Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of menstrual cycle phase on diffusing capacity of the lung Bacon, Catherine 1997

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

Item Metadata

Download

Media
831-ubc_1997-0503.pdf [ 8.51MB ]
Metadata
JSON: 831-1.0077155.json
JSON-LD: 831-1.0077155-ld.json
RDF/XML (Pretty): 831-1.0077155-rdf.xml
RDF/JSON: 831-1.0077155-rdf.json
Turtle: 831-1.0077155-turtle.txt
N-Triples: 831-1.0077155-rdf-ntriples.txt
Original Record: 831-1.0077155-source.json
Full Text
831-1.0077155-fulltext.txt
Citation
831-1.0077155.ris

Full Text

THE EFFECT OF MENSTRUAL CYCLE PHASE ON DIFFUSING CAPACITY OF THE LUNG by CATHERINE BACON B.Sc, The University of Otago, 1991 B.Phed. (Hons.)/ The University of Otago, 1993 A THESIS SUBMITTED IN PARTIAL FULFILLMENT 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 September 1997 © Catherine Jane Bacon In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library, shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department, or by his or her representatives. It is understood that copying or publication of; this thesis for financial gain shall not be allowed without my written permission. Department of HoMAW ICtfJCTiCS The University of British Columbia Vancouver, Canada Date ^ ScSpreMfe^ i ^ f DE-6 (2/88) ABSTRACT Pulmonary diffusing capacity (DL) has been observed to decrease during menses. Nonetheless, a descriptive study of alterations in this parameter wi th menstrual cycle phase has not been completed and the mechanism of change is not clear. Changes in resting single-breath diffusing capacity of carbon monoxide (DLQO), and in its two components: pulmonary capillary blood volume (Vc ), and membrane diffusing capacity (DM ) were measured in 13 normally menstruating women at points within the menstrual cycle chosen to best discriminate between the effects of oestradiol, progesterone and prostaglandins. In addition, haemoglobin concentration ([Hb]), packed cell volume ( P C V ) , and percent of carboxyhaemoglobin (COHb) were measured. Measurements of DLQO, V C , D M , and [Hb] were undertaken at five testing points throughout three menstrual cycles, whilst C O H b and P C V were assessed at four points within one cycle. The phase of the menstrual cycle was determined by quantitative analysis of basal body temperature recorded daily by subjects. N o changes in resting Dlco, D L C O divided by alveolar volume ( V A ) , C O H b , P C V or for [Hb] corrected D L C O , D L C O / V A , D M or V c were found using one-way repeated measures analyses of variance ( A N O V A s ) of the most representative ovulatory menstrual cycle for each subject. Two-way repeated measures A N O V A s of D L C O and D L C O / V A ; and [Hb] corrected D L C O and D L C O / V A , which separated the effects of the five testing points and the ovulatory or anovulatory status of a menstrual cycle were also performed and no significant changes were observed. When the effect of the large hormonal differences between an ovulatory and an anovulatory cycle were removed, a trend towards an increase in D L independent of the effects of [Hb] at mid-cycle and during the luteal phase compared to the early follicular phase were observed. ii Notwithstanding the extreme variability of hormonal changes within the human menstrual cycle, without the benefit of hormonal analysis we have not found consistent alterations in D L C O with menstrual phase in normally menstruating women. This is despite careful effort to time diffusion test with points in the cycle that should best discriminate between the hormonal effects of oestradiol, progesterone and prostaglandins. iii TABLE OF CONTENTS Abstract ii Table of Contents iv List of Tables vii List of Figures viii Acknowledgment ix CHAPTER ONE: INTRODUCTION 1 Cyclic Alterations in Diffusing Capacity 1 Respiratory Changes Within the Menstrual Cycle 3 Circulatory Changes Within the Menstrual Cycle 3 Further Explanations of D L C O Changes During Menses 5 Limitations of Previous Work 6 Research Questions 6 CHAPTER TWO: METHODOLOGY 9 Ethical Approval 9 Subject Recruitment and Selection 9 Documentation of Menstrual Data 10 Timing of Testing 11 Documentation of Exercise 15 Testing Protocol 15 Measurement of Height, Weight and Anthropometric Variables 16 Measurement of Resting Forced Vital Capacity and Forced Expired Volume in One Second 18 Measurement of Resting Diffusing Capacity 18 Calculation of Diffusing Capacity 20 Correction for Haemoglobin Concentration 21 Quantification of D M and Vc 21 The Reliability of Diffusing Capacity and its Components 22 Collection and Batching of Blood Samples 22 iv Measurement of Packed Cel l Volume and Haemoglobin Concentration 23 Data Analysis 23 Criteria for the Selection of the "Best" Cycle for Analysis 24 Statistical Analysis 25 C H A P T E R T H R E E : RESULTS 27 Recruitment of Subjects 27 General Subject Characteristics 27 Subject Act iv i ty 28 Lung Function Characteristics 28 Analysis of Cycles 30 Analysis of Test Points 30 Blood Sample Analysis Testing Dates 32 Best Ovulatory Cycle Analysis 34 Ovulatory Versus Anovulatory Cycles 38 Changes in Carboxyhaemoglobin with Menstrual Cycle Phase 38 Changes in Packed Cel l Volume With Menstrual Cycle Phase 42 Reliability of the Packed Cel l Volume Measurements 42 Reliability and Val idi ty of the Haemoglobin Concentration Measurements 42 C H A P T E R FOUR: DISCUSSION 43 Recruitment and Cycle Characteristics of Subjects 43 Diffusion Changes Within the "Best" Ovulatory Menstrual Cycle 44 Haemoglobin Concentration Changes With in the "Best" Ovulatory Menstrual Cycle 45 Diffusion Changes in Ovulatory Versus Anovulatory Cycles 46 Further Mechanisms of Diffusion Change 48 Reliability and Validi ty of Packed Cel l Volume and Haemoglobin Concentration Measurements 49 C H A P T E R FIVE: C O N C L U S I O N 51 v References Abbreviations 52 56 APPENDICES Appendix I: Pulmonary Diffusion: Its Measurement, Components and Determinants 57 Appendix II: The Human Menstrual Cycle 67 Appendix III: Calculation of Intraclass Correlation Coefficient 70 Appendix IV: Ethical Approval 72 Appendix V: Poster Advertisement for Subjects 76 Appendix VI: Menstrual Cycle Diary 77 Appendix VII: Daily Exercise Record 79 Appendix VIII: Initial Questionnaire 81 Appendix IX: Data Sheets 88 Appendix X: Final Questionnaire 91 Appendix XI: Timetables of Testing Dates 104 Appendix XII: Raw Data Summary 115 vi LIST OF TABLES Table 1. Lung Function Characteristics of Subjects. 29 Table 2. Summary of Menstrual Cycles. 31 Table 3. A n Analysis of Test Points. 33 vii LIST OF FIGURES Figure 1. The Timing of Testing at Points in the Menstrual Cycle. 13 Figure 2. Diffusing Capacity Measured at Five Test Points Wi th in Ovulatory Menstrual Cycles of 13 Women. 35 Figure 3. Diffusing Capacity/Alveolar Ventilation Ratio Measured at Five Test Points Wi th in Ovulatory Menstrual Cycles of 13 Women. 35 Figure 4. Haemoglobin Corrected Diffusing Capacity Measured at Five Test Points Wi th in Ovulatory Menstrual Cycles of 13 Women. 36 Figure 5. Haemoglobin Corrected Diffusing Capaci ty/Alveolar Ventilation Ratio Measured at Five Test Points With in Ovulatory Menstrual Cycles of 13 Women. 36 Figure 6. Haemoglobin Changes Measured at Five Test Points Wi th in Ovulatory Menstrual Cycles of 13 Women. 37 Figure 7. Diffusing Capacity Measured at Five Test Points Wi th in Ovulatory and Anovulatory Menstrual Cycles of 6 Women. 39 Figure 8. Haemoglobin Corrected Diffusing Capacity Measured at Five Test Points Wi th in Ovulatory and Anovulatory Menstrual Cycles of 6 Women. 39 Figure 9. Diffusing Capacity/Alveolar Ventilation Ratio Measured at Five Test Points With in Ovulatory and Anovulatory Menstrual Cycles of 6 W o m e n . 40 Figure 10. Haemoglobin Corrected Diffusing Capaci ty/Alveolar Ventilation Ratio Measured at Five Test Points Throughout Ovulatory and Anovulatory Menstrual Cycles of 6 Women. 40 Figure 11. Percent Carboxyhaemoglobin Measured at Five Test Points Wi th in the Menstrual Cycles of 7 Ovulating and 3 Non-Ovulat ing Women. 41 Figure 12. The Human Menstrual Cycle. 68 v i i i ACKNOWLEDGMENT I wou ld like to thank Drs. Jerilynn Prior, Don McKenzie , Raja Abboud and A l a n Mar t in for their advice in the various stages of putting this thesis together. I wou ld also like to acknowledge the assistance of Bi l l Sheel and Drs Jim Potts and Angelo Belcastro. Finally, my deepest gratitude goes to the subjects involved in this study: those who provided insight with their questions, those who maintained enthusiasm for the project even when mine was waning, and those who waited expectantly for the appropriate stages of their menstrual cycles, left messages of their progress for me on answer machine and voice mail, and turned up to the laboratory at all hours of the day for testing. ix CHAPTER ONE: INTRODUCTION The movement of metabolic and other gases across the alveolar membrane of the lung is a vital physiological process which takes place solely via passive diffusion. The physical properties of the alveolar membrane, the respiratory-cardiovascular interface, in particular its extreme thinness (around 1pm) and vast surface area (approximately 70m 2) allow this exchange to take place freely in a healthy lung. It is doubtful in fact that lung diffusion limits the transfer of oxygen (O2) from the atmosphere to metabolising tissues in healthy, untrained individuals, either in rest or during maximal exercise. Despite this, under certain circumstances lung diffusion may limit oxygen delivery, for example in diseased lungs (Crapo and Forster, 1989), at high altitude (Hastala and Berger, 1996), or in elite athletes during maximal exercise (Dempsey, 1986). The capacity of the lungs to transfer O 2 from the lung alveoli to the red blood cell haemoglobin (Hb) in the pulmonary capillaries is called pulmonary diffusing capacity ( D L ) . D L is usually measured clinically by determining the rate of disappearance of a known concentration of carbon monoxide from a single breath of a mixed gas: hence the name for the measurement: "diffusing capacity of the lungs for carbon monoxide" ( D L C O ) " - For a substantial explanation of the basis of lung diffusion measurement and calculation refer to Appendix I: Lung Diffusion: Its Measurement, Components and Determinants (p48). C Y C L I C A L T E R A T I O N S I N D I F F U S I N G C A P A C I T Y Because D L C O is an important clinical measurement, often measured serially in one patient to assess changes in pulmonary function occurring as a result of pulmonary disease, it is important to be aware of other factors that may alter lung diffusion on a day to day basis as well as any cyclic variations in D L C O D L C O has 1 traditionally been thought to undergo circadian variation, because it has been shown to decrease in individuals of both sexes by 1 to 2 percent per hour from morning to night (Cinkotai and Thompson, 1966). Nonetheless, a cyclic circadian pattern of D L c o variation has not been demonstrated. Moreover, a more recent study (Frey et al., 1987) found no change in D L c o at different times of the day after correcting it for COHb backpressure (an artifact of repeated diffusion testing), and for small circadian changes in haemoglobin concentration ( [Hb] ). There are two ovarian phases of the human menstrual cycle The first, the follicular phase, begins on the first day of menses and lasts until ovulation. The subsequent luteal phase lasts from ovulation until the beginning of the next cycle and is characterised by vastly increased production of ovarian steroid hormones by the corpus luteum, a remnant of the ruptured ovarian follicle. For a complete review of the human menstrual cycle refer to Appendix II: The Human Menstrual Cycle. A change in D L c o with menstrual cycle phase might be expected from the recent results of Sansores et al. (1995). These researchers noted a 9.5% decrease on the third day of menses relative to a premenstrual baseline measurement. In this study, 14 healthy women (including one smoker and eight who were taking oral contraceptives), underwent six single-breath carbon-monoxide lung diffusion (DLco) measurements. A baseline measurement was obtained premenstrually (1 to 7 days before menses), then repeated measurements made on each of the first 4 days of menses, and on one occasion following menses (5 to 10 days after the onset of bleeding). In ten of the 14 women the two components of DLco, pulmonary capillary blood volume (Vc ), and membrane diffusing capacity (DM), were determined via duplicate D L c o measurements at two different O 2 fractions. The authors observed no significant changes in Vc, D M , or [Hb]. The mechanisms of this diffusion reduction are difficult to explain. D L c o is a function of both pulmonary and circulatory parameters and will alter with changes 2 in the diffusing properties of the alveolar membrane (DM), the pulmonary capillary blood volume (Vc), or the reaction kinetics of carbon monoxide (CO) and Hb (expressed as theta or 6). These might alter as a consequence of respiratory alterations in the luteal phase or as a direct hormonal effect. RESPIRATORY CHANGES WITHIN THE MENSTRUAL CYCLE The luteal phase of the menstrual cycle is associated with an increased resting and exercise minute ventilation (VE), (Schoene et al., 1981 and Jurkowski et al., 1981), raised resting hypercapnic and hypoxic drives (Schoene et al., 1981; Dombovy et al., 1987) and greater inspiratory muscle endurance (Chen and Tang, 1989) compared to the follicular phase. A potential mechanism for DLco changes is not apparent from established alterations in any of these respiratory parameters. It is possible that mild respiratory alkalosis occurring as a result of increased V E might have a direct effect on 0. Deep breathing also has the potential to elevate DLco by raising the compliance of previously under inflated alveoli and consequently increasing the exchange surface area. Whilst large inspirations prior to a single breath might therefore raise DLco measurement, it seems unlikely that the small increase in tidal volume occurring in the luteal phase of the menstrual cycle would have a physiologically significant or even detectable effect on lung diffusion. CIRCULATORY CHANGES WITHIN THE MENSTRUAL CYCLE A substantial (17%) elevation of Vc (Seaton, 1972), and small (2.6%) increases in [Hb] (Jurkowski et al., 1981) have been observed in the luteal compared to the follicular phase of the menstrual cycle. The cause of these alterations is unclear but the timing of changes implicates a hormonal mechanism. 3 Progesterone mediated increases in V c may have occurred during the premenstrual measurement in the Sansores et al., (1995) study. These researchers regarded hormonal changes as a possible explanation for the D L C O reduction they noted during menses. Although alterations in V c in the same study were non significant, the sample size for the assessment of V c changes was small and a corresponding alteration in V c might not have been detected. Progesterone levels increase around 10 to 40 fold in the luteal phase of a menstrual cycle in which normal ovulation occurs. A progesterone mediated change in D L C O wou ld therefore be expected to occur in the luteal phase of a normal ovulatory cycle, but not in an anovulatory cycle. The difference in diffusion alterations between menstrual cycles of differing ovulatory status has not been investigated. H i g h levels of oestrogen which occur immediately prior to ovulation in a normal menstrual cycle and following the administration of synthetic oestrogen, prevent the plasma volume drop associated with bed-rest (Fortney et al., 1988). Al though unsubstantiated, a potential mechanism for this oestrogenic effect on plasma volume may involve the cardiovascular vasodilator nitric oxide (NO). Expired N O levels were shown in one study to increase almost threefold from menses until days 13 to 19 of the menstrual cycle and rapidly drop again later in the cycle (Kharitonov et al., 1994). Although ovulation was not documented in this study and the timing of reproductive hormone changes therefore unknown, the profile of N O fluctuations suggest that oestradiol, luteinising hormone or follicle stimulating hormone are the most likely hormones to initiate the response. Mid-cycle increases in plasma volume would result in similar increases in V c and therefore D L C O by definition. Moreover, depending on the ability of the blood to restore [Hb] to normal levels, plasma volume alterations might affect the measured D L C O through changes in [Hb]. Plasma volume, if it were raised too high, might also have the potential to induce pulmonary oedema via increased pressure and stress failure of pulmonary capillaries thus increasing the thickness of the 4 diffusion membrane and lowering D M (and DLCO ) - If raised oestradiol is responsible for menstrual alterations in DLco, a large increase in DLco wou ld be expected in the late follicular phase prior to ovulation compared to moderate elevations during the luteal phase. F U R T H E R E X P L A N A T I O N S O F D L c o C H A N G E S D U R I N G M E N S E S Alterations in the endogenous production of C O with menstrual phase have been reported in the past but not well studied (Coburn, 1970). A change in production would affect the assessment of D L C O via changes in the diffusion gradient for carbon monoxide. Nonetheless, the expected magnitude of endogenous C O increase during menses as a result of red blood cell breakdown was calculated by Sansores et al. (1995), to result in only negligible declines in Dlco, wel l short of the observed reduction. Direct effects of steroid hormones on 6, the reaction rate of C O and haemoglobin might be considered as a potential mechanism. Reductions in DLco occurring around menstruation might also be mediated by vasoconstrictive agents (prostaglandins for example), which might alter the vasomotor tone of pulmonary capillaries thus affecting pulmonary capillary blood volume and hence D L c o Prostaglandin A2 has a known vasoconstrictor action on pulmonary blood vessels (Patton et al., 1989) and may be involved in uterine vasoconstriction occurring just before and around the time of flow. Although its exact role in the human endometrium in vivo has not been established, Prostaglandin F2a has also been implicated in the process of luteolysis or break-down of the corpus luteum occurring at the end of the luteal phase (Adashi et al., 1996). 5 L I M I T A T I O N S OF PREVIOUS W O R K In addition to the small number of subjects who were able to undergo the diffusion partitioning procedure, the Sansores et al. (1995) study is l imited in its ability to either characterise changes in D L C O over the menstrual cycle or determine a l ikely mechanism for the observed change because it considered only the effect of a single menstrual cycle phase (menses itself) on D L C O and because only one cycle for each subject was investigated. Furthermore, 8 of the 14 subjects were taking oral contraceptives and the post-menses measurement (obtained between Day 5 and Day 10 of the menstrual cycle) may have corresponded with a point in time when some of these 8 women were still taking placebo tablets, while others had resumed hormone tablets for the new cycle. R E S E A R C H QUESTIONS The main purpose of this study is to describe changes in DLco and the two components of D L C O ( V C and DM), within an ovulatory menstrual cycle of healthy regularly menstruating women. D L C O expressed as a ratio to alveolar ventilation (VA) is also analysed in an attempt to reduce error due to varying inspired volumes in the D L C O procedure. Haemoglobin concentration [Hb], may also vary with menstrual cycle phase and values of D L C O , DLco / V A , V c and D M are reported corrected for haemoglobin. Measurement of packed cell volume (PCV or haematocrit), which reflects alterations in plasma volume, and C O H b percent provide further explanatory power for significant alterations in diffusion. The following research questions provide a specific basis for the investigation. 6 1. Is there a change in DLco within an ovulatory menstrual cycle in regularly menstruating women? 2. Is there a change in D L C O / V A within an ovulatory menstrual cycle in regularly menstruating women? 3. Is there a change in [Hb] corrected D L C O within an ovulatory menstrual cycle in regularly menstruating women? 4. Is there a change in [Hb] corrected D L C O / V A within an ovulatory menstrual cycle in regularly menstruating women? 5. Is there a change in [Hb] corrected Vc within an ovulatory menstrual cycle in regularly menstruating women? 6. Is there a change in [Hb] corrected D M within an ovulatory menstrual cycle in regularly menstruating women? If a significant alteration in DLco is found, the following research questions will be investigated as explanatory variables of the above. 7. Is there a change in percent COHb within an ovulatory menstrual cycle in regularly menstruating women? 8. Is there a change in PCV within an ovulatory menstrual cycle in regularly menstruating women? The characteristics of diffusion changes between ovulatory and anovulatory menstrual cycles of individuals who display both within the study period will be compared. 7 9 -12. Is the change in D L C O / D L C O / V A , [Hb] corrected D L C O / [Hb] corrected D L C O / V A with menstrual cycle phase different for ovulatory versus anovulatory cycles of the same regularly menstruating woman. Finally, the reliabilities of the P C V measurement and [Hb] measurement procedures used in the study were assessed. A further comparison of the validity of the portable [Hb] analyser compared to a hospital spectometry unit was undertaken. 8 CHAPTER TWO: METHODOLOGY E T H I C A L A P P R O V A L Ethical approval for the study was obtained from the University of British Columbia Cl in ica l Screening Committee for Research and other Studies Involving H u m a n Subjects. The ethical approval certificate and approved subject consent form is included (Appendix IV). SUBJECT R E C R U I T M E N T A N D S E L E C T I O N Volunteers were obtained through advertising on bulletin boards around the University of British Columbia (Appendix V) and through word of mouth. Initial contact wi th people indicating their interest in participating in the study was normally made by telephone. Subjects included in the study had not been taking oral contraceptives for the previous 3 months, had no history of respiratory medical conditions and had not smoked regularly in the last 2 years since smoking has been shown to be associated wi th a decline in D L C O (Frans et al., 1975). A subject who is an ex-smoker but has not smoked regularly for 2 years or more was thought to be less likely to resume smoking during the study. A l l subjects had also menstruated regularly for at least 5 years and a normal average cycle length of 21-36 days over the previous 12 months (Barr and Prior, 1994). Subjects selected for the study also reported the recognition of cyclic alterations of physiological or psychological parameters during the menstrual cycle from which they believed they were able to estimate the time of ovulation and predict the occurrence of menstruation. Examples of commonly reported moliminal changes are increased viscosity of cervical mucous just prior to 9 ovulation and cramps just prior to menses. Subjects who had no experience with completing The Menstrual Diary (© Prior, 1996; Appendix VI) and who were not totally familiar with the timing of key events in their menstrual cycle, kept a record using this instrument for at least one month before testing. An appointment was then made with subjects who met all selection criteria for the study. At this appointment, subjects read and signed the consent form. The timing of the testing points in the cycle was carefully explained and tentative bookings were made for the next four or five tests. DOCUMENTATION OF MENSTRUAL DATA The Menstrual Cycle Diary (© Prior, 1996) that allows daily recording of menstrual flow, molimina, mood fluctuations, and sub-lingual temperature was used by subjects in the study. The Diary was modified to include a measurement of supine resting heart-rate and hours of sleep (Appendix VI). Subjects were instructed to measure sub-lingual basal body temperature at approximately the same time each morning before rising using a Becton Dickenson digital thermometer. Daily basal body temperature measurements were used to provide an index of ovulation and luteal phase onset using the method of least mean squares (Prior et al., 1990b). Luteal phase onset was defined as the number of days from the first day of the quantitatively determined mean temperature rise to the day before the onset of bleeding inclusive. Daily reports of menstrual cycle experiences (molimina), were used to increase subject interest in the project and their adherence to the study, identify the best time of testing and in the analysis to help confirm possible cycle-related changes in diffusion. Results from the menstrual charts were explained to subjects to help clarify the testing points and an attempt was made to answer any questions they had relating to their own menstrual cycle. 10 TIMING OF TESTING Testing of subjects occurred at five points during the menstrual cycle (figure 1) for three cycles. Day 1 of the cycle was defined as beginning at midnight preceding the day that menstrual bleeding began irrespective of the exact time at which bleeding started. Subjects were instructed to report to the lab at the following five test points outlined below. Test Point a. Early Menstrual This took place on the first or second day of flow or the last day of the previous cycle. It was preferable that this Test Point occurred while menstrual cramps were were present. This time point was chosen as a time in the cycle where levels of oestrogen and progesterone are low but prostaglandins (as indicated by cramping) are present. Test Point b. Late Menstrual This testing point took place a few days later between Day 3 to Day 4 of the cycle (inclusive), when menstrual cramps if they had been present had subsided. At this time point levels of oestrogen, progesterone and prostaglandins are minimal. Test Point c. Early Follicular This took place a few days later again on Day 5 to Day 8 inclusive. At this time point follicular oestrogen levels would be expected to be moderate whilst levels of progesterone and prostaglandins are very low. Test Point d. Midcycle This testing point took place as close as possible to subject observed and reported thickening of cervical mucous in The Daily Menstrual Diary©. This time point would be expected to coincide with peak levels of oestrogen and low levels of 1 1 progesterone and prostaglandins that occur just prior to ovulation. Daily monitoring of cervical mucous was crucial in determining the correct time for this test point as the timing of ovulation within the cycle is quite variable (Landgren et al., 1980). Test Point e. Mid Luteal This took place between 3 to 7 days after a subject observed an increase in basal body temperature of around 0.35°C (Prior et al., 1990b) or between Day 17 to Day 22 of the cycle if a basal body temperature increase was not clear. An increase in temperature signifies the luteal phase onset and was quantitatively determined from basal body temperature measurements using the least mean squares analysis of Prior et al. (1990b). Analysis of the ovulatory pattern of the subject's past cycles was also used to help identify when ovulation was likely to occur and estimate the best time for mid-luteal testing for that individual. A mid-luteal test occurring within an ovulatory cycle would be expected to coincide with moderate levels of oestrogen, high levels of progesterone and low levels of prostaglandins. 12 TEST POINTS a 8) .C. LU z § LU r— LU i 22 24 26 23 DAYS OF MENSTRUAL CYCLE Figure 1: The Timing of Testing at Points in the Menstrual Cycle 13 Subjects began their first test at any of the five points depending upon the stage of their cycle they were at when they were first available for testing. For the purposes of this study, all test points were given a number and letter code. The number corresponds to the cycle number while the letter corresponds to the testing point (eg: "2a" refers to Cycle Number 2, Test Point a). Apart from a fingerprick blood sample, which was used for [Hb] measurement, and haematocrit measurement if there was enough blood, no blood or urine samples were collected at Test Point a (Early Menstrual). At the completion of the study, the timing of visits was checked retrospectively against the subjects Menstrual Diary to ensure that the timing corresponded with the criteria. While advance warning of Test Point d (Midcycle) was provided by alterations in cervical mucous or other menstrual cycle experiences that the individual associated with ovulation (eg: mid-cycle cramps), it was possible to retrospectively compare the actual laboratory testing day with the quantitatively determined day of luteal phase onset. Ovulation and peak oestrogen would be expected to preceed the basal temperature rise by 2 to 3 days (Prior et al., 1990b) and peak progesterone levels might be expected around midway between the basal body temperature rise and the onset of the following cycle. The criteria for these two test points was clarified based on basal temperature analysis and was as follows. Test Point d. Mid Cycle This test point should have occurred between 5 days before basal temperature rise and the day before basal temperature rise to correspond with expected peak oestrogen. 14 Test Point e. M i d Luteal This test point should have occurred between the second day of increased body temperature and three days before the onset of menses for the next cycle to correspond with expected peak progesterone. D O C U M E N T A T I O N OF EXERCISE Subjects used the Daily Exercise Record (Appendix VII) to record the amount of time spent engaging in either vigorous or strenuous exercise (sufficient to elevate heart-rate to greater than 150 beats per minute) or mi ld or moderate exercise (sufficient to elevate heart-rate to between 90 and 150 beats per minute). The primary mode of exercise each day was also recorded. Daily exercise was measured so that a reterospective assessment of exercise patterns could be used to explain aberrant DLco measurements and to help explain changes that occurred in menstrual hormonal status, particularly those changes sufficient to alter DLco or its components. TESTING P R O T O C O L O n an initial visit to the laboratory subjects completed a questionnaire detailing descriptive data including age, occupation, a brief menstrual, exercise and health history as wel l as a 24 hour recall of the consumption of foods high in calcium (Appendix VIII). Basic anthropometric variables (height, body mass and sum of six skinfolds) were measured and the subject's date of birth was recorded. The timing of subsequent visits to the lab was explained to subjects and tentative dates were recorded. Where possible, each subject came into the lab at the same time for each of the 15 laboratory tests in order to minimise the effect of diurnal variation in lung diffusion (Cinkotai and Thompson, 1966). Subjects were 15 asked to be as consistent as possible with the timing of meals and caffeinated drinks prior to each session. Because pulmonary diffusing capacity of CO has been found to decline from 1 hour to 24 hours following maximal exercise (Sheel, 1995), subjects were also instructed not to exercise intensively 24 hours prior to their laboratory visit. The time of last vigorous exercise, last meal and last coffee or caffeinated drink was recorded. Upon arrival in the laboratory, the subject's menstrual diary was reviewed, the date (or tentative date) and time of the next testing point were confirmed and any questions they had relating to their menstrual cycle were addressed. At Test Points b, c, d, and e for the first two cycles only blood was drawn and overnight urine collected from 11 subjects willing to undergo this procedure. At each Test Point body weight, forced vital capacity (FVC) and the forced expired volume in 1 second (FEV1) were measured prior to assessment of diffusing capacity of the lung (DLco) and its two components D M and Vc. At Test Point "a." when no venous blood was drawn, [Hb] was measured from a finger capillary sample. At the final testing session, skinfolds were again measured, items of the questionnaire which might have changed over the study period were readministered and selected items from the Canadian Multicentre Osteoporosis Study (CaMOS) were administered via interview (Appendix X). Sheets for raw data entry are included in Appendix IX. MEASUREMENT OF HEIGHT, WEIGHT A N D ANTHROPOMETRIC VARIABLES Height The subject stood erect without shoes against a Holtain stadiometer. Subjects were instructed look straight ahead and visual inspection was used to determine if 16 their head was held in the Frankfort Plane: the position where an imaginary line joining the orbitale (most inferior point on the margin of the eye socket) to the tragion (notch superior to the flap of the ear at the superior aspect of the zygomatic bone) is horizontal. Upon a full inspiration, the measurement was taken as the maximum distance from the floor to the most superior point on the skull. Height was measured both on the initial and final testing session and, if the measurement differed, an average was recorded. Weight Body weight was determined with subjects wearing light clothing, using a Horns beam scale and measured to the nearest 0.05 kg. An average of the measurement at the initial and final testing session was recorded in the descriptive data. The subjects weight was also measured at all other test sessions to ascertain that no large fluctuations over the duration of the study occurred. Skinfolds Skinfold measurements were obtained following the procedure described by Ross and Marfell-Jones (1982), with the exception that all measurements, including the abdominal skinfold, were made on the right side of the body. The Triceps, Subscapular, Iliac Crest, Supraspinal, Abdominal, Anterior Thigh and Medial Calf skinfolds were each measured twice. If the two measurements varied more than 10% a third measurement was taken. The average between the two measurements varying less than 10% was recorded. If no measurement differed from the next closest by more than 10% an average of all three was used. Six skinfolds (all the above excluding Iliac Crest) were summed and recorded as the sum of six skinfolds. As skinfold assessment also took place at at the initial and final testing session, the average of the two sum of six skinfolds was reported. 17 MEASUREMENT OF RESTING FORCED VITAL CAPACITY (FVC) A N D FORCED EXPIRED VOLUME IN ONE SECOND (FEV1) Forced Vital Capacity (FVC) and Forced Expired Volume in Is (FEV1) were measured at each testing session using the spirometry functions of a Collins/DS pulmonary function analyser. The subject was instructed to inspire deeply and then expire forceably: "as hard and fast as possible" until all air was "squeezed out" of their lungs. A flow versus volume loop for a whole breath cycle was completed when the subject reinspired to vital capacity. The largest FVC and FEV1 from at least two maximal tests not varying by more than 10% for either variable was recorded. The forced expired volume from 25% to 75% of the expiration (FEV25-75) and the peak expired flow rate (PEFR) were recorded from the test with the greatest summed FVC and FEV1. MEASUREMENT OF RESTING DIFFUSING CAPACITY ( D L C O ) Resting pulmonary diffusing capacity was determined via the single-breath method first developed by Krogh (1915) and modified by Ogilvie et al. (1957) using the Collins analyser. Single breath methods as opposed to steady state measurements of diffusion are thought to better reflect the alveolar membrane and pulmonary capillary characteristics of the ventilated parts of the lungs (Forster et al., 1986). Moreover, they do not require the measurement of arterial pC02 for their most accurate determination. According to Forster et al. (1986) however, single-breath techniques may be less sensitive to unevenness of gas distribution and probably to non-uniformity of diffusion throughout the lung. It was thought that in healthy subjects unevenness and non-uniformity are likely to be minimal hence the use of the single-breath measurement in this investigation. The single-breath method used to measure D L C O in our laboratory consists of a rapid inspiration, a 10s breath-hold and a rapid expiration of the test gas containing 18 about 21% oxygen, 10% helium, 0.3% carbon monoxide and the balance nitrogen. A sample of expired gas is collected in a collection bag attached to the Collins analyser. Breath-hold during the test is timed from the beginning of inspiration to the beginning of sample collection as outlined by Ogilvie et al. (1957). While the latest American Thoracic Society (A.T.S.) recomendations (A.T.S., 1995) suggest the use of the Jones and Meade (1961) protocol which adjusts the calculation of breath hold to better reflect the CO concentration profile in the alveolar space, the results of Graham et al., (1981), suggest that in healthy subjects measures of diffusing capacity calculated via the Ogilvie method are likely to be similar to those measured via the Jones and Meade method. Both carbon dioxide ( C O 2 ) and water ( H 2 O ) were removed from the expired gas sample prior to analysis as the gas sample passed through a divided canister containing calcium sulphate and barium hydroxide. Concentrations of CO were measured using an infrared analyser. The concentration of expired O 2 is assumed for the purposes of calculation. In the measurement of D L C O , subjects were encouraged to relax against a closed glottis and remain calm during the breath-hold to avoid performing either a Valsalva or Muller manoeuver that could under- or overestimate DLco respectively. Each diffusion measurement was also examined to ensure that the inspired volume was at least 85% of the FVC, the total time of inspiration was less than two seconds and the breath-hold time was between nine and eleven seconds as the accuracy of DLco measurements are increased in this range (Graham et al, 1981). The American Thoracic Society (A.T.S., 1995) recommends that 90% FVC be attained for each inspiration. However, subjects in this study were unable to attain this standard consistently and so the criterion was lowered to 85%. On a few occasions for 9 subjects, only one of the two tests reached 85%. If the two tests were above 80% and acceptably close to each other (below), then an average of these two tests was recorded. In line with A.T.S. recommendations (1995), DLco was determined in duplicate and repeated a third time if the initial two measurements varied from each other by more than 10% of their average. The 19 average of tests differing from the mean by 10% or less was reported. A n interval of at least 4 minutes was allowed between tests to ensure elimination of the test gas from the lungs. C A L C U L A T I O N O F DIFFUSING C A P A C I T Y D L C O was calculated automatically by the Collins system using equation 1 (below). The rationale for its use is outlined in Appendix I (p57). Equation 1: Calculation of Diffusing Capacity D L C O = V A X 60 x L n [(FEHe/FECO) x (FiCO/FiHe)] x STPD correction 713 x t where D L C O = diffusing capacity for C O ( m l C O ( S T P D ) / m i n / m m H g ) V A = alveolar volume ( A T P S in ml) = V i x FiHe x 1.05 FEHe FiHe = inspired He fraction FEHe = expired He fraction F i C O = inspired C O fraction F E C O = expired C O fraction V i = volume inspired t = breath-hold time (s) A T P S = ambient body temperature and pressure S T P D = standard temperature and pressure dry 1.05 = correction factor for 5% carbon dioxide in expired air removed prior to analysis 713 = PB of 760mmHg - Pwater vapour at 37° c of 47mmHg 20 CORRECTION FOR HAEMOGLOBIN CONCENTRATION The diffusion value was adjusted for haemoglobin concentration using the formula below recommended by the American Thoracic Society in its 1995 recommendations (equation 2). Both adjusted and unadjusted diffusion measurements are reported and analysed. Equation 2: D L C O [Hb]adjusted = D L C O measured (9.38 + [Hb]) / 1.7 [Hb] where D L C O [Hb]adjusted = diffusing capacity for CO corrected for [Hb] (g/dl) (mlCO(STPD)/min/mmHg) D L C O measured = unadjusted diffusion measurement QUANTIFICATION OF D M AND Vc The two components of pulmonary diffusing capacity ( D M and Vc) were measured using the single-breath method of Roughton and Forster (1957) as modified by Ogilvie et al. (1957). Two measurements of resting diffusing capacity were made using two different inspired fractions of 02 (21% and 90%). Subjects breathed for 5 minutes through a low resistance valve (Hans Rudolph, #2700B) attached to a Douglas bag filled with a gas mixture of 90% ± 5% O 2 , and the balance N 2 . The D L C O 90% 02 test was performed in the same manner as the 21% 02. The reciprocal of D L C O (1 /DLCO) or total resistance to diffusion, is the sum of two component resistances ( 1 / D M and l /Vc). Mean pulmonary capillary oxygen partial pressure required for a calculation of 1/6 was estimated using the alveolar gas equation. The calculation of D M and Vc and the formula and underlying 21 assumptions for the use of the alveolar gas equation are outlined in Appendix I (p57). THE RELIABILITY OF DIFFUSING CAPACITY A N D ITS COMPONENTS The reliability of the single breath procedure in measuring DLco and its components has been determined in our laboratory in a study of 9 individuals measured twice on separate days (Sheel et al., 1996). Pearson's product-moment correlation coefficients between the two measurements were r = 0.98, r = 0.84 and r = 0.92 for D L C O / D M and Vc respectively. COLLECTION A N D BATCHING OF BLOOD AND URINE SAMPLES On the first cycle a total of approximately 16ml of blood was drawn from those subjects willing to undergo this procedure into 2, 6ml SST® vacutubes containing clot activator for later serum production, and into a 5ml vacutube (3 to 4 ml only) containing 15% K3 EDTA for haemoglobin (Hb) and packed cell volume (PCV) analysis. On the second cycle, approximately 20ml of blood was drawn as above with an additional 3 to 4ml drawn into an airtight vacutube containing EDTA for later carboxyhaemoglobin concentration [COHb] analysis. In addition, eight subjects who were willing collected a 45ml urine sample from their first morning excretion of the day that they were scheduled to come into the laboratory. Upon arrival at the lab, the urine sample was transferred to a standard household freezer in the laboratory. At the end of the week urine samples were transferred to -70°C. 22 M E A S U R E M E N T OF P A C K E D C E L L V O L U M E A N D H A E M O G L O B I N C O N C E N T R A T I O N Haemoglobin was measured at each testing session. For sessions when venous blood was not drawn, a pin-prick capillary blood sample was used. When venous samples were already taken, the whole, unclotted blood was slowly drawn into a 3ml syringe from the gently rotated vacutube and a drop from the syringe was allowed to saturate the measuring cuvette. Total [Hb] was then analysed using a HemoCue A13 portable (J-hemoglobin photometer in the laboratory. In addition, [COHb] and total [Hb] were measured for one cycle (four time-points) in each subject at Vancouver General Hospital using an OSM3 hemoximeter. Samples were taken from the refrigerated storage in the laboratory and transported on ice-packs for this analysis. Both same day and backdated samples were checked to ascertain that a delay d id not affect percent C O H b determination. A sample from the remaining whole unclotted blood was injected into two microcapillary tubes and spun in an International Electric Company (IEC) microcapillary centrifuge model M 8 centrifuge for 3 minutes. P C V was determined from the spun samples using a Sherwood Micro-Hematocrit Tube Reader and the average from the two microcapillary tubes recorded. D A T A A N A L Y S I S In order to answer the research questions, the data analysis was divided into three parts. Firstly to investigate changes in lung diffusion and diffusion-related variables, a "best" ovulatory cycle was selected for each subject and a difference over the five test points investigated. Secondly, the different profiles of diffusion change between ovulatory and anovulatory cycles in the same subject were investigated by selecting the best ovulatory and best anovulatory cycle. 2 3 Finally, analyses of reliability of the [Hb] and P C V measurements were made in two ways. Firstly, intraclass correlation coefficients for four repeated measurements (consecutive where possible), from the Test Points at which blood was drawn were determined. Secondly, at a random Test Point for each subject, duplicate measurements were analysed. Thus the stability (or reliability of the measurements over time), controlled for changes in [Hb] and P C V with menstrual cycle phase, were compared to a same day reliability of the measurement procedures. The validity of the [Hb] measurements made at our laboratory was determined from cross-measurements of venous blood total [Hb] on our laboratory portable HemoCue [Hb] analyser and the total [Hb] measurement made on the O S M hemoximeter located at Vancouver General Hospital. One Test in which both measurements were made w i l l be selected at random for each subject. Although any spectrometry measurement uses only a very small sample of whole blood (cf. blood chemistry analysis), this O S M hemoximeter was deemed as a suitable criterion for the determination of validity of our equipment. The manufacturer reports the total standard error for measurements of total [Hb] and [HbCO] made on the equipment to be 0.4 g / d l and 0.6% respectively for a 13.8 g / d l sample of oxygenated blood. N o comprehensive study of Hemo-Cue reliability is reported in the manufacturer's manual although the stated accuracy of the apparatus is ±0.3 g / d l . CRITERIA F O R T H E S E L E C T I O N OF T H E "BEST" C Y C L E FOR A N A L Y S I S One cycle was chosen for each subject for investigation of diffusion alterations. This procedure prevented the overrepresentation of one subject in the investigation whilst providing a solution to effectively deal wi th missing test points and anovulatory menstrual cycles. The criteria for selecting the best cycle were as follows. 1. Diffusion tests met the criteria outlined previously (pl8-20). 24 2. Cycle data was complete. N o more than 33% of temperature readings (or any 3 at mid-cycle) were missing or affected by reported illness or taking at the wrong time (Prior et al., 1990b). 3. The cycle was ovulatory as confirmed by basal body temperature methods. 4. The menstrual period was between 3 and 6 days in length and the cycle (defined by the luteal phase cycle point (point "e") was of normal length: between 21 and 36 days (Prior, 1996). The cycle chosen normally began at the same cycle point that the subject originally started participating in the laboratory sessions, unless it was impossible to select five sequential testing points meeting the criteria above starting at this test point. Ovulat ion was assessed using The Menstrual Diary© which begins each cycle wi th the onset of menstruation. Whether a cycle for analysis was to be classed as ovulatory or not depended on the basal body temperature analysis of Test Points d and e. (Mid Cycle and Mid-Luteal Test Points) irrespective of where the cycle began. If the cycle began at Test Point e, both the preceding and proceeding cycle needed to be ovulatory. S T A T I S T I C A L A N A L Y S I S Descriptive statistics of subject characteristics including age, height, weight and lung function parameters were calculated using Microsoft™ Excel 5.0. The first analysis of changes in diffusion parameters during a normal ovulatory cycle was made via a One-Way Repeated Measures Analysis of Variance ( R M - A N O V A ) over the five Test Points. Significant trends across the Test Points were also investigated using trend analysis. A l l pair-wise Tukey's post-hoc tests were chosen as the statistic used to investigate the nature of any significant changes in the overall F statistics that were observed. The second analysis of ovulatory and anovulatory cycles was investigated using a Two-Way (Ovulatory Status (2 levels) x 25 Test Point (5 levels)) R M - A N O V A and testing for a significant interaction between the groups. The significance of the regression equation determined was also tested. Finally for the reliability and validity studies, Intraclass Correlation Coefficients were calculated for the repeated measurements using Statistical Package for the Social Sciences, Version x (SPSSx) for Windows. The Intraclass Correlation Coefficient is a more appropriate measurement of reliability than Pearson's (interclass) Correlation Coefficient (Vincent, 1995, pl78). Refer to Appendix III for the formula and an explanation. For all analyses a was set at 0.05. 26 CHAPTER THREE: RESULTS RECRUITMENT OF SUBJECTS Subjects were largely self-selected because the criteria for entry were outlined on the advertisement, however three individuals who were ineligible to participate contacted by telephone. A total of 19 subjects agreed to enter the study. Four withdrew from the study before the first laboratory visit due to other time commitments. Two subjects withdrew after their first laboratory visit, one because of other commitments and one for feelings of discomfort with the laboratory procedure. One subject agreed to participate but was not willing to have blood drawn and no blood was drawn from another subject after medical staff experienced undue difficulty in drawing blood from her. One subject (subject 21) was only able to participate for two rather than three menstrual cycles and. Thus, a total of 13 subjects (11 from whom blood was drawn) are included in the analysis. GENERAL SUBJECT CHARACTERISTICS All the subjects were associated with the university either as staff (n=3), students (n=9) or in one case, a recently graduated student. They were 168.8±4.2cm in height and weighed 67.1±21.2kg. In accordance with the selection criteria, no subjects smoked regularly, none had a respiratory or endocrine medical condition and none had taken oral contraceptives in the last year. All subjects had a history of regular menstruation and all reported that they were normally able to predict the onset of menses. They reported a cycle length of 29.6±2.0 days and a 4.9±1.3 day duration of menses. Their average age at menarche was 12.5±1.3 years. None of the subjects had been pregnant in the past. Summary data of subject characteristics is included in Appendix XII (p59). 27 SUBJECT A C T I V I T Y A l l subjects were healthy and active. Three were competitive athletes at the time of the study. Analysis of the open question for subjects to list the vigorous activity they performed regularly revealed that they performed 6.0±3.1 hours per week of vigorous activity ranging from 0 to 10.1 hours per week at the time of the study. A l l had spent more than one hour per week vigorously active in the past for an average of 9.3+4.2 years. The question from the Canada Multicentre Study of Osteoporosis (CaMOS), which divides activity into strenuous sport, vigorous work and moderate activity elicited values of 6.5±4.9,1.5±1.3, and 6.1+3.9 hours per week respectively. Hours of sitting per week ranged from 1.5 to 13.5 hours per week (7.2±3.3 hours/week). L U N G F U N C T I O N C H A R A C T E R I S T I C S Since all subjects were active and had no respiratory conditions, all lung function variables were within the normal range (Table 1). Blood haemoglobin concentration (recorded as an average between the initial and final laboratory reading was 13.1+1.0 m g / d l . The packed cell volume was 39.0±2.9%. Ten of the thirteen subjects experienced difficulty in consistently attaining an inspired volume of 90% of forced vital capacity (FVC) in al l four diffusion tests during each test session. Ten subjects were able to attain 85% in at least one of the two tests at both O 2 fractions at every session. For the three subjects who could not, care was taken in the selection of the "best" cycle for analysis to include test sessions that met this criterion. 28 TABLE ONE: Lung Function Characteristics of Subjects showing forced vital capacity (FVC), forced expired volume in Is (FEV1), forced expired volume between 25% and 75% of expiratory time (FEV25-75), and peak expiratory flow rate (PEFR) SUBJECT FVC (1) FEV1 (1/s) FEV 25-75 (1) PEFR (1/s) 00 4.39 3.67 3.58 9.29 01 4.14 3.40 3.11 7.69 03 3.76 3.43 4.17 7.04 04 5.77 4.91 5.18 9.59 08 3.95 3.25 3.22 6.58 12 4.47 3.92 4.35 9.20 13 3.76 3.46 4.59 7.02 16 4.08 3.74 4.47 6.99 18 3.35 2.72 2.55 5.30 19 3.89 3.50 4.40 9.52 20 3.96 3.50 4.07 8.36 21 4.46 3.85 4.48 9.93 22 4.56 3.82 3.76 7.72 M E A N 4.19 3.63 3.99 8.02 S.D. 0.59 0.49 0.72 1.42 29 A N A L Y S I S O F C Y C L E S The 13 subjects completed testing over a total of 38 full menstrual cycles (two for Subject 21 and three for all the rest). The cycle characteristics for each subject are shown in Table 2. Because subjects began attending the lab at different times throughout their cycle, but kept their Menstrual Diary for the period of time leading up to and following their initial and final laboratory visit, there are more than 3 cycles listed for some subjects. In addition one subject, (Subject 12), was not able to attend laboratory sessions for one month, but kept her Menstrual Diary during the intervening time. One subject (Subject 19), missed a menstrual period. This is represented in the table below as a particularly long cycle. Basal body temperature analysis confirmed that 6 (16%) of the 37 menstrual cycles with sufficient data recorded were anovulatory. A N A L Y S I S O F TESTING POINTS A timetable which shows the dates of diffusion testing for each subject, and corresponding cycle day for the diffusion measurements is included (Appendix XI). The timetable also shows the cycle length (CL) for the cycle beginning at Point "a" and the day of luteal phase onset as determined by the basal body temperature rise. The Test points chosen for the "best" ovulatory cycle and the anovulatory cycle are also shown in Appendix XI on a separate timetable. When the best ovulatory cycles were chosen for each subject, data from four Test Points (in three subjects) were missing because the subject was unable to attend at that time. For the anovulatory cycles recorded by six subjects, there were two missing Test Points. Missing values were replaced in the statistical analysis wi th the grand cell mean corrected for the subject mean and for the test point mean. In two cycles chosen as the best ovulatory cycle, the length of menses was more than 6 days (7 days for Subject 13 and 11 days for Subject 19). Subject 04, whose best ovulatory 30 TABLE TWO: Summary of Menstrual Cycles showing cycle number in study, cycle length (days), length of menses (days), and length of the luteal phase*1. S U B J E C T C Y C L E N O . C Y C L E L E N G T H M E N S E S L E N G T H L U T E A L L E N G T H (t value) 00 1 26 5 11 (2.07) 2 23 6 0 3 30 5 9 (7.94) 01 1 27 6 14 (5.75) 2 26 5 11 (6.57) 3 28 5 13 (11.67) 03 1 38 6 9 (3.86) 2 27 6 9 (5.26) 3 41 7 0 31 6 8 •„ (4.64) 04 1 26 5 6 (2.49) 2 27 5 11 (2.29) 3 28 7 I.D.*2 4 26 7 ID. 08 1 28 4 6 (2.53) 2 28 3 10 (5.99) 3 34 5 9 (3.78) 12 1 29 4 15 (2.14) 2 26 I.D. 0 3 27 I.D. I.D. 4 33 I.D. I.D. 5 ID. I.D. I.D. 13 1 27 7 10 (3.49) 2 29 5 7 (4.76) 3 25 7 0 16 1 29 7 14 (2.07) 2 27 5 11 (4.96) 3 28 5 14 (3.44) 4 27 5 12 18 1 32 7 13 (8.06) 2 32 6 14 (9.30) 3 34 8 0 4 31 5 17 19 1 29 11 6 (3.75) 2 59 10 0 3 I.D. I.D. I.D. 20 1 38 6 14 (7.92) 2 31 6 15 (8.27) 3 35 5 13 (6.31) 4 ID. I.D. I.D. 21 1 30 I.D. 10 (6.16) 2 34 5 9 (4.56) 3 33 4 I.D. 22 1 28 I.D. I.D. 2 29 I.D. 9 (6.45) 3 27 4 14 (12.25) 4 26 4 13 *1 Luteal length was measured from the day of onset of basal body temperature rise (inclusive) to the day of onset of menses (exclusive). A luteal length of 0 means that the menstrual cycle was anovulatory. The t value is for the difference in basal temperatures between the follicular and luteal phase (greater than 2.0 is considered acceptable). *2 Insufficient data to determine this value 3 1 cycle was 26 days in length, had recorded only 15 basal temperatures for that cycle and therefore fell just short of the criteria for sufficient data. Despite there being inadequate recordings, the mean difference between temperatures in the follicular and luteal phase for that subject was allowable (t = 2.49) and was included as an ovulatory cycle. The anovulatory cycles appeared to be relatively more disturbed in terms of cycle length and length of bleeding. Two of the six anovulatory cycles were longer than 36 days and four exhibited menstrual periods greater than 6 days. One anovulatory cycle wi th only 16 (out of 26) recorded temperatures was also used. The proportion of cycle Testing Days that occurred outside the strict timing criteria previously outlined (pll-13) is shown below (Table 3). The details of this information may be examined by comparing Diffusion Testing Timetable (Appendix XI) wi th Table 2. A total of 10 tests for ovulatory cycles d id not occur at the correct time. Nine were only one day outside the criteria and one (Subject 20, Test Point d) occurred two days late. Three tests originally scheduled as Test Point e met the criteria for Test Points d (2) and a (1) and were thus analysed as such. A l l tests occurred wi th in the correct time frame for the anovulatory cycles. Test Points "d" and "e" took place between Day 15 and 19, and between Day 19 and 23 of the menstrual cycle respectively. B L O O D S A M P L E A N A L Y S I S TESTING D A T E S Two timetables showing the date of testing, and corresponding cycle day for collection of blood samples for H b and C O H b assessment are included (Appendix XI). In addition to the smaller number of samples, they differ from the diffusion time-table in minor details when blood testing took place on the previous or subsequent day. 32 TABLE THREE: An Analysis of Test Points BEST O V U L A T O R Y CYCLE a b c d e Number of Missing Test Points 1 1 0 0 2 Timing of Tests Outside Criteria 2 2 0 5 1 Total Test Points 13 13 13 13 13 A N O V U L A T O R Y CYCLE a b c d e Number of Missing Test Points 1 0 0 0 1 Timing of Tests Outside Criteria 0 0 0 0 0 Total Test Points 6 6 6 6 6 33 BEST C Y C L E A N A L Y S I S N o significant change across the five test points was noted in D L C O or D L C O / V A (Figure 2 and 3). Adjustment for [Fib] alterations made no difference to the results for DLQO, and D L C O / V A (Figures 4 and 5), although a significant difference in [Hb] between Test Points of an ovulatory cycle was observed (Figure 6). Tukey's post-hoc test for all pairwise comparisons at the a=0.05 significance level showed that only the largest difference from Test Point a to Test Point e was significant. The change in [Hb] was characterised by an average 5% increase from the early follicular to the mid luteal phase. N o changes were found in [Hb] adjusted or unadjusted V c or D M over 5 Test Points. For these parameters however, 16 missing values (25%), were replaced with mean corrected values. 34 Figure 2: Diffusing Capacity Measured at Five Test Points Within Ovulatory Menstrual Cycles of 13 Women. Bars represent standard deviations from the mean. No significant alterations were observed. 5.00 T 4.50 + 1) X E i E E 4.00 + 3.50 + 3.00 + 2.50 + 2.00 < TEST POINTS Figure 3: Diffusing Capacity/Alveolar Ventilation Ratio Measured at Five Test Points Within Ovulatory Menstrual Cycles of 13 Women. Bars represent standard deviations from the mean. No significant alterations were observed. 3 5 Figure 4: Haemoglobin Corrected Diffusing Capacity Measured at Five Test Points Within Ovulatory Menstrual Cycles of 13 Women. Bars represent standard deviations from the mean. No significant alterations were observed. Figure 5: Haemoglobin Corrected Diffusing Capacity/Alveolar Ventilation Ratio Measured at Five Test Points Within Ovulatory Menstrual Cycles of 13 Women. Bars represent standard deviations from the mean. No significant alterations were observed. 36 15.0 14.0 + £ 12.0 1 11.0 < -10.0 I I I I 1 a b c d e TEST POINTS Figure 6: Haemoglobin Changes Measured at Five Test Points Within Ovulatory Menstrual Cycles of 13 Women. Bars represent standard deviations from the mean. Test Point e is significantly different from Test Point a. 37 O V U L A T O R Y V E R S U S A N O V U L A T O R Y C Y C L E S When diffusion measurements of the 6 subjects who showed a variation in menstrual cycle status over the 3 months were analysed, [Hb] adjusted D L c o changes over the 5 test points approached significance (p=0.066). There was also a slight trend towards a difference in D L C O / V A and D L C O / V A adjusted for [Hb]. A l l three measurements tended to fall from the late menstrual measurement to mid-cycle. [Hb] adjusted D L c o tended to undergo steady decline from Test Point a to d and then a small rise at Test Point e. Figures 7 to 10 show trends in the changes of these variables in ovulatory and anovulatory cycles of subjects who recorded both during the study. There was no change in non [Hb] adjusted D L c o , nor in adjusted or unadjusted D M or V c over the 5 test points. N o changes were observed in any of the diffusion variables measured from the ovulatory to the anovulatory cycle. Addi t ional ly , there were no significant interactions between menstrual status and Test Point for any of the diffusion variables. C H A N G E S I N C A R B O X Y H A E M O G L O B I N W I T H M E N S T R U A L C Y C L E P H A S E There was no change in C O H b percentage measured at 4 Test Points for all 11 subjects. Seven of the subjects exhibited ovulatory cycles for this analysis. Separating out the potential differences in C O H b changes in ovulatory subjects from the three subjects who d id not ovulate during that cycle by using a M i x e d Method Repeated Measures A N O V A , revealed no change in C O H b over the test points and no interaction between menstrual status and Test Point (Figure 11). 38 Figure 7: Diffusing Capacity Measured at Five Test Points Within Ovulatory and Anovulatory Menstrual Cycles of 6 Women. Bars represent standard deviations from the mean. No significant alterations were observed. Figure 8: Haemoglobin Corrected Diffusing Capacity Measured at Five Test Points Within Ovulatory and Anovulatory Menstrual Cycles of 6 Women. Bars represent standard deviations from the mean. No significant alterations were observed. 39 Figure 9: Diffusing Capacity/Alveolar Ventilation Ratio Measured at Five Test Points Within Ovulatory and Anovulatory Menstrual Cycles of 6 Women. Bars represent standard deviations from the mean. No significant alterations were observed. 5.00 x | 4.50 f i 4.00 < ^ 3.50 + 3.00 + I 2.50 + 2.00 - O V U L A T O R Y - - A N O V U L A T O R Y TEST POINT Figure 10: Haemoglobin Corrected Diffusing Capacity/Alveolar Ventilation Ratio Measured at Five Test Points Within Ovulatory and Anovulatory Menstrual Cycles of 6 Women. Bars represent standard deviations from the mean. No significant alterations were observed. 40 Figure 11: Percent Carboxyhaemoglobin Measured at Five Test Points Within the Menstrual Cycles of 7 Ovulating and 3 Non-Ovulating Women. Bars represent standard deviations from the mean. No significant changes between test points, differences between the two groups or interactions found. 41 C H A N G E S I N P C V W I T H M E N S T R U A L P H A S E N o alteration in P C V over 4 Test Points was observed when measurements from all 11 subjects were analysed. Data from nine subjects who exhibited an ovulatory cycle during the phase of P C V measurement were also analysed and no changes over the menstrual cycle were found. RELIABILITY OF T H E P A C K E D C E L L V O L U M E M E A S U R E M E N T S The test-retest reliability of two measurements of P C V , drawn from the same whole blood sample and spun at the same time in our laboratory was very high and significant (r intraclass = 0.99; r 2 = 0.98; p<0.01). Over four tests measured at different phases in the menstrual cycle and after the separation of any consistent change across the Test Points, the reliability of P C V measurement in one subject was much lower and non-significant (r intraclass = 0.25; r 2 = 0.06). RELIABILITY A N D V A L I D I T Y OF T H E H A E M O G L O B I N C O N C E N T R A T I O N M E A S U R E M E N T S Test retest reliability for the measurement of [Hb] i n our laboratory was also very high (r = 0.98; r 2 = 0.96; p<0.01). The reliability of this measurement procedure undertaken at four points in the menstrual cycle after removing consistent changes wi th menstrual phase was still high (r = 0.96; r 2 = 0.92; p<0.01). The Intraclass Correlation Coefficient for the relationship between our measurement procedure and hospital based spectroscopy was (r = 0.89; r 2 = 0.85; p<0.01). There was a significant trial effect (p<0.01) and our laboratory procedure overestimated [Hb] compared to the O S M unit by an average of 0.63 g/100ml or 5%. 42 CHAPTER FOUR: DISCUSSION RECRUITMENT AND CYCLE CHARACTERISTICS OF SUBJECTS Of the subjects who were initially recruited for the study, only 68% completed the entire project. As only two of subjects who later withdrew attended the first laboratory visit, there is little information available to assess a difference in characteristics between those who completed the study and those who did not. The subjects who withdrew were of a similar educational and occupational background to the subjects who remained in the study and no obvious selection bias is apparent. Moreover, since the initial sample was non-random, the consequences of a selection bias as a result of subject withdrawal is likely to be relatively unimportant. Only 72% (81 out of 113) carefully selected women met the further screening criterion of two consecutive ovulatory cycles in the study of Prior et al. (1990a). Moreover, of the 81 women who remained in their 12 month study, only 66 (81%) completed it and only 13 (20%) of these women had consistently ovulatory menstrual cycles throughout the year. Screening for the current study was not as rigid as for the Prior et al. study. In the current study, subjects were not required to demonstrate two consecutive ovulatory cycles before entering the study. Subjects in the current study did however report a good knowledge of the timing and nature of events within their cycle and it was assumed that this was likely to be related to hormonal consistency. It was anticipated that at least 35% of subjects recruited for the study would either withdraw or not exhibit consistent ovulatory cycles for the three cycles. In fact, 32% of recruited subjects withdrew from the study and of the 13 who remained only six ovulated consistently for the time they were in the study. Two subjects recorded insufficient data to determine if they ovulated consistently or not. 4 3 DIFFUSION CHANGES WITHIN THE "BEST" OVULATORY MENSTRUAL CYCLE Neither D L C O nor its components was found to alter during ovulatory menstrual cycles in this group of women. While a change in [Hb] was observed, correction of the diffusion measurements for [Hb] did not affect the result. Differences in V A from test to test or over the three month period could not explain the lack of significance in D L c o either because there was no change in D L c o expressed as a ratio of V A . In addition, V A was very consistent. The mean intra-subject standard deviation of V A was 0.23 litres, less than 4% of the mean. For a One Way Repeated Measures ANOVA, power is correctly determined using the noncentrality parameter lambda (k) (Winer et al., 1991). X is essentially an effect size measure, analogous to Cohen's "d", whose variance term in the denominator is equivalent to the mean squared error ( M S e r r o r ) from the A N O V A table. In turn, M S e r r o r can be estimated from the mean intratrial variance and the mean correlation coefficient of all pairwise trials as shown in Equation 3. Equation 3: X = n X (Lti - It) 2 C 2 e (1 - p) where A, (lambda) = the noncentrality coefficient = number of subjects = the sum of the squared differences between n X(Lti-Ll)2 each trial mean and the overall mean = the common within cell variance (estimated from the average variance among scores of the dependent variable within a group) p (rho) = the average of the correlation coefficients between all pairs of trials 44 In the current study, the power to detect a 10% difference in D L C O from the highest to the lowest Test Point, assuming an even spread of DLco at other Test Points, was greater than 99.9%. The power to detect the same magnitude difference was even greater for [Hb] adjusted D L C O and [Hb] adjusted D L C O / V A , but was lower (75.5%) for [Hb] adjusted Vc. The power to detect a 5% change in DLQO, [Hb] adjusted D L C O and [Hb] adjusted D L C O / V A was still high: 59%, 74% and 70% respectively. More than the required number of subjects to detect with confidence the 9.5% difference in D L C O noted by Sansores et al (1995) were recruited for this study in order to ensure enough consistently ovulating women. HAEMOGLOBIN CONCENTRATION CHANGES WITHIN THE "BEST" OVULATORY MENSTRUAL CYCLE The luteal phase increase in [Hb] is in agreement with the findings of Jurkowski et al. (1981). While these researchers noted a 3% increase in the luteal phase of healthy subjects in whom ovulation was hormonally confirmed, in the current study a 5% alteration was observed. Past studies of [Hb] changes throughout the menstrual cycle have produced spurious results. While Vellar (1994) and Dombovy et al. (1987) noted luteal phase decreases in [Hb] neither of these studies confirmed the timing or existence of a luteal phase using basal body temperature or hormonal analysis. Lebrun et al. (1993), in a study quantitatively confirming ovulation, observed no change in [Hb] from the follicular to luteal phase. In the past, cyclic changes in [Hb] have been thought to occur as a result of increases in plasma volume and lowered PCV. Both progesterone and oestradiol may mediate fluid retention probably through stimulation of the renin-angiotensin system and increased production or activity of anti-diuretic hormone. Exogenous oestrogen also prevents plasma volume reduction with bed rest and the positive influence of endogenous oestradiol on plasma volume appears to be greater around the time of 45 ovulation when progesterone levels are low (Fortney et al., 1988). Luteal increases in plasma volume and concomitant reduction in PCV would result in a lowered [Hb] and cannot explain the luteal increases in [Hb] noted in this study. Further evidence that [Hb] changes occur independently of hormonally mediated haemostatic alterations in this study are the lack of change noted in PCV and the lack of difference in [Hb] changes over the Test Points between ovulatory and anovulatory cycles. The [Hb] rise from menses to luteal phase in this study may simply reflect the gradual restoration of red blood cells lost during menstruation. A 16% increase in caloric consumption in the luteal phase compared to the follicular phase has been previously noted (Barr et al., 1995). Although this study did not analyse differences in micronutrient intake, it is reasonable to suggest that iron intake is also higher in this phase and may accelerate this restoration process. Menstrual phase alterations in iron intake would be likely to have an important effect in the [Hb] of individuals with a suboptimal iron status. DIFFUSION CHANGES IN OVULATORY VERSUS ANOVULATORY CYCLES While no alterations in any of the diffusion measurements were noted with this smaller group of subjects, a non-significant tendency for [Hb] corrected D L C O to change existed, as did weak trends in both corrected and uncorrected D L C O / V A . At Test Point c in the ovulatory cycle, one subject (00), recorded a value much higher than normal. Removal of this subject from the analysis and correction of this data point to an average value, resulted in a significant change in [Hb] corrected DLco with menstrual cycle phase. However, no reason for this discrepancy could be determined from the diffusion test results. The subject had not exercised intensely in the previous day and had not consumed a large meal or recently drunk coffee. The test was completed at the same time of day as usual. The elevated record could 46 be due in part to the unusually high V A but correction for V A fluctuations in all the subjects did not produce significant results. The pattern of these trends in D L fluctuation was towards a small (5%) increase from early in menses until later in menses followed by a drop later in the follicular and luteal phases. Seaton (1972) noted a substantial decline (14.8%) in Vc from 7 to 10 days following menstruation compared to 2 to 4 days prior to menstruation. Although there was on average a corresponding 4.4% drop in D L C O / this difference was not statistically significant. The author suggests that progesterone mediated premenstrual distension of the pulmonary capillary bed, possibly secondary to respiratory arteriole or venous distension is the most likely explanation of his findings. The results of this study which carefully documented rises in basal body temperature, a secondary effect of progesterone, and found no corresponding change in D L C O do not support this rationalisation. An alternative interpretation of Seaton's results would be that peaking oestradiol (around day 7 to 10 of the menstrual cycle) caused the alteration in Vc. This would be in agreement with the trend of a midcycle rise in D L C O observed in the current study, although subjects in this study tended to ovulate much later than Day 7 to 10. Sansores et al. (1995), in contrast, noted a decline in diffusion during menses. These researchers investigated D L C O change during menses itself and recorded daily changes during this time of the cycle thus investigating more closely a specific phase of the menstrual cycle, menses itself. Mean intra-individual coefficients of variation for the parameter D L C O in the Sansores et al. (1995) study was lower than in the current study: 4.8% as opposed to 8.8%. This is not surprising given that in the Sansores et al. (1995) study, the trials were conducted over a few days compared to the current study which analysed trials conducted over a monthly cycle. The current study has carefully controlled for the effects of hormonal changes by selecting testing points to coincide with peak oestradiol and progesterone levels and assessing menstrual cycles for the occurrence of ovulation. Accordingly, the 47 lack of significant change in diffusion would appear to make the possibility of hormonally induced diffusion changes unlikely. FURTHER MECHANISMS OF DIFFUSION CHANGE Percentage COHb changes with menstrual cycle phase might be at least partially responsible for the DLco changes during menstruation observed by Sansores et al. (1995). These researchers reject COHb changes as an explanation of their observed DLco decline as the magnitude of COHb increase required to lower DLco by 10% was too great to be accounted for by red cell break-down. Coburn et al. (1970) report a doubling of CO production in the luteal phase of the menstrual cycle. Excess endogenous CO production may be a result of the breakdown of haemoglobin from senescent blood cells but may also reflect increased degradation of other haem compounds particularly of hemoproteins in the liver and other organs (Coburn, 1970). This excess production is unlikely to be reflected to any great extent in the blood as it is cleared from the body via the lungs and is masked by uptake from the environment. No change in whole blood COHb percent with menstrual cycle phase was noted in the current study. Repeated measurements of DLco will raise blood [COHb] and thus lower successive determination of DLco. We have found that percentage COHb increases around 1.5% following a partitioned diffusion test in our laboratory. Following the test or repeated tests it declines at a rate of approximately 1% per hour (Stewart et al., 1997). Complete clearance of raised [COHb] may therefore take some hours and depends upon alveolar ventilation, pulmonary capillary P O 2 and upon DLco itself. It seems very unlikely that any of the decrease in DLco noted by Sansores et al. (1995) was explained by [COHb] accumulation with repeated testing, given that tests were completed 24 hours apart. 48 Prior to this investigation we felt that levels of prostaglandins associated with menstrual cycle cramps might have been responsible for the observed changes in D L C O / perhaps acting via pulmonary arteriole vasoconstriction or alveolar membrane f luid retention. The incidence of menstrual cramps although it was assessed in The Menstrual Diary© was very unpredictable in both its occurrence for a particular cycle and in its timing. The few testing points that coincided wi th subjects reported cramping seemed to elicit D L C O measurements within the subject's normal range. Moreover, neither changes in V c nor D M wi th menstrual cycle phase (including the early and late menses testing points) were significant and this does not provide support for the prostaglandin initiated mechanisms described. Hormone independent alterations in N O are a possible alternative explanation for D L C O changes during menses. If this were the case however, it is hard to explain why hormone dependent changes in N O production (Kharitonov et al., 1994) were not observed. RELIABILITY A N D V A L I D I T Y OF P A C K E D C E L L V O L U M E A N D H A E M O G L O B I N C O N C E N T R A T I O N M E A S U R E M E N T S Both P C V and [Hb] measurements obtained in our laboratory were highly reproducible. Approximately 98% and 96% of the variation in one measurement of P C V and [Hb] respectively could be explained by the variation in a measurement just preceding it. In addition, 92% of the variation in [Hb] was consistent over time after the effect of systematic alterations over the menstrual cycle were removed. Over time P C V determination was much less reliable reflecting changes in P C V over time that were not consistent between subjects and were not related to menstrual cycle changes since this was controlled. Only 6% of the variation in a P C V measurement was retained in repeated test over a month. It seems therefore 49 that PCV undergoes fluctuation with time that is not systematic between individuals. The portable Hemocue analyser used in our lab for [Hb] determination appears to overestimate [Hb] compared to hospital based spectometry. The overestimation for a subject with a [Hb] of approximately 12 g/dl would be in the order of about 0.6 g/dl. This is outside the stated accuracy of both pieces of equipment (0.4 and 0.3 g/dl for the OSM and Hemocue respectively). In assessing the validity of the Hemocue, there is obviously a question of faith regarding the authenticity of manufacturer accuracy reports. The OSM spectometer in Vancouver Hospital outlined comprehensive laboratory trials in the determination of its reliability and accuracy compared to blood chemistry analysis. It therefore seems to be a more trustworthy standard and it could be assumed that the Hemocue in our laboratory did indeed consistently overestimate [Hb] by at least 0.2 g/dl. 50 CHAPTER FOUR: CONCLUSION This study has resulted in the following findings. 1. There is no change in D L C O or D L C O / V A either corrected or uncorrected for [Hb] with menstrual cycle phase of an ovulatory cycle in regularly menstruating healthy women. 2. There is no change in [Hb] corrected Vc or D M ; or in [COHb] or PCV with menstrual cycle phase of an ovulatory cycle in regularly menstruating healthy women. 3. There is a small (5%) increase in [Hb] from the early follicular to mid luteal phase of an ovulatory menstrual cycle in regularly menstruating healthy women. 4. There is no difference in menstrual cycle related D L C O or D L C O / V A changes between ovulatory and anovulatory cycles in the same subject. 5. There is no difference in menstrual cycle related [COHb] changes between ovulatory and anovulatory cycles in different subjects. Given that the timing of diffusion testing was designed to maximise reproductive hormone changes and that no difference in diffusing capacity exists between ovulatory and anovulatory cycles, a hormonally mediated alteration in D L over the menstrual cycle is unlikely. The avoidance of female subjects in studies related to pulmonary diffusion because of supposed phase related changes appears to be unfounded. 5 1 References Adashi, E.Y., Rock, JA. and Rosenwaks, Z. Reproductive Endocrinology, Surgery, and Technology. Raven Press Ltd, New York, 1996. American Thoracic Society. Single breath carbon monoxide diffusing capacity (transfer factor). Recommendations for a standard technique - 1995 update. American Journal or Respiration and Critical Care Medicine 152: 2185-2198, 1995. Barr, S.I., Janelle, K.C. and Prior, J.C. Energy intakes are higher during the luteal phase of ovulatory menstrual cycles. American Journal of Clinical Nutrition 61(1): 39-43,1995. Barr, S.I. and Prior, J.C. The menstrual cycle: Effects on premenopausal women. In Draper, H.H. (ed.) Advances in Nutritional Research 9; Chapter 17, 1994. Chang, S.C., Chang, H.I., Liu, S.Y., Shiao, G.M. and Perng, B.P. Effects of body position and age on membrane diffusing capacity and pulmonary capillary blood volume. Chest 102(1): 139-142, 1992. Chen, H.I. and Tang, Y.R. Effects of the menstrual cycle on respiratory muscle function. American Reviews of Respiratory Diseases 140; 1359-1362, 1989. Cinkotai, F.F. and Thompson, M.L. Diurnal variation in pulmonary diffusing capacity for carbon monoxide. Journal of Applied Physiology 21(2); 539-542, 1966. Coburn, R.F. Endogenous Carbon Monoxide Production. Medical Intelligence 282(4); 207-209,1970. Crapo, R.O. and Forster, R.E. Carbon monoxide diffusing capacity. Clinics in Chest Medicine 10(2); 187-197, 1989. Dempsey, J.A. Is the lung built for exercise? Medicine and Science in Sport and Exercise 18(2): 143-155,1986. Dombovy, M.L., Bonekat, H.W., Williams, T.J. and Staats, B.A. Exercise performance and ventilatory response in the menstrual cycle. Medicine and Science in Sports and Exercise 19(2); 111-117, 1987. Forster, R.E., Dubois, A.B., Briscoe, W.A., and Fisher, A.B. In: The Lung: Physiological Basis of Pulmonary Function Tests. Year Book Medical: Chicago, 3rd edition, 1986. Fortney, S.M., Beckett, W.S., Carpenter, A.J., Davis, J., Drew, H. , LaFrance, N.D., Rock, J.A., Tandersley, C.G. and Vromen, N.B. Changes in plasma volume during bed rest: effects of menstrual cycle and estrogen administration. Journal of Applied Physiology 65(2): 525-533, 1988. 52 Frans, A., Stanescu, D.C., Veriter, C. Smoking and pulmonary diffusing capacity. Scandinavian Journal of Respiratory Disease 56; 165, 1975. Frey, T.M., Crapo, R.O., Jensen, R.L. and Elliott, C.G. Diurnal variation of the diffusing capacity of the lung: Is it real?American Reviews in Respiratory Diseases 136; 1381,1987. Graham, B.L., Mink, J.T. and Cotton, D.J. Improved accuracy and precision of single-breath CO diffusing capacity measurements. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology 51(5); 1306-1313, 1981. Guyton, A . C Textbook of Medical Physiology 9th Edition. Guyton, A.C. and Hall, J.E. (eds). W.B. Saunders Company, 1996 Hastala, M.P. and Berger, A.J. Physiology of Respiration. Oxford University Press, 1996. Jones, F.S. and Meade, F.A. A theoretical and experimental analysis of anomalies in the estimation of pulmonary diffusing capacity by the single breath method. Quarterly Journal of Experimental Physiology 46; 131-43, 1961. Jurkowski, J.E.H., Jones, N.L., Toews, C.J., and Sutton, J.R. Effects of menstrual cycle on blood lactate, 02 delivery and performance during exercise. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology 51(6); 1493-1499, 1981. Kharitonov, S.A., Logan-Sinclair, R.B., Busset, C M . and Shinebourne, E.A. Peak expiratory nitric oxide differences in men and women: relation to the menstrual cycle. British Heart Journal 72; 243-245, 1994. Krogh, M . The diffusion of gases through the lungs of man. Journal of Physiology 49(4); 271-300,1915. Landgren, B.M., Unden, A.L. and Diczfalusy, E. Hormonal profile of the cycle of 68 normally menstruating women. Acta Endocrinologica 94: 89-98, 1980. Lebrun, C M . , McKenzie, D . C , Prior, J.C and Taunton, J.E. Effects of menstrual cycle phase on athletic performance. Medicine and Science in Sports and Exercise 27(3): 437-444,1995. Manier, G., Moinard, J. and Stoicheff, H . Pulmonary diffusing capacity after maximal exercise. Journal of Applied Physiology 75(6); 2580-2585, 1991. Miles, D.S., Christopher, C.E., Doerr, E., Schonfeld, S.A., Sinks, D.E. and Gotshall, R.W. Changes in pulmonary diffusing capacity and closing volume after running a marathon. Respiratory Physiology 52; 349-359, 1983. 53 Ogilvie, C M . , Forster, R.E., Blake, W.S. and Morton, J. A standardized breath holding technique for the clinical measurement of the diffusing capacity of the lung for carbon monoxide. Journal of Clinical Investigation 36; 1-17, 1957. Prior, J .C, Vigna, Y.M., Schechter, M.T. and Burgess, A.E. Spinal bone loss and ovulatory disturbances. New England Journal of Medicine 323; 1221-1227, 1990a. Prior, J .C, Vigna, Y.M., Schulzer, M. , Hall, J.E. and Bonen, A. Determination of luteal phase length by quantitative basal temperature methods: validation against the midcycle L H peak. Clinical and Investigative Medicine 13(3); 123-131, 1990b. Ross, W.D. and Marfell-Jones, M.J. Kinanthropometry In McDougall, J.D., Wenger, H.A. and Green, H.J. (eds). Physiological Testing of the High Performance Athlete 2nd Edition.. Human Kinetics Books, 1982. Roughton, F.J.W. and Forster, R.E. 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. Journal of Applied Physiology 11; 290, 1957. Sansores, R.H., Abboud, R.T., Kennell, C. and Haynes, N. The effect of menstruation on the pulmonary carbon monoxide diffusing capacity. American Journal of Respiratory and Critical Care Medicine 152; 381-384. Schoene, R.B., Robertson, T., Pierson, D.J. and Peterson, A.P. Respiratory drives and exercise in menstrual cycles of athletic and nonathletic women. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology 50: 1300-1305,1981. Scoggin, C.H., Doekel, R.D., Kryger, M.H. , Zwillich, C.W. and Weil, J.V. Familial aspects of decreased hypoxic drive in endurance athletes. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology 44; 464-468, 1978. Seaton, A. Pulmonary capillary blood volume in women: normal values and the effect of oral contraceptives. Thorax 27, 75-79, 1972. Sheel, A.W. The time-course of pulmonary diffusion capacity changes following maximal exercise. Unpublished Master's Thesis, University of British Columbia, 1995. Sheel, A.W., Potts, J., Lama, I., Coutts, K. and McKenzie, D . C Reliability of measurement of diffusion capacity of the pulmonary membrane and pulmonary capillary blood volume. The Physiologist 39(5): A-47, abstract 21:3, 1996 Stewart, I.B., Bacon, C.J., McKenzie, D.C. Carboxyhaemoglobin accumulation and clearance following repeated pulmonary diffusion testing. Unpublished data. 54 Turcotte, R.A., Perrault, H. , Marcotte, J.E. and Beland, M. A test for the measurement of pulmonary diffusion capacity during high-intensity exercise. Journal of Sport Sciences 10; 229-235, 1991. Vellar, O.D. Changes in hemoglobin concentration and hematocrit during the menstrual cycle. Acta Obstetrica Gynecologica Scandivnavica 53: 243-246, 1974. Vincent, W.J. Statistics in Kinesiology. Human Kinetics, 1995 Winer, B.J., Brown, D.R. and Michels, K.M. Statistical Principles in Experimental Design. 3rd Edition. McGraw-Hill Inc, 1991. 55 Abbreviations C O Carbon Monoxide C O H b Carboxyhaemoglobin [COHb] Carboxyhaemoglobin Concentration DLco Diffusing Capacity of Carbon Monoxide Units: m l C O (STPD) per m m H g Partial Pressure Change D M Diffusing Capacity of the Alveolar Membrane Hb Haemoglobin [Hb] Haemoglobin Concentration N O Ni t r ic Oxide O 2 Molecular Oxygen P A C O Partial Pressure of Alveolar Carbon Monoxide PcCO Partial Pressure of Pulmonary Capillary Carbon Monoxide pC02 Partial Pressure of Carbon Dioxide P C V Packed Ce l l Volume (haematocrit) p02 Partial Pressure of Molecular Oxygen V A Alveo la r Volume Units: m l (BTPS) V c Pulmonary Capil lary Blood Volume V C O Rate of Carbon Monoxide Production V C 0 2 Rate of Carbon Dioxide Production V 0 2 Rate of Oxygen Production 5 6 Appendix I: Lung Diffusion: Its Measurement, Components and Determinants. DIFFUSING CAPACITY OF THE LUNG Diffusing capacity of the lung entails the transfer of O 2 from the atmospheric side of the alveolar epithelium to its binding with erythrocytic haemoglobin. While both steady-state and a number of variations of single-breath techniques are available for the measurement of DLco, lung diffusion is most commonly assessed clinically using a single-breath method originally developed by Krogh (1915). This method quantifies the rate of disappearance of CO from the alveolar space within the lung following the inspiration and breath hold (usually 10s) of a known concentration of CO and assumes that the diffusion gradient for CO across the exchange surface does not limit its transport. This assumption is valid because carbon monoxide has a very high affinity to haemoglobin and at low concentrations is effectively removed from the blood plasma upon its diffusion across the membrane. This ensures that a large CO concentration driving force is maintained between the epithelial layer of the alveolar membrane (on the atmospheric side) and the pulmonary capillary blood plasma even when pulmonary capillary blood perfusion is relatively low. In general, D L C O is the volume of carbon monoxide diffusing across the alveolar membrane per mmHg partial pressure change from mean alveolar air to mixed pulmonary capillary blood (equation 1). 57 Equation 1: Basic Equation for Diffusing Capacity D L c o = V C O P A C O - P C C O where D L c o = diffusing capacity of CO V C O = volume (ml) of CO transferred per minute and P A C O - PcCO = difference between mean alveolar and capillary CO partial pressure In the single breath procedure, the lungs may be considered as a closed bag (of volume V A : the alveolar volume) from which CO is removed at an exponential rate proportional to its concentration gradient. Under normal circumstances, the pulmonary capillary CO partial pressure (PcCO) is negligible and ignored. D L c o at breath-hold time (t) is a function of the initial and final alveolar fraction of CO and the alveolar volume ( V A ) (equation 2). 58 Equation 2: The Krogh Equation for Single Breath Diffusing Capacity D L C O = V A X 60 x L n [ F A C O o / F A C O t ] x STPD correction 713 x t where D L C O ( m l C O / m i n / m m H g ) STPD V A = alveolar volume (ATPS in ml) t = breath-hold time (60 in the numerator converts seconds to minutes) 713 constant reflecting C O transfer L n = natural logarithm F A C O O = initial alveolar C O concentration F A C O C = alveolar C O concentration at the end of breath-hold STPD = standard temperature and pressure dry 1.05 = correction factor for 5% carbon dioxide in expired air removed prior to analysis 713 = the correction factor for conversion from concentration gradient to partial pressure difference = P B Of 760mmHg - Pwater vapour at 37° c of 47mmHg In the single breath procedure, an inert gas that w i l l not diffuse across the alveolar membrane (usually helium (He)) is present in the inspired mixture and has two purposes: to assess the initial alveolar C O fraction from the inspired fraction; and to determine the V A via its dilution in the total lung volume. Because helium does not diffuse to any great extent, the ratio of inspired He fraction to alveolar He fraction (assumed to be equivalent to the expired He ~ fraction), w i l l equal the ratio of inspired C O fraction to alveolar C O fraction after inspiration but before any diffusion has occurred. Thus, the initial fraction of C O 59 ( F A C O O ) may be determined from the change in helium fraction from inspired to alveolar air multiplied by the inspired C O fraction (equation 3). Equation 3: Calculation of the Initial Alveolar CO Fraction. F A C O O = FiCO x FEHe FlHe where FiCO = inspired CO fraction FEHe = expired helium fraction FlHe = inspired helium fraction If both sides of the equation are divided by the expired CO fraction (assumed equivalent to the alveolar CO concentration at the end of breath-hold time ( F E C O = F A C O O , the following equation is obtained (equation 4). Equation 4: Calculation of the Initial to Final Alveolar CO Fraction Ratio. F A C O O = FiCO x FEHe FlHe FACOt F E C O = FEHe x F I C O F E C O x FlHe where F E C O = expired CO fraction As long as the change in the FCO:FHe ratio from inspired to expired gas remains proportional to changes in either CO or He fraction, the relationship is independent of the actual quantites of inspired CO or He. It is convenient to assume 60 that FiCO is equal to FiHe cancelling these variables out of the equation. The ratio [ F A C O O / F A C O I ] may thus be regarded as equal to [(FEHe/FECO) x (FiCO/FiHe)] and is used in the calculation of D L C O in this study (Equation 2, p20). V A may also be determined via He dilution in the total lung volume (which includes the inspired volume and the residual volume) with a correction made for the dead-space of the diffusion instrument and for anatomical dead-space (equation 5). Equation 5: Calculation of the Alveolar Volume. V A = (Vi-VD) x FiHe x 1.05 FEHe where V A = alveolar volume Vi = inspired volume V D = dead space (anatomical and instrument) FiHe = inspired He fraction FEHe = expired CO fraction 1.05 = dilution factor for the C02 fraction which is normally chemically removed from the sample D L C O a s a ratio of V A may be reported to correct for variations in inspired ventilation and lung size (equation 6). Equation 6: Calculation of Diffusing Capacity to Alveolar Volume Ratio D L C O / V A = D L C O / V A (1 BTPS) where BTPS = Body Temperature and Pressure Saturated 6 1 COMPONENTS OF LUNG DIFFUSION Because diffusing capacity of the lung may be considered as a conductance of the lung for CO, its inverse, the total resistance of the lung (1/DLco)/ can be partitioned into two component resistances as shown (equation 7). Equation 7: Partitioning of Diffusing Capacity into its Membrane Component and Pulmonary Capillary Volume Component + DLco D M 6 X V C where DLco = diffusing capacity of the lung for carbon monoxide D M = diffusion across the alveolar membrane 0 = the reaction rate of carbon monoxide with haemoglobin and Vc = volume of the pulmonary capillaries In essence, l/0Vc may be considered as the component of resistance that varies with changes in PA02 and 1 / D M as all other resistance (Crapo and Forster, 1989). These authors describe a useful anatomical model for the partitioning which though not strictly correct is helpful in understanding the factors affecting diffusion. Using this model, the first resistance (1/DM), can be thought of as the resistance due to the movement of CO across the alveolar membrane to the surface of a red blood cell. This movement requires the crossing of 3 cell layers (the alveolar epithelium, the alveolar basement membrane and the capillary endothelium), the interstitial fluid within the alveolar membrane and a layer of blood plasma. Aveolar membrane diffusion ( D M ) therefore depends on the thickness and composition of 62 the involved pathways, which may be altered in disease (as for example in pulmonary fibrosis); and the surface areas of air-tissue and tissue-blood interfaces, which may in turn be affected by changes in the pulmonary capillary blood supply or alveolar ventillation. D M may also be decreased by changes in the back pressure of C O caused by C O bound haemoglobin already circulating in pulmonary arterial blood. The second resistance (1 / 9 V c ) , is a product of two resistances: one due to the reaction of C O with the red blood cell haemoglobin (1/0), and the other to the pulmonary capillary blood volume (1/Vc) . The combined resistance, depends upon the surface area of the erythrocytes, diffusion across the erythrocyte membrane and wi th in the blood cell as well as the chemical reaction of C O with haemoglobin. The chemical reaction term (0) refers to a predetermined rate that 1ml of normal [Hb] can pick up C O per I m m H g concentration gradient. Consequently, changes in the concentration of erythrocytic haemoglobin and factors which affect the saturation kinetics of the reaction (namely the partial pressures of 02 and C O 2 (p02 and pC02), p H , and temperature) all affect the combined resistance. Because 0 varies in a standard way with alterations in p02, both 1 /Vc and 1 / D M may be calculated from the slope and y-intercept respectively of a linear line of 1 /DLco plotted against 1/0. Specifically, the slope of the line estimates 1 /Vc , while the Y-intercept represents 1 / D M . Each value of 1/0 is normally calculated as described by Forster et al. (1986). It is the mean capillary O 2 tension ( P c 0 2 ) which determines 0 and this is very difficult to quantify. While the alveolar partial pressure of oxygen ( P A O 2 ) is typically the same as the end capillary partial pressure of oxygen (PecC02), Pc02 is assumed to be 15mmHg lower. Note that 0 is also a function of [Hb], as its uptake depends on the number of red cells present (equation 8). 63 Equation 8: Determination of Theta. 0.34 + f0.006 x P0O2I iHbJ 15 where 9 (theta) = the reaction rate of CO and red blood cell haemoglobin Pc02 = mean capillary partial pressure of oxygen = PA02-15 [Hb] = haemoglobin concentration P A 0 2 can be estimated by using the alveolar gas equation (equation 9), assuming a respiratory exchange ratio (RER) of 0.8 and that the arterial pressure of carbon dioxide (PaC02) is equal to an alveolar P A C O 2 of 40 mmHg. Equation 9. Alveolar Gas Equation PA02 = [Fi02 x (PB - 47)] - P A C 0 2 [FI02 + (1 - FiQ2)] RER where PA02 = partial pressure of alveolar oxygen F1O2 = fraction of inspired oxygen PB = barometric pressure P A C 0 2 = partial pressure of alveolar carbon monoxide F1O2 = fraction of inspired oxygen RER = respiratory exchange ratio 64 FACTORS AFFECTING LUNG DIFFUSION An individuals pulmonary diffusing capacity is largely determined by anthropometric factors. Larger D L C O is correlated with greater body dimensions including weight, height, surface area and lung volume (Crapo and Forster, 1989). Reduction in D L C O results from both obstructive and fibrotic lung disorders, marijuana and cigarette smoking, pulmonary oedema, and anaemia. Increases may occur in association with asthma, pulmonary haemorrhage, and left to right circulatory shunts (Crapo and Forster, 1989). In healthy individuals, D L C O also depends on body position, increasing 15% from the upright to supine or prone position (Chang et al., 1992). In addition, D L C O increases during exercise. Turcotte et al. (1992), noted 42% and 65% increases in diffusing capacity during moderate and intense exercise respectively. Paradoxically, a number of researchers have noted reductions in D L C O following maximal exercise (reviewed by Sheel, 1995). The magnitude of the post-exercise decrease reported in the literature varies from 2% to 19% depending on intensity and duration of the exercise bout and the time following exercise that diffusion capacity is measured. Miles et al. (1983) observed 2% reductions in DLQO, while Manier et al. (1991), found 10% reductions in D L C O / 24 hours and 28 minutes respectively after the completion of a marathon. Sheel (1995), in a study designed to follow the timecourse of post-exercise alterations in DLCO, observed a peak (13%), reduction in pulmonary diffusion six hours following maximal exercise which was still 6% below baseline 24 hours after the exercise bout. Although exercise training induced changes in D L C O have not been reported in the literature, physically active people generally have a higher D L C O value (Crapo and Forster, 1989). An increase in D L C O is likely to occur in trained athletes as a result of increased total blood volume and hence pulmonary capillary blood volume. Sheel (1995), found small (7.5%) but statistically non-significant differences 65 between the D L C O of a group of highly trained and a group of moderately trained males. 66 Appendix II: The Human Menstrual Cycle THE HUMAN MENSTRUAL CYCLE The menstrual cycle of humans and old world primates is characterised by circa-lunar fluctuations in the levels of hormones of the hypothalamic-pituitary-ovarian (HPO) axis, early-cycle bleeding (menstruation), and mid-cycle release of a mature ovum (ovulation) (figure 2). The phases of the menstrual cycle may be categorised by ovarian hormone changes or changes in the uterine lining. During the early part of the ovarian "follicular" phase, reproductive hormone levels are low and menstruation occurs. Increasing levels of oestrogen cause the cessation of menstruation and the onset of development of an encapsulated ovum (the follicle). At this time, the uterine lining (endometrium) undergoes proliferation: it begins to thicken and develop a new blood supply. A late follicular shift from the negative feedback regulatory action of oestrogens on the hypothalamus to a positive feedback mechanism results in a rapid but short-lived increase in oestrogen levels and a slightly delayed gonadtrophin level rise. This results in the rupture of the mature follicle and exocytosis of the ovum from the ovary marking ovulation. The remainder of the follicle becomes a steroid hormone producing body known as the corpus luteum. During the ovarian "luteal" phase, the corpus luteum produces large amounts of both progesterone and oestradiol which act to maintain the uterine lining. In the latter part of the luteal phase corpus luteal function declines. Lowered steroid hormonal levels, mediated by the action of locally produced prostaglandins, initiate transient vasoconstriction of arteries leading to the endometrium. Necrotic erosion of the superficial layer of the ishaemic endometrium by proteolytic enzymes released following leukocyte and macrophage invasion, results in the shedding of outer cell layers of the endometrium and rupture of blood vessels causing the onset of menstruation. 67 0 2 4 6 8 10 12 14 16 13 20 22 24 26 28 DAYS OF MENSTRUAL CYCLE Figure 12: The Human Menstrual Cycle (Adapted from Guyton, 1996) 68 CHANGES IN RESPIRATORY FUNCTION OVER THE H U M A N MENSTRUAL CYCLE Menstrual cycle phase is already known to influence respiratory function. Both resting (Shoene et al., 1981) and maximal exercise (Jurkowski et al., 1981) minute ventilation (VE) increases in the luteal phase of the menstrual cycle compared to the follicular phase. Resting hypercapnic and hypoxic drives are also higher in the luteal phase (Schoene et al., 1981; Dombovy et al., 1987). Menstrual phase dependent alterations in ventilatory drive are thought to be a result of progesterone mediated stimulation of central respiratory centres. Firstly, the synthetic progestin medroxyprogesterone acetate (MPA), which increases ventilatory drive, has been previously identified in the cerebrospinal fluid (Scoggin et al., 1978). Secondly, the changes in oral occlusion pressure found by Schoene and coworkers (1981) to coincide with changes in ventilatory drive are generally thought to reflect total neural drive. Peripheral factors may also be involved in ventilatory changes over the menstrual cycle; Chen and Tang (1989) have noted a higher inspiratory muscle endurance in the mid to late luteal phase compared to the mid follicular phase. 69 Appendix III: Calculation of Intraclass Correlation Coefficient The Intraclass Correlation Coefficient may be calculated from a Repeated Measures A N O V A using the following formula (Equation 1, from Vincent, 1995, pg 179). Equation 1: Intraclass Correlation Coefficient Tintraclass = MSsubject - MStrial+error MSsubjec t where rmtraclass = Intraclass Correlation Coefficient MStrial+error = Mean Square of Trial Effect and Error MSsubject = Between Subjects Mean Square Because both P C V and [Hb] repeated measurements were made on separate days at different points of the menstrual cycle, day to day variation and/or menstrual cycle effects w i l l cause changes in the means of the separate trials. In order to assess only the measurement error (or reliability of the measurement procedure), the trial effect should be ignored. Vincent (1995) recommends the use of the following modified formula in this situation (Equation 2). 70 Equation 2: Modif ied Intraclass Correlation Coefficient ^intraclass = MSsubject - M S e r r o r MSsubjec t where r2intraclass = Modified Intraclass Correlation Coefficient M S e r r o r = Error Mean Square MSsubject = Between Subjects Mean Square 7 1 Appendix IV: Ethical Approval Certificate and Consent Form 72 Appendix VI: Menstrual Cycle Diary Name: Menstrual Cycle Diary M o n t h : Year: Cycle Day 1 2 3 . 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Date Tampons/pads/day Record 0 = none, 1 = minimal, 2 = moderate, 3 = moderately intense, 4 = very intense Amount Flow Cramps Breast Sore: Front Breast Sore: Side Fluid Retention Mucous Secretion Constipation Headache Sleep Problems Feeling Frustrated Feeling Depressed Feeling Anxious Record M = much less, L = a little less, U = usual, Y = a little increased, Z = much increased Appetite Breast Size Interest in sex Feeling of energy Feeling of self-worth Outside stresses Basal Temperature SUPIN£ n£ARr«AT£ Comments (temperature taken late, feeling sick, poor sleep, etc) JC Prior Copyright 1990 78 Appendix VII: Daily Exercise Record 79 Name: DAILY EXERCISE RECORD Month: Year: Date Men Day Vigorous/ Strenuous Exercise (HR 10s >25) (hrs:mins) Mild/ Moderate Exercise (HR 10s 15-25) (hrs:mins) Mode (eg: running, walking, swimming, resistance training) Comments 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 80 Appendix VIII: Initial Questionnaire 8 1 I N I T I A L VERSION GENERAL AND HEALTH QUESTIONNAIRE C o d e Number ~~~ Date/Time Birthday 1. How would you describe your work? 2- What 's your partnership status? 3. W h o is at home with you? n ° - ° n e family D a r t n e r / partner roammate(s} 4, H o w m a n y years of schooling do you have? less than S 8 ~ 1 2 « 1 f t m m _ _ _ ™ i l l l * ? L more than 16 5 " A m y D U o n a ^ ^ r m o n e s (including the pill) or other med ica t ions?Yes N o w ^ J f y ^ p t e a $ e J i s t i h e m Sx Do you have asthma, other lung problems or significant Hlness? Y e s N o 7. Is there anyone related to you who has/had a broken bone (without a fall or accident) or who is getting/became shorter and developing/developed a hunched back? 8, Do you currently $moke? Y e s N o & Are you a past smoker? Y e s N o If yes, how long since you stopped? mnnthi'V^aro 8 2 10, Do you currently drink alcohol? Y e s N o if yes, how many glasses of wine or beer do you drink per week? 1 o r 2 3 - 7 8 - 12 more than 12 If yes, how many glasses of spirits do you drink per week? l o r 2 3 - 7 8 - 1 2 more than 12 11. Do you currently drink caffeine containing drinks (non-herbal tea, coffee, or coke)? If yes, how many glasses do you drink per week? 1 o r 2 3 * 7 8 - 1 2 more than 12 12. Are you on a diet for a health concern or to lose weight? Y e s No If yes, p lease explain? 13, Has your weight changed in the past 5 years? Y e s No If yes, has it? Increased decreased fluctuated How much has It changed/fluctuated? ^ & ^ s 14, How much do you feel you should weigh? gpp lbs 15, What was your highest non-pregnant weight? kgs tos 16- What was your lowest non-pregnant weight? kgs lbs 83 NUTRITION 17. Are you taking any multiple vitamins? Y e s N o 18, A re you taking any calcium supplements? Y e s N o If yes: How many mg are there in each pill? How many pills do you take each day? How long have you been taking them? 1 0 , Have you had any calcium containing foods (milk, yoghurt, cheese, creamed cheese , canned sa lmon, etc) over the last 24 hours? Pood Amount : Breakfast Lunch Dinner 8 4 E X E R C I S E 20. Do you do an hour or more of vigorous exercise per week? Y e s No If yes, please indicate the type of activity, the average frequency and duration. T y p e o f A c t i v i t y F r e q u e n c y ( n o s e s s i o n s p e r w e e k ) D u r a t i o n ( t i m e p e r s e s s i o n ) 21. How many months over the last year have you exercised a similar amount? 22. How many years (over the last 5) have you exercised a similar amount? 23. Have you done more than 1 hour of aerobic exercise Y e s N o per week in the past? If yes , how many years did you do this for? years M E N S T R U A L C Y C L E 24. Are you having regular periods? Y e s . N o 25. How long is your cycle length? days £6, How many days Jong is your flow? days 27. C a n you usually tell, by the way you feel, that your Y e s N o period is coming? 8 5 2$ , Do you usual ly experience the following symptoms? If yes, p lease indicate the time in your cycle that you notice them (respond "intermittently" if this seems the most appropriate answer). T i m e i n c y c l e Breast Tenderness Y e s No Appetite Changes Y e s No Mood Changes Y e s No Fluid Retention Y e s No Stretchy Vag ina l Mucous Y e s No 29. Are there any changes in your symptoms after your flow starts? Illlllllllllllllllillllllllllllillllllll l l l l l l l l l If yes, p lease explain? 30. How many times did you menstruate in the last year? n o n e 1-2 3-6 7-9 10-17 more than 17 31. Have you menstruated in the last 6 months? Y e s No 32. How many periods have you missed in the last 5 years? (normal = I0>i4/year) periods 33. Have you been pregnant in the last 5 years (full « 10 months)? Y e s No If yes : How many months after delivery until your period started? periods 8 6 34. Have you recently had P M S ? Y e s N o If yes, p lease describe your symptoms and their timing in relation to flow? 36, Are you currently taking oral contraceptives? y e s No 36. Have you taken them in the past? Y e s No If yes : How long did you use them for? months/years How long ago s ince you used them last? 37* Did you have regular periods when you stopped the pills? y e s N o If no: How long were they irregular for? months/years 8 7 Appendix IX: Data Sheets INITIAL S U B J E C T D A T A S H E E T Date / / Height (m) Code DOB / _ _ / ( ) day month year Triceps SupSpi Subscap Abdom AntThi MedCalf Suplliac (IliacCrest) S of Six Any Respiratory or Endocrine Medical Conditions Y/N Current Smoking Y/N Past Smoking Y/N Length ago No per day Oral Contraceptives Y/N Past Oral Contraceptives Y/N Length ago Type Menstruate Regularly Y/N Able to predict onset of bleeding Y/N No of times menstruated in the last year None 1-3 4-7 8-10 10-17 >17 Training/Sport Hrs/Week (average training or intensive exercise) Months/Year 8 8 SUBJECT DATA SHEET Code Date / / day month year Time P B (mmHg) T R O O M (°C)_ Weight (kg) Humidity (%) Test Time 1 2 3 4 5 Day in menstrual cycle Time at: Last Vig Ex Last Meal Last Coffee F V C (ml) FEV1 (ml) % FEV25-75 PEFR Hb (mg/lOOdl) PCV (haematocrit) C O H b Oes Prog Fi02 (between) 21% Tests I and II Fi02 VinBTPS He C O Hold 21% D L C O (ave) V A (ave) D / V A (ave) 90% Tests III and IV Fi02 VinBTPS He C O Hold 90% D L C O (ave) V A (ave) D / V A (ave) D M V C 8 9 FINAL SUBJECT DATA SHEET Code Date / _ _ / Triceps SupSpi AntThi Height (m) DOB _ _ / _ _ / ( ) day month year Subscap Abdom MedCalf . ._ Suplliac (IliacCrest) S of Six Any Respiratory or Endocrine Medical Conditions Y/N Current Smoking Y/N Oral Contraceptives Y/N Menstruation During Study (circle) Did not Menstruate Irregular Cycles Regular Cycles Training/Sport Hrs/Week (average training or intensive exercise) Months/Year 9 0 Appendix X: Final Questionnaire 9 1 FINAL V E R S I O N GENERAL AND HEALTH QUESTIONNAIRE C o d e Number Date/Time 1. A re you on any hormones (including the pill) or other medicat ions?Yes N o Jf yes , p lease list them 2. Do you have asthma, other lung problems or significant i l lness? Y e s No 3. Do you currently smoke? Y e s N o 4, Do you currently drink alcohol? Y e s N o If yes , how many g lasses of wine or beer do you drink per week? If yes, how many glasses of spirits do you drink per week? more than 12 more than 12 5. D o you currently drink caffeine containing drinks {non-herbal tea, coffee, or coke)? Jf yes , how many g lasses do you drink per week? 1 or 2 3 * 7 8 - 1 2 more than 12 9 2 NUTRITION 6, Are you taking any multiple vitamins? Y e s N o 7. Are you taking any calcium supplements? Y e s N o If yes: H o w many mg are there in each pill? H o w many pills Oo you take each day? How long have you been taking them? S, Have you had any calcium containing foods (milk, yoghurt, cheese , c reamed cheese , canned sa lmon, etc) over the last 24 homsl Food Amount Breakfast Lunch Dinner 9 3 9. Are you on a diet for a health concern or to lose weight? Y e s No If yes, p lease explain? EXERCISE 10. Do you do an hour or more of vigorous exercise per week? Y e s N o If yes, p lease indicate the type of activity, the average frequency and duration. T y p e o f A c t i v i t y F r e q u e n c y ( n o s e s s i o n s p e r w e e k ) D u r a t i o n ( t i m e p e r s e s s i o n ) 11. How many months over the last year have you exercised a similar amount? 12. How many years (over the last 5) have you exercised a similar amount? If yes, how many years did you do this for? years 13. Have you done more than 1 hour of aerobic exercise Y e s N o per week in the past? ^^^^II^^^^BIIIIIIIiHliili^^HiHB^Bi^HBiBI .111111 9 4 MENSTRUAL CYCLE 14. A re you having regular periods? Y e s N o 15. How long is your cycle length? days 16. How many days long is your flow? days 17. C a n you usually tell, by the way you feel, that your Y e s N o period Is coming? 18. Do you usually experience the following symptoms? If yes, please indicate the time in your cycle that you notice them (respond "intermittently" if this seems the most appropriate answer). T i m e I n c y c l e Breast Tenderness Y e s N o Appetite Changes Y e s No Mood Changes Y e s No Fluid Retention Y e s No Stretchy Vaginal Y e s No Mucous 19. Are there any changes in your symptoms after your flow starts? If yes , please explain? 20. Over the study have you? not menstruated at all menstruated menstruated normally abnormally 21 . Over the study have you taken oral contraceptives? y e s N o 95 Respondent I.D. # In this section I would like to ask you questions that will help us understand how women's hormones relate to bone structure. We ask everyone these questions. 5* REPRODUCTIVE HISTORY (FEMAU3S) 5.1 * Before menopause, have you ever gone 3 months or more without a menstrual period? (not including pregnancy or during breastfeeding) • Yes • No '—* Go to 5.2 What was the longest single period of time without a menstrual flow? months If you count all the periods you have missed throughout your menstruating years, how many months would that be? months (this question asks for the cumulative time) 5.2* Have your menstrual periods stopped for more than one year? (No period one year or more after last menstruation) • Yes • No l—» At what age? .-- years 5.12 How old were you when you had your first menstrual period? years 5.3 Have you had your uterus removed (hystereaomyf! • Yes • No I—» At what age? years 5.4* Have you ever had one or both ovaries removed? • Yes, one ovary removed at what age? • Yes, both ovaries removed at what age? (if ovaries were removed on separate occasions, write the age al which the second ovary was removed) • Yes, do not know how many at what age? • No See notes in manual 9 6 Respondent I.D. # I'm going to ask you a few questions on your eating habits. 8.7 a) I am going to read two sentences for you. Please answer True (T) or False (F) for each statement as it pertains to you. I enjoy eating too much to spoil it T • F • by counting calories of watching my weight. I consciously hold back at meals in order not to gain weight. T • . F • b) Which of these best describes you? On a scale of 0 to 5, where 0 means no restraint in eating (eating whatever you want, whenever you want it) and 5 means total restraint (constantly limiting food intake and never "giving in"), what number would you give yourself? 0 Eat whatever you want, whenever you want it 1 Usually eat whatever you want, whenever you want it 2 Often eat whatever you want, whenever you want it 3 Often limit food intake, but often "give in" 4 Usually limit food intake, rarely "give in" 5 Constantly limiting food intake, never "giving in" Now the questions I will ask will relate to the use of tobacco. 9, TOBACCO 9.1 Have you ever used any of the following tobacco products daily for at least 6 months? Cigarettes • Yes • No Pipes • Yes • No Cigars • Yes . • No Chewing tobacco • Yes • No 9 7 Respondent I.D. # o ^3 Q S no c o c u «-» cu U 3 O >> <o > CU JS ' i ? CTJ u c o c u <£ o == o 33 As a child? More As a child? Same As a child? Less As a child? Never In your teens? More In your teens? Same • In your teens? Less In your teens? Never In your 30's (If subject 40 years or over) More In your 30's (If subject 40 years or over) Same In your 30's (If subject 40 years or over) Less In your 30's (If subject 40 years or over) Never Serving Size • 125 ml (0.5 cup) • 250 ml (1.0 cup) • 375 ml (1.5 cup) • 60 ml (.25 cup) • 125 ml (0.5 cup) • 250 ml (1.0 cup) • 15 ml (1 tbsp) • 30 ml (2 tbsp) • 60 ml (4 tbsp) • 125 ml (0.5 cup) • 250 ml (1.0 cup) • 15 g (0.5 oz) • 30 g (1 oz) • 60 g (2oz) • 125 ml (0.5 cup) • 175 ml (single) • 250 ml (1 cup) • 125 ml (0.5 cup) • 250 ml (1.0 cup) • 375 ml (1.5 cup) • 125 ml (0.5 cup) • 160 ml (.67 cup) • 250 ml (1.0 cup) During the last 12 months? I 1 J? u During the last 12 months? •vings p week During the last 12 months? sei month During the last 12 months? Never Fold Milk to drink incl. choc, milk & hot cocoa w/milk •a 0J o a o. M 9 § Milk/cream in tea/coffee Milk desserts (tapioca, rice pudding) Hard cheese (to eat, in sandwich or mixed dish) Yogurt Ice-cream, ice milk or frozen yogurt Cream soups made with milk 8 Respondent I.D. # o 2 "O B CQ C/3 13 O a > CD z 2. 5-1 o S a " et) Vi cu Z 01 .a 1 / 3 - em . 3 g g g 00 00 60 S 8 8 a 1 1 • a l l O D D • a l l 8 SP.S := .o a. _ * s a1 .s is ^ a, ~ S S §• g - ^ • o o g. " <=L a • a l l • n o • • • • o £ 1 cu JS M C j , I CD z o o o ft. 2§ a & 3$ U 1.9 cd O U ii ja ^ ° a 5 o •-a e •a =9 5g cd S J 3 3 a cS o H a, p. 3 .Is ed t_ U -Respondent I.D. # 43 cu O •s -s; a o 4 3 cu -5 © a cu a cu 5 <u £ 3 c/3 B O o o >» ."2 •3 c/3 M C 60 C o c S o o c CO CU K ) 8 -I 3 toe too cu cu c-s c-i £^ ©^  .2 •S a £ >» S © © a "5 C3 -Ct © © ©" 8 k a © ^ cu 5 © cu — 1| ° -a £ s £ © .?<> oo s; * 2 When in your teens? More When in your teens? Same When in your teens? Less When in your teens? None In your 30's (If subject is 40 years or over) More In your 30's (If subject is 40 years or over) Same In your 30's (If subject is 40 years or over) Less In your 30's (If subject is 40 years or over) None During the past 12 months? Serving /day During the past 12 months? Serving /week During the past 12 months? Serving /month During the past 12 months? None 8 2 CU cu inn caffeinated decaffeinated caffeinated Tea 1——:—-—-decaffeinated caffeinated rnlaq -—— 1 )^  i t 3 <u i Alcoholic beverages Respondent I.D. # In this section I will ask you about your physical activities and exercise. 11. PHYSICAL ACTIVITY 11.1 During a typical week in the past 6 months, how much time did you usually spend walking to work or school or while doing errands? 11.2 Which of the following describes the paid work you usually do or what you consider your job? Or if retired or unemployed, which best describes your (past or longest) job? • I am usually sitting during the day and do not walk around very much • I stand or walk quite a lot during the day but I do not have to lift or carry heavy things • I usually lift or carry light loads or I often have to climb stairs or hills • I do heavy work or have to carry loads 11.3 Do you currently participate in any regular activity or programme (either on your own or in a formal class)! • None • Less than 1 hour • Between 1-5 hour • Between 6-10 hours • Between 11-20 hours • More than 20 hours • Yes • No —• How many times a week? L—* How long per session ? minutes 1 0 1 Respondent I.D. # 11.4* On the average, during the last year, how many hours in a week did you spend in the following activities? Never, 1/2-1 hr 2-3 hrs 4-6 hrs 7-10 hrs 11-20 hrs 21-30 hrs 31 hrs + STRENUOUS SPORTS (such as jogging, bicycling on hills, tennis, racquetball, swimming laps, aerobics) VIGOROUS WORK (such as moving heavy furniture, loading or unloading trucks, shovelling, weight lifting, or equivalent manual labour) -MODERATE ACTIVITY (such as housework, brisk walking, golfing, bowling, bicycling on level ground, gardening) c U. of Hawaii Cancer Research Center 11.5 * On the average, during the last year, how many hours in a day did you spend in the following sitting activities? Never Less than 1 hr 1 to 2 hrs 3 to 4 hrs 5 to 6 hrs 7 to 10 hrs 11 hrs or more Sitting in car or bus Sitting at work Watching T V Sitting at meals Other sitting activities (such as reading, playing cards, sewing) •U. of Hawaii Cancer Research Center 11.6 On the average, during the last year, how many hours in a day did you sleep (include naps)! • 5 hours or less • 7 hours • 9 hours • 6 hours • 8 hours • 10 hours or more See notes in manual 1 0 2 Respondent I.D. # 11.7 * Rate your overall level of physical activity compared to your peers during certain times in your past life. When you were about 50 if subject 60 y. and over When you were about 30 if subject 40 y. and over Teenager Child A lot less active Somewhat less active About the same Somewhat more active A lot more active Now I want to ask you questions about being in the sunlight 12.1 * Did you ever expose a considerable part of your body to direct sunlight? A. During the past 12 months? • never • seldom • regularly • often If 60 years old or more. B. When you were about 50 years old? • never • seldom • regularly • often If 40 years old or more. C. When you were about 30 years old? • never • seldom • regularly • often For all. D. When you were a child or teenager? • never • seldom • regularly • often See notes in manual 1 0 3 Appendix XI: Timetables of Testing Dates 1 0 4 o co Q O O o z t-co LU t— z o to CO S Q Q in Q O co a o o O CO O O O Si o o CM CO CM o co 3 105 3 o O CD CO in g "xT co in o 9 o i w o to "5 o CM Q io CO I D X3 CO CO co C O I C O C M TO ' co •<* m C M CO ! C M p O CM Q CM > re Q Q TO C O I T ) 00 C M a> co in CM T J CO in in in o -O CM 10 CO T J C M 9 C M Q <3 | CM co e. co CO C M C O Q co C M in i f f l cn CM Q .3 >H co Q Q l Q , C M | m o O) o o co CD CM TT C 3 CD CM T J CM s o CM O co CO CD o o o T - CO >v CO CM TO Q ,3 T J CO o C O C M n co CD o C M TO CO CD CO •a 1 : IB TO a re C M C M > > CO O in i ° > o co a C O II _ l O a> co CO o CO o T J C M in o co CM-in co C M n ' C M in re C M in o co C M m o m o re O m CM CM co TO TO C O CD co r~ o o T J C O r>-o CO o CO g o C M Xi co CD CO C M C M C M ,1 3 II _ J o CM ,1 C M CD C M T ->- <tf . re o o CO CO m o C M C O C M C M > , o co JO C M co in a TO D TO C M CO "co CM in i o T J CM C M C O C M C M .O C M CO co ml O H I >< D co CM CM I O en CO m o CM > TO Q ro co CO Q CM CP co co p CO o co > TO Q .9 ,9 CO I CM CO IS' CO p c^  CO co o Xi C M in o co TO C M JH a> CM m p m co II — i O "J o T J C M C O C M o o C M m co CM Xi ' C M in a 00 co co o TO ' C O CO g m o a" m o C M in g o C M C M in g m T J CM in p TO CM CO CO S5 II u C M in g CM O m I C M > . Q CD C M m o o TO Q 1^ | o eo ' S I co II _j O a I g o o co C O C M O C M CO m o CO eo II _ i O C M CM CM I oo TO Q ,1 I o C M CO ffi o co o co II _ i O TO Q CM C O C M I O Xi CM m o ~ T O " CM in g CO o CD C M o C O EM O C M C M I CO CM o C M C M CO CD o o 1 0 6 o o o o CM 00 Q co Q a o o CO CM 1 P * Jf, o O gi cn £ -tt f g « 3 «d CO i l l o 1 0 7 -Day 13 20/05 2d Day 2 30/04 2a CO CM II - J O oo >> CO Q 15/05 2c Day 25 26/04 2e Day 23 19/05 2e Day 3 10/05 2b Day 17 ; 18/04 2d Day 19 14/05 2d Day1 08/05 2a CM II _l o Day 7 08/04 2C Day 7 02/05 2c Day 22 30/04 26 Day 20 20/03 1e CO > CQ Q 04/04 1b Day 15 28/04 1d Day 2 16/04 1a o> to II _l O Day 5 30/04 1b Day 25 07/05 1d Day 15 15/03 1d Day 8 21/04 1c Day 22 07/04 1e Day 3 28/04 1a CO II _i O Day 7 19/04 1c Day 5 05/03 1c Day 3 16/04 1b Day 16 01/04 1d Day 2 14/04 1b Day 3 03/03 1b >. CO Q 14/04 1a CM co II _i O 0 0 > , CO Q 24/03 1c Day-2 11/04 1a CL = 34 Day -1 28/02 1a CM n _ i O Day 22 03/04 1e CM co II _i O 1 0 >> CO O 21/03 1b CL = 29 Day 22 04/041e CL = 30 CBMS13 CBMS16 CBMS18 CBMS19 CBMS20 CBMS21 CBMS22 1 0 8 co CM Q CO a Q O O CM CM in co of Q CD O o O CM il t i l o O g. » 2 J fit 1 0 9 o o CJJ z _ l CL 2 % 3 CC LU CO UL o CO < 5T< Q O co L O o O 3 C D Q u C O O O O O O 1 1 0 C5 Z _J CL 2 < CO Z> rr in co u_ O CO g Q Day 15 04/07 3d C M II _i O Day 25 07/07 38 Day 27 25/05 38 Day 18 16/05 3d CL=31 C O C M II - J o CL = 25 Day 11 09/05 2c Day 4 16/06 3b miss 2b Day 15 30/05 2d CL = ? miss 2d cs CNJ II —1 O Day 8 23/05 2c Day 22 06/05 28 in C O II _i O Day 7 23/05 2c Day 25 26/04 28 Day 4 19/05 2b Day 16 30/04 2d Day 23 19/05 28 Day 17 18/04 2d CL = 34 Day 7 1 C M Day 19 14/05 2d C O C O u o CL = 27 C M II —j O Day 7 08/04 2c Day 22 05/0528 Day 7 02/05 2c Day 25 07/0528 miss 1b Day 15 28/04 1d CL = 59 Day 5 30/04 1b Day 19 01/05 1d C M II —1 O Day 8 21/04 1c Day 22 07/0418 co u _ i O Day 7 19/04 1c Day 23 26/03 1e Day 3 16/04 1b Day 16 01/04 1d miss E2 Day 22 09/0418 Day 2 14/04 1b Day 3 03/03 1b ! Day 18 21/03 1d CM co II _1 o co > CO Q 24/03 1c Day 17 04/04 1d 3 n _i o CL = 29 h-C M II —1 o Day 9 12/03 1c cn C M II _j O 05/06 3c Day 22 03/0418 CL = 32 Day 5 21/03 1b CL = 29 C O > l CO o 26/03 1c CL=38 Day 22 04/0418 o C O II _ l O C O C M II _l O CBMS13 CBMS16 CBMS18 CBMS19 CBMS20 CBMS21 CBMS22 1 o 18 CM CO s to CO >-_J < z < z CO 2 8 LU < I > X o m cc 3 LL o CO LU a CO CM oi T 3 I CM a iffl " O CM I O in o 0 ' CM 1 CM o CN Q ft a CM CO oo CM I o a CO CM O 1^ CO CM II _ l O o o o CO 2 CO O o CO 2 ca o CO CM 3 co -O. CO co o CO CM O co n ,8 a CD CO LO O CO ft w1 o TJ I CM I f) CM O CM a CD CM | CM CM i Q CD CM s CO o CM CM Q CM 0 0 CM o CM s II - J o 8 CD CM § O 8 II _ l O CO a LO P 3 o 0 0 CM 5 O CM II o. co co CM >. CO Q CO ft co Q a o CM I O 0 0 o O CO CM II _l O 0 0 CM II _l O CO CO O CO o CO 2 m O o to 2 ca O 0 0 o CO 2 ca o to 2 ca O 1 1 2 CO —I < z < LU 3 _ l 9 LU O Q LU o < Q. U . o co LU Q co Q S o o o o eg O O O a o CO CM O O o o o o o 1 1 3 C O co >-< U i 2 _l _ j LU o Q LU 0. LL o CO Q CL = 27 Day 27 25/05 3e Day 17 15/05 3d CL = 31 Day 21 16/06 3d C O C M II _ l O m C M II _i o Day 11 09/05 2c •Day 4 16/06 2b Day 8 03/06 3c Day 15 30/05 2d CL=? Day 4 30/05 2b Day 17 02/06 2d C O C M II _ l O Day 8 23/05 2c •Day 22 06/05 2e m C O II _i O >< CO Q * 23/05 2c Day 25 26/04 2e Day 4 19/05 2b •Day 16 30/04 2d •Day 23 19/05 26 Day 3 19/05 2b C O II - J O t--> . CO Q * 21/04 2c O ) >l CO O * 14/05 2d miss 2a C O C O II _ l O CL = 27 8 II _i o Day 7 08/04 2c Day 22 05/05 2e •Day 7 02/05 2c m CO Q * 28/04 1d cn m il - j O •Day 25 07/05 1d CL = 27 C O >> CO Q * 21/04 1c Day 22 07/0416 C O II _i O Day 23 26/03 1e C O > , CO Q * 16/04 1b Day 22 09/04 1e •Day 2 14/04 1b Day 18 21/03 1d C M C O II _1 o Day 17 04/04 1d Tt C O II _ l o o> CM II _ l O r-. C M II _ l o Day 9 12/03 1c O) C M II _ l O Day 10 05/06 3c Day 22 03/0416 CM C O II _ l O o> C M II _l O Day 8 26/03 1c co co II _i O •Day 22 04/0416 CL = 30 C O C M II _ l O CBMS13 CBMS16 CBMS18 CBMS19 CBMS20 CBMS21 CBMS22 4 Appendix XII: Raw Data Summary 1 1 5 LU _ l O >-o >-LT O > o z < z < o z < >-O > o z < C O o C L I-co LU K— UJ > < Q UJ DC LU LU Q X o z o Z o co rr < o LL o o < 0. < o z g C O C O 00 CO CO CO CM CM CM CM CM CM CM CM s CM CM 8 8 cn LO co CM CM CM CM CO CO 1 1 6 CD CO O 0 5 oo s CD CO CO 8 o CO co 7 CP CO T3 O CO 0 0 03 CO CD CM CO CO o 8 cn in 55 8 UJ _ l o >-o >-rx O 3 > O Z < Z < Q Z < >-cr o 3 > o Z < z X CO I-z o a. i -co LU I-UJ > < Q 111 cr 3 CO < 111 UJ O > a o o —i CO rr _ J CL < o >-cr < z O 2 CO CD o re CO re 9 CM 83 8 C O CM CO 83 cr> CM co L O i n 1 1 9 CM CO o 00 CM CM CO 00 CM CO CM 05 CM CM CO CM CO CO CM co 1 2 0 CO co CO m cn m cn O) CM T t AN AN CO CO CO CO T t CO co z LU X h- CM 92 T t i n z o T t 55 CO CM CO o T t T t CO h» CO CO d CO o h- Q L miss z 1— miss E TEST POI SIS CO CD miss E TEST POI SIS TE T t 8 CM o o o 9 CM CO CO E TEST POI ANALY ! OVER T J co CO CO CO T t CO CO d > HANGES HANGES CM O) cn o i n T t LO LL HANGES O CO ^— T t CM i n DAT CYCLE HANGES co T t CO CO T t CO co d UJ q* CYCLE O O r> CO < o T t LO T t cn i n co co r - CO CO CO T t < UJ O co T t T t CO T t CO CO d UJ -1 >• o s> cc _J o Q O o n X) I- >• < LU CM 0> f>. CO o CM < cc NOVUL o CM CO i n CO T t * o NOVUL DJUS CO T t SSII CO T t CO CO d 11 < < ro E 83 *- m T— T t CO o ^— o r> CO T t cn T— CO CM T t CM CO 00 v— K-^ z CO CO T t CM CO CO CO T t T t T t CO CO T t tz < Z S z EA O < UJ Q ^ 5 OINTS OINTS o to CO CM 56 CO CM co S3 8 c^ l CM CO o CM i n OINTS CO CO CO CM CO CO T t T t CO CO T t CO d O cc o_ C O C O IS CD mis mis Is UJ CM CO CO T t o LO o T t i n CO CO t — T t T t co CO o CM Y— o CM CO o m co i n CAPA OVL LYSIS OVER CO CO T t CM CO CO T t T t T t CO CO co T t co d < CO "O T t O z < LU ,NGE: §8 LO ^— 3 CO OJ T™ co o i n i n T t 8 CO m CM O T t CO ,NGE: CO CO CO CM CO CO CO T t T t T t CO CO ?' co d u_ _l <t LL o X Q IYCY o o o LU IYCY 1 •<t co O T— cjj- cn CO 8 8 o t>-T t cn CO CO 8 m I- EC o CO CO T t CO CM CO T t T t LO CO T t CO CO d co O o liss 1— _ l liss ~> < Q liss o —1 n XJ 1= z 1A0 LU i — N. rr T— o a LO CO T t o 8 CO v— T t LO CM cn o CO 8 o i n OBI 1A0 CO C O CO iri CO CO T t CO Tj" CO CO CO T t CO d OBI ST =3 — i CO ,EMOGL BE AD, CO mis ,EMOGL o T— CO T t CO CM CO CO CO O) O CM > ,EMOGL o o o o o T— •»— TJ— CM CM CM LU < CO CO CO CO CO CO CO CO CO CO CO CO CO Z Q < 2 2 2 2 2 2 2 2 2 2 2 < CD CD CO CD CD CD CD CD CD CD CD CD CD LU r -o o o O o o o o o o O O O 2 CO 1 2 1 cn CM T t OJ rv <& CO T t CM o LO CO CO z T t yZ CO CO CO gj < z co CO CO CM CM co z u5 I T t i— o rv § CO T t OJ CO to CO CO CO CM rv' CM CO CO CM T t 1— co to z \— CO CO o CO z CD E E 0 . O 0 . CO CM T— CO r*. T t CO i — CO T t O o> CO CM CO CO LU LY IS CO CO CO CM CM gj CM iri ^/E TI <z z TE CO CO ^/E TI < rr TJ 'E LL VEI LO o 0 ) 8 OJ CO 00 CJ 3 < CHANGES 0 CO CO CO CO CM CO CM gj CO RED CYC CHANGES 0 O miss ASU E CHANGES 0 4.40 2.57 2.05 0.43 2.64 0.42 4.89 LU _J > CHANGES 0 CO CO CM CO CO co 2 O cc miss o Q miss r\ S3 miss < > CC cc TEI CO cn co LO rv o co o NOVl DJUS iri CO iss iss CM ss| co CO o CO rv < < CO E E E cc > rv CO in CM OJ o 8 co CM <; o LO o LO rv T t LO 00 o OJ CO "HE ALVEOL AND AN AN MEAN LO CO CO CM CM" rv CM gl CO CM CO CM CO CO CO CO 3 "HE ALVEOL AND AN AN to LO h- 00 to IO o O O 3 OJ 00 "HE ALVEOL AND AN AN iri T t s> CO iri l< CM T t T t CO CO CM CM CM CM CO CO LL rr CO to CO CO (0 1 CAPACITY 0 OVULATOI r— z CD mis mis mis mis mis 1 CAPACITY 0 OVULATOI o CL co in £ CO CM m CO CM to t— in s r>» CM OJ T t rv o s 1 CAPACITY 0 OVULATOI CO K— CO CO rv' CO IV iri to cvi o iri 1 CAPACITY 0 OVULATOI CO CO in CO CM to CO CO CM CM CM CM to CM T t CO 1 CAPACITY 0 OVULATOI R TI TJ mis mis mis mis o • ANJ LU CO O |v 3 N. CM 00 CM CO o to rv co • ANJ > rv in CO LO T— r- cn OJ T t CM 00 o CO • ANJ O o CO O £ in y-i iri iri to iri cvi LO o iri : FU o ES T t CM CO CM CO CM CO CO CM co CM co p co Q CY U z O I CY CY CHAI CO CO CM © 22.28 i— CM 00 o LO OJ USTEC ORY CHAI 34.0 25.0 35.7 35.8 22.28 28.8 28.9 38.1 29.5 33.1 31.1 OS USTEC H 3 DM miss miss miss DJ ovuu 3 DM S3 miss miss miss < ovuu LU 1— 8 rv o CO OJ CM cn in o |v O in o T t rv «_ ovuu CO CO tv rv to o c\i CO CO OJ yZ T t CQ r- ZJ CO CM CO CM CO CO CO CM CM CO CO o co —} CO CO CO _1 UI Q to to to MOG CQ < CO 'E E E MOG o T™ CO T t CO CM CO CO 00 OJ o CM > UJ o o o o o T — CM CM LU < CO CO CO CO CO CO CO CO CO CO CO CO CO Z Q < 2 2 2 2 2 2 2 2 2 2 2 2 2 CQ CQ CO CO CO CO CO CO CO CO CQ CQ CO LU I-O o o o o o o o o o O O o 2 CO I i s l i 51 > i l l 111 If 122 5 CO CO C O CO CM CO cd eo CO CO CD CO TJ c o s L O CO 1 2 3 LU _ l ' o >-o >-cc o I-3 > O z < z < Q z < > cc o 3 > o Z < Z X z O 0_ r -co LU I-LU > < Q UJ CC 3 CO < UJ CO LU CD Z < X o < cc r -Z LU o z o o z CO o - J CD O 2 UJ < X CO JEj CO cj CO C O CO CM C O 1 2 4 LU I o >-o > cc o > o z < < o z < >-cc o > o 5 CO o CL t-co LU I -UJ > < o UJ cc ZD CO < LU o tt cc I-z LU o z o o z CQ 2 o 0 LU < 1 >-X o C Q cc < o status I AN | 1 AO! I NV/AO ! AO I AO g I AO I AO AN | AO AO z 0.66 0.97 0.98I 0.88 1.08 0.79 0.55 1.00 0.45 0.88 0.77 0.82 MEA 0.65 0.90 1.00 1.50 0.75 0.40 0.90 0.25 0.95 0.81 0.36 WARY CD miss miss WARY 0.55 1.00 0.85 1.00 0.80 0.50 0.90 0.55 0.90 0.80 0.79 0.19 SUMI TJ miss COHB 0.70 1.20 1.10 0.90 0.90 0.85 0.75 1.15 0.50 0.65 1.00 0.88 0.22 COHB O 0.75 0.80 0.85 0.75 0.90 0.75 miss 1.05 0.50 1.00 0.50 0.79 0.18 JO miss ICBMS00 1 CBMS01 CBMS03 CBMS04 ICBMS08 iCBMS12 ICBMS16 ICBMS18 ICBMS19 ICBMS20 ICBMS21 MEAN STDEV in i 1 j?-8 1 2 5 CO LU - J o >-o >-cc o > o Q z < CO LU _ l o >-o < u. o LU _ l o >-o _ l < 3 CC I-co z LU LU X r -cc LU > o CO LU o z < x o LU o > LU o Q LU O < CL CD O CD O s CO CO 55 s I i I f ° 11 6* E S I S I •a 5 1' f "8 1 2 6 I I 8 8- 8 co 1 2 7 CO CO 03 co CO CO in co LU I D O 1 I f ?3 1 2 8 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0077155/manifest

Comment

Related Items