UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Endothelial selectins and pulmonary gas exchange in female aerobic athletes Hunte, Garth Stephen 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-0432.pdf [ 6.93MB ]
Metadata
JSON: 831-1.0077067.json
JSON-LD: 831-1.0077067-ld.json
RDF/XML (Pretty): 831-1.0077067-rdf.xml
RDF/JSON: 831-1.0077067-rdf.json
Turtle: 831-1.0077067-turtle.txt
N-Triples: 831-1.0077067-rdf-ntriples.txt
Original Record: 831-1.0077067-source.json
Full Text
831-1.0077067-fulltext.txt
Citation
831-1.0077067.ris

Full Text

ENDOTHELIAL SELECTINS AND PULMONARY GAS EXCHANGE IN FEMALE AEROBIC ATHLETES by GARTH STEPHEN HUNTE B.Sc. (Hon.), The University of Calgary, 1986 M.D., The University of Alberta, 1990 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (School of Human Kinetics) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 2000 © Garth Stephen Hunte, 2000 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood that copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n permission. Department of The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada Date Abstract Demonstration of a greater elevation in the (ideal) alveolar/arterial oxygen difference in habitually active female subjects with exercise-induced arterial hypoxemia, at equivalent submaximal levels of oxygen uptake compared to inactive controls, suggests functional or structural compromise of the blood-gas interface may occur with chronic-recurrent intensive exercise. Mechanical and/or chemically mediated pulmonary endothelial dysfunction during heavy exercise may alter vascular tone and permeability, leading to interstitial edema and accentuation of ventilation-perfusion mismatch and/or diffusion limitation. Elevated plasma levels of soluble endothelial cell adhesion molecules E- and P-selectin have been demonstrated in acute lung injury and have been used as indirect markers of endothelial activation or injury. Therefore, plasma levels of these selectins were measured by enzyme immunoassay in fourteen habitually active, eumenorrheic female subjects (mean±SD: age = 2 8 . 9 ± 5 . 5 1 ; l / 0 2 p e a k = 49.4±8.2 ml.kg"1.min"1, range 32.3 to 63.7 ml.kg"1.min"1; TLC = 5.41±0.68 L, 101±9.3% predicted) before and after an incremental maximal exercise test during the follicular phase of their menstrual cycle (cycle day = 6.2±1.2, serum progesterone = 80+100 pmolL"1). Arterial partial pressure of oxygen (Pa02) was measured and corrected for esophageal temperature, arterial oxyhemoglobin saturation (%Sa02) was calculated from blood gas variables and measured with pulse oximetry, and the (ideal) alveolar/arterial oxygen gradient was calculated from the ideal gas equation. Pulmonary gas exchange efficiency was maintained at peak exercise in ten subjects, while decrements in arterial partial pressure of oxygen during exercise of greater than 1.3 kilopascals (10 mmHg) were seen in three of the remaining four subjects. One subject displayed a minimal %Sa0 2 of 94% and was included in the mild hypoxemia group. Maximum likelihood ANOVA procedures, used on account of missing data, showed significant differences between groups averaged over time for Pa0 2 (p<0.01) and %Sa0 2 (p=0.04), while the group by time interaction for the (ideal) A-aD0 2 approached significance (p=0.07). Averaged over time, changes in alveolar P0 2 , arterial PC0 2 , pH and temperature were not significantly different between groups. Plasma concentrations of soluble E-selectin were not significantly different before or after exercise (p=0.16), but plasma concentrations of P-selectin rose significantly (mean increase ± SD; 21.5±24.8 ngmL'1, p=0.007). No significant group by time interaction was noted in pre-post exercise concentrations of either E-selectin (p=0.74) or P-selectin i i (p=0.42) between subjects who demonstrated normal gas exchange and subjects who displayed mild to moderate exercise-induced gas exchange impairment. The correlation between absolute (ngmL"1) and relative (%) change in soluble E- and P-selectin, and V /02peak, maximal A-aD0 2 and PaC0 2 was not significant, nor was the correlation between minimal exercise Pa0 2 and either absolute (r=0.16, p=0.61) or relative (r=0.18, p=0.57) change in soluble E-selectin. However, absolute change in plasma concentration of soluble P-selectin was significantly correlated with minimal Pa0 2 (r=-0.60, p=0.04), while the correlation between the relative change in P-selectin and minimal Pa0 2 approached significance (r=-0.46, p=0.14). The increase in plasma P-selectin induced by heavy exercise may represent platelet and/or endothelial activation. Correlation with impairment of arterial oxygenation is compatible with the hypothesis that pulmonary endothelial dysfunction may occur during intense exercise in some habitually active female subjects. iii Table of Contents Abstract ii List of Tables v List of Figures vii List of Abbreviations ix 1. Introduction 1 2. Methods 4 2.1 Participants 4 2.2 Pulmonary function 4 2.3 Blood sampling 4 2.4 Exercise protocol 5 2.5 Blood gas analysis 5 2.6 Enzyme immunoassays 6 2.6.1 Sex hormones 6 2.6.2 Soluble selectins 7 2.7 Statistical analysis 7 3. Results 7 3.1 Baseline characteristics 7 3.2 Exercise performance 10 3.3 Pulmonary gas exchange 10 3.4 Soluble selectins 22 3.5 Soluble selectins and pulmonary gas exchange 22 4. Discussion 22 4.1 Pulmonary gas exchange 22 4.2 Cell adhesion molecules 26 4.2.1 Soluble E-selectin 27 4.2.2 Soluble P-selectin 28 4.2.2.1 Endothelium 29 4.2.2.2 Platelets 30 4.3 Cell adhesion molecules and pulmonary gas exchange 30 4.4 Vascular remodeling 31 4.5 Regional perfusion heterogeneity 32 4.6 Conclusion 35 5. References 36 Appendix I. Literature Review: Exercise-induced arterial hypoxemia 73 Appendix II. Literature Review: Exercise and the pulmonary circulation 97 Appendix III. Raw Data 126 iv List of Tables Table 1. Aerobic exercise habits of individual subjects 8 Table 2. Physical characteristics at baseline 8 Table 3. Lung volumes and resting spirometry 9 Table 4. Performance and ventilatory parameters at peak exercise 9 Table 5. Differences in blood gas variables between groups 15 Table 6. Age, height and mass of individual subjects 126 Table 7. Individual hematologic and hormonal parameters 127 Table 8. Lung volumes and expiratory airflow of individual subjects 128 Table 9. Ventilatory parameters at peak exercise of individual subjects 129 Table 10. Performance at peak exercise of individual subjects 130 Table 11. Ambient testing conditions 131 Table 12. Individual PePz data calculated from the 'ideal' alveolar gas equation 132 Table 13. Individual temperature corrected Pa02 data measured during 133 progressive exercise Table 14. Individual calculated (ideal) alveolar/arterial PO2 gradient data 134 Table 15. Individual PaC02 data measured during progressive exercise 135 Table 16. Individual calculated %Sa02 from Pa02 data measured during 136 progressive exercise Table 17. Individual SpO^data measured during progressive exercise 137 Table 18. Individual HCO3" data measured during progressive exercise 138 Table 19. Individual BE data measured during progressive exercise 139 Table 20. Individual pH data measured during progressive exercise 140 Table 21. Individual core temperature data (degrees Celsius) measured during 141 progressive exercise v Table 22 . Individual pre- and post-exercise plasma sample concentrations 142 of sE-selectin Table 23 . Individual pre- and post-exercise plasma sample concentrations 143 of sP-selectin Table 24. Individual serum sample concentrations of 17B-estradiol 144 Table 25 . Individual serum sample concentrations of progesterone 145 v i List of Figures Figure 1. Individual PaC»2 data versus absolute VO2 11 Figure 2. Individual %SaC>2 data versus absolute VO2 11 Figure 3. Individual PAC>2 data versus absolute VO2 12 Figure 4. Individual (ideal) A-aD02 data versus absolute VO2 12 Figure 5. Individual P a C 0 2 data versus absolute VCO2 13 Figure 6. Individual pH data versus absolute VCO2 13 Figure 7. Individual pulse oximetry data versus absolute VO2 14 Figure 8. Individual temperature data versus absolute VO2 14 Figure 9. Differences in mean PaOz between groups 16 Figure 10 Differences in mean (ideal) A-aDC»2 between groups 16 Figure 11 Differences in mean PA02 between groups 17 Figure 12 Differences in mean PaCC>2 between groups 17 Figure 13 Differences in mean % S a 0 2 between groups 18 Figure 14 Differences in mean pulse oximetry between groups 18 Figure 15 Differences in mean pH between groups 19 Figure 16 Differences in mean temperature between groups 19 Figure 17 Relationship between PaC>2 and (ideal) A-aDC"2 20 Figure 18 Relationship between PaC>2 and PAC"2 21 Figure 19 Relationship between Pa02 and PaCC>2 21 Figure 20 Individual pre- and post-exercise plasma concentrations of sE-selectin 23 Figure 21 Individual pre- and post-exercise plasma concentrations of sP-selectin 23 Figure 22 Relationship between absolute change in soluble sE-selectin 24 concentration and minimal PaC>2 v i i Figure 23. Relationship between relative change in soluble sE-selectin 24 concentration and minimal PaC»2 Figure 24. Relationship between absolute change in soluble sP-selectin 25 concentration and minimal PaC»2 Figure 25. Relationship between relative change in soluble sP-selectin 25 concentration and minimal PaC>2 Figure 26. EIA standard curves for: A. soluble E-selectin, B. soluble P-selectin 146 Figure 27. EIA standard curves for: A. 17p-estradiol, B. progesterone 147 v i i i List of Abbreviations BE base excess HCCV bicarbonate CAM cell adhesion molecule [ C a 2 ! cytosol C a 2 + concentration EC endothelial cell E 2 17B-estradiol EIAH exercise-induced arterial hypoxemia FEFmax maximal forced expiratory flow FEVT forced expiratory volume in 1 second FVC forced vital capacity HAPE high altitude pulmonary edema HAPE-S HAPE susceptibility Hct hematocrit Hgb hemoglobin (ideal) A-aD0 2 (ideal) alveolar/arterial P 0 2 gradient IL interleukin NO nitric oxide P 0 2 partial pressure of oxygen P A 0 2 alveolar partial pressure of oxygen Pa0 2 arterial partial pressure of oxygen P a C 0 2 arterial partial pressure of carbon dioxide PH (log10)hydrogen ion concentration RER respiratory exchange ratio %Sa0 2 percent arterial oxyhemoglobin saturation Sp0 2 pulse oximetry sE-selectin soluble E-selectin (CD62E) sP-selectin soluble P-selectin (CD62P) SMC smooth muscle cell TLC total lung capacity VSMC vascular smooth muscle cell VC vital capacity V E minute ventilation VA/Q ventilation-perfusion ratio V 0 2 rate of oxygen consumption V C 0 2 rate of carbon dioxide production v'02peak peak oxygen consumption vWF:Ag von Willebrand factor antigen ix 1 1 Introduction Decrements in arterial P 0 2 and oxyhemoglobin saturation (%Sa02) have been demonstrated during intense exercise in a subset of high aerobic power male athletes (V02peak>65 mlkg"1min"1) (Dempsey ef a/., 1984; Powers et a/., 1988; Hopkins and McKenzie, 1989; Rice et al., 1999a; Edwards et al., in press), male masters athletes (Prefaut al., 1994), and female athletes (Harms et al., 1998a; Hopkins er al., 2000). Gas exchange impairment has also been noted at submaximal exercise intensity in some habitually active male and female subjects (Dempsey et al., 1984; Harms ef al., 1998a). Compared to control subjects with similar and substantially less aerobic power, female subjects with exercise-induced arterial hypoxemia (EIAH) displayed a greater increase in the alveolar/arterial 0 2 gradient and a greater reduction in PaC>2 and %SaC>2 at the same relative submaximal exercise intensity (Harms er al., 1998a). This discrepancy at submaximal exercise intensity suggests ventilation-perfusion ( V V Q ) inequality may be accentuated by a functional or structural difference, possibly induced by recurrent, intensive exercise. Ventilation/perfusion inequality (Hammond ef al., 1986; Shaffartzik et al., 1992; Hopkins et al., 1994; Rice et al., 1999a), end-capillary diffusion limitation (Torre-Bueno ef al., 1985; Hopkins et al., 1996; Rice ef al., 1999a), and relative hypoventilation (Dempsey ef al., 1984; Johnson ef al., 1992; Harms and Stager, 1995) have each been suggested to account for some portion of EIAH. Although increases in (ideal) A-aD0 2 occur in all subjects during exercise, reductions in Pa0 2 reflect greater gas exchange impairment. Defined as a minimal exercise Pa0 2 < 1.3 kPa (10 Torr) below resting values or absolute %SaG"2 <95% (Dempsey and Wagner, 1999), EIAH has been shown to correlate most strongly with excessive widening of (ideal) A-aD0 2 (Hopkins and McKenzie, 1989; Harms ef al., 1998a). Rice ef al. (1999a) failed to demonstrate any significant differences in measures of PaC0 2 or V V Q inequality as estimated by multiple inert gas elimination technique (MIGET) between athletes with EIAH versus matched controls, whereas diffusion limitation accounted for more of the observed rise in A-aD0 2 in athletes with EIAH. Further analysis with stepwise multiple linear regression explained 90% of the variance in Pa0 2, with 0 2 diffusion capacity, log SD of the perfusion distribution, and PaC0 2 each accounting for approximately 30%. In conflict with earlier studies (Hammond ef al., 1986; Wagner ef al., 1986; Shaffartzik ef al., 1992; Hopkins ef al., 1994), a rise in V V Q inequality with increasing exercise was not demonstrated. Even so, these MIGET 2 data do support a multifactorial etiology of EIAH comprised of some combination of relative hypoventilation, V V Q inequality, and diffusion limitation. Variance between studies reflects clear intersubject differences in both VA/Q dispersion and ventilatory response during exercise. A satisfactory explanation for inter-subject differences in V V Q inequality during exercise at sea level remains incomplete, and several theories were recently reviewed (Dempsey and Wagner, 1999). Although subclinical bronchoconstriction, airway secretions and/or dynamic expiratory airway compression could impair ventilation distribution (Johnson ef al., 1992), persistence of V V Q inequality past recovery of VE and Q to pre-exercise levels (Schaffartzik ef al., 1992), normalization of V V Q matching during exercise at altitude with administration of 100% O2, and lack of significant spirometric changes following exercise, lend more support to a circulatory mechanism (Gale et al., 1985; Wagner, 1992). Mediators of airway or vascular tone could also alter ventilation and/or perfusion distributions, and this is supported by demonstration of increased plasma histamine levels in athletes with EIAH (Anselme et al., 1994) and improvement in gas exchange impairment after administration of nedocromil (Prefaut ef al., 1997). In addition, increased transcapillary fluid flux from enhanced pulmonary blood flow and perfusion pressure could promote accumulation of mild interstitial edema. Regional blood flow heterogeneity, a consequence of fixed and variable components, could promote segmental overperfusion, while variation in the vascular response to increases in pressure and flow, with potential for pressure-induced endothelial dysfunction/injury and release of inflammatory mediators, could have a significant impact on V V Q ratios and diffusion. Vascular remodeling in response to endothelial injury may contribute to V V Q dispersion and gas exchange impairment at submaximal exercise. Increases in pulmonary artery pressure in excess of 40 mmHg have been recorded during upright sea level exercise in healthy human subjects (Groves ef al., 1985; Wagner ef al., 1986; Eldridge ef al., 1996). At similar pulmonary perfusion pressures, capillary stress failure has been detected in several animal models (West ef al., 1991; Zhenxing ef al., 1992; Elliot ef al., 1992). Attempts to demonstrate exercise-induced interstitial edema in humans using imaging techniques have been inconsistent. Increases in radiographic opacities consistent with interstitial lung water have been seen with computerized tomography (Cauillaud ef al., 1995) and magnetic resonance imaging (McKenzie ef al., 3 1999) following prolonged submaximal exercise. On the other hand, McKenzie et al. (1996) failed to detect increases in extravascular lung water exercise with computerized tomography following maximal exercise. Similarly, despite measured reductions in Pa02 during maximal exercise in high aerobic power male cyclists, Edwards ef al. (in press) failed to demonstrate a significant difference in pulmonary clearance of Technetium 99m DTPA before and after a progressive cycle ergometry exercise test. While these data may not support alveolar epithelial injury and leakage, they do not rule out altered integrity of the pulmonary endothelium. Endothelial dysfunction is an early feature of acute lung injury (ALI). Endothelial cell adhesion molecules (CAMs) E- and P-selectin are central to neutrophil-endothelial and platelet-neutrophil interactions. E-selectin is required in experimental rat models of neutrophil-dependent ALI induced by ischemia-reperfusion (Seekamp ef al., 1994) and intrapulmonary IgG immune complex deposition (Mulligan et al., 1993), whereas P-selectin is required for ALI induced by complement activation with cobra venom factor (Mulligan ef al., 1993). Elevated cell-specific plasma markers of endothelial activation/injury, including CAMs and von Willebrand factor antigen (vWF:Ag), have been detected in patients at risk for and with ALI (Carvalho ef al., 1982; Rubin ef al., 1990; Newman ef al., 1993; Kayal ef al., 1998). Grissom ef al. (1997) demonstrated elevated plasma levels of soluble E-selectin in subjects with hypoxic acute mountain sickness (AMS) and high-altitude pulmonary edema (HAPE), and noted a significant correlation between levels of E-selectin and the degree of hypoxemia. However, plasma levels of P-selectin were neither altered with ascent to altitude nor increased in subjects with AMS or HAPE. Eldridge ef al. (1998) also reported significantly increased plasma levels of soluble E-selectin in company with an elevated A-aD02 and elevated bronchoalveolar lavage RBC and WBC following repeated heavy exercise at altitude (3810 metres). These data imply alveolar-capillary structural failure may occur with exercise at altitude, but the contribution of exercise alone cannot be determined. To address the potential compromise of the blood-gas interface in habitually active subjects, we sought to demonstrate a relationship between endothelial dysfunction and failure of blood gas homeostasis during exercise. We hypothesized plasma levels of soluble E- and P-selectin would increase with maximal exercise in habitually active subjects and would correlate with measures of pulmonary gas exchange impairment. 4 2 Methods 2.1 Participants A convenience sample of eighteen nonsmoking eumenorrheic women, aged 22-38 years, with self-reported levels of aerobic exercise of no less than 2.5 hours per week for the preceeding 6 months, was recruited by posted notice and third party referral. All subjects presented on day 5 to 9 of their menstrual cycle having abstained from high intensity exercise, caffeine or other stimulants for twenty-four hours, and from food for four hours prior to testing. Testing was completed at least 6 hours following awakening to reduce the effect of the morning Cortisol surge. No subject reported a history of pulmonary, cardiac, vascular or autoimmune disease. Informed consent was obtained in writing. Study procedures were approved by the Clinical and Behavioral Sciences Research Ethics Board of the University of British Columbia. 2.2 Pulmonary function Following measurement of height and mass, pulmonary function was determined by spirometry. Forced expiratory volume in 1 second (FEV-i), forced vital capacity (FVC) and maximal forced expiratory flow (FEFm ax) were measured and compared to predicted normal values. Functional residual capacity was determined by helium wash-in (Collins DS/PLUS II, Braintree, MA) then added to the inspiratory capacity to calculate total lung capacity (TLC). One subject was excluded from further testing after demonstrating reduced FEVi and F E F m a x consistent with a diagnosis of small airway obstruction. 2.3 Blood sampling Arterial blood was sampled from an indwelling 20-gauge catheter inserted under sterile conditions and 1% lidocaine local anaesthetic into the non-dominant radial or brachial artery. Two subjects withdrew after failure to cannulate the radial artery. The arterial catheter was connected to a three-way stopcock and patency was maintained by intermittent flushing with a 1 mL heparin in 500 mL normal, saline solution at no greater than 3 mL per hour. Prior to nasal insertion of the esophageal thermistor to the midsternal line, blood was sampled for resting blood gas variables, hemoglobin and hematocrit, 17f3-estradiol (E2), progesterone, von Willebrand factor antigen (vWF:Ag), soluble E-selectin (CD62E) and soluble P-selectin (CD62P). Immediately following incremental maximal exercise, blood was sampled again for measurement of vWF:Ag, CD62E and 5 CD62P. Samples were collected in 7 mL EDTA vacutainer tubes for plasma endothelial markers and in plain 7 mL tubes for serum hormone levels. All blood samples were centrifuged at 2000 x g for 15 minutes. Plasma and serum were then aspirated and frozen within one hour at -20°C for later analysis. 2.4 Exercise protocol Subjects rested for 30 minutes following insertion of the arterial catheter. Then, after insertion of the nasal thermistor, each participant underwent an incremental maximal exercise test on an electronically braked cycle ergometer (Quinton Excalibur, Lode, Groningen, Netherlands). One subject withdrew from the test after feeling faint at the onset of exercise. Each participant was encouraged to pedal at a consistent cadence above 80 rpm during a ramped increase in workload beginning at 0 watts and progressing at 25 wattsmin"1. Subjects inspired using a two-way non-rebreathing valve (Hans-Rudolph, model 2700B, Kansas City, KS). Gas samples were analyzed for oxygen and carbon dioxide concentrations at a rate of 300 mL-min"1 (S-3A oxygen analyzer and CD-3A carbon dioxide analyzer, Applied Electrochemistry, Pittsburgh, PA). Analyzers were calibrated with known gases prior to each test. Heart rate was measured continuously with a portable heart rate monitor (Polar Vantage XL, Kempele, Finland). Following application of a topical vasodilator nitrate cream (Finalgon, Boehringer Ingelheim, Burlington, ON), an oximeter (Ohmeda Biox 3740, Louisville, CO) was attached to the left pinna for measurement of %Sa02- Ventilatory, gas exchange, heart rate and %Sa02 parameters were averaged and recorded every 15 seconds using a computerized data system (Rayfield, Waitsfield, VT). V0 2 max was noted by attainment of 3 of 4 criteria, including volitional fatigue, peak heart rate within 10% of predicted, RER greater than 1.10, and a plateau in V02 with increasing workload. 2.5 Blood gas analysis Blood gas samples were obtained anaerobic-ally at rest, at 3-minute (75 watt) intervals during exercise, and at peak exercise. Arterial blood was analyzed for Pa02, PaC02, pH, base excess and H C O 3 - within 10 minutes of collection (iSTAT, Abbott Laboratories), and corrected for esophageal temperature. The iSTAT analyzer was calibrated using an electronic module prior to each testing session. Measured Pa02 data that calculated an (ideal) A-aD02 of greater than or equal to minus 1 kPa (7.5 Torr) as per the (ideal) gas 6 equation (Lumb, 2000; derived from Riley ef al., 1946) were deleted. Peak exercise Pa02 values in 2 subjects predicted a negative (ideal) A-aDC>2 and these O2 values were excluded from further analysis. Similarly, Pa0 2 , %Sa02 and (ideal) A-aD0 2 data were deleted on more than two samples in four subjects. However, simultaneous measurements of %SaC>2 by ear oximetry did not fall below 95%, so the remaining blood gas data from these participants was retained for analysis. Minimal and maximal blood gas data was taken as the lowest or highest value measured during exercise. Thus, for comparisons of PaCb, %SaC>2 and (ideal) A-aD02 over the course of exercise, n=10, for minimal PaC>2, minimal %SaC>2 and maximal (ideal) A-aDC>2 values, n=12, and for all other blood gas variables and measurements, n=14. 2.6 Enzyme linked immunoassays 2.6.1 Sex hormones Serum concentrations of 17p-estradiol (E2) and progesterone were measured by enzyme immunoassay (EIA, Assay Designs Inc., Ann Arbor, Ml; #90008 and 90011). Minimal detection limits for each assay were 37 pmol-L'1 (10.1 pgmL"1) for 17p-estradiol (E2) and 8 pmol-L"1 (2.45 pg mL"1) for progesterone. One part of steroid displacement reagent was added per 99 parts of sample prior to sample dilution. Standards and samples were run in duplicate as per manufacturer's instructions. Optical density was read at 405nm with correction at 570nm. Average net optical density was calculated, and percent binding of each cell compared to the maximum binding cells was plotted against the standard concentrations of E2 and progesterone. Sex hormone serum concentrations were determined by interpolation from a straight-line equation (Appendix III, Figure. 27). E2 data from the EIA demonstrated poor reproducibility between duplicate samples with coefficients of variation (CVs) ranging from 0.9-14.9% (mean±SD, 5.2±4.0%). Moreover, all calculated E2 values were 2.5-17 fold greater than predicted ovulatory peak values of 1650 pmol-L"1. The E2 data was therefore discarded. Duplicate sample CVs for the progesterone EIA ranged between 0.2-8.4% (3.8±2.8%), with one pair exceeding the reported intraassay CV of 7.6%. This result was also deleted. The manufacturer's reported interassay CV for the progesterone EIA was 6.8%. 7 2.6.2 Soluble selectins Plasma concentrations of soluble E- and P-selectin were also measured by EIA (Bender MedSystems, Vienna; #BMS205 and BMS219/2) as per manufacturer's instructions. Minimal limits of detection for each assay were <0.5 ngmL"1 for sE-selectin and <1.3 ng mL"1 for sP-selectin. Standards and samples were run in duplicate. Optical density was read at 450 nm with correction at 620 nm. A standard curve was plotted and selectin concentrations were determined by interpolation from a straight-line equation (Appendix III, Figure. 26). Duplicate sample CVs for sE-selectin ranged from 2.0-36.5% (14.7±9.0%) pre-exercise and from 2.7-60% (22.0±15.1%) post-exercise. The reported intraassay CV was 7.1% and the interassay CV was 6.1%. Duplicate sample CVs for sP-selectin ranged from 1.8-10.2 (6.2±2.5) pre-exercise and from 0.4-8.2 (3.9±2.9) post-exercise. The reported intraassay and interassay CVs for sP-selectin were 3.5% and 12.9%, respectively. 2.7 Statistical analysis On account of missing blood gas data, a 2x4 factor repeated measures ANOVA utilizing maximum likelihood estimation procedures, and assuming compound symmetry, was used to determine differences among means on measures of pulmonary gas exchange obtained during exercise (BMDP5V). Differences between pre- and post-exercise concentrations of sE- and sP-selectin were analyzed by a one-way 2x2 factor ANOVA with repeated measures (Statistica™). Strength of the linear relationship between V02Peak, minimal Pa02 (n=12) during exercise, and absolute and relative change (%) in the plasma concentration of soluble E- and P-selectin was determined using Pearson's product moment coefficient. Significance was set at a <0.05. 3 Results 3.1 Baseline characteristics Each subject had been involved in regular aerobic physical activity for at least 3 hours per week for the preceding 6 months. Many were considerably more physically active (Table 1). Anthropometric and baseline hematologic and hormone values are shown in Table 2. Values reported throughout are means±SD. All subjects had serum progesterone levels less than 80 pmolmL"1 compatible with the mid-follicular phase. Resting lung volumes and expiratory flow rates are shown in Table 3. All subjects were 8 Table 1. Aerobic exercise habits of Individual subjects Subject Primary mode of aerobic Average volume Duration of habitual exercise or sport per week (hrs) exercise (months) MK Hiking/Mountaineering 6 9 RJ1 Running 6 >24 RJ2 Running 3 >6 DW Ice hockey 3 18 OM Commuter cycling 4 36 MS Commuter cycling/hiking 7 >60 KM Bike courier/racing 30 18 NE Running/boxing 8 18 WK Tri 12 >12 WJ Cycling/swimming 3.5 12 HJ Tri/h iking 9 >60 IS Tri 7 >24 CL Cycling/racing/running 10 36 TS Swimming/water polo 10 36 Tri, triathlon sports: swimming, cycling and running Table 2. Physical characteristics at baseline (n=14) Age, yrs 28.9 ± 5.51 Height, m 1.7 ± 0.06 Mass, kg 60.3 ± 7.51 Hgb, g-L 1 133 ± 14.5 Hct 0.39 ± 0.04 progesterone, pmolL"1 80 ± 100 Values are mean+SD; yrs, years; m, metres; kg, kilograms; g-L'1, grams per litre; pmol-L"1, picomoles per litre 9 Table 3. Lung volumes and resting spirometry (n = 14) TLC, litres 5.41 ± 0.68 (101 ± 9.3) VC, litres 4.09 ± 0.47 (111 ± 8.7) FVC, litres 4.18 ± 0.6 (114 ± 9.3) FEV1, litres 3.46 ± 0.5 (111 ± 11.0) FEV1/FVC, % 83.1 ± 6.31 (97 ± 7.2) FEFmax, Lsec-1 7.25 ± 0.93 (109 ± 14.6) Values are mean±SD; TLC, total lung capacity; VC, vital capacity; FVC, forced vital capacity; FEV-i, forced expiratory volume in 1 second; FEVi/FVC, ratio of FEVi to FVC; FEFmax, maximal forced expiratory flow in liters per second; bracketed numbers are percent predicted values±SD Table 4. Performance and ventilatory parameters at peak exercise (n = 14) HR, bpm 181 ± 10.6 Power, watts 279 ± 47.1 VE, litres-min"1 86.7 ± 17.3 V02peak, litres-min"1 3.1 ± 0.6 V02Peak, mlkg"1min"1 49.4 ± 8.2 V C 0 2 , litres-min'1 3.3 ± 0.6 RER 1.13 ± 0.1 Values are mean±SD; HR, heart rate in beats per minute; VE, minute ventilation in litres per minute; V02peak, peak rate of oxygen consumption in litres per minute; V02peak, peak rate of oxygen consumption in millilitres per kilogram per minute; VCO2, rate of carbon dioxide production at peak exercise in litres per minute; RER. respiratory exchange ratio 10 within 10% of predicted values for lung volumes and within 15% of predicted expiratory flows. 3.2 Exercise performance Peak exercise variables are shown in Table 4. A range of aerobic power was demonstrated from 1.9 to 4.2 Lmin"1 (32.3 to 63.7 mLkg"1 min"1), representing a cross-section from normally active to elite levels. Seven of the fourteen participants reached or exceeded 300 watts on the cycle ergometer. Peak exercise was marked by volitional fatigue. The respiratory exchange ratio (RER) exceeded 1.10 in all but 3 subjects, and all but one subject achieved a peak heart rate within 10% of predicted (226-age). Therefore, 13 of 14 subjects met at least 3 of 4 criteria for attainment of V02max-3.3 Pulmonary gas exchange Individual blood gas variables during progressive exercise are displayed in Figures 1-8. Minimal %Sa0 2 during exercise of <95% was used to distinguish subjects with mild to moderate EIAH from normals (Dempsey and Wagner, 1999). Pulmonary gas exchange efficiency was maintained at peak exercise in ten subjects, while decrements in arterial partial pressure of oxygen during exercise of greater than 1.3 kilopascals (10 mmHg) were seen in three of the remaining four subjects. One subject displayed a minimal %Sa0 2 of 94% and was included in the mild to moderate hypoxemia group (EIAH), although her reduction in Pa02 was less than 1.3 kPa. Averaged over the four blood gas sampling intervals from 75 watts to peak exercise, significant differences between groups were noted for Pa0 2 (p<0.01) and %Sa02 (p=0.04), while the group by time interaction for the (ideal) A-aD02 approached significance (p=0.07). However, averaged over time, alveolar P 0 2 (p=0.11), arterial P C 0 2 (p=0.36), pH (p=0.61) and temperature (p=0.99) were not significantly different between groups. Maximal and minimal blood gas data and the 2x4 ANOVA group by time interactions are shown in Table 5. Changes in blood gas variables between groups over the course of progressive exercise are shown in Figures 9-16. Pa02 was significantly correlated with (ideal) A-aD02 (r=-0.75, p<0.01), but the correlations between Pa0 2 and P A 0 2 (r=0.09, p=0.58), and Pa0 2 and P a C 0 2 (r=-0.16, p=0.32) were not significant (Figures 17-19). Figure 1. Individual Pa0 2data versus V02 \ 6 y 0^-Oxygen Consumption (L-min"1) Figure 2. Individual %Sa0 2 data versus V02 12 18 17 16 15 4 14 Q- 13 re 1 12 Q. ^ 11-1 D> s? o *-o Q> k. 3 (0 </> 0> re o 0) > ra •o V 130 C C 120 $ o 110 a> </> 100 £ 90 80 re ra 0. o > ra Q) •o Oxygen Consumption (L-min") Figure 3. Individual PA02data versus VO ra Q. "> 5 C I > CM O a. 1 3 i_ t ra o a> > re •a 2 J r 100 - 90 - 80 - 70 - 60 oo - 50 Axis - 40 >-- 30 - 20 - 10 - 0 Oxygen Consumption (L-min"1) Figure 4. Individual (ideal) A-aD02 data versus V02 < 0 1 2 3 4 Carbon Dioxide Production (L/min"1) Figure 5. Individual PCO.data versus VCO, 7.6 7.5 7.4 7.3 7.2 7.1 \ V 0 A 1 2 3 4 Carbon Dioxide Production (L-min"1) Figure 6. Individual pH data versus VCO, 100 -, Oxygen Consumption (L-min"1) Figure 7. Individual pulse oximetry data versus V02 39 -, Figure 8. Individual temperature data versus V02 15 c o o CO L-0 c E -t—' X Q . o o 0 N co » o ILU o X < UJ _C0 ro E i o 0 II Q CO +1 c ro 0 E O in cn Q CO •H c ro 0 E "3-o o V o C D T- O ) C O C O O o o o o o C N cd in od co oo o m C N Oi CNI co iri co T— C O C D m C O CM C O c\i i o o d o r- 0 0 iri CM in iri CM 0 0 cp T— in C O iri O C O C O 4 ob C N T— CM a> T— h-' C O m C O in T— C D o in o d d d d +i -H +1 +i -H -H C O in o O C O C O C O C O iri iri CM o> T— C O r- C N m m o CM iri o C O C O cp C N iri C O ob r-i- C O C D 4 C D csi iri C N i^ cn T _ C O +- +- -1-++ ++ ++ ++ C D N " N " O in d c i d d d d •H -H +1 -H -H -H +i C O m r-- C O C O r-h~ C O : iri co C D T— T— C O CM CO c 1 i ^ ro ro Q-o_ * -s C N C N O Q ro ro _ O ro -— O. C N o o < o 0 o_ ro 4 o. X ro E 0 Q . E 0 •4-* X ro E "fr C N d d d CM ^ d 9 C O C D C O T _ C D C D LT) iri ^ C D ° . d C D C N CM O -H d co co co m co C N CO C D CO -H CO m C N +1 o 0 0 E E C D C D c c c c o o _0 _0 0 0 W CO UJ DL to co < < CD C c ~ c 0 CD >1 X o 3 "a co 0 _3 ro > 0 E ° ro 2 E ^ . . to x co -k 0 O 0 _ CO CU CO CU 0 .5 i- t -ro ro CO C L _g ro 5 0 j _ - ro ro _ 0 E c c 0 'E c ^ 0 O -a ro I | o E ro LU Q . ro a; ro .1 - Q -3 2. E § ro I o ro c "ro to" E ro x ° J2 ro E CL . o c5^ ^ • ^ X cu ro -g t x .. o ro o c co o ro a. ±. Jg s c o o 0 0 J= 0 •= to = E 0 Q . CO E ro C D O c ro c ^ •£ o -B c o O 0 ° 0 ro co E ol ro 0 " 5 . - Q CU CO ro c 0 ro o 0 to o — - CU 0 C L ^ c 1 in — |OJ c o II x ro o w x: X) _ o l-S o 0 g Zi — •!= to ro 0 £ c to 0 to 0 2 fc C L ^ ro s i X _ o ro "ro 0 c t : ^ ro ro " r o xi ro .g ro "5 E 0 CO X Qj O ro CO cf .g II c -H- CU CO oo 'P. II CD C X +- 0 ro o 0 > ro . CM ro O E O x .ro 2 'x fc ro * E O a « ro s < to x ro ro Q . E . - ^ to _ro o. 0 cn c ro x: _ o s§ II 0 XJ cn ro 0 -T3 .C 0 CO LU co < in to o 0 0 to 0 ro ol 0 ^ a. ^ E 0 0 j= "*-* —^ "ro = E E 'ro 0 ro o t  co E ro ro 0 a . cn o = x ^ o ro >- o 0 c Q . ro E c 16 16 120 ra 0. i . 15 -I 0) U) O o p 14 13 J in (A I 12 ra '€ ra Q. 11 ra "SJ 10 75 150 225 300 Power Output (W) PEAK 110 100 90 V 80 c a> ST O w w a> ra '€ ra a. < Figure 9. Differences in mean Pa0 2 betwen groups ra Q. * 4 0) o c Q> J? o 0) ra o > o 30 H o c 0) 20 10 U -10 c a> S? o a> CO o CD > 75 150 225 300 Power Output (W) PEAK Figure 10. Differences in mean (ideal) A-aD0 2 between groups 17 o 0 75 150 225 300 PEAK Power Output (W) Figure 11. Differences in mean PAQ2 between groups Power Output (W) Figure 12. Differences in mean PaC0 2 between groups 18 100 Power Output (W) Figure 13. Differences in mean %Sa0 2 between groups 100 -, Power Output (W) Figure 14. Differences in mean pulse oximetry between groups 7.50 -, Power Output (W) Figure 15. Differences in mean pH between groups 39 n Power Output (W) Figure 16. Differences in mean temperature between groups 20 Arterial Partial Pressure of Oxygen (kPa) Figure 17. Relationship between Pa0 2 and the (ideal) A-aD0 2 0 10 120 r- 110 O) S? O o CD L_ 3 (0 W Q) 100 _ ro t (0 Q. 90 « o 0) > re O) 11 12 13 14 15 16 Arterial Partial Pressure of Oxygen (kPa) Figure 18. Relationship between Pa0 2 and PA02 t 11 12 13 14 15 Arterial Partial Pressure of Oxygen (kPa) o 50 t X 4= | c o 40 "2 CO o o 35 a> 3 (0 w 30 fi 25 « Q. "fo V 16 < Figure 19. Relationship between Pa0 2 and PaC0 2 22 3.4 Soluble selectins A 2x2 ANOVA failed to demonstrate a significant difference in plasma concentrations of soluble E-selectin before and after exercise (pre-exercise = 27.7±12.5 ngmL"1; post-exercise = 31.7+17.1 ngmL"1, F112=2.29, p=0.16), but plasma concentrations of soluble P-selectin rose significantly (pre-exercise = 76.6+19.8 ngmL"1; post-exercise = 98.1 ±28.0 ng mL"1, F1|12=10.62, p=0.007). No significant group by time interaction was noted in pre-post exercise concentrations of either sE-selectin (F1i12=0.11, p=0.74) or sP-selectin (Fi,i2=0.70, p=0.42) between subjects who demonstrated normal gas exchange and subjects who displayed mild to moderate exercise-induced gas exchange impairment (Table 5). The changes with exercise of sE- and sP-selectin concentrations are shown in Figures 20 and 21. 3.5 Soluble selectins and pulmonary gas exchange Correlations between absolute change (absA, ngmL"1) and relative change (relA, %) in sE- and sP-selectin and V02peak in mLkg"1 min"1 (absAsE, r=-0.37, p=0.19; relAsE, r=-0.38, p=0.19; absAsP, r=-0.07, p=0.81; relAsP, r=-0.20, p=0.50), were not significant, and neither were the correlations between minimal exercise Pa0 2 and either absolute (r=0.16, p=0.61) or relative (r=0.18, p=0.57) change in sE-selectin. However, absolute change in plasma levels of soluble P-selectin was significantly correlated with minimal Pa0 2 (r=-0.60, p=0.04), while the correlation between the relative change in sP-selectin and minimal Pa0 2 approached significance (r=-0.46, p=0.14). Correlations between absolute and relative change in sE- and sP-selectin concentrations and minimal Pa0 2 are shown in Figures 22-25. 4 Discussion 4.1 Pulmonary gas exchange In this descriptive study of a small convenience sample of female habitual exercisers with a range of aerobic power from active to elite, we have demonstrated mild to moderate impairment in pulmonary gas exchange during progressive exercise in 4 of 7 female subjects with a VO^eak^O mLkg"1 min"1, but in no subject with a VO^eak^O mLkg"1 min"1. This apparent threshold of EIAH in women is consistent with recent data (Hopkins et al., 2000). Nevertheless, our sample size is insufficient to make any estimates of incidence or prevalence of EIAH among female athletes. Reduction in arterial oxygenation was not as severe in these subjects compared with previously reported values in 29 female runners 2 3 90 -, ~ 80 D) C 70 UJ 60 CM <o Q O 50 t3 40 -I a> «» 30 111 « J3 20 J O (0 10 PRE-EXERCISE POST-EXERCISE Figure 20. Individual mean pre- and post-exercise plasma concentrations of sE-selectin Figure 21. Individual mean pre- and post-exercise plasma concentrations of sP-selectin 12 11 10 -15 -10 -5 10 15 A sE-selectin (CD62E) (ngmL' 1) 20 25 Figure 22. Relationship between absolute change in sE-selectin concentration and minimal PaO„ 16 15 14 13 12 -I 11 10 -40 -20 0 20 40 A sE-selectin (CD62E) (%) 60 80 Figure 23. Relationship between relative change in sE-selectin concentration and minimal Pa0 2 11 J 10 -40 — i — -20 0 20 40 AsP-selectin (CD62P) (ng-mL'1) 60 80 16 15 Figure 24. Relationship between absolute change in sP-selectin concentration and minimal PaO, ^ 14 ro Q. 13 12 = 11 10 V -40 -20 0 20 40 60 80 A sP-selectin (CD62P) (%) 100 120 Figure 25. Relationship between relative change in sP-selectin concentration and minimal Pa0 2 26 (Harms et al., 1998a), but was more marked than data from 17 female athletes during a progressive cycle ergometry exercise test (Hopkins et al., 2000). Aerobic power was also lower than in the group of runners (V02peak = 49.4±8.2 vs. 57±6 mLkg"1 min"1), but approximated the range of aerobic power in the 17 female subjects who underwent both treadmill and cycle exercise testing (V02peak= 49.4±8.2 vs. 51 ±2 (mean±SE) mLkg"1 min"1). Differences in mode of exercise (cycle vs. treadmill) may contribute to the discrepancy in gas exchange impairment between studies (Gavin and Stager, 1999; Hopkins etal., 2000; Rice et al., in press). Reduction in Pa0 2 corresponded to an increase in the (ideal) A-aD0 2 , but was not significantly related to measures of alveolar ventilation. Moreover, 6 subjects displayed an absent hyperventilatory response with peak exercise PaC0 2 values >5.1 kPa (38 Torr). Failure to demonstrate a statistically significant difference in the (ideal) A-aD0 2 between groups is accounted for by a small sample size and insufficient statistical power with repeated measures ANOVA. A Student's t test comparing mean peak (ideal) A-aD02 between normals and subjects with mild to moderate EIAH demonstrates a significant difference (normals 1.7±0.7 kPa (95%CI:1.1-2.2), EIAH 3.5±1.1 kPa (95%CI:2.7-4.3), effect size=-1.9, p=0.04). Furthermore, statistical significance aside, a physiologically significant difference of >1.3 kPa (10 Torr) was seen between mean values at peak exercise. Hence, gas exchange inefficiency, not ventilatory embarrassment, seemed to account more for the degree of EIAH in these subjects. In addition, differences in arterial oxyhemoglobin desaturation between subjects with and without mild to moderate EIAH were explained primarily by reduction in Pa0 2 . Temperature and pH changes were not significantly different between groups. 4.2 Cell adhesion molecules Cell adhesion molecules mediate interactions between cells, and between cells and the extracellular matrix. The selectins are a group of highly conserved lectins expressed on the surface of leukocytes, endothelial cells (EC) and platelets, of which E-selectin and P-selectin are expressed by the endothelium. They are central to neutrophil-endothelial and neutrophil-platelet interactions and play a significant role in the inflammatory response. 27 Sex hormones appear to effect CAM expression, but are least likely to interfere during the follicular phase. E 2 did not significantly effect serum levels of sE-selectin (CD62E), slCAM-1(CD54) or sVCAM-1(CD106) in 18 males administered 10 mg of E 2 valerate intramuscularly in a double blind, randomized, placebo controlled, cross-over study (Jilma ef al., 1994). E 2 administration was associated with a decrease in sP-selectin (CD62P). However, sP-selectin levels measured during the follicular phase were not significantly different from males (Jilma ef al., 1996a). 4.2.1 Soluble E-selectin E-selectin is a 140 kDa transmembrane protein expressed by EC in response to IL-ip and TNF-ct (Bevilacqua ef al., 1987, 1989). E-selectin is formed de novo and requires 2 to 6 hours to reach peak levels. Soluble forms are released when EC are activated, making sE-selectin a potentially useful marker of inflammation. Jilma et al. (1997a) measured a <11% rise in slCAM-1, sVCAM-1, and soluble sE-selectin in twelve untrained men after both maximal cycle ergometry and sixty minutes at 60%VO 2 m a x. This failed to reach a preset clinically relevant difference of 15%, based on a previously reported day-to-day variability of 8% (Jilma ef al., 1994). They concluded that recreational activity at sea level has insignificant effects on these circulating CAMs, but specifically excluded subjects who swam >1 km/week, ran >5 km/week or cycled >50 km/week. In contrast, soluble ICAM-1 was elevated by 20% following an endurance training program in middle-distance runners (Baum ef al., 1994), and by 23% after a roundtrip ascent and descent from 750m to 2350m over a five hour period in eighteen untrained males (Tilz et al., 1993). Levels rose coincidental^ with increased SIL-2R and following a rise in sTNF-R (Tilz ef al., 1993) suggesting a possible cytokine induction. These changes compare to a 20-30% increase seen in patients with insulin-dependent diabetes mellitus, atherosclerosis (Ridker ef al., 1998) and inflammatory vascular disease. The discrepancy in results between the studies of Jilma ef al. (1994) and Tilz ef al. (1993) may reflect differences in levels of activity and aerobic fitness and/or the influence of altitude. Eldridge ef al. (1998) reported significantly increased levels of sE-selectin at altitude (3810 metres) (28.7+8.9 ngmL"1 pre-exercise, 39.4±14.2 ngmL'1 24-hours post-exercise, p<0.05), in company with an elevated A-aD0 2, and elevated bronchoalveolar lavage (BAL) RBC (p=0.03) and WBC (p=0.08) at 24 hours, in 5 subjects who exercised at 85%V0 2 m a x for three 5 minute intervals with a 5 minute recovery at 30%VO 2 m a x between. Likewise, 28 alterations in pulmonary endothelial function on exposure to hypobaric hypoxia is supported by demonstration of increased levels of plasma E-selectin in 6 subjects with hypoxemic acute mountain sickness (AMS) and 8 climbers with high altitude pulmonary edema (HAPE) presenting to the National Park Service medical camp at 4200 m on Denali (Grissom et al., 1997). Plasma E-selectin levels increased significantly in 17 control subjects on ascent from sea level to 4200 m (mean+SD, 12.9±8.2 ngmL"1 versus 17.2+8.2 ng-mL'1, p=0.001), but were significantly higher compared to sea level control values in subjects with hypoxemic AMS (30.6+13.4 ngmL"1) and HAPE (23.3+9.1 ngmL"1) (p=0.009). Significant correlation was also seen between plasma E-selectin levels and the degree of hypoxemia (p=0.006). However, differences in plasma E-selectin levels between control subjects at altitude and subjects with hypoxemic AMS and HAPE, perhaps a more telling comparison, was not reported. Plasma P-selectin levels were unchanged on ascent to altitude or in subjects with either AMS or HAPE. This data implies alveolar-capillary structural failure with exercise at altitude, but the contribution of exercise alone cannot be determined, and this conclusion may not be applicable at sea level. Despite previously reported increases in pro-inflammatory cytokines with exercise, failure to demonstrate a significant rise in soluble E-selectin levels during progressive exercise in the current study is compatible with an insufficient time course for induction and expression. The cycle ergometry exercise test lasted between 9 and 13 minutes, whereas ascent to high altitude with onset of AMS or HAPE generally takes several days. The difference in time course would allow for cytokine induction and likely explains the discrepant rise in sE-selectin with exercise at altitude but not at sea level. Background levels of sE-selectin were also lower than previously reported levels in men (Jilma ef al., 1994; Grissom et al., 1997). The large sample CVs in the current data raises questions of reliability and accuracy, and any interpretation is necessarily guarded. 4.2.2 Soluble P-selectin The CAM P-selectin is a 140 kDa glycoprotein stored in EC Weibel-Palade bodies and a granules of platelets. It is expressed by Ca2+-dependent exocytosis and can be induced by histamine (Lorant ef al., 1991), activated complement (Mulligan ef al., 1997), and TNF-a upregulation (Weller ef al., 1992). Expression of P-selectin is undetectable in resting blood vessels, but is enhanced by several injury stimuli. Its expression, therefore, has also been used as a marker of early inflammation. The source of elevated levels of 29 soluble P-selectin remains uncertain, with evidence for a contribution from both EC and platelets. 4.2.2.1 Endothelium High molecular weight multimers of von Willebrand factor (vWF) are also stored in EC Wiebel-Palade bodies and form the pool of protein that is most likely released at the time of vascular activation or injury (Wagner, 1990). As such, vWF.Ag has been considered a sensitive marker of endothelial activation (Hamilton ef a/., 1987). In healthy subjects, levels of vWF have been shown to rise in an intensity-dependent relationship with exercise (Andrew ef a/., 1986; Wheeler ef a/., 1986), while normal resting levels of vWF:Ag did not vary with sex (Blann, 1990). Whether or not the rise in vWF reflects endothelial activation or injury is not known. Lack of correlation between an increase in soluble P-selectin and vWF in studies of patients at risk for or with cardiovascular disease (Lip etal., 1995; Blann etal., 1995), does not appear to support an EC source for sP-selectin (Blann and Lip, 1997). Nailin ef al. (1999) reported a significant rise in both sP-selectin and vWF following exhaustive exercise in 15 male subjects but did not report correlation. They did not show an effect on exercise-induced increases in sP-selectin, vWF or platelet P-selectin expression with pretreatment with 500mg/day of aspirin in an open crossover trial, although aspirin did attenuate platelet P-selectin expression at rest. The lack of an effect of aspirin on exercise-induced rise in vWF and sP-selectin would be compatible with an EC source. Alternately, using the styryl dye FM1-43, which fluoresces brightly upon binding cell membranes, Kuebler ef al. (1999) determined FM1-43 fluorescence in conjunction with intravital quantifications of P-selectin expression and EC [Ca2+]j in venular capillaries of isolated, blood perfused rat lung. FM1-43 colocalized with P-selectin and was inhibited by blockade of mechanogated Ca2+-channels. These animal data support a role for pressure-induced alterations in permeability via Ca2+-mediated processes, and potentially by initiating CAM expression, neutrophil adhesion and inflammation. If mechanical pressure induces EC P-selectin expression in lung venules, elevation of sP-selectin levels may reflect a rise in pulmonary perfusion pressure. However, the lack of rise in sP-selectin at altitude in subjects with HAPE, in whom pulmonary hypertension is considered a sine qua non, does not appear to support this hypothesis. 30 4.2.2.2 Platelets Exercise has been shown to induce platelet activation, coagulation, and fibrinolysis (Cash, 1966; Davis ef al., 1976) relative to exercise intensity (Andrew ef al., 1986). Platelet function was desensitized by short-term moderate exercise (50% V0 2 m ax) , but potentiated by strenuous exercise (V0 2 m a x) (Wang et al., 1994). While this pattern also held true for women in the mid-follicular phase, platelets adhesiveness and aggregation were not enhanced by acute heavy exercise at mid-luteal phase (Wang ef al., 1997). Weiss ef al. (1998) measured markers of thrombin, fibrin and plasmin formation in 12 male subjects who underwent 1 hour of treadmill running at 68% and 83% of V 0 2 m a x and demonstrated a balanced rise in both fibrinolytic and thrombogenic activity. However, Mockel ef al. (1999) demonstrated a significant increase in platelet expression of P-selectin in 15 male triathletes after exhaustive exercise. The rise in P-selectin expression was more marked after intensive exercise but was independent of the platelet count. Levels returned to baseline within 30 minutes. Fibrin monomer and pro-thrombin was also elevated and maximal levels were attained at 30 minutes post exercise, suggesting the potential for a vulnerable coagulation window following intensive exercise. 4.3 Cell adhesion molecules and pulmonary gas exchange Demonstration of a significant association between sP-selectin and minimal Pa0 2 in the current study does not distinguish between either an EC or platelet source, although correlation with a rise in vWF may aid in this distinction. However, it need not be one or the other. Activated platelets may interact with neutrophils in a cross over response between hemostasis and inflammation in aid of vascular repair, and neutrophil activation may contribute to vascular permeability through release of vasoactive mediators. The potential relationship of exercise, platelet function, P-selectin and 17p-estradiol, and their influence on neutrophil traffic through the lung, remains incompletely addressed. A significant rise with progressive exercise of soluble P-selectin, but not soluble E-selectin, suggests endothelial or platelet activation without cytokine induction. Although no significance was demonstrated between groups, there was a trend towards a greater rise in sP-selectin in those subjects who demonstrated the most gas exchange impairment. If 31 related, the increase in sP-selectin may reflect exercise-induced mechanical and/or inflammatory endothelial activation/injury. Strenuous exercise incites an acute inflammatory response, marked by leukocytosis and neutrophil activation, release of inflammatory mediators and acute phase proteins, and activation of complement, coagulation and fibrinolytic cascades, that is akin to, but quantitatively different from the acute phase and systemic inflammatory response accompanying sepsis, major burns and trauma (Weight et al., 1991; Camus et al., 1994; Northoff et al., 1994; Moyna ef al., 1996; Pedersen et al., 1997; Skek and Shephard, 1998). All subtypes of natural killer, B, and T lymphocytes are recruited, and neutrophils, monocytes and platelets increase in numbers. An extensive arsenal of inflammatory mediators, including eicosanoids, reactive oxygen species, cytokines and vasoactive amines, are elevated by exercise (Cannon ef al., 1986; Sprenger ef al, 1992; Anselme ef al., 1994; Camus ef al., 1994; Northoff ef al., 1994; Weinstock ef al., 1997; Pedersen, ef al., 1997). Although the physiological and clinical significance of these alterations remains undetermined (Smith, 1997), the pulmonary endothelium may be at risk for inflammatory injury. Recognizing soluble selectins are systemic measures, which may not reflect the pulmonary circuit, the significance and relevance of an increase in sP-selectin with exercise is uncertain. However, its association with gas exchange impairment, if not spurious, cautiously points toward platelet and endothelial function and/or interaction at the blood-gas interface. If reflective of endothelial activation or injury, it would suggest the potential for vascular remodeling and repair, which may, or may not, have implications on ventilation-perfusion inequality. 4.4 Vascular remodeling Intersubject variability in vyo. dispersion during exercise may betray variations in vascular remodeling and perfusion distribution. The potential for induction of pulmonary endothelial activation/injury by recurrent increases in pulmonary blood flow may affect regional vascular response to flow. Structural changes of the pulmonary circulation occur in response to elevations in pulmonary arterial pressure or long-term increases in flow, and endothelial dysfunction forms the early response. EC injury, muscular layer development, increases in SMC, and medial thickening appears to occur irrespective of the inciting 32 event. Genes or gene products characteristic of earlier development are expressed by SMC in response to injury, and phenotypic heterogeneity of SMCs adds to the complexity of response (see Stenmark and Mecham, 1997 for review). Local vascular injury induces expression and release of varying growth factors, while reiteration of fibroblast mRNA induces collagen deposition and protein production. Protein kinase C and C a 2 + may be involved in mechanotransduction in response to shear stress, although pressure elevation alone may not be sufficient to induce the remodeling process. The presence of inflammatory signals may be required (Tanaka et al., 1996). Remodeling of the vascular endothelium, with impact on functional and/or structural perfusion distribution, may, in turn, influence regional perfusion heterogeneity and VA/Q matching. 4.5 Regional perfusion heterogeneity 4.5.1 Structural heterogeneity Heterogeneity of blood flow distribution is enhanced by structure of the vascular bed. Regional blood flow heterogeneity exceeds predicted variability based on gravity alone (Glenny et al., 1991; Prisk et al., 1994; Hlastala et al., 1996) and can be characterized more completely by fractal methods (Glenny and Robertson, 1990; Glenny and Robertson, 1991). Studies utilizing 15 pm fluorescent microspheres have demonstrated stability of high flow and low flow areas over time in dogs (n=5, Glenny ef al., 1997), at rest and during maximal exercise in thoroughbred racehorses (n=4, Bernard et al., 1996), and, in a preliminary study, with changes in posture in baboons (n=2, Glenny et al., 1997). In the latter, a strong association was demonstrated between blood flow and height up the lung in the upright posture (^ =0.65), whereas in the head-down position, height up the lung accounted for only 3% of blood flow variability. Together, these data support the hypothesis that regional pulmonary perfusion is determined primarily by a fixed structure and suggest preference of blood flow for basal regions is based more on vascular anatomy than hydrostatic forces. 4.5.2 Functional heterogeneity Local factors and perfusion pressures may also influence variability of blood flow distribution. Phenotypic heterogeneity for nitric oxide (NO) production has been demonstrated between microvascular and macrovascular endothelial cells in rats (Geiger et al., 1997). Regional heterogeneity of NO release has also been demonstrated in thoroughbred racehorses, with preference for caudodorsal vessels (Pelletier et al., 1998). 33 This latter observation is in keeping with previously documented preferential distribution of pulmonary blood flow to dorsal lung regions in exercising horses (Bernard ef al., 1996). In addition, using flourescent-labelled aerosol and radioactive-labelled microspheres in standing sheep, and blocking endogenous NO production with L-NAME in the presence or absence of exogenous NO (30 ppm), Melsom et al. (2000) noted no alteration in perfusion along horizontal levels when varying access to NO, but did demonstrate increased homogeneity of blood flow along the gravitational axis when NO availability was reduced. Delivery of NO restored the vertical perfusion distribution, but variable access to NO in this animal model had no affect on arterial oxygenation. 17p-estradiol has been shown to exhibit differential effects on NOS, up-regulating endothelial NOS (eNOS) in cultured EC (Hayashi ef al., 1995), but down-regulating inducible NOS (iNOS) in rat aorta (Duckies et al., 1996). Heterogeneity has also been seen in estrogen modulation of endothelin-1 induced vasoconstriction in dog coronary conduit or microvessels depending on vessel size (Lamping and Nuno, 1996). Furthermore, significant between-gender effects have been demonstrated for whole body measures of NO even if normalized for body weight or surface area. Exhaled NO and plasma nitrate (N03-), a stable NO end product, were significantly lower in women than men (p<0.001 and p<0.0001, respectively), but did not vary significantly across the menstrual cycle (Jilma et al., 1996b). Plasma NO metabolites did, however, display a different response to exercise depending on the menstrual phase. The percent increase in nitrite to nitrate (N02-/N03) was greatest after moderate exercise at the mid-follicular phase, but was significantly less (p<0.05) following similar exercise intensity at the mid-luteal phase, or after maximal exercise at either phase (Wang et al., 1997). The effects of E 2 on the adult human pulmonary circulation are not well understood, but data from animal models (sheep and rat) suggest it may influence hypoxic vasonstriction (Wetzel ef al., 1984; Gordon ef al., 1986) and pulmonary vascular remodeling (Farhat et al., 1993). An influence on endothelial function is further supported by the interaction between E 2 and NO. Hence, the pulmonary vascular tree appears to demonstrate structural (fixed) and functional (variable) heterogeneity. Regional perfusion, in turn, is determined by the combination of their relative contributions. Furthermore, the heterogeneity of perfusion, 34 neither uniform nor random, does not appear to change substantially with exercise: high flow areas remain high flow and low flow areas remain low flow (Bernard et al., 1996). Increased blood flow and pressure during near maximal and maximal exercise may overcome local redistribution factors leaving flow distribution dependent on underlying structural heterogeneity. If these relationships seen in animal data are also applicable to the human pulmonary circulation, excessive perfusion of relatively poorly ventilated areas could accentuate venous admixture and contribute to development of hypoxemia. The potential influence of recurrent exercise and sex hormones on the pulmonary endothelium and the corresponding effect on functional perfusion heterogeneity awaits further study. A comparison of gas exchange measures during exercise across phases of the menstrual cycle, in conjunction with measures of endothelial function and inflammation, may help elucidate the interplay of these factors. 4.6 Conclusion Exposed to the entire blood volume, and central to inflammatory and hemostatic reactions, control of vascular tone, and processes of vascular injury, repair and remodeling, the pulmonary endothelium is situated to play a major role in blood-gas barrier integrity and gas exchange. Alterations in membrane permeability via mechanotransduction and increases in [Ca2+]j, and/or EC activation/injury in response to mechanical stress and inflammatory mediators, may effect an increase in interstitial lung water during exercise. Injury repair and remodeling may also influence the vascular response to flow, thereby initiating a cycle. Altered hemodynamics could also influence functional blood flow redistribution and VA/Q matching. Whether or not any of these mechanisms play a role in EIAH is open to further research, but one or more could potentially explain the finding of altered gas exchange impairment seen during submaximal exercise in some habitually active subjects. Demonstration of a significant relationship between impairment of arterial oxygenation during exercise and a rise in soluble P-selectin is compatible with this hypothesis. In summary, we have demonstrated exercise-induced gas exchange impairment in 4 of 7 female subjects with a v/O2peak>50 mLkg"1 min"1, but in none of 7 subjects with a VO 2p eak <50 mLkg"1 min"1. Reduction in Pa0 2 corresponded to an increase in the (ideal) A-aD0 2 , but was not significantly related to measures of alveolar ventilation. Differences in arterial oxyhemoglobin desaturation between subjects with and without mild to moderate 35 EIAH were primarily explained by reduction in Pa0 2 . Temperature and pH changes over the course of progressive exercise were not signficantly different between groups. A significant rise in soluble P-selectin, but not soluble E-selectin, suggests endothelial or platelet activation may occur without cytokine induction. Demonstration of a significant association between a rise in sP-selectin and minimal exercise Pa0 2 cautiously points toward platelet and endothelial function and/or interaction at the blood-gas interface with potential for vascular remodeling and repair, and impact on ventilation-perfusion inequality. 36 5. References Aarron, E.A., K.G. Henke, D.F. Pegelow, and J.A. Dempsey. Effects of mechanical unloading of the respiratory system on exercise and respiratory muscle endurance. Med. Sci. Sports Exerc. 17:290, 1985. [Abstract] Aaron, E.A., K.C. Seow, B.D. Johnson, and J.A. Dempsey. Oxygen cost of exercise hyperpnea: implications for performance, d. Appl. Physiol. 72:1818-1825, 1992. Ahmed, S.A., W.J. Penhale, and N. Talal. Sex hormones, immune responses, and autoimmune diseases. Am. d. Pathol. 121:531-551, 1985. Andrew, M., C. Carter, H. O'Brodovich, and G. Heigenhauser. Increases in factor Vlll complex and fibrinolytic activity are dependent on exercise intensity. d. Appl. Physiol. 60:1917-1922, 1986. Anselme, F., C. Caillaud, I. Couret, M. Rossi, and C. Prefaut. Histamine and exercise-induced hypoxemia in highly trained athletes. J. Appl. Physiol. 76:127-132, 1994. Aoki, N., M. Siegried, and A.M. Lefer. Anti-EDRF effect of tumor necrosis factor in isolated, perfused cat carotid arteries. Am. d. Physiol. 256:H1509-H1512, 1989. Arora, S., A. Veves A, E. Caballaro, P. Smakowski, and F.W. LoGerfo. Estrogen improves endothelial function, d. Vase. Surg. 27:1141-1147, 1998. Arnett, F.C. High-altitude pulmonary edema: an immunogenetically mediated disease? Circulation 97:1111 -1113, 1998. [Editorial] Asmussen, E., and M. Neilsen. Alveolar-arterial gas exchange at rest and during work at different oxygen tensions. Acta Physiol. Scand. 50:153-166, 1960. Babcock, M.A., D.F. Pegelow, B.H. Taha, and J.A. Dempsey. High frequency diaphragmatic fatigue detected with paired stimuli in humans. Acta Physiol. Scand. 50:153-166, 1998. 37 Bake, B., J . Bjure, and J . Widimsky. The effect of sitting and graded exercise on the distribution of pulmonary blood flow in healthy subjects studied with the 133xenon technique. Scand. J. Clin. Lab. Invest. 22:99-106, 1968. Bartsch, P. High altitude pulmonary edema. Med. Sci. Sports Exerc. 31(1 Suppl):S23-27, 1999. Bartsch, P., U. Eichenberger, P.E. Ballmer, J.S. Gibbs, C. Schirlo, O. Oelz, and E. Mayatepek. Urinary leukotriene E(4) levels are not increased prior to high-altitude pulmonary edema. Chest 117:1393-8,2000. Bauer, J.A., J.A. Wald, S. Doran, and D. Soda. Endogenous nitric oxide in expired air: effects of acute exercise in humans. Life Sciences 55:1903-1909, 1994. Baum, M., H. Liesen, and J . Enneper. Leucocytes, lymphocytes, activation parameters and cell adhesion molecules in middle-distance runners under different training conditions. Int. d. Sports Med. 15:S122-S126, 1994. Beck, L., Jr., and P.A. D'Amore. Vascular development: cellular and molecular regulation. FASEBJ. 11:365-373,1997. Bender, P.R., and B.J. Martin. Maximal ventilation after exhaustive exercise. Med. Sci. Sports Exerc. 17:164-167, 1985. Bernard, S.L., R.W. Glenny, H.H. Erickson, M.R. Fedde, N. Polissar, R.J. Basaraba, and M.P. Hlastla. Minimal redistribution of pulmonary blood flow with exercise in race horses. d. Appl. Physiol. 81:1062-1070, 1996. Bevilacqua, M.P., J.S. Pober, D.L. Medrick, R.S. Cotran, and M.A. Gimbrone, Jr. Identification of an inducible endothelial-leukocyte adhesion molecule. Proc. Natl. Acad. Sci. USA 84:9238-9242, 1987. 38 Bevilacqua, M.P., S. Stengelin, M.A. Gimbrone, Jr., and B. Seed. Endothelial leukocyte adhesion molecule-1: an inducible receptor for neutrophils related to complement regulatory proteins and lectins. Science 243:1160-1165, 1989. Blann, A.D. Normal levels of von Willebrand factor antigen in human body fluids. Biologicals 18:351-353, 1990. Blann, A.D. von Willebrand factor and the endothelium in vascular disease. Br. J. Biomed. Sci. 50:125-134, 1993. Blann, A.D., M. Dobrotova, P. Kubisz, and C.N. McCollum. von Willebrand factor, soluble P-selectin, tissue plasminogen activator and plasminogen activator inhibitor in athersclerosis. Thromb. Haemostasis. 74:626-630, 1995. Blann, A.D., and G.Y.H. Lip. Hypothesis: is soluble P-selectin a new marker of platelet activation? Atherosclerosis 128:135-138, 1997. Bone, R.C. Toward a theory regarding the pathogenesis of the systemic inflammatory response syndrome: what we do and do not know about cytokine regulation. Crit. Care Med. 24:163-172, 1996. Bosenberg, A.T., J .G. Brock-Utne, S.L. Gaffin, M.T. Wells, and G.T.W. Blake. Strenuous exercise causes systemic endotoxemia. J. Appl. Physiol. 65:106-108, 1988. Boutellier, U., and P. Piwko. The respiratory system as an exercise limiting factor in normal sedentary subjects. Eur. J. Appl. Physiol. 64:145-152, 1992. Boutellier, U., R. Buchel, A. Kundert, and C. Spangler. The respiratory system as an exercise limiting factor in normal trained subjects. Eur. J. Appl. Physiol. 65:347-353, 1992. Brenner, O. Pathology of the vessels of the pulmonary circulation. Arch. Int. Med. 56:211, 1935. 39 Brigham, K.L. Mechanisms of endothelial injury. In: Ryan US, ed. Pulmonary Endothelium in Health and Disease New York: Marcel Dekker, 207-236, 1987. Brock-Utne, J.G. , S.L. Gaffin, M.T. Wells, P. Gathiram, E. Sohar, M.F. James, D.F. Morrell, and R.J. Norman. Endotoxinaemia in exhausted runners after a long-distance race. S. Afr. Med. J. 73:533-536, 1988. Bryne-Quinn, E., J.V. Weil, I.E. Sodal, G.F. Filley, and R.F. Grover. Ventilatory control of the athlete. J. Appl. Physiol. 30:91-98, 1971. Buck, C.A., J.M. Edelman, C E . Buck, G. Kennedy, and H.S. Baldwin. Expression patterns of adhesion receptors in the developing mouse lung: functional implications. Cell Adhesion & Comm. 4:69-87, 1996. Buono, M.J., J.H. Wilmore, and F.B. Roby, Jr. Indirect assessment of thoracic fluid balance following maximal exercise in man. J. Sport Science 1:217-226, 1983. Buono, M.J., and R. Maly. Augmented hyperventilation via normoxic helium breathing does not prevent exercise-induced hypoxemia. Can. J. Appl. Physiol. 21:264-270, 1996. Bury, T.B., R. Louis, M.F. Radermecker, and F. Pirnay. Blood mononuclear cells mobilization and cytokines secretion during prolonged exercise. Int. J. Sports Med. 17:156-160, 1996. Caillaud, C , F. Anselme, J . Mercier, and C. Prefaut. Pulmonary gas exchange and breathing pattern during and after exercise in highly trained athletes. Eur. J. Appl. Physiol. 67:431-437, 1993. Caillaud, C , O. Serre-Cousine, F. Anselme, X. Capdevilla, and C. Prefaut. Computerized tomography and pulmonary diffusion capacity in highly trained athletes after performing a triathlon. J. Appl. Physiol. 79:1226-1232, 1995. 40 Caillaud, C.F., F.M. Anselme, and C.G. Prefaut. Effects of two successive maximal exercise tests on pulmonary gas exchange in athletes. Eur. J. Appl. Physiol. & Occup. Physiol. 74:141-147, 1996. Camus, G., G. Deby-Dupont, J . Duchanteau, C. Deby, J . Pincemail, and M. Lamy. Are similar inflammatory factors involved in strenuous exercise and sepsis? Int. Care Med. 20:602-610, 1994. Cannon, J .G. , and M.J. Kluger. Endogenous pyrogen activity in human plasma after exercise. Science 220:617-619, 1983. Cannon, J .G. , W.J. Evans, V.A. Hughes, C.N. Meredith, and C A . Dinarello. Physiological mechanisms contributing to increased interleukin-1 secretion. J. Appl. Physiol. 61:1869-1874, 1986. Carvahlo, A.C.A., S.M. Bellman, V.J. Saullo, D. Quinn, and W.M. Zapol. Altered factor Vlll in acute respiratoryfailure. N. Engl. d. Med. 307:1113-1119, 1982. Cash, J.D. Effect of moderate exercise on the fibrinolytic system in normal young men and women. Br. Med. d. 5512:502-506, 1966. Castell, L.M., J.R. Poortmans, R. Leclercq, M. Brasseur, J . Duchateau, and E.A. Newsholme. Some aspects of the acute phase response after a marathon race, and the effects of glutamine supplementation. Eur. d. Appl. Physiol. & Occup. Physiol. 75:47-53, 1997. Catravas, J.D., A. Cziraki, A. Chester, A.F. Maizza, S. Tajkaremi, A. Kaghani, and M. Yacoub. Early detection of lung endothelial dysfunction in man. In: Catravas JD, Callow AD, Ryan US, eds. Vascular endothelium: responses to injury New York: Plenum Press, 11-21, 1996. Celermajer, D.S., K.E. Sorensen, V.M. Gooch, D.J. Spiegelhalter, O.I. Miller, LD. Sullivan, J.K. Lloyd, and J.E. Deanfield. Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 340:1111-1115, 1992. 41 Celermajer, D.S., S. Cullen, and J.E. Deanfield. Impairment of endothelium-dependent pulmonary artery relaxation in children with congenital heart disease and abnormal pulmonary hemodyamics. Circulation 87:440-446, 1993. Chang, S-W., and N.F. Voelkel. Inflammatory mediator effects on pulmonary blood flow, edema, and the vascular endothelium. In: Bray MA, Anderson WH, eds. Mediators of Pulmonary Inflammation New York: Marcel Dekker, 403-453, 1991. Chirpaz-Oddou, M.F., A. Favre-Juvin, P. Flore, J . Eterradossi, M. Delaire, F. Grimbert, and A. Therminaris. Nitric oxide response in exhaled air during an incremental exhaustive exercise. J. Appl. Physiol. 82:1311-1318, 1997. Cines, D.B., E.S. Pollak, C A . Buck, J . Loscalzo, G.A. Zimmerman, R.P. McEver, J.S. Pober, T.M. Wick, B.A. Konkle, B.S. Schwartz, E.S. Barnathan, K.R. McCrae, B.A. Hug, A-M. Schmidt, and D.M. Stern DM. Endothelial cells in physiology and in the pathophysiology of vascular disorders. Blood 10:3527-3561,1998. Coates, G.H., H. O'Brodovich, A.L. Jerreries, and G.W. Gray. Effects of exercise on lung lymph flow in sheep and goats during normoxia and hypoxia. J. Clin. Invest. 74:133-141,1984. Cooper, T.K., S.R. Hopkins, K.D. Coutts, and D.C. McKenzie. Peripheral chemoresponsiveness and exercise-induced hypoxemia in highly trained athletes. Med. Sci. Sports Exerc. 26:S207, 1994 [Abstract]. Davies, P.F. Flow-mediated endothelial mechanotransduction. Physiological Reviews. 75:519-560, 1995. Davis, G.L., C P . Abildgaard, E.M. Bernauer, and M. Britton. Fibronolytic and hemostatic changes during and after maximal exercise in males. d. Appl. Physiol. 40:287-292, 1976. 42 Day, B., T. Oury, and J . Crapo. Interactions of superoxide dismutases with nitric oxide. In: Catravas JD, Callow AD, Ryan US, eds. Vascular endothelium: responses to injury New York: Plenum Press, 23-30, 1996. Dejana, E., and J-L. Plantier. Molecular organization of endothelial cell to cell junctions. In: Catravas JD, Callow AD, Ryan US, eds. Vascular endothelium: responses to injury New York: Plenum Press, 167-171, 1996. Dempsey, J.A., P. Hanson, D. Pegelow, A. Claremont, and J . Rankin. Limitations to exercise capacity and endurance: pulmonary system. Can. d. Appl. Sport Sci. 7:4-13, 1982. Dempsey, J.A., P.G. Hanson, and K.S. Henderson. Exercise-induced hypoxaemia in healthy human subjects at sea level, d. Physiol. 355:161-175, 1984. Dempsey, J.A. Is the lung built for exercise? Med. Sci. Sports Exerc. 18:143-155,1986. Dempsey, J.A., and P.D. Wagner. Exercise-induced arterial hypoxemia, d. Appl. Physiol. 87:1997-2006, 1999. DeRijk, R., D. Michelson, B. Karp, J . Petrides, E. Galliven, P. Deuster, G. Paciotti, P.W. Gold, and E.M. Sternberg. Exercise and circadian rhythm-induced variations in plasma Cortisol differentially regulate interleukin-1 p (IL-1 f3), IL-6, and tumor necrosis factor-a (TNF-a) production in humans: high sensitivity of TNF-a and resistance of IL-6. d. Clin. Endo. Metab. 1997;82:2182-2191, 1997. Dirnberger, E., H. Lucan, H.G. Eichler, J . Kastner, T. Pernerstorfer, and B. Jilma. Effects of nitroglycerin and sodium nitroprusside on endexpiratory concentrations of nitric oxide in healthy humans. Life Sciences 62:PL103-108, 1998. Donnelly, S.C., C. Haslett, I. Dransfield, C E . Robertson, D.C. Carter, J.A. Ross, I.S. Grant, and T.F. Tedder. Role of selectins in the development of adult respiratory distress syndrome. Lancet 344:215-219, 1994. 43 Doubler K.J. , J.D. Catravas, and C.N. Gillis. Early detection of oxygen-induced lung injury in conscious rabbits. Am. Rev. Resp. Dis. 126:534-539, 1982. Druml, W., H. Steltzer, W. Waldhausl, K. Lenz, A. Hammerle. H, Vierhapper, S. Gasic, and O.F. Wagner. Endothelin-1 in adult respiratory distress syndrome. Am. Rev. Resp. Dis. 148:1169-73, 1993. Duckies, S.P., D.N. Krause, and V.M. Miller. Effects of gonadal steroids on vascular function. J. Pharm. Exp. Ther. 279:1-3, 1996. Dufaux, B., and U. Order. Plasma elastase-a1 -antitrypsin, neopterin, tumor necrosis factor, and soluble interleukin-2 receptor after prolonged exercise. Int. J. Sports Med. 10:434-438, 1989. Dufaux, B., U. Order, and H. Liesen. Effect of a short maximal physical exercise on coagulation, fibrinolysis, and complement system. Int. d. Sports Med. 12 Suppl 1:S38-42, 1991. Durand, F., P. Mucci, L. Safont, and C. Prefaut. Effects of nitric oxide inhalation on pulmonary gas exchange during exercise in highly trained athletes. Acta Physiol. Scand. 165:169-176, 1999. Edwards, A.T., A.D. Blann, V.J. Suarez-Mendez, A.M. Lardi, and C.N. McCollum. Systemic responses in patients with intermittent claudication after treadmill exercise. Br. d. Surg. 8:1738-1741, 1994. Edwards, M.R., G.S. Hunte, A.S. Belzberg, A.W. Sheel, D.F. Worsley, and D.C. McKenzie. Blood-gas barrier permeability in athletes with exercise-induced hypoxemia. d. Appl. Physiol, (in press), 2000. Ehrhart, I.C., L.L. McCloud, S.E. Orfanos, J.D. Catravas, and W.F. Hofman. Effect of high blood flow on pulmonary vascular permeability to protein. J. Appl. Physiol. 76:2342-2347, 1994. 44 Ehrhart, I.e., S.E. Orfanos, L.L. McCloud, D.W. Sickles, W.F. Hofman, and J.D. Catravas. Vascular recruitment increases evidence of lung injury. Crit. Care Medr 27:120-129, 1999. Eldridge, M.E., A. Podolsky, R.S. Richardson, D.H. Johnson, D.R. Knight, E.C. Johnson, S.R. Hopkins, H. Michimata, B. Grassi, J . Feiner, S.S. Kurdak, P.E. Bickler, P.D. Wagner, and J.W. Severinghaus. Pulmonary hemodynamic response to exercise in subjects with prior high-altitude pulmonary edema. J. Appl. Physiol. 81:911-921,1996. Eldridge, M.W., R.K. Braun, K.Y. Yoneda, W.F. Walby, S. Bennett, and D.M. Hyde. Lung injury after heavy exercise at altitude. Chest 114:S66-S67, 1998. Elliot. A.R., F. Zhenxing, K. Tsukimoto, R. Prediletto, O. Mathieu-Costello, and J.B. West. Short-term reversibility of ultrastructural changes in pulmonary capillaries caused by stress failure. J. Appl. Physiol. 73:1150-1158, 1992. English, J.L. , L.O. Jacobs, G. Green, and T.C. Andrews. Effect of the menstrual cycle on the endothelium-dependent vasodilation of the brachial artery in normal young women. Am. J. Cardiol. 82:256-258, 1998. Epstein, S.E., G.D. Beiser, M. Stampfer, B.F. Robinson, and E. Braunwald. Characterization of the circulatory response to maximal exercise in normal subjects and patients with heart disease. Circulation 35:1049-1062, 1967. Espersen, G.T., A. Elbaek, E. Ernst, E. Toft, S. Kaalund, C. Jersild, and N. Grunnet. Effect of physical exercise on cytokines and lymphocyte subpopulations in human peripheral blood. APMIS 98:395-400, 1990. Fantone, J.C., and P.A. Ward. Oxygen-derived radicals and their metabolites: relationship to tissue injury. In: Current Concepts Kalamazoo, Michigan: Upjohn Company, 1985. 4 5 Fairbarn, M.S., K.D. Coutts, R.L. Pardy, and D.C. McKenzie. Improved respiratory muscle endurance of highly trained cyclists and the effects on maximal exercise performance. Int. J. Sports Med. 12:66-70,1991. Farhat, M.Y., M-F. Chen, T. Bhatti, A. Iqbal, S. Cathapermal, and P.W. Ramwell. Protection by oestradiol against the development of cardiovascular changes associated with monocrotaline pulmonary hypertension in rats. Br. d. Pharmacol. 110:719-723, 1993. Flick, M.R., A. Perel, and N.C. Staub. Leukocytes are required for increased lung microvascular permeability after microembolization in sheep. Circ. Res. 48:344-351, 1981. Freund, B.J., E.V. Shizuru, G.M. Hashiro, and J.R. Claybaugh. Hormonal, electrolyte, and renal responses to exercise are intensity dependent. J. Appl. Physiol. 70:900-906, 1991. Fung, Y.C., and S.S. Sobin. Pulmonary alveolar blood flow, in Engineering aspects of the lung, J.B. West, ed., Marcel Dekker, New York, 267-359, 1977. Gale, G.E., J.R. Torre-Bueno, R.E. Moon, H.A. Saltzman, and P.D. Wagner. Ventilation-perfusion inequality in normal humans during exercise at sea level and simulated altitude. J. Appl. Physiol. 58:978-988, 1985. Gallagher, C.G., W. Huda, M. Rigby, D. Greenberg, and M. Younes. Lack of radiographic evidence of interstitial pulmonary edema after maximal exercise in normal subjects. Am. Rev. Resp. Dis. 137:474-476, 1988. Gavin T.P., A.P. Derchak, and J.M. Stager. Ventilation's role in the decline in V02max and Sa02 in acute hypoxic exercise. Med. Sci. Sports. Exerc. 30:195-199, 1998. Gavin, T.P., and J.M. Stager. The effect of exercise modality on exercise-induced hypoxemia. Respir. Physiol. 115:317-323,1999. 46 Gehr, P., M. Bachofen, and E.R. Weibel. The normal human lung: ultrastructure and morphometric estimation of diffusion capacity. Respir. Physiol. 32:121-140, 1978. Geiger, M., A. Stone, S.N. Mason, K.T. Oldham, and K.S. Guice. Differential nitric oxide production by microvascular and macrovascular endothelial cells. Am. J. Physiol. 273 (Lung Cell. Mol. Physiol. 17j:L275-L281, 1997. Giesbrecht, G.G., F. Ali, and M. Younes. Short-term effect of tidal pleural pressure swings on pulmonary blood flow during rest and exercise. J. Appl. Physiol. 71:465-73, 1991. Giaid, A., and D. Saleh. Reduced expression of endothelial nitric oxide synthase in the lungs of patients with pulmonary hypertension. N. Engl. J. Med. 333:214-221, 1995. Gimbrone, M.A., Jr. Biomechanical activation: an emerging paradigm in endothelial adhesion biology, d. Clin. Invest. 100Suppl:61S-65S, 1997. Gledhill, N., A.B. Froese, and J.A. Dempsey. Ventilation to perfusion distribution during exercise in health. In: Muscular Exercise and the Lung, J. Dempsey and C E . Reed, eds., Madison, Wl: Univ. of Wisconsin Press, 325-343, 1977. Glenny, R.W., and H.T. Robertson. Fractal properties of pulmonary blood flow: characterization of spatial heterogeneity, d. Appl. Physiol. 69:532-545, 1990. Glenny, R.W., W.J. Lamm, R.K. Albert, and H.T. Robertson. Gravity is a minor determinant of pulmonary blood flow distribution, d. Appl. Physiol. 71:620-629, 1991. Glenny, R.W., and H.T. Robertson. Fractal modeling of pulmonary blood flow heterogeneity, d. Appl. Physiol. 70:1024-1030,1991. Glenny, R.W., S. McKinney, H.T. Robertson. The spatial pattern of pulmonary blood flow distribution is stable over days. d. Appl. Physiol. 82:902-907, 1997. 47 Glenny, R.W., S. Bernard, H.T. Robertson, and M.P. Hlastala. Gravity is an important but minor determinant of pulmonary blood flow heterogeneity in the baboon. FASEB J. 11:A344, 1997 [Abstract]. Glenny, R.W. Blood flow distribution in the lung. Chest 114:8S-16S, 1998. Gluecker, T., P. Capasso, P. Schnyder, F. Gudinchet, M.D. Schaller, J.P. Revelley, R. Chiolero, P. Vock, and S. Wicky. Clinical and radiologic features of pulmonary edema. Radiographics 19:1507-1531, 1999. Gordon, J.B., R.C. Wetzel, M.L. McGeady, N.F. Adkinson, Jr., and J.T. Sylvester. Effects of indomethacin on estradiol-induced attenuation of hypoxic vasocontriction in lamb lungs. J. Appl. Physiol. 61:2116-2121, 1986. Gore, C.J. , A .G. Hahn, G.C. Scroop, D.B. Watson, K.I. Norton, R.J. Wood, D.P. Campbell, and D.L. Emonson. Increased arterial desaturation in trained cyclists during maximal exercise at 580 m altitude. J. Appl. Physiol. 80:2204-2210, 1996. Gralnick, H.R., L.P McKeown, O.M. Wilson, S.B. Williams, and R.J. Elin. von Willebrand factor release induced by endotoxin. J. Lab. Clin. Med. 113:118-122, 1989. Grissom, C.K., G.A. Zimmerman, and R.E. Whatley. Endothelial selectins in acute mountain sickness and high-altitude pulmonary edema. Chest 112:1572-1578, 1997. Groves, B.M., J.T. Reeves, J.R. Sutton, P.D. Wagner, A. Cymerman, M. Malconian, P.B. Rock, P.M. Young, and C.S. Houston. Operation Everest II: elevated high-altitude pulmonary resistance unresponsive to oxygen. J. Appl. Physiol. 63:521-530, 1987. Hall, S.M., and S.G. Haworth. Onset and evolution of pulmonary vascular disease in young children: abnormal postnatal remodeling studied in lung biopsies. J. Pathol. 166:183-193, 1992. 48 Hamilton, K.K., and P.J. Sims. Changes in cytosolic C a 2 + associated with von Willebrand factor release in human endothelial cells exposed to histamine. J. Clin. Invest. 79:600-608, 1987. Hammond, M.D., G.E. Gale, K.S. Kapitan, A. Ries, and P.D. Wagner. Pulmonary gas exchange in humans during exercise at sea level. J. Appl. Physiol. 60:1590-1598, 1986. Hanel, B., I. Teunissen, A. Rabol, J . Warberg, and N.H. Secher. Restricted post-exeercise pulmonary diffusing capacity and central blood volume depletion. J. Appl. Physiol. 83:11-17, 1997. Hanaoka, M., K. Kubo, Y. Yamazaki, T. Miyahara, Y. Matsuzawa, T. Kobayashi, M. Sekiguchi, M. Ota, and H. Watanabe. Association of high-altitude pulmonary edema with the major histocompatibility complex. Circulation 97-1124-1128, 1998. Hanaoka, M., M. Tanaka, R-L. Ge, Y. Droma, A. Ito, T. Miyahara, T. Koizumi, K. Fujimoto, T. Fujii, T. Kobayashi, and K. Kubo. Hypoxia-induced pulmonary blood redistribution in subjects with a history of high-altitude pulmonary edema. Circulation 101:1418-1422, 2000. Harms, C.A., and J.M. Stager. Low chemoresponsiveness and inadequate hyperventilation contribute to exercise-induced hypoxemia. J. Appl. Physiol. 79:575-580, 1995. Harms, C M . , M.A. Babcock, S.R. McClaran, D.F. Pegelow, G.A. Nickle, W.B. Nelson, and J.A. Dempsey. Respiratory muscle work compromises leg blood flow during maximal exercise. J. Appl. Physiol. 82:1573-1583, 1997. Harms, C.A., S.R. McClaran, G.A. Nickele, D.F. Pegelow, W.B. Nelson, and J.A. Dempsey. Exercise-induced arterial hypoxaemia in healthy young women. J. Physiol. 507:619-628, 1998a. 49 Harms, C.A., T.J. Wetter, S.R. McClaran, D.F. Pegelow, G.A. Nickele, W.B. Nelson, P. Hanson, and J.A. Dempsey. Effect of respiratory muscle work on cardiac output and its distribution during maximal exercise. J. Appl. Physiol. 85:609-618, 1998b. Harris, P., D. Heath, and A. Apostolopoulos. The extensibility of the human pulmonary trunk. Br. Heart J. 27:651-659,1965. Harrop, G.A. The oxygen and carbon dioxide content of arterial blood in normal individuals and in patients with anemia and heart disease. J. Exp. Med. 30:241-257, 1919. Haskard, D.O. Proinflammatory adhesins. In: Catravas JD, Callow AD, Ryan US, eds. Vascular endothelium: responses to injury New York: Plenum Press, 145-154, 1996. Hayashi, T., K. Yamada, T. Esaki, M. Kuzuya, S. Satake, T. Ishikawa, H. Hidaka, and A. Iguchi. Estrogen increases endothelial nitric oxide by a receptor-mediated system. Biochem. Biophys. Res. Comm. 214:847-855, 1995. Hinder, F., M. Brooke, L.D. Traber, and D.L. Traber. Nitric oxide and endothelial permeability, d. Appl. Physiol. 83:1941-1946,1997. Hlastala, M.P., S.L. Bernard, and H.H. Erickson, Pulmonary blood flow distribution in standing horses is not dominated by gravity, d. Appl. Physiol. 81:1051-1061, 1996. Hogg, J.C. , B.A. Martin, S. Lee, and T. McLean. Regional differences in erythrocyte transit in normal lungs. d. Appl. Physiol. 59:1266-1271, 1985. Holmgren, A., and H. Linderholm. Oxygen and carbon dioxide tensions of arterial blood during heavy and exhaustive exercise. Acta Physiol. Scand. 44:203-215, 1958. Hopkins, S.R., and D.C. McKenzie. Hypoxic ventilatory response and arterial desaturation during heavy work. d. Appl. Physiol. 67:1119-1124, 1989. 1 50 Hopkins, S.R., D.C. McKenzie, R.B. Schoene, R.W. Glenny, and H.T. Robertson. Pulmonary gas exchange during exercise in athletes I. Ventilation-perfusion mismatch and diffusion limitation. J. Appl. Physiol. 77:912-917, 1994. Hopkins, S.R., A.S. Belzberg, B.R. Wiggs, and D.C. McKenzie. Pulmonary transit time and diffusion limitation during heavy exercise in athletes. Resp. Physiol. 103:67-73, 1996. Hopkins, S.R., R.B. Schoene, W.R. Henderson, R.G. Spragg, T.R. Martin, and J.B. West. Intense exercise impairs the integrity of the pulmonary blood-gas barrier in elite athletes. Am. J. Respir. Crit. Care. Med. 155:1090-1094, 1997. Hopkins, S.R., T.P. Gavin, N.M. Siafakas, L.J. Haseler, M. Olfert, H. Wagner, and P.D. Wagner. Effect of prolonged, heavy exercise on pulmonary gas exchange in athletes. J. Appl. Physiol. 85:1523-1532, 1998. Hopkins, S.R., R.C. Barker, T.D. Brutsaert, T.P. Gavin, P. Entin, I.M. Olfert, S. Veisel, and P.D. Wagner. Pulmonary gas exchange during exercise in women: effects of exercise type and work increment. J. Appl. Physiol. 89:721-730, 2000. Howell, J.B.L., S. Permutt, D.F. Proctor, and R.L. Riley. Effect of inflation of the lung on different parts of the pulmonary vascular bed. d. Appl. Physiol. 16:71-76, 1961. Hultgren, H.N. High altitude pulmonary edema. Adv. Cardiol. 5:24- ,1970. Hultgren, H.N., R.F. Grover, and L.H. Hartley. Abnormal circulatory repsonses to high altitude in subjects with previous history of high altitude pulmonary edema. Circulation 54:759-770, 1971. Huupponen, M.R.H., L.H. Makinen, P.M. Hyvonen, C.K. Sen, T. Rankinen, S. Vaisanen, and R. Rauramaa. The affect of ^-acetylcysteine on exercise-induced priming of neutrophils: a chemiluminescence study. Int. J. Sports Med. 16:399-403, 1995. Iwamoto, J . , D.R. Pendergast, H. Susuki, and J . Krasney. Effect of graded exercise on nitric oxide in expired air in humans. Resp. Physiol. 97:333-345, 1994. 51 Janosi, A., P. Apor, J. Hankoczy, and A. Kadar. Pulmonary artery pressure and oxygen consumption measurement during supine bicycle exercise. Chest 93:419-421, 1988. Jilma, B., H-G Eichler, H. Breiteneder, M. Wolzt, M. Aringer, W. Graninger, C. Rohrer, M. Veitl, and O.F. Wagner. Effects of 17p-estradiol on circulating adhesion molecules. J. Clin. Endo. Metab. 79:1619-1624, 1994. Jilma, B., J. Hildebrandt, S. Kapiotis, O.F. Wagner, E. Kitzweger, C. Mullner, B. Monitzer, K. Krejcy, and H-G. Eichler. Effects of estradiol on circulating P-selectin. d. Clin. Endocrinol. Metab. 81:2350-2355, 1996a. Jilma, B., J. Kastner, C. Mensik, B. Vondrovec, J. Hildebrandt, K. Krejcy, O.F. Wagner, and H-G. Eichler. Sex differences in concentrations of exhaled nitric oxide and plasma nitrate. Life Sciences 58:469-476, 1996b. Jilma, B., H-G. Eichler, P. Stohlawetz, E. Dirnberger, S. Kapiotis, O.F. Wagner, W. Schutz, and K. Krejcy. Effects of exercise on circulating vascular adhesion molecules in healthy men. Immunobiol. 197:505-512, 1997a. Jilma, B., Dirnberger, E., H-G. Eichler, B. Matulla, L. Schmetterer, S. Kapiotis, W. Speiser, and O.F. Wagner. Partial blockade of nitric oxide synthase blunts the exercise-induced increase of von Willebrand factor antigen and of factor Vlll in man. Thromb. Haemost. 78:1268-1271, 1997b. Jilma, B., T. Pernerstorfer, E. Dirnberger, P. Stohlawetz, L. Schmetterer, E.A. Singer, U. Grasseli, H-G. Eichler, and S. Kapiotis. Effects of histamine and nitric oxide synthase inhibition on plasma levels of von Willebrand factor antigen. d. Lab. Clin. Med. 131:151-156, 1998. Johnson, B.D., K.W. Saupe, and J.A. Dempsey. Mechanical constraints on exercise hyperpnea in endurance athletes, d. Appl. Physiol. 73:874-886, 1992. 52 Jungersten, L., A. Ambring, B. Wall, and A. Wennmalm. Both physical fitness and acute exercise regulate nitric oxide formation in healthy humans. J. Appl. Physiol. 82:760-764, 1997. Kamiya, A., and T. Togawa. Adaptive regulation of wall shear stress to flow change in the canine carotid artery. Am. J. Physiol. 239 (Heart Circ. Physiol. 8):H14-H21, 1980. Kappel, M., N. Tvede, H. Galbo, P.M. Haahr, M. Kjaer, M. Linstow, K. Klarlund, and B.K. Pedersen. Evidence that the effect of physical exercise on NK cell activity is mediated by epinephrine. J. Appl. Physiol. 70:2530-2534,1991. Kappel, M., M.B. Hansen, M. Diamant, J.O. Jorgensen, A. Gyhrs, and B.K. Pedersen. Effects of acute bolus growth hormone infusion on the human immune system. Hormon. Metab. Res . 25:579-585, 1993. Kayal, S., J-P. Jais, N. Aguini, J . Chaudiere, and J . Labrousse. Elevated circulating E-selectin, intracellular adhesion molecule 1, and von Willebrand factor in patients with severe infection. Am. J. Respir. Crit. Care Med. 157:776-784, 1998. Kelly, J .J. , T.M. Moore, P. Babal, A.H. Diwan, T. Stevens, and W.J. Thompson. Pulmonary microvascular and macrovascular endothelial cells: differential regulation of C a 2 + and permeability. Am. d. Physiol. 274(Lung Cell. Mol. Physiol. 18):L810-L819, 1998. Ker, J.A., and C M . Shultz. Respiratory muscle fatigue after an ultra-marathon measured as inspiratory task failure. Int. d. Sports Med. 17:493-496, 1996. Ketai, L.H., and J.D. Godwin. A new view of pulmonary edema and acute respiratory distress syndrome. J. Thoracic Imaging 13:147-171, 1998. Kjaer, M. Epinephrine and some other hormonal responses to exercise in man: with special reference to physical training. Int. d. Sports Med. 10:2-15, 1989. 53 Klein, J.P., H.V. Forster, R.D. Stewart, and A. Wu. Hemoglobin affinity for oxygen during short-term exhaustive exercise. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 48:236-242, 1980. Koizumi, T., R. Gupta, M. Banerjee, and J.H. Newman. Changes in pulmonary vascular tone during exercise. J. Clin. Invest. 94:2275-2282,1994. Koskolou M.D., and D.C. McKenzie. Arterial hypoxemia and performance during intense exercise. Eur. J. Appl. Physiol. 68:80-86, 1994. Kubes, P., M. Suzuki, and D.N. Granger. Nitric oxide: an endogenous modulator of leukocyte adhesion. Proc. Natl. Acad. Sci. USA 88:4651-4655, 1991. Kubo, K., M. Hanaoka, T. Hayano, T. Miyahara, T. Hachiya, M. Hayasaka, T. Koizumi, K. Fujimoto, T. Kobayashi, and T. Honda. Inflammatory cytokines in BAL fluid and pulmonary hemodynamics in high-altitude pulmonary edema. Respir. Phsyiol. 111:301-310, 1998. Kuchan M.J., and J.A. Frangos. Shear stress regulates endothelin-1 release via protein kinase C and cGMP in cultured endothelial cells. Am. d. Physiol. 64(Heart Circ. Physiol. 33):H150-H156, 1994. Kuebler, W.M., X. Ying, B. Singh, A. Issekutz, and J. Bhattacharya. Pressure is proinflammatory in lung venular capillaries, d. Clin. Invest. 104:494-502, 1999. Kuhnle G.E.H., W.M. Kuebler, J . Groh, and A.E. Goetz. Effect of blood flow on the leukocyte-endothelium interaction in pulmonary microvessels. Am. d. Respir. Crit. Care Med. 152:1221-1224, 1995. Kuhns, D.B., W.G. Alvord, and J.I. Gallin. Increased circulating cytokines, cytokine antagonists, and E-selectin after intravenous administration of endotoxin in humans. J . Infect. Dis. 171145-152,1995. 54 Kuikka, J.T., and E. Lansimies. A fractal approach for evaluation of pulmonary circulation in man at rest and during exercise. Clin. Physiol. 19:107-110, 1999. Kurdak, S.S., Y. Namba, Z. Fu, B. Kennedy, O. Mathieu-Costello, and J.B. West. Effect of increased duration of high perfusion pressure on stress failure of pulmonary capillaries. Micro. Res. 50:235-248, 1995. Lama, I.L., L.A. Wolski, K.D. Coutts, and D.C. McKenzie. The relationship between test protocol and the development of exercise-induced hypoxemia (EIH) in highly trained athletes. Eur. J. Appl. Physiol. 74:330-335, 1996. Lamping, K.G., and D.W. Nuno. Effects of 17 beta-estradiol on coronary microvascular responses to endothelin-1. Am. J. Physiol. 271:H1117-H1124, 1996. Lawler, J . , S.K. Powers, and D. Thompson. Linear relationship between V02max and V02max decrement during exposure to acute hypoxia. J. Appl. Physiol. 64:1486-1492,1989. Leith, D.E., and M. Bradley. Ventilatory muscle strength and endurance training. J. Appl. Physiol. 41:508-516, 1976. Lien, D . C , G.S. Worthen, R.L. Capen, W.L. Hanson, L.L. Checkley, S.J. Janke, P.M. Henson, and W.W. Wagner, Jr. Neutrophil kinetics in the pulmonary microcirculation. Effects of pressure and flow in the dependent lung. Am. Rev. Resp. Dis. 141:953-959, 1990. Lip, G.Y.H., A.D. Blann, J . Zarifis, M. Beevers, P-L. Lip, and D.G. Beevers. Soluble adhesion molecule P-selectin and endothelial dysfunction in essential hypertension: implications for atherogenesis? A preliminary report. J. Hypertension 13:1674-1678, 1995. 55 Loesberg, C , M.D. Gonsalves, J . Zandbergen, C. Willems, W.G. van Aken, H.V. Stel, J.A. van Mourik, and P.G. de Groot. The effects of calcium on the secretion of factor-VIM-related antigen by cultured human endothelial cells. Biochem. Biophys. Acta 763:160-168, 1983. Loke, J . , D.A. Mahler, and J.A. Virgulto. Respiratory muscle fatigue after marathon running. J. Appl. Physiol. 52:821-824, 1982. Lorant, D.E., K.D. Patel, T.M. Mclntyre, R.P. McEver, S.M. Prescott, and G.A. Zimmerman. Co-expression of GP-140 and PAF by endothelium stimulated by histamine or thrombin: a juxtacrine system for adhesion and activation of neutrophils. J.Cell. Biol. 115:223-234, 1991. Lumb, A., and J.F. Nunn, eds. Nunn's applied respiratory physiology, 5 th Ed., Oxford: Butterworth-Heinemann, 2000. Lundberg, J.O.N., J . Rinder, E. Weitzberg, K. Alving, and J.M. Lundberg. Heavy physical exercise decreases nitric oxide levels in the nasal airways in humans. >Acfa Physiol. Scand. 159:51-57, 1997. Lynch, E.A., C A . Dinarello, and J .G. Cannon. Gender differences in IL-1a, IL-ip receptor antagonist secretion from mononuclear cells and urinary excretion. J. Immunol. 154:300-306, 1994. MacNee, W., B.A. Martin, B.R. Wiggs, A.S. Belzberg, and J . C Hogg. Regional pulmonary transit times in humans. d. Appl. Physiol. 66:844-850, 1989. MacNee, W., and C. Selby. Neutrophil traffic in the lungs: role of hemodynamics, cell adhesion, and deformability. Thorax 48:79-88, 1993. Manier, G., J . Moinard, P. Techoueyres, N. Varene, and H. Guenard. Pulmonary diffusion limitation after prolonged strenuous exercise. Resp. Physiol. 79:1219-1225, 1991. 56 Maroun, M., S. Mehta, R. Turcotte, M. Coslo, and S. Hussain. Effects of physical conditioning on endogenous nitric oxide output during exercise. d. Appl. Physiol. 79:121-125, 1995. Marshall, J .C. The gut as a potential trigger of exercise-induced inflammatory responses. Can. J. Physiol. Pharmacol. 76:479-484, 1998. Martin, B.A., J.V. Weil, K.E. Sparks, R.E. McCollough, and R.F, Grover. Exercise ventilation correlates positively with ventilatory chemoresponsiveness. J. Appl. Physiol. 45:557-564, 1978. Mathur, S., R.S. Mathur, J.M. Goust, H.O. Williamson, and H.H. Fudenberg. Cycle variations in white cell subpopulations in the human menstrual cycle: correlations with progesterone and estradiol. Clin. Immunonol. Immunopath. 13:246-253, 1979. Matsumoto, A., Y. Hirata, S-l. Momomura, H. Fujita, A. Yao, M. Sata, and T. Serizawa. Increased nitric oxide production during exercise. Lancet 343:849-850, 1994. McConnell, A.K., M.P. Caine, and G.R. Sharpe. Inspiratory muscle fatigue following running to volitional fatigue: the influence of baseline strength. Int. J. Sports Med. 18:169-173, 1997. McKechnie, J.K., W.P. Leary, T.D. Noakes, J.C. Kallmeyer, E.T.M. MacSearraigh, and L.R. Oliver. Acute pulmonary edema in two athletes during a 90-km running race. S. African Med. d. 56:261-265, 1979. McKenzie, D . C , J.R. Mayo, J . Potts, I. Lama, D.K. Jespersen, K.P. Whittall, and A.L. MacKay. Changes in lung water content and pulmonary diffusion capacity following intense exercise in well-trained athletes. Physiol. 39:A47, 1996. [abstract] McKenzie, D . C , J.R. Mayo, A.L. MacKay, K.P. Whittall, and T. O'Hare. Sustained heavy exercise and the development of pulmonary edema in trained male cyclists. Med. Sci. Sports Exerc. 31:S330, 1999. [abstract] 57 McKenzie, D.C., I.L. Lama, J .E. Potts, A.W. Sheel, and K.D. Coutts. The effect of repeat exercise on pulmonary diffusing capacity and EIH in trained athletes. Med. Sci. Sports Exerc. 31:99-104,1999. McLaran, S.R., C A . Harms, D.F. Pegelow, and J.A. Dempsey. Smaller lungs in women affect exercise hyperpnea. d. Appl. Physiol. 84:1872-1881,1998. McLaran, S.R., T.J. Wetter, D.F. Pegelow, and J.A. Dempsey. Role of expiratory flow limitation in determining lung volumes and ventilation during exercise. d. Appl. Physiol. 86:1357-1366, 1999. Melsom, M.N., T. Flatebo, and G. Nicolaysen. No apparent effect of nitirc oxide on the local matching of pulmonary perfusion and ventilation in awake sheep. Acta Physiol. Scand. 168:361-370, 2000. Miller, V.M., L.L. Aarhus, and P.M. VanHoutte. Modulation of endothelium-dependent responses by chronic alterations of blood flow. Am. d. Physiol. 251 {Heart Circ. Physiol. 20):H520-H527, 1986. Miyachi, M., and H. Yano. Thinning of the capillary endothelium with physical activity shortens the diffusion distance of the pulmonary air-blood barrier. Acta Physiol. Scand. 157:513-514, 1996. Mochel, M, N-V Ulrich, L. Rocker, A. Ruf, F. Klefisch, H. Patscheke, H. Eichstadt, T. Stork, and U. Frei. Exhaustive cycle exercise induces P-selectin expression, coagulation, and fibrinolysis activation in ultraendurance athletes. Thrombosis Res. 94:263-269, 1999. Moncada, S., R. Grtglewsky, S. Bunting, and J.R. Vane. An enzyme isolated from arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature Lond. 263:663-665,1976. Moncada, S., R.M.J. Palmer, and E.A. Higgs. Nitric oxide: physiology, physiopathology, and pharmacology. Pharmacol. Rev. 43:109-142,1991. 58 Moore, G.E., M.E. Holbein, and J.P. Knochel. Exercise-associated collapse in cyclists is unrelated to endotoxemia. Med. Sci. Sports Exerc. 27:1238-1242,1995. Morgan, D., W. Kohrt, B. Bates, and J . Skinner. Effects of respiratory muscle endurance training on ventilatory and endurance performance of moderately trained cyclists. Int. d. Sports Med. 8:88-93,1987. Moss, M., L. Ackerson, M.K. Gillespie, F.A. Moore, E.E. Moore, and P.E. Parsons. von Willebrand factor antigen levels are not predictive for the adult respiratory distress syndrome. Am. d. Respir. Crit. Care Med. 151:15-20, 1995. Moss, M., M.K. Gillespie, L. Ackerson, F.A. Moore, E.E. Moore, and P.E. Parsons. Endothelial cell activity varies in patients at risk for the adult respiratory distress syndrome. Crit. Care Med. 24:1782-1786,1996. Mota, S., P. Casan, F. Drobnic, J . Giner, O. Ruiz O, J . Sanchis, and J . Milic-Emili. Expiratory flow limitation during exercise in competition cyclists. J. Appl. Physiol. 86:611-616, 1998. Moyna, N.M., G.R. Acker, J.R. Fulton, F.L. Goss, R.J. Robertson, D.J. Tollerud, and B.S. Rabin. Lymphocyte function and cytokine production during incremental exercise in active and sedentary males and females. Int. d. Sports Med. 17:585-591, 1996. Muir, A.L., M. Cruz, B.A. Martin, H. Thommasen, A. Belzberg, and J.C. Hogg. Leukocyte kinetics in the human lung: role of exercise and catecholamines, d. Appl. Physiol. 57:711-719, 1984. Mulligan, M.S., M.J. Polley, R.J. Bayer, M.F. Nunn, J .C. Paulson, and P.A. Ward. Neutrophil-dependent acute lung injury. Requirement for P-selectin (GMP-140). J. Clin. Invest. 90:1600-1607, 1992. Mulligan, M.S., C W . Smith, D . C Anderson, R.F. Todd, M. Miyasaka, T. Tamatani, T.B. Issekutz, and P.A. Ward. Role of leukocyte adhesion molecules in complement-induced lung injury, d. Immunol. 150:2401-2406,1993. 59 Mulligan, M.S., E. Schmid, G.O. Till, T.E. Hugli, H.P. Friedl, R.A. Roth, and P.A. Ward. C5a-dependent up-regulation in vivo of lung vascular P-selectin. J. Immunol. 158:1857-1861, 1997. Mure, M., K.B. Domino, S.G. Lindahl, M.P. Hlastala, W.A. Altemeier, and R.W. Glenny. Regional ventilation-perfusion distribution is more uniform in the prone position. J. Appl. Physiol. 88:1076-1083,2000. NaDaud, S., M. Phillipe, J.F. Arnal, J.B. Michel, and F. Soubrier. Sustained increase in aortic endothelial nitric oxide synthase expression in vivo in a model of chronic high blood flow. Circ. Res. 79:857-863, 1996. Nailin, L , H. Wallen, and P. Hjemdahl. Evidence for prothrombotic effects of exercise and limited protection by aspirin. Circulation 100:1374-1379, 1999. Natelson, B.H., X. Zhou, J .E. Ottenweller, M.T. Bergen, S.A. Sisto, S. Drastal, W.N. Tapp, and W.L. Gause. Effect of acute exhausting exercise on cytokine gene expression in men. Int. J. Sports Med. 17:299-302, 1996. Nathens, A.R., R. Bitar, R.W.G. Watson, T.B. Issekutz, J.C. Marshall, A.P.B. Dackiw, and O.D. Rotstein. Thiol-mediated regulation of ICAM-1 expression in endotoxin-induced acute lung injury, d. Immunol. 160:2959-2966,1998. Newman, W., L.D. Beall, C W . Carson, G.G. Hunder, N. Graben, Z.I. Randhawa, T.V. Gopal, J . Wiener-Kronish, and M.A. Matthay. Soluble E-selectin is found in supernatants of activated endothelial cells and is elevated in the serum of patients with septic shock, d. Immunol. 150:644-654,1993. Nilius, B., F. Viana, and G. Droogmans. Ion channels in vascular endothelium. Ann. Rev. Physiol. 59:145-170, 1997. Northoff, H., and A. Berg. Immunologic mediators as parameters of the reaction to strenuous exercise. Int. d. Sports Med. 12:9-15, 1991. 60 Northoff, H., C. Weinstock, and A. Berg. The cytokine response to strenuous exercise. Int. J. Sports Med. 15:S167-S171, 1994. O'Brodovich, H., and G. Coates. Effect of isoproterenol or exercise on pulmonary lymph flow and hemodynamics. J. Appl. Physiol. 60:38-44, 1986. Oelz, O., M. Maggiorini, M. Ritter, U. Waber, R. Jenni, P. Vock, and P. Bartsch. Nifedipine for high altitude pulmonary edema. Lancet 2:1241-1244,1989. Paintal, A.S. Vagal sensory receptors and their reflex effects. Physiol. Rev. 53:159-227, 1973. Pedersen, P.K., H. Mandoe, K. Jensen, C. Andersen, and K. Madsen. Reduced arterial 02 saturation during supine exercise in highly trained cyclists. Acta Physiol. Scand. 158:323-331, 1996. Pedersen, B.K., H. Bruunsgaard, M. Klokker, M. Klappel, D.A. MacLean, H.B. Nielsen, T. Rohde, H. Ullum, and M. Zacho. Exercise-induced immunomodulation - possible roles of neuroendocrine and metabolic factors. Int. d. Sports Med. 18:S2-S7, 1997. Pelletier, N., N.E. Robinson, L. Kaiser, and F.J. Derksen. Regional differences in endothelial function in horse lung: possible role in blood flow distribution? d. Appl. Physiol. 85:537-542, 1998. Persson, M.G., N.P. Wiklund, and L.E. Gustafsson. Endogenous nitric oxide in single exhalations and the change during exercise. Am. Rev. Resp. Dis. 148:1210-1214, 1993. Peters, A.M., P. Allsop, A.W.J. Stuttle, R.N. Arnot, N. William, and G.M. Hall. Granulocyte margination in the human lung and its response to strenuous exercise. Clin. Sci. 82:237-244, 1992. 61 Pittet, J.F., D.R. Morel, A. Hemsen, K. Gunning, J.S. Lacroix, P.M. Suter, and J.M. Lundberg. Elevated plasma endothelin-1 concentrations are associated with the severity of illness in patients with sepsis. Ann. Surg. 213:261-4, 1991. Pittet, J.F., R.C. Mackersie, T.R. Martin, and M.A. Matthay. Biological markers of acute lung injury: prognostic and pathogenetic significance. Am. J. Respir. Crit. Care Med. 155:1187-1205, 1997. Podolsky, A. M.W. Eldridge, R.S. Richardson, D.R. Knight, E.C. Johnson, S.R. Hopkins, D.H. Johnson, H. Michimata, B. Grassi, J . Feiner, S.S. Kurdak, P.E. Bickler, J.W. Severinghaus, and P.D. Wagner. Exercise-induced V A / Q inequality in subjects with prior high-altitude pulmonary edema, d. Appl. Physiol. 81:922-932, 1996. Pohlman, T.H., K.A. Stanness, P.G. Beatty, H.D. Ochs, and J.M. Harlan. An endothelial cell surface factor induced in vitro by lipopolysaccharide, interleukin-1 and tumor necrosis factor-a increases neutrophil adherence by a CDW 18-dependent mechanism, d. Immunol. 136:4548-4553, 1986. Powers, S., J . Lawler, S. Dodd, M. Kirtley, and G. Landry. Incidence of exercise-induced hypoxemia in elite athletes at sea level. Eur. d. Appl. Physiol. 58:298-302, 1988. Powers, S.K., J . Lawler, J.A. Dempsey, S. Dodd, and G. Landry. Effects of incomplete pulmonary gas exchange on V02max. d. Appl. Physiol. 66:2491-2495, 1989. Powers, S., D. Martin, M. Cicale, N. Collop, D. Huang, and D. Criswell. Exercise-induced hypoxemia in athletes: role of inadequate hyperventilation. Eur. d. Appl. Physiol. 65:37-42, 1992. Powers, S.K., D. Martin, and S. Dodd. Exercise-induced hypoxemia in elite endurance athletes. Incidence, causes and impact on V02max. Sports Medicine 16:14-22,1993. Prefaut, C , F. Anselme, C. Caillaud, and J . Masse-Biron. Exercise-induced hypoxemia in older athletes. J. Appl. Physiol. 76:120-126, 1994. 62 Prefaut, C , F. Anselme, and C. Caillaud. Inhibition of histamine release by nedocromil sodium reduces exercise-induced hypoxemia in masters athletes. Med. Sci. Sports Exerc. 29:10-16, 1997. Presson, R.G., Jr., C.C. Hanger, P.S. Godbey, J.A. Graham, T.C. Lloyd, Jr., and W.W. Wagner, Jr. Effect of increasing flow on distribution of pulmonary capillary transit times. d. Appl. Physiol. 76:1701 -1711, 1994. Prisk, G.K., H.J.B. Guy, A.R. Elliot, and J.B. West. Inhomogeneity of pulmonary perfusion during sustained microgravity on SLS-1. J. Appl. Physiol. 76:1730-1738, 1994. Rasmussen, B.D., B. Hanel, K. Saunamki, and N.H. Secher. Recovery of pulmonary diffusion capacity after maximal exercise, d. Sport Sciences 10:525-531, 1992. Reeves, J.T., R.E. Moon, R.F. Grover, and B.M. Groves. Increased wedge pressure facilitates decreased lung vascular resistance during upright exercise. Chest 93:97S-99S, 1988. Reeves, J.T., B.M. Groves, A. Cymerman, J.R. Sutton, P.D. Wagner, D. Turkevich, and C.S. Houston. Operation Everest II: cardiac filling pressures during cycle exercise at sea level. Resp. Physiol. 80:147-154,1990. Rice, A.J . , G.C. Scroop, C.G. Gore, A.T. Thornton, M. J . Chapman, H.W. Grevill, M.D. Holmes, and R. Scicchitano. Exercise-induced hypoxemia in highly trained cyclists at 40% peak oxygen uptake. Eur. d. Appl. Physiol. 79:353-359, 1999a. Rice, A.J . , A.T. Thornton, C.J. Gore, G.C. Scroop, H.W. Greville, H. Wagner, P.D. Wagner, and S.R. Hopkins. Pulmonary gas exchange during exercise in highly trained cyclists with arterial hypoxemia. J. Appl. Physiol. 87:1802-1812, 1999b. Rice A.J . , G.C. Scroop, A.T. Thornton, N.S. McNaughton, K.J. Rogers, M.J. Chapman, H.W. Greville, R. Scicchitano, and C.G. Gore. Arterial hypoxemia in endurance athletes is greater during running than cycling. Respir. Physiol, (in press), 2000. 63 Ridker. P.M., C H . Hennekens, B. Roitman-Johnson, M.J. Stampfer, and J . Allen. Plasma concentration of soluble intracellular adhesion molecule 1 and risks of future myocardial infarction in apparently healthy men. Lancet 351:88-92, 1998. Riley, R.L., J.L. Lilienthal, D.D. Proemmel, and R.E. Franke. On the determination of the physiologically effective pressures of oxygen and carbon dioxide in alveolar air. Am. J. Physiol. 147:191-198, 1946. Riley, M.S., J . Porszasz, J . Miranda, M.P.K.J. Engelen, B. Brundage, and K. Wasserman K. Exhaled nitric oxide during exercise in primary pulmonary hypertension and pulmonary fibrosis. Chest 111:44-50, 1997. Rivier, A., J . Pene, P. Chanez, F. Anselme, C. Caillaud, C. Prefaut, P. Godard, and J . Bousquet. Release of cytokines by blood monocytes during strenuous exercise. Int. J. Sports Med. 15:192-198, 1994. Roach, J.M., S.R. Muza, P.B. Rock, T.P. Lyons, C M . Lilly, J.M. Drazen, and A. Cymerman. Urinary leukotriene E4 levels increase upon exposure to hypobaric hypoxia. Chest 110:946-951, 1996. Roger, N., J.A. Barbera, J . Roca, I. Rovira, F.P. Gomez, and R. Rodriguez-Roisin. Nitric oxide inhalation during exercise in chronic obstructive pulmonary disease. Am. d. Respir. Crit. Care Med. 156:800-806, 1997. Roughton, F.J.W., and R.E. Forster. Relative importance of diffusion and chemical reaction rates in determining rate of exchange of gases in the human lung, with special reference to true diffusing capacity of pulmonary membrane and volume of blood in the lung capillaries. d. Appl. Physiol. 11:290-302, 1957. Rowell, L., H. Taylor, V. Wang, and Y. Carlson. Saturation of arterial blood with oxygen during maximal exercise, d. Appl. Physiol. 19:284-286, 1964. 64 Rubin, D.B., J.P. Weiner-Kronish, J.F. Murray, D.R. Green, J . Turner, J.M. Luce, A.B. Montgomery, J.D. Marks, and M.A. Matthay. Elevated von Willebrand factor andtigen is an early plasma predictor of acute lung injury in nonpulmonary sepsis syndrome. J. Clin. Invest. 86:474-480, 1990. Rumpf, K.W., A. Brat, V. Armstrong, and F. Scheler. Increased angiotensin converting enzyme in end-stage renal disease. Nephron 40:248-249,1985. Sadler, T.W. Langman's Medical Embryology, 5th Ed., Baltimore: Williams & Wilkins, Chps. 12 and 13, 1985. St. Croix, C M . , C A . Harms, S.M. McClaran, G.A. Nickle, D.F. Pegelow, W.B. Nelson and J.A. Dempsey. Effect of prior exercise on exercise-induced hypoxemia in young women, d. Appl. Physiol. 85:1556-1563,1998. St. Croix, C M . , T.J. Wetter, D.F. Pegelow, K.C. Meyer, and J.A. Dempsey. Assessment of nitric oxide formation during exercise. Am. d. Respir. Crit. Care Med. 159 1125-1133, 1999. Schaffartzik, W., D.C. Poole, T. Derion, K. Tsukimoto, M.C. Hogan, J.P. Arcos, D.E. Bebout, and P.D. Wagner. V A / Q distribution during heavy exercise and recovery in humans: implications for pulmonary edema, d. Appl. Physiol. 72:1657-1667, 1992. Scherrer, U., L. Vollenweider, A. Delbays, M. Savcic, U. Eichenberger, G-R. Kleger, A. Fikrle, P.E. Ballmer, P. Nicod, and P. Bartsch. Inhaled nitric oxide for high-altitude pulmonary edema. N. Eng. d. Med. 334:624-629, 1996. Scnitzner, J.E. , A. Sifflinger-Birnboim, P.J. Del Vecchio, and A.B. Malik. Segmental differentiation of permeability, protein glycosylation, and morphology of cultured bovine lung vascular endothelium. Biochem. Biophys. Res. Comm. 199:11-19, 1994. Schwartz, J . , S.A. Katz, R.W. Fegley, and M.S. Tockman. Analysis of spirometric data from a national sample of healthy 6- to 24-year-olds (NHANES II). Am. Rev. Resp. Dis. 138:1405-1414, 1988. 65 Schwartz, J . , S.A. Katz, R.W. Fegley, and M.S. Tockman. Sex and race differences in the development of lung function. Am. Rev. Resp. Dis. 138:1415-1421, 1988. Seekamp, A., J.S. Warren, D.G Remick, G.O. Till, and P.A. Ward. Requirements for tumor necrosis factor-a and interleukin-1 in limb ischemia/reperfusion injury and associated lung injury. Am. J. Pathol. 143:453-463, 1993. Seekamp, A., G.O. Till, J .C. Paulson, M.S. Mulligan, D.C. Anderson, M. Miyasaka, and P.A. Ward. Role of selectins in local and remote tissue injury following ischemia and reperfusion. Am. J. Pathol. 144:592-598, 1994. Shek, P.N., and R.J. Shephard. Physical exercise as a human model of limited inflammatory response. Can. J. Physiol. Pharmacol. 76:589-597, 1998. Sheridan, B.C., R.C. Mclntyre, D.R. Meldrum, and D.A. Fullerton. Phosphodiesterase inhibition overcomes pulmonary vasomotor dysfunction in acute lung injury. Surg. Res. 71:145-149, 1997. Sheridan, B.C., R.C. Mclntyre, D.R. Meldrum, and D.A. Fullerton. L-arginine attenuates endothelial dysfunction in endotoxin-induced lung injury. Surgery 125:33-40, 1999. Sheel, A.W., K.D. Coutts, J .E. Potts, and D.C. McKenzie. The time course of pulmonary diffusiomg capacity for carbon monoxide following short duration high intensity exercise. Resp. Physiol. 111:271-281,1998. Sheel, A.W., M.R. Edwards, and D.C. McKenzie. Relationship between decreased oxyhemoglobin saturation and exhaled nitric oxide during exercise. >Acfa Physiol. Scand. 169:149-156,2000. Sheel, A.W., M.R. Edwards, G.S. Hunte, and D.C. McKenzie. Influence of inhaled nitric oxide on gas exchange during normoxic and hypoxic exercise in highyl trained cyclists. J. Appl. Physiol, (in press), 2000. 66 Singhal, S., R. Henderson, K. Horsfield, K. Harding, and G. Cumming. Morphometry of the human pulmonary arterial tree. Circ. Res. 33:190-197,1973. Small, M., A.C. Tweddel, A.C. Rankin, G.D. Lowe, C R . Prentice, and C D . Forbes. Blood coagulation and platelet function following maximal exercise: effects of beta-adrenoceptor blockade. Haemostasis 14:262-268,1984. Smith, J.A., A.B. Gray, D.B. Pyne, M.S. Baker, R.D. Telford, and M.J. Weidemann. Submaximal exercise causes both priming and activation of neutrophils. Am. J. Physiol. 270:R838-R845, 1996. Smith, J.A. Exercise immunology and neutrophils. Int. J. Sports Med. 18:S46-S55, 1997. Smits, H.H., K. Grunberg, R.H. DeRijk, P.J. Sterk, and P.S. Hiemstra. Cytokine release and its modulation by dexamethasone in whole blood following exercise. Clin. Exp. Immunol. 111:463-468, 1998. Sprague, R.S., C. Thiemermann, and J.R. Vane. Endogenous endothelium-derived relaxing factor opposes hypoxic pulmonary vasoconstriction and supports blood flow to hypoxic alveoli in anesthetized rabbits. Proc. Natl. Acad. Sci. USA 89:8711-8715, 1992. Sprenger, H., C. Jacobs, M. Nain, A.M. Gressner, H. Prinz, W. Wesemann, and D. Gemsa. Enhanced release of cytokines, interleukin-2 receptors, and neopterin after long-distance running. Clin. Immunol. Immunopathol. 63:188-195,1992. Starling, E.H. On absorption of fluids from connective tissue spaces, d. Physiol. (London) 19:312-326, 1896. Staub, N . C , H. Nagano, and M.L. Pearce. Pulmonary edema in dogs, especially the sequence of fluid administration in the lungs, d. Appl. Physiol. 22:227-240, 1967. Staub, N.C. Steady state pulmonary transvascular water filtration in unanesthetized sheep. Circ. Res. 28:Suppl 1:135-139, 1971. 67 Stenmark, K.R., and R.P. Mecham. Cellular and molecular mechanisms of pulmonary vascular remodeling. Ann. Rev. Physiol. 59:89-144,1997. Stevens, T., B. Fouty, D. Cornfield, and D.M. Rodman. Reduced P 0 2 alters the behavior of fura-2 and indo-1 in bovine pulmonary artery endothelial cells. Cell Calcium 16:404-412, 1994. Strieter, R.M., N.W. Lukacs, T.J. Standiford, and S.L. Kunkel. Cytokines and lung inflammation: mechanisms of neutrophil recruitment to the lung. Thorax 48:765-769, 1993. Sutton, J.R., J.T Reeves, P.D. Wagner, B.M. Groves, A. Cymerman, M.K. Malconian, P.B. Rock, P.M. Young, S.D. Walter, and C S . Houston. Operation Everest II: oxygen transport during exercise at extreme simulated altitude. J. Appl. Physiol. 64:1309-1321, 1988. Tanaka, Y., D.P. Schuster, E . C Davis, G.A. Patterson, and M.D. Botney. The role of vascular injury and hemodynamics in rat pulmonary artery remodeling. J. Clin. Invest. 98:434-442, 1996. Thadani, U., and J . Parker. Hemodynamics at rest and during supine and sitting bicycle exercise in normal subjects. Am. J. Cardiol. 41:52-59, 1978. Thompson, J.M., J.A. Dempsey, L.W. Chosy, NT. Shahidi, and W.G. Reddan. Oxygen transport and oxyhemoglobin dissociation during prolonged muscular work. d. Appl. Physiol. 37:658-664, 1974. Tilz, G.P., W. Domej, A. Diez-Ruiz, G. Weiss, R. Brezinschek, H.P. Brezinschek, E. Hutel, H. Pristautz, H. Wachter, and D. Fuchs. Increased immune activation during and after physical exercise. Immunobiol. 188:194-202,1993. 68 Tisi, P.V., M. Hulse, A. Chulakadabba, P. Gosling, and C P . Shearman. Exercise training for intermittent claudication: does it adversely affect biochemical markers of the exercise-induced inflammatory response? Eur. J. Vase. Surg. 14:344-350, 1997. Torre-Bueno J.R., P.D. Wagner, H.A. Saltzman, G.E. Gale, and R.E. Moon. Diffusion limitation in normal humans during exercise at sea level and simulated altitude. J. Appl. Physiol. 58:989-995, 1985. Treppo, S., S.M. Mijailovich, and J.G. Venegas. Contributions of pulmonary perfusion and ventilation to heterogeneity in V&/Q measured by PET. J. Appl. Physiol. 82:1163-1176, 1997. Tronc, F., M. Wassef, B. Esposito, D. Henrion, S. Glagov, and A. Tedgui. Role of NO in flow-induced remodeling of the rabbit common carotid artery. Arterioscler. Thromb. Vase. Biol. 16:1256-1262, 1996. Tuder R.M., and N.F. Voelkel. Pulmonary hypertension and inflammation. J. Lab. Clin. Med. 132:16-24, 1998. Turcotte, R., L. Kiteala, J .E. Marcotte, and H. Perrault. Exercise-induced oxyhemoglobin desaturation and pulmonary diffusing capacity during high-intensity exercise. Eur. d. Appl. Physiol. 75:573-582, 1997. van der Poll, T., S.J. van Deventer, G. Pasterkamp, J.A. van Mourik, H.R. Buller, and J.W. ten Cate. Tumor necrosis factor induces von Willebrand factor release in healthy humans. Thromb. Haemost. 67:623-626, 1992. Varani, J . , S.E.G. Fligiel, G.O. Till, R.G. Kunkel, U.S. Ryan, and P.A. Ward. Pulmonary endothelial cell killing by human neutrophils. Possible involvement of hydroxyl radical. Lab. Invest. 53:653-663, 1985. Varani, J . , M.J. Bendelow, D.E. Sealey, S.L. Kunkel, D.E. Gannon, U.S. Ryan, and P.A. Ward. Tumor necrosis factor enhances susceptibility of vascular endothelial cells to neutrophil-mediated killing. Lab. Invest. 59:292-295, 1988. 69 Vitvitsky, E.V., J.P. Griffin, M.H. Collins, T.L. Spray, and J.W. Gaynor. Increased pulmonary blood flow produces endothelial cell dysfunction in neonatal swine. Ann. Thorac. Surg. 66:1372-1377, 1998. Wagenvoort, C.A., and N. Wagenvoort. Pathology of pulmonary hypertension. Wiley & Sons, New York, 1977. Wagner, D.D. Cell biology of von Willebrand factor. Ann. Rev. Cell Biol. 6:217-246, 1990. Wagner, P.D. Diffusion and chemical reaction in pulmonary gas exchange. Physiol. Rev. 57:257-312, 1977. Wagner, P.D., G.E. Gale, R.E. Moon, J.R. Torre-Bueno, B.W. Stolp, and H.A. Saltzman. Pulmonary gas exchange in humans exercising at sea level and simulated altitude. J. Appl. Physiol. 61:260-270, 1986. Wagner, P.D.. Ventilation-perfusion matching during exercise. Chest 101:192S-198S, 1992. Wagner, W.W., Jr., L.P. Latham, W.L. Hanson, S.E. Hofmeister, and R.L. Capen. Vertical gradient of pulmonary capillary transit times. J. Appl. Physiol. 61:1270-1274, 1986. Wang, J-S., C.J. Jen, H.C. Kung, L.J. Lin, T.R. Hsiue, and H-l. Chen. Different effects of strenuous exercise an moderate exercise on platelet function in men. Circulation 90:2877-2885, 1994. Wang, J-S, C.J. Jen, and H-l. Chen. Effects of chronic exercise and deconditioning on platelet fucntion in women. J. Appl. Physiol. 83:2080-2085,1997. 70 Wang, J-S, C.J. Jen, H-L. Lee, and H-l. Chen. Effects of short-term exercise on female platelet function during different phases of the menstrual cycle. Arterioscler. Thromb. Vase. Biol. 17:1682-1686, 1997. Ward, P.A., G.O. Till, J.R. Hatherill, T.M. Annesley, and R.G. Kunkel. Systemic complement activation, lung injury and products of lipid peroxidation. J. Clin. Invest. 76:517-527, 1985. Ward, P.A., A. Seekamp, and M.S. Mulligan. Acute inflammatory lung injury: mechanisms and interventions. In: Catravas JD, Callow AD, Ryan US, eds. Vascular endothelium: responses to injury New York: Plenum Press, 131-137, 1996. Ward, P.A. Recruitment of inflammatory cells into lungs: roles of cytokines, adhesion molecules , and complement, d. Lab. Clin. Med. 129:400-404, 1997. Ward, M., J.S. Milledge, and J.B. West. High altitude medicine and physiology. 2 n d Ed., London: Chapman&Hall, 1995. Warren, J.S., K.R. Yabroff, D.G. Remick, S.L. Kunkel, S.W. Chensue, R.G. Kunkel, K.J. Johnson, and P.A. Ward. Tumor necrosis factor participates in the pathogenesis of acute immune complex alveolitis in the rat. d. Clin. Invest. 1873-187, 1989. Warren, G.L., K.J. Cureton, W.F. Middendorf, C A . Ray, and J.A. Warren. Red blood cell pulmonary transit time during exercise in athletes. Med. Sci. Sports Exerc. 23:1353-1361, 1991. Warren R.S., K.D. Coutts, J .E. Potts, and D . C McKenzie. The effect of exercise intensity on post-exercise lung diffusion in highly trained athletes. Med. Sci. Sports Exerc. 3LS332, 1999 [Abstract]. Weibel, E.R. Morphometry of the human lung. New York: Springer Verlag, Berlin and Academic Press, 1963. 71 Weibel, E.R., and J . Gil. Structure function relationships of the alveolar level, in Engineering aspects of the lung, J.B. West, ed., New York: Marcel Dekker, 1977. Weight, L.M., D. Alexander, and P. Jacobs. Strenuous exercise: analogous to the acute-phase response? Clin. Sci. 81:677-683,1991. Weiler-Rarrel, D., A. Shupak, I. Goldenberg, P. Halpern, O. Shoshani, G. Hirschborn, and A. Margulis. Pulmonary oedema and hemoptysis induced by strenuous swimming. Br. Med. d. 311:361-362, 1995. Weinstock, C , D. Konig, R. Harnischmacher, J . Keul, B. Berg, and H. Northoff. Effect of exhaustive exercise stress on the cytokine response. Med. Sci. Sports Exerc. 29:345-354, 1997. Weiss, S.J. Tissue destruction by neutrophils. New Engl. d. Med. 320:365-376, 1989. Weiss, C , G. Seitel, and P. Bartsch. Coagulation and fibrinolysis after moderate and very heavy exercise in healthy male subjects. Med. Sci. Sports Exerc. 30:246-251, 1998. Weitzberg E., J.M. Lundberg, and A. Rudehill. Elevated plasma levels of endothelin in patients with sepsis syndrome. Circ. Shock. 33:222-227, 1991. Weller, A., S. Isenman, and D. Vestweber. Cloning of the mouse endothelial selectins: expression of both E- and P-selectin is inducible by tumor necrosis factor a . d. Biol. Chem. 267:15176-15183, 1992. West, J.B., K. Tsukimoto, O. Mathieu-Costello, and R. Predilleto. Stress failure in pulmonary capillaries. d. Appl. Physiol. 70:1731-1742, 1991. West, J.B., and O. Mathieu-Costello. Pulmonary blood-gas barrier: a physiological dilemma. NIPS 8:249-253, 1993. 72 Wetzel, R . C , H.A. Zacur, and J.T. Sylvester. Effect of puberty and estradiol on hypoxic vasomotor response in isolated sheep lungs. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 56:1199-1203, 1984. Wheeler, M.E., G.L. Davis, W.J. Gillespie, and M.M. Bern. Physiological changes in hemostasis associated with acute exercise. J. Appl. Physiol. 60:986-990, 1986. Wiedermann, C.J. , M. Niedermuhlbichler, and H. Braunstiner. Priming of polymorphonuclear neutrophils by atrial natriuretic peptide. J. Clin. Invest. 89:1580-1586, 1992. Williams, J.H., S.K. Powers, and M.K. Stuart. Hemoglobin desaturation in highly trained athletes during heavy exercise. Med. Sci. Sports Exerc. 18:168-172, 1986. Wu, D.X.Y., E.R. Weibel, H. Bachofen, and S. Schiirch. Lung lesions in experimental hydrostatic pulmonary edema: an electron microscopic and morphometric study. Exp. Lung Res. 21:711-730, 1995. Younes, M., and J . Burks. Breathing pattern during and after exercise of different intensities. J. Appl. Physiol. 59:168-172,1985. Younes, M., Z. Bshouty, and J . AM. Longitudinal distribution of pulmonary vascular resistance with very high pulmonary blood flow. J. Appl. Physiol. 62:344-358, 1987. Zhenxing, F., M.L. Costello, K. Tsukimoto, R. Prediletto, A.R. Elliot, O. Mathieu-Costello, and J.B. West. High lung volume increases stress failure in pulmonary capillaries. d. Appl. Physiol. 73:123-133, 1992. 73 Appendix I. Review of the Literature: Exercise-induced arterial hypoxemia 1.11ntroduction Intense exercise presents a challenge to oxygen homeostasis. Muscular work results in elevations of hydrogen ion concentration ([H+]), body temperature, and erythrocyte 2,3 diphosphoglycerate (DPG), which combine to produce a rightward shift in the oxyhemoglobin (HbC»2) dissociation curve and a reduction in arterial oxyhemoglobin saturation (SaC»2). In subjects with high aerobic power, decreased mixed venous oxygen saturation (SvC»2) returning to the pulmonary circulation (mean ± SD, SvC>2 = 15.8+1.5%; maximal oxygen consumption (V0 2 max) = 62+5 mlkg"1min"1, n=8 males; Harms ef al., 1998b) as a result of enhanced peripheral oxygen extraction by working muscle, and reduction in red blood cell pulmonary transit times secondary to elevated cardiac output (Q) (Hopkins ef al., 1996), together impose a severe demand on the ability of the lung to re-saturate hemoglobin during heavy exercise. Redistribution of pulmonary blood flow during exercise at high work levels (VO2 >3 L-min"1) further impairs gas exchange efficiency by increasing ventilation-perfusion (VA/Q) mismatch (Gale ef al., 1985; Hammond ef al., 1986; Shaffartzik ef al., 1992; Hopkins ef al., 1994; Hopkins ef al., 1996), while accentuation of venous admixture by overperfusion of alveoli with low V A /Q ratios contributes to reduction in arterial partial pressure of oxygen (Pa02). Nonetheless, despite these demands on pulmonary gas exchange, Pa02 remains within 1.3 kPa (10 Torr) of resting values and minimal %Sa02 generally exceeds 94% in most normal subjects during near maximal exercise (Asmussen and Neilsen, 1960). 1.2 Incidence and prevalence of exercise-induced arterial hypoxemia In many otherwise healthy subjects, however, and particularly in those capable of high levels of muscular work and cardiac output, pulmonary gas exchange is significantly impaired. Reductions in Pa02 and Sa02 during heavy exercise in normal subjects have been demonstrated for over 80 years (Harrop, 1919; Holmgren and Linderholm, 1958; Rowell ef al., 1964), but these early data were largely ignored. In the past two decades, exercise-induced arterial hypoxemia (EIAH), variably defined as a decrement in Pa02 of greater than 1.3 kPa (10 Torr) below resting values (Dempsey ef al., 1984; Harms ef al., 1998; Dempsey and Wagner, 1999), or a reduction of Sa0 2 below 91% (Williams ef al., 1986; Powers ef al., 1989), has been detected in up to half of healthy high aerobic power 74 male athletes (mean VC>2max >65 mlkg"1min'1) in whom arterial blood has been sampled during near maximal exercise (Dempsey et al., 1984; Powers ef al., 1988). In their pivotal study, Dempsey, Hanson and Henderson (1984) measured a decrease in Pa02 of greater than 1.3 kPa (10 Torr) from resting values in 12 of 16 high aerobic power male runners (V02max = 72±2 mlkg"1min"1) during a treadmill exercise test, with 8 of those 12 subjects demonstrating reductions in Pa02 of 2.8-4.7 kPa (21-35 Torr). Varying but similar degrees of impairment in arterial oxygenation during near sea level exercise have subsequently been replicated in several different laboratories (Johnson et al., 1992; Powers et al., 1992; Prefaut et al., 1994; Gore ef al., 1996; Rice et al., 1999a; Edwards et al., in press). However, despite reproducibility of EIAH, the paucity of exercise studies in which temperature corrected arterial blood gas variables have been measured, and the small sample sizes of studies with blood gas measures, does not allow for a reliable estimation of either incidence or prevalence of EIAH in the normal habitually active population. 1.3 Relationship between aerobic power and EIAH 1.3.1 Aerobic power and incidence of EIAH A significant inverse relationship has been noted between V0 2max and %SaC>2 measured at 1.5 minutes into constant load treadmill exercise at 95% VC>2max (r=-0.77, p<0.05; Williams et al., 1986). Other data demonstrate less impressive correlation (Pa02 at v'02max versus V02max, r=-0.49, p<0.05; Harms ef al., 1998). Many exceptions exist, and nonsignificant correlation between %Sa02 and V02max has been reported (Harms and Stager, 1995; Turcotte et al., 1997). The discrepancy is likely explained by differences in study populations. Correlation is significant in heterogeneous groups but nonsignificant in homogeneous groups. Thus, while significant reductions in %Sa02 and Pa02 are uncommonly seen in young healthy male athletes with weight adjusted V0 2max of less than 60-65 mlkg"1min"1 or absolute V02max <4.0 L-min"1 (Dempsey, 1986), aerobic power alone is insufficient to explain the intersubject difference in development of EIAH. Although a relationship between habitual activity, aerobic power and EIAH is cautiously supported by recent data (Harms et al., 1998a), the underlying (patho)physiological characteristic(s) capable of discriminating between similarly matched subjects with and without EIAH remain(s) incompletely understood. 75 13.2 Impact of EIAH on aerobic power and work performance Alterations in arterial oxyhemoglobin saturation have a much greater impact on arterial oxygen content and oxygen delivery than changes in Pa0 2 . Hence, reductions in %Sa0 2 have a potentially greater impact on exercise performance. In their recent review, Dempsey and Wagner (1999) proposed stratifying EIAH according to decreases in absolute Sa0 2 , with mild EIAH corresponding to an absolute Sa0 2 of 93-95%, moderate EIAH to a Sa0 2 of 88-93%, and severe EIAH to Sa0 2 values of <88%. Significant impairment in power output has been demonstrated in trained cyclists working under an artificially induced %Sa0 2 of 87% (Koskolou and McKenzie, 1994). Although an artificially induced %Sa0 2 of 90% failed to produce a significant reduction in power output, a downward linear trend in performance was observed beginning with mild hypoxia. Moreover, a linear change in V 0 2 m a x has been demonstrated below a threshold %Sa0 2 value of 95%, resulting in a 1-2% drop in V0 2max for each 1% reduction in %Sa0 2 (Lawler ef al., 1989). Therefore, it appears athletes with moderate or severe EIAH may potentially be at a performance disadvantage compared to their non-desaturating or minimally desaturating peers. 1.4 Influence of exercise duration, intensity, mode and repetition on EIAH 1.4.1 Exercise duration and intensity Progressive VVQ mismatch has been demonstrated during prolonged constant-load submaximal exercise (~65% V 0 2 m a x ) without concurrent progressive increased impairment in gas exchange as measured by Pa0 2 or the (ideal) alveolar/arterial oxygen difference (A-aD02) (Hopkins et al., 1998). Following an initial reduction in Pa0 2 and increase in (ideal) A-aD0 2 , further changes were not seen, although, theoretically, %Sa0 2 (not reported) may have continued to fall with shifts in oxyhemoglobin (Hb02) dissociation. Again, during brief (4-5 minutes), intense constant-load exercise, Pa0 2 has been shown to fall within the first minute and then remain relatively static (Dempsey et al., 1984; St. Croix et al., 1998; Sheel et al., in press). EIAH is more marked and often is only present during near maximal and maximal exercise (Dempsey et al., 1984; Powers et al., 1988), but many subjects display onset of gas exchange impairment with minimal or no hyperventilatory compensation beginning at moderate, submaximal work intensities (Harms et al., 1998a; Rice et al., 1999b). Demonstration of a greater (ideal) A-aD0 2 and impairment of arterial oxygenation in athletes with EIAH, at equivalent submaximal V 0 2 levels compared to 76 subjects with both similar and substantially less aerobic power, suggests gas exchange is functionally and/or structurally compromised in these habitually active subjects (Harms et al., 1998a). 1.4.2 Exercise mode Comparing the effect of uphill treadmill or cycle ergometer incremental maximal exercise on %Sa0 2 as determined by ear oximetry, Gavin and Stager (1999) measured significantly lower %Sa02 during treadmill running than cycling (mean±SE, 88.6±0.6% versus 92.6±0.6%, p>0.05) in thirteen male athletes (absolute V 0 2 = 4.83±0.11 L-min"1 versus 4.61±0.14 L-min'1, p>0.05). Combining treadmill and cycling data, a significant relationship was noted between Sa0 2 and the ratio of minute ventilation (VE) to carbon dioxide production (V E /VC0 2 ) (r=0.54, p>0.05), but no relationship was seen between differences in S a 0 2 and V E / V C 0 2 between modes of exercise (r=0.01). Hence, although a significantly greater V E / V C 0 2 was demonstrated during near maximal and maximal cycle ergometry, differences in %Sa0 2 between running and cycling could not adequately be explained by differences in ventilation. Significant differences in arterial oxygenation between exercise modes were noted in 17 female subjects who underwent progressive treadmill (R) and cycle (C) exercise (Hopkins et al., 2000). At near maximal exercise, Pa0 2 was significantly lower (p<0.0001) with running (mean±SE, 12.5±0.3 kPa (94±2 Torr)) versus cycling (14±0.3 kPa (105± 2 Torr)), and the (ideal) A-aD0 2 was higher (R:2.9±0.3 kPa (22+2 Torr) versus C:2.1+0.3 kPa (16±2 Torr), p<0.0001). Significant differences in ventilation were also noted between exercise modes. VE was significantly less during running (85.2±3.8 L-min"1) than cycling (98.2+4.4 L-min"1), and arterial P C 0 2 was significantly higher on the treadmill than the bike (4.3+0.07 kPa (32±0.5 Torr) versus 4±0.08 kPa (30+0.6 Torr), p<0.001. Therefore, these data support differences in gas exchange and ventilation between exercise modes by an as yet incompletely defined mechanism, but likely including differences in V A / Q matching and ventilatory response. Rice et al. (in press) similarly demonstrated significant differences in exercise-induced gas exchange impairment between cycle and treadmill exercise in a group of runners (mean±SE; n=7, V 0 2 p e a k = 72.7±1 mlkg"1min"1) and cyclists (n=6, V 0 2 p e a k = 67.3±3.3 77 mlkg"1-min"1). Each subject completed randomized, consecutive 5-minute exercise tests at -95% of their lower previously determined V02peak on each ergometer separated by 90 minutes of recovery. Significantly higher ratios of minute ventilation to oxygen uptake (VE/V0 2 ) were seen in runners and cyclists on the cycle ergometer compared to the treadmill, but repeated measures ANOVA revealed no significant interaction of training discipline and exercise modality on Pa0 2 . However, when data from both groups were pooled, significant differences were seen between treadmill exercise and cycle ergometry in Pa0 2 (meantSE; cycle: 12.0+0.3 kPa (90.2±2.5 Torr) versus treadmill: 10.8+0.2 kPa (80.8±1.8 Torr), p<0.05), PaC0 2 (cycle: 4.8±0.1 kPa (35.6±0.9 Torr) versus treadmill: 5.2±0.1 kPa (38.9±0.7 Torr), p<0.05), alveolar P 0 2 (PA0 2) (cycle: 15.3±0.1 kPa (114.4±0.9 Torr) versus treadmill: 14.5±0.1 kPa (108.9±0.8 Torr), p<0.05), [H+] (cycle: 59.7±1.5 nmolL"1 versus treadmill: 50.5±1.5 nmolL"1, p^O.05), and lactate (cycle: 15.9±0.7 mmol-L"1 versus treadmill: 9.7±0.9 mmol-L"1, p<0.05). Data reported were measured at 5 minutes of exercise. No significant differences were seen between exercise modalities in %Sa02, (ideal) A-aD0 2 or V0 2 p eak, although a trend towards a greater (ideal) A-aD0 2 was seen with running. Greater relative ventilation during cycling, likely induced in response to a greater lactate and [H+] load, appeared to preserve Pa0 2 , while running may have provoked more V V Q inequality. In addition, it also appears the greater ventilatory response seen during cycling offset any detrimental rightward shift of the Hb0 2 dissociation curve induced by an increase in [H+]. Despite differences between cycle and treadmill exercise and severity of EIAH, arterial desaturation develops independent of cycle ergometer protocol (Lama ef al., 1996) and position (Pedersen ef al., 1996), but may be exacerbated by 1 minute versus 5 minute increments on the treadmill (Hopkins ef al., 2000). Pedersen ef al. (1996) compared Pa0 2 and S a 0 2 in high aerobic power cyclists during recumbent and upright cycle exercise and failed to detect any significant difference in gas exchange between positions. While they argued that these data did not support a substantial role for V V Q inequality in the development of gas exchange impairment during exercise, they did not measure V V Q distribution. Their conclusion is based on the assumption that pulmonary blood flow distribution is altered by postural changes in hydrostatic gradients. If blood flow were determined primarily by gravity, then more homogeneous distribution and improvement in V V Q matching might be expected in the recumbent position. The lack of change in Pa0 2 78 and Sa02 between exercise positions does not support this conclusion. Instead, development of EAIH independent of exercise position is more consistent with the hypothesis that anatomical heterogeneity rather than gravity is a greater determinant of blood flow distribution (Glenny, 1998). 14.3 Repeated exercise Arterial oxyhemoglobin saturation measured by ear oximetry was not different between repeated maximal cycle exercise tests separated by 1-hour in 13 high aerobic power male athletes (McKenzie et al., 1999). Gas exchange variables measured during repeated episodes of maximal exercise also do not demonstrate increased gas exchange inefficiency in the second bout (Hanel et al., 1994; Caillaud et al., 1996; St. Croix ef al., 1998). Hanel et al. (1994) detected no difference in Pa0 2 , PaC0 2 , or %Sa0 2 between two 6-minute constant-load maximal rowing ergometer tests separated by a 2-hour recovery period. Again, no differences in Pa0 2 , A-aD0 2 , or PaC0 2 at maximal exercise were demonstrated between two incremental maximal cycle ergometer tests separated by a 30-minute recovery period, although reduction in Pa0 2 and increase in (ideal) A-aD0 2 were more marked during submaximal exercise in the second test (Caillaud ef al., 1996). However, these Pa0 2 values were likely underestimated since blood gas values were not corrected for temperature. Similarly, St. Croix et al. (1998) failed to demonstrate exacerbation of gas exchange impairment in 28 female subjects with varying levels of aerobic fitness during a constant-load maximal treadmill exercise test performed 20 minutes after an incremental treadmill test to exhaustion. On the contrary, they demonstrated a reduction in (ideal) A-aD0 2 and improvement in Pa0 2 and %Sa0 2 compared to end-point values measured during the progressive exercise test. Time spent at maximal workload was significantly longer in the constant-load test (p<0.004), while V 0 2 was not significantly different (p>0.004). These data suggest a functionally based mechanism present only during exercise, and support the inability of successive bouts of heavy exercise to induce or accentuate EIAH. They do not, however, address the effects of chronic, recurrent, intensive exercise on development of EIAH. 1.5 Influence of gender and age on EIAH Comparing 10 male masters endurance cyclists (age = 65.3±2.6 years; V0 2 m ax = 37.8±2.1 mlkg~1min~1), 10 age matched control subjects (age = 68.3±2.2; V 0 2 m a x = 27.9±1.5 79 mlkg"1min"1), and 10 young high aerobic power male cyclists (age = 23.3±1.1 years; V/02max = 66.9±1.3 mlkg~1min"1), Prefaut ef al. (1994) demonstrated significant reductions in PaG-2 in all masters cyclists, in 8 of 10 young male cyclists, but in none of the control subjects during an incremental cycle ergometer exercise test. Furthermore, the reduction in Pa02was greater in masters athletes compared to younger athletes at all levels of work, suggesting an accentuation of EIAH with increasing age in habitual exercisers. Similarly, 22 of 29 young eumenorrheic female athletes tested during their mid-follicular phase (age = 27±7 years; V02max = 57±6 ml-kg"1min"1) demonstrated decrements in PaC>2 of greater than 1.3 kPa (10 Torr) during an incremental treadmill exercise test, with 15 of these 22 subjects displaying decreases of greater than 2.6 kPa (20 Torr) (Harms ef al., 1998a; St. Croix ef al., 1998). Harms ef al. (1998a) further noted that 40% of their female subjects with V02max levels within 15% of predicted normal values (VC^max = 35-50 mlkg'1min"1) demonstrated EIAH, often with onset at submaximal exercise intensity. Comparing minimal PaC<2 data in these female subjects with previously published data from young males suggests female subjects develop similar reductions in Pa0 2 during exercise, but at much lower levels of V0 2 . Together, these few data support the hypothesis that female and older male subjects are more prone to EIAH, although recent data from female subjects do not suggest a greater degree of EIAH than in male subjects (Hopkins ef al., 2000). Mechanical differences in lung volume and expiratory flow may contribute to this potential difference (McClaran et al., 1998), but additional descriptive and experimental data are required to test the relationships between age and gender and EIAH. 1.6 Oxyhemoglobin dissociation during exercise The relationship between percentage SaC»2 and PaC>2 is non-linear and is described instead by a sigmoidal dissociation curve with flat and steep portions. Depending on position on the curve, alterations in PaC»2 will have minor (flat portion) or major (steep portion) influences on SaC»2. Elevations of [H+], partial pressure of carbon dioxide ( P C O 2 ) , body temperature, and DPG shift the HbC»2 dissociation curve rightward, which is reflected in an increase in the P 0 2 required for 50% saturation (P50). Furthermore, the combined effects of temperature and [H+] elevation during exercise are additive (Thompson ef al., 1974). Therefore, combined decrements in PaG"2 and rightward shift in the HbC>2 dissociation curve will affect reductions in SaC»2 during exercise. 80 In severe arterial hypoxemia, PaC>2 is on the steep portion of the dissociation curve, and any improvements in oxygen release at the tissues where PO2 is low are outweighed by deficiencies in arterial oxygenation and oxygen delivery. Under such severe conditions a rightward shift is disadvantageous. It is of interest then that a leftward shift in HbC>2 dissociation, induced by elevated DPG levels and respiratory alkalosis, has been demonstrated at atmospheric pressures equivalent to the summit of Mount Everest (32 kPa (240 Torr)), where Pa0 2 may fall as low as 3.7 kPa (27.6 Torr) (Sutton ef al., 1988; Ward et al., 1995). However, minimal Pa02 levels demonstrated during intense exercise do not approach those seen at altitude, and the primary effect of sea level exercise on the HbC>2 dissociation is rightward displacement. Short-term exhaustive exercise has been shown to affect HbC>2 dissociation independent of temperature and [H+], and alterations in DPG may account for up to 30-50% of the variability in %Sa02 during very intense exercise (Klein et al., 1980). Thus, in the economy of oxygen supply and demand during exercise, shifts in the Hb02 dissociation curve allow increased oxygen consumption at the periphery to affect a price on pulmonary gas exchange to benefit muscular work. 1.7 Mechanisms of EIAH Exercise-induced arterial hypoxemia potentially belies impairments in ventilation and/or gas exchange, and may result from hypoventilation, venous admixture, diffusion limitation, or some combination. Venous admixture comprises a continuum of VVQ ratios between zero and infinity and describes a spectrum between shunt and dead space ventilation. Diffusion is limited either by increasing thickness of the blood-gas barrier, by decreasing time for gas transfer from air to blood, or both. Such inefficiencies in gas exchange are reflected in a nonspecific increase in the (ideal) A-aD0 2. Maximal A-aD02 values have been noted to correlate strongly with minimum values of %Sa02 and PaC>2 during heavy exercise (%Sa02, r=-0.87, p<0.001, Hopkins & McKenzie, 1989; Pa0 2 , r=0.93, p<0.05, Harms et al., 1998), whereas relationships between PaG-2 and PaCC>2 at V02max have been less impressive (r=0.62, p<0.05, Harms ef al., 1998). Rice et al. (1999) utilized multiple inert gas elimination technique (MIGET) in 12 healthy male cyclists with high aerobic power (absolute V02>5 L-min"1) under normoxic and hypoxic conditions and failed to demonstrate any significant differences in measures of PaC0 2 or VVQ inequality -measured by the log SD of the perfusion distribution (logSDo.) - between athletes with EIAH (n=7) versus matched controls (n=5). Diffusion limitation was estimated from the 81 difference between observed and predicted A-aDC>2 and accounted for more of the observed rise in A-aD0 2 in athletes with EIAH than in controls (p<0.01). Further analysis including 0 2 diffusion capacity (DL02), logSDQ and PaC0 2 values at 90% V0 2 p e ak in stepwise multiple linear regression explained 90% of the variance in Pa0 2 , with DL0 2 , logSDo. and PaC0 2 each accounting for approximately 30%. These data, in conflict with earlier studies (Gale et al., 1985; Hammond et al., 1986; Wagner et al., 1986; Shaffartzik et at, 1992; Hopkins et al., 1994), do not demonstrate a rise in Vp/Q inequality with increasing exercise. Even so, these MIGET data do support a multifactorial etiology of EIAH comprised of some combination of hypoventilation, VVQ inequality, and diffusion limitation. 1.7.1 Relative hypoventilation Forming the initial step in the oxygen cascade from atmosphere to mitochondria, alveolar ventilation affects the driving pressure of gas transfer across the blood-gas membrane. Arterial P 0 2 reflects the sum of mixed venous P 0 2 and the contribution of P A 0 2 , of which the latter is the primary determinant under conditions of normal gas exchange. At sea level, at rest, P A 0 2 approximates 13.3 kPa (100 mmHg). As VE increases with exercise hyperpnea, P A 0 2 rises towards the partial pressure of inspired oxygen (P|02) of approximately 20 kPa (150 mmHg). Almost all athletes show a reduction in P a C 0 2 during exercise (Dempsey et al., 1984; Harms and Stager, 1995; Harms et al., 1998), but failure to mount an adequate increase in VE to maintain blood gas homeostasis betrays a relative hypoventilation (Dempsey et al., 1984; Powers, ef al., 1993). The effect on oxygenation is minimal, but insufficient exercise VE may indirectly play a permissive role in the development of hypoxemia in some subjects by failure to compensate for an increased (ideal) A-aD0 2 (Johnson et al., 1992). Dempsey and Wagner (1999) have suggested that P a C 0 2 values of 4.7-5.1 kPa (35-38 Torr) reflect a borderline hyperventilatory response to exercise, while PaC0 2 values of greater than 5.1 kPa (38 Torr) mark lack of compensatory hyperventilation. Decreases in ventilatory response, encroachment upon the maximal flow-volume loop, and/or respiratory muscle fatigue have each been suggested as potential explanations for an inadequate alveolar hyperventilatory response seen in some subjects during heavy exercise. 82 1.7.1.1 Ventilatory response Ventilatory response to chemical stimuli is diminished in endurance athletes (Byrne-Quinn et al., 1971; Martin et al., 1979), while subjects with the least ventilatory response to exercise display the greatest reduction in arterial oxygenation (Dempsey et al., 1984). Yet, whether or not low chemoresponsiveness leads to development of hypoxemia remains undetermined. Hopkins and McKenzie (1989) failed to detect significant correlation between hypoxic ventilatory response (HVR) and VE (r=0.08), the ratio of minute ventilation to oxygen consumption (VE/V02) (r=0.1), or arterial desaturation (r=0.06) at maximal treadmill exercise in 12 male subjects (V02max = 63.0±2.2 ml-kg"1-min"1) despite significant intersubject differences in HVR (normal HVR, n=6: A V E / A S a 0 2 = 1.02+0.15 L-min"1 %"1 versus blunted HVR, n=6: A V E / A S a 0 2 = 0.33±0.1 L-min"1 %"1; p value not reported). Furthermore, no difference was observed in minimal %Sa0 2 between subjects with normal (minSa02 = 92.2+2.36%) and blunted HVR (minSa02 = 91.3+2.07%) (p=NS), and P A 0 2 values were consistently above 110 mmHg after the first minute of exercise in both groups. Non-significant correlation between Pa0 2 and indirect measures of ventilatory response, including P A 0 2 at 90%VO 2 m a x (r=0.51, p>0.05), P A 0 2 (r=0.17, p>0.05), V E / V 0 2 (r=0.17, p>0.05), and PaC0 2 (r=0.34, p>0.05) at V 0 2 m a x seen in 12 high aerobic power cyclists, supports a minor contribution to EIAH from hypoventilation (Powers etal., 1992). On the other hand, Harms and Stager (1995) reported significant correlation between ear oximetry %Sa02, end-tidal P 0 2 (P E T0 2 ) (r=0.84, p<0.05) and P C 0 2 ( P E T C 0 2 ) (r=-0.70, p<0.05), and V E / V 0 2 (r=0.72, p<0.05) at V0 2 max in 36 male subjects ( V 0 2 m a x = 64.1+7.8 ml-kg"1-min"1) during maximal treadmill exercise. In contrast to Hopkins and McKenzie (1989), Harms and Stager also noted significant positive relationships between %Sa0 2at V 0 2 m a x and HVR (r=0.63, p<0.05) and %Sa0 2 at V 0 2 m a x and hypercapnic ventilatory response (HCVR; r=0.62, p<0.05). They argued the apparent discrepancy between studies was related in part to different definitions of EIAH and suggested a significant difference in HVR could be demonstrated between subjects with %Sa0 2 greater than or less than 92%. However, statistical comparison demonstrates a non-significant difference between groups (HVR, min%Sa02 >92%: 0.69±0.11 L-min"1 %"1 versus HVR, min%Sa02 <92%: 0.66±0.18 L-min"1 %"1, p>0.05; data from Hopkins and McKenzie, 1989). 83 Dividing 13 healthy active males into two groups based on normoxic maximal exercise V E / V 0 2 (low < 27.7; high s 30.2) and ideal estimates of P A 0 2 (low < 14.3 kPa (107 Torr); high > 14.7 kPa (110 Torr)), Gavin ef al. (1998) demonstrated significantly lower ventilatory response (VE, V E /V0 2 , and V E /VC0 2 ) and greater decline in %Sa0 2 during hypoxic exercise (F|0 2 = 0.13) in the low group. During normoxic exercise, however, the low group did not demonstrate EIAH. In agreement with Hopkins and McKenzie (1989), they also failed to detect significant correlation between resting HVR and any ventilatory response variables or %Sa02. Cooper (1993) measured a significantly greater hypercapnic peripheral chemoresponse at rest and during mild to moderate exercise in high aerobic power cyclists without EIAH compared to subjects with minimal exercise %Sa02<92% (p=0.004), but detected no significant difference in hyperoxic peripheral chemoresponse (p=0.988) between groups. Therefore, a blunted hypercapnic peripheral chemoresponse at rest and during mild to moderate exercise may permit accumulation of C 0 2 and predispose to relative hypoventilation, but blunted HVR or peripheral hyperoxic response at rest and during exercise does not consistently predict exercise-induced gas exchange impairment. Clearly, intersubject variabilty exists in the compensatory hyperventilatory response to exercise, and this difference contributes to EIAH. Its relative impact on arterial oxygenation, however, is insufficient to explain gas exchange impairment demonstrated during submaximal and maximal exercise. 1.7.1.2 Expiratory flow limitation Mechanical expiratory flow limitation may occur in some athletes at maximal exercise (Johnson etal., 1992; McClaran etal., 1998; McClaran etal., 1999; Mota etal., 1999), but maximal reserves are reached coincident with, not prior to, V 0 2 m a x . Dempsey (1986) estimated an athlete exercising at a V C 0 2 of 5 or 6 L-min"1 would require a VE in excess of 200 L-min"1 to achieve a PaC0 2 of 4 kPa (30 Torr) and P A 0 2 of 15.3 to 16 kPa (115 to 120 Torr). Tidal breaths required to reach this extreme level could intrude upon the mechanical limits of inspiratory and expiratory pressure and flow. Johnson ef al. (1992) measured pleural pressures in 8 competitive male endurance runners (mean±SE; V 0 2 m a x = 73±1 mlkg"1min"1) at rest over a range of lung volumes and flow rates. Capacity to generate inspiratory pressure (Pcapi) and the maximal effective pleural pressure beyond which flow rates no longer increased (Pmaxe) were determined. The athletes were then 84 tested at maximal exercise. Peak tidal inspiratory pressure reached an average of 89% of Pcapi, and end-expiratory lung volumes averaged 86% of total lung capacity. Mean blood gas measurements during maximal exercise showed a moderate decrease in P a C 0 2 to 4.8 kPa (36 Torr), a P A 0 2 of 14.7 kPa (110 Torr), and variable hypoxemia (Pa0 2 range 8.7-11.1 kPa (65-83 Torr)). Attempts to further increase the stimulus to breathe during maximal exercise via hypercapnia or hypoxemia failed to increase either VE or inspiratory and expiratory pressures. They concluded mechanical limitation occurred during maximal exercise, but their data would also be consistent with lack of chemoresponsiveness. Moreover, P a C 0 2 at maximal exercise never rose above resting values, nor did it rise with increasing work levels. This suggests mechanical limitation did not severely limit the ventilatory response. In addition, the effects on arterial oxygenation were minimal, and, as in other studies, the reduction in Pa0 2 was more closely related to the rise in (ideal) A-aD0 2 (r=-0.96, p<0.01). Applying negative expiratory pressure at the mouth during exercise, and comparing the subsequent flow-volume curve to the preceeding breath, Mota et al. (1999) failed to demonstrate expiratory flow limitation (EFL) in 9 of 10 male competition cyclists (mean V 0 2 m a x = 72 mlkg"1min'1) during a progressive maximal exercise test. Expiratory flow increased with negative expiratory pressure over the entire tidal volume range during exercise in all but one subject. End-expiratory lung volume (EELV) at rest and peak exercise was not significantly different. Attempts to increase VE during exercise, by enhancing laminar flow with normoxic helium (He02) and reducing flow resistance, have variably been reported to increase Pa0 2 (Dempsey et al., 1984) or have no effect on %Sa0 2 reduction (Buono and Maly, 1996). In 4 of 5 runners breathing either room air or He0 2 during 3 to 4 minutes at 75-90% V 0 2 m a x , Pa0 2 increased between 0.7 and 2 kPa (5 and 15 Torr) during He0 2 breathing; the effects were reversed on return to room air (Dempsey et al., 1984). However, 4 other subjects either failed to increase Pa0 2 above 8.7 kPa (65 Torr) despite hyperventilation with He0 2 or developed EIAH despite a PaC0 2 less than 4.7 kPa (35 Torr). Likewise, although VE increased from 139 L-min"1 to 168 L-min"1 (p<0.05) with He0 2 in 7 male cyclists (mean±SE, VC^max = 65±2 ml-kg"1-min"1) during an incremental maximal exercise test, there was no corresponding change in %Sa0 2 (Buono and Maly, 1996). Mean %Sa0 2at V0 2 m a xwas 85 90% in room air and 89% with He0 2 (p>0.05). Nonetheless, He0 2 breathing does prevent the reduction in V E and increase in EELV seen when tidal expiratory flows abut resting maximal flow-volume loops during maximal exercise, lending support to the presence of EFL in some athletes (McLaran et al., 1998). Furthermore, the oxygen cost of breathing may account for up to 10% of V 0 2 m a x (Aaron et al., 1992) and up to 15% of the total cardiac output (Harms et al., 1998b). The importance of these requirements is suggested by recorded improvements in performance times with He0 2 breathing (Aaron et al., 1985). In all, these He0 2 studies suggest encroachment on mechanical limits to ventilation may occur in some athletes during heavy exercise. Modest reductions in PaC0 2 and increases in Pa0 2 with He0 2 result in 30-40% improvement in EIAH (Dempsey et al., 1984; McClaran etal., 1998; McClaran etal., 1999). Expiratory flow limitation (EFL) during exercise may, however, have a more significant impact in females (McClaran ef al., 1998). Females have smaller lung volumes and reduced expiratory flow rates than males, even when compared to height adjusted values (American Thoracic Society, 1991). Sitting height may account for some of this discrepancy (Schwartz et al., 1988). Greater EFL, with increased end-expiratory and end-inspiratory lung volumes, was seen in higher aerobic power females ( \ /0 2 m ax >57 mlkg"1min"1) compared to less fit women (V0 2 m a x <56 mlkg"1min"1) (McClaran et al., 1998). He0 2 versus room air breathing again demonstrated increased maximal flow rates, breathing frequency, VY, and VE when EFL was reduced with He0 2 , but no change in VE with He0 2 was seen if EFL was not present. Despite EFL, PaC0 2 values were less than 5.1 kPa (38 Torr) in all subjects during room air exercise, nor was any correlation seen between the degree of EFL and magnitude of the hyperventilatory response (data not reported). Thus, while EFL may constrain ventilation, it does not appear to induce hypoventilation. Although some of these same female subjects also demonstrated significant EIAH (Harms ef al., 1998), the contributing role of EFL seems peripheral or permissive at best. Arguably, mechanical limitation on the hyperventilatory response could restrain ventilatory compensation for a widened (ideal) A-aD0 2 , but this is not well supported by a measured reduction in PaC0 2 . However, under conditions of impaired gas exchange, as seen during exercise, any reduction in P A 0 2 would permit increased contribution of mixed venous P 0 2 to arterial oxygenation, and potentially could enhance arterial hypoxemia. 86 1.7.1.3 Respiratory muscle fatigue Perhaps the greater impact of EFL is on work of breathing and/or respiratory muscle fatigue. Although a high density of slow twitch fibers makes the diaphragm resistant to fatigue, respiratory muscle fatigue has been demonstrated on volitional tests following both prolonged submaximal exercise (Loke ef al., 1982; Ker and Schultz, 1996) and heavy short-term exercise (Bender and Martin, 1985; McConnell ef al., 1997). Bilateral phrenic nerve stimulation (BPNS) at low frquency (10-20 Hz) detected up to a 40% reduction in transdiaphragmatic pressure (Pdi) in the majority of twelve healthy subjects immediately following and up to 1 hour after exercise at 90-95%VC^max (Johnson ef al., 1992). Similarly, BPNS at low and high (50-100 Hz) frequency has detected reduction in Pdi following exercise at 95% VC^max (Babcock ef al., 1998). The impact of respiratory muscle fatigue on exercise VE is, however, unknown. Given the demands on respiratory muscles during heavy exercise, competition between respiratory and peripheral locomotor muscles may exist, and this hypothesis is supported by demonstration of reduced leg blood flow under conditions of increased respiratory work (Harms ef al., 1997). Thus, exercise hyperpnea may impose a cost to the balance between central and peripheral oxygen demands, and could reduce performance, but the implications for development of EIAH are not yet clear. Apart from He02 breathing, other efforts to increase ventilation during exercise, including respiratory muscle training (Leith and Bradley, 1976; Morgan ef al., 1987; Fairbarn ef al., 1991; Boutellier and Piwko, 1992; Boutellier ef al., 1992) and external assist devices (Krishnan ef al., 1996), have produced mixed results on exercise performance. Despite significant changes in maximal voluntary ventilation after 4 weeks of respiratory muscle training with isocapnic hyperventilation (p<0.01), no change in V02max, VE at V02max, or endurance time at 90% of peak power output (p>0.05) was seen between experimental (n=5) and control groups (Fairbarn ef al., 1991). Morgan ef al. (1987) also failed to detect a significant increase in performance following respiratory muscle training. In contrast, large increases in cycle endurance time at threshold (approximately 77% V02max), upwards of 50% in non-athletes, have been reported in sedentary (Boutellier and Piwko, 1992) and trained subjects (Boutellier ef al., 1992) following 4 weeks of respiratory muscle training. Other investigators have not yet replicated such dramatic increases. 87 In summary, relative hypoventilation, due in part to decreased chemoresponsiveness, EFL, or respiratory muscle fatigue, may play a permissive role in development of EIAH by failure to adequately compensate for increasing gas exchange impairment. However, it cannot explain the widened (ideal) A-aDG"2 seen during exercise. 1.7.2 Elevated alveolar/arterial gradient Increases in (ideal) A-aDC>2 occur in all subjects during exercise, largely as a result of background VVQ inequality, a rise in the respiratory exchange ratio (RER), and a reduction in PaC02 from hyperventilation. Assuming P A C 0 2 equals PaC02, measured reduction in PaC02 to 4.8 kPa (36 Torr) predicts an increase in (ideal) P A02 to approximately 16 kPa (120 Torr) or greater. Failure to increase Pa02 above resting levels (approximately 13 kPa (98 Torr) at sea level) reflects a rise in (ideal) A-aD02 from 0-0.7 kPa (0-5 Torr) towards 3-3.3 kPa (20-25 Torr). Similar measures of change in gas exchange efficiency are commonly seen during exercise in normal subjects. Thus, the (ideal) A-aD02 will be elevated with exercise without necessarily an accompanying alteration in Pa02- Reductions in PaC>2 reflect greater widening of (ideal) A-aD02, with values of up to 6.7 kPa (50 Torr) reported in high aerobic power athletes during heavy exercise (Hopkins and McKenzie, 1989). Noted previously, reduction in Pa02 during exercise correlates most strongly with excessive widening of (ideal) A-aD02, which points to inefficiencies in gas exchange - one or both of VVQ inequality or diffusion limitation -as the primary etiology of EIAH. 17.2.7 VVQ inequality Intrapulmonary or extrapulmonary venoarterial shunts form one extreme on the spectrum of VVQ mismatch whereby arterial PO2 equals mixed venous PO2. Venous blood from bronchial and coronary veins normally enters the left ventricle having bypassed the lung. Contribution of deoxygenated blood to the arterial circulation may account for up to half of the difference between predicted and measured A-aD02 at rest (Gledhill et al., 1977), but calculated shunt fractions during heavy exercise when breathing 100% O2 are less than 2% (Wagner et al., 1986). Theoretically, even without increasing fractional perfusion, shunt could have a greater impact on PaC"2 during exercise by contribution of considerably reduced mixed venous PO2. However, shunt effect on PaC>2 during exercise is minimal. 88 This is demonstrated by a proportionate rise in P A 0 2 and PaO"2, with restoration of PaC>2 to normal levels in athletes with EIAH on administration of mildly hyperoxic gas (F1O2 = 0.24-0.26) (Dempsey etal., 1984; Powers etal., 1992). Vfi/Q ratios, as determined by MIGET, have been variably shown to disperse during exercise, reaching statistical significance above VO2 of approximately 3 L-min"1 in male subjects (Gale ef al., 1985; Hammond et al., 1986; Wagner et al., 1986; Shaffartzik ef al., 1992). Topographic Vp/Q relationships have appeared more uniform at moderate VO2 levels of 1-2 L-min"1 (Bake ef al., 1968), consistent with more homogeneous blood flow distribution. An initial rise in flow and perfusion pressure likely overcame some portion of resting regional blood flow heterogeneity. Paradoxically, this initial trend towards homogeneity occurred as (ideal) A-aDC>2 widened. As exercise intensity increased, Vp/Q dispersion was seen in more, but not all subjects, despite a universal increase in (ideal) A-aD02 (Gale et al., 1985; Hammond ef al., 1986). Greater Vp/Q dispersion would allow any increase in arterial blood from alveolar units with low Vp/Q ratios, operating as they are on the steep portion of the Hb02-dissociation curve, to have an adverse impact on oxygen content out of proportion to their relative contribution. Therefore, the greater the dispersion of Vp/Q ratios, the wider the (ideal) A-aD02, and the lower Pa02 and %Sa02-However, alveolar ventilation increased relatively more than Q during exercise and contributed to a shift in the Vp/Q distribution towards higher VA/Q ratios. As such, the component of (ideal) A-aD02 attributed to Vp/Q inequality remained relatively constant up to an absolute VO2 of approximately 3.7 L-min"1 (Wagner ef ai, 1986). The magnitude of dispersion was small, with logSDo. rising from a mean of 0.35±0.12 at rest to an average of 0.44±0.10 at an absolute VO2 of 3.22±0.84 L-min'1. Combining data from their earlier study (Gale et al., 1985), Wagner et al. (1986) calculated a rise of 0.05 in logSDQ per litre VO2 and predicted a logSDQ of 0.54 at V0 2 of 4 L-min"1. At this intensity, MIGET predicted 1/3 of the calculated A-aD02 was accounted for by VVQ inequality, which suggests the remaining difference was attributable to diffusion limitation plus or minus a post-pulmonary shunt. At higher work intensities (absolute VO2 >5 L-min"1), Vp/Q inequality has been reported to account for between 30% (Rice ef al., 1999a) and 60% (Hopkins et al., 1994) of the 89 increase in (ideal) A-aDC»2. Comparing the studies of Hammond et al. (1986), Wagner et al. (1986), Hopkins et al. (1994) and Rice et al. (1999), which to date are the only studies where MIGET and blood gas data have been collected during moderate to heavy exercise, reveals distinct differences between measurements and/or subjects. At a comparable absolute VO2 of approximately 4 L-min"1, Hammond et al. (1986) reported a mean predicted A-aD02(p) of 1.4±1 kPa (10.7±7.8 Torr) and a difference between mean observed and predicted A-aD02(o-p) of (3.1 ±1.1 - 1.4±1 = 1.7 kPa, or 23.0+8.0 - 10.7±7.8 = 12.3 Torr) in 5 male subjects, whereas Hopkins et al. (1994) recorded a mean A-aD02(p) of twice that value and a mean A-aD02(o-p) of approximately one-half in 10 male high aerobic power cyclists. Comparison of mean logSDQ values also reveals differences, with Hammond ef al. (1986) reporting a resting value of 0.28±0.13 and a value during heavy exercise of 0.58±0.30, Hopkins ef al. (1994) reporting a resting value of 0.54±0.13 (the same value Wagner ef al. (1986) predicted for a VO2 of 4 L-min"1) and a value during heavy exercise (300W) of 0.64±0.10, and Rice ef al. (1999a) recording values of 0.38±0.08 at rest and 0.37±0.05 during heavy exercise. Interestingly, in the later study, logSDQ values were not significantly different at rest or during heavy exercise between subjects with and without EIAH (minimal Pa02 11.6±0.5 kPa (87±4 Torr) versus 12.9±0.8 kPa (97±6 Torr), p<0.05), nor was there a significant difference between groups in A-aD02(o-p) during hypoxic exercise (Rice ef al., 1999a). These data, as Wagner (1992) has previously pointed out, reveal considerable inter-subject variability in the amount of V V Q inequality demonstrated over varying exercise intensities. A satisfactory explanation for inter-subject differences in V V Q inequality remains incomplete, and several theories were recently reviewed (Dempsey and Wagner, 1999). First, underlying structural differences in airways and/or blood vessels could become unmasked with elevated flow rates during exercise, although persistence of V V Q inequality past recovery of V E and Q to pre-exercise levels does not support this (Schaffartzik ef al., 1992). Second, subclinical bronchoconstriction, airway secretions and/or dynamic expiratory compression of airways could impair ventilation distribution (Johnson ef al., 1992), but normalization of V V Q matching during exercise at altitude by administration of 100% O2, and lack of significant spirometric changes following exercise, lend more support to a circulatory mechanism (Gale ef al., 1985; Wagner, 1992). Third, mediators of airway or vascular tone could alter ventilation and/or perfusion distributions. 9 0 Support for this potential mechanism is offered by demonstration of increased plasma histamine levels in athletes with EIAH (Anselme et al., 1994) and improvement in gas exchange impairment after administration of nedocromil (Prefaut ef al., 1997). Fourth, increased transcapillary fluid flux from enhanced pulmonary blood flow and perfusion pressure could promote accumulation of mild interstitial edema, which, by collecting around bronchioles, vessels or the alveolar wall, could alter compliance, resistance and/or alveolar-capillary architecture and change local V V Q distribution (Hopkins et al., 1998). Although supporting evidence is largely indirect, transient interstitial edema has been suggested as the most attractive hypothesis, not only because increased extravascular lung water could potentially explain both W Q inequality and diffusion limitation, but because it could also account for accentuation of V V Q inequality under hypoxic conditions where pulmonary vascular pressures are raised due to hypoxic vasoconstriction (Wagner, 1992). Moreover, these potential mechanisms are not mutually exclusive, and one or more of them could contribute to alterations in W Q dispersion during exercise. For example, regional blood flow heterogeneity, a consequence of fixed and variable components, could promote segmental overperfusion. Variation in the vascular response, with potential for pressure-induced endothelial dysfunction/injury, release of inflammatory mediators, increase in membrane permeability and accumulation of interstitial fluid, could have a significant impact on V V Q ratios and diffusion. Accentuation of perfusion heterogeneity at altitude by hypoxic vasoconstriction probably contributes to development of high-altitude pulmonary edema (HAPE), a condition known to be associated with pulmonary hypertension and loss of blood-gas membrane integrity. Hence, it is tempting, as Wagner (1992) speculated, to consider EIAH during sea level exercise as part of a continuum of pulmonary vascular response to increased flow and perfusion pressure that ends in fulminant alveolar edema. The potential relationship between susceptibility to HAPE (HAPE-S) and EIAH is discussed in the context of the pulmonary circulation and exercise in further detail in Appendix II. 1.7.2.1.1 Exercise-induced interstitial edema Reports of frank pulmonary edema following exercise at sea level are rare. McKechnie et al. (1979) described acute pulmonary edema in two participants following a Commrades Marathon (90-km), and documented plain film radiographic evidence. Weiler-Raller et al. 91 (1995) also reported pulmonary edema and hemoptysis in 8 swimmers involved a swimming meet. Symptom onset occurred within 45 minutes of the start of exercise and forced 5 of these participants to withdraw from further competition. In 2 athletes, plain film radiographs taken upon hospitalization demonstrated evidence of interstitial and alveolar edema. Although attributed to heavy exercise, the dramatic presentation of several competitors in a pool environment suggests chlorine or other toxic gas inhalation as a more probable explanation. Evidence for exercise-induced pulmonary interstitial edema is suggested indirectly by persistence of VVQ mismatch into recovery despite normalization of Q and VE , and by post exercise reduction in vital capacity without concurrent alteration in expiratory flow rates (Schaffartzik ef al., 1992). In addition, rapid shallow breathing (RSB), possibly a result of stimulation of pulmonary J receptors by interstitial edema (Paintal, 1973), has been noted during recovery from maximal exercise (Younes and Burks, 1985). Plotting exercise and recovery VE as a function of tidal volume (VY) to estimate RSB, with both values normalized to vital capacity (VC), Caillaud ef al. (1993) documented a significantly greater change in Vj(%VC) (AVj(%VC)) during recovery from maximal exercise in high aerobic power male athletes (mean±SEM; VC»2max = 68.6±1.6 mlkg"1min"1) with persisting elevation of (ideal) A-aDC»2 than in untrained controls (V0 2 m a x = 46.3±2.2 mlkg"1min"1) (p<0.002). However, despite a trend, no significant relationship was seen between AVT(%VC) and (ideal) A-aD0 2 . Attempts to measure mild, transient interstitial edema have unfortunately been limited by current technology. Transthoracic electrical impedance (TEI) has been reported to fall (Buono ef al., 1983) or rise (Rasmussen ef al., 1992) following exercise. Changes in TEI may reflect changes in lung water, but the technique is nonspecific and could as likely reflect alterations in chest wall or skin perfusion. Post-exercise reduction of lung diffusing capacity to carbon monoxide (D|_co) was previously thought to reflect an increase in interstitial lung water (Buono ef al., 1983; Manier ef al., 1991; Rasmussen et al., 1992), but documentation of similar reductions in D|_co 1 hour following 10 minutes of exercise at 30%VO2max or 90%VC>2max in fit male cyclists (Warren ef al., 1999), or following maximal exercise in males with disparate levels of aerobic power (mean±SD; VC»2max = 51.6±4.7 ml-kg"1min"1 versus 68.0±4.9 mlkg"1min"1) (Sheel ef al., 1998), lends more support to a 92 post-exercise reduction in pulmonary capillary blood volume (Vc) than to a decrease in the diffusion capacity of the alveolar membrane. Hanel et al. (1997), from estimates using ""Technetium pertechnetate labelled blood, suggested up to one half of the post-exercise reduction in DLco could be accounted for by a shift in blood volume from the core to the periphery. Therefore, post-exercise measures of DLco would potentially be insensitive to changes in the alveolar membrane and are not a reliable estimate of interstitial edema. Imaging studies have likewise been inconsistent. Plain film radiographs were not diagnostic for interstitial edema after short-term intense exercise in subjects with moderate aerobic power (range of absolute VO2 = 3.07-3.77 L-min"1; Gallagher ef al., 1988). Similarly, McKenzie ef al. (1996) failed to detect increases in extravascular lung water with computerized tomography following near maximal exercise. Conversely, Cauillaud ef al. (1995) and McKenzie ef al. (1999), using computerized tomography (CT) and magnetic resonance imaging (MRI), respectively, demonstrated significant increases in radiographic opacities after prolonged submaximal endurance exercise, consistent with an accumulation of interstitial lung water. Although sensitivity of CT and MRI is not sufficient to distinguish between increases in capillary blood volume or interstitial fluid, demonstration of decreases in Vc post-exercise would suggest the imaging findings reflect interstitial edema. Even so, gas exchange variables were not measured in any of these studies, and the potential relationship between exercise-induced increases in extravascular lung water and pulmonary gas exchange remains incompletely addressed. 1.7.2.1.2 Airway and/or vascular mediators and EIAH Anselme ef al. (1994) demonstrated a significant correlation (r=0.8, p<0.01) between the change in plasma histamine and the decrease in Pa02 in 7 young male and 7 masters male athletes with EIAH compared to age matched controls. In a follow-up study, they inhibited histamine release with nedocromil sodium in seven male masters athletes with EIAH in a randomized, double-blind placebo controlled trial and demonstrated a significant improvement in pulmonary gas exchange in the treatment group (Prefaut ef al., 1997). Rivier ef al. (1994) measured plasma levels of IL-6 and TNF-a in 10 young male high aerobic power athletes and 6 masters endurance athletes immediately following an incremental cycle exercise test. They demonstrated significant correlation between in vitro release of interleukin-6 (IL-6) or tumor necrosis factor-a (TNF-a) and V02max or power 93 output, but did not find significant correlation between IL-6 or TNF-a and A-aDC>2or Sa0 2 . Although this is the only study to compare cytokine levels with an outcome measure of gas exchange, the minimum saturation recorded did not drop below 92%, and in vitro rather than in vivo responses were measured. Thus, the potential relationship between soluble inflammatory mediators and severe EIAH has not been tested. The potential relationship between nitric oxide (NO) and EIAH has also been investigated. Nine high aerobic power male athletes ( V 0 2 m a x = 65±0.5 mlkg"1min"1) repeated a maximal cycle ergometry exercise test with and without 15 parts per million (ppm) of inhaled NO. NO provoked a reduction in histamine release but, although A-aD0 2 and Pa0 2 did not change significantly between 75 and 100% V0 2 m a x , gas exchange inefficiency and hypoxemia were accentuated beginning at rest (Durand et al., 1999). Although inhaled NO induces pulmonary vasodilatation and a selective increase in pulmonary perfusion of ventilated alveoli, which, under conditions of dead space ventilation reduces V V Q ratios toward unity, NO also opposes hypoxic pulmonary vasoconstriction. By redirecting blood flow to hypoxic alveoli, inhaled NO can increase venous admixture and hypoxemia (Sprague ef al., 1992). These differences were demonstrated in patients with chronic obstructive pulmonary disease (COPD) by estimating V V Q distributions with MIGET at rest and during exercise with inhaled NO (Roger et al., 1997). At rest, inhaled NO accentuated gas exchange impairment by increasing W Q mismatch, but inhaled NO during exercise attenuated the rise in pulmonary artery pressure and redistributed blood flow from alveolar units with low W Q ratios toward units with normal ratios. Gas exchange correspondingly improved. Thus, in the study by Durand et al. (1999) it appears venous admixture was accentuated by excessive capillary flow at rest which persisted until near maximal exercise when increasing V E could elevate W Q ratios towards normal. More recently, 7 high aerobic power male cyclists performed four 5 minute exercise sessions at V 0 2 m a x under randomized conditions of normoxia, normoxia plus inhaled NO (20 ppm), hypoxia (F|0 2 = 0.14), and hypoxia plus inhaled NO. No significant differences were detected between normoxic conditions with or without inhaled NO, or between hypoxic conditions with or without inhaled NO, for measures of Pa0 2 , Sa0 2 , or (ideal) A-aD0 2 (p>0.05) (Sheel ef al., in press). The apparent lack of response to inhaled NO at 20 ppm is consistent with data from exercising sheep (Koizumi et al., 1994), but is 94 not consistent with other studies in humans (Roger et al., 1997). Moreover, failure to alter gas exchange under hypoxic conditions with administration of NO was unexpected. While failure to demonstrate an effect of inhaled NO during heavy exercise might suggest further vasodilation is not possible due to maximal vascular recruitment and distention, the lack of an effect at rest is difficult to explain. 1.7.2.2 Diffusion limitation Determinants of alveolar-capillary diffusion, defined by Fick's law, include the surface area available for diffusion (varies with pulmonary blood volume), diffusion distance, red cell capillary transit time, and the equilibration rate of mixed venous blood and alveolar gas. The presence of alveolar-end-capillary diffusion limitation during heavy exercise is suggested by MIGET data of VVQ inequality that only accounts for up to 1/3 of the calculated (ideal) A-aD02 (Torre-Bueno ef al., 1985). Because inert gases approach equilibrium more quickly than oxygen, any difference between observed and predicted A-aD02 would be attributable to diffusion limitation plus or minus a post-pulmonary shunt. Direct measurement of diffusion limitation was not possible, and the argument for diffusion limitation was made instead based on the estimate that an extrapulmonary shunt would have to be 10-20% of the cardiac output to explain measured data during hypoxia (Torre-Bueno etal., 1985; Wagner, 1992). Increases in interstitial lung water would be expected to increase diffusion distance and the potential for this mechanism of diffusion limitation during exercise has been discussed. A reduction in Vc would also predict a decrease in diffusion and could contribute to diffusion limitation in subjects with smaller lungs and vascular cross-sectional area. A significant negative correlation between perfusion distribution (logSDo.) at rest and at 90% V0 2 m ax, and total lung capacity corrected for body surface area (p=0.07 and p<0.01, respectively) has been shown by Hopkins ef al. (1998). The equilibration rate of mixed venous blood and alveolar gas is not considered to be a limiting factor. In theory, any increase in pulmonary blood flow through a fully dilated and recruited vascular bed could reduce red blood cell capillary transit times (Dempsey et al., 1982). Warren et al. (1991) inferred mean pulmonary transit time during exercise by dividing Vc, as determined by single breath diffusion of carbon monoxide, by Q, determined by echocardiography, but failed to demonstrate a reduction in mean transit time or a 95 significant correlation between mean transit time and the (ideal) A-aD0 2 (r^O.OSO. However, maximal work levels were not achieved and 14 of 16 high aerobic power male athletes (V02max= 67.4 mL-kg"1-min"1) maintained Pa0 2 > 10.7 kPa (80 Torr). In contrast, Hopkins et al. (1996), using first-pass radiocardiography and deconvolution analysis in 10 athletes who had previously demonstrated evidence of a diffusion limitation by MIGET (Hopkins et al., 1994), reported a significant negative correlation between the observed and predicted A-aD0 2 (mean±SD, 1.3±1.3 kPa (10±10 Torr)) and whole lung transit time (2.91 ±0.30 seconds) (r=-0.58, p<0.05). Over 40% of whole lung transit times were less than 2 seconds during peak exercise (V0 2 = 5.13±0.50 L-min"1). Assuming the pulmonary capillary volume was approximately 2.9mL-kg"1 as per Gehr et al. (1978), they estimated the corresponding capillary transit times would be less than 0.3 seconds. While probable that some capillary transit times fell below the estimated 0.25 seconds required for complete saturation to occur (Wagner, 1977), minimal Pa0 2 during maximal exercise only reached 12.5±1.1 kPa (94.1 ±8.0 Torr). Hence, to date the potential relationship between pulmonary capillary transit times, diffusion limitation and EIAH remains incompletely addressed. 1.8 Future directions The literature on EIAH is marked by conflicting results, varying exercise protocols, differences in definitions, methodologies, and theories, and struggles under attempts to find a single defining physiological mechanism. The recent data of Rice et al. (1999a), limitations aside, serve to re-emphasize the multifactorial and heterogeneous nature of EIAH. Indeed, in a word, EIAH is perhaps best described by heterogeneity: heterogeneity of ventilatory response, VVQ relationships and pulmonary perfusion. Although an explanation for intersubject differences in gas exchange, and particularly the onset of gas exchange impairment during submaximal exercise in some habitually active subjects, awaits further study, the role of the pulmonary circulation, dominant in both VVQ inequality and diffusion limitation, stands at the forefront. Most pressing is the potential that chronic, recurrent intensive exercise may induce adverse structural or functional alterations. As such, data on potential differences in pulmonary capillary volume, the hemodynamic response of the pulmonary vascular bed to the demands of increased flow and pressure, and the potential role of the inflammatory response, via airway or vascular mediators, is awaited. 96 1.9 Conclusion Exercise-induced arterial hypoxemia occurs in the habitually active and describes a limitation in pulmonary gas exchange comprised of a combination of relative hypoventilation, ventilation-perfusion inequality and diffusion limitation. The underlying etiology of these contributing mechanisms is incompletely understood. Demonstration of a greater (ideal) alveolar/arterial O2 difference and accompanying impairment of oxygen homeostasis in habitually active subjects with EIAH, at an equivalent level of oxygen consumption compared to subjects with similar and substantially less aerobic power, suggests gas exchange is functionally and/or structurally compromised. Whether or not chronic recurrent intensive exercise induces alteration of the blood-gas interface remains to be shown. 97 Appendix II. Review of the Literature: Exercise and the Pulmonary Circulation 2.1 Introduction In series with the systemic circulation, and recipient of the entire cardiac output, the pulmonary circulation is situated to impact oxygen delivery via its central role in gas exchange and left ventricular preload. More than a passive low-pressure fluid conduit, the vascular bed of the lung responds to a number of biomechanical and biochemical stimuli, and similarly is susceptible to mechanical and inflammatory injury. Its endothelium comprises the largest endothelial surface in the body and functions at the critical interface between environment and internal milieu. Serving as barrier, filter and sophisticated regulatory surface, it is capable of transforming humoral signals, amplifying and/or initiating inflammatory and hemostatic reactions, modulating vascular tone, and playing a primary role in vascular injury, repair and remodeling. Under conditions of increased blood flow, pulmonary endothelial dysfunction, secondary to a complex of elevated perfusion pressure, inflammation, and/or oxidative stress, could alter vascular tone, impair blood-gas membrane integrity, permit interstitial edema, accentuate ventilation-perfusion mismatch, and promote development of arterial hypoxemia. Pulmonary endothelial activation/injury and/or vascular remodeling may result from chronic, recurrent, intermittent increases in lung blood flow, and this potentially could account for the gas exchange impairment seen in some athletes at equivalent submaximal levels of oxygen consumption compared to subjects with both similar and substantially less aerobic power. 2.2 Structure and function Ventilation-perfusion matching and gas diffusion across the blood-gas interface are the primary mechanisms of impaired gas exchange efficiency during exercise; these mechanisms of exercise-induced arterial hypoxemia have been discussed in detail in Appendix I. Integration of ventilation-perfusion (VVQ) matching and diffusion, in turn, depends upon airway and vascular structure, blood flow distribution, maintenance of blood-gas barrier integrity and preservation of a dry interstitial space. 2.2.1 Embryology Pulmonary vessels develop alongside the bronchial tree to allow for an intimate interface between air and blood. Arising from the paired sixth aortic arch during the fifth week of 98 embryogenesis, and separated from the aorta by the spiral aorticopulmonary septum, the pulmonary artery grows toward the developing lung bud. The lung, on the other hand, develops from the foregut, shares its venous drainage through the splanchnic plexus, and later meets up with an outpouching of the heart in the form of the common pulmonary vein. Endodermally derived cells of the branching airway provide directional cues for the advancing arteries and promote development of mesodermal angioblasts that become incorporated into the pulmonary vasculature (Buck et al., 1996). Molecular events in endothelial cell (EC) differentiation from mesoderm remain uncertain, but based on mouse embryo studies, initiation of early vascular tree development requires interaction between epithelial vascular EC growth factors (VEGF) and vascular EC growth factor receptors (VEGFR) (Beck and D'Amore, 1997). Cell-matrix and cell-cell interactions permit vessel formation and are mediated by various cell adhesion molecules (CAMs) including integrins and cadherins (Cines et al., 1998). Formation of larger vessels involves smooth muscle cell (SMC) precusor cell recruitment and replication. Understanding of these early processes of angiogenesis offers insight into vascular remodeling and repair as discussed below. Branching morphogenesis of bronchioles with a corresponding increase in vascularity marks further development in the canalicular phase between 15 and 25 weeks of gestation. By the seventh month an adequate number of capillaries have formed to allow for gas exchange (Sadler, 1985). After birth, resting at sea level, the pulmonary circulation operates as a low-pressure, low-resistance link between the right and left ventricles at a perfusion pressure approximately one-sixth that of the systemic circulation. It accepts the entire blood volume, serves the vital function of gas exchange, and generates, activates or inactivates biological compounds that influence vascular tone and permeability of both the pulmonary and systemic vessels. In service of these functions, several key structural differences in conducting vessels and capillary networks exist between halves of the cardiopulmonary circuit. 2.2.2 Conducting vessels Unlike their systemic counterparts, the pulmonary arteries are thin, comprised of a minimal medial layer sandwiched between internal and external elastic laminae. The transition from elastic to muscular arteries occurs much later and is not seen until arteries are less than 1mm in diameter (Brenner, 1935; Wagenvoort and Wagenvoort, 1977). The media of 9 9 the pulmonary artery is also more extensible than that of the aorta, a factor explained in part by differences in the relative proportions of collagen and elastin (Harris et al., 1965). These features allow for greater distention of the lung arterial circulation without a substantial rise in pressure. Weibel and Gil (1977) have suggested that all connective tissue compartments in the lung are part of a fibrous continuum from the hilum to the pleura within which bronchi and extraalveolar blood vessels are embedded. Such perivascular connective tissue cuffing allows extraalveolar and corner vessels to be relatively protected from perturbations in alveolar pressure. This interstitial cuff also provides a reservoir for collection of extravascular lung water until drainage by the lymphatics proves sufficient. As lung volumes increase and close alveolar vessels, increasingly subatmospheric interstitial pressures tether open extraalveolar vessels and permit continued flow (Howell ef al., 1961). Pulmonary vascular resistance (PVR) increases at high lung volumes while vascular capacitance is maximal at lung volumes close to functional residual capacity (FRC). However, increasing pulmonary driving pressures with rising cardiac output (Q) during exercise may mediate a decrease in PVR and help to maximize vascular capacitance (Reeves et al., 1988). These hemodynamic affects likely overcome any changes based on lung volume. This conclusion is further supported by failure of short term loading and unloading of pleural pressure to impact on alveolar oxygen consumption (VO2) during mild exercise (Giesbrecht ef al., 1991). Nevertheless, large swings in pleural pressure during near maximal and maximal exercise may promote overperfusion of extraalveolar vessels. 2.2.3 Tfje microvasculature The dense hexagonal alveolar capillary network of the lung differs from the systemic circulation, likened more to an underground-parking garage than the spreading branches of a tree. Capillary networks serve more than one alveolus and allow for varying degrees of venous admixture. Flow through this network has been described as tubular (Weibel, 1963) or sheet-like (Fung and Sobin, 1977), with anatomical evidence compatible with either option largely dependent on the fixation method. Microvascular flow is affected by the interrelationship of three pressures - arterial inflow pressure, venous return pressure and alveolar pressure - the relative values of which define three classical lung zones. In Zone 1 alveolar pressure exceeds arterial and venous pressures, in Zone 2 alveolar 100 pressure lies between arterial and venous pressures, and in Zone 3 alveolar pressure is less than arterial and venous pressures. Flow in Zone 1 is confined to corner vessels, in Zone 2 capillary flow more closely approximates Fung and Sobin's sheet flow theory, while in Zone 3 capillary flow is more tubular. A rise in perfusion pressure during exercise increases the portion of the lung described by Zone 3 conditions and initially enhances V A / Q matching. Increased V A / Q homogeneity is supported by topographical measures of V V Q distribution at low intensity exercise (Bake et al., 1968). However, as exercise intensity increases, considerable intersubject differences are noted (Wagner et al., 1986), and dispersion of V A / Q ratios contributes up to 60% of the elevated (ideal) alveolar/arterial oxygen gradient seen during exercise above VO2 > 3 L-min"1 (Hopkins et al., 1994). The etiology of this variable intersubject V A / Q ratio dispersion remains an unsolved question, but structural or functional blood flow distribution heterogeneity and/or interstitial edema find indirect support from animal and human data. 2.2.4 Blood flow distribution 2.2.4.1 Structural heterogeneity Pulmonary hydrostatic pressures increase from apex to base under influence of gravity. This gradient promotes pulmonary blood flow distribution heterogeneity with preference for basal regions. Heterogeneity of blood flow distribution is further enhanced by structure of the vascular bed. From post-mortem casts, branching of the human pulmonary arterial tree has been compared to the tributaries of a river system, with 17 orders, a branching ratio of 3.0, a diameter ratio of 1.6, and a length ratio of 1.5 (Singhal ef al., 1973). Like a river system, the geometric branching of the pulmonary vessels can be described mathematically by fractal geometry. Regional blood flow heterogeneity exceeds predicted variability based on gravity alone (Glenny ef al., 1991; Prisk et al., 1994; Hlastala et al., 1996) and can be characterized more completely by fractal methods (Glenny and Robertson, 1990; Glenny and Robertson, 1991). Studies utilizing 15 pm fluorescent microspheres have demonstrated stability of high flow and low flow areas over time in dogs (n=5, Glenny et al., 1997), at rest and during maximal exercise in thoroughbred racehorses (n=4, Bernard et al., 1996), and, in a preliminary study, with changes in posture in baboons (n=2, Glenny et al., 1997). In the latter, a strong association was demonstrated between blood flow and height up the lung in the upright posture (^=0.65), whereas in the head-down position, height up the lung accounted for only 3% of blood flow variability. 101 Studies comparing V A / Q relationships between prone and supine postures in dogs and pigs have similarly demonstrated perfusion heterogeneity that is incompletely explained by gravitational gradients. Using positron emission tomography in dogs to image local distributions of VA and Q per unit of gas volume content (sVA and sQ, respectively), Treppo et al. (1997) detected fairly uniform spatial distributions of SVA, S Q , and V A / Q in the prone position. However, these variables were more heterogeneous in the supine position, in part, because of gravitational gradients. They argued that gravitational and structural influences effectively cancelled each other out in the prone position, while the effects were additive in the supine position. More recently, Mure ef al. (2000) used fluorescent-labeled aerosols and radioactive-labeled microspheres in anaesthetized, mechanically ventilated pigs to measure regional VA and Q, respectively. In so doing, they demonstrated increased VA homogeneity (p=0.03) and V A / Q correlation (correlation coeff ic ients .82±0.06) in the prone position to conclude that improvement of V A / Q matching in the prone position is due to enhancement of VA distribution. Together, these data support the hypothesis that regional pulmonary perfusion is determined primarily by a fixed structure and suggest preference of blood flow for basal regions is based more on vascular anatomy than hydrostatic forces. 2.2.4.2 Functional heterogeneity Local factors and perfusion pressures may also influence variability of blood flow distribution. Phenotypic heterogeneity for nitric oxide (NO) production has been demonstrated between microvascular and macrovascular endothelial cells in rats (Geiger ef al., 1997). Regional heterogeneity of NO release has also been demonstrated in thoroughbred racehorses, with preference for caudodorsal vessels (Pelletier ef al., 1998). This latter observation is in keeping with previously documented preferential distribution of pulmonary blood flow to dorsal lung regions in exercising horses (Bernard ef al., 1996). In addition, using flourescent-labelled aerosol and radioactive-labelled microspheres in standing sheep, and blocking endogenous NO production with L-NAME in the presence or absence of exogenous NO (30 ppm), Melsom ef al. (2000) noted no alteration in perfusion along horizontal levels when varying access to NO, but did demonstrate increased homogeneity of blood flow along the gravitational axis when NO availability was reduced. 102 Delivery of NO restored the vertical perfusion distribution, but variable access to NO in this animal model had no affect on arterial oxygenation. Hence, the pulmonary vascular tree appears to demonstrate structural (fixed) and functional (variable) heterogeneity. Regional perfusion, in turn, is determined by the combination of their relative contributions. Furthermore, the heterogeneity of perfusion, neither uniform nor random, does not appear to change substantially with exercise: high flow areas remain high flow and low flow areas remain low flow (Bernard ef al., 1996). Increased blood flow and pressure during near maximal and maximal exercise may overcome local redistribution factors leaving flow distribution dependent on underlying structural heterogeneity. If these relationships are also applicable to the human pulmonary circulation, excessive perfusion of relatively poorly ventilated areas, that is, alveolar units with low V A / Q ratios, could accentuate venous admixture and contribute to development of hypoxemia. 2.2.5 Pulmonary blood volume and transit times The pulmonary circulation has an enormous capacity to accommodate increasing blood flow through distention and recruitment of pulmonary capillaries. Human pulmonary capillary blood volume (Vc) has been estimated by post-mortem morphometric measures (Gehr ef al., 1978) and diffusion capacity for carbon monoxide (Roughton and Forster, 1957). Functional measures suggest Vc approximates 100-150 mL whereas anatomical measures estimate Vc at 2.9 mLkg"1 (Gehr ef al, 1978). In dogs, at rest, longer apical than basal capillary transit times have been demonstrated, with a skewed frequency distribution of pulmonary transit times towards the right (Hogg ef al., 1985; Wagner ef al., 1986). A vertical gradient has also been demonstrated in humans (MacNee ef al., 1989). The relative sparing of apical capillaries serves to create a reserve that can be recruited by increasing blood flow and perfusion pressure during exercise. As flow increases, pulmonary red cell transit times are dependent on sufficient expansion of the vascular bed. Any additional increase in Q through a fully dilated and recruited lung vascular bed reduces mean pulmonary capillary red cell transit time to produce a more homogeneous and left shifted transit time frequency distribution. This relationship has been demonstrated using videomicroscopy of canine lungs (Presson et al., 1994). Human pulmonary transit time data, however, remain limited by whole lung measures. 103 2.2.6 Lymphatics The pulmonary lymphatics are arranged in superficial and deep networks. Superficial lymphatics are found within the pleura, whereas deep lymphatics are located around bronchi, arteries and veins, and within the connective tissue. These two systems connect at the pleura and the hilum and drain into the tracheobronchial nodes. Lymph flow in unanaesthetized sheep at rest has been measured at 10 mL-hr"1 (Staub, 1971). During exercise, lymph flow increased 5 to 7 times (Coates ef al., 1984; O'Brodovich ef al., 1986) and fell to baseline levels immediately upon exercise cessation. Lymph-to-plasma protein ratios also decreased, which further suggests vascular permeability was not altered. Human lymph flow during exercise is at best an estimate. 2.2.7 The interstitium The perivascular interstitium forms a cuff around extra-alveolar blood vessels and bronchi and provides a temporary storage area for extravascular lung water (Staub ef al., 1967). The forces affecting fluid flux between the pulmonary microcirculation and the interstitium are described in Starling's (1896) equation - JMV = KFC [(PMV-PINT) - ofaMv-rciNT)] ~~ w n e r e JMV represents the net microvascular filtration, KFC is the microvascular filtration coefficient, a is the osmotic pressure of all plasma proteins, PMV and PINT are the hydrostatic pressures of the microvasculature and interstitium, respectively, and 7TMV and rciNT are osmotic pressure of the microvasculature and interstitium, respectively. Thus, net microvascular filtration reflects a balance between filtration and absorptive pressures. Edema accumulation is opposed by the osmotic pressure gradient and lymph flow, and under resting conditions, the balance lies in favor of absorption. However, during exercise, increasing pulmonary artery pressure (PAP) and pulmonary artery occlusion pressure ( P A O P ) increase microvascular driving pressures (see section 2.5). This is partially offset by a fall in P V R , but the potential for interstitial edema formation and capillary stress failure exists (see sections 2.4.7 and 2.5). 2.2.8 The blood-gas barrier Central to gas exchange, the blood-gas barrier (BGB) must be thin to allow for efficient gas transfer by diffusion according to Fick's law. Between 0.2-0.4 um thick over half its area (Gehr ef al., 1978), the BGB is comprised at times of only epithelial and endothelial cells 104 on either side of a common basement membrane. Strength of the BGB is primarily provided by type IV collagen in the lamina densa at the center of the extracellular matrix (ECM). On opposing sides of the ECM, epithelial intercellular gap junctions are ten fold tighter than those between endothelial cells, leaving the endothelial layer the most permeable. Integrins, immunoglobulins and cadherins coordinate at endothelial cell-cell junctions to regulate vascular permeability and leukocyte extravasation (Dejana and Plantier, 1996). Electron micrographs of the BGB in exercising rats compared to sedentary controls suggest exercise may thin the BGB by reducing thickness of the endothelium (Miyachi and Yano, 1996). Mechanical shear stress and pressure has similarly been demonstrated to alter endothelial cell shape through flow-mediated mechanotransduction (Davies, 1995). 2.3 The pulmonary endothelium Forming the largest endothelial surface in the body, the pulmonary endothelium plays a major role in modulating vascular tone, permeability, inflammation and coagulation. The lung is a common site of injury in systemic inflammatory processes and pulmonary endothelial injury is an early and significant factor. The rise in inflammatory signals during exercise, discussed in section 2.3.3, may promote or mark one end of a spectrum of pulmonary endothelial dysfunction. 2.3.7 Vasomotor tone The pulmonary circulation is a uniquely low-pressure, high volume circuit that is highly sensitive to oxygen tension. In this, it responds in opposite to the systemic circulation when exposed to arterial hypoxemia. Whereas systemic arteries vasodilate, pulmonary vessels constrict. Hypoxic pulmonary vasoconstriction (HPV) is mediated through increases in intracellular Ca 2 + , although the intermediary stages of this response remain uncertain. The effect of HPV serves to preserve V A/Q equality by redistribution of blood flow away from hypoxic alveoli. Vessel tone is also affected by various mediators, including nitric oxide, prostacyclin and endothelin-1. 2.3 .7 .7 Nitric oxide Nitric oxide (NO), an endogenous reactive gas synthesized from L-arginine by a family of flavoproteins known as NO synthases (NOS), induces pulmonary vasodilation. 105 Constitutive NOS (cNOS) is present in pulmonary EC and produces NO in response to changes in C a 2 + and calmodulin. Shear stress can also activate cNOS via Ca2+-dependent potassium channels. NO diffuses from the EC to the SMC to act on guanylate cyclase, converting GTP to cGMP, which then activates G kinase. G kinase, in turn, effects vasorelaxation by modifying cytosolic C a 2 + ([Ca2+]|) and myosin activity. The apparent difference between pulmonary and systemic flow-induced endothelial-dependent vasorelaxation may partially be explained by a differential response of endothelial NO production. Koizumi et al. (1994) measured blood flow and pressure in the pulmonary circulation of sheep at rest and during treadmill exercise with a NOS agonist or antagonist, or a combination of both. They failed to demonstrate an enhancement of basal NO vasodilation during exercise, and suggested the reduction in PVR with onset of exercise would be more likely related to an increase in vascular pressure. In neonatal swine, Vitvitsky et al. (1998) evaluated endothelial-dependent (acetylcholine (ACH) infusion) and endothelial-independent (inhaled NO) pulmonary vasodilation following left pulmonary artery ligation and demonstrated loss of endothelial-dependent vasodilation without morphological evidence of vascular remodeling. They concluded that increased pulmonary blood flow in the immature swine produced endothelial dysfunction before the onset of remodeling and suggested endothelial dysfunction may be one of the earliest responses to increased pulmonary blood flow. This supports an earlier study in humans by Celemajer ef al. (1993) who demonstrated diminished flow velocity to ACH with maintained response to nitroprusside in patients with congenital heart disease and increased pulmonary blood flow but no established pulmonary vascular disease. Unfortunately, more chronic increases in pulmonary blood flow in congenital heart disease with left-to-right shunts often leads to vascular remodelling, characterized by subintimal smooth muscle hyperplasia and medial thickening, and irreversible pulmonary hypertension if surgical correction is not performed early (Hall and Haworth, 1992). These studies on pulmonary vessels lie in contrast to the response of systemic arteries to increased flow. Kamiya and Togawa (1980) demonstrated vessel dilation after creating an arteriovenous (A-V) fistula between the carotid artery and jugular vein in dogs, and Miller ef al. (1986) showed enhanced vessel relaxation in response to ACH in isolated arterial 106 rings from dogs with femoral A-V fistulas. NaDuad ef al. (1996) subsequently detected increased expression of NOS in systemic arteries subjected to increased flow, while Tronc et al. (1996) showed decreased carotid artery vessel dilation in an A-V fistula model when NO production was blocked. Thus, it seems elevated blood flow induces NO upregulation in the systemic circulation, but may instead produce endothelial dysfunction in the pulmonary circulation. Attempts to non-invasively measure NO production by the pulmonary circulation have been unsuccessful. Although NO has been measured in expired air (VNO) (Bauer et al., 1994; Maroun et al., 1995; Persson ef al., 1993; Chirpaz-Oddou et al. 1997), its origin is uncertain. Chirpaz-Oddou et al. (1997) continuously measured the concentration of NO in exhaled air (CNO) and demonstrated a significant reduction if exercise intensity exceeded 65% V02max- They also showed significant correlation between VNO, calculated from the product of CNO and minute ventilation (VE), and VO2, VCO2, V E , heart rate, and V02 p eak, leading them to conclude that VNO is related to the magnitude of aerobic metabolism. These data agree with Maroun ef al. (1995), who demonstrated significantly higher VNO in high aerobic power athletes (V02max >60 mlkg"1min"1) during exercise compared with controls. In addition, Riley et al. (1997) showed a significantly diminished exercise VNO response in patients with pulmonary hypertension or pulmonary fibrosis compared to healthy controls. Since endothelial NOS expression is decreased in the pulmonary arteries of patients with pulmonary hypertension (Giaid and Saleh, 1995), some component of the exercise VNO response may have been related to the pulmonary vasculature. However, NO is produced in the upper airway, the lower respiratory tract, and both the pulmonary epithelium and endothelium. Although nasal NO levels fall during heavy exercise (Lundberg et al., 1997), the considerable intersubject variability of CNO in exhaled air during exercise (Bauer ef al., 1994; Iwamoto ef al., 1994; Matsumoto et al., 1994; Persson et al., 1994; Maroun ef al., 1995) may be an artifact of the expiratory flow rate (St. Croix ef al., 1999). Furthermore, intravenous administration of nitroglycerin, sodium nitroprusside, or placebo to 24 healthy subjects in a randomized, double blind cross-over trial failed to demonstrate any significant differences in end expiratory CNO (p=0.328) between the treatment conditions despite significant changes in heart rate and blood pressure (p<0.01) (Dirnberger et al., 1998). Thus, the specificity of this method is 107 insufficient to reliably detect changes in pulmonary vascular NO. Nonetheless, and irrespective of the source of VNO, whole body NO turnover, as measured by NO metabolites, appears to be increased in subjects with greater aerobic power (Jugensten ef al., 1997). However, as with VNO, systemic measures fail to isolate the contribution of the pulmonary vascular bed. 2.3.12 Prostacyclin Prostacyclin, or prostaglandin I2 (PGI2) is an inferior vasodilator compared to NO, but is the most potent inhibitor of platelet aggregation (Moncada et al., 1976). Like NO, prostacyclin is released from the endothelium in response to shear stress, but its effects are longer (t-1/2 ~3min) and more subdued. Moreover, it is neither selective for the pulmonary circulation, nor is it inactivated by transit through the vascular bed of the lung. Formed by prostacyclin synthase from products of the arachadonic acid cascade, PGI2 acts on vascular SMC (VSMC) via increases in cAMP and protein kinase modulation of actin-myosin coupling. In this way it is synergistic with NO. 2.3.1.3 Endothelin-1 In contrast, endothelin-1 (ET-1) is a potent EC derived vasoconstrictor. NO has some effect on shear stress-induced ET-1 mRNA downregulation as suggested by reversal with /V-nitro-L-arginine, a competitive inhibitor of L-arginine (Kuchan and Fragos, 1994). Acting either through voltage-dependant C a 2 + channels or specific VSMC receptors, ET-1 provokes a rapid rise in [Ca2+], and, via phospholipase C, the phosphoinositol pathway. Interest in ET-1 as a modifier of vascular tone in the setting of acute lung injury (ALI) is supported by finding elevated plasma levels in patients with sepsis, systemic inflammation and the acute respiratory distress syndrome (ARDS) (Pittet ef al., 1991; Weitzberg ef al., 1991;Druml etal., 1993). 2.3.1.4 Influence of sex hormones 17(3-estradiol (E2) demonstrates a beneficial effect on systemic arterial endothelial-dependent and independent vasorelaxation (Arora ef al., 1998), which appears to be antagonized by progesterone (English ef al., 1998). E2 has been shown to exhibit differential effects on NOS, up-regulating endothelial NOS (eNOS) in cultured EC (Hayashi ef al., 1995), but down-regulating inducible NOS (iNOS) in rat aorta (Duckies et al., 1996). Heterogeneity has also been seen in estrogen modulation of ET-1 induced 108 vasoconstriction in dog coronary conduit or microvessels depending on vessel size (Lamping and Nuno, 1996). Furthermore, significant between-gender effects have been demonstrated for whole body measures of NO even if normalized for body weight or surface area. Exhaled NO and plasma nitrate (NO3-), a stable NO end product, were significantly lower in women than men (p<0.001 and p<0.0001, respectively), but did not vary significantly across the menstrual cycle (Jilma et al., 1996). Plasma NO metabolites did, however, display a different response to exercise depending on the menstrual phase. The percent increase in nitrite to nitrate (NO2-/NO3) was greatest after moderate exercise at the mid-follicular phase, but was significantly less (p<0.05) following similar exercise intensity at the mid-luteal phase, or after maximal exercise at either phase (Wang ef al., 1997). The effects of 17B-estradiol on the adult human pulmonary circulation are not well understood, but data from animal models (sheep and rat) suggest it may influence hypoxic vasonstriction (Wetzel ef al., 1984; Gordon ef al., 1986) and pulmonary vascular remodelling (Farhat ef al., 1993). 2.3.2 Permeability Endothelial cytosolic C a 2 + concentration and cAMP produce opposing effects on endothelial paracellular gaps and barrier permeability. Elevations of [Ca2+]j induce activation of Ca2+/calmodulin-dependent myosin light chain kinase, F-actin reorganization, and decrease cell-cell and cell-ECM interaction, any of which may lead to formation of EC gaps and enhanced permeability. However, while this effect of [Ca2+]j on EC permeability has been seen in preparations of rat pulmonary artery ECs (RPAEC), in contrast, increased extracellular C a 2 + did not enhance [Ca2+]i in rat pulmonary microvascular ECs (MVECs), nor did it increase permeability (Kelly ef al., 1998). This differential C a 2 + regulation is in keeping with comparative studies of bovine pulmonary MVECs which demonstrated less permeability to macromolecules than either pulmonary artery or vein ECs (Schnitzer ef al., 1994; Stevens ef al., 1994). Direct activation of mechanosensitive Ca2+-permeable channels and indirect activation of mechanosensitive K+-channels by vascular shear stress can alter endothelial cell permeability by increasing C a 2 + entry and influencing cAMP levels. C a 2 + entry can also increase through agonist-activated nonselective cation channels via ATP, ET-1, 109 bradykinin, histamine, platelet activating factor, serotonin, substance P and thrombin (Nilius era/., 1997). Several in vitro studies lend support to induction of proinflammatory responses by mechanically stretched endothelial cells, via increases in [Ca2+]i, endothelin-1, cytokines, superoxide production, activation of tyrosine kinases and transcription factors. These responses may underlie inflammation or permeability increase in pressure-induced lung disease. Furthermore, stretch-induced effects of pressure may lead to Ca2+"dependent proinflammatory events such as the expression of the neutrophil-trafficking receptor P-selectin. The CAM P-selectin (CD62P) is stored in EC Weibel-Palade bodies and is expressed by Ca2+-dependent exocytosis. Expression of P-selectin is undetectable in resting blood vessels, but is enhanced by several injury stimuli. Its expression, therefore, has been used as a marker of early inflammation. Using the styryl dye FM1-43, which fluoresces brightly upon binding cell membranes, Kuebler et al. (1999) determined FM1-43 fluorescence in conjunction with intravital quantifications of P-selectin expression and EC [Ca2+]i in venular capillaries of isolated, blood perused rat lung. FM1-43 colocalized with P-selectin and was inhibited by blockade of mechanogated Ca2+-channels. These data support a role for pressure-induced alterations in permeability via Ca2+mediated processes, and potentially by initiating CAM expression, neutrophil adhesion and inflammation. 2.3.3 Inflammation and the inflammatory response to exercise Inflammation is a local tissue or systemic response to injury capable of repair or accentuation of tissue damage. It is initiated by a progression of leukocyte-endothelial interactions - rolling, sticking and transendothelial migration - that results from interaction of shear and adhesive forces. The latter depend upon two categories of leukocyte and endothelial adhesion receptor-ligand pairs: leukocyte and endothelial selectins that bind to their specific carbohydrate counter-structures, and leukocyte integrins that bind to endothelial immunoglobulins. The selectins (leukocyte: L-selectin and sialylated Lewis X; endothelial: E- and P-selectin) and their carbohydrate counter-structures mediate leukocyte rolling, while sticking and diapedesis depend on the interaction between leukocyte integrins (CD11a/CD18, CD11b/CD18, CD11c/CD18 and VLA-4) and 110 endothelial members of the immunoglobulin superfamily (IGSF) (intracellular adhesion molecule 1 and 2 - ICAM-1 and -2 - and vascular cell adhesion molecule - VCAM-1). Each of these adhesion molecule interactions in turn is regulated by various activating stimuli including cytokines, platelet activating factor, leukotrienes, complement components and oxidants (see Ward, 1997 for review). Strenuous exercise incites an acute inflammatory response, marked by leukocytosis and neutrophil activation, release of inflammatory mediators and acute phase proteins, and activation of complement, coagulation and fibrinolytic cascades, that is akin to, but quantitatively different from the acute phase and systemic inflammatory response accompanying sepsis, major burns and trauma (Weight ef al., 1991; Camus et al., 1994; Northoff et al., 1994; Moyna ef al., 1996; Pedersen ef al., 1997; Skek and Shephard, 1998). All subtypes of natural killer (NK), B, and T lymphocytes are recruited, and neutrophils and monocytes increase in numbers. An extensive arsenal of inflammatory mediators, including eicosanoids, reactive oxygen species (ROS), cytokines and vasoactive amines, are elevated by exercise (Cannon ef al., 1986; Sprenger et al, 1992; Anselme ef al., 1994; Camus ef al., 1994; Northoff ef al., 1994; Weinstock et al., 1997; Pedersen, ef al., 1997), although the physiological and clinical significance of these alterations remains undetermined (Smith, 1997). 2.3.3.1 Leukocyte traffic Adherence of leukocytes, and particularly neutrophils, is central to both immune defense and the inflammatory response. Rolling and sticking of cells in pulmonary arterioles and venules and retention in alveolar capillaries, in turn, is dependent on lung inflation, endothelial adhesion molecules and hydrodynamic forces. Increased Q and PAP produce a significant increase in microvascular blood flow and a reduction in leukocyte -endothelial interaction (Kuhnle ef al., 1995). However, transit times of fluorescently labelled neutrophils in pulmonary capillaries fell when Lien ef al. (1990) administered epinephrine or hypoxic conditions in a dog lung model. Both conditions increased PAP, but Q was only increased with epinephrine. Subsequent balloon inflation in the vena cava reduced Q by 40% without altering PAP and produced an increase in neutrophil transit time. Exercise studies in humans have been conflicting (Muir ef al., 1984; Peters ef al., 1992), but the discrepancies suggest factors other than local flow velocity may influence neutrophil transit in the pulmonary microcirculation (MacNee and Shelby, 1993). I l l Growth hormone and catecholamines mediate the immediate increase in white blood cell lines (Kappel ef al., 1991; Kappel ef al., 1993), while Cortisol likely contributes to the neutrophilia and lymphopenia after prolonged, intense exercise (Kjaer, 1989; Pedersen ef al., 1997). Neutrophils are primed via neuroendocrine hormone and cytokine release (Smith ef al., 1996; Smith, 1997), but this can be abolished by the antioxidant /V-acetylcysteine (NAC) (Huupponen ef al., 1995) or dexamethasone (Smits ef al., 1998). Increased atrial natriuretic factor (ANF) with exercise (Freund ef al., 1991) may prime neutrophil superoxide (O2) release (Wiedermann ef al., 1992) and contribute to oxidant damage. 2.3.3.2 Cytokines Muscle and connective tissue damage from eccentric work (Camus ef al., 1992), rather than the rise in stress hormones (Pedersen ef al., 1997), primes monocytes and macrophages, which, in turn, release pleiotropic cytokine immunoregulatory proteins to help promote local repair. Regulation by anti-inflammatory cytokines and receptor antagonists maintains the protective effect of these immune proteins (Bone, 1996), but failure of homeostasis disturbs the balance, allowing proinflammatory cytokines such as tumor necrosis factor a (TNFa), interleukin 1p (IL-ip), IL-6 and IL-8, to inflict tissue injury directly or through activation of neutrophils. Although variability exists, significant increases in plasma levels of IL-ip, IL-2 receptor (IL-2R), IL-6 and TNF-a have been detected immediately after prolonged strenuous exercise (Cannon ef al., 1983; Cannon et al., 1986; Dufaux ef al., 1989; Espersen ef al., 1990; Northoff and Berg, 1991; Weight ef al., 1991; Sprenger ef al., 1992; Tilz ef al., 1993; Bury ef al., 1996). Conflicting results may partially be explained by differences in exercise protocols (concentric versus eccentric exercise), intensity of exercise, time of day and possibly subject fitness. Leukocyte TNF-a gene expression decreased (p<0.008) following a maximal walking treadmill exercise test in 7 moderately fit men, but no difference was seen for IL-1-a, IL-ip, or interferon y (IFN-y) (Natelson ef al., 1996). Alternately, significant elevations in urine of IL-ip, IL-2R, IL-6, TNF-a and IFN-y have been shown following a 20-kilometre run (Sprenger ef al., 1992). Urinary cytokines, although more unstable, likely reflect enhanced cytokine turnover and may prove more sensitive markers of exercise-induced immune modulation (Northoff ef al., 1994). 112 Hypothalamic-pituitary-adrenal (HPA) axis activation, under circadian, exercise and/or pharmacologic induction of Cortisol, acts as a physiological restraint on cytokine production. Glucocorticoids demonstrate a differential effect on TNF-a, IL-ip, and IL-6, with a hierarchy of sensitivity greatest for TNF-a and least with IL-6 (DeRijk ef al., 1997). These influences on cytokine production may interfere with exercise measures. 2.3.3.4 Cell adhesion molecules Cell adhesion molecules (CAMs) mediate interactions between cells, and between cells and the ECM. They are central to neutrophil-endothelial interactions and thereby to the inflammatory response. Soluble ICAM-1 (CD54) was elevated by 20% following an endurance training program in middle-distance runners (Baum et al., 1994), and by 23% after a roundtrip ascent and descent from 750m to 2350m over a five hour period in eighteen untrained males (Tilz et al., 1993). Levels rose coincidental^ with increased SIL-2R and following a rise in sTNF-R (Tilz et al., 1993) suggesting a possible cytokine induction. These changes compare to a 20-30% increase seen in patients with insulin-dependent diabetes mellitus, atherosclerosis (Ridker ef al., 1998) and inflammatory vascular disease. In contrast, Jilma et al. (1997) measured a <11% rise in slCAM-1, soluble VCAM-1 (CD106), and soluble E-selectin (sCD62E) in twelve untrained men after both maximal cycle ergometry and sixty minutes at 60%VO2max. This failed to reach a preset clinically relevant difference of 15%, based on a previously reported day-to-day variability of 8% (Jilma et al., 1994). They concluded that recreational activity at normal altitude has insignificant effects on these circulating CAMs, but they specifically excluded subjects who swam >1 km/week, ran >5 km/week or cycled >50 km/week. Hypoxic vasoconstriction and elevated PAP induced by altitude may account for the discrepancy in results, and the physiologic or clinical significance of these studies remains undetermined. Eldridge et al. (1998) reported significantly increased levels of sE-selectin at altitude (3810 metres) (28.7±8.9 ngmL"1 pre-exercise; 39.4±14.2 ngmL"1 24-hours post-exercise, p<0.05), in company with an elevated A-aD0 2, and elevated bronchoalveolar lavage (BAL) RBC (p=0.03) and WBC (p=0.08) at 24 hours, in 5 subjects who exercised at 85%V0 2 max for three 5 minute intervals with a 5 minute recovery at 30%VC>2max between. This data clearly implies alveolar-capillary structural failure with exercise at altitude, but the contribution of exercise alone cannot be determined, and this conclusion may not be 113 applicable at sea level. Likewise, alterations in pulmonary endothelial function on exposure to hypobaric hypoxia is supported by demonstration of increased levels of plasma E-selectin in 6 subjects with hypoxemic acute mountain sickness (AMS) and 8 climbers with high altitude pulmonary edema (HAPE) presenting to the National Park Service medical camp at 4200 m on Denali (Grissom et al., 1997). Plasma E-selectin levels increased significantly in 17 control subjects on ascent from sea level to 4200 m (mean±SD, 12.9±8.2 ngmL 1 versus 17.2±8.2 ngmL"1, p=0.001), but were significantly higher compared to sea level control values in subjects with hypoxemic AMS (30.6+13.4 ngmL"1) and HAPE (23.3±9.1 ngmL"1) (p=0.009). Significant correlation was also seen between plasma E-selectin levels and the degree of hypoxemia (p=0.006). However, differences in plasma E-selectin levels between control subjects at altitude and subjects with hypoxemic AMS and HAPE, perhaps a more telling comparison, was not reported. Plasma P-selectin levels were unchanged on ascent to altitude or in subjects with either AMS or HAPE. 2.3.3.3 Complement The complement system comprises a family of over 30 proteins that contribute to inflammation by induction of the classical or alternate pathways. Complement fragments, C3a and C5a in particular, trigger release of inflammatory mediators, and, in the case of C5a, acts as a chemotactic factor. Opsonization and enhanced removal of immune complexes are also aided by complement. Elevated serum concentrations of C3, C4 and C5a, have been demonstrated following progressive maximal exercise (Dufaux ef al., 1991), short-term submaximal exercise (Camus ef al., 1994) and prolonged submaximal exercise (Dufaux and Order, 1989; Castell ef al., 1997). These increases were likely in response to muscular injury. 2.3.3.5 Nitric oxide In addition to its effects on vascular tone, NO is also involved in the inflammatory response. Inducible NOS (iNOS) generates NO in inflammatory cells, endothelium and smooth muscle cells in response to cytokines and endotoxin and contributes to host defense and inflammatory diseases (Moncada ef al., 1991). Using a cat in situ superior mesenteric artery model, Kubes ef al. (1991) demonstrated increased endothelial neutrophil adherence after pretreatment with L-arginine analogues, NG-monomethyl-L-arginine (L-NMMA) and NG-nitro-L-arginine methyl ester (L-NAME), that inhibit NO 114 production, while neutrophil adherence was completely reversed by a CD18-specific antibody. These data suggests NO potentially limits neutrophil adherence through some unknown affect on the neutrophil adhesion glycoprotein CD11/18, and reduction in NO could permit enhanced neutrophil-endothelial adherence with potential for neutrophil priming by circulating cytokines, leading to inflammation and endothelial injury. 2.3.3.6 Influence of gender and fitness Immune responsiveness is generally greater in women than men and may account for the greater susceptibility to autoimmune disease in females (Ahmed ef al., 1985). Leukocyte subpopulations have been shown to vary over the normal menstrual cycle, but the significance of this alteration is uncertain. Lymphocytes reached a nadir at midcycle coincident with E 2 peak, whereas monocytes and neutrophils were significantly higher in the luteal versus the follicular phase (p<0.05), and correlated with progesterone levels (r not reported; p<0.05) but not E 2 (Mathur ef al., 1979). IL-1 activity was increased during the luteal phase, but IL-1 response to endotoxin was more vigorous during the follicular phase (Lynch et al., 1994). Implications of these findings on the role of varying cytokines, inflammation and immune responsiveness in women are as yet undetermined. Increased in vitro production of IL-1 following 6 minutes of interval exercise corresponding to 55%, 70% and 85%V0 2 m a x , was independent of gender and fitness in a group of 64 sedentary and moderately fit males and females (Moyna et al., 1996). Normal resting levels of vWF:Ag did not vary with sex (Blann, 1990). Nor did E 2 significantly effect serum levels of sE-selectin, slCAM-1 or sVCAM-1 in 18 males administered 10 mg of E 2 valerate intramuscularly in a double blind, randomized, placebo controlled, cross-over study (Jilma ef al., 1994). E 2 administration was, however, associated with a decrease in sP-selectin, but levels measured during the follicular phase were not significantly different from males (Jilma etal., 1996). 2.3.4 Coagulation Exercise has also been shown to induce platelet activation, coagulation, and fibrinolysis (Cash, 1966; Davis ef al., 1976) relative to exercise intensity (Andrew ef al., 1986). Platelet function was desensitized by short-term moderate exercise (50% V 0 2 m a x ) , but potentiated by strenuous exercise (V0 2 m a x ) (Wang ef al., 1994). While this pattern also 115 held true for women in the mid-follicular phase, platelets adhesiveness and aggregation were not enhanced by acute heavy exercise at mid-luteal phase (Wang et al., 1997). Weiss etal. (1998) measured markers of thrombin, fibrin and plasmin formation in 12 male subjects who underwent 1 hour of treadmill running at 6 8 % and 8 3 % of V 0 2 m a x and demonstrated a balanced rise in both fibrinolytic and thrombogenic activity. However, Mockel ef al. (1999) demonstrated a significant increase in platelet expression of P-selectin in 15 male triathletes after exhaustive exercise. The rise in P-selectin expression was more marked after intensive exercise and was independent of the platelet count. Levels returned to baseline within 30 minutes. Fibrin monomer and pro-thrombin was also elevated and maximal levels were attained at 30 minutes post exercise. These data suggest the potential for a vulnerable coagulation window following intensive exercise. What is more, platelet activation may interact with neutrophils in a cross over response between hemostasis and inflammation in aid of vascular repair. The potential relationship of exercise, platelet function, P-selectin and 17p-estradiol, and the potential influence on neutrophil traffic through the lung, remains incompletely addressed. E 2 may exert a confounding influence on the measurement of cell-specific markers of acute lung injury, but this appears least likely to occur at the mid-follicular phase. 2.3.5 Acute lung injury Acute lung injury describes a constellation of clinical features associated with increased capillary-alveolar permeability, diffuse alveolar damage, and severe gas exchange impairment that all too commonly culminates in ARDS. Endotoxin-induced ALI is mediated by release of TNF -a and IL-1p (see Streiter ef al., 1993 for review). In humans, intravenous challenge with RE-2 endotoxin produced a systemic inflammatory response with fever, neutrophilia, and cytokine release in a temporally differentiated and dose-dependent manner (Kuhns ef al., 1995). TNF -a and soluble TNF receptor have been demonstrated to rise 1 hour after endotoxin challenge, IL-6 increased by 1.5 hours, IL-1 receptor antagonist increased by 2 hours, and sE-selectin rose linearly at 4, 6 and 24 hours in response to 1-4 ng-kg"1 of endotoxin. These temporal changes suggest sE-selectin may help predict severity of endotoxemia. Pretreatment of rats with a glutathione depleting agent (diethylmaleate) prevented lipopolysaccharide (LPS)-induced increase in lung permeability and neutrophil content, which seemed to be mediated by prevention of ICAM-1 upregulation (Nathens ef al., 1998). 116 Muscular effort, thermoregulation and preservation of cerebral and coronary perfusion promote competition between vascular beds. Blood redistribution and increased adrenergic tone during exercise further reduce splanchnic blood flow and may lead to potential bowel ischemia and release of endotoxin by gut bacterial translocation (Marshall, 1998). Elevated plasma levels of LPS and anti-endotoxin IgG have been detected after an international distance triathlon (Bosenberg et al., 1988) and an ultra-distance marathon (Brock-Utne ef al., 1988), but increased levels of LPS were not demonstrated after a half-marathon (Brock-Utne et al., 1988) or in cyclists who collapsed after a race (Moore et al., 1995). Increased levels of anti-endotoxin IgG may have confounded the negative results. TNF -a and IL-1 (3 have been shown to enhance neutrophil-endothelial adherence (Pohlman ef al., 1986) by upregulation of E-selectin and ICAM-1, while TNF -a also enhanced neutrophil-mediated endothelial injury (Varani ef al., 1988), and following incubation in vitro, inhibited NO release from vascular endothelium (Aoki et al., 1989). Neutophil recruitment, with release of toxic neutrophil products such as proteases and ROS, mediates EC injury (Ward ef al., 1985; Varani ef al., 1985) in lung microvascular injury from complement (Fantone and Ward, 1985), and microembolism (Flick et al., 1981). Neutrophil O2" release reacts with NO in extracellular spaces to produce peroxynitrite (ONOO), a potent and toxic proinflammatory mediator. Extracellular superoxide dismutase (EC-SOD) defends against the formation of O2" and ONOO", and may control the critical balance of whether NO acts as an inflammatory regulator or is transformed into a pro-inflammatory species (Day ef al., 1996). Although Hinder ef al. (1997) failed to detect an increase in pulmonary endothelial permeability after increasing PVR by reversing endotoxin-induced systemic vasodilation with L-NAME in a sheep model, Sheridan et al. (1999) recently attenuated endotoxin-induced dysfunction of pulmonary endothelial-dependent vasodilation in rats by L-arginine supplementation. They had earlier reduced endotoxin-induced vasoconstriction in isolated rat pulmonary arterial rings by inhibiting phosphodiesterase, an inactivator of cGMP (Sheridan et al., 1997). While inflammation can induce pulmonary vasoconstriction acutely, chronic interactions between inflammatory cells, EC and VSMC, with derangement of growth and differentiation of EC, SMC and perivascular fibroblasts, can also be a cause, result or modulator of pulmonary vascular remodeling leading to pulmonary hypertension (Tuder and Voelkel, 1998). 117 Three experimental rat models illustrate the mechanistic complexity and variance of ALI (see Ward et al., 1996 for review). Infusion of cobra venom factor (CVF), a potent activator of the alternate complement pathway, produces intravascular neutrophil activation, upregulation of CD11b/CD18, oxidant-mediated damage of endothelial and alveolar epithelial cells, and a morphological pattern similar to ARDS. This model of injury is neutrophil- and complement-dependent, in particular C5a, but is cytokine-independent (IL-1p and TNF-a) (Mulligan et al., 1992; Mulligan ef al., 1997). p2 integrins (CD11a/CD18, CD11b/CD18), and constitutively expressed ICAM-1 and L- (CD62L) and P- selectins are required (Mulligan et al., 1993). Intrapulmonary IgG immune complex deposition produces a second pattern of injury which is also neutrophil- and complement-dependent, and requires p2 integrin (CD11a/CD18), ICAM-1, E- and L-selectins (Mulligan ef al., 1993), and cytokines (IL-ip and TNF-a) for neutrophil recruitment (Warren ef al., 1989). Injury develops over four hours and the subsequent morphological pattern caused by combined release of oxidants and proteases by alveolar macrophages and recruited neutrophils is similar to inflammation secondary to bacterial infection. A third model is produced by ischemia-reperfusion. Like the immune complex model it again is both neutrophil- and complement-dependent, and requires cytokine (IL-ip and TNF-a) (Seekamp ef al., 1993) and adhesion molecule (CD11a/CD18, CD11b/CD18, E- and L-selectin) participation (Seekamp ef al., 1994). Thus, p2 integrin (CD11a/CD18) and L-selectin are universally involved, while cytokine dependence and E-selectin independence differentiate between the immune complex/ischemia-reperfusion models and the complement activation model. Furthermore, the time course of injury is more rapid in the complement model, whereas injury in both the immune complex and ischemia-reperfusion models is delayed, a temporal difference that may reflect cytokine dependence. 2.3.5.1 Cell markers of acute lung injury The effort to better predict and prognosticate ARDS has lead to a search for sensitive and specific cellular and biochemical markers of ALI. Apart from direct measurement of BAL and pulmonary edema fluid, several cell-specific markers of endothelial and epithelial 118 injury, as well as markers of acute inflammation, have been measured in plasma of patients at risk for and with ALI (see Pittet et al., 1997 for review). Potentially useful markers of endothelial injury include von Willebrand factor antigen (vWF:Ag) and soluble CAMs. von Willebrand factor is synthesized primarily by vascular endothelial cells and to a lesser extent by megakaryocytes and platelets. It functions in hemostasis by mediating adhesion of platelets to exposed collagen and subendothelium. Consitutive secretion of dimers or small multimers of vWF can be induced by calcium or vasopressin without concomitant endothelial injury (Loesberg ef al., 1987), or by inflammatory mediators like endotoxin or TNF-a (Gralnick et al., 1989; van der Poll ef al., 1992). On the other hand, high molecular weight multimers are stored in endothelial specific Wiebel-Palade bodies and form the pool of protein that is most likely released at the time of vascular activation or injury (Wagner, 1990). vWF:Ag is considered a sensitive marker of endothelial activation (Hamilton ef al., 1987) and has been used as a marker of vascular endothelial damage after exercise in patients with intermittent claudication (Edwards et al., 1994). In healthy subjects, levels rose in an intensity-dependent relationship with exercise (Andrew et al., 1986; Wheeler et al., 1986), but could be blocked with propanolol (Small et al., 1984), and blunted by partial blockade of NOS with L-NMMA (Jilma ef al., 1997). In contrast, at rest, histamine induction of vWF:Ag was accentuated by pretreatment with L-NMMA, which by itself had no effect (Jilma ef al., 1998). Together, these results suggest p-adrenergic and histamine induced release which is at least partially mediated by NO in an as yet undefined mechanism. Elevated levels of vWF:Ag have been reported in patients with acute respiratory failure (Carvalho ef al., 1982) and were predictive of acute lung injury in patients with nonpulmonary sepsis (Rubin ef al., 1990). However, studying a more heterogeneous population of patients with ALI, Moss et al. (1995) found poor sensitivity and specificity for predicting ARDS in both septic (sensitivity 70%, specificity 47%) and nonseptic (sensitivity 64%, specificity 52%) patients. Differences in the pattern of lung injury and variance in the response of the endothelium may account for the discrepancy. Prospectively following a cohort of 36 trauma patients and 19 patients with sepsis, two at risk diagnoses for ARDS, Moss ef al. (1996) demonstrated significant differences between sepsis and trauma 119 patients in endothelial cell activity. Levels of vWF:Ag, slCAM-1 and sE-selectin were significantly higher in the septic patients (p<0.001), and neither slCAM-1 (p=0.17) nor sE-selectin (p=0.24) in trauma patients was different from normal controls. Twenty-six percent of the septic patients and 25% of the trauma patients developed ARDS, but significant differences between groups in all three endothelial cell markers persisted (vWF:Ag p=0.008, slCAM-1 p=0.003, and sE-selectin p=0.003). In view of their central role in neutrophil-mediated lung injury, soluble vascular selectins have also been measured as indirect markers of endothelial activation or injury. E-selectin is produced exclusively by endothelium following cytokine activation, whereas P-selectin is preformed and stored in a granules of platelets or in company with vWF:Ag in endothelial Weibel-Palade bodies. P-selectin release has been stimulated by thrombin, activated complement, hydrogen peroxide or histamine (Lorant ef al., 1991) without de novo synthesis, as well as by TNF-a upregulation (Weller et al., 1992). It likely plays crossover role between inflammation and thrombosis through mediation of neutrophil-platelet interactions. Relative contributions of EC or platelets to circulating levels of P-selectin are unknown. For 17 patients with positive blood cultures, plasma levels of sE-selectin were elevated in those with septic shock but not bacteremia without hypotension (Newman et al., 1993). Levels of sE-selectin, along with vWF:Ag and slCAM-1, were similarly elevated in 25 patients with severe sepsis compared to healthy volunteers (n=9) and nonsepsis controls (n=7), were correlated with simplified acute physiology and multiple organ failure scores, but not ALI scores, and, unlike markers of neutrophil activation, were able to predict survival with relatively high sensitivity (80-90%) and specificity (73-87%) (Kayal ef al., 1998). However, unlike levels of sL-selectin, levels of sP-selectin and sE-selectin were unable to predict the development of ALI in a more heterogeneous population of at risk patients with multiple trauma, pancreatitis and bowel perforation (Donnelly ef al., 1993). Hence, the selectins have yet to find an established role to monitor disease in critically ill patients with acute lung injury. 2.4 Hemodynamics Although a small linear increase in pulmonary artery pressure occurs with exercise in humans, values do not usually exceed the 20 mmHg threshold pressure that defines pulmonary hypertension (Epstein ef al., 1967; Thadani and Parker, 1978; Janosi ef al., 1988). However, evidence for pulmonary hypertension during exercise in humans comes 120 from hemodynamic data acquired in the Operation Everest II environmental physiology study (Groves et al., 1985; Reeves ef al, 1990). Eight healthy male volunteers, all former competitive endurance athletes (mean V0 2max = 51.3 mlkg"1min"1) underwent right heart catheterization in a decompression chamber at barometric pressures equivalent to sea level, 6100m and 7620m. Hemodynamic, blood gas and metabolic parameters were measured at rest and during exercise at each altitude and at 8840m. At sea level, a positive linear relationship was demonstrated between PAP and Q (r=0.76, p<0.05). Of interest, the three subjects with Q greater than 30 L-min"1 all had mean peak PAP of greater than 40 mmHg. Similar measures of PAP and PAOP were also obtained from 8 subjects at a VO2 of approximately 3.3 L-min"1 (peak workload 271 watts) with mean PAP values reaching 37.3+6.1 mmHg and mean PAOP values reaching 21.0+3.7 mmHg. Following the reasoning of West et al. (1991), pulmonary capillary pressures at the base of the exercising lung would potentially have exceeded pressures of 36 mmHg and might have been within the range of mechanical stress failure seen in their rabbit model. However, this value was predicated on a hydrostatic column, which may or may not hold true if vascular perfusion is determined more by fractal structure. Furthermore, pulmonary capillary pressure was thought to approximate the half way point between PAP and PAOP, although, based on data from exercising dogs, pulmonary capillary pressures under conditions of pulmonary blood flow may approach arterial pressures (Younes et al., 1987). Thus, stress failure could potentially result from regional flow heterogeneity and hyperperfusion. 2.4.1 Capillary stress failure West and Mathieu-Costello (1993) have brought attention to the bioengineering dilemma of the alveolar-capillary membrane. At once thin enough to allow for efficient gas exchange, yet strong enough to retain its structural integrity, the blood-gas interface is vulnerable during high flow and/or pressure states such as exercise. The pulmonary capillary wall, supported in part by alveolar surface tension, is subject to circumferential tension from transmural pressure, and longitudinal tension from lung inflation. As well, wall structures are under stress as defined by the Laplace relationship of tension over thickness. West ef al. (1991) investigated ultrastructural changes of pulmonary capillaries in rabbits at various transmural pressures and consistently found electron micrograph evidence of capillary endothelial and alveolar epithelial disruption at perfusion pressures exceeding 39 mmHg. The incidence of stress failure was increased further at high lung 121 volumes (Zhenxing et al., 1992). Comparative images, three minutes after returning to normal pressures, showed significantly fewer breaks, suggesting cell adhesion molecules may be able to disengage and reattach along an intact basement membrane (Elliot et al., 1992), which lends support to a mechanical recovery process. Kurdak et al. (1995) perfused rabbit lungs at a pressure of 26 mmHg for 10 and 100 minutes and demonstrated no significant difference between conditions in the degree of BGB disruption. These data support the potential for vascular damage consequent to brief high intensity insults, but not following prolonged moderate levels of perfusion pressure. Hopkins et al. (1997) compared BAL fluid of sedentary controls to BAL fluid of athletes following a maximal exercise challenge at sea level and reported increased quantities of red blood cells and protein in the exercise group, but no increase in leukotrienes. Unfortunately, they did not measure Sa02 or V0 2 . However, their results suggest the integrity of the blood-gas barrier may be impaired during intense exercise in some subjects. These data are in contrast to results from a group of 12 high aerobic power male athletes (mean±SD, V02max = 5.30±0.37 L-min"1) who underwent Technetium 99m DPTA lung scanning at rest and following maximal exercise. No significant difference in resting or post-exercise pulmonary clearance of Tc-99m DTPA was observed (p>0.05), and correlation between pulmonary clearance rate and minimum Pa02 was not significant (r=-0.26, p>0.05) (Edwards et al., in press). Failure to increase pulmonary clearance post-exercise does not support loss of epithelial integrity despite exercise-induced reductions in gas exchange. However, the integrity of the endothelium is not assessed by Tc-99m DTPA. It is therefore possible that BGB injury short of epithelial leakage may still have occurred. 2.4.2 Vascular remodeling Structural changes of the pulmonary circulation occur in response to elevations in pulmonary arterial pressure or long-term increases in flow. EC injury, muscular layer development, increases in SMC, and medial thickening characterizes remodeling. These alterations appear to occur irrespective of the inciting event. Genes or gene products characteristic of earlier development are expressed by SMC in response to injury, and SMC development provides insight into vascular injury and remodeling. As well, phenotypic heterogeneity of SMCs adds to the complexity of response to increases in pressure (see Stenmark and Mecham, 1997 for review). Local vascular injury induces 122 expression and release of varying growth factors including VEGF, platelet derived growth factor (PDGF), and transforming growth factor-p (TGF-p). Protein kinase C and C a 2 + may be involved in mechanotransduction in response to shear stress, while reiteration of fibroblast mRNA induces collagen deposition and protein production. Pressure elevation alone may not be sufficient to induce remodeling processes, but may require the presence of inflammatory signals (Tanaka ef al., 1996). 2.5 Pulmonary edema The dichotomous description of pulmonary edema as either hydrostatic or increased permeability edema no longer reflects the underlying pathophysiological processes. Newer radiographic classifications have been introduced and divide pulmonary edema into one of four groupings: increased hydrostatic pressure edema, permeability without diffuse alveolar damage (DAD), permeability edema with DAD, and mixed edema (Ketai and Godwin, 1998; Gluecker et al., 1999). Yet, even this modification does not describe the overlapping continuum of pulmonary vascular response to mechanical and inflammatory signals. Pressure may induce inflammation, and inflammation may effect a rise in pressure. In addition, hydrostatic edema is not simply a result of an imbalance in Starling's forces, but like permeability edema, has been associated with barrier leaks (Wu et al., 1995). In the rabbit lung model, breaks have typically been seen in the epithelial layer. However, the rabbit lungs were fixed after perfusion pressures had returned to normal. It is probable, therefore, that endothelial cell repair and recovery had already taken place. The potential for exercise-induced/exacerbated interstitial edema is best provided by the example of high altitude pulmonary edema. Although confounded by the influence of hypobaric hypoxia, there remain some similarities that may be applicable to gas exchange impairment during sea level exercise. 2.5.1 High altitude pulmonary edema The pulmonary circulation is central to the development of high permeability pulmonary edema seen on exposure to hypobaric hypoxia in susceptible subjects. Marked by abnormal pulmonary hemodynamics, gas exchange impairment, and severe arterial oxyhemoglobin desaturation, HAPE usually manifests with tachypnea, cough and dyspnea 2-5 days following acute exposure to altitudes above 2500 to 3000 metres. Plain film 123 chest radiography and computerized tomography demonstrate patchy airspace consolidation with a predominantly peripheral distribution (Bartsch, 1999). Resting hemodynamic measures reveal a marked rise in PAP without a corresponding elevation of PAOP and reflect an increase in the driving pressure across the vascular bed. 2.5.1.1 Pulmonary hemodynamics Pulmonary hypertension is considered a sine qua non of HAPE. Eldridge et al. (1996) invasively measured the pulmonary vascular response to exercise at sea level in 7 HAPE susceptible (HAPE-S) subjects compared to 9 nonsusceptible controls, and demonstrated a significantly greater PAP and PAOP response to exercise in the HAPE-S subjects at an equivalent Q . In a companion study, Podolsky ef al. (1996) noted enhanced exercise-induced V A / Q inequality in the HAPE-S subjects. Hultgren (1970) proposed HAPE may result from segmental overperfusion as a consequence of patchy HPV, a hypothesis that is supported by improvement in HAPE with pulmonary vasodilators such as calcium channel blockers (Oelz ef al., 1989) and NO (Scherrer ef al., 1996). Whether or not this difference in response is an innate feature of the pulmonary endothelium in HAPE-S subjects or a result of vascular remodelling is unknown. 2.5.1.2 Inflammation However, although seen in all subjects with HAPE, the elevation of PAP alone may be insufficient to induce edema formation; additional factors, such as inflammatory mediators and/or altered fluid clearance may be required. Brochoalveolar lavage in 7 patients with early HAPE demonstrated increases in cell counts, protein, and pro-inflammatory cytokines (IL-1(3, TNF-a, IL-6, and IL-8), but no increase in the anti-inflammatory cytokine IL-10 (Kubo ef al., 1998). In addition, concentrations of IL-6 and TNF-a in BAL fluid were significantly correlated with driving pressure (PAP-PAOP) and room air Pa02. As previously noted, alterations in pulmonary endothelial function on exposure to hypobaric hypoxia is further supported by demonstration of increased levels of plasma E-selectin in subjects with hypoxemic AMS and climbers with HAPE (Grissom ef al., 1997). Whereas evidence exists for both an inflammatory component and endothelial cell activation or injury in HAPE, the particulars of which inflammatory mediators are involved remains incompletely determined. Although urinary leukotriene E4 (uLTE4) levels were increased significantly from sea level to high altitude (4300 m) (mean+SEM, 67.9±13.2 124 pg-mg"1 creatinine versus 134.8±19.4 pg-mg"1 creatinine, p<0.05) in 8 healthy young male subjects (Roach et al., 1996), no significant difference was noted between 7 HAPE-S subjects and 5 nonsusceptible controls subjected to slow decompression to 4000 m over 4 hours (Bartsch ef al., 2000). However, HAPE did not occur in any subjects. In a companion field test, Bartsch ef al. (2000) failed to demonstrate any increase in uLTE4 in 5 subjects who developed HAPE at 4559 m compared to their pre-HAPE levels. These data do not support a leukotreine mediated inflammatory response in susceptibility or development of HAPE. 2.5.1.3 HAPE susceptibility HAPE tends to recur in individuals on re-exposure to altitude, whereas others are seemingly unaffected despite repeated ascents. Susceptibility to HAPE may reflect a constitutional predisposition (Hultgren et al, 1971), and Hanaoka et al. (1998) have demonstrated an association between HAPE and human leukocyte antigens (HI.A) in Japanese climbers. HLA-DR6 was positive in 14 of 30 patients (46.7%) with HAPE but only in 16% of controls (p=0.0005, OR=4.6). Furthermore, PAP in 5 HLA-DR6 positive patients with HAPE was significantly greater (p<0.05) than in 5 DR6-negative patients. These data support an interaction between HI.A class II alleles and HAPE-S, but cannot exclude the potential involvement of another non-HLA gene that is linked to the major histocompatibility complex (MHC) (Arnett, 1998). 2.6 EIAH and HAPE-S Distinct intersubject differences in HAPE-S, associated with an elevated pulmonary vascular hemodynamic response to exercise, accentuation of V A / Q inequality, release of inflammatory mediators and soluble CAMs, and, potentially an association with MHC alleles, may prove instructive to understanding exercise-induced gas exchange impairment at sea level in a subset of habitually active people. Similarities exist, although comparative hemodynamic measures, and measures of immune response and genetic phenotyping have not been reported in subjects with EIAH. It remains possible, and seems probable, that EIAH at sea level represents in subtle form a continuum of response of the pulmonary vasculature to environmental and physical stress that culminates in florid alveolar flooding on ascent to altitude. 125 2.7 Future directions The picture that emerges from this review is one of the pulmonary endothelium as a thin, vital, but vulnerable cell layer located at the critical interface of air and blood. Under demand as modifier, regulator and protector, it nonetheless withstands large increases in blood flow and pressure. For, despite its apparent fragile nature, it does not routinely fail. Or, if it does, its repair processes are rapid and effective in restoring and maintaining function. Faced with the stress of exercise, and particularly the repetitive and sustained increases in flow seen in habitual aerobic athletes, it is perhaps more remarkable that every aerobic athlete does not demonstrate exercise-induced arterial hypoxemia. Clear intersubject variabilities exist, and the reasons for these differences will continue to provoke. Heterogeneity of vascular structure, perfusion, V A / Q dispersion, ventilatory response, mechanical limitation, inflammation and immune response are among the many candidate mechanisms that require integration. Our understanding of the genetic, immune and cellular processes that impact on the pulmonary endothelium remains limited. So too our understanding of vascular repair and remodeling processes that may be central to deciphering the impact of exercise on the pulmonary circulation, as well as the impact of the pulmonary circulation on exercise performance. 2.8 Conclusion Exposed to the entire blood volume, and central to inflammatory and hemostatic reactions, control of vascular tone, and processes of vascular injury, repair and remodeling, the pulmonary endothelium is situated to play a major role in BGB integrity and gas exchange. Alterations in membrane permeability via mechanotransduction and increases in [Ca2+]i, or EC injury from mechanical and inflammatory stressors, may effect an increase in interstitial lung water during exercise. Injury repair and remodeling may also influence the vascular response to flow, thereby initiating a cycle. Altered hemodynamics could also influence functional blood flow redistribution and V A / Q matching. Whether or not any of these mechanisms play a role in EIAH is open to further research, but one or more could potentially explain the finding of altered gas exchange impairment seen during submaximal exercise in some habitually active subjects. 126 Appendix III. Raw Data Table 6. Age, height and mass of individual subjects Subject Age (yrs) Height (m) Mass (kg) MK 27 1.74 64.4 RJ1 22 1.69 66.3 RJ2 38 1.62 58.2 DW 32 1.69 58.8 OM 23 1.56 45.5 MS 35 1.63 67.7 KM 25 1.64 57.7 NE 34 1.63 62.0 WK 29 1.62 62.0 WJ 28 1.62 57.5 HJ 23 1.75 62.3 IS 36 1.73 73.8 CL 31 1.56 46.5 TS 22 1.64 62.1 Mean 29.3 1.7 61.4 SD 5.57 0.06 6.92 yrs, years; m, metres; kg, kilograms Table 7. Individual subject hematologic and hormonal parameters Subject Hgb Hct Day of Progesterone (gL-1) cycle (nmol-L-1) MK 146 0.43 6 0.24 RJ1 126 0.37 6 0.17 RJ2 136 0.40 5 0.40f DW 129 0.38 6 0.14 OM 146 0.43 9 <0.008 MS * * 6 0.21 KM 150 0.44 5 <0.008 NE 102 0.30 5 <0.008 WK 116 0.34 DP <0.008 WJ 139 0.41 7 <0.008 HJ 122 0.36 7 0.42t IS 133 0.39 6 <0.008 CL 153 0.45 8 0.20 TS 136 0.40 5 <0.008 Mean 133 0.39 6 0.08 SD 14.5 0.04 1.2 0.1 DP, Depoprovera (at the end of a three month cycle) nmol-L"1, nanomoles per litre *denotes missing data due to a faulty analyzer cartridge tCoefficient of variation >6%, excluded from mean 128 CD T J E £ Q. |LU i £ co o E <» u_ « LU - i O > T J LL CD CL > LU O > > LU T J | > 2> LU Q. Lu 3 O > o > LL o > o > CD CD iri O) 00 CD T J CD TJ CD i— D_ T J CD P. 2 o _l t-O CD O T - O •<fr eg CO CM CM O T - O 00 OJ o i n co CO oo o "fr 00 T - T -N 00 CO CD CD 00 O CO O CO CO CO OJ o CO CO i LO CM CO OJ 1^  CD iri oo CO c i oo OJ CO CO CO o T - T - 00 CO OJ OJ OJ OJ m o i OJ i n 00 OJ OJ r - " f r -e t OJ 00 co CO CM CM OJ CO CO CO CO CM 00 CO CO " I " O) O) c o N c o s c o n r O T f O O O T - T - O C M T -CM T -CO OJ CM i n m OJ o •<-00 o CO T -CD m CD "fr "fr o CO o co CO i n m o CD CD CO O) o CM i CD •>fr o i n CO CO CO CO CO CO CO CO 00 o CO CO s T " T ~ O CM OJ CO CO CO CO CO CM N CM O) r~- "fr oo T - CM T -o co I f l t C O t C O C O ^ C O f C O r r t "f 00 o 00 , CO CO o in OJ 00 CO o CM LO "fr OJ C O C D O J C O C O C O O J O O t - T - O r - T - T - O T -00 CO OO o OJ 00 "fr CM OJ CD CM CM N-CO CO "fr co "fr "fr "fr o o o o o j g c o ^ ^ c o g g ^ O) CO 00 CM O CO O LO CD CO "fr "fr i n LO N- 00 oo CO "fr 00 LO OJ oo CO CO o CM CO CM CM m OJ 00 CD "fr iri CD iri iri CO iri iri iri c i CO CO CO § 2: •>fr m co OJ in o co co i n co co CM ^ <P OJ O r-i o O CD "fr ™ T - OJ o o cd OJ O "fr "fr c i OJ CO 00 CM o o OJ c >» -*—' o CD Q. CD o ro •4—» CD CD Q. in CD .t± o 0 3 CL C CD O ° ro 3-a > s O T J ^ o in o cu <•= CD E X CO E X CO E CD LL § LU 2 LL CD •*£ O (D ~ o O" CD I D_ ro > o •4-* TJ , _ B > O LU T J LL CD « _ Q. ° — 2 c -~ CD ro o CD O" ° - > T> ^ 2 > LL O CD CO CM CO LU ^ CO 1-C CD Q ^ CO 129 Table 9. Ventilatory parameters at peak exercise of individual subjects Subject (L-min-1) V C 0 2 (L-min1) VE IVCO2 vE /vo 2 RER MK 93.3 3.54 26.3 33.4 1.31 RJ1 88.7 3.76 23.6 28.0 1.20 RJ2 89.3 3.10 28.8 32.4 1.11 DW 103.2 3.51 29.8 32.8 1.11 OM 46.8 2.00 23.4 25.1 1.10 MS 70.6 2.50 28.4 32.4 1.13 KM 91.8 3.71 25.0 28.1 1.14 NE 73.5 2.71 26.6 27.3 1.05 WK 109.9 4.05 27.2 30.1 1.11 WJ 80.5 3.19 25.3 28.5 1.13 HJ 91.7 3.51 26.6 29.2 1.13 IS 107.6 4.12 26.2 27.1 1.02 CL 68.1 3.10 21.5 22.6 1.06 TS 98.4 3.86 25.6 29.1 1.16 Mean 86.7 3.3 26.0 29.0 1.1 SD 17.3 0.6 2.2 3.1 0.1 VE, minute ventilation in litres per minute; VCO2, CO2 production in litres per minute; VE /VCO2 and VE /VO2, ventilatory equivalents for CO2 and O2, respectively; RER, respiratory exchange ratio 130 Table 10. Performance at peak exercise of individual subjects Subject HR Power VC^peak V02peak bpm (W) (L-min-1) (ml-kg"1-min-1) MK 187 269 3.4 42.0 RJ1 187 300 3.2 . 47.7 RJ2 160 250 3.1 47.4 DW 191 310 3.5 53.4 OM 177 174 1.9 41.0 MS 173 195 2.5 32.3 KM 186 300 3.7 . 56.3 NE 165 264 2.6 42.0 WK 176 311 3.7 58.9 WJ 195 277 2.8 49.2 HJ 187 333 3.2 50.6 IS 176 325 4.2 53.9 CL 184 271 2.9 63.7 TS 195 320 3.3 53.6 Mean 181.4 278.5 3.1 49.4 SD 10.6 47.1 0.6 8.2 HR, heart rate in beats per minute; Power, power output in watts; VC^peak, peak O2 consumption in liters per minute and millilitres per kilogram of body mass per minute 131 Table 11. Ambient testing conditions Subject Atmospher ic Pressure Temperature RH (%) kPa Torr (°C) MK 100.7 755 22 64 RJ1 101.1 758 24 66 RJ2 101.1 758 24 70 DW 101.2 759 22 70 OM 100.7 755 22 66 MS 100.9 757 24 66 KM 101.2 759 22 70 NE 100.7 755 23 62 WK 100.4 753 22 65 WJ 100.7 755 23 64 HJ 101.1 758 26 68 IS 100.4 753 23 64 CL 100.9 757 22 68 TS 99.9 749 22 60 Mean 100.8 755.8 22.9 65.9 SD 0.37 2.81 1.21 3.05 kPa, kilopascals; °C, degrees Celsius; RH, relative humidity in percent 132 c o CO cr cu to co OJ I— _ro o cu > CO TO cu T J 0 E o TJ 0 ••-» ro _o ro o ro -4—» ro TJ CM o ro -g > T J c CN 0 ro 0 o D_ ro 0 D_ 0 co E" 0 x 0 ro 0 Q. to . g o c co CM to 0 > C ID • F CM E CM OJ to c o E £ CD CO 1 z CO to 0 a: cn o o o St LO o 1^ - CO LO o CD o LO T— CD o CD T— CO CM CM CN CO CN CO CO CN CO CM CO CO CM CO CM o CM CM CD CN CO CD CO o CD CM o o h-ro CO co CD o CD CD CD CD LO ,— r - CO CD CM D_ •sf LO LO CO •sf LO LO LO LO •sf •sf LO LO r>- o CD CO CO •sr OJ CD o CM T— o LO 1-x— T— T— T— ^— — ro CO o OJ CD CD CO 0- LO CD LO LO LO CO LO LO T T— ro o_ ro CL ro DL ro CM O •Sf CO T - T ~ T - O LO C N C D C D C O O O U n c O C N r ^ T - O O O O O O O O CD I— CM •sr 4 ir i •sr C D L n L n - r - T f o r ^ - c D c o C O C O L O C D L O C N O T - C D C J J O O O O O O T - o o o o CD OJ OJ OJ CN T]" CO o o o OJ 00 OJ OJ CN CD •sr o o CO OJ OJ OJ CD OJ •sr -sr O T - * T - CD o o CO OJ CM o OJ CO OJ OJ o o oo OJ •sr o CN O c o c r i ^ r c o c o ^ c N c o c r i c N c r i - ^ c M c o OJ o CO CD CM L O T - T J C D C M I - C O T - L O i r i * CO LO * d - s f C N O O C O C O C O C O ' S r + CO CO ^ in •sr 4 OJ CO CD CO CO CO CO iri CO d in CO O CO in d CD "sf o •sj-CD d OJ co o iri : N -co CO ^ O CD '. OJ o 0 xT CO c ro Q u CO 133 CD o CO CD CL o o o o - ^ - L o o ^ r T - r - c o L O T - o c D O L O t - N . o ) o c o t - i ^ - c o c N j r - c \ i C N C O C N C O T - T - C O C N C O C N O O C O C N C O CD w CD X CD CO CD CL CO c O C CO |CN CO = * C LO F CM E CM IO) co CD > C O " LO £ ICO CO CD ^ 1 £ I £ CO CO CO CL CO ISz ,ro 5E i CO ,ro 5E CO |0_ LO CO LO * LO CO LO o * CO CM 0) "fr CM CO CO " i r ^ o i o c n o i c o c o c o O L O N - " f r C O O ) 0 0 > L O C O " f r C O ' f r C M C M C M C M T ^ T - ^ O CO "fr "fr "fr CO CD "fr LO CM CO CO LO CM LO * "fr CO CO o * CO c n CO CM CO CO c n c o "fr LO CM ^ ° o r-~ c n c n O LO o o L O T - O O C O O O C N C O L O C M C M C O " f r C M T - : CO CD "fr CM CD c n o o CM _ "fr 5 2 00 00 CO " - c n r - " -CM CO "fr CM CO c n o O "fr CO ^ O ) CO CO LO CO "fr T -0 ) 0 ) 0 ) 0 co r^ - "fr LO T -CO CM CM CM CM CO K CM o o5 o T- e n CD "fr CM CO CO o CD "fr CD <D c n T -c n CM c n c n CM O) CM O ) CD CO CO CM CD cn CM CO CO CO o o r-' o ' c n T -CO 00 CD CO CO TJ c CO aT CL "«» & -= CJ) s ? CL g — z. 12 CD c CO CD £ cn ro "fe .ro o 0 3 0 CO 3 ro 5 t : ro °> D_.£ — CO ro co CD E w CO CD -4—• CN O o c o j 0 I CO c ro Q £ CO CO CD 134 0 5 o CL CD 0 CL c j > o o o ' s j - u o o ' s f T - r - c o u o T - o C D O L O T — I ^ C D O C D T — r ^ - c o c \ i r ^ - c \ i C N C O C N C O T - T - C O C N C O C M C O O O C N C O 0 CO o 1_ 0 0 0 > co CO 0 1— CD o CD c T J ro ro T J " c 0 X J ro L _ O J CM 8 ro 0 n ro i o 0 j> ro "ro 0 T J T J 0 -*—» _ro _o ro o " r o T J ' > T J C 0 X J ro 0 CO E 0 x 0 ro 0 CL CO .9 o CM CO 0 > C L O F C N E CM O J CO 0 > "5 > c o E £ CO CO ! * . £ LO E r-00 co 0 a: ro CL ro CL A : ro CL ro CL o ro CL ro CL * CM * * r~- CD C D I— C D C O T ~ T— C O t - T - CM CM C N C O 00 00 ^ CO ^ C D LO CM LfJ CO LO O C ) T ^ ^ C M C M C N C 0 C M C N L O CM C O O O J CN * * r~- * O J d co O J C O L O LO LO C N CM CD <*> CO CM O 00 CM i -CM CN LO LO LO CM T " 9 ( N 00 O J LO CO LO LO o 9 LO O CD LO r - - CO °9 co O J •sf t ^ o j o o o ^ ^ r t i r- -s f CM T - CM C O CO CM r- -sr r - r-^ 6 d d i-co •sf CD O J C O C O CM T-^ CM CO 1^  C O C D CM T-^ LO '. CD CO C O C O C O C D LO O J d d 00 CD CN -sf •sr co d d co co co r-^  •sr o d T -o 0 xT CO or CU Q O 2 z £ <: i o c ro Q CO 135 o CO 0 Q. co o CD O CM CO O O L O T -CM CO •sr 1^-LO CO o o CO •sr CD CM CO r-CM CO CO CO LO CM CO CM o CM CO 0 to o \ cu X 0 0 > to to 8> O J O CD c 3 T J T J 2> 3 to CO 0 E CO -*—' CO T J CM o o CO 3 -g > T J _ C LO 0 X J CO 0 X 0 CO 0 0_ co 0 3 £ O E ° CM E CO CO 0 cr: CO to I £ C LO = CM E CM CO CO 0 > 3 ^ c o LO to 0 _ 1 2 I £ CO CD X CO CL CD X E E co CL O J X E E co CL O J X E E CO CL O J X E E co CL O ) X E E CO CL O •sf N ( O O O ) CO CO CO CO co OJ •sf O T -•sr -sr •sf CM CO CO LO O C O C D r ^ - C O O O C O C O C M • s T C O C O C O C O C O C O ' s r c O C O O J C O O C N C O ' s r c O C 3 J - s r T - T - r - - C O i r i ^ ^ ^ u i i o ' ^ ^ ^ ^ i r i i r i i r i ^ - ' CO CO o CO o •sr o •sr CO CO •sr •sr co OJ •sf h~ CO co -sr OJ i^-•sr iri CO LO 00 CO LO LO LO LO CO CD LO CO CO CO CO oo r -•sf -sf •sr •sr OJ LO CO CO •sr •sf LO LO O co co -sr LO LO r~- I— CO T— <sf -sf LO LO •sr -sr •«-CO CO r j -LO LO LO •sf "ST LO CM CO CO •sf O O J O J O O O J T - O C O O J • s r c o c o c o c o - s r - s r - s r c o C O C M C M T - C M L O C O T -L O L O L O L O L O L O L O C D CM iri C O C M O O C O C M N - N - T - r ^ - ' s r i - C M C D O c o - s r c o c o - s r c o c o - s r c o c o - s r - s r - s r ' s r CO O LO O J co co -sr co O O CM •sr co iri C 0 C O r - L O O J r ~ - L O T - L O c o c o c o c o c o c o c o - s r c o • s r c O O J r - C M O J r - L O r -• s f - s r - s r - s f i r i - s f - s f L r i ' s f CO CD CO CD iri CO CD CO CO CO iri co CD CO CD CO O J LO •sf Ci CO •sf r- CD •sf d °9 co CO ^ CO CO LO' d o CO •sr d O CD iri d 00 CM oo -sr •sf d o 0 x7 3 CO * ^ % % 5 CO 2 K Q O 2 LU X. £ z £ ^ x w o CO c CO Q J> CO cu o Q_ CD CU o o o o -sr to C D o L O T— r^ - co C N C O C N C O T - v -O - S J - T - N - C O L O T - O 0 ( D i - N n ( N N N c o c N c o c N c o c o r g c o cu CO o 1 cu X cu 03 0 0-co 0 _ | I co c L O C N C N to 0 > c O L O | C D T _ CO 0 1 * = L O E C O co or o 0 IS" CO C O C O O O N S S N 0 i n s f ^ M 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) ° > C O L O • C O ^ C O C O co co r - L O O ) C O •sr co M . L O C D • C O ^ C O C O N C O N N N C O N W © 0 ) 0 ) 0 ) 0 0 ) 0 ) 0 ) 0 ) 0 ) O ) co co co oo oo C O C O CO r- r^ - h- r~-0 ) 0 ) 0 ) 0 ) 0 ) O ) r-- oo C O C O CO C O N C O C O C O C O L O C O L O 00 C O 00 C O C O C O S OO C O O ) , C O C O O ) O ) C O C O 00 C O CD CN CO LU ^ CO c CO Q 0 CO 137 CD .co CO $ .2 co co & O) 2 CL O) . C 41 co CO CD CO •§ CM o "cc T3 s •5 c 0 -O CO r-0 5 o 0_ CO CD CL CD CO O i_ 0 X 0 co 0 CL CO 0 -11 CN O) o CO LO CN CN CO c o co co 0 _ 1 * .b LO CO 0 o 0 157 3 , C O S S C D C O N ( D l O m C D i n i f l L O C M 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0) O O O "fr CO O LO T -CN CO CN CO T -L O O " f r t - r - - c o L O T - o O I O C D T - N O N N C M T - C O C N C O C N C O C O C N C O O) CO O) CO O) LO O) CO LO O) O) "fr O) CO CO N N O) O) O) O) N N C O N N C O C O C O C O 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) C D C O S C O O O S N N C O C O C O C O C O C O 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) C D f f i C O C O C O N C O N C O C O C O N N C O 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) C O C O O ) N N ( O C O C O C O C O O ) C O 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) 0 ) CM CO LU ^ CO O) LO C) CO CO LO O) O) CO O) 00 CO O) O) O) ^ O) O) ° c CO Q ® CO CO CO 5 c g ro ro co ' c 0 o L— CD 0 E X o 0 w 3 Q. CN o O-138 CO .co 00 oo > 55 co Co 8 CL Co .c C: 41 To co CO 00 s CO 41 CO o o "co 41 s T5 c CO _CD CO o CO CD CL CD W o k CD X CD CO CD CL CO p E o CN ° ° CO 1 s C L O E CN t CM CD CO CD > c o E L O CO CD _ 1 ^ CO co CD CL" o 0 id? 3 CO 0 ) 0 0 0 " t L O O " f T - h ~ C O L O T - 0 fflOlOrSOOCD'-NCONSCM C N C O C N C O T - T - C O C N C O C N C O C O C N C O T - C O ( D C O T - C M ^ O ) O ^ S I O C O C N T - T - T - C N C N T - T - C N T - T - T - T -co CN CO CO CO O CN CN CN T - CO T - LO O ) (D CN CN T - CN C N T - C O C O O L O T J - C O C O C N C N C N C N T - C N C N C N C N CO CN CN CN CN \ r C D ^ C D I O C O ^ W l O O ( D l O S C D C N C N C N C N C N C N C N C N C N C N C N C N C N C N CN \ t l O i n N ( O T - C O I f i ( M C O C O O ) N N C N C N C N C N C N C N C N C N C N C N C N T - C N C N LO •sr CN CO CN CN C I C D S C O C O C D C O ^ m N C D t C N C N C N C N C N C N C N C N C N C N C N C N LO •sf CN O ) CN CO LU l i s » d t Q CO co -*—• -•—* CO 5 CD CD Q. CO _CD O E c 0 •+-» CO <i ro o IJ3 CO o o I 139 0 .co 0 $ CO co Cn 8 Q. CO .c c 41 "CJ e CO co 0 E CO 41 UJ QQ "co 41 s c CJ) 0 .o. CO 0 o D_ CO 0 0 CO o 1— 0 X 0 ^ CO 0 Q . (O £ o (O c m E CN C= CN O) if) 0 > c o 1 £ CD CO 0 _ 1 * CO CO 0 o 0 CO o o o o ' d - i o o ' s i - T - h - . c o m T - o ( O O l f l i - N O l O l O r - N n N N N CNCOCNCOi-t-COCNCOCNCOCOCNCO m CD in "fr £2 CD co i i i ~ • • • f r O CN O 00 CD CO ll CD m m £2 i i T m O CD CN It) N v- N o CO CN CN CO O^rCN^COT-CNCNCNCOcNCO CO i CN CN CO CN CN O CO CN L t L t • O CO LU ^ | i x ^ d CO " f r T -9 ^ °? co co r^ . <N CN T - CO O CN O CD CD CN CN o 9 CN c CO Q £> CO co -§-» CO 5 0 0 Q. CO 0 o E CO CO 0 o X 0 0 CO ro .Q LU CQ 140 CD 5 o D_ CO CD D_ CD O O O "t LO CO O LO T— I— o > CM CO CM CO T - T - o t i - N c o m i - o O C D T - S C O N N C M C O C N C O C N O O C O C N C O CD CD 3 .2 co co Co 8 Q. CO .C c: T3 £ 3 co CO 00 s CO H—1 41 a: a "co 3 T3 s •5 c o CM Q) XI CO CD CO CO l _ CD X CD co CD CO CD < 3 c E CM CO CD > C LO E CM = CN OO CO CD > 3 > c o 1 £ CO co CD _ 1 Z I K CO co CD a: CO LO h-CO CM CM •sr •sr h-l 10 00 -sr •sr -sr -sr h-l h-l o •sr h~ h~ h~ h-CO 00 CO •sr •sr O CO CM h- LO CO h-CO CO CN CO CO CN CN CN CN CN O h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l CD LO O O CN •sr O h-l h-l CO CO •sr CD 0 CD CO CM CO CO CO CN CO O h-l h-l h-l h-l h-l h-l CD co o -sr co co -sr co -sr r-l h-' h-l h-l LO CO CJ) CN O OO T CO h~ •sr CO CO CO CO CO •sr CO ro CO CO 0 h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l 0 •sr r-l •sr h-l 00 0 CO O) •sr CO O) cn CN CN CO •sr •sr CO •sr CO •sr CO CO •sr •sr O h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l CD cn CO 0 •sr CO •sr LO •sr CM •sr h-00 O) CO 5 O) CO LO •sr CN •sr CO 0 h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l c i •sr •sr h-l co r-l •sr h-l •sr h-l CO •sr LO •sr 0 •sr •sr •sr CO •sr O) CO CO •sr O) •sr 5-O •sr CO •sr CO 0 h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l h-l 0 ' o CD 15? 3 CO •>- CN —i —> a o co LU ^ CO c CO Q £ CO 141 co o D_ A : co 0 Q_ cn o o o -sr i n co o L O T - r - o> oj co C N ro T - T -O ^ f T - r - C O L O T - o O ( D T - S C 0 N S C M O O C M C O C M O O C O C M O O CD .CO CD $ 55 co & CD e CL CD .C -§ S CO CO CD £ "to CD O CO CD 8> CD CD T3 CO e =5 2 CD CL s CD O O "co :§ s "5 c CN 0 XI CO 0 CO CJ s_ 0 X 0 co 0 CO 0 _ I! ^ 8 to I £ c LO E CN «= CN O) to 0 > c o C L O CO 0 _ 1 £ I 12 CO to 0 or c N L q L q c o o c q c o c o c j » L o c o c o c o c o N N N c d o D N N N N O D N N N t O c o c o c o c o c o c o c o c o c o c o c o c o c o c o L O CO CO CO CO T - T j - C N co co co o oo CO O O ^ N O l t N l f l N C O r - ^ r - ' r ^ r ^ r - ^ r ^ r ^ r ^ c b C O C O C O C O C O C O C O C O C O c o q s c o o ) N ( 3 ) 0 ) ^ ( D O c o c \ i i n c b s d s N s d c o s N N s s c b C O C O C O C O C O C O C O C O C O C O C O C O C O C O C D O ) C N L O r - 0 ) C D O C O C O L O T - C O O CD c d c O N N C O C D C O N S c b s N c d c o c o c o c o c o c o c o c o c o c o c o c o c o c o c o r ^ r ^ L o r ^ r ^ - s r o o o L O - s r c N N -I d i r i i o ' c b s c d c d d s s i r i c D N C D c o c o c o c o c o c o c o c o c o c o c o c o c o c o CO CO CD d CO CD LO r- 00 CD CD CO CO CO CO CO CO L O d £ ° CM co d CD d co CD d LO CD co d o 0 x7 3 CO T - CM or or Q o co 2 UJ * z iS2 2 o {2 Q CO 142 o _CD 0 CO i LU CO co c g -*—. ca i_ c 0 o c o o _0 CL E CO CO CO E CO _ro Q . 0 co o i_ 0 X 0 i "co o CL TJ c CO 0 1 " r o 3 TJ > TJ C M C M _0 X5 CO 15 0 v IQ _ CJ) c X LO 0 CO X c <P CT) "Jo C o C L ^ C M X LO 0 CO o -I— 0 X 0 CL > O Q o Q o CM o 0 1Q D CO O 00 CO -sr CM CM CO T -^LO ^ T -CO^CN-s fCOcO •sr^ l ^ r o L O L O 'mCbr^LO^} : • V ^ ' C N ^ L O T - C M C V , LncDiococoLncn^cocnT-i^cooo C O r ^ C M p ^ ^ C M c ? i r j O ) 0 0 T l - C T | h ~ •sr O O CO CO CD O) h~ CM x h~ CD LO 00 i - CN OO h- LO •sr •sr CO CN CO •sr •sr •sr •sf h-l CO CO 00 •sf •sf h-l CO h-l CO CD •sf CM CM CN CN "~ LO CN h~ CM CN CN CN •sr CO O O CD LO LO h~ •sr oo CM •sr CO CN CO CN •sr h- CN h- x— CD •sr CN LO h- 00 CD CD OO •sr h~ CD CD oo LO CO •sr CD 00 OO •sf •sf LO •sf CO T~ CM •sf LO LO •sf LO 00 CM CN CD CN h- CO CO O o h- 00 CD h-l CD CN CM CM CO CD ^—'• CD CD ci CM CN CN T _ T _ I N CN CN T _ CO •sr CO oo CD CD h- co CO o CD LO CD CO •sr CO CM T— CO h~ LO LO CM h- CO CO T— LO LO x— v- T— O CO o T— •sr T— x— T— CN CD CD CD CD CD CD CD CD CD CD CD CD CD CD LO O O CD •sr CO •sr •sr CD LO OO OO CN CO O •sr T— CN CO CO LO •sr CO •sr CD O CO CN T— T— T— o CO o T— LO T— T— T— CO CD CD CD CD CD CD CD CD CD o CD CD o CD O LO h~ CO o CN CN CD CD •sr O h~ LO CO O) CO O h- CN LO CD CO CN 00 T— h- 00 O 1— o CO O O CO i — O T— T— T— CD CD CD CD CD CD CD CD CD CD CD ci CD CD CD ^ _ •sr ^ _ CD O CD O 00 CN CD o h- 00 "- CO CD CD LO CO CN 00 h- CD •sr CD CO CD CD CD LO •sf CD h-l •sf LO •sf h-l CD CD LO CN CN CN CN CO T~ CN CO CN T ~ "~ CO CO 00 CD CN CN O oo CD CO CO co O) CO CD CO CO CO 00 CO LO LO CO CN CD h- h- CO O 00 T— CD "- •sr oo CN O) •sr CD 00 •sr CO LO •sf •sf h-l CN LO •sf CO CO h-l h-l 00 CO CO CN •sr CD T - LO CN CO LO h- T -T— IT* h-l /-vi 1— CN LO LO cri CD o" CN 1 T _ T _ T _ CN CO t — LO o CO •sT h- CD co ^ CO CD CO CD oo CO CN CD o LO CM h- CN h- 00 CN CD O o "- O CN O co o O CM CD CD CD CD CD CD CD CD CD CD CD CD CD CD h- CN h- CN •sr LO O CN •sr 00 CD CD O h- CD CO CO CO o LO •sr CO CO LO CD O i — O O O CN O CO o O CM CN CD CD CD CD CD CD CD CD CD CD CD CD ci CD LO CO CO LO LO o h- oo •sr OO h- CN CO CN O h- CN x— O T— •sr T— O 00 T— CO CO T— O T— T— CN o T— o T— CM r — CD CD CD ci ci CD CD CD CD CD CD ci CD CD CN ^ 2 CO ^ LU ^ 2 L L L L Q 0 5 ^ Z $ 5 I - O | -co CO CD CO CN •<- CD •sf h-l L O -sr h- T -h-l CO T -CN LO CN t -LO T -C D -sr h-l C N C N f -O) c ro Q °> CO c g TJ LO X LO 0 = 0 E 22 c o 0 ^ °- 0 co ^ P <o o C CL 2 «= -.2 c" o o " 5? ° ro c 0 0 l " g "o O ' , E <? ro ~—-CO fl) > ^ O o co *J ro S .E 0 CD TJ D) S 5 o ro O Q 143 o _0 CD CO i CL CO co c o "«+-< CO i _ c cu o c o o 0 Q. E CO CO CO E CO _ro Q. 0 co o 1_ 0 X 0 I CO o CL TJ c CO 0 \ C L "co TJ ' > TJ C CO CM 0XJ CO CO 0 v Q 1. OJ C 0 co 0 X 0 co o CL X o OJ Q O Q O x o 0 co o ' 0 X 0 0 ICL > O Q O CM Q O o 0 x i CO °° £ <9 O J o» c _ OJ O CM ,ri OJ K J CO CO CM CD 1 0 OJ ^ V O J O O I T - N ^ CO CN LO CO CM •sf + ( D N •sf co CO CO CO CO CO CO CD LO CM CO - £ O L 0 r a c r i 0 S C 0 L 0 C 0 l 0 r s j 0 ) ' S f CM CM co CD 1 r v l ^ — CO CM CO CD CO CM 00 LO CM oo LO CO LO o 00 LO r- o 00 LO o /—» OJ CM CO •sf o CO CM LO CD T— LO o O CD CO d •sf CM CO •sf d 1^  CM •<— •sf d CO 00 • 1 CO CO CO 00 OJ CO OJ OJ CD NI OJ CN CO -sf CO CD CN CO CO LO CD CM LO CO 00 o T— •sf o r"- •sf o CO 00 d CO CO d OJ CM OJ d r- oo CO -sf > CO CM CO CM CN CN in o CM •sf •sf o o •sf CM CO CM d •sf T~ •sf iri d •sf CO CM c CO CO LO •sf CO OJ CD 00 CN OJ CD CM •sf r-CO r - CD CD OJ CN •sf r- CN in oo •sf CO •sf 0 O O o o T— o o o o o o T— E d d d d d d d d d d d d d d LO CM •sf •sf CM CO m OJ 00 _^ CM r- CD CD co •sf 00 CD 00 CO T— in o O o o o o o * o o o d d d d d d d d d d d d d oo 00 _^ CN CO in CO CN •sf CD •sf o CM CD CD CD OJ •sf CO •sr CN in CO •sf m •sf o O o o T— o o o o o o T— d d d d d d d d d d d d d d in CO in CM CO r- CO m in CO in O •sf CM OJ 00 CM •sf CO CD 00 CM CO h~ O d •sf LO 00 d 00 d •sf CO •sf 00 d d d CD OJ •sf OJ CO OO CD OJ CD CD in CO LO CO in CM CO f- CO in m CO LO o o CM r- 00 CM •sf CO co CO CM CO r- o CO -sf *- 00 OJ CO CD •sf CO •sf T— CO CO d d •sf d d d d d d d iri T— d CO h-d iri t - co iri d CM CM O CO CD CO CM CM r - CD OJ OJ 00 CD CO CO C O r o o O N O j T - C O C D O J C O C O O J T - C i ) • s f O O C N C O - s f N - L O - s f l ^ - s f T f C O T - C D O O O O O O O O O O O O T - O •sf •sf OJ LO CM •sf m CO •sf o m OJ CO LO •sf CO CM OO in r- LO •sf •sf •sf CO CD O o o o o o o O o O o o T— O d d d d d d d d d d d d o d CN OJ in o OJ •sf m CO CO CO •sf OJ CN CO •sf CO CO •sf 00 •sf •sf CO o O o o o o o o o o o O o ^ — o d d d d d d d d d d d d d d ^ Z Z ^ ^ ^ C O ^ L U ^ ^ - O c n - I CO CD CO CO ^ in oo ™ CN OJ OJ d CN CN LO d CM c CO Q £ CO c o _3 TJ O X o 0 — 0 E o 0 g a- co co F OT 2 g. o £ C CL « c -.2 E c g>8 c o o o >2 H— • — ° to c 0 0 '-O "O o ^ ° L i E !? co ^ co cu „ -I—« > ^  O o co -w CO 10 0 l ] ° O J o 2 o. " ° . i2 Q ^ O Q 144 Table 24. Individual serum sample concentrations of 17(3-estradiol Subject 1 2 MEAN OD CV [MEAN] 100X pmol-L-1 MK 413 382 398 5.5 22.16 2216 81353; RJ1 430 405 418 4.2 <MIN <MIN <MIN RJ2 376 392 384 2.9 20.95 2095 76911 DW 390 397 394 1.3 14.73 1473 54073; OM 409 464 437 8.9 <MIN <MIN <MIN MS 400 427 414 4.6 <MIN <MIN <MIN KM 406 414 410 1.4 <MIN <MIN <MIN NE 442 472 457 4.6 <MIN <MIN <MIN WK 421 427 424 1.0 <MIN <MIN <MIN WJ 447 398 423 8.2 12.22 1222 44863; HJ 424 343 384 14.9 79.67 7967 292473; IS 425 393 409 5.5 14.88 1488 54633; CL 395 400 398 0.9 13.75 1375 5048$ TS 455 399 427 9.3 11.75 1175 4314$ Mean 417 408 412 5.2 23.6 2359 8658 SD 22.6 32.5 20.6 4.0 21.4 2143.8 7870.0 OD, optical density at 405 nm; CV, coefficient of variation; [mean], mean concentration in picograms per millilitre; 100X, correction for 1:100 dilution; pmol-L-1, picomoles per litre; <MIN, less than minimal detection limit of 37 pmol-L"1; $ poor replicate 145 Table 25. Individual serum sample concentrations of progesterone Subject 1 2 MEAN OD cv [MEAN] 10X nmol-L"1 MK 474 435 455 6.1 7.39 73.9 0.24* RJ1 460 440 450 3.1 5.42 54.2 0.17* RJ2 403 442 423 6.5 12.69 126.9 0.404: DW 443 441 442 0.3 4.5 45 0.14 OM 449 487 468 5.7 <MIN <MIN <MIN MS 437 470 454 5.1 6.68 66.8 0.21+. KM 459 458 459 0.2 <MIN <MIN <MIN NE 456 497 477 6.1 <MIN <MIN <MIN WK 481 494 488 1.9 <MIN <MIN <MIN WJ 476 456 466 3.0 <MIN <MIN <MIN HJ 447 397 422 8.4 13.35 133.5 0.42$ IS 460 455 458 0.8 <MIN <MIN <MIN CL 439 438 439 0.2 6.16 61.6 0.20 TS 449 487 468 5.7 <MIN <MIN <MIN 452 457 455 3.8 8.0 80.3 0.08 19.6 28.0 18.8 2.8 3.5 35.4 0.10 OD, optical density at 405 nm; CV, coefficient of variation; [mean], mean concentration in picograms per millilitre; 10X, correction for 1:10 dilution; nmol-L"1, nanomoles per litre; <MIN, less than minimal detection limit of 0.008 nmol-L"1; +. poor replicate 146 SCD62E (ng/mL) 2.0 n 0 50 100 150 SCD62P (ng/mL) Figure 26. EIA standard curves for: A. soluble E-selectin (sCD62E), B. soluble P-selectin (sCD62P). 100 90 80 70 E 60 o 50 m m 40 30 20 10 0 I t J • t t *v s j N j | 10 10000 100 1000 17B estradiol (pg/mL) (y=-10.69 Ln(x) + 116.96, ^=0.9511) 100000 100 90 80 70 60 o 50 m m 4 0 30 20 10 0 S < V < > > 10 100 1000 Progesterone (pg/mL) (y=-12.656Ln(x) + 103.22, ^=0.9387) 10000 Figure 27. EIA standard curves for: A. 17p-estradiol, B. progesterone. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0077067/manifest

Comment

Related Items