UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of prelatent and latent iron deficiency on physical work capacity Newhouse, Ian Joseph 1987

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

Item Metadata


831-UBC_1987_A1 N48_5.pdf [ 6.32MB ]
JSON: 831-1.0076963.json
JSON-LD: 831-1.0076963-ld.json
RDF/XML (Pretty): 831-1.0076963-rdf.xml
RDF/JSON: 831-1.0076963-rdf.json
Turtle: 831-1.0076963-turtle.txt
N-Triples: 831-1.0076963-rdf-ntriples.txt
Original Record: 831-1.0076963-source.json
Full Text

Full Text

THE EFFECTS OF PRELATENT AND LATENT IRON DEFICIENCY ON PHYSICAL WORK CAPACITY By IAN JOSEPH NEWHOUSE B.P.E., The University of Alberta, 1980 M.Sc, The University of Alberta, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES INTERDISCIPLINARY STUDIES (PHYSICAL EDUCATION/MEDICINE/ZOOLOGY) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1987 ©Ian Joseph Newhouse, 1987 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 The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 D E - 6 ( 3 / 8 1 ) ABSTRACT In order to examine the effects of prelatent/latent iron deficiency on physical work capacity and selected muscle enzyme activities, forty female subjects were studied before and after eight weeks of supplementation with either oral iron or a matching placebo. Initially, female volunteers engaged in regular endurance running were screened for iron deficiency by blood analysis (serum ferritin and hemoglobin). Forty non-anemic subjects with deficient iron stores underwent physiological and anthropometric tests to obtain a comprehensive profile. The specific physical work capacity tests were alactic and lactacid power on the Wingate cycle ergometer test, lactacid capacity on the anaerobic speed test, anaerobic (ventilatory) threshold using gas exchange variables, V 0 2 max. and the max. treadmill velocity during the v"02 max. test. Muscle biopsy samples pre-, and post- treatment were assayed for citrate synthase and alpha-glycerophosphate dehydrogenase activity. Treatment was oral iron supplementation (320 mg ferrous sulfate = 100 mg elemental iron taken as SLOW-Fe® twice a day) or a matching placebo. The subjects were randomly assigned to one of the treatment groups and a double-blind method of administration of the supplements was used. ii It was hypothesized that work capacity would be enhanced following oral iron supplementation, possibly due to the repletion of iron containing oxidative enzymes important in energy production. Results could not strongly support this hypothesis with the difference between the two groups on the work capacity and enzyme activity variables being statistically nonsignificant. Serum ferritin values rose from a mean of 12.4+4.5 to 37.7+19.7 n g m l " 1 for the experimental group and 12.2±4.3 to 17.2±8.9 for the controls; (p=0.0025). Hemoglobin levels remained fairly constant for both treatment groups; 13.4±0.6 to 13.5±0.5 gd l " 1 (experimental), and 13.0±0.6 to 13.1+0.5 (control); (p=0.6). Pre to post values on the work capacity variables, experimental vs control respectively were: Alactic power, 8.8 to 8.4 watts-kg"1 body wt. vs 8.4 to 8.2; lactacid capacity, 6.9 to 6.9 watts-kg"1 body wt. vs 7.0 to 6.0; anaerobic speed test, 41.3 to 45.1 seconds vs 43.7 to 44.8; anaerobic threshold, 7.4 to 7.5 mileshour"1 vs 7.2 to 7.2; V 0 2 max, 51.3 to 52.7 ml-kg" 1min" 1 vs 50.6 to 50.6; max velocity during V 0 2 max, 9.8 to 9.8 mileshour"1 vs 9.6 to 9.5. Except for alactic power, the change in work capacity favored the iron treated group. Noting this trend, further study may be warranted. Prelatent/latent iron deficiency appeared not to depress the activities of the two enzymes measured. Cytoplasmic alpha-glycerophosphate dehydrogenase activity rose from 0.066 to 0.085 units for the experimental group (p=.58) vs .058 to .066 for the control group and citrate synthase activity changed from 0.047 to 0.048 (experimental) vs 0.039 to 0.042 (control). iii It can be concluded that eight weeks of iron supplementation to prelatent/latent iron deficient, physically active females does not significantly enhance work capacity nor the activity of 2 oxidative muscle enzymes (citrate synthase and cytoplasmic alpha-glycerophosphate dehydrogenase). Within the limitations of this study the presence of a serum ferritin below 20 ng-ml"1 does not pose a significant handicap to anaerobic or aerobic capacity. iv TABLE OF CONTENTS Page Abstract i i List of Tables v i i List of Figures v i i i Acknowledgements ix Introduction 1 Methods 5 Results 19 Discussion 35 References 49 Appendix A. Review of Literature 65 Overview of Iron Metabolism 65 Negative Iron Balance in Athletes 69 Iron Deficiency With Anemia 81 Iron Deficiency Without Anemia 85 B. Definitions 99 C. Limitations 104 D. Delimitations 106 v Page E. Subject Consent Form 107 F. Muscle Biopsy Consent Form 110 G. Dietary Analysis 111 H. Schedules of Examinations and Tests 11 6 I. Iron Status Raw Data 117 J. Work Capacity Raw Data 11 8 K. Enzyme Activity Raw Data 11 9 L. Descriptive Statistics of Placebo Treated Subjects who became Anemic 1 20 vi LIST OF TABLES Page Table I. Descriptive Statistics of Iron Status 22 Table II. Descriptive Statistics of Work Capacity 24 Table III. MANOVA on Iron Status 27 Table IV. MANOVA on Work Capacity 28 Table V. Comparison between the Latent Iron Deficient Females within the Iron Treated Group and the Placebo Group on the Anaerobic Speed Test 30 vii List of Figures Page Figure 1. Distribution of Serum Ferritin Levels among the Females initially Screened for Inclusion in the Study 19 Figure 2. Distribution of Hemoglobin Levels among the Females Initially Screened for Inclusion in the Study 20 Figure 3. Iron vs Placebo Summary 26 viii Acknowledgements The efforts, advice and expertise of a large number of people have made this thesis possible. I would especially like to thank CIBA-Geigy pharmaceuticals who funded this research and helped formulate the proposal. The commitment required from the subjects was considerable. I am deeply indebted to these volunteers for their time, exhaustive physical efforts and muscle biopsy specimens donated to the cause of the study. The guidance provided by Dr. Doug Clement and my research supervisor, Dr. Don McKenzie was invaluable, as was the input of each of my supervisory committee members in their own areas of expertise. I am also grateful to the technical assistance provided by Dr. Gord Matheson, Dr. Tom Momson, Lynne Sawchuk, Lois Culling and Dusan Benicky. ix 1 INTRODUCTION Iron deficiency is the most common form of nutritional deficiency in many populations (Siimes et al. , 1980). Although it reaches its greatest prevalence and severity in developing countries, it is also frequently encountered in affluent societies. According to the Nutrition Canada Survey of 1973, 76% of Canadian women had iron intakes below the recommendation for the general population. Recent literature suggests that regular exercisers may be at increased risk of iron deficiency (Clement and Asmundson, 1982; Dickson et al., 1982; Ehn et al., 1980; Wishnitzer et al., 1983). Based on serum ferritin levels, which are believed to accurately reflect the size of the body iron stores (Jacobs, 1977), it was observed that 82% of female, elite Canadian distance runners were iron deficient (Clement and Asmundson, 1982). Another report involving long distance runners found that despite normal hemoglobin and serum iron values, the bone marrow showed either an absence or only traces of iron (Ehn et al. , 1980). Several other investigators have confirmed this surprisingly high incidence of iron deficiency in active persons, particularly endurance runners (Dickson et al., 1982; Nickerson and Tripp, 1983; Wishnitzer et al., 1983). As women are already at increased risk due to the superimposed requirements related to menstruation, the possibility of an increased iron demand associated with exercise is of particular concern to those engaged in regular physical activity. As a result, a variety of supplementation 2 regimens and monitoring schedules are recommended to ensure adequate iron status (Bothwell et al., 1979) These intervention programmes, aimed at reducing the prevalence of iron deficiency, are predicated on the assumption that nutritional iron deficiency does indeed lead to significant disability. In the case of iron deficiency with anemia, there is no doubt that the condition is undesirable. Several studies in rats and in humans (Anderson and Barkve, 1970; Edgerton et al., 1972; Viteri and Torun, 1974) have shown a significant correlation between the hemoglobin concentration and work capacity, with some results indicating that even minor decrements affect performance. This decreased physical work capacity in anemic subjects was presumably due to the reduction in oxygen delivery to the musculature (Anderson and Barkve, 1970). However, more recently, the effects of iron deficiency without anemia on physical work capacity have been examined and this condition has been shown to reduce exercise tolerance or blood lactate levels at maximal exercise (Finch et al., 1976; Finch et al., 1979; Schoene et al., 1983). Finch et al's. studies used exchange transfusions to bring the hemoglobin levels of anemic rats back to normal levels. This iron deficient non-anemic state would not replicate the prelatent/latent iron deficiency observed in many athletes for in the rat studies this state was achieved artificially and the prior state of iron deficiency was severe. The studies did show though that despite normal hemoglobin levels, iron deficient rats had a marked impairment of running ability as compared to controls. This 3 defect in physical performance was attributed to reduced concentrations of an iron containing enzyme (mitochondrial alpha-glycerophosphate dehydrogenase: a-GPDH) important for energy metabolism (Finch et al., 1976). Mitochondrial a-GPDH is a component of the electron transport system and is thought to be the limiting enzyme of the alpha-glycerophosphate shuttle for NADH between cytosol and mitochondrion. Furthermore, it appears that when the activity of this enzyme is decreased, there is an accumulation of lactic acid in the muscles and this contributes to the muscle dysfunction (Finch et al., 1979). Schoene et al., (1983) studied the effect of two weeks of iron therapy on exercise performance and exercise induced lactate production in trained, mildly iron deficient women athletes. Exercise performance was unchanged after therapy although blood lactate levels at maximum exercise decreased significantly. The short period of iron repletion may explain the non-significant change in exercise performance. Valberg (1980) noted that oral iron treatment needs to be continued for two months to replenish iron stores after the hemoglobin concentration has reached normal values. The need for investigation involving non-anemic, iron-deficient, human subjects is a pressing one since the results have far-reaching implications in relation to the well-being and productivity of large population groups, particularly exercising females in their reproductive years. 4 Thus the purpose of this investigation was to determine if iron deficiency without anemia limits physical work capacity. The specific aim was to assess the anaerobic/aerobic work capacities of prelatent/latent iron deficient, physically active females before and after eight weeks of oral iron supplementation. Assuming that changes in exercise tolerance may be due to repletion of iron-containing oxidative enzymes important in energy production (Finch et al., 1976, 1980), the activity of cytoplasmic alpha-glycerophosphate dehydrogenase was measured before and after the treatment period to give an estimate of the maximum possible flux through the NADH-glycerophosphate shuttle system. This shuttle system serves two purposes; 1. regeneration of nicotinamide adenine dinucleotide (NAD + ) necessary for the continued operation of the glycolytic cycle and 2. generation of adenosine 5'-triphosphate (ATP) via the oxidation of a-glycerophosphate by oxidative phosphorylation (Finch et al., 1976). Citrate synthase, which was also assayed to assess change in activity, is a regulatory enzyme which catalyzes the first reaction of the Krebs cycle, the condensation of acetly-CoA with oxaloacetate to form citrate (Lehninger, 1982). It was hypothesized that physical work capacity would be enhanced following eight weeks of oral iron supplementation, possibly due to the repletion of iron dependant enzymes important in aerobic metabolism. 5 METHODOLOGY 1. Subjects Healthy, physically active female runners from the greater Vancouver area formed the population from which a sample was drawn. Forty volunteers between the ages of 18 and 40 years were selected after screening 155 females for prelatent/latent iron deficiency (i.e., serum ferritin below 20 ng-ml"1 and hemoglobin level of 12 gdl" 1 or greater). It should be noted that no condition except iron deficiency has been reported to produce a low serum ferritin concentration (Valberg, 1980). The subjects signed informed consent forms (See Appendices E and F) and all procedures were approved by the University of British Columbia Ethics Committee. Criteria for exclusion from the study were occult blood loss in the stool as determined from serial fecal specimen analysis (Hemocult, Smith-Kline Diagnostics, Inc.) or blood loss in urine as determined by color changes on a dip and read chemstrip (Chemstrip, Boehringer Mannheim Canada Ltd.). Other reasons for exclusion were: ingestion of acetylsalicylic acid (aspirin) or medication which may cause acute or chronic blood loss from the gastro- intest inal tract (such as other non-steroidal 6 anti-inflamatory agents, steroids or solid potassium supplements) one week prior to the pre-test or during the treatment period; pregnancy, blood donation or loss of more than 500 ml of blood through injury within one month of the pre-test or during the treatment period; dietary intakes of vitamin C (ascorbic acid) of less than 60 mgday" 1 as determined from three day dietary analysis; taking of a birth control pill coincidental with a low serum folate (below 3.2 nmol'L"1); and finally a fever within 2 weeks of the hematological pre-test. Data from protocol deviators was excluded from the analysis. Protocol deviations included: failure to take at least 75% of the medication (as determined by pill count); failure to return for post-testing within five days of scheduled date; and development of a fever (oral temperature greater than 38.5° C) within two weeks of the post-test. Subjects who failed to complete the full eight weeks of the study for reasons of toxicity, lack of compliance, lack of cooperation or reasons other than treatment failure were considered dropouts. Seven subjects dropped out with the reasons being injury (3), lack of compliance (2), and relocation (2). Patient accession continued until the requisite number of evaluable study subjects were completed. Continuous enrollment ensured that 40 subjects completed the experimental procedures. 7 2. Test Items or Instruments Subjects underwent hematological, anthropometric and physiological tests to obtain a comprehensive profile. Three work capacity tests were chosen to assess the range of highly anaerobic to highly aerobic function. The test items included: a) Hematological data: Hemoglobins were performed on the ELT-8 multiparameter instrument (Ortho Diagnostic systems, Inc., Raritan, New Jersey) by a colorimetric method. Hemoglobin is converted to cyanmethemoglobin by the addition of a reagent containing potassium ferricyanide and sodium cyanide (cyanac). The density of the color produced is directly proportional to the amount of hemoglobin present (Sonnenwirth and Jarett, 1980). The ELT-8 uses a modified cyanmethemoglobin method to measure hemoglobin. Light is transmitted through a cuvette containing a sample of blood diluted with cyanac reagent. The amount of light that passes through the sample is measured by a light sensor. The output of the sensor when measuring the specimen is compared to that of the reagent blank. The difference represents the amount of hemoglobin in the sample. 8 Serum ferritin concentration was determined by the 2 site Quantimune® Ferritin immunoradiometric assay (IRMA) (Bio-Rad Laboratories, Mississauga, Ontario). IRMA is based on principles described by Bothwell et al. (1979) and Addison and Hales (1971). After raising antisera in rabbits against human recrystallized ferritin, an antibody to ferritin is isolated by mixing immune serum with ferritin that has previously been rendered insoluble by allowing it to react with diazocellulose. After incubation, the specific IgG molecules extracted in this way are labelled with 1 2 5 l while still coupled to the insolubile ferritin. The 1 2 5 l labelled antibodies are then eluted by acidification for use in the routine assay. In the routine assay ferritin standards are incubated with the unknown serum and 1 2 5 l - label led antiferritin IgG. A ferritin immunoadsorbent (ie. Ferritin Immunobeads®) is added which binds and insolubilizes any 1 2 5 l - label led antibody that has not reacted with ferritin and enables it to be removed by centrifugation. Supernatant radioactivity is proportional to the amount of ferritin in the sample being tested. Based on repeated measurements of quality controlled sera in 20 consecutive assays, the between assay variation is about 8 to 10 % (Miles et al., 1974). Total iron (serum iron) measurement is based on the principle that in an acidic medium, transferrin bound iron dissociates into free ferrous and ferric ions. Ascorbic acid is then used to reduce the ferric iron to the ferrous state. Ferrozine® (3-(-pyridyl)-5,6-bis 9 (4-phenylsulfonic acid)-1,2,4 triazine, disodium salt) reacts with ferrous iron to form a magenta colored complex which absorbs at 560 nm. The absorbance is directly proportional to the amount- of iron in the serum (Sonnenwirth and Jarett, 1980) Reagents for both total iron and unsaturated iron binding capacity (UIBC) were obtained from Diagnostic Chemicals Ltd., Charlottetown, P.E.I., Canada. Performance characteristics as reported by the kit manufacturer list the coefficient of variation as between 1.8 and 0.6% for total iron and between 1.4 and 3.7% for UIBC. Unsaturated iron binding capacity (UIBC) is determined as follows: A known ferrous iron standard containing 70 u.mol-L"1 (391 jigdl"1) incubated with the serum at a pH of 7.9 saturates the available binding sites on serum transferrin. The unbound excess iron is then complexed with ferrozine and the UIBC is determined. The UIBC is equal to the total iron added less the excess iron. Total iron binding capacity (TIBC) is equal to serum iron plus UIBC. % Transferrin saturation is arrived at mathematically by dividing total iron by TIBC. All hematological work was done at the University of British Columbia Health Sciences Centre Hospital. Subjects reported to the lab in the morning having avoided physical activity in the preceding 24 hours. 1 0 b) Anthropometric data: Percent body fat (densitometry), height (Holtain Ltd), and unclothed weight (Detecto Scales) were recorded. This data was recorded for descriptive rather than statistical purposes. c) Physiological tests: The research facility used for physiologic (and anthropometric) testing was the Buchanan Fitness Laboratory located on the University of British Columbia campus. The physiological tests were spread over 2 visits with the Wingate and the anaerobic speed test performed on the first visit ( „ 45 minute recovery between the two tests) and the incremental treadmill test performed on the second visit. The Wingate anaerobic power test measured alactic anaerobic capacity and lactacid capacity on the cycle ergometer. A Monark bicycle ergometer equipped with toe clips and an electrically triggered counter were used in this test. Subjects were instructed to pedal as fast as possible for the 30 second test duration. Within 2 to 3 seconds of fast pedalling, the resistance was adjusted to 75 g-kg"1 body weight. At that instant the timer and counter were activated. Vocal encouragement was given. The number of pedal revolutions was recorded every 5 seconds. 11 Alactic power corresponded to the maximal power averaged over a 5 second period (Watts-kg" 1 body wt.). Lactacid power corresponded to the average power output over 30 seconds (Watts -kg" 1 body wt.),(Bouchard et al., 1982). Test-retest reliability is reported to be 0.90 to 0.93 for a 1 or 2 week interval (Bouchard et al., 1982). The Anaerobic speed test on the treadmill measured lactacid capacity and was usually of a longer duration than the Wingate test (ie. 30-60 seconds). As most subjects were runners, this test was more sport specific. A Quinton 24-72 treadmill set at 7 miles-hour"1 and 20% grade was used. Subjects were instructed to run as long as possible with results recorded as time to exhaustion. Reliability is 0,76 to 0.91 (Bouchard et al., 1982) The progressive workload treadmill test assessed both the anaerobic/ventilatory threshold using expired gas analysis and vT^max as the workload increased to the point of exhaustion. The final treadmill velocity achieved was also noted as a work capacity variable. Subjects warmed up on the treadmill for at least 5 minutes. The initial treadmill speed was 5.0 miles-hour"1 and increased by 0.5 miles-hour"1 every minute; the treadmill grade remained at 0% for the duration. Heart rate was monitored by direct chest lead EKG (Burdick EK/5/A electrocardiograph). Expired gases were sampled and analyzed by a Beckman Metabolic Measurement Cart interfaced to a Hewlett Packard 3052 A Data 1 2 Aquisition System. Two major criteria were used to determine the anaerobic threshold: 1) a non-linear increase in excess C 0 2 (ie. V C 0 2 - (RQ X V 0 2 ) ) and 2) a non-linear increase in ventilation (ie. Ve/V0 2 ) . Ventilatory and gas exchange variables were tabulated every 15 seconds for the duration of the treadmill test allowing for plots of these variables against the speed of the treadmill. Results were expressed as the velocity of the treadmill (mph) at the anaerobic threshold. Two independent evaluators determined the anaerobic threshold. A third evaluator arbitrated irreconcilible differences of opinion. A test-retest correlation for this variable was found to be 0.91 (Davis et al., 1979). V 0 2 m a x was determined by the mean of the four highest consecutive 15 second values. Thoden et al. (1982) report a test-retest correlation of 0.95. d) Muscle Biopsies This was an optional procedure for all subjects; 17 of 40 volunteered and utilized a separate consent form (see Appendix F). Information gained from this aided in understanding possible biochemical effects of prelatent and latent iron deficiency. 1 3 Two biopsy samples were thus taken (one pre- and one post-treatment). The biopsies were taken from the lateral portion of the right quadriceps muscle (vastus lateralis) about 20 cm above the knee. First a small area was anesthetized with local anesthetic (2% xylocaine). An incision was made large enough to allow the biopsy needle to be put into the muscle. Approximately 50 mg of muscle was excised with the needle and was immediately frozen in liquid nitrogen and stored at -80°C for further analysis. The biopsy excision was conducted at the University of British Columbia Health Science Centre Hospital and the assays were performed in a laboratory of the Zoology department at the University of British Columbia. Assays were performed for citrate synthase activity and alpha-glycerophosphate dehydrogenase (a-GPDH) activity. After thawing, samples were homogenized with Polytron homogenizer (2 bursts of 10 seconds) in 10 volumes imidazole buffer (50 mM Imidazole, adjusted with HCI to pH 7.8 at room temperature). Homogenates were centrifuged briefly (60 seconds at 13,000 x g) and aliquots of the supernatants were used for the enzyme assays without further dilution. 1 4 Citrate synthase (E.C., citrate oxaloacetate-lyase [CoA-acetylating]) was assayed at 412 nm, utilizing the reduction of DTNB (5,5'-dithiobis- (2-nitrobenzoic acid) by the HSCoA liberated as the indicator reaction. Final concentrations of the reactants and cofactors in the reaction medium were: 50 mM Tris-HCl (pH 8.1), 0.1 mM DTNB, 0.25 mM acetyl-Coenzyme A, 0.5 mM oxaloacetate (omitted for control). The extinction coefficient of DTNB at the chosen pH and wavelength is 13.6 cm 2-|iMol" 1. Alpha-glycerophosphate dehydrogenase (E.C., sn-glycerol 3-phosphate: NAD 2-oxidoreductase) was determined at 340 nm, following the oxidation of NADH upon the addition of dihydroxyacetonephosphate (DHAP). The reaction medium consisted of: 50 mM Imidazole-HCI (pH 7.8), 0.15 mM NADH and the dimonocyclohexylamine salt of DHAP at a final concentration of 1 mM (omitted for control). The extinction coefficient of NADH at the chosen wavelength is 6.22 cm 2-u.Mol" 1. Final reaction volumes were 1.0 ml and reaction rates were followed for 1 - 5 min. It was ascertained that rates were linear with time and with volume of homogenate used. Determinations were done using a Unicam SP6-500 recording spectrophotometer fitted with a thermostatted cell holder and a Soltec single channel recorder. Cells were maintained at 25°C; all reactions were done in duplicates. One unit is defined as the amount of substratye converted to product per minute at 25°C in the presence of 1 5 saturating concentrations of substrates (Vmax). All steps in the preparation of extracts were performed at 0 - 4°C. Enzyme activities were measured within 2 hours of homogenization. Protein concentration in the supernatants of the muscle homogenates was determined following the procedure outlined by Bradford (1976). The technique employs Coomassie Blue as the color reagent and 5 JLLI of homogenate were used in a final assay volume of 1 ml. Defatted bovine serum albumin was used as a standard. After initial testing, individuals conforming to entry criteria were randomly assigned to one of two treatment groups. The two treatment groups were: a) Experimental: The experimental treatment consisted of oral iron supplementation (320 mg ferrous sulfate = 100 mg elemental iron) taken as SLOW-Fe® 1 tablet twice a day, as tolerated. This tablet appeared different from SLOW-Fe® that is available commercially. b) Control: The control group took a placebo which appeared identical to the active agent and was taken 1 tablet twice a day as 1 6 tolerated. Ingredients: Cornstarch, lactose, gelatin, talc, stearic acid, magnesium stearate, green #3 dye, silicon dioxide, edible shelac, hydroxypropylmethyl cellulose, PEG 4000. A double blind method of administration and random assignment to treatment groups was used. This involved computer generated randomization with each enrolled participant assigned to the next consecutive number. Treatment continued for eight weeks with a follow-up visit at four weeks and retesting upon completion of treatment. During the treatment period subjects monitored training (type, duration, intensity) and menstrual blood loss (estimated from number of tampons/pads used each cycle). Subjects were instructed to make no changes in their training programme or their diet during the study. A three day dietary analysis prior to both the pre-test and post-test aided in monitoring diet. (See Appendix G) The post-testing procedures were identical to the initial testing. See Appendix H for schedules of examinations and tests. 3. Data Analysis The experimental design is a 2 x 2 (treatment x time) factorial experiment with repeated measures on the second factor as shown on the following page. 1 7 S 1 EXPERIMENTAL j S 20 n=20 n=20 S 21 CONTROL I S 40 n=20 n=20 PRE-TEST POST-TEST (8 WEEKS) TIME The independent variable was oral iron supplementation. There were 11 dependent variables with two domains (iron storage and work capacity) and thus two multivariate analyses of variance were used. This provided experiment-wise control over the type one error rate. A significance probability level of 0.05 was chosen. The dependent variables were: IRON STATUS: a) Serum ferritin levels b) Hemoglobin levels c) Total iron d) Unsaturated iron binding capacity (UIBC) e) % Transferrin saturation 1 8 WORK CAPACITY: f) Alactic capacity on Wingate cycle ergometer test g) Lactacid capacity on Wingate cycle ergometer test h) Lactacid capacity on anaerobic speed test i) Anaerobic threshold on treadmill j) V 0 2 max on treadmill k) Max treadmill velocity during V0 2 max test. A cannonical correlation was used to obtain a measure of the linear association between improvement scores on the domains of iron storage (serum ferritin and hemoglobin) and work capacity (variables f-k above). The loadings associated with each variable in the cannonical variate indicate the relative contribution that each variable within the set makes toward the relationship between the two sets (Schutz et al., 1983). The statistical programs used were from BMDP Statistical Software, Inc. , Los Angeles. 19 RESULTS One hundred and fifty-five physically active women (age 18-40) underwent blood testing to select a sample population who fulfilled the criteria of prelatent/latent iron deficiency. (Figures 1 and 2) The slightly smaller number of subjects shown in these figures is due to subjects who obtained a serum ferritin determination but no hemoglobin value or vice versa. Number Of Subjects (n= 152) 40 30 25 I I I I I I I I I l l l l l l I 50 | 70 I 90 | 1 1 0 | 130| 150| 170| 190 40 60 80 100 120 140 160 180 Mean (29.9) Serum Ferritin (ngml -1) FIGURE 1: Distribution of serum ferritin levels among the females initially screened for inclusion in the study. 20 The inclusion criteria for serum ferritin levels was 20 ngml" 1 or less. Sixty-one of the subjects (40%) fit this classification while 94 females (62%) had serum ferritin values below 30 ngml" 1 . Two subjects were above the lab's normal range of 10 to 160 ngml" 1 while 20 subjects were below this range. FIGURE 2: Distribution of hemoglobin levels among the females initially screened for inclusion in the study. 21 Hemoglobin values among the females were centered around the mean of 13.0 g d l " 1 . Thirteen females (8.5%) could be classified as anemic (Hb below 12.0 gdl" 1). Twelve of these women had low iron stores as reflected by serum ferritin values below 30 ngml" 1 . On the basis of hemoglobin and ferritin measurements, 47 females qualified to continue with the study. Forty of the selected females completed the eight week study, 19 of whom received iron supplements and 21 of whom received matching placebos. Pre and post measurements of iron status (Table I) and work capacity (Table II) were taken. 22 TABLE I Variate Serum ferritin (ng-mr1) Hemoglobin (gdl"1) Iron Total (ug-oT1) UIBC (ugdr1) % Saturation (%) Group Iron Placebo Iron Placebo Iron Placebo Iron Placebo Iron Placebo Count 19 18 19 18 13 14 13 14 13 14 Pre-Test 12.4±4.5 12.2±4.3 13.4±0.6 13.0±0.6 79.8±48.7 84.1143.0 251.8±81.0 252.4161.5 25.1112.8 25.6112.9 Post-Test 37.7119.7 17.218.9 13.510.5 13.110.5 98.5140.1 70.5145.7 195.7156.1 231.9143.9 34.1114.4 21.4113.0 Table I: Descriptive statistics of iron status, iron group vs placebo group (count, mean and l S.D.)-23 Serum ferritin values rose from a mean of 12.4 to 37.7 for the iron supplemented group, but also noteworthy is the ±19.7 standard deviation during the post-test. Hemoglobin values remained constant for both treatment groups. Due to missing pre-test values, (the blood samples were discarded) the count for the iron profile measurements was decreased to 13 (iron) and 14 (placebo). Alactic power dropped slightly for both treatment groups while only the placebo group showed a performance decrement on the lactacid capacity variable. The iron treated group improved their performance on the anaerobic speed test by almost 4 seconds compared to an improvement of just over 1 second for the controls. The incremental speed test measurements of anaerobic threshold and max. treadmill velocity remained almost constant for both treatment groups. On the V 0 2 max variable a modest improvement was posted (51.3 to 52.7 mlkg" 1min* 1) for the iron supplemented group, with no mean change noted for the placebo treated group. Tables of raw data for iron status, physical work capacity and enzyme activity can be found in appendices I, J , and K. 24 TABLE II Variate Alactic Power (Watts-kg - 1 body wt) Lactacid Power (Watts-kg"1 body wt) Anaerobic Speed Test (Seconds) Anaerobic Threshold (mileshr 1) V 0 2 Max (mlkg" 1min" 1) Max Treadmill Velocity (during V0 2max) (mi leshr - 1 ) Group Count Pre-Test Post-Test Iron 19 Placebo 18 Iron 19 Placebo 18 Iron 19 Placebo 18 Iron Placebo Iron Placebo Iron Placebo 19 18 19 18 19 18 8.8±1.0 8.411.0 6.910.4 7.010.7 41.3115.2 43.7113.5 7.411.0 7.2+0.7 51.315.7 50.614.9 9.811.0 9.610.7 8.411.1 8.211.4 6.910.9 6.010.9 45.1116.9 44.8112.9 7.510.9 7.210.7 52.713.8 50.615.5 9.810.8 9.510.7 Table II: Descriptive statistics of work capacity, Iron group vs. Placebo group. 25 Change scores for each dependent variable were calculated and then standardized (Figure 3 ) by dividing mean improvement by the average standard deviation of change. This is essentially an effect size measurement which permits valid comparisons between the two groups. The largest positive change was the iron groups' serum ferritin levels. Unsaturated iron binding capacity (UIBC) and % transferrin saturation also showed favorable changes for the iron group. Smaller changes appeared with the work capacity variables. For both treatment groups, there was a performance decrement on the Wingate bike test measurements of alactic and lactacid power. Anaerobic speed test scores exhibited the largest improvement with this being the only work capacity measurement on which the control group improved. Except for alactic power, the change in work capacity over the treatment period appeared to favor the iron group. 26 2.0 1.5 Standard Change Score 1.0 -0.5 • Iron Group m Placebo Group 0.5 0.0 — i l — = 1 — H — 1 Iron _| — — Max JFer. Hb. Total UIBC %Sat AlacP LacP AST AT 702 Vel. Dependent Variables Figure 3: Iron vs Placebo summary. (Fer.-serum ferritin, Hb.-hemoglobin, UIBC=unsaturated iron binding capacity, %Sat.=%transferrin saturation, Alac.P=alactic power, Lac.P= lactacid power, AST-anaerobic speed test, AT-anaerobic threshold, V02=V02Max, Max Vel.-maximal velocity of treadmill during V02Max test) Standard change score = mean change score divided by standard deviation of change. Two Manova's (Multivariate analysis of variance) were conducted to test the hypothesis that a difference exists between the two treatment groups. In this case the dependent variables are difference scores, pre to post. Table III presents the Manova on change scores with the iron status variables while Table IV presents the Manova on the work capacity variables. A probability (p) value of 0.05 or less has been chosen to represent significance. 27 TABLE 111 Effect Overall: (ie. pre-post) Variate Grand Mean - A l l -Change in Serum Ferritin Change in Hemoglobin Change in Iron Total Change in UIBC Change in % Saturation - A l l -Change in Serum Ferritin Change in Hemoglobin Change in Iron Total Change in UIBC Change in % Saturation Table III: MANOVA on Iron Status, highlighted with an asterisk (*). Group: (ie. GXT) Statistic T-squared F ratio F ratio F ratio F ratio F ratio T-squared F ratio F ratio F ratio F ratio F ratio P-value 0.0013* 0.0000* 0.9476 0.5962 0.0111* 0.2417 0.0442* 0.0025* 0.6003 0.3482 0.0701 0.1399 Significant P-values are Serum ferritin levels and UIBC levels were significantly altered for both groups combined (overall effect). From the group effect, though, the only significant difference between iron treated and placebo treated groups was the change in serum ferritin levels. 28 TABLE IV Overall Variate Statistic p-value Overall: Grand Mean (ie. pre- - A l l T-Squared 0.2409 post) Change in Alactic power F ratio 0.1401 Change in Lactacid power F ratio 0.4265 Change in Anaerobic Spd Test F ratio 0.0239 Change in Anaerobic Threshold F ratio 0.8375 Change in V 0 2 Max F ratio 0.3192 Change in Max Velocity F ratio 0.7068 Group: - A l l T-Squared 0.7691 (ie. GxT) Change in Alactic power F ratio 0.6315 Change in Lactacid power F ratio 0.6291 Change in Anaerobic Spd Test F ratio 0.2135 Change in Anaerobic Threshold F ratio 0.5083 Change in V 0 2 Max F ratio 0.3116 Change in Max Velocity F ratio 0.7086 Table IV: MANOVA on Work Capacity 29 There were no significant p-values among the work capacity variables. Although the Anaerobic Speed Test (AST) scores (for both groups combined) improved with an associated p-value of 0.024, lack of a significant T-Squared p-value precludes attributing a significant change in any one variable. A post hoc ANOVA was added to the analysis to determine if a subgroup of latent iron deficient females within the iron treated group significantly improved their work capacity when compared to the placebo group (Table V). This subgroup consisted of only 5 females with the added criterion being a % transferrin saturation less than 20% (the normal range is 20 to 55%). UIBC (321±44u.gd|-1) and total iron (58±11 jj.g d! 1 ) values were generally supportive of this criterion, although most total iron values could still be considered normal (40 to 130u,gdl"1). The AST performance variable was used because the previous MANOVA (Table IV) showed it to be the work capacity measurement closest to significance when comparing treatments. 30 TABLE V Variate Group Count Anaerobic Speed Test Iron 5 (seconds) Placebo 18 P re-test Post-test P-value 34.3±5.6 38.1 ±8.0 0.38 43.6±13.5 44.8±12.9 Table V: Comparison between the latent iron deficient females within the iron treated group and the placebo group on the anaerobic speed test variable. A significant difference between the groups could not be found, (p-value = 0.38). This may be due to the small sample and the large standard deviation, (Table V). The group means of AST measurements showed an improvement from 34.3 seconds to 38.1 seconds for the iron treated latent iron deficient females and a more modest improvement from 43.6 seconds to 44.8 seconds for the placebo controls. 31 To further define the nature and strength of the relationship between the domain of iron status and work capacity, two cannonical correlations on change scores were performed. One evaluation included all iron status measurements and used a count of 29 while the other evaluation excluded measurements of total iron, UIBC, and % transferrin saturation so as to increase the count to 40. (Note; there were some missing values in these supplementary iron status measurements.) In either case the Chi-Squared test was not significant. This indicates that the two sets of variables are independent. Of primary interest from the three day dietary records, (Appendix G), was iron intake. The mean intake of subjects enrolled was 11.8 ± 3.7 mg/day during the pre-test period and 11.0 ± 3.0 mg/day during the post test period. Correlations were computed amongst all variables. Selected correlations are presented in table VI. Table VI Selected Correlations n Serum ferritin with total iron(pre) 30 (post) 39 " UIBC (pre) 30 " UIBC (post) 39 " % trans, satn.(pre) 30 (post) 39 Alactic with lactacid power 40 AST " . . « 4 0 AST " % body fat 40 Change scores Serum ferritin with total iron 30 " UIBC 30 " % trans, satn. 30 " AST 40 " AT 40 " V0 2 Max 40 Control group only Serum ferritin pre with post 1 8 Hemoglobin " " 18 Iron total 13 UIBC 13 % Trans. Satn. " " 13 Alactic power " " 18 Lactacid 18 AST 18 AT 18 V0 2 max. 18 Max. Workload 18 r p (probability) .03 > .20 .26 < .20 .33 <.10 .47 < .02 .07 > .20 .36 <.10 .80 < .001 .61 < .001 -.67 < .001 .02 > .20 .23 > .20 .09 > .20 .25 < .20 .15 > .20 -.07 > .20 .24 > .20 .43 <.10 .18 > .20 .20 > .20 .19 > .20 .39 < .20 .69 < .01 .89 <.001 .90 <.001 .79 <.001 .81 <.001 Table VI: Selected correlations 33 Seventeen of the forty enrolled subjects volunteered to have muscle biopsy samples taken pre and post treatment from the lateral portion of the quadricep (vastus lateralis). The extremely small size of two of the biopsy specimens (ie. <20 mg), prevented analysis of all 17 subjects, while the small size of the samples in general (between 20 to 70 mg) allowed measurement of only 2 relevant enzymes. The enzymes chosen were cytoplasmic alpha-glycerophosphate dehydrogenase (a-GPDH) and citrate synthase. Table VII Variate Group Count Pre-test Post-test P-value a-GPDH Iron 1 0 0.066±.024 0.085±.021 0.58 (unitsg 1 protein) Placebo 5 0.058+.029 0.066±.011 Citrate synthase Iron 1 0 0.047+.011 0.048±.010 0.76 (unitsg 1 protein) Placebo 5 0.039±.017 0.042±.012 Table VII: Descriptive statistics of enzyme activity, Iron group vs. placebo group.(1 unit = Ijimole of product produced under saturating conditions in 1 minute at 25° C) There was not a significant difference between the two treatments with regards to either enzyme (table VII). Probability values obtained from ANOVAs were .58 for the change in a-GPDH and .76 for the change in citrate synthase. 34 Finally, it can be noted that three women who were on the placebo treatment became anemic (hemoglobin levels below 12.0 gdl" 1). The data from these subjects was excluded from the MANOVA's but are presented (Appendix L) for descriptive purposes. 35 Discussion Eight weeks of iron supplementation ((a daily dose of 100 mg elemental iron (320 mg FeSo4 taken as Slow Fe® twice a day, as tolerated)) to prelatent/latent iron deficient physically active females did not significantly enhance work capacity nor the activity of 2 oxidative muscle enzymes (citrate synthase and a- GPDH). Within the limitations of this study the presence of a serum ferritin below 20 ng-ml"1 does not pose a significant handicap to anaerobic or aerobic capacity. Serum Ferritin From the 155 females involved in the initial screening, it was found that the mean serum ferritin was 29.9 ngml" 1 which is similar to that reported for Canadian female distance runners; 27.9 ngml" 1 (Clement and Asmundson, 1982) and U.S. cross country skiers; 32.8 ngml"1 (Haymes et al., 1986). A random sampling of 95 females (age 20 - 39) across Canada found a mean serum ferritin of 23 ngml" 1 , but in 44 controls who were selected on the basis of normal hemoglobin the mean ferritin value was 35 ngml" 1 Valberg et al., 1976). The distribution about the mean in the present study is also similar to these previous studies in that the majority of the females had serum ferritins below 30 ngml"1 (ie. 60% < 30 n g m l " 1 ) . Clement and Asmundson (1982) found an even greater skewedness to the distribution. Using a criterion of 25 ngml" 1 , 82% of the 36 females in their study were prelatent iron deficient. The reason for this difference may be training volume. The subjects in the present study were mostly recreational runners while Clement and Asmundson studied elite level runners. Magnusson et al. (1984) question the validity of low serum ferritin levels in distance runners. They observed normal erythropoiesis and low serum ferritin concentrations in a group of middle and long distance runners and it thus appeared that a lack of iron had not limited erythropoisis. Magnusson et al. concluded that a true iron deficiency did not exist. Instead, increased intravascular hemolysis, as evidenced by the formation of the haptoglobin-hemoglobin complex, causes a shift in red cell catabolism from the reticuloendothelial system to the hepatocytes. The fact that 3 of the 18 placebo treated subjects in the present study became anemic is suggestive that at least in recreational female runners the presence of a low serum ferritin concentration does indicate a true iron deficiency. Inadequate dietary iron intake appears to be a major contributing factor to the prevalency of iron deficiency. Iron intakes of the prelatent/latent iron deficient athletes who completed the 8 week study averaged 11.8 ± 3.7 mg/day during the pretest period and 11.0 ± 3.0 mg/day during the post test period (based on 3 day dietary records; Appendix G). Approximately 80% of the females had intakes below the Canadian recommended intake of 14 mg/day (Health and Welfare Canada, 1983). The Nutrition Canada survey (1975) also showed inadequate iron intakes for the females in the 37 20 - 39 age category; ie. 11.1 mg/day (Health and Welfare Canada, 1975). British Columbians participating in this national survey were even lower in iron intakes with a mean of 10.0 mg/day. These results reinforce the common finding that it is difficult for the typical Western diet to meet the iron demands of the menstruating female (Clement and Sawchuck, 1984). The only significant difference between the two treatment groups was the change in serum ferritin levels (Table III). Both groups started with mean serum ferritins of near 12.3 ngml" 1 . The control group's mean level rose to 17.2 ngml" 1 while the iron group's rose to 37.7 ngml" 1 . Although statistically significant, this rise in serum ferritin is still modest when one considers the normal range extends to 160 ngml" 1 . Schoene et al., (1983), supplemented with a similar dosage (300 mg/day) but for only 2 weeks and found an increase in ferritin levels from 10.0 ± 2.0 to 22.1 ±7.8 n g m l " 1 . In the present study it was hoped that 8 weeks of supplementation would be sufficient to bring the mean ferritin level to greater than 60 n g m l " 1 . Ferritin levels below 64 ngml" 1 may still indicate an iron deficient state (Heinrich et al., 1977). The inability of this study to show statistically significant changes in work capacity was arguably handicapped by the still sub optimal post treatment ferritin levels. Heinrich et al. (1977) correlated iron absorption with serum ferritin concentration. Diagnostic 5 9 F e + 2 absorption appeared to be a more sensitive indicator of depleted iron stores. It was concluded that serum ferritin values up to 64 ngml" 1 could still be suspect of representing prelatent iron deficiency. Exhausted iron stores cannot be definitely 38 excluded as a possibility until serum ferritin concentration rises above this level. Combining the data of Schoene et al. (1983) with the present data it is apparent that replenishment follows an asymptotic curve. A linear extension of Schoene et al's data would result in a mean serum ferritin level of 60 ng-ml" 1 after 8 weeks. Supplementation of prelatent/latent iron deficient females should ideally be continued for perhaps 16 weeks to ensure mean levels reach the 60 to 70 ng-ml"1 range. The large standard deviation of post test serum ferritins for the iron supplemented group in both Schoene et al's (1983) study (22.1 ±7.8 ng-ml"1) and the present research (37.7 ± 19.7 ng-ml"1) is noteworthy. One individual in this study increased their ferritin from 7 to 12 ng-ml"1 while another similarly iron treated subject responded with an increase from 18 to 83 ng-ml" 1. There is thus a great deal of variability in how subjects respond to iron therapy. Differences in absorption are likely the largest contributing factor to this variability. Controls were in place to limit the variability imposed by other factors. These included pill counts at 4 and 8 weeks, checks for blood loss in urine and stool and standardized instructions for when and how the supplements should be taken. The absorptive mechanism of allowing the amount of iron crossing the small intestinal epithelial barrier to be related to internal iron status can be defective in many people (Ehn et al., 1980, Jacobs, 1985). The problem is probably genetic and/or due to interrelationships that iron has with other nutrients. 39 Considering this latter alternative, it is conceivable that multiple mineral deficiencies can occur. Attempts at repleting one mineral may not meet with success but instead could compound the imbalance of the internal system. As a speculative example: The phytates, oxalates and alkali that inhibit iron absorption will also inhibit copper and zinc absorption (Kreg and Murray, 1986). Copper, zinc and iron could be further synchronously depleted due to their similar food sources (ie liver, meat). A vegetarian diet may not only contribute to iron deficiency but to copper and zinc deficiency. Typically, only the iron deficiency is diagnosed and megadoses of iron are prescribed. Absorption of copper and zinc are both inhibited by iron and these two minerals could be further depleted. Copper is essential for hemoglobin synthesis and iron mobilization, while zinc is a component of numerous enzymes. A deficiency in either will likely affect iron metabolism and absorption. Support for this concept can be found in a study on copper deficient swine (Lee et al., 1968). Copper deficient pigs developed a severe hypochromic, microcytic anemia and had low serum iron concentration. In these pigs there was a defect in the transfer of iron from the intestinal mucosa to the blood and from the reticuloendothelial cells to the plasma and also in the incorporation of iron into hemoglobin. The oxidation of F e + 2 by caeruloplasmin, of which copper is a prime constituent, may explain these defects (Jacobs and Worwood, 1974). Although the above argument is speculative it is clear that a broad focus to the diagnosis of iron and other trace mineral deficiencies is necessary. 40 Hemoglobin Hemoglobin levels among the 155 females initially screened averaged 13.0 g d l - 1 (Figure 2). The normal range is defined as 12.0 to 16.0 g d l " 1 . Thirteen females had hemoglobin levels below this normal range and could be classified as anemic. The observation that 12 of the thirteen had ferritin levels below 30 ngml" 1 (13.6 ± 8.1 ng-ml"1) underlines the fact that of the anemias iron deficiency anemia is the most prevalent in this group. The change in hemoglobin values pre to post was negligible for both treatment groups (Tables I and III). This is in agreement with Pate et al. (1979) who found that oral iron supplements when administered to normal, non-anemic women athletes, have no statistically significant effects on hemoglobin levels. Pate et al. did note though a modest improvement from 14.38 to 14.96 gdl" 1 over their treatment period (5 to 9 weeks with 50 mg elemental iron/day). There was thus some question prior to the present study whether changing hemoglobin levels would influence our results. If there was a difference between the treatment groups it would have confused the issue for a rise or fall in hemoglobin levels would affect oxygen carrying capacity and thereby affect work capacity (Haymes, 1973). Edgerton et al. (1972) suggests that the capacity for work is directly related to hemoglobin concentration. To differentiate the effects of iron depletion from hemoglobin levels it was essential that hemoglobin levels remained constant. 41 Total Iron, U.I.B.C., and % Transferrin Saturation These variables were added to the iron status measurements to help discriminate between prelatent and latent iron deficiency. They do not reflect decreasing iron stores until the stores are exhausted. A low serum ferritin (< 20 ngml" 1) coupled with an abnormally low total iron and % transferrin saturation, high UIBC but normal hemoglobin levels would indicate the latent iron deficient stage. It was hypothesized that a subtle difference in work capacity may only become apparent when this subgroup is isolated. Only 5 women of the iron treated group initially had latent iron deficiency. This was unfortunate from a scientific viewpoint in that the small 'n' makes a finding of statistical significance less likely. Such was the case here. From a single post hoc ANOVA on this subgroup vs. the placebo group on the AST performance variable a significant difference could not be found. Based on this datum latent iron deficiency does not affect work capacity. Pakarinen (1980) has criticized the reliability of total iron and transferrin saturation measurements. Evidence supporting this criticism could be found in this study. Correlations were computed between all variables (table VI). Of interest would be whether serum ferritin levels were correlated to the other parameters of iron status. Only UIBC exhibited strong correlations with serum ferritin levels on both the pre and post tests (r = -.33 and -.47 respectively, with associated p values of 42 p<.1 and p<.02). The post test % transferrin saturation correlated well with serum ferritin (r - .36; p<.1) but the pre test didn't, nor did the total iron values. A low correlation means that only a small portion of the variance in one variable is predictable from the variance in the other. In other words, the serum ferritin concentration does not aid in the prediction of the total iron. These results are not totally unexpected. In healthy persons a weak inverse relationship was observed between plasma ferritin concentration and TIBC, but not the serum iron concentration or the transferrin saturation (Cook et al., 1974). Jacobs et al. (1972) found no correlation between serum ferritin concentration and % transferrin saturation in healthy males, although for females the correlation was significant (r= 0.66, p<.001). This sex difference may be explained by the larger number of women with low serum ferritins. In both sexes in Jacob et al's (1972) study, a serum ferritin concentration below 10 ngml" 1 is associated with a transferrin saturation of less than 16%. The same can be said in the present study. On the pretest measures six of the seven subjects with ferritins below 10 ngml" 1 also had a transferrin saturation below 16% (Appendix I). It can be presumed that iron delivery to the plasma pool is almost certain to be impaired when serum ferritin concentration drops below this 10 ngml" 1 level. Above this level the relationship between serum ferritin concentration and transferrin saturation is not well defined. Also of interest would be whether the change in serum ferritin pre to post was correlated to the change in the other iron status parameters. In this case none of the 3 correlations were significant (r < 23; p>.2). In many instances serum ferritin levels changed independantly of changes in total iron, UIBC, and % transferrin saturation. 43 One, or a combination of 3 possible conclusions can be made; 1) Concurring with Pakarinin (1980) total iron and transferrin saturation are not good tests to show depleted iron stores. Only serum ferritin is reliable. 2) Serum ferritin measurements are unreliable. Bannister and Hamilton (1985) performed repeat tests on hemoglobin, total iron, TIBC and serum ferritin and found coefficients of variation which were 0.7%, 1.78%, 1.99%, and 10% respectively. 3) The pattern of depletion/repletion has a high degree of individual variability. All three possibilities likely contribute in that lability of blood measurements is considerable and the two sets of data can behave independantly due to individual patterns of depletion/repletion. Work Capacity None of the work capacity measurements improved significantly (Table IV). The anaerobic speed test elicited the greatest difference between the two groups, but statistically this was not significant. Under the conditions of this study it can be concluded that the prelatent/latent stage of iron deficiency does not pose a significant handicap to either aerobic or anaerobic performance. Supporting this conclusion derived from a MANOVA was the lack of a significant correlation between the change scores for serum ferritin and the work capacity variables (Table VI). 44 Unlike the iron status measurements, the work capacity tests showed good test-retest reliability as evidenced by the placebo groups fairly stable scores and pre to post correlations (table VI). The least reliable measurements were from the Wingate bicycle ergometer test (r = 0.39 for alactic power, r = 0.69 for lactacid power). The other exercise tests had test-retest correlations of over 0.78. It is interesting to speculate on the fact that except for alactic power, the direction of change in the physical work capacity measurements was in favor of the iron supplemented group. To view the results in light of the limitations would suggest that further research is warranted. Iron supplementation, if at all advantageous, would most likely affect oxidative energy production, needed in some degree in exercise tests of longer duration than that of alactic power. The anaerobic speed test showed the greatest differentiation (albeit statistically nonsignificant) between the two treatment groups (p=.21). Although the contribution of the aerobic pathways is small in the approximately 45 second effort of the AST, aerobic capacity is critical to the achieved performance. If, as hypothesized, iron supplementation enhances aerobic capacity, the build up of lactic acid is delayed and running time is extended. Outside the focus of this study it was noted that % body fat correlated highly to AST performance (r=.67, p<.001). Also noteworthy is the large standard deviation (eg. 45.1 ± 16.9) associated with this test. A more homogeneous sample population or adjustment of treadmill speed to narrow the standard deviation may aid in it's discriminating ability. 45 A significant difference between the groups on the V0 2 max variable was not evident (p=.31; table IV). Work by Schoene et al. (1983), Finch et al. (1976, 1979), Davies et al. (1982) and Vellar and Hermanson (1971) would not lead one to expect significant changes in this variable as long as hemoglobin levels stayed within the normal range. In the study by Davies et al. (1982), rats were raised on an iron deficient diet so that hemoglobin levels and muscle mitochondrial content were depressed. Iron was then restored in the rat's diet to assess whether V 0 2 m a x or muscle mitochondrial content better predicted exercise endurance. Hemoglobin levels increased in parallel with V0 2 max, while mitochondrial capacity and running endurance improved in parallel but more slowly. These results suggest that V0 2 max is a function of 0 2 transport whereas endurance is more dependant on the ability of the mitochondria to consume oxygen. This can also be interpretted to mean that the mitochondrial capacity to consume 0 2 is much greater than the ability of the circulation to supply 0 2 . Although V0 2 max was not expected to improve, peripheral changes (ie. increased enzyme activity) had been hypothesized to improve endurance capacity. The anaerobic threshold and maximal treadmill velocity during the V0 2 max test should reflect this improvement but once again the change was not statistically significant. Endurance capacity is not significantly associated with serum ferritin levels. 46 Enzyme Activity Measurement of enzyme activity changes represented the most objective evaluation as to the effects of prelatent/latent iron deficiency in this study. No significant difference could be found in the two enzymes measured (cytoplasmic alpha glycerophosphate dehydrogenase and citrate synthase). The mean activity of a-GPDH increased from .066 to .085 unitsg" 1 protein for the iron group and only .058 to .066 for the placebo group but the associated p value of .58 does not tempt further speculation (see table VI). Citrate synthase activity also showed a non significant change over the treatment period for both groups. It can be concluded that a more severe degree of iron deficiency is necessary to affect cytoplasmic a-GPDH and citrate synthase activity. The present study lends support to the theory that iron dependent enzyme activity drops in concert with a drop in hemoglobin and not before it. This conclusion is in agreement with Celsing et al. (1986), who studied different enzymes but also concluded that it is possible to have intact endurance capacity and maximal enzyme activity despite diminished or depleted iron stores. The NAD-linked form of a-GPDH was assayed instead of the mitochondrial form as assayed by Finch et al., (1976, 1979). Their activities are comparable as they both participate in the NADH-glycerol phosphate shuttle. The fact that work capacity was not significantly affected by the prelatent/latent stage of iron deficiency makes it difficult to comment on 47 the debate over the relative importance of a-GPDH (Finch et al., 1976,1979 and McLane et al., 1981). Conclusions Oral iron supplementation (320 mg ferrous sulfate = 100 mg elemental iron taken as Slow-Fe® twice a day for 8 weeks) was successful in raising serum ferritin (12.4 ± 4.5 to 37.7 ± 19.7 ngml" 1) in prelatent/latent iron deficient females but was not associated with significant improvements in work capacity, nor the activity of 2 muscle enzymes, cytoplasmic a-GPDH and citrate synthase. 48 Recommendations for future study 1. Treatment should induce higher serum ferritin response. Consider longer treatments and/or higher dosages. 2. Improve the discriminating ability of work capacity measurements. For example, consider familiarization tests to determine the ideal treadmill speed for each subject during the AST (ie. to bring exhaustion between 35 - 55 seconds). 3. Examine factors leading to the variable response to iron supplementation. For example multiple mineral deficiency. 4. Examine a wider range of muscle enzymes. 5. Select a larger group of latent (as opposed to prelatent/latent) iron deficient females for study. 49 REFERENCES Addison G., C. Hales, "Two site assay of human growth hormone", Hormone Metabolism Research. 3: 59, 1971. Anderson, H.T., H. Barkve, "Iron deficiency and muscular work performance", Scandinavian Journal of Clinical and Laboratory  Investigation. 25(suppl. 114): 1-62, 1970. Apte, S.V., P.S. Venkatachalom, "Factors influencing dermal loss of iron in human volunteers", Indian Journal of Medical Research. 50: 817-822, 1962. Aronson, V., "Ancient Man's Meat", Runners World. Rodale Press, Emmaus, Penn., August: 21, 1985. Baird, I.M., R.L. Walters, D.R. Sutton, "Absorption of slow-release iron and effects of ascorbic acid in normal subjects after partial gastrectomy", British Medical Journal. 4: 505-508, 1974. Banister, E.W., C.L. Hamilton, "Variations in iron status with fatigue modelled from training in female distance runners", European  Journal of Applied Physiology. 54: 16-23, 1985. Bannerman, R.M., B.R. Beveridge and L.J. Witts, "Anaemia associated with unexplained occult blood loss", British Medical Journal. 1: 1417-1419, 1964. Barcikowski, R.S. (ed.), Computer Packages and Research Design Volume 1. BMDP. University Press of America, Lanham, Md., 1983. Beutler, E., S. Larsh, F. Tanzi, "Iron enzymes in iron deficiency: Oxygen comsumption measurements in iron deficient subjects", American Journal of Medical Science. 239: 759-765, 1960. Blacklock, N., "Bladder trauma in the long distance runner: '10 000 meters haematuria'", British Journal of Urology. 49: 129-132, 1977. 50 Blayney, L, R. Bailey-Wood, A. Jacobs, A. Henderson and J. Muir, "The effects of iron deficiency on the respiratory function and cytochrome content of rat heart mitochondria", Circulatory  Research. 39: 744-748, 1976. Boileau, M., E. Fuchs, J . Barry, C. Hodges, "Stress hematuria: Athletic pseudo-nephritis in marathoners", Urology. 15(5): 471-474, 1980. Bothwell, T.H., R.W. Charlton, J.D. Cook, C.A. Finch, Iron Metabolism in  Man. Blackwell Scientific Publications, Oxford, 1979. Bouchard C , A.W. Taylor, S. Dulac, "Testing maximal anaerobic power and capacity", from Physiological testing of the elite athlete. J.D. MacDougall, H.A. Wenger, H.J. Green eds., Canadian Assoc of Sports Sciences, Mutual Press, Canada, 1982, pp 61-74. Brooks, S.M., C F . Sandborn, B.M. Albrecht and W.W. Wagner Jr., "Diet in athletic amenorrhaea", Lancet. 1: 559-560, 1984. Brotherhood, J. , B. Brozovic and L.G.C. Pugh, "Haematological status of middle and long distance runners", Clinical Science and Molecular Medicine. 48: 139-145, 1975. Brune, M.,B. Magnusson, H. Persson and L. Hallberg, "Iron losses in sweat", American Journal of Clinical Nutrition. 43: 438-443, 1986. Buckman, M.J., "Gastrointestinal bleeding in long distance runners", (Editorial), Annals of Internal Medicine. 101: 127-128, 1984. Casoni, I., C. Borsetto, A. Laviechi, S. Martinelli, F. Conconi, "Reduced hemoglobin concentration and red cell hemoglobinization in Italian marathon and ultramarathon runners", International Journal of  Sports Medicine. 6(3): 176-179, 1985. Cavill, I., "Diagnostic Methods", Clinics in Haematoloov. 11(2): 259-273, 1982. 51 Celsing, F., E. Blomsted, B. Werner, P. Pihlstedt, B. Ekblom, "Effects of iron deficiency on endurance and muscle enzyme activity in man", Medicine and Science in Sports and Exercise. 18(2): 156-161, 1986. Charlton, R.W., D. Derman, B. Skikne, J.D. Torrance, S.R. Lynch, M.H. Sayers, S. Zwi, H.I. Goldman, A. Van As, G. Margo, J.T. Schneider and T.H. Bothwell, "Anemia, iron deficiency and exercise: Extended studies in human subjects", Clinical Science and Molecular  Medicine. 53: 537-541, 1977. Christensen, D.J., "Differentiation of iron deficiency and the anemia of chronic disease", Journal of Family Practise. 20(1): 35-39, 1985. Clark, N., "Increasing dietary iron", The Physician and Sports Medicine. 13(1): 131-132, 1985. Clement, D.B., R. C. Asmundson, C.W. Medhurst, "Hemoglobin values: Comparative survey of the 1976 Canadian Olympic team", Canadian  Medical Association Journal. 117: 614-616, 1977. Clement D.B., "Anemia and iron deficiency in athletes", Sports. The Coaching Assoc. of Canada, Nov. 1981. Clement, D.B., R.C. Asmundson, "Nutritional intake and hematological parameters in endurance runners", The Physician and Sports  Medicine. 10(3): 37-43, 1982. Clement, D.B., J.E. Taunton, K. Poskitt, "Iron deficiency anemia in a distance runner", Canadian Family Physician. 28: 1008-1012, 1982. Clement, D.B. and L.L. Sawchuk, "Iron status and sports performance", Sports Medicine. 1: 65-74, 1984. Colt, E. and B. Heyman, "Low ferritin levels in runners", Journal of  Sports Medicine. 24: 13-17, 1984. 52 Committee on Dietary Allowances, Food and Nutrition Board, National  Research Council Recommended Dietary Allowances. 9—- Review  Edition. (National Academy of Sciences), Washington, 1980. Conrad, M.E., J.C. Barton, "Factors affecting iron balance", American Journal of Hematology. 10: 199-225, 1981. Cook, J.D., D.A. Lipschitz, E.M. Laughton, C.A. Finch, "Serum ferritin as a measure of iron stores in normal subjects", American Journal of  Clinical Nutrition. 27: 681-687, 1975. Cook, J.D., S.S. Watson, K.M. Simpson, D.A. Lipschitz, B.S. Skikne, "The effect of high ascorbic acid supplementation on body iron stores", Blood. 64(3): 721-726, 1984. Dallman, P.R. and H.C. Schwartz, "Distribution of cytochrome C and myoglobin in rats with dietary iron deficiency", Paediatrics. 35: 677-686, 1965. Dallman, P.R., E. Beutler and B.A. Finch, "Effects of iron deficiency exclusive of anemia", British Journal of Haematologv. 179-184, 1978. Dallman, P.R., "Manifestations of iron deficiency", Seminars in Hematology. 19(1): 19-30, 1982. Davidson, R.J.L., "Exertional haemoglobinuria: A report on three cases with studies on the haemolytic mechanism", Journal of Clinical  Pathology. 17: 536-540, 1964. Davidson, R.J.L., "March or exertional haemoglobinuria", Seminars in  Hematology. 6: 150-161, 1969. Davies, K.J.A., J.J. Maguire, G.A. Brooks, P.R. Dallman and L. Packer, "Muscle mitochondrial bioenergetics, oxygen supply and work capacity during dietary iron deficiency and repletion", American  Journal of Physiology. 242(Endocrinol Metab. 5): E418-E427, 1982 53 Davis, J.A., M.H. Frank, B.J. Shipp and K. Wasserman, "Anaerobic threshold alterations caused by endurance training in middle-aged men", Journal of Applied Physiology. 46:6, 1039-1046, 1979. DeWijn, J.F., J.L. DeJongste, W. Mosterd and D. Willebrand, "Hemoglobin, packed cell volume, serum iron and iron binding capacity of selected athletes during training", Nutritional  Metabolism. 13: 129-139, 1971. Dickson, D.N., R. L. Wilkinson and T.D. Noakes, "Effects of ultra-marathon training and racing on hematologic parameters and serum ferritin levels in well trained athletes". International  Journal of Sports Medicine. 3: 111-117, 1982. Diehl, D.M., T.G. Lohman, S.C. Smith and R.J. Kertzer, "The effects of physical training on iron status of female athletes", Medicine and  Science in Sports and Exercise. (Abstract), 14: 164, 1982. Dressendorfer, R.H., E. W. Askew, J.E. Rimar, C.L. Keen, D.E. Wade, S. Gordon and G.C. Timmis, "Development of sports anemia in men without iron deficiency", Abstract only; presented at annual meeting of the American College of Sports Medicine, Nashville, Tennessee,1985. Dufaux, B., A. Hoederath, I. Streitberger, W. Hollman and G. Assmann, "Serum ferritin, haptoglobin and iron in middle and long distance runners, elite rowers, and professional racing cyclists", International Journal of Sports Medicine. 2: 43-46, 1981. Edgerton, V.R., S.L. Bryant, C A . Gillespie and G.W. Gardner, "Iron deficiency anemia and physical performance and activity of rats", Journal of Nutrition. 102: 382-400, 1972. Edgerton, V.R., Y. Ohira, J . Hettiarachi, B. Senewirante, G.W. Gardner, R.J. Barnard, "Elevation of hemoglobin and work tolerance in iron deficient subjects", Journal of Nutritional Science and  Vitaminology. 27: 77-86, 1981. 54 Ehn, L, B. Carmark and S. Hoglund, "Iron in young sportsmen" in Eriksson, B., B. Furburg (eds.), Swimming Medicine IV. University Park Press, Baltimore, pp. 85-88, 1977. Ehn, L, B. Carmark and S. Hoglund, "Iron status in athletes involved in intense physical activity", Medicine and Science in Sports and Exercise. 12(1): 61-64, 1980. Eichner, E.R., "Runners macrocytosis: A clue to footstrike hemolysis", The American Journal of Medicine. 78: 321-325, 1985. Eichner, E.R., "The anemias of athletes", The Physician and Sports  Medicine. 14(9): 122-130, 1986. Ekblom, B. A. Goldburg and B. Gullbring, "Response to exercise after blood loss and reinfusion", Journal of Applied Physiology. 33: 175-180, 1972. El-Hawary, M.F.S., F.A. El Shobaki, T. Kholeif, R. Sakr and M. El Bassovssy, "The absorption of iron, with or without supplements of single amino acids, and of ascorbic acid in healthy and Fe-de-ficient children", British Journal of Nutrition. 33: 351-355,1975. Ericsson, P., "The effect of iron supplementation on the physical work capacity in the elderly", Acta Medica Scandinavia. 188: 361-374, 1970. Eston, R.G., "The regular menstrual cycle and athletic performance", Sports Medicine. 1: 431-445, 1984. Falsetti, H.L, E.R. Burke, R.D. Feld, E.D. Frederick, D. Ratering, "Hematological variations after endurance running with hard- and soft-soled running shoes", The Physician and Sports Medicine. 11 (8):118-127, 1983. Finch, C.A., L. R. Miller, A. R. Inamdar, R. Person, K. Seiler and B. Mackler, "Iron deficiency in the rat: Physiological and biochemical studies of muscle dysfunction". Journal of Clinical  Investigation. 58: 447-453, 1976. 55 Finch, C.A., P.D. Gollnick, M.P. Hlastala, L.R. Miller E. Dillman and B. Mackler, "Lactic acidosis as a result of iron deficiency", Journal of Clinical Investigation. 64: 129-137, 1979. Finch, C.A., "Drugs effective in iron deficiency and other hypochromic anemias" in The Pharmacological Basis of Therapeutics. 6 t h ed., Gilman, A.G., L.B. Goodman and A. Gilman (eds.), Chapter 56, 1980. Finch, C A . and H. Huebers, "Perspectives in iron metabolism", New  England Journal of Medicine. 306(25): 1520-1528, 1982. Ganong, W.F., Review of Medical Physiology. 1 0 t n ed., Lange Medical Publications, Los Altos, California, 1981, pp 378, 379, 246, 416. Gardner, G.W., V.R. Edgerton, B. Senewiratne, R.J. Barnard and Y. Ohira, "Physical work capacity and metabolic stress in subjects with iron deficiency anemia", American Journal of Clinical Nutrition. 30: 910-917, 1977. Gomez, G. and J. Gomez, "Anemia in general practise: Controlled release ferrous sulphate and Vitamin C compared with ferrous sulphate B.P., The British Journal of Clinical Practise. 23(10): 421-424, 1969. Grebe, G., C. Martinez-Torres, M. Layrisse, "Effect of meals and ascorbic acid on the absorption of a therapeutic dose of iron as ferrous and ferric salts", Current therapeutic research. 17(4): 382-397, April 1975. Griner, P.F., et al, "Changes in posture may affect hemoglobin values", American Family Physician. 24: 189, 1981. Guthrie, H.A., Introductory Nutrition. 3 r d ed., C V . Mosby Co., St. Louis, 1975, pp 145-160. Guyton, A . C , Basic Human Physiology: Normal Function and Mechanisms of Disease. 2 n c ' ed., W.B. Saunders Co., Philadelphia, 1977, pp 59-60. 56 Hagler, L, E.W. Askew, J.R. Neville, P.W. Mellick. R.I. Coppers Jr., J.F. Lowder Jr., "Influence of dietary iron deficiency on hemoglobin, myoglobin, their respective reductases, and skeletal muscle mitochondrial respiration", American Journal of Clinical Nutrition. 34: 2169-2177, 1981. Hallberg, L, "Iron", from Nutrition Reviews: Present Knowledge in  Nutrition. 5 t n ed., Nutrition Foundation Inc., Washington, D.C, 1984, p 53. Hallberg, L, B. Magnusson, "The etiology of 'sports anemia'" editorial, Acta Medical Scandinavia. 216: 145-148, 1984. Harju, E. and H. Lindberg, "The increase of body iron stores estimated by the increase of serum ferritin concentration during ferrous sulphate treatment after gastrointestinal surgery", Journal of  Parenteral and Enteral Nutrition.10: 66-69, 1986. Hatefi, Y. and D.L. Stiggall, "Metal-containing flavoprotein dehydrogenases", from The Enzymes. 3rd ed., Vol. XIII, Ed. by P. Boyer, Academic Press, 1971. Hayashi, M., T. Kume, H. Nihira, "Abnormalities of renal venous system and unexplained renal haematuria", Journal of Urology. 124(1): 12-16, 1980. Haymes, E.M., "Iron deficiency and the active woman", in DGWS Research Reports: Women in Sport. Harrif, D.V. (ed.), Vol. II, pp. 91-97, Washington AAHPER Press, 1973. Haymes, E.M., J.L. Puhl & T.E. Temples, "Training for cross country skiing and iron status", Medicine and Science in Sports and Exercise. 18(2): 162-167, 1986. Health and Welfare Canada, "Dietary Standard for Canada", Supply and Services, 1976, pp 32, 33, 36, 37, 58-63, 67. Health and Welfare Canada, "Nutrition Canada: the British Columbia survey report", Supply and Services,1975, pp 107-115. 57 Health and Welfare Canada, "Recommended nutrient intakes for Canadians", Supply and Services, 1983, pp 120-136. Heinrich, H.C, "Intestinal iron absorption in man: methods of measuring dose relationship, diagnostic and therapeutic applicatons", in Hallberg, I., H.G. Harworth, and A. Vannetti (eds.), Iron Deficiency: Pathogenesis. Clinical Aspects. Therapy. Academic Press, New York, pp. 213-296, 1970. Heinrich, H.C, J . Bruggemann, E.E. Gabbe, M. Glaser, "Correlation between diagnostic F e 2 + absorption and serum ferritin concentration in man", Zeitschritt fur Naturforschung. 32(c): 1023-1025, 1977. Hoederath, A., B. Dufaux, W. Hollmann, "Serum ferritin during a four day footrace", International Journal of Sports Medicine. 4: 64, 1983. Hollmann, W., A. Mader, H. Leesen, B. Dufaux, R. Rost, H. Heck, K. Weber, R. Volker, "Haemodynamic and metabolic aspects in top class competetive sport", Current Topics in Sports Medicine. Bachl, Prokop and Sucker, eds., Baltimore, 137-150, 1984. Hunding, A., R. Jordal, P.E. Pauler, "Runners anemia and iron deficiency", Acta Medical Scandinavia. 209: 315-318, 1981. Jacobs, A. and M. Worwood, Iron in Biochemistry and Medicine. Academic Press, London, 1974, pp 335-362. Jacobs, A., E. Miller, M. Worwood, M.R. Beamish, C A . Wardrop, "Ferritin in the serum of normal subjects and patients with iron deficiency and iron overload", British Medical Journal. 4: 206-208, 1972. Jacobs, A., "The non-haematological effects of iron deficiency", Editorial Review, Clinical Science and Molecular Medicine. 53: 105-109. 1977. Jacobs, A., "Serum ferritin and iron stores". Federaton Proceedings. 36(7): 2024-2027, 1977. 58 Jacobs, A., "Ferritin: An Interim Review", Current Topics in  Hematology. Alan R. Liss Inc., New York, 5: 25-62, 1985. Jacobs, P., G. Johnson and L. Wood, "Oral iron therapy in human subjects, comparative absorption between ferrous salts and iron polymaltose", Journal of Medicine. 15(5&6): 367-377, 1985. Jeschke, D., H.C. Heitkamp, R. Locher and D. Schneider, "Effects of different work duration in gradual bicycle and treadmill ergometer tests on aerobic/anaerobic capacity and anaerobic threshold determinations", Current Topics in Sports Medicine. Bachl, Prokop and Sucker, eds., Baltimore, 151-160, 1984. Johnston, F.A., T.S. McMillan, E.R. Evans, "Perspiration as a factor influencing the requirement for calcium and iron", Journal of  Nutrition. 42: 285-295, 1950. Jurkowski, J.E.H., "Hormonal and physiological responses to the menstrual cycle", Canadian Journal of Applied Sport Science. 7(2): 85-89, 1982. Kirschman, J.D., Nutrition Almanac. 2 n c l ed., Nutrition Search Inc., McGraw-Hill Book Co., New York, 1984, pp 75-76, 120-121, 152. Klausner, R.D., J.B. Harford, K. Rao, E. Mattia, A.W. Weissman, G. Ashwell, and J. Van Renswoude, "Molecular aspects of the regulation of cellular iron metabolism", Proteins of Iron Storage  and Transport. (Spik, Montrevil, Chrichton and Mazur, eds.), Elsevier Science Publishers, Amsterdam, 1985, pp 111-122. Kreg, S.H., & R.L. Murray, "Dynamics of nutrition support: Assessment, Implementation, Evaluation", Appleton-Century Crofts, Norwalk, Connecticut, 1986, pp 234-246. Lee, P.C, J.R. Ledwich, D.C. Smith, "Large and small doses of ascorbic acid in the absorption of ferrous iron", Canadian Medical  Association Journal. 97: 181-184, 1967. 59 Lee, G.R., S. Nacht, J.N. Lukens and G.E. Cartwright, "Iron metabolism in copper deficient swine", Journal of Clinical Investigation. 47: 2058-2069, 1968. Lehninger, A.L., Principles of Biochemistry. Worth Publishers, Inc., New York, 1982. Lindemann, R., R. Ekanger, P.K. Opstad, M. Nummestad and R. Ljostand, "Haematological changes in normal men during prolonged severe exercise", American Corrective Therapy Journal. 32(4): 107-111, 1978. Mackler, B., R. Grace and C.A. Finch, "Iron deficiency in the rat: effects on oxidative metabolism in distinct types of skeletal muscle", Pediatric Research. 18(6): 499-500, 1984. Magnusson, B., L. Hallburg, L. Rossander, B. Swolin, "Iron metabolism and 'sports anemia'", Acta Medical Scandinavia. 216: 149-164, 1984. McClure, S., E. Custer, J.D. Bessman, "Improved detection of early iron efficiency in non-anemic subjects", Journal of American Medical  Association. 253: 1021-1023, 1985. McKenzie, D.C, W.S. Parkhouse and W.E. Hearst, "Anaerobic performance characteristics of elite Canadian 800m runners", Canadian Journal of Applied Sport Science. 7(3): 158-160, 1982. McLane, J.A., R.D. Fell, R.H. McKay, W.W. Winder, E.B. Brown, and J.O. Wolloszy, "Physiological and biochemical effects of iron deficiency on rat skeletal muscle", American Journal of  Phvsioloov. 241 (Cell Physiology 10): C47-C54, 1981. McMahon, L.F., M.J. Ryan, D. Larson and R.L Fisher, "Occult gastro-intestinal blood loss in marathon runners", Annals of Internal  Medicine. 101: 846-847, 1984. 60 Miles, L.E.M., C P . Bieber, L.F. Eng and D.A. Lipschitz, "Properties of two-site immunoradiometric (labelled antibody) assay systems. Reprint from "Radioimmunoassay and related procedures in medicine", Vol.1, International atomic aaencv. Vienna, 1974. Monsen, E.R., J.L. Balintfy, "Calculating dietary iron bioavailability: refinement and computerization", Journal of American Dietetic  Association. 80: 307-311, 1982. Newsholme, E.A. & C. Start, Regulation in metabolism. John Wiley and Sons, London, 1973. Nickerson, H.J. and A. Tripp, "Iron deficiency in adolescent cross country runners", The Physician and Sports Medicine. 11(6): 60-66, 1983. Nickerson, H.J., M. Holubets, A.D. Tripp and W.E. Pierce, "Decreased iron stores in high school female runners", American Journal of  Diseases of Children. 139: 1115-1119, November, 1985. Nienhuis, A.W., "Vitamin C and Iron", The New England Journal of  Medicine. 304 (3),170-171, 1981. Ohira, Y., V.R. Edgerton, G.W. Gardner, B. Senewirante and D.R. Simpson, "Non hemoglobin related effects on heart rate in iron deficiency anemia", Nutrition Reports International. 18(6): 647-651, 1978. Ohira, Y., V.R. Edgerton, G.W. Gardner, B. Sinewirante, R.J. Barnard and D.R. Simpson, "Work capacity heart rate and blood lactate response to iron treatment", British Journal of Haematoloov. 41: 365-372, 1979. Ohira, Y., V.R. Edgerton, G.W. Gardner, K.A. Gunawardena, B. Sinewirante, "Characteristics of blood gas in response to iron treatment and exercise in iron deficient and anemic subjects", Journal of Nutritional Science and Vitaminology. 29: 129-139, 1983. 61 Overall, J.E. and J.A. Woodward, "Unreliability of difference scores: A paradox for measurement of change", Psychological Bulletin. 82(1): 85-86, 1975. Pakarinen, A., "Ferritin in Sports Medicine", NordiLab Newsletters. 4: 20-28, 1980. Parr, R.B., "Iron deficiency in female athletes", The Physician and  Sports Medicine. 12(4): 81-86, 1984. Passmore, R. and M.A. Eastwood, Human Nutrition and Dietetics. Churchill Livingstone, Edinburgh, 1986, pp 121-126. Pate, R.R., M. Maguire and J.V. Wyk, "Dietary iron supplementation in women athletes". The Physician and Sports Medicine. 7 (9): 81-88, 1979. Plowman, S.A. and P.C. McSwegin, "The effect of iron supplementation on female cross country runners", Journal of Sports Medicine. 21: 407, 1981. Porter, A.M.W., "Do some marathon runners bleed into the gut?", British Medical Journal. 287: 1427, 1983. Potera, C , "Gastro-intestinal bleeding found in long distance runners", The Physician and Sports Medicine. 12(12): 29, 1984. Puhl, J.L. and W.S. Runyan, "Hematological variations during aerobic training of college women", Research Quarterly for Exercise and  Sport. 51(3): 533-541, 1980. Refusm, S.B. and B.B. Schreiner, "Regulation of iron balance by absorption and excretion: A critical review and new hypothesis", Scandinavian Journal of Gastroenterology. 19: 867-874, 1984. Robinson, C.H., E.S. Weigley, Basic Nutrition and Diet Therapy. 5 i n ed., Macmillan Publishing Co., New York, 1984, pp 91-93. 62 Roeser, H.P., "The role of ascorbic acid in the turnover of storage iron", Seminars in Hematology. 20(2): 91-100, 1983. Ruckman, K. and A. Sherman, "Effects of exercise on iron and copper metabolism in rats", Journal of Nutrition. 111: 1593-1601, 1981. Schoene, R.B., P. Escourrov, H.T. Robertson, K.L. Nilson, J.R. Parsons and N.J. Smith, "Iron repletion decreases maximal exercise lactate concentrations in female athletes with minimal iron deficiency anemia", Journal of Laboratory and Clinical Medicine. 102(2): 306-312, 1983. Schutz, R.W., F.L. Small and M.E. Gessaroli, "Multivariate Statistics: A self test and guide to their utilization", Research Quarterly for  Exercise and Sport. 54(3): 255-263, 1983. Schutz, R.W. and M.E. Gessaroli, "The analysis of repeated measures designs involving multiple dependent variables", unpublished paper, 1986. Shiraki, K., H. Yoshimura and T. Yamada, "Anemia during physical training and physical performances",. World Congress in Sports Medicine Handbook. (Melbourne, Australia), 410-415, 1974. Siegel, A., C. Hennekens, H. Solomin and B. Vanboeck, "Exercise-related hematuria findings in a group of marathon runners", Journal of the American Medical Association. 241(4): 391-392, 1979. Siimes, M.A., C. Refino, P.R. Dallman, "Manifestation of iron deficiency of various levels of dietary iron intake", American Journal of  Clinical Nutrition. 33: 570-574, 1980. Sonnenwirth, A . C , L. Jarett, Gradwohl's Clinical Methods and Diagnosis. 8 t h ed., CV . Mosby Co., St. Louis, 1980 pp 850-859. 63 Steel, J.E., "A Nutritional study of Australian Olympic athletes", Medical Journal of Australia. 7 ,119, 1970. Steenkamp, I., C. Fuller, J . Graves, T.D. Noekes, P. Jacobs, "Marathon running fails to influence RBC survival rates in iron-replete women", The Physician and Sports Medicine. 14(5): May 1986. Stewart, J.G., D.A. Ahlquist, D.B. McGill, D.M. Ilstrup, S. Schwartz and R.A. Owen, "Gastrointestinal blood loss and anemia in runners", Annals of Internal Medicine. 101: 843-845, 1984. Strausse, L, J . Hegenauer and P. Saltman, "Effects of exercise on iron metabolism in rats", Nutrition Research. 3: 70-89, 1983. Strauzenberg, S.E., R. Kassner, R. Bohm, F. Schneider, "Iron deficiency in female athletes", Medicine and Sport. 14: 200-208, 1981. Thoden J.S., B.A. Wilson, J.D. Macdougall, "Testing aerobic power", in Physiological testing of the elite athlete. J.D. Macdougall, H.A. Wenger, H.J. Green eds., Canadian Association of Sports Sciences, Mutual Press, Canada.1982, pp 39-52. Valberg, L.B., "Plasma ferritin concentration: their clinical significance and relevance to patient care", Canadian Medical  Association Journal. 122: 1240-1248, June 1980. Valberg, L.S., J . Sorbie, T. Ludwig and D. Pelletier, "Serum ferritin and the iron status of Canadians", Canadian Medical Association  Journal. 114, 417-421, 1976 Van der Beek, E.J., "Vitamins and endurance training food for running or faddish claims?", Sports Medicine. 2: 175-197, 1985. Vellar, O.D., "Studies on sweat losses of nutrients. Iron content of whole body sweat and its association with other sweat constituents, serum iron levels, hematological indices, body surface area, and sweat rate", Scandinavian Journal of Clinical  Investigation. 21: 157-167, 1968. 64 Vellar, O.D. and L. Hermansen, "Physical performance and hematological parameters", Acta Medica Scandinavia. 522(suppl.): 1-40, 1971. Viteri, F.E. and B. Torun, "Anemia and physical work capacity", Clinical Haematology. 3: 609-626, 1974. Volpicelli, N.A., "Effect of sports on the gastrointestinal tract and liver", Sports Medicine: Fitness. Training. Injuries. (O. Appenzeller and AT. Kinson, eds.), 2 n d ed., Urband and Schwarzenberg, Baltimore, 1983, pp. 91-98. Weib, M., T. Krauter, K.H. Graff, W. Abmayr and W. Weicker, "Iron pathway and iron loss in different forms of training and the origin of sports anemia", International Journal of Sports Medicine. 4: 64, 1983. Williamson, M.R., "Anemia in runners and other athletes", Physician  and Sports Medicine. 9(6): 73-79, 1981. Winick, M., "Beating iron deficiency anemia: A family physicians guide", Modern Medicine of Canada. 36(11): 1571-1581, 1981. Wirth, J.C., T.G. Lohman, J.P. Avallone Jr., T. Shire and R.A. Boileau, "The effect of physical training on the serum iron levels of college-age women", Medicine and Science in Sports. 10(3): 223-226, 1978. Wishnitzer, R., E. Vorst and A. Berrebi, "Bone marrow iron depression in competitive distance runners", International Journal of Sports  Medicine. 4: 27-30, 1983. Wood, M. and P. Elwood, "Symptoms of iron deficiency anemia. A community survey", British Journal of preventive and social medicine . 20: 117,1966. Yoshimura, H., "Anemia during physical training (sports anemia)", Nutrition Reviews. 28 (10): 251-253, 1970. 65 Appendix A Review of Literature OVERVIEW OF IRON METABOLISM Iron is a trace element found in all cells of the body and is essential for life. In fact, virtually all living matter must procure a certain amount of iron for normal development (Finch, 1980). Many plants and animals have developed mechanisms for acquiring this crucial element. The requirement of a hemoglobin/myoglobin oxygen transport system expands Man's requirements tenfold compared to animals without this transport system. This fact and the progressive change in diet away from meats that began with the cultivation of grains about 10,000 years ago has made the maintenance of proper iron balance difficult. The 3-5 g of total body iron in the adult allows many key biochemical reactions to take place. It is responsible for electron transport (cytochromes), activation of oxygen (oxidases and oxygenases) and for the transport of oxygen and carbon dioxide (hemoglobin and myoglobin). Hemoglobin accounts for about 2.5 gs (2500 mg) of the 4 g total; myoglobin values are approximately 150 mg and the metalloenzymes account for 6-8 mg. There is 66 about 3 mg of iron in the plasma and this is bound to a carrier protein, transferrin. Tissues and organs such as the liver, spleen, and bone marrow store iron as ferritin, or the more insoluble hemosiderin, in varying amounts (Haymes, 1973, Valberg, 1980). Measurement of the degree of saturation of transferrin (and red cell protoporphyrin) can detect iron deficient erythropoiesis, while quantitation of iron in plasma ferritin and in bone marrow reveals the status of the body's stores (Sonnenwirth and Jarett, 1980). Stores in adult men may be 500 to 1000 mg while the iron stores of women seldom exceed 500 mg (Bothwell et al., 1979). Internal exchange of iron is accomplished by the plasma, B-glycoprotein, transferrin (Finch, 1980). Normally transferrin is about 35% saturated with iron and the normal plasma iron level is about 130 ug-dl"1 (23 umolT1) in men and 110 ugdl"1 (19 umoll"1) in women. There is a continuous reutilization of iron catabolized in the body. Most of this is in red blood cells which circulate for approximately 120 days before being catabolized by the reticuloendothelial system, which is comprised of liver, bone marrow, and spleen. Iron absorbed from the small intestine or from hemoglobin in the reticuloendothelial system is taken up by transferrin and is transported to the bone marrow for formation of red blood cells. The recycling of iron in the red cell mass helps maintain iron balance. The following diagram shows the pathways of iron metabolism. 67 Dietary Iron (14±4 mg/day: about 6 mg/IOOO kcal) I Intestinal Mucosa (Absorption about 1 mg/day) I Plasma Iron '(pool about 3 mg: turnover about 10X/day)^ Erythroid Marrow (uptake about 25 mg/day) Circulating Erythrocytes (pool about 210Omg: daily turnover 18 mg) Interstitial Fluid I Parenchymal Exchange-Especially Liver (about 6 mg/day) I Ferritin Stores Reticuloendothelium (25 mg/day from erythron) Figure #1. Pathways of iron metabolism in man (excretion omitted). (From Finch, 1980). 68 The most remarkable feature of iron metabolism in humans is the degree to which the body store is conserved. For normal men only 1 mg/day is lost. Two thirds of this iron is excreted from the gastrointestinal tract as extravasated red blood cells, iron in bile, and iron in exfoliated mucosal cells. The other third is accounted for by small amounts of iron in sweat, desquamated skin, and in the urine (Bothwell et al., 1979). Iron losses in the male vary less than in the female. About 0.5 mg may be lost in the iron deficient male and as much as 1.5 to 2 mg/day can be lost when excessive iron is consumed (Bothwell et al., 1979). Additional losses of iron occur in the female due to menstruation. Menstrual losses average 0.5 mg/day although 10% of the normal menstruating females lose over 2 mg/day. Estrogen containing oral contraceptives reduce menstrual loss by about one-half, while intra-uterine devices increase losses (Finch,1980). Athletes may have increased requirements for iron and increased losses. Endurance athletes, especially females in their reproductive years, are at a high risk of becoming iron deficient. In other words, the average amount of iron derived from their diet may be less than the losses from their body. 69 A negative iron balance would cause a depletion of iron stores, increased absorption from the gut, a fall in plasma iron concentration, impairment of erythropoieses, and finally microcytic hypochromic anemia (Winick, 1981). One could, however come into balance at any of these stages. The stage at which iron deficiency affects work capacity has not been fully elucidated nor has the biochemical locus of this detriment been determined (Strauss, 1983). This review of iron metabolism will address two pertinent areas: 1) the causes for a negative iron balance in athletes; and, 2) the effects this may have on work capacity, especially in the prelatent/latent iron deficiency stages. NEGATIVE IRON BALANCE IN ATHLETES In endurance athletes, the sites of jeopardized iron balance may be inadequate dietary intake, malabsorption from the gut, or increased iron losses in the sweat, urine or feces. 70 Inadequate Dietary Intake It is difficult for the typical western diet to meet the iron demands of the menstruating female. The average western diet supplies only 5-7 mg of the iron per 1000 Kcal. Since the recommended dietary intake in Canada for females is 14 mg based on an average 10% absorption (Health & Welfare Canada, 1983; ) or even 18 mg as is the U.S. Recommended Dietary Intake (Committee on Dietary Allowances, 1980), intakes below 2000 Kcal are often associated with iron deficiency (Clement & Sawchuk, 1984). In fact, nutritional analysis of female distance runners revealed an average intake of 12.5 mg (Clement and Asmundson, 1982). Female athletes in sports such as gymnastics, ballet, and distance running, where low body fat is an asset, run a high risk of developing iron deficiency due to a low caloric intake. Much of the iron in food is absorbed inefficiently. The absorption of iron depends on both the amount and bioavailability of dietary iron. The bioavailability of food iron is quite varied. Heme iron, which is in the reduced, ferrous (Fe + 2) form and is found in meats, is easily absorbed; (i.e. 10-37% of intake is absorbed) (Bothwell et al., 1979; Finch, 1980; Hallberg, 1984). Non-heme iron, which represents by far the largest amount of dietary iron (i.e. vegetable foods), has poor bioavailability as usually less than 10% of intake is absorbed (Bothwell et al., 1979; 71 Finch, 1980). In vegetable foods, ferric iron (Fe + 3) is bound in insoluble complexes and its' absorption is altered to a great extent by the other ingredients of the meal (Passmore and Eastwood, 1986). Meat, apart from supplying heme iron, appears to promote the absorption of non-heme iron, probably by meat stimulating the production of gastic acid (Finch, 1980). Most of the non-heme iron ingested is in the ferric form and must be reduced to the ferrous state to be easily absorbed. The acid secretions of the stomach aid in this conversion, hence absorption is mainly from the upper portion of the small intestine. Ascorbic acid (Vitamin C) also reduces F e + 3 to F e + 2 . Antacids negate this effect. Interfering with the absorption of non-heme iron is phytic acid found in cereals. Phytates, carbonates, phosphates, tannins and oxalates react with iron to form insoluble compounds in the intestine (Ganong, 1981; Passmore and Eastwood, 1986). Calcium may counteract the effect of excess phosphorus for in sufficient amounts calcium will combine with phosphates and free iron for use (Kirschmann, 1984) ^ The high prevalence of iron deficiency is largely a function of the limited availability of iron in the contemporary diet (Bothwell et al., 1979). Paleolithic man had a diet with meat comprising up to 80% of the total diet (Aronson, 1985). Iron, folic acid, and protein were ample. The agricultural revolution drastically reduced animal protein consumption. 72 Malabsorption The limited physiological losses of iron point to the primary importance of absorption as the determinant of the body's content of iron. The major regulator of iron repletion is that of keeping the absorptive processes attuned to body requirements (Haymes, 1973). Thus the amount of iron crossing the small intestinal epithelial barrier is related to the internal iron status. The extent to which this mechanism operates in a normal population is quite variable since there is a wide range in the amount of storage iron found in individual subjects (Jacobs, 1985). Absorption may be through passive diffusion and active transport, although most iron is passed through the mucosal cells by binding to the protein apoferritin, forming ferritin molecules. If there is a depletion of the body's iron stores there is a compensatory increase in the amount of iron absorbed into the blood. The rate at which iron enters the blood is dependent upon the quantity of ferritin present in the intestinal mucosal cells, which in turn is dependent on the amount and saturation of transferrin. The ferritin level in the mucosal cells will decrease in the event of an iron deficiency which will consequently allow more iron to enter the plasma (Klausner et al.,1985). 73 Luminal Phase Mucosal Phase Blood Food ^/\\ Digestive/ \ \ J u . i c e s / \ Heme i J Non-heme v Compounds ligands ^^Ava i lab le ^^•Complex 1 l Unavailable Complex i •. % n r m r ^ %; ^s* i Transport ! Protein Noil '| Ferritin! 1 \ —•Transferrin Figure #2. The effects of various intraluminal factors on the absorption of heme and non-heme iron. (From Bothwell, 1979). 74 In spite of this mechanism for maintaining iron balance, absorption disturbances appear to contribute to a negative balance in a number of athletes. Ehn et al., (1980), using radioactive iron, found that absorption in iron deficient runners was 16.4%, while the iron deficient controls absorbed 30% of the labelled iron. Similar discrepancies between the absorption values of athletes versus non-exercisers have been reported by Clement et al., (1983), Heinrich, (1970), Heinrich et al., (1977), and Bannister and Hamilton, (1985). The mechanism for these absorption disturbances has not been determined. It is known that any cause of intestinal hurry can be associated with iron deficiency (Passmore and Eastwood, 1986), but whether this is a factor in athletes is not known. Bannister and Hamilton, (1985), proposed that heavy training causes the plasma transferrin saturation to be elevated. This would retard the release of iron from the mucosal cells of the intestine. Not until periods of relative rest, such as peaking, would the % saturation of transferrin decline, allowing iron to be absorbed, and the RBC count and Hb concentration to increase. 75 Gastrointestinal Blood Loss Long distance running may cause gastrointestinal bleeding that contributes to iron deficiency in some runners (Porter, 1983; Potera, 1984; Stewart et al., 1984). Four recent studies have shown blood loss through the stool after long distance races (10 km to marathon) (McMahon et al. 1984; Porter, 1983; Stewart et al., 1984; Volpicelli, 1983). Using the Hemoccult test to detect gastrointestinal bleeding after racing, Porter (1983) found an incidence of 8%, Volpicelli (1983) 20% and McMahon et al. (1984) 22%. Stewart and colleagues used the Hemoquant assay (a more sensitive assay, yet not at present commercially available) and found nearly 30% of stool specimens collected after a race contained more than 3 mg of hemoglobin per gram of stool. In Stewart et al.'s (1984) study, the mean fecal hemoglobin rose from 0.99 mgg"1 before the race to 3.96 mgg" 1 after racing, with 20 of the 24 runners showing increased levels (p < 0.01). Seven of these runners had equivalent iron losses of more than 2 mg. The significance of gastrointestinal bleeding depends on the quantity of blood loss and the frequency of occurrence (Buckman, 1984). It appears that bleeding is associated with the degree of 76 running stress. McMahon et al. (1984) noted a possible relationship between bleeding and effort. Gastrointestinal bleeding may thus be a phenomenon of distance racing or very strenuous training and not distance training per se (Eicher, 1985). In most runners, the bleeding resolves within 72 hours (McMahon, 1984). Exercise associated gastrointestinal bleeding remains to be, though, a potential contributor to iron deficiency (Buckman, 1984). The cause of running induced gastrointestinal blood loss is obscure but is speculated as being due to transient gut ischemia (Porter, 1983; Potera, 1984). and/or a repetitive traumatic jarring effect on the intra-abdominal organs (Buckman, 1984; Porter, 1983). Causes of gastrointestinal bleeding not associated with exercise and their relative frequency can be obtained from a retrospective study on 371 patients with iron deficiency anemia (Bannerman et al., 1964): gastrointestinal hemorrhage 60%, hemorrhoids 10%, salicylate ingestion 8%, hiatus hernia 7%, peptic ulceration 7%, diverticulosis 4%, neoplasm 2%, ulcerative colitis 2%, and miscellaneous causes in 4%. No explanation for the occult gastrointestinal blood loss was found in 16%. In adult males and post-menopausal females, iron deficiency is nearly always due to gastrointestinal bleeding (Bothwell et al., 1979). A daily blood loss of 7-10 ml is sufficient in causing a negative iron balance. Unless the bleeding is caused by hemorrhoids, such bleeding will usually escape notice by the patient. 77 Acetylsalicylic acid is a common drug known to cause both acute and chronic blood loss from the gastrointestinal tract. Corticosteroids, phenylbutazone, indomethacin and other non-steroidal anti-inflammatory agents and anticoagulants can also cause gastrointestinal bleeding. Blood loss due to aspirin usually parallels dosage. Typical losses may be 0.5 ml daily, while patients who take acetylsalicylic acid regularly may lose 2 ml/day and even larger doses (2-6 tablets daily) could result in daily blood losses of more than 5 ml. Salicylates do their damage through increased exfoliation of the gastric mucosal cells, erosive or hemorrhagic gastritis, and inhibition of platelet aggregation which may prolong bleeding. The variety of compound preparations containing acetylsalicylic acid and the frequency in which athletes ingest these and other anti-inflammatory medications makes it apparent that drug use is an important consideration in the diagnosis of iron deficiency in athletes. Increased fecal iron loss could also be due to an elevated secretion of endogenous iron in the bile. This explanation was offered by Ruckman and Sherman (1981) in noting increased fecal iron losses in exercising rats. 78 Iron Loss in Urine The observation of increased hemolysis with exercise has been made by a number of researchers (Davidson, 1964; Ekblom et al., 1977; Lindemann et al., 1978; Puhl and Runyan, 1980; Shiraki et aL, 1974; Yoshimura, 1970; Dufaux et al., 1981; Ehn et al., 1980; Eichner, 1985; Hunding et al., 1981; Magnusson, 1984; and Porter, 1981). A low serum haptoglobin indicates increased intravascular hemolysis. It is debatable whether this contributes to an increased iron loss. Hallburg and Magnusson (1984) did not see an increased iron loss in urine suggesting that hemolysis was slight and that the excretion of heme iron was prevented by the formation of the haptoglobin-hemoglobin complex. This complex prevents excretion of free hemoglobin in the urine. The complex is then taken up by hepatocytes and the iron is recycled. Others (Clement and Sawchuk, 1984; Davidson, 1964, 1969; Siegal et al. 1979; Dufaux, 1981; Eichner, 1985; Hunding et al., 1981; and Porter, 1981) would argue that hemolysis (possibly due to the traumas of increased circulatory rate, increased body temperature, compression of red blood cells by muscular activity or with foot strike in activities such as running, or acute exercise acidosis) causes excessive amounts of free circulating hemoglobin which can be excreted in the urine. This trauma could be 79 compounded in exercise as elevated levels of catecholamines increase both the osmotic and mechanical fragility of erythrocytes (Puhl and Runyan, 1980; Yoshimura, 1970). Hunding et al. (1981) calculated that the amount of free hemoglobin found in the plasma of some of their studies' competition runners was large enough (if not balanced by diet) to effect an iron loss that would eventually (over months) lead to iron deficiency and anemia. Eichner (1985) did follow the hematological adaptations over eighteen months of a healthy long distance runner. He concluded that hemolysis, probably involving small numbers of the oldest red cells, does occur, and even this low-grade runner's hemolysis could create and sustain a negative iron balance. In accordance with the above theories, Eichner found the hemoglobin and haptoglobin levels would increase when mileage was decreased. There is also a possibility that red blood cells could be released in the urine following physical exertion as a result of bladder trauma or bleeding from the kidney pelvis. This may be present as grossly red urine or occur microscopically and undetected (Blacklock, 1977; Boileau et al., 1980; Hayashi et al., 1980). 80 Iron Loss in Sweat There is a relatively minute amount of iron in sweat, but with profuse sweating some researcers believe this is a plausible route for excessive iron losses in endurance athletes. Vellar, (1968), reported an iron content in sweat of up to 40 ug-100 ml" 1. As the sweat production during one day of endurance running may well amount to 1-3 litres, an additional iron loss of 0.4-1.0 mg/day is possible, thereby potentially doubling one's iron loss (Clement and Asmundson, 1982). More recent work by Brune et al. (1986) using a very careful cleaning procedure and repeated consecutive sampling periods of sweat in the sauna found that iron losses in sweat are much lower. Steady state sweat content was only 22.5 u,g iron T1 sweat and this should have marginal effects on total body iron losses. 81 IRON DEFICIENCY WITH ANEMIA The most apparent physiological consequences of iron deficiency are those that can be attributed to anemia. The criteria for anemia in females has been set at hemoglobin (Hb) levels below 12 gdl" 1 (Anderson and Barkve, 1970). In males, the cut-off value for anemia is 13.0 or 14.0 gdl" 1 (Bothwell et al., 1979). Hemoglobin, carried in the red blood cells, transfers oxygen from the lungs to the tissues so that oxidative reactions can take place in the cells. From the cells, hemoglobin carries C 0 2 to the lungs to be exhaled. Both delivery of 0 2 and clearance of C 0 2 can be reduced in an anemic individual. Ohira et al. (1983) noted that although C 0 2 is twenty times more soluble than 0 2 , the red blood cell is essential for carrying or converting to bicarbonate, approximately 90% of the C 0 2 An inability to adequately clear C 0 2 from the working muscle because of a reduced Hb level or red blood cell count could cause a reduced muscle pH, which would limit the metabolic processes of skeletal muscle. One of the major effects of anemia is on the circulatory system where a greatly increased workload is placed on the heart (Anderson and Barkve, 1970). The viscosity of the blood is dependent upon the concentration of red blood cells. Normal blood viscosity is about three times that of water. With severe anemia 82 the blood viscosity may drop to as low as one and one-half times that of water. This causes a decreased peripheral resistance so that far greater than normal quantities of blood are returned to the heart. Cardiac output may increase two or more times than normal (Guyton, 1977). The increased cardiac output allows many of the symptoms of anemia to go unnoticed at rest (Hallberg, 1978). Most individuals with mild iron deficiency anemia have no complaints that are dramatic enough to make them seek medical attention (Dallman, 1982). With exercise, though, the heart is not capable of pumping much greater quantities of blood than it is already pumping. Tissue hypoxia thus develops as circulation cannot meet oxygen demands. Acute cardiac failure may ensue (Guyton, 1977), although less dramatic outcomes are an impairment in work capacity. Work capacity is also reflected in an elevated heart rate response during submaximal work, increased post exercise lactate levels, and a prolonged length of recovery (Anderson and Barkve, 1970). There is a significant correlation between hemoglobin concentration and work capacity (Anderson and Barkve, 1970; Dallman, 1982; Edgerton et al., 1972; Gardner et al., 1977; Ohira et al. 1983; Viteri and Torun, 1974). The greater the severity of anemia in humans and rats, the greater the decrement in work capacity (Ohira et al., 1983). The hemoglobin levels need not change markedly before there is a significant change in physical work tolerance (Gardner et al., 1977). 83 In studying female Sri Lankan tea estate labourers with varying hemoglobin levels, Gardner (1977) found that subjects with Hb concentrations between 11.0 and 11.9 gdl" 1 showed approximately a 20% decrease in their work tolerance on an incremental treadmill test compared to those subjects with Hb concentrations above 13 gd l " 1 . The continuing parallel between Hb levels and work capacity can be seen with higher than normal Hb levels. Ekblom et al. (1972) and Edgerton et al. (1972) have shown that work performance can be immediately improved when the oxygen carrying capacity is simply increased through blood transfusion. Ekblom's data indicated that work performance as measured by V 0 2 max could be improved if the Hb levels were increased above the 13.0 g d l - 1 level through reinfusion. Parr (1984) considers any Hb level below 14 gdl" 1 in females and under 16 gdl" 1 in males to be suboptimal. The hematocrit for trained athletes may not be truly ideal for maximum oxygen transport. Heart rate response during moderate work was examined during the incremental treadmill test in Gardner's (1977) study. There was an approximate 4.7% elevation in heart rate with each 1 gdl" 1 decrease in Hb. Charlton et al. (1977) likewise showed the same relationship between Hb and heart rate during exercise. Post exercise lactate levels are also directly related to the degree of anemia, which indicates the degree of reliance on the anaerobic metabolic pathways. Compensating for this is increased 84 levels of 2-3 diphosphoglycerate in the red blood cells which enhances oxygen release to the tissues (i.e. the Hb-02 dissociation curve shifts to the right) (Gardner et al., 1977; Charlton et al., 1977). The length of recovery is prolonged with anemia. Anderson and Balkve (1970) reported that with manipulated Hb through repeated phlebotomy, a drop of 1.5 gdl" 1 extended the time to restore resting cardiorespiratory functions by 10%. The data on work capacity and iron deficiency anemia can be summarized with the statement than any decrement in function will be proportional to the degree of anemia. 85 IRON DEFICIENCY WITHOUT ANEMIA In the past there has been a pre-occupation with anemia in the study of iron deficiency. Anemia is just part of the iron deficiency picture. This portion of the review will first establish the non-hematological effects of iron deficiency and then address the central question as to where in the stages of iron deficiency these tissue abnormalities occur. The negative health effects of iron deficiency anemia can be partly ascribed to anemia and the impaired delivery of oxygen to the working muscles and partly to a deficiency of iron containing compounds, especially enzymes, in various sites (Dallman et al., 1978). A wide range of tissue abnormalities have been described in both iron deficient patients and experimental animals. The general assumption is that these occur as a result of defective synthesis of iron enzymes. The total iron content of tissue compounds is small but their distribution and specific metabolic functions are large. Heme iron compounds in muscles include myoglobin, the cytochromes, catalase and peroxidase. Cytochrome a, b, and c are located within mitochondria and are responsible for the oxidative production of energy (ATP). Other cytochromes are present in the membranous structure of the endoplasmic reticulum 86 and these assist functions such as oxidative degradation of drugs and endogenous substances (cytochrome P450 in the liver) or protein synthesis (cytochrome b5). Non heme iron compounds include iron-sulphur proteins of the electron transport system in the mitochondria. Examples of these are NADH dehydrogenase, succinic dehydrogenase and xanthine oxidase. There are also enzymes such as aconitase of the tricarboxylic acid cycle and tryptophan pyrrolase which do not contain iron but require the metal as a cofactor. With iron therapy to iron deficient, anemic humans and rats, Ohira (1978,79) noted an improvement in physical work capacity and a reduction in heart rate at a given workload before a significant rise in hemoglobin, suggesting an effect unrelated to the elevation in hemoglobin that normally occurs. Finch et al., (1976, 1979), Davies et al., (1982), Mackler, (1984) and McLane et al., ((1981) have proven that in rats iron deficiency results in defective oxidative energy metabolism. Finch et al. (1976) were able to dissociate the effects of tissue iron deficiency on work performance from those of anemia. Iron deficient rats had their anemia corrected with exchange transfusion of erythrocytes. Even with their anemia corrected these animals had an impaired work performance. Davies et al. (1982), using a similar protocol, found that the recovery of the diminished V 0 2 max paralleled the rate of reversal of the anemia 87 but the recovery of endurance lagged behind, supposedly following the repair of muscle oxidative capacity. Davies et al. (1982) and Dallman (1982) thus hypothesized that the capacity to deliver 0 2 is closely related to the peak workload that can be attained in a brief, intense exercise (V0 2 max). On the other hand, the capacity to utilize 0 2 for oxidative phosphorylation is critical to the ability to perform submaximal exercise for prolonged periods (endurance). In an attempt to pinpoint the biochemical locus for this impairment to endurance, Finch et al. (1976) monitored both running ability and various enzyme activities as iron deficient rats underwent iron therapy. In the rats Finch studied, it appeared as though mitochondrial a-glycerophosphate dehydrogenase (a-GPDH) may be the limiting factor to performance as iron therapy brought parallel increases in the rate of phosphorylation with a-glycerophosphate as substrate and the recovery in work performance. The effect of lowered a-GPDH is two-fold: one, involving oxidative phsophorylation within the mitochondria and the other a coupling with cytoplasmic NADH to regenerate cytoplasmic NAD. Excess lactate could result from conversion of pyruvate to lactate as a means of regenerating NAD and could also come from impaired mitochondrial phosphorylation with accumulation of pyruvate. Lactic acidosis causes impairment of muscle work (Finch et al., 1979). 88 Myoglobin and the cytochromes were excluded as a cause of muscle dysfunction by Finch et al. (1976) because their concentrations, although decreased, did not improve with iron therapy during the time required to normalize running. Other investigators do not agree with this interpretation. McLane et al. (1981) re-evaluated Finch's studies, questioning in particular the role of a-GPDH and the apparently negligible effects that depressed respiratory compounds myoglobin and the cytochromes had on fatigue resistance. According to Mclane et al., a-GPDH activity is secondary to the malate-aspartate shuttle in transporting reducing equivalents into the mitochondria. Results did not support the hypothesis that a-glycerophosphate dehydrogenase is a key enzyme of energy metabolism in mammalian muscle. Instead Mclane et al. (1981) concluded that, "iron deficiency results in a decrease in mitochondrial enzymes in skeletal muscle with a decline in the maximum ability to utilize oxygen. This reduction in the capacity for aerobic metabolism is associated with an increased susceptibility to fatigue as a more rapid decline in contractile force and a greater increase in lactate concentration during contractile activities." The search for a biochemical locus of a rate limiting step to work capacity in iron deficiency flounders due to the complexity of iron's effects. Isolated iron enzyme defects can affect various tissues and there is no consistent pattern thoughout the body (Jacobs, 1977). Blayney et al. (1976) showed that in simultaneous 89 studies of myocardium and liver in iron deficient rats there was a depletion of myocardial cytochromes (and thus impaired activity thoughout the electron transport pathway), but in the liver only succinate-cytochrome reductase activity was affected. In skeletal muscle, myoglobin concentration may be reduced by one-half though the proportional reduction of myocardial myoglobin is much less (Dallman and Schwartz, 1965; Hagler et al., 1981). Similarly, in skeletal muscle the cytochromes become more depleted than in cardiac muscle which has an increased obligatory workload with anemia (Dallman, 1978). Within individual tissue and subcellular compartments, some compounds are affected far more than others. In the liver parenchymal cell, the mitochondrial cytochromes are affected more than the cytochromes in the endoplasmic reticulum, cytochrome P450 is more readily depleted than cytochrome b5 (Dallman, 1978). There is a preferential utilization of iron, not only between tissues but within tissues and tissue specific adaptive responses to iron deficiency (Hagler et al., 1981). The identification of a rate limiting step in the skeletal muscle may be dependent upon the intensity required of the muscle, making the locus fiber type specific (Mackler et al., 1984; Mclane et al. 1981). Mackler et al. (1984) found the oxidative energy production by mitochondria from red and intermediate muscle from iron deficient rats to be markedly reduced, with production of ATP greatly decreased. According to McLane et al. (1981), it is the 90 intermediate (red, fast-twitch) muscle fibers which are most affected. The a-glycerophosphate shuttle is not affected in white muscle although ATP formation by NADH and succinate is markedly reduced (Mackler et al. , 1984). The workload of individual tissues is an important factor in determining where the effect of iron deficiency is greatest. Also important is the rate of tissue growth, turnover of specific iron compounds, time of onset during development and the degree of anemia (Dallman, 1978). These factors may explain why some tissues and iron compounds are vulnerable to iron deficiency, whereas others are quite resistant. It also explains why a range of exercise tests is warranted in studying the effects of tissue level iron deficiency. Apart from a drop in work capacity through a depletion of certain enzymes, iron deficiency will affect cell morphology and growth which may also compromise exercise tolerance. In addition, iron deficiency can affect behavior, catecholamines, thermogenesis (Dallman, 1982), cellular immunity and intestinal function (Siimes et al., 1980). It is important to determine the significance of prelatent/latent iron deficiency in order to evaluate the extent to which its detection, treatment and prevention deserve a high priority. The significance of low iron stores in the non anemic individual was previously believed to be solely a predisposition of the individual to a more severe anemia. This alone has relevance 91 to athletes who may have an increased iron requirement. Without stores, a sudden loss of blood cannot be replaced at a rate greater than 3 mg iron per day, which can be obtained from the diet. With adequate stores, though, the replacement rate may be 25-50 mg iron per day (Bothwell et al., 1979). The vulnerability of the situation is illustrated by the fact that women who have no iron stores cannot go through pregnancy without becoming anemic (Bothwell et al., 1979). The hypothesis that prelatent/latent iron deficiency affects performance is dependent on establishing the sequential changes in the development of iron deficiency. Stage Serum fer r i t in (np/ml) Total Iron (ug/dl) UIBC (ug/dl) Transferrir Saturation <*) Hemo-qlobin (q/dl) Normal 20-160 40-130 105-282 20 -55 12-16 Stage 1: Iron depletion from stores 1-20 40-130 105-282 20 -55 12-16 Stage 2: Iron-deficient 1-20 <60 >280 <20 12-16 erythropoisis Stage 3: Iron-deficiency 1-20 <40 >280 <20 <12 anemia Figure #3 Sequential changes (top to bottom) in the development of iron deficiency. (Values for females.) 92 It can be seen that iron restriction first affects iron stores, then transferrin saturation and lastly hemoglobin production. Not shown in the preceding diagram is the stage at which iron loss affects tissue enzymes or myoglobin. Clouding the issue of iron status evaluation are the limitations of the commonly used indices. The diagnosis of iron deficiency is based on indirect methods of measurement. The wide variation in the individual variables (eg. ferritin, serum iron, transferrin saturation, bone marrow iron) even in apparently healthy persons, combined with the fact that they form part of kinetic systems which are influenced by many factors other than the availability of iron, means that definitions of the term iron deficiency based on these variables must be taken with considerable caution (Ericsson, 1970). For example, physiological variation in the concentration of hemoglobin is so great that only about half of the individuals with iron deficient erythropoiesis are identified by recognizable anemia (Cook et al., 1976). As well, Finch (1979) notes that with iron deficient rats, the dissociation between pre-exchange hemoglobin concentration and work performance indicates that hemoglobin concentration per se is an unreliable indicator of the severity of tissue iron deficiency. Serum ferritin measurements have also been put in question. At the high end of the scale, serum ferritin levels can be elevated 93 in clinical situations not related to iron stores. These situations include inflamation, chronic liver disease, and malignancy. Magnusson et al. (1984) re-evaluated the interpretation of reduced serum ferritin values in runners. Their theory was that athletes are not truly iron deficient, but that there is a shift in red cell catabolism from the reticuloendothelial system to the hepatocytes. The hemoglobin-haptoglobin complex formed due to intravascular hemolysis is taken up by hepatocytes. The basis of their argument was the finding in their study that although hematocrit, serum iron, transferrin saturation and serum ferritin values were low in athletes compared to controls; sideroblast counts, red cell indices and red cell protoporphyrin were normal, indicating that lack of iron had not limited erythropoiesis. To confirm this theory a measurement of hepatocyte iron (liver biopsy) would be necessary. The present investigation has a direct bearing on Magnusson et al's theory. If iron supplementation had improved performance of prelatent/latent iron deficient athletes, it would indicate a true iron deficiency existed. A significant difference in performance could not be found and by this measure Magnusson et al's theory could not be disproven. On the other hand, the fact that 3 of the 21 placebo treated subjects dropped into anemia is indicative of a true iron deficiency in runners. Similar evidence that iron supplementation will help prelatent/latent iron deficient athletes is scant (Pate et al., 1979; Vellar and Hermansen, 1971; Celsing et 94 al., 1986; Schoene et al., 1983). The only study to date to measure human muscle enzyme activities when the body iron stores were severely diminished or depleted was that of Celsing et al. (1986). These authors used repeated venesections and reinfusion to induce an artificial latent iron deficiency; they found that this state did not affect endurance, nor the maximal activities of various enzymes in human skeletal muscle. The findings of Celsing et al. are in apparent conflict with the previous rat studies of Davies et al. (1982) and Finch et al. (1976,79). One possible reason for the discrepancy could be that extrapolation of findings from weanling rats to adult humans is not accurate. As well, the rats were rendered severely iron deficient with a low iron diet while Celsing et al. used less extreme experimental conditions (repeated venesections) with his nine male subjects. Mean values of cytochrome oxidase activity, the only iron dependent enzyme investigated by Celsing et al. (1986), were slightly lower after retransfusion than in the control state, but this difference did not reach significance (3.6 ±1.7 vs. 4.4 ±1.8 u,mol 0 2 ( m i n g wet weight) - 1 ) . Certain symptoms which occur in other chronic anemias are often observed in iron deficient subjects in the absence of anemia. These nonspecific syptoms include headache, fatigue, heartburn, changes in appetite, vasomotor disturbances, muscular cramping, dyspnea, palpitation, menorrhagia and diminished work performance (Conrad and Barton, 1981). Most investigators have 95 attributed these functional disturbances to decreases in iron containing enzymes in various body tissues, although a direct cause and effect relationship has been difficult to establish (Conrad and Barton, 1981). Wood and Elwood (1966) found no significant difference between treated (iron supplemented) and control (placebo) patients in regards to influence upon these symptoms. Repeated measurement of work capacity would have aided the study of Wood and Elwood (1966). There may be a difference between clinical symptoms and physical work capacity as reference symptoms. There is higher precision in the determination of physical work capacity. Ericsson (1970) reported a greater increase in physical work capacity (based on bicycle ergometer heart rate measurements) with iron treatment (120 mg/day) in normal elderly subjects compared to controls of similar age. This significant increase was 4% greater than controls in the men and 12% greater in the women. Interestingly, this increase in performance was not matched by an increase in hemoglobin levels. In the placebo (control) group, a relationship approaching significance (r=.493; 0.01 < p < 0.05) was found between iron storage (stainable iron in the bone marrow) and work capacity. In other words, 25% of the variance in work capacity can be explained by the variance in iron storage. In the control group, physical work capacity increased in some subjects and decreased in others. Those control subjects who initially had greater iron depots were in an advantageous 96 position for improving work capacity. In the presence of such a relationship, it was also expected and found (r=0.484;0.01 < p < 0.05) that the amount of stainable bone marrow iron would have decreased in those whose physical work capacity had increased. Depot iron had less influence in the treatment group with high iron intake. Ericsson (1970) concluded that in apparently healthy individuals the increase in physical work capacity during moderate training is related to the availability of iron. Studies by Ohira et al. (1978, 1979, 1981) indicated that there are beneficial effects of iron treatment that cannot be related to changes in Hb. Ohira et al. (1981) noted from unpublished data that even normal rats absorb greater amounts of 5 9 F e in cardiac muscle within four days after oral iron treatment. Using human subjects with a wide range of hemoglobin (6.4-14.1 gdl" 1) and serum iron levels (28.5-98 ugdl" 1 ) , Ohira et al. (1981) showed that this effect is more significant in iron deficient groups than in normal iron groups, even when both groups had similar Hb (i.e. 11.9 vs. 11.8 gdl" 1). Work capacity was measured before and after one week of iron treatment (Imferon IV) in the study by Ohira et al. (1981). There was a tendency for all groups to increase their work time, although this could merely reflect some degree of adaptation to the treadmill tests. The post exercise venous blood lactate data of Ohira et al. (1981) suggests a non Hb related effect. Three groups of subjects: high Hb-high serum iron, marginal Hb-high serum iron and marginal Hb-low serum iron, had almost identical 97 work capacities, thereby eliminating work time as a variable in affecting blood lactate. The coefficient of correlation (r=0.41; p < 0.05) between serum iron and lactate levels after exercise suggested that the lower the serum iron, the higher the blood lactate after exercise. Looking at the post exercise lactate levels of the two marginal Hb groups (11.9 and 11.8 gdl" 1), but one with low serum iron and the other with high serum iron values, significantly higher post exercise blood lactate occurred in the low serum iron group even though intensity and duration of exercise were similar. Changes in mean heart rate at a given heavy workload showed the same relationship. Schoene et al. (1983) examined the effects of iron therapy on work capacity and exercise induced lactate production in trained, mildly iron deficient female athletes. Although exercise performance on a cycle ergometer with a progressive work-exercise protocol did not change with two weeks of oral iron therapy, lactate production after exercise decreased significantly. Two weeks of ferrous sulfate supplementation (300 mg/day) brought hemoglobin values from an average 12.2 gdl" 1 to 12.7(p < 0.05), and ferritin from 10.0 to 22.1 ngml" 1 (p < 0.002). Blood lactate levels at maximum exercise in this iron deficient group dropped from 10.3 u,moll"1 before therapy to 8.42 after therapy (p < 0.03). The non-iron deficient controls had no significant changes for any of these parameters. The results suggested that work capacity might have improved. A protocol entailing an endurance 98 test instead of a short, intensive test might have illuminated this possibility (Schoene et al . , 1983). Continuation of iron supplementation for two months may also have led to a significant improvement in exercise performance as this period of time may be needed for replenishment of iron stores after the hemoglobin concentration has reached normal, non-anemic levels (Jacobs, 1985; Valberg, 1980). The rapid rise in ferritin values noted by Schoene et al. (1983) probably does not represent increased iron storage levels, but instead a transient increase in ferritin synthesis (Jacobs, 1985). Because evidence concerning the effects of prelatent/latent iron deficiency is limited and the complexity of iron usage and depletion is immense, the significance of iron deficiency without anemia is speculative. Jacobs (1985) states that iron deficiency anemia, resulting from an impaired iron supply for erythropoiesis, occurs at the same time as the iron supply to other organs is reduced. Dallman (1978) acknowledges that deficiencies of tissue iron compounds generally occur in conjunction with anemia, but notes that they may also be observed in its absence. Clement (1981) speculates that the non-anemic iron deficient symptoms reported are an indication that iron deficiency will impair initially certain enzymes (and perhaps myoglobin) and later hemoglobin. 99 Appendix B Definitions WORK CAPACITY: Alactic power - is a measure of explosive power, relying mainly on stored adenosine triphosphate (ATP) and phosphocreatine (PC) in muscle without a significant build up of lactic acid. On the Wingate bicycle ergometer test it is the maximal power output observed averaged over a 5 second period and is expressed in watts-kg"1 body wt. Lactacid power - is a measure of power output when lactic acid tolerance is maximally stressed. On the cycle ergometer test it is the average power output over 30 seconds with units being watts-kg"1 body wt. Lactacid capacity - is the total energy output during the anaerobic speed test and is expressed as time to exhaustion in seconds. Anaerobic threshold (AT) - is the inflection point of a blood lactate curve in an incremental exercise test. It can be expressed as a %v"0 2 max or, as in this study, a treadmill speed. Instead of blood lactate sampling, the protocol utilized gas exchange variables to indirectly determine the anaerobic threshold and technically this inflection point is called the ventilatory threshold. 100 V02max. - indicates the highest oxygen uptake (mlkg" 1min" 1) an individual can attain during physical work while breathing air at sea level. On the progressive workload treadmill test this usually occurs close to the point of exhaustion. Maximal treadmill velocity - is the highest treadmill velocity (mileshr" 1) achieved during the incremental treadmill test. IRON STATUS: Serum Ferritin: is a sensitive index of the earliest stage of iron deficiency-depletion of body iron reserves (Valberg, 1980). One ng/ml of serum ferritin corresponds to approximately 5-9 mg of storage iron. Levels below 20 ng-ml"1 are associated with absent bone marrow iron (Pakarinen, 1980). Levels up to 64 ng-ml" 1 may still represent an iron deficient state (Heinrich, 1977). Hemoglobin (Hb): is an iron containing conjugated protein found in red blood cells and capable of transporting oxygen and carbon dioxide to and from the working tissues. Total Iron: (also called plasma iron and serum iron) is useful in the clinical evaluation of iron deficient erythropoiesis. Exhaustion of the iron stores affects the supply to the plasma pool and hence to the erythroid bone marrow (Bothwell et al.,1979). 101 Unsaturated Iron Binding Capacity (UIBC): is also a measure of iron deficient erythropoiesis. It is determined by incubating a known ferrous standard with serum to saturate the available binding sites on serum transferrin. UIBC is the amount of iron that can be taken up by the transferrin. This figure is added to the total iron to yield the total iron binding capacity (TIBC). % Transferrin Saturation: is a function of both the total iron and the transferrin concentration. A saturation of less than 16% has been shown to be inadequate to meet basal erythropoietic needs (Bothwell et al , 1979). The figure is arrived at mathematically from the values above; total iron/TIBC. IRON DEFICIENCY: A state which develops if the loss of iron is greater than the amount obtained (Pakarinen, 1980). Iron deficiency is commonly divided into three stages: (a) prelatent, (b) latent, and (c) manifest (Clement and Sawchuk, 1984). a) Prelatent iron deficiency: is the earliest stage of iron deficiency characterized only by a decrease in storage iron (and decreased serum ferritin) and increased intestinal absorption (Strauzenberg et al., 1981). 102 b) Latent iron deficiency: in addition to the indices of prelatent iron deficiency, the levels of total serum iron is decreased, the unsaturated iron binding capacity is increased, and the % transferrin saturation is decreased (Strauzenberg et al., 1981). The lack of reliability in these measurements (Pakarinen, 1980) makes differentiation between prelatent and latent iron deficiency difficult. For this study, these measurements did not affect the screening procedures nor were they utilized in the stated statistical analysis. Instead, they provided back-up information on each subject, which may help in the interpretation of the data. The criterion for prelatent/latent iron deficiency was serum ferritin levels below 20 ngml" 1 . Levels below this are associated with absent bone marrow iron (Pakarinen, 1983; cited in Clement and Sawchuk,1984). c) Manifest iron deficiency: refers to an advanced stage of iron deficiency characterized by a drop in hemoglobin levels (Clement and Sawchuk, 1984). In this study, the criterion will be levels below 12 gdl" 1 for women (Williamson, 1981). Manifest iron deficiency will also be referred to as anemia. 103 PHYSICAL ACTIVITY: This criterion for inclusion in the study was defined as regular (three or more times per week) participation in relatively intense (three-quarters maximal effort) physical activity of at least 120 minutes total duration per week. For most subjects this involved running. 104 Appendix C L i m i t a t i o n s : SUBJECT SELECTION: The subjects selected were physically active but mostly non-elite, as compliance to the treatment procedures may have sacrificed an optimal training/diet regimen. SAMPLE SIZE: Primarily for practical reasons, the sample size was set at 40 completed subjects. Calculations using work capacity data from a similar population supported a size of 40 as being reasonable in obtaining a statistically significant result. USE OF SERUM FERRITIN: A serum ferritin assay allows accurate assessment of prelatent iron deficiency but not differentiation between prelatent and latent iron deficiency. USE OF EIGHT WEEK TREATMENT PERIOD: While this length of treatment was felt necessary to induce a significant repletion of iron stores, the number of dropouts (due to various uncontrollable circumstances) increases with the length of study. 105 PHYSICAL FACTORS: Training, diet, menstrual blood loss and other physical factors which may affect the testing were not rigidly controlled or precisely measured. Instructions were given to the subjects regarding exercise and diet prior to blood tests and work capacity tests. As well, they were instructed to maintain their typical training and diet regimen during the treatment period. A log book of training, menstrual blood loss, and any other physical factors which may have influenced the testing was completed by each subject. A three day dietary analysis prior to both the pre-test and post-test aided in monitoring diet (See Appendix G). 106 Appendix D D e l i m i t a t i o n s : This study was delimited to physically active females, age 18 to 40, with prelatent/latent iron deficiency. Subjects received placebo or iron supplementation (320 mg ferrous sulfate = 100 mg elemental iron taken as SLOW-Fe® one tablet BID as tolerated) for a treatment duration of eight weeks. The only blood parameters measured were serum ferritin, hemoglobin, total iron, UIBC, and % transferrin saturation. The only physical performance measures were the Wingate bicycle ergometer test, anaerobic speed test, anaerobic threshold, and V 0 2 max. and max. velocity. The subjects selected were all volunteers who resided in the greater Vancouver area. They underwent a brief medical examination and history review prior to enrollment in the study, including temperature, stool and urine examination, blood pressure and heart rate determinations. 107 Appendix E THE UNIVERSITY OF BRITISH COLUMBIA FACULTY OF MEDICINE DIVISION OF SPORTS MEDICINE John Owen Pavilion • Ian Newhouse Subject Consent Form The purpose of this research is to assess the effects of prelatent/latent iron deficiency in physically active females (see attached abstract). As a subject you will undergo the following tests: (requiring approx. 10 hours). 1. anthropometric tests: underwater weighing for % body fat, height, weight. 2. urinalysis: for hemocyturia (blood loss in urine). 3. stool analysis: for occult blood loss. 4. blood tests: samples will be analysed for serum ferritin, hemoglobin, total iron, unsaturated iron binding capacity, and % transferrin saturation. 5. Wingate anaerobic power test: a 30 second bicycle ergometer test. 6. Anaerobic speed test: a treadmill run (at grade of 20% and speed of 7 mph) to exhaustion (usually between 30 seconds and 1 minute). 7. A progressive workload treadmill test (a treadmill run with velocity increasing every minute up to exhaustion). Continuous sampling of expired gases will allow measurement of VC^max. Testing will be be performed pre and post treatment. All testing will be conducted by skilled technicians. 108 The treatment will be 8 weeks of either iron supplementation (a daily dosage of 320 mg ferrous sulfate (100 mg elemental iron) taken as Slow Fe® twice a day, or matching placebo tablets taken as one tablet twice a day). Placebo subjects will receive iron supplementation upon completion of the post tests. A medical examination will be provided prior to entry into the study. You will be asked to record all training, an estimate of menstrual blood loss, and any other factors which may influence testing for the duration of the testing and treatment period. As well, a 3 day dietary record will be filled out prior to the pre and post tests. Training and diet should be kept consistent with previous habits. Ingestion of anti-inflamatories or medication which may cause acute or chronic blood loss from the upper gastro intestinal tract will not be allowed. The chief agent in this category is acetylsalicylic acid (aspirin). Although you will be exercising to the point of temporary exhaustion, and taking large doses of iron supplementation, there is very little risk involved if you are a healthy, mildly iron deficient individual. The discomfort associated with testing and treatment, apart from temporary exhaustion in the physiological tests, should be mild. At the point of exhaustion on the treadmill tests you will grasp the safety rail and the treadmill will be stopped. If you stumble or fall while running, the treadmill will be stopped and the consequences would be similar to if you had fallen on a track. Blood sampling will be done by skilled technicians. The amount of blood drawn will be small and there will be little discomfort with the procedure. There may be slight bruising at the point of puncture. Side effects of iron supplementation have been reported to affect a small percentage of the population and these include nausea, constipation, and abdominal pain (Gomez & Gomez, 1969). Publication of results will not reveal subject identity as subjectsd will be referenced by number. -2-109 I have read and understand the above explanation of the purpose and procedures for this test and agree to participate. I also understand that I am free to withdraw my consent at any time. Signature Witness Date -3-110 Appendix F THE UNIVERSITY OF BRITISH COLUMBIA FACULTY OF MEDICINE DIVISION OF SPORTS MEDICINE John Owen Pavilion Ian Newhouse Subject Consent Form: Muscle Biopsy This project is designed to investigate the response of iron containing skeletal muscle enzymes to iron supplementation. To do this you will be required to have two needle biopsies: one will be taken at the beginning of the study and the second after 8 weeks of iron supplementation. There are minimal risks associated with this procedure. A 1/2cm incision will be made under local anaesthetic in the lateral thigh and a needle inserted in the vastus lateralis. A small portion of muscle (100mg) will be taken with the needle. The location is free from underlying deep structures such as arteries, veins, or nerves. Complications, which are rare, include infection, bleeding, hematoma or an allergic reaction to the local anaesthetic. All procedures will be performed using sterile technique by experienced physicians. Pressurized bandaging will prevent post operative bleeding. There will be some discomfort associated with this procedure at the time of needle insertion but most subjects report this as mild. You will be able to resume your normal activities immediately. All information derived from this project is confidential. In reporting the data, names of the subjects will not be used. Information about your results is available to you and these can be discussed with one of the investigators. We will be happy to answer any questions that you have concerning the project and you are free to withdraw from the project at any time without prejudice to future care. I have read the above consent form and understand fully the procedure and risks involved. I consent to participate in this research project and acknowledge receipt of a copy of this form. Witness Subject Date 111 Appendix G Dietary Analysis A computerized nutrition analysis was provided by the YMCA. Three days of dietary intake and physical activity were recorded by each subject. The instruction sheets and recording forms are shown on the following pages (pp 112-115). COMPUTERIZED NUTRITION ANALYSIS: Personal Form YMCA This evaluation is only as accurate as the information you provide on these forms. To obtain the best results, follow the instructions carefully and complete ALL sections as accurately as possible. Complete: 1. ONE green personal form only. 2. ONE white computer form FOR EACH DAY to be analyzed. SECTION A: PERSONAL DATA No. OF DAYS TO BE ANALYZED. 1 10 7 NAME ADDRESS GROUP. PHONE NUMBER SEX: MALE • FEMALE • HEIGHT (without shoes) .inches OR. lbs. OR _ .cm WEIGHT (without clothes) FRAME SIZE: SmallD MediumD LargeD BIRTH DATE day month year -kg DO YOU WISH TO LOSE WEIGHT? YESD NOD ARE YOU ON A SPECIAL DIET? YESD. NOD ARE YOU TAKING VITAMIN/MINERAL SUPPLEMENTS? Y E S D NOD ARE YOU PREGNANT? YESD NOD AREY0U NURSING? Y E S D NOD SECTION B: FOOD RECORD t. Use a worksheet (see sample) to list EVERYTHING YOU ftTE OR DRANK for EACH DAY to be analyzed. Be sure to include: a) ALL FOOD AND DRINKS, for example: snacks, sugar and cream in coffee, mayonnaise in a sandwich, sauce on ... vegetables, candies, soft drinks, wine, etc. • b) THE TIME OF DAY these foods were consumed. c) THE AMOUNT OF FOOD that you ate (ounces, slices, cups, teaspoons, etc., in whole or decimal numbers). 2. On the WHITE COMPUTER SHEET, enter the NUMBER OF SERVINGS of ALL foods and drinks ACCORDING TO LISTED SERVING SIZE. a) Break down all combination foods. (For example, cheese omelette in food record above.) TIME OF DAY? HOW MUCH? FOOD OR DRINK? 10-16 am IT-CO r\oov\ lJi "U»p • . coffee. * b) NOTE THE SERVING SIZE c) WRITE the number of servings you had in the CORRECT TIME SLOT FOODS EATEN Serving S i ze^ . ft N 0 Q ^ R V I N G ^ CODE Office Use Only ^5 am # 11 anrlL 11 am -5 am MEAT, FISH, EGG OR POULTRY EGG 1 1 oz. 2. 10129 Cheese: Cheddar I.S 10009 Vegetable Oil 1 tbsp. as 40518 Program originally developed by Action B.C. Copyrighted 1983. This form may be reproduced with the consent of the Vancouver YMCA, 955 Burrard Street, Vancouver, B.C. V6Z1Y2 (604)681-0221 SECTION C: ACTIVITY RECORD 1. Use a worksheet (see sample) to RECORD YOUR ACTIVITY FOR EACH DAY, (that is, tor 24 HOURS.) Activity levels explained below. 2. TOTAL THE NUMBER OF HOURS SPENT AT EACH LEVEL. The total must EQUAL 24 hours. TIME OF DAY LENGTH OF TIME ACTIVITY LEVEL* 0 . - O 5 - 0 M 5 '/2 IAX. /CP wi/H. 5 3 / 4 rvK office wov'l* z 2. 3 2 WRITE THE TOTAL HOURS spent at EACH LEVEL in the ACTIVITY section at the top of the WHITE COMPUTER FORM. ACTIVITY: Hours at each level 1 e.s ? \2-0 , 9.0 4 0.5 5. 0 = 24.0 Hours Sleeping/Resting Sitting/Standing Light Activity Active Very Active •EXAMPLES OF LEVELS OF ACTIVITY LEVEL 1 SLEEPING, resting LEVEL 2 Sitting, eating, watching TV. classwork, office work, driving, playing quietly, playing a musical instrument, sailing, STANDING, personal toilet, teaching, laboratory work, HOUSEWORK, SHOPPING, cooking, bowling, pool LEVEL 3 (Count only the TIME that you are ACTIVELY PARTICIPATING) BRISK WALKING (2-3m.p.h.), BICYCLING (5.5m.p.h.), farming, gardening, mechanical work, electrical trades, house-painting, restaurant trades, loading/stacking, carpentry, dancing, 4 baseball, weight training, golf, l^y-W^-i RECREATIONAL swimming, table tennis tennis, volleyball, canoeing or rowing LEVELS 4 and 5: Activities at these levels should cause you to raise * your HEART RATE, to BREATHE HARD, to PERSPIRE, to use ALL YOUR EFFORT. Count only the TIME that you are ACTIVELY PARTICIPATING. / LEVEL 4 WALKING/JOGGING (4.5-5m.p.h.), hiking, BICYCLING (10m.p.h.), digging, skipping, downhill skiing, recreational basketball, badminton or skating, canoeing (4m.p.h.), competitive table tennis or tennis LEVEL 5 RUNNING (more than 5m.p.h), football, hockey, soccer, lacrosse, bicyling (13m.p.h.), COMPETITIVE badminton, volley ball or basketball, racquet sports, vigorous skating or rowing, cross-country skiing, martial arts, swimming (45 yards/min. or more CHECK THAT YOU HAVE COMPLETED ALL SECTIONS OF THE GREEN PERSONAL FORM AND WHITE COMPUTER FORM(S) BEFORE RETURNING THEM. COMPUTERIZED NUTRITION ANALYSIS: Computer Form N A M E . DAT.E_ Day Month ACTIVITY: Hours at each level 1. Year GROUP EVALUATION DAY N U M B E R . 1 to 7 S leep ing/Res t ing S i t t ing/Stand ing Light Act iv i ty Act ive .5._ Very Act ive 24.0 Hours FOODS EA1EN Serv ing S ize NO 5 am • It am OF SERVI 11 am -5 p m NGS 5 pm • 5 am CODE Ol l ice Use Only FOODS EAIF.N Serv ing S ize NO 5 am -11 am OF SEItVI 11 am • 5 p m NGS 5 pm -5 am CODE Ol l ice Use Only MILK AND MILK PRODUCTS F R E S H : apple 1 m e d . 13 CHEESE : camembert , so i l 1 07. 10007 np i i cots (3). mango ('/<) 1 seiv . 30 Cheddar, hard 1 oz 10009 banana 1 med . 141 cottage cheese V? cup 10015 perr ies : b l a ckbe r r i e s , b lueberr ies , raspberr ies Vi cup 1849 cream cheese 1 oz 10017 Chen ics 1 cup 663 p tocessed (1 s l ice ) , sptead (2 tbsp. ) 1 serv. 10042 g iapes 1 cup 1085 Sw i s s or gouda 1 oz 10040 grapelrui t Vi 1053 Ice cream Vi cup 10061 me lons : cantaloupe (orange) Vi 1358 Ice milk Vi cup 10164 honeydew (pale green) 2 " wedge 1360 Instant breakfast : with milk 1 cup 4046 watermelon 1 0 " x 1 " sl ice 2424 M I LK : chocolate condensed , sweetened, canned 1 cup 1 tbsp . 10103 10095 orange, tangerine peach , nectar ine ('/:), papaya (V?_cyd) _ l i n e d , 1 serv. 1420 1479 evaporated, who le , canned t tbsp . 11096 pear 1 1502 s k i m , buttermilk 1 cup 10085 pineapple V i " s l ice 1611 ?."/. l c up 10079 p lums . prunes 1 1641 whole , homogenized 1 cup 10077 s t i a w b e m e s Vi cup 2217 M i l k s h a k e 10 oz 10111 DRIED: apr icots 2 ha lves 39 YOGURT: fruit f lavoured 6 oz 10122 dates 2 952 plain 6 oz 10116 prunes 2 1818 BREADS AND CEREALS ra is ins '/< cup 1846 B I SCU ITS : bak ing powder , scone (Z "x i 'A") 1 410 JU ICES (unsweetened) : apple Vi cup 27 B R E A D S : European, d a r k ' 1 s l ice 456 grape Vi cup 1088 trui l b read , rais in 1 s l ice 452 grapefruit Vi cup 1071 ' w h i l e , enr iched t s l ice 461 orange Vi cup 1437 whole wheat , whole grain 1 sl ice 471 pineapple Vi cup 1619 B U N S , ROLLS : p la in , hamburger 1 1902 prune '/; CUp 1821 whole wheat 1 1906 VEGETABLES C E R E A L S : cooked , whole grain Vi cup 1391 COOKED: asparagus 4 s p r s . 47 dry, l l aked , enr iched 1 cup 4005 beans , g reen , yellow Vi cup 183 dry, sh redded , whole grain 1 cup 4033 beets Vi cup 385 dry, sugar coated 1 cup 4037 broccol i 1 cup 484 granola Vi cup 4101 B i u s s e l s s p i o u l s 8 490 C R A C K E R S : plain 4 916 cabbage Vi cup 513 melba toast 2 4063 carrots Vi c up 620 whole grain 2 919 caul i f lower 1 cup 631 M U F F I N S : fruit 1 med . 1345 Chinese g reens , bok choy 1 cup 520 whole gra in , bran 1 m e d . 1346 c o r n , kerne l i'/i c u p ) , cob (1 sm. ) 1 serv. 857 PASTA: macaron i , noodles, spaghett i 1 cup 1300 eggplant Vi cup 987 P A N C A K E S . W A F F L E S : 4 " d iam. 1 1457 mixed vegetables Vi cup 2404 P I E C R U S T : p la in , l o w e r c r u s l wedge 1598 onions Vi cup 1413 RICE: brown Vi cup 1870 parsn ips Vi cup 1474 white , converted, parboi led Vi cup 1874 peas Vi cup 1530 whi te , unenr iched Vi cup .. 1878 potatoes: baked , boi led, mashed 1 med . 1788 M I S C E L L A N E O U S : bran '/< cup 2446 trench d i e s . Iried potatoes 1 cup 1806 wheat germ 1 tbsp . 244 7 sca l loped Vi cup 1795 yeast, b rewer ' s 1 tbsp . ! 2478 sweet, yam 1 med . 2249 FRUITS sp inach Vi cup 2170 C A N N E D : (sweetened) : app lesauce Vi cup s q u a s h : w in te i . yel low Vi cup 2200 apricots V: cup 35 summer, zucch in i Vi cup 2192 cherr ies cup 671 tomatoes, baked , c a n n e d Vi cup 2284 fruit cockta i l peaches V: cup 77 cup 1023 Ml! . ' ! turnip:'. HAW s;.l«.h r i ? . , ^ w , " , ' ' ' ; , v n , , , , ! , K o Vi c up Vi cup 2353 803 pear s Vi c up 150? p i t ied vegetable '/) cup 4059 p lums , prunes Vi c up IB.-. 5 pnfalu. with d ress ing Vi cup 1812 rhubarb Vi cup 1866 avncarln 1 'T 64 others - choose FRESH + 3 tsp. sugar pei serv ing c a n o l s 1 med . 619 CoDvrinht 1983. This form mav be reproduced with the consent of th o re l e i v 1 stalk 637 Vancouver. Y M C A . 955 Burrard Street. Vancouver, (604) 681 -0221 . 3 C. V6Z 1Y2 C ' . icumbei g ieen neppe i . du.j:d 7 s l i ces '/< cup 943 1545 FOODS EATEN Serv ing S ize NO OF SERV INGS CODE FOODS EAI EN Serv ing S ize NO . OF SERV INGS CODE 5 am -t l am 11 am -5 pm 5 pm • 5 am Ol l ice Use Only 5 am -11 am 11 am -5 pm 5 pm -5 am Ol l ice Use Only lettuce, chopped 1 cup 1258 DESSERTS, SWEETS, BAKED GOODS mush i ooms '/< cup 1351 J a m . jelly, hnney. s y m p 1Ibsp. 1148 onions, green 1 t -115 Mola s se s 1 i b s p . 1341 radishes 2 T 8-1-1 Sugar: white, b iown 1 t sp. 2230 sprouts 1 bean al la l la 'A r .up 1IM S w - e i sauce l imping 2 Ibsp 761 tomato 1 med . 2282 CANUV: hard(f i ) caranie ls (3) rnarshmnllows(4 1 serv. 608 JU ICES : tomato, vegetable 'h cup 2288 chewing gum 1 p iece 680 MEAT. FISH, EGG OR POULTRY chocolate bar 1 587 EGG 1 11)1?'.! chocolates 1 594 F ISH: cod. halibut, sole: baked 3 o z 795 CAKE : angel lood 2vr wedge 521 deep Iried 6 oz 1 100 Inn! 2"x " ' I W ' x ' / V 532 l ist i s l i c k s 1 1 (11 / plain, n i l ic ing 2'/r-" s q . 534 tresh water f ish 3 o z •1089 t i c l i . with icing 2'/;" s q . 526 sa lmon, tuna: baked 3 o z 1958 COOKIES: brownies , squares 1 813 canned cup 2324 chocolate chip 1 818 S H E L L F I S H : c lams(5) . oysters(9) I serv. 773 g iano la bats 2 4101 c rab, lobster 3 o z 905 oatmeal 1 4069 shr imp, scal lops 3 o z 20-15 other, n s so i l e i l 1 812 MEAT: beel , vea l : g round, hamburger patlie 3 o z 370 Doughnut. Danish pastry, eclair 1 958 roast 3 o z 2-187 F i u i l c r i sp Vt cup 25 steak 6 oz 310-1 Je l l ied d e s s e i l . jello Vi cup 1032 stew meat 3 o z 353 Fie (3'/,-' wedge), far ! , turnover 1 1566 del i -sty le meats (1 s l ice) 1 oz 1982 Pops ic le. ices 1 1144 lamb: roast, chop 3 o z 1230 PUDDING, e g g c u s t a i d Vi c up 948 liver, organ meal s 3 oz 12G7 milk Vt cup 1826 liver pate 1 oz 19815 rice, fapioca Vi cup 1891 luncheon meat, canned 1 oz 2006 Sherbet Vt cup 10066 pork: bacon (side) 2 s l ices 126 BEVERAGES OR DRINKS ham, hack bacon 3 o z 1769 A I X U M t l l : beer 12 oz 394 roast, chops (2 small) 3 o z 1717 lioueur(1 oz), dessert wine( 1V? oz) 1 400 sausages ( 4 " l ink) 1 2014 liquor 1 Vi oz 398 sparer ibs 6 r ibs 1762 wine, lali le 4 oz 401 weiner (frankfurter) t 1999 Carbonated pop moi i-diel) 1 0oz 404 POULTRY: c h i c k e n ; roast 3 o z 50019 Colfee 1 c up 4051 Iried 3 oz 50008 Frui l flavourer! dr ink (crystals) 1 cup 4043 turkey: roast 3 oz 50166 Hot chocolate (no milk) 1 cup 779 MEAT ALTERNATES Lemonade 1 cup 1252 BEANS , COOKED: kidney, red Vt cup 160 Postuni, coffee subs t i l u l e s 1 cup 4070 l ima Vt cup 177 lea. (il iced, add 2 Isp. sugar) 1 c up 4050 soy Vt cup 2140 SNACKS whi le Vi cup 155 Corn ch ip s 01 s nack s , cheez ies 1 sm. pkg . 4080 Lenti ls '/? cup 125-1 Popcorn (wi lh oil a n d sai l) 1 c up 1655 Split peas Vt C u p 1533 Potato ch ips 1 sm p k g . 1809 NUTS : sa l ted, roasted Vt cup 1-196 Pretzels 1 sm. pkg . 1814 unsa l ted, raw V. cup 1-195 COMBINATION DINNERS Peanut butter 2Ibsp. 1-199 Baked beans w i l h tomato sauce 1 cup 156 Seeds: sunflower, sesame 2 tbsp. 2236 Chil i w i lh beans 1 cup 756 Tofu: bean cu rd 2 o z 21-15 Chop si.iey with mea l 1 cup 762 SOUPS, SAUCES, CONDIMENTS, OTHER Macaron i and cheese 1 cup 1304 SOUPS : broth: consomme, plain 1 cup 2065 Mea l loal 3 oz 2007 with vegetables 1 cup 2 I 08 Mea l pie (1 smal l . 3 W wedge) 1 382 with noodles t cup 2067 P izza with cheese, tomato sauce 6 " wedge 1628 chowder, c l am (no mi lk) t cup 2085 Spaghett i with cheese, tomato sauce 1 cup 2163 cream soups, all (with mi lk) t cup 2090 S lew meal with vegetables 1 cup 371 lenti l, split pea t cup 2097 FATS, CREAMS AND OILS -S A U C E S : gravy % cup -108-1 Duller. 1 pal 1 i sp . 10001 tomato Vi cup 229G C R E A M : coffee, t pkg. 1 tbsp. 10049 wh i le , (if cheese, add V? oz C h e d d a r ) V* cup 2-170 SHIM 1 Ibsp. 10056 Mus ta rd 1 IS|, 1:17:1 •.-.•iiilUi-'d '/< cup 10053 P ick les (2 s l i ces ) , re l i sh (1 thsp.) 1 1 •.;,'> Coltfi.! .-.iril'm-.-r. i n . - a r n -.ubslilufe 1 tsp. 10069 Soy sauce I Ibsp. ? 1 Mi Margar ine 1 i s p . 43130 tomato ketchup 1 Ibsp. V 'melnhl ' - ml 1 Ibsp. 40518 Lemon juice 1 Ibsp S A L A D DRESS INGS, blue cheese 1 Ibsp. 45539 bench , oil and vinegar 1Ibsp. 45120 low 1,-11. d ie l 1 Ibsp. 45023 mayonnaise 1 I b sp. 45018 1 housaud Island 1 Ibsp. I 45017 116 Appendix H Schedules of examinations and tests Visit # 1 2 3 4 Week - 1 0 4 8 Pretreatment Evaluation X 3 Day Dietary Analysis X X Blood Loss Studies X X Hematological Tests X X Patient Consent X Physician Examination X Randomization X Anthropometric Tests X X Physiological Tests X X Dispense Study Medication X X Collect Unused Medication X Dispense Patient Diary* X X Collect Patient Diary X X Study Termination Form X * Patient diary monitored menstrual blood loss, training including type, duration and intensity, use of concomitant medications and study medication. Appendix I Iron Status Raw Data 117 # Ferritin Hemoglobin Iron Total U.I.B.C. % Saturation Norms: : 10-160 na/ml 12-16 a/dl 41-130 iia/dl 105-282 ua/dl 20-55 % Pre Post Pre Post Pre Post Pre Post Pre Post Iron Group 1 8 63 13.5 13.6 166 158 51 2 7 35 12.9 12.6 66 118 384 295 15 29 3 10 14 13.8 13.1 130 87 209 210 38 29 4 18 83 12.9 13.2 63 107 199 164 24 39 5 5 19 12.5 13.3 60 95 348 219 15 30 6 10 29 13.9 14.5 86 387 18 7 8 30 12.6 13.4 114 197 37 8 7 12 13.1 14.1 33 250 12 9 18 26 13.1 13.1 67 296 18 10 14 33 12.9 12.9 53 260 17 11 18 34 12.9 13.3 60 196 278 143 18 58 12 19 40 14.4 13.6 47 45 243 221 16 17 13 11 44 14.7 14.3 103 87 291 235 26 27 14 11 37 13.2 13.5 39 70 309 202 11 26 15 19 27 12.9 13.4 66 76 286 222 19 26 16 14 29 13.7 13.2 69 144 214 96 24 60 17 10 62 13.7 13.4 221 60 208 240 52 20 18 16 75 13.3 13.8 66 71 250 194 21 27 12 12 24 13.9 13.4 48. 124 54 103 47 55 Mean 12.4 37.7 13.4 13.5 79.8 98.5 251.8 195.7 25.1 34.1 Placebo Group 20 15 23 14.5 13.7 69 68 230 198 23 26 21 19 25 13.1 13.8 85 158 307 205 22 44 22 10 8 13.0 12.9 55 44 294 253 16 15 23 8 12 13.0 12.8 137 74 225 263 38 22 24 10 27 13.4 14.3 179 59 164 231 52 20 25 10 17 13.4 13.5 102 233 30 26 15 14 12.2 12.2 133 330 29 27 8 5 12.1 13.4 54 385 12 28 7 6 12.3 12.7 27 385 7 29 17 19 12.6 13.1 84 94 209 277 29 25 30 14 12 13.1 12.8 81 45 274 257 24 14 31 18 21 12.7 12.7 54 81 246 224 18 27 32 8 40 13.7 12.7 16 60 387 276 4 18 33 15 20 13.4 13.6 122 153 44 34 19 15 13.5 12.9 95 25 269 282 26 8 35 11 26 12.6 12.8 112 160 220 181 34 47 36 7 12 12.2 12.9 50 91 312 263 14 26 37 9 8 12.7 13.6 38 28 243 336 14 8 38 13 13 12.2 10.9 116 155 292 241 28 39 39 5 7 12.2 11.1 29 30 389 382 7 7 4Q 1Q 5 12.7 11,2 as 52 272 2ZZ 22 13. Mean 12.2 17.2 13.0 13.1 84.1 70.5 252.4 231.9 25.6 21.4 118 Appendix J Work Capacity Raw Data # Alactic P. Lactacid A.S.T AT- VOoimax Max. , VeL Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post 1 7.7 7.2 6.4 6.4 29.0 29.0 6.5 6.5 45.2 45.4 9.0 8.5 2 7.1 7.9 5.0 6.0 26.0 25.0 8.5 8.0 43.5 46.7 9.5 9.5 3 8.0 9.1 6.4 7.3 21.0 28.5 8.0 7.5 47.8 57.8 9.0 9.0 4 9.3 9.0 6.9 8.0 71.0 94.5 9.5 9.5 63.5 54.7 12.0 12.0 5 8.8 9.8 7.3 8.2 36.0 45.0 8.0 8.0 51.1 56.3 10.5 10.0 6 7.6 7.5 6.3 5.9 47.5 48.0 9.0 9.0 58.5 52.8 11.0 11.0 7 8.1 8.8 6.6 7.0 39.0 46.0 6.0 6.5 50.6 52.3 9.5 9.5 8 8.9 8.2 8.0 7.5 51.0 59.5 7.0 7.0 53.8 55.0 9.0 10.0 9 8.6 8.1 6.4 6.6 42.0 40.5 7.0 7.5 47.3 54.8 10.0 9.5 10 9.4 7.8 7.4 6.3 35.0 47.0 7.5 7.5 54.0 56.0 10.0 10.0 11 8.8 7.9 6.9 6.3 40.0 44.0 6.0 6.0 51.7 49.7 8.5 9.0 12 9.7 9.6 7.2 7.7 38.0 38.0 7.0 7.5 48.6 51.6 10.0 9.5 13 10.1 7.3 7.1 6.6 33.0 33.5 7.0 6.5 51.8 50.2 8.5 9.0 14 8.8 8.8 6.9 6.6 31.5 37.0 7.5 7.5 52.4 53.6 9.5 10.0 15 10.0 7.3 7.9 6.7 38.0 39.5 7.0 7.5 45.0 46.5 9.0 9.0 16 7.7 6.1 5.6 5.2 29.0 29.5 6.5 6.5 43.4 51.0 9.0 9.5 17 8.0 9.7 6.5 7.3 34.5 41.5 6.5 7.5 48.2 53.7 9.5 9.5 18 10.8 10.0 8.9 8.0 63.5 67.5 8.0 8.0 60.9 59.7 11.0 10.5 ia 2A &2 IA 79-0 74.0 M 57.3 53.2 11.0 10.5 Mean 8.8 8.4 6.9 6.9 41.3 45.1 7.4 7.5 51.3 52.7 9.8 9.5 Placebo Group 20 7.6 8.3 6.2 6.5 33.0 34.5 7.0 7.0 50.0 55.5 9.5 9.5 21 8.7 7.5 7.1 6.7 54.0 57.5 7.5 7.5 56.6 59.3 9.5 9.0 22 6.4 6.3 5.8 5.6 29.0 31.5 7.0 7.0 47.3 50.3 9.5 9.5 23 8.8 9.6 7.1 7.4 66.0 66.0 8.5 8.5 54.9 57.2 10.5 10.5 24 7.8 7.6 6.6 6.2 29.0 31.0 7.5 7.0 51.4 52.9 9.0 9.0 25 9.0 8.0 6.9 6.5 35.0 44.5 7.5 7.0 49.5 47.0 10.0 9.0 26 7.6 7.5 7.8 7.0 51.5 60.0 7.5 8.0 56.2 56.0 10.0 10.5 27 7.6 6.7 7.2 6.1 37.5 47.5 6.0 6.5 51.1 52.0 8.5 9.0 28 8.4 7.5 7.3 6.9 45.5 48.0 6.5 6.5 50.0 47.6 9.0 9.5 29 10.0 10.0 8.6 8.8 58.0 65.0 7.5 8.0 53.0 55.6 10.0 10.5 30 9.2 8.5 7.8 7.2 50.0 47.5 7.0 6.5 53.0 49.7 9.0 9.0 31 6.9 7.8 6.2 6.9 36.5 33.0 8.0 8.0 50.9 48.8 10.0 10.0 32 10.4 7.0 7.4 6.3 38.0 39.5 7.0 7.0 43.9 48.1 9.0 9.0 33 8.4 8.6 6.7 7.5 50.0 39.0 6.0 6.0 42.0 39.1 8.5 8.5 34 7.4 8.2 6.7 7.8 45.0 48.0 8.0 8.0 49.4 51.8 10.0 10.0 35 8.4 8.5 6.6 5.9 26.5 26.0 6.0 6.0 39.7 38.8 9.0 8.0 36 9.1 12.2 8.0 8.6 73.0 60.0 8.0 7.5 57.4 52.2 11.0 10.5 37 8.9 6.9 6.7 6.1 28.0 28.0 7.5 7.5 55.6 48.8 10.0 10.0 38 6.7 5.9 5.6 5.7 40.5 31.0 7.5 7.0 49.1 47.7 10.0 10.0 39 6.7 8.9 6.4 7.3 42.0 38.5 7.0 7.0 45.1 45.9 10.0 8.5 42 IA 9J2 M ZZ 58,5 55,0 IA IA 56,5 55,4 10.0 10.0 Mean 8.4 8.2 7.0 6.0 43.7 44.8 7.2 7.2 50.6 50.6 9.6 9.5 119 Appendix K Enzyme Activity Raw Data (units g 1 protein) Citrate synthase a-GPDH Subject # Pre-test Post-test Pre-test Post-test Iron Group 34 0.052 0.031 0.038 0.135 8 0.043 0.048 0.064 0.078 12 0.037 0.050 0.061 0.066 19 0.057 0.041 0.086 0.082 21 0.065 0.052 0.059 0.083 24 0.048 0.043 0.073 0.056 31 0.039 0.055 0.045 0.098 36 0.041 0.040 0.050 0.078 37 0.033 0.057 0.059 0.087 as 0,059 0,065 0,12Q 0,087 Mean+S.D. 0.047 0.048 0.066 0.085 ±.011 ±.010 ±.024 ±.021 Placebo Group 11 0.054 0.056 0.061 0.079 20 0.059 0.047 0.087 0.060 22 0.021 0.046 0.021 0.070 23 0.036 0.036 0.084 0.072 35 0.025 0.026 0.036 0.051 Mean ±S.D. 0.039 0.042 0.058 0.066 ±.017 ±.011 ±.029 ±.011 1 unit = 1 umole of product produced under saturating conditions in 1 minute at 25°C. 120 Appendix L Descriptive statistics of placebo treated subjects who became anemic (n = 3) Variate Count Pre-Test Post-Test Serum ferritin 3 9.3±4.0 8.3±4.2 (ngml"1) Hemoglobin 3 12.4±0.3 11.1±0.2 (gd"1) Iron Total 3 75.0±43.7 81.3±65.4 (wrdr1) UIBC 3 317.7±62.6 300.0±73.3 (^gdr1) % Transferrin Saturation 3 19.3±11.0 21.3±16.3 (%) Alactic Power 3 7.1±0.7 7.9± 1.8 (watts-kg"1 body wt) Lactacid Power 3 6.1±0.5 6.9±1.1 (wattskg"1 body wt) Anaerobic Speed Test 3 47.0±10.0 41.5±12.3 (seconds) Anaerobic Threshold 3 7.3±0.3 7.0±0.0 (mh"1) V0 2 max 3 50.2±5.8 49.7±5.0 (mlkg"1min'1) Max Velocity 3 10.0±0.00 9.5±0.9 (mh"1) 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items