MAW, GRAEME J.; MACKENZIE, IAN L.; COMER, DES A. M.; TAYLOR, NIGEL A. S.
A commonly reported, though not universally observed(27,33), adaptation accompanying habitual physical exercise is an expansion of blood volume (BV). This is typically reflected by a supranormal plasma volume (PV)(25,28), with trained male runners, for example, having BV of up to 107 ml · kg-1, including 66 ml · kg-1 of plasma, compared with BV of between 75 and 85 ml · kg-1 in untrained males (10,19).
The mechanisms of exercise-induced hypervolemia are presently unclear, although both hydrostatic and osmotic causes have been suggested(2,6). Brotherhood et al. (2) suggested that BV was expanded hydrostatically in order to fill the enlarged volume of the trained heart and circulation. Harrison et al.(14) proposed that plasma expansion was stimulated by a post-exercise intravasation of protein, which osmotically drew water from the extravascular compartment. Thus, PV expands in response to physical training, with the possible consequence of reduced mixed-venous hematocrit(Hct[horizontal bar over]v) and hemoglobin concentrations(25,28).
Depressed Hct[horizontal bar over]v and hemoglobin concentrations are not consistent consequences of physical training. For example, when investigating exercise-induced anemia, Magnusson et al.(23) found no difference between the hemoglobin concentrations of long-distance runners and inactive males. When measured directly, hemoglobin mass or erythrocyte volume (RCV) has been found to be enlarged following endurance training, or in comparison with nonathletic counterparts (2,19). Dill et al.(10) reported a mean RCV of 41.4 ml · kg-1 in 12 middle-distance runners compared with 35.2 ml · kg-1 in a similar, more sedentary, population. Such a trend has been attributed to erythropoietin release following localized hypoxia during physical exertion(2,28). The hypoxia may result from reduced blood flow, from increased oxygen demand, or from the initial hemodilution associated with exercise training. Regardless of the process, it appears that prolonged habitual physical training may provoke an expansion of BV that, while initially confined to the plasma phase, is later reflected by a proportionate expansion of RCV as well.
The proportionate expansion of PV and RCV is not surprising in light of the freedom-of-water movement around the body. Acute reductions in PV during dehydration, for example, have been shown to rapidly deplete RCV due to the osmotic transfer of water across the cellular membrane(8). Similarly, it would be surprising if chronic hypervolemia were maintained at the expense of depleted extravascular fluids. However, there is presently a paucity of information regarding the volume and distribution of extravascular fluids in endurance-trained individuals, so that speculation regarding such fluid balance often lacks empirical support.
Measurements of whole-body fluid distribution, which could shed light on this balance, have previously been hampered by methodological limitations including time constraints, subject safety, and internal validity. For example, the initial method of simultaneously measuring total-body water(TBW), extracellular fluid, plasma, and RCV required 36 h to complete, and the withdrawal of 200 ml of blood (24). A later modification allowed similar volume measurements within 3 h (31), but was limited by the use of radioiodinated serum albumin (RISA) to measure PV. RISA dilution has consistently been shown to exaggerate PV in comparison with the dilution of larger tracers due to the constant exchange of albumin between the plasma and interstitial fluids (1,22). In contrast, radiolabeled fibrinogen (RISF) is slow to leave the vascular space(1,9), and, hence, in 1980 the International Committee for Standardization in Haematology (ICSH) (18) suggested that the dilution of RISF may provide a more accurate measure of PV than presently afforded by the use of RISA. It was, therefore, the purpose of this study to develop an accurate and convenient method for measuring the overall distribution of body fluid using a method of simultaneous radionuclide dilution incorporating RISF and to subsequently determine the distribution of body fluids in endurance-trained males.
Body-fluid distribution was measured in eight adult males(Table 1), using the simultaneous dilution of four radionuclides. Subjects were asymptomatic, trained endurance athletes, five of whom had achieved National- or State-level representation in their chosen sport. Subjects were fully informed of the experimental procedures, which were approved by the University of Wollongong Human Experimentation Ethics Committee, and subsequently provided informed consent. Body-fluid distribution was assessed on three occasions, each separated by a minimum of 28 d, to permit the decline of residual radioactivity.
Subjects arrived at the laboratory 1 h prior to assessment, in a rested state, following a 12-h overnight fast. Following a standard breakfast (38 kJ· kg-1 of body mass plus 5 ml · kg-1 of water), 20 ml of blood were collected, without stasis, from an antecubital vein. 10 ml were used in the preparation of a radiochromate injection and 10 ml were stored as a background reference. A urine sample was collected, from which a sample was stored as a background reference. Subjects were first seated at rest for 30 min to stabilize circulation. Our clinical procedures for such measurements typically use a seated posture, and since postural changes affect fluid distribution, we kept subjects seated during both the preparatory and the data collection phases of each trial.
The radiochromate injection, used to measure RCV, was prepared in accordance with ICSH (18) recommendations. Packed erythrocytes from 10 ml of blood were incubated for 20 min at 37°C with 8μCi of sodium-radiochromate (Amersham Australia, Na51Cr), washed three times, and resuspended to an approximate volume of 10 ml in isotonic saline. Four radionuclide injections were then administered into a second antecubital vein, via a Teflon cannula, within a 30-s period. Two μCi of radioiodinated human serum fibrinogen (Amersham Australia, 125I Human Fibrinogen; RISF), the 51Cr-labeled autologous erythrocytes, 20 μCi of sodium-radiobromide (Australian Radioisotopes, Na82Br), and 500μCi of tritiated water (Amersham Australia, 3H2O) were injected sequentially to enable measurement of PV, RCV, extracellular fluid(ECFV), and TBW volumes, respectively. The mid-time of the fibrinogen injection was considered as the commencement of assessment (t0). The exact quantity of each injection was determined gravimetrically, and small quantities were used to prepare the respective radionuclide standards. The cannula was immediately flushed with sufficient heparinized saline to render it suitable for subsequent blood sampling (21).
Blood samples (10 ml) were collected, without stasis, at t30, t60, and t180, and treated with ethylenediamine tetra-acetic acid (EDTA). Collection of each sample was preceded by removal from the cannula of 5 ml of supernatant, and followed by a 10-ml flush of heparinized saline. A urine void was collected and measured at t180. Blood samples, including the background reference, were centrifuged for 40 min at 1700× g to separate erythrocytes, and the volume of erythrocytes adjusted to account for 2% trapped plasma (3). On seven occasions, a 3-ml aliquot of whole blood was removed prior to blood separation to assist in the calculation of ECFV. Aliquots (3 ml) of plasma and erythrocytes from each sample were placed into glass vials and refrigerated at 4°C pending γ-radiation counting; a 3-ml aliquot of urine was similarly stored. Erythrocyte aliquots were hemolyzed prior to storage, using a trace of powdered saponin (Sigma Chemical Company, S-1252). Further 0.5-ml aliquots of plasma and urine were placed into glass vials in preparation forβ-radiation counting. The preparation involved vigorously mixing each aliquot with 0.05 ml of 1 M · l-1 hydrochloric acid to solubilize all solid tissues, followed by 9 ml of liquid scintillation cocktail (Packard Instruments, Emulsifier-Safe). The combination of the acid and the cocktail effected clear solutions with homogeneous radiation quench, regardless of the original fluid.
Radiobromine activity was counted on the day of assessment using a calibrated well-type γ scintillation counter (Abbott Laboratories, Auto-LOGIC). Counting of all other radionuclide samples was delayed for 14 d, pending the decay of 82Br, since its activity interfered with the detection of 125I and 51Cr. Spears et al.(31) previously suggested allowing between 5 and 7 d to elapse prior to the measurement of 125I and 51Cr in order to permit the decay of 82Br; however, in the present study, considerable82 Br was still detectable in both plasma and erythrocytes 10 d after their collection, making it impossible to differentiate the respective activities. It was therefore necessary to allow 82Br to decay for 14 d prior to the measurement of 125I and 51Cr, at which time only 0.14% of the original 82Br remained, due to its half-life of 36 h(16). 3H activity was counted using a liquid scintillation counter (LKB Wallac, 1219 Rackbeta), while 125I and51 Cr were counted using the γ scintillation counter. Aliquots were counted twice for 82Br due to its short physical half-life; and three times for 3H, 125I, and 51Cr, for a minimum aggregate of 10,000 counts each. For each radionuclide, the sequence of vials was reversed between counts. Radionuclide concentrations were then averaged and expressed as counts · min · ml-1. After radiation counting, plasma aliquots were assessed in triplicate for protein concentration using a refractometer (Otago, 93032). Plasma protein concentration was considered as the mean of the three refractometer readings.
Three-ml dose standards were prepared for each radionuclide from dilution samples. For 51Cr, this was achieved by combining 0.5 ml of the radiochromated erythrocyte preparation with distilled water to produce a total sample volume of 250 ml. Tritium, 82Br, and 125I standards were similarly prepared, diluting approximately 0.1, 0.1, and 0.2 ml of the respective preparations with distilled water to provide dilution volumes of 500, 500, and 250 ml, respectively. A 3-ml distilled water sample was also stored as a background reference. These standards were subsequently used to determine the exact injected doses of the four radionuclides in the calculation of respective compartmental fluid volumes.
Compartmental fluid volumes were determined using equations described by Chien and Gregersen (4). Thus, in determining TBW, it was necessary to correct plasma 3H concentration ([3H]) for the presence of protein and 125I, and for 3H loss in urine. Similarly, 82Br concentration ([82Br]) was corrected for the presence of protein, and for 82Br loss in erythrocytes and urine;51 Cr concentration ([51Cr]) was corrected for 51Cr loss in urine; and 125I concentration ([125I]) was corrected for the gradual loss of 125I from the vascular space. This latter extravasation was accounted for using semilogarithmic extrapolation of the 125I elution curve to estimate the theoretical [125I] at t0(18). Equation
where d = protein displacement factor (see below); SH =[3H] of the 3H standard; Sd = dilution of the 3H standard; SV = volume of the 3H2O injection; UV = volume of urine collected at t180; UH = [3H] in the urine collected at t180; i = ratio of 125I detected in the3 H and 125I energy ranges; UI = [125I] in the urine collected at t180; PH = [3H] in the t180 plasma aliquot; PI = [125I] in the t180 plasma aliquot.Equation
where d = protein displacement factor (see below); r = Gibbs-Donnan ratio (taken as 1.02; 4); SB [82Br] of the82 Br standard; Sd = dilution of the 82Br standard; SV = volume of the Na82Br injection; UV = volume of urine collected at t180; UB = [82Br] in the urine collected at t180; c = the ratio of 51Cr detected in the82 Br and 51Cr energy ranges; UC = [51Cr] in the urine collected at t180; RCV = erythrocyte volume; EB =[82Br] in the t180 erythrocyte aliquot; EC =[51Cr] in the t180 erythrocyte aliquot; PB =[82Br] in the t180 plasma aliquot; PV = plasma volume.Equation
where SC = [51Cr] of the 51Cr standard; Sd = dilution of the 51Cr standard; SV = volume of the Na51Cr injection; 0.25 accounts for the first 45 min of the t180 urine collection; UC = [51Cr] in the urine collected at t180; UV = volume of urine collected at t180; EC = mean of[51Cr] in the t30 and t60 erythrocyte aliquots.Equation
where SI = [125I] of the 125I standard; Sd = dilution of the 125I standard; SV = volume of the RISF injection; PIO = theoretical [125I] in plasma at t0.
The protein displacement factor (d) was calculated as:Equation
where [PP] = plasma protein concentration.
On seven occasions, when a whole-blood aliquot was preserved, the extent of82 Br loss into erythrocytes was determined both directly from measuring[82Br] in erythrocytes and from calculating erythrocyte-[82Br] from the whole-blood concentration adjusted for Hct[horizontal bar over]v. Erythrocyte-[82Br] was then calculated as:Equation
where PB = [82Br] in the t180 plasma aliquot; WB = [82Br] in the t180 whole-blood aliquot; c= ratio of 51Cr detected in the 51Cr and 82Br energy ranges; WC = [51Cr] in the t180 whole-blood aliquot; Hct[horizontal bar over]v = hematocrit of the t180 blood sample.
Further calculations were made to determine intracellular water volume (ICW= TBW - ECW, where ECW was ECFV-corrected for the total volume of plasma solutes (4), interstitial fluid volume (IFV = ECFV - PV), and BV (BV = RCV + PV).
Compartmental fluid volumes were then considered for each subject as the mean of three assessments; retest reliability was considered as the mean of each subject's coefficient of variation for each fluid volume. The three assessments were compared using analyses of variance, while comparisons of two related measures were made using paired t-tests. Alpha was set at the 5% level, and specific differences were then examined using Tukey's test of Wholly Significant Difference (TukeyWSD). Data are reported as means with standard error of the means, unless otherwise stated as standard deviations (σ).
Body-fluid volumes did not differ significantly within subjects between repeat assessments, with the mean intrasubject coefficients of variation generally being less than or equal to 6% (Table 2). The exception was for PV, with a coefficient of variation of 8.44%, for which considerable physiological variation was possible during the 56-d course of the study. The coefficients of variation embraced both physiological and methodological variation, and hence, for all volumes, variation due to methodological error could be considered to be less than the reported coefficient. The compartmental fluid volumes for TBW, ECFV, RCV, and PV averaged 51,822 (±5022), 20,420 (±1893), 2627 (±242), and 3673 (±365) ml, respectively (means ± σ;Table 2). Hence, the respective ICW, IFV, and BV were 30,784 (±4630), 16,747 (±1621), and 6302 (±529) ml (means± σ; Table 2).
Measurement variability was probably unaffected by the presence of residual radiation as, at the time of reassessment, less than 0.01% of the previous3 H, 125I, and 51Cr doses remained in the blood; no residual 82Br was detected, reflecting the 36-h half-life of the82 Br nuclide (16). The clearance of residual radiation confirmed the report of Fortney et al. (12) that the multiple radionuclide method could safely be used for sequential determinations of human body-fluid distribution, with the total radiation dose imposed by three assessments amounting to approximately one-fifteenth of the annual dose permitted for an industrial worker (16).
The measurement of ECFV was dependent on the method of calculating intracellular losses of 82Br in the seven assessments when erythrocyte-[82Br] was measured both directly from erythrocytes and calculated from the plasma and whole-blood [82Br]. The apparent erythrocyte-[82Br] was higher in the latter measure (P = 0.001). The corresponding ratio between erythrocyte-[82Br] and that in plasma was 0.804 (±0.019), compared with 0.703 (±0.011) when erythrocyte and plasma concentrations were both measured directly. Consequently, ECFV was 532 (±54) ml smaller when erythrocyte-[82Br] was derived from plasma and whole-blood concentrations compared with the volume determined directly from plasma and erythrocytes (P = 0.001).
Analysis of urine voided showed that 1.25% (±0.08%) of the82 Br dose was excreted during the 3-h assessments. Similarly, an average of 0.85% (±0.10%), 15.60% (±0.97%), and 0.44%(±0.07%) of respective 3H, 125I, and 51Cr doses was also excreted during the course of assessment.
The distribution of body fluid was measured in eight endurance-trained males. Previous studies have examined the body-fluid distribution of normal or bed-ridden populations (11,12,31), or the intravascular fluid volumes of athletes(10,19,23). However, no studies were found documenting the distribution of both intra and extravascular fluids in endurance-trained subjects.
The body-fluid volumes of the present sample were all slightly enlarged compared with previously reported reference values(17,18; see Table 2), but were considered normal in light of the subjects' athletic history and body composition (26,30). The subjects were all well-trained, participating in endurance exercise daily, and, on the basis of their estimated maximal aerobic power, were among the top 10% of age-matched Australian males (13). Similarly, based on their skinfold measurements, subjects were among the leanest 5% of aged-matched Australian males (13).
Low adiposity is known to be accompanied by a large volume of TBW relative to body mass due to the higher concentration of water in lean tissue than in fat (26). Hence, in comparison with less lean populations, the present subjects possessed enlarged TBW, resulting in similarly expanded ICW and ECFV. Fluid volumes were still comparable to values at the upper end of the reference data, determined using similar dilution methods on more sedentary male populations(12,31), confirming the validity of the present simultaneous dilution method. For example, Dyrbye and Kragelund(11) reported values of 609, 352, and 265 ml · kg-1 for TBW, ICW, and ECFV, respectively.
Despite the present absolute volumes being enlarged, the relative contributions of both the ICW and ECFV to TBW remained similar to previously reported values. ICW and ECFV accounted for 61% and 39% of TBW, respectively, compared with respective values ranging from 57% to 63% and from 37% to 43% in more sedentary populations (11,17,31). Similarly, the relationships between TBW and the intravascular fluid volumes were essentially maintained, with BV and PV accounting for 12.3% and 7.1% of TBW compared with previously reported values ranging from 11.2% to 12.4% and 6.9% to 7.7%, respectively (11,17,31).
Intravascular volumes were also larger, in absolute terms, than those considered normal for more sedentary populations, with the BV averaging 80 ml· kg-1; 47 ml · kg-1 for plasma, and 33 ml· kg-1 for erythrocytes. In comparison, the ICSH(18) reported normal BV, measured using the simultaneous dilution of RISA and radiochromated erythrocytes (51Cr), to be between 65 and 75 ml · kg-1. Of this BV, plasma accounted for 40 ml· kg-1 and erythrocytes for 25-35 ml · kg-1. Similarly, Sawka et al. (27) found that the blood volume of 22 healthy young males was 69 ml · kg-1, including 43 ml· kg-1 of plasma and 26 ml · kg-1 of erythrocytes. We suggest that the normal BV determined using RISF may be slightly smaller than these values, as PV measured using RISA may be an overestimate of the true circulating plasma, due to its relatively greater leakage from the vascular space. Consequently, it has been shown that PV is enlarged by approximately 6% when measured using RISA dilution compared with that measured using the dilution of RISF (1,22).
The present intravascular fluid volumes were comparable to values obtained from trained subjects. For example, BV between 79 and 85 ml · kg-1 have been reported for trained cyclists and runners(23,30), while even larger volumes are found in highly trained athletes (93-104 ml · kg-1) with maximal aerobic power around 75 ml · kg-1 · min-1(2,10,19). The relationship between BV and maximal aerobic power has previously been identified(15), and it was considered that the present mild hypervolemia was the result of training history of the subjects rather than methodological error. It was therefore concluded that the radionuclide dilution method employed in the current investigation provided a reliable means of determining the distribution of intravascular and extravascular fluid volumes.
In the present study, IFV was not decreased compared with normal values(17), and Hct[horizontal bar over]v was normal(23). It is therefore likely that hypervolemia was a reflection of whole-body hyperhydration (5), including a proportionate expansion of all body-fluid compartments. Proportionately expanded PV and RCV have previously been reported following prolonged periods of exercise training (19,28), while depressed Hct[horizontal bar over]v is primarily evident during the early stages of training, before RCV has responded (6,25). Increased RCV has been attributed to the stimulation of erythropoietin production (2,28), although, this does not always increase following exercise (20). In light of the present data, expansion of the RCV may also be explained by homeostatic balancing of the body-fluid compartments, since it is unlikely that whole-body hyperhydration would be maintained without a proportionate increase in RCV.
The increased volume of TBW, necessary to support such a whole-body hyperhydration, may be accounted for by increased water retention due to increased plasma renin, aldosterone, and vasopressin activity following exercise (7). In particular, increased sodium retention, due to increased renal sensitivity to aldosterone, may be an important factor in exercise-induced water retention (5). In addition, part of this hyperhydration may be associated with reduced adiposity. Dill et al. (10) suggested that at least one-third of exercise-induced hypervolemia could be explained by reduced adiposity accompanying habitual exercise. In fact, the present data indicate that differences in adiposity between the current trained subjects and more sedentary reference males may actually account for more than one-third of the apparent hyperhydration. For example, PV adjusted for adiposity (estimated from TBW) (26) equated to approximately 52 ml · kg-1 of fatfree mass compared with a volume of 50 ml · kg-1 from normal values (18; based on body fat content of 20%). It is therefore suggested that, when considering the chronic status of body-fluid volumes, consideration must be given to the relative adiposity of the subject while recognizing the inaccuracies of estimating body fat content.
The present method for measuring body-fluid volumes, using the simultaneous dilution of 3H2O, Na82Br, RISF, and 51Cr-labeled erythrocytes, produced results that were not only considered normal for the appropriate sample, but were also reliable between repeated assessments. Seven subjects underwent three repeated assessments, spaced at 28-d intervals, and the resultant coefficients of variation between repeated assessments was less than or equal to 6% for measured and derived fluid volumes(Table 2). The exception was PV, which produced a coefficient of variation of 8.4%, which may be attributed to physiological changes between repeat assessments, as well as to methodological error. Considerable physiological variation is possible in PV, and other fluid volumes, due to changes in environmental conditions and health status(12). Such variations may well have occurred during the 56-d course of the present study. Spears et al. (31) reported a variation in plasma volume of just 1.2% during a 7-d period, and variations of 0.5%, 1.3%, and 3.9% in TBW, ECFV, and RCV, respectively. In addition, in the present study, the accuracy of RISF and 51Cr dilution was found to be ±2.50% and ±2.82%, respectively, when measuringin vitro volumes of between 2 and 71. Hence, it was considered that, in the present study, the variation in volume measurements due to methodological error was probably less than the calculated coefficients of variation, and may well have approached 3% for all compartmental volumes.
The accuracy of ECFV measurement was further influenced by the method of its calculation, specifically the method of accounting for the loss of82 Br into erythrocytes. The discrepancy between directly measured erythrocyte-[82Br] and that calculated from plasma and whole blood was difficult to explain. Had the direct measure been influenced by the presence of 82Br in trapped plasma, it would probably have exceeded the indirect[82Br]. In practice, the reverse was true, making the directly determined ECFV smaller than that derived from indirect calculations. In the former case, the ratio between directly measured erythrocyte-[82Br] and plasma-[82Br] was similar to values previously reported during the calculation of extracellular fluid volume (32). Hence, it was considered that erythrocyte-[82Br] should be measured directly from packed cells, rather than calculated as a function of whole blood concentration, during the measurement of the extracellular space. This procedure is recommended when adopting similar methods to determine ECFV.
It has been shown that the simultaneous dilution of 3H2O, Na82Br, RISF, and 51Cr-labeled erythrocytes provides a reliable method for measuring body-fluid distribution in humans. Each of the respective dilution volumes required several corrections, but, having done so, retest variance was generally around 6%. The compartmental fluid volumes of endurance-trained males were generally larger than normally observed in sedentary adult males, revealing a whole-body hyperhydration, but with normal between-compartment fluid distribution. While hypervolemia has previously been observed in trained subjects, this study has shown that such hyperhydration is also found in all other fluid compartments.
1. Bent-Hansen, L. Initial plasma disappearance and distribution volume of [131
I]albumin and [125
I] fibrinogen in man. Acta Physiol. Scand.
2. Brotherhood, J., B. Brozovic, and L. G. C. Pugh. Haematological status of middle- and long-distance runners. Clin. Sci. Mol. Med.
3. Chaplin, H. and P. L. Mollison. Correction for trapped plasma in the red cell column of the hematocrit. Blood
4. Chien, S. and M. I. Gregersen. Determination of body fluid volume. In: Physical Techniques in Biological Research, IV, Special Methods
, W. L. Nastuk (Ed.). New York: Academic Press, 1962, pp. 1-105.
5. Convertino, V. A. Blood volume: its adaptation to endurance training. Med. Sci. Sports Exerc.
6. Convertino, V. A., P. J. Brock, L. C. Keil, E. M. Bernauer, and J. E. Greenleaf. Exercise training-induced hypervolemia: role of plasma albumin, renin, and vasopressin. J. Appl. Physiol.
7. Convertino, V. A., L. C. Keil, E. M. Bernauer, and J. E. Greenleaf. Plasma volume, osmolality, vasopressin, and renin activity during graded exercise in man. J. Appl. Physiol.
8. Costill, D. L. and B. Saltin. Changes in the ratio of venous to body hematocrit following dehydration. J. Appl. Physiol.
9. De Lanne, R., J. R. Barnes, and L. Brouha. Changes in concentration of plasma protein fractions during muscular exercise. J. Appl. Physiol.
10. Dill, D. B., K. Braithwaite, W. C. Adams, and E. M. Bernauer. Blood volume of middle-distance runners: Effect of 2,300-m altitude and comparison with non-athletes. Med. Sci. Sports
11. Dyrbye, M. O. and E. Kragelund. Simultaneous determination of the apparent 3
I human albumin and 51
Cr red cell volumes in human subjects. Scand. J. Clin. Lab. Invest.
12. Fortney, S. M., K. H. Hyatt, J. E. Davis, and J. M. Vogel. Changes in body fluid compartments during a 28-day bed rest.Aviat. Space Environ. Med.
13. Gore, C. J. and D. Edwards. Australian Fitness Norms: A Manual for Fitness Assessors
. Adelaide, Australia: Health Development Foundation, 1992.
14. Harrison, M. H., R. J. Edwards, and D. R. Leitch. Effect of exercise and thermal stress on plasma volume. J. Appl. Physiol.
15. Holmgren, A. and P-O Åstrand. DL
and the dimensions and functional capacities of the O2
transport system in humans. J. Appl. Physiol.
16. International Commission on Radiological Protection.Evaluation of Radiation Doses to Body Tissues from Internal Contamination due to Occupational Exposure
(Publication No. 10). Oxford, England: Pergamon Press, 1968.
17. International Commission on Radiological Protection.Report of the Task Group on Reference Man
(Publication No. 23). Oxford, England: Pergamon Press, 1975.
18. International Committee for Standardization in Haematology. Recommended methods for measurement of red-cell and plasma volume. J. Nucl. Med.
19. Kjellberg, S. R., U. Rudhe, and T. Sjostrand. Increase in the amount of hemoglobin and blood volume in connection with physical training. Acta Physiol. Scand.
20. Klausen, T., L. Breum, N. Fogh-Andersen, P. Bennett, and E. Hippe. The effect of short and long duration exercise on serum erythropoietin concentrations. Eur. J. Appl. Physiol.
21. Ladegaard-Pedersen, H. J. and H. C. Engell. Blood volume determination using one catheter located in a central vein: an analysis of sources of error in blood sampling. Acta Chir. Scand.
22. Larsen, O. A. Studies of the body hematocrit phenomenon: dynamic hematocrit of a large vessel and initial distribution space of albumin and fibrinogen in the whole body. Scand. J. Clin. Lab. Invest.
23. Magnusson, B., L. Hallberg, L. Rossander, and B. Swolin. Iron metabolism and “sports anemia”, II: a hematological comparison of elite runners and control subjects. Acta Med. Scand.
24. Moore, F. D., J. D. McMurrey, H. V. Parker, and I. C. Magnus. Total body water and electrolytes: intravascular and extravascular phase volumes. Metabolism
25. Oscai, L. B., B. T. Williams, and B. A. Hertig. Effect of exercise on blood volume. J. Appl. Physiol.
26. Pace, N. and E. N. Rathbun. Studies on body composition, III: The body water and chemically combined nitrogen content in relation to fat content. J. Biol. Chem.
27. Sawka, M. N., A. J. Young, K. B. Pandolf, R. C. Dennis, and C. R. Valeri. Erythrocyte, plasma, and blood volume of healthy young men.Med. Sci. Sports Exerc.
28. Schmidt, W., N. Maassen, F. Trost, and D. Boning. Training induced effects on blood volume, erythrocyte turnover and haemoglobin oxygen binding properties. Eur. J. Appl. Physiol.
29. Senay, L. C., D. Mitchell, and C. H. Wyndham. Acclimatization in a hot, humid environment: body fluid adjustments. J. Appl. Physiol. 40:786-796, 1976.
30. Sjostrand, T. Blood volume. In: Handbook of Physiology: Circulation I.
Bethesda, MD: American Physiological Society, 1962, pp. 51-62.
31. Spears, C. P., K. H. Hyatt, J. M. Vogel, and S. B. Langfitt. Unified method for serial study of body fluid compartments.Aerosp. Med.
32. Staffurth, J. S. and I. Birchall. The measurement of the extracellular fluid volume with radioactive bromine. Clin. Sci.
33. Young, A. J., M. N. Sawka, and M. D. Quigley. Role of thermal factors on aerobic capacity improvement with endurance training.J. Appl. Physiol.
BLOOD VOLUME; ERYTHROCYTE VOLUME; EXTRACELLULAR FLUID VOLUME; INTRACELLULAR FLUID VOLUME; PLASMA VOLUME; TOTAL BODY WATER