnic blood volume. Am J Physiol Heart Circ Physiol 292(3): H1420‐ 1426, 2007. Taylor MD, Tracy JK, Meyer W, Pasquale M, Napolitano LM. Trauma in the elderly: intensive care unit resource use and outcome. J Trauma 53(3): 407‐ 414, 2002. Thomson HL, Wright K, Frenneaux M. Baroreflex sensitivity in patients with vasovagal syncope. Circulation 95(2): 395‐400, 1997. Tollan A, Kvenild K, Strand H, Oian P, Maltau JM. Increased capillary permeability for plasma proteins in oral contraceptive users. Contraception 45(5): 473‐481, 1992. Tomiyama H, Yamashina A, Arai T, Hirose K, Koji Y, Chikamori T, Hori S, Yamamoto Y, Doba N, Hinohara S. Influences of age and gender on results of noninvasive brachial‐ankle pulse wave velocity measurement‐‐ a survey of 12517 subjects. Atherosclerosis 166(2): 303‐309, 2003. Toraman F, Evrenkaya S, Yuce M, Turek O, Aksoy N, Karabulut H, Demirhisar O, Alhan C. Highly positive intraoperative fluid balance during cardiac surgery is associated with adverse outcome. Perfusion 19(2): 85‐91, 2004.

69 Cardiovascular responses to hypovolemic circulatory stress in women

Tsutsui Y, Sagawa S, Yamauchi K, Endo Y, Yamazaki F, Shiraki K. Cardiovascular responses to lower body negative pressure in the elderly: role of reduced leg compliance. Gerontology 48(3): 133‐139, 2002. van Brummelen P, Buhler FR, Kiowski W, Amann FW. Age‐related decrease in cardiac and peripheral vascular responsiveness to isoprenaline: studies in normal subjects. Clin Sci (Lond) 60(5): 571‐577, 1981. Wasmund SL, Smith ML, Takata TS, Joglar JA, Li JM, Kowal RC, Page RL, Hamdan MH. Sympathoexcitation is attenuated during low level lower body negative pressure in subjects who develop pre‐syncope. Clin Auton Res 13(3): 208‐213, 2003. Well D, Yang H, Houseni M, Iruvuri S, Alzeair S, Sansovini M, Wintering N, Alavi A, Torigian DA. Age‐related structural and metabolic changes in the pelvic reproductive end organs. Semin Nucl Med 37(3): 173‐184, 2007. Wesly RL, Vaishnav RN, Fuchs JC, Patel DJ, Greenfield JC, Jr. Static linear and nonlinear elastic properties of normal and arterialized venous tissue in dog and man. Circ Res 37(4): 509‐520, 1975. White DD, Gotshall RW, Tucker A. Women have lower tolerance to lower body negative pressure than men. J Appl Physiol 80(4): 1138‐1143, 1996. White DD, Montgomery LD. Pelvic blood pooling of men and women during lower body negative pressure. Aviat Space Environ Med 67(6): 555‐559, 1996. Wijeyaratne CN, Moult PJ. The effect of alpha human atrial natriuretic peptide on plasma volume and vascular permeability in normotensive subjects. J Clin Endocrinol Metab 76(2): 343‐346, 1993. Wijeysundera DN, Butler GC, Ando S, Pollard M, Picton P, Floras JS. Attenuated cardiac baroreflex in men with presyncope evoked by lower body negative pressure. Clin Sci (Lond) 100(3): 303‐309, 2001. Wohltmann CD, Franklin GA, Boaz PW, Luchette FA, Kearney PA, Richardson JD, Spain DA. A multicenter evaluation of whether gender dimorphism affects survival after trauma. Am J Surg 181(4): 297‐300, 2001. Wolf MB, Porter LP, Watson PD. Effects of elevated venous pressure on capillary permeability in cat hindlimbs. Am J Physiol 257(6 Pt 2): H2025‐ 2032, 1989. Yamanouchi Y, Jaalouk S, Shehadeh AA, Jaeger F, Goren H, Fouad‐Tarazi FM. Changes in left ventricular volume during head‐up tilt in patients with vasovagal syncope: an echocardiographic study. Am Heart J 131(1): 73‐80, 1996.

70