Journal Logo

BASIC SCIENCES: Original Investigations

Physical activity, cardiorespiratory fitness, and the primary components of blood viscosity


Author Information
Medicine & Science in Sports & Exercise: February 2000 - Volume 32 - Issue 2 - p 353
  • Free


Blood viscosity (the intrinsic resistance of blood to flow in vessels) has been positively associated with the incidence of ischemic heart disease (IHD) events (22). The primary determinants of blood viscosity are the Hct concentration (volume fraction of red blood cells) and plasma viscosity, which together explain up to 70% of inter-individual variation (22). Hct concentration and plasma viscosity have each been independently associated with IHD events in middle-aged British men (22,37,41,42). Relative viscosity (a measure of red cell deformability and aggregation) accounts for the residual inter-individual variation, but it does not appear to be a major determinant of the predictive value of blood viscosity (22).

Epidemiological evidence has associated blood viscosity and its determinants with conventional IHD risk factors (3,7,22–24,31). Regular physical activity significantly reduces the risk of IHD, but limited data exist on its relationship with plasma viscosity or Hct from population studies (8,13,40). Plasma fibrinogen is the main protein determinant of plasma viscosity and its association with IHD appears partially dependent on its mutual association with plasma viscosity (37). Several cross-sectional studies have reported an inverse association between leisure time physical activity (LTPA) and plasma fibrinogen (8,19,25). Age, smoking, and social class appear to confound this association within some representative British investigations (19,25). In the present study we investigated the association between both LTPA and predicted maximum oxygen consumption (V̇O2max) with plasma viscosity levels and Hct concentrations in employed middle-aged men of similar high social class.

Study participants.

The cross-sectional sample in this study consisted of 740 asymptomatic middle-aged men. Data collection was conducted between 1992 and 1996 on individuals who presented themselves for routine preventive assessment at a Leeds-based private hospital. All subjects were free of diagnosed IHD. Subjects with possible asymptomatic IHD had been excluded using sensitive exercise electrocardiographic criteria (≥1 mm ST segment depression on exercise testing). All subjects were in current employment and within socio-economic groups I and II. Written informed consent was obtained from all subjects before the clinical examination. The study protocol was approved by the hospital institutional review committee.

Clinical examination.

Subjects stature (centimeters) and body mass (to the nearest 0.2 kg) were measured using Seca balance scales and metal stadiometer (Model 713, Seca Ltd., Birmingham, U.K.). Body Mass Index (BMI) was calculated as the ratio of body mass to stature squared (kg·m-2).

Subjects were instructed to adhere to an overnight fast (minimum 12 h). Venous blood was sampled between 9:00 a.m. and 1:00 p.m. A tourniquet was applied lightly for minimal duration and blood was drawn into monovette (Sarstedt, Numbrecht, Germany) collection tubes. Plasma viscosity was added to the blood profile from July 1992—6 months following commencement of data collection. Plasma viscosity was available in 666 men (595 nonsmokers). Samples for rheological parameters were collected in 2.7-mL EDTA tubes. Plasma viscosity assays were measured on a Coulter-Harkness capillary viscometer at 25°C (11). Coefficients of variation for plasma viscosity measurement by capillary viscometer are 1–2% (21). Full blood count, including Hct concentration, was measured on a Coulter electronic particle counter (Coulter Instruments, Luton, UK). Hct concentrations were available in 710 men (638 nonsmokers). Missing cases (N = 26) were attributable to technical difficulties with venepuncture or assay. Serum lipids were determined with standard enzymatic methods on a Greiner G450 automated analyzer (Langethal, Switzerland). Serum high-density lipoprotein was measured in all normolipemic samples (fasting triglycerides < 4.0 mmol·L1) by the heparin manganese-chloride method.

Information on subjects’ medical and social history was obtained during a structured interview by a physician. The interview included standard questions to ascertain information on medical and social history. Smoking questions were based on standard epidemiological definitions (19,34), and current cigarette smokers were distinguished from former smokers and those who had never smoked cigarettes. Current pipe/cigars were classified as smokers, irrespective of previous cigarette smoking habit. In this analysis respondents were classified as current or nonsmokers. Recordings of weekly alcohol units based on self-assessed drinking habits were available for 97% of the entire cohort.

Self-reported leisure-time physical activity habits (4-wk reference period) were classified in 98% of the entire study sample using a modification of the Physical Activity Index (PAI) developed for the British Regional Heart Study (BRHS) (34). On interview subjects were asked standard physical activity questions and indicated their physical activity pattern under the prompting of regular walking or cycling, recreational activity, and vigorous activity. Full details of the calculation of composite scores for overall physical activity, and the original derivation of the scores for leisure time physical activities have been described (34). Golfing activity was modified to a recreational rather than a vigorous activity in the present study. For classification purposes, the original six PAI categories (sedentary, occasional, light, moderate, moderately vigorous, and vigorous) were regrouped into a four-category PAI based on total physical activity scores: sedentary (0–2), occasional/light (3–8), moderate/moderately vigorous (9–20), and vigorous (≥21). Similar interviewer administered surveys of usual physical activity appear reliable in middle-aged and older men (12,32). The initial validation of this physical activity index was based primarily on its inverse association with resting heart rate, following adjustment for age, BMI, smoking, and social class within the BRHS cohort of middle-aged men (34).

Maximum oxygen consumption (V̇O2max) is generally accepted as a valid and reliable measure of cardiorespiratory fitness (1). Cardiorespiratory fitness was assessed in all subjects using a modification of a submaximal (85% age-predicted maximum heart rate) cycle ergometer (Monark 818, Varberg, Sweden) protocol recommended for epidemiological surveys (35). V̇O2max (L·min1) and weight-related V̇O2max (mL·kg1·min1) were indirectly predicted using well-established methods (2). The continuous incremental protocol consisted of a cycling cadence of 50 rpm, with successive workload increments of 25 W at each exercise stage, following an initial workload of 75–100 W (depending on body mass). Each exercise stage was performed for 2 min if a steady-state heart rate was attained (steady state defined as heart rate increase < 5 beats·min1) or up to a maximum of 3 min. Heart rates were determined throughout the exercise test from a continuous electrocardiogram (CR7, Cardiorater, London, UK) and 12 lead ECG tracings (Cardioscript CD6000, Picker, Germany) were taken at the end of each exercise stage. Despite limitations in the submaximal prediction of V̇O2max (6) this method has continued applicability in discriminating cardiorespiratory fitness within epidemiological investigations (26,27).

Data analysis.

Descriptive statistics were computed and distributions of all variables were assessed for normality. The distribution of plasma viscosity was non-normal among the nonsmokers (skewness value 1.275, kurtosis 4.675). Since log transformation of plasma viscosity did not affect the significance of the findings, untransformed data are presented. Because the distributions of physical activity score, alcohol consumption, and serum triglycerides were substantially skewed, these variables were analyzed following logarithmic transformation. Data for variables were compared between physical activity categories and predicted V̇O2max quartiles using one-way analysis of variance (ANOVA).

Pearson product moment correlation coefficients and partial correlation coefficients (controlling for age) were calculated to determine the association between variables. ANCOVA was used to adjust for significant age-adjusted partial correlates of plasma viscosity and Hct. Student’s t-test for independent samples was utilized to examine general characteristics between subjects with and without the relevant rheological assays and between smokers and nonsmokers. Data are presented as mean ± SD and 95% confidence intervals (C I) for mean. For all statistical tests the alpha level adopted for significance was P < 0.05. Statistical analyses were performed on SPSS for Windows version 6.1 (SPSS Inc., Chicago, IL).


Plasma viscosity was evaluated in 666 men and Hct in 710 men (90% and 96% of entire study cohort, respectively). The prevalence of smoking within the entire cohort was 9.7%. Data on smokers were not included because of the small sample size and low variability in physical activity habits. The mean (± SD) plasma viscosity was 1.650 (± 0.084) mPa·s1 among the nonsmokers (95% C I 1.6433, 1.6568). The mean (± SD) Hct concentration was 45.55% (± 3.26%) among the nonsmokers (95% C I 45.29, 45.80). Nonsmokers without plasma viscosity determinations were not significantly different in terms of general and IHD risk characteristics. The small proportion of nonsmokers without Hct determinations (N = 26) had higher mean BMI and predicted V̇O2max (mL·kg1·min1) (P = 0.008, P = 0.001, respectively). Systolic blood pressure (BP) (N = 26) and fasting blood glucose (N = 10) were significantly lower (P = 0.001 and P = 0.017, respectively). The general and IHD risk characteristics of the nonsmokers for both rheological cohorts are shown in Tables 1 and 2. The prevalence of IHD risk factors among nonsmokers was similar for both rheological cohorts (data not shown).

Table 1:
Descriptive data on nonsmokers with plasma viscosity determinations.
Table 2:
Descriptive data on nonsmokers with haematocrit determinations.

Tables 3 and 4 show correlation coefficients and age-adjusted partial coefficients between both plasma viscosity and Hct concentration with IHD risk variables among nonsmokers. Plasma viscosity and Hct concentration were both inversely associated with both log-PAI and predicted V̇O2max (L·min1 and mL·kg1·min1) scores following adjustment for age. Significant age-adjusted partial correlation coefficients were also found between both plasma viscosity and Hct concentration with BMI. Hct showed a significant inverse association with height rather than a positive association with body mass. Both rheological parameters were significantly associated with total cholesterol, log-triglycerides, systolic and diastolic BP, and blood leukocytes. Plasma viscosity was associated with fasting blood glucose concentration.

Table 3:
Pearson correlation and partial (age-adjusted) correlation coefficients between plasma viscosity and independent variables among nonsmokers.
Table 4:
Pearson correlation and partial (age-adjusted) correlation coefficients between haematocrit concentration and independent variables among nonsmokers.

Among nonsmokers (N = 590) significant differences were found in mean plasma viscosity levels between PAI categories (P = 0.0001). The unadjusted mean (95% C I) plasma viscosity levels across the inactive, occasional/light, moderate/moderately vigorous, and vigorous PAI groups were 1.671 (1.652, 1.689), 1.660 (1.648, 1.671), 1.640 (1.629, 1.651), and 1.620 (1.604, 1.636 mPa·s1). Highly significant decreases (P = < 0.00005) in plasma viscosity levels with higher PAI categories remained following adjustment for age. Adjustment for age, BMI, total cholesterol, log-triglyceride concentration, glucose, and leukocyte count (N = 564) attenuated this trend (P = 0.184).

Significant differences in mean plasma viscosity levels (N = 595) were found between quartiles of predicted V̇O2max (L·min1 and mL·kg1·min1) (both P = < 0.000005). Mean age-adjusted plasma viscosity levels were also significantly lower with higher quartiles of predicted V̇O2max (L·min1 and mL·kg1·min1) (both P = < 0.00005). Following adjustment for age, BMI, total cholesterol, diastolic BP, log-triglyceride, and fasting glucose (N = 569) significantly lower mean plasma viscosity levels were found with higher predicted V̇O2max (L·min1 and mL·kg1·min1) scores (P = < 0.00005). These relationships were not attenuated following further adjustment by blood leukocyte count (P = < 0.00005). The fully adjusted relative differences in mean plasma viscosity between the highest and lowest quartiles of predicted V̇O2max scores (L·min1 and mL·kg1·min1) were 0.052 and 0.059 mPa·s1, respectively.

Within nonsmoking subjects (N = 632) significant differences in mean Hct concentrations were evident across PAI categories (P = 0.0066). The unadjusted mean (95% C I) Hct concentrations for the inactive, occasional/light, moderate/moderately vigorous, and vigorous PAI were 46.32 (45.73, 46.92), 45.72 (45.31, 46.13), 45.05 (44.60, 45.50), and 45.19% (44.43, 45.95%), respectively. Age-adjusted mean Hct concentrations were significantly lower with higher PAI categories (P = 0.014). Following further adjustment for BMI, total cholesterol, log-triglyceride, diastolic BP, and leukocyte count, mean Hct concentrations (N = 628) remained significantly lower with higher levels of physical activity (P = 0.044).

Hct concentrations were significantly lower with higher levels of predicted V̇O2max (L·min1 and mL·kg1·min1) (P = 0.0186 and P = 0.0004, respectively). This inverse association remained significant following age-adjustment for predicted weight-adjusted V̇O2max (mL·kg1·min1) categories (P = 0.001). A trend was also evident for lower Hct with higher predicted V̇O2max (L·min1) categories (P = 0.051).

The relationship between predicted V̇O2max (mL·kg1·min1) and lower Hct concentrations remained following adjustment for age, BMI, total cholesterol, log-transformed triglyceride concentration, diastolic BP, and blood leukocyte count (P = 0.047). The relationship between predicted V̇O2max (L·min1) and lower Hct concentrations did not achieve statistical significance following adjustment for the above confounding variables (P = 0.065).


Increased blood viscosity has been associated with the incidence of IHD events. Plasma viscosity level and Hct appear most important in determining the association between blood viscosity and IHD risk (22). A reduction in these rheological parameters may be an important mechanism through which physical activity protects against IHD events (8,10).

Plasma viscosity has been shown to be a strong independent long-term predictor of IHD risk. Within the Caerphilly and Speedwell Collaborative Heart Disease study a representative sample of >4800 middle-aged men were prospectively followed for up to 10 yr for incident cases of IHD (37). An age-standardized mean plasma viscosity level of 0.032 mPa·s1 discriminated between men with and without incident IHD events during long-term follow-up of this cohort (95% CI 0.017–0.047 mPa·s1 inclusive). At the first follow-up, the adjusted relative risk of a first major incident IHD event for an increase in plasma viscosity by 1SD (0.096 mPa·s1) was 1.33 (P. Sweetnam, personal communication cited in (16)).

The present study has examined cross-sectionally the effects of LTPA and predicted V̇O2max on plasma viscosity levels in employed middle-aged men. Age-adjusted plasma viscosity levels were inversely associated with higher levels of overall physical activity among nonsmoking middle-aged men without evidence of IHD. Mean age-adjusted plasma viscosity levels were significantly lower (0.049 mPa·s1) among the vigorously active compared with inactive nonsmokers. The association between PAI categories and plasma viscosity in the present study was attenuated following adjustment for all IHD risk indicators. Plasma viscosity levels (following adjustment for age, smoking, and pre-existing IHD) were also inversely associated with overall and high intensity leisure time energy expenditure within the first reexamination of the Caerphilly Prospective Heart Disease Study (8). Within the above study, the association between total or high intensity energy expenditure showed an interaction with employment status. Significant inverse associations with overall or high intensity energy expenditure (plasma viscosity 0.026 mPa·s1 lower in the most active third) were restricted to unemployed men who exhibited larger variability in physical activity.

Recent results from the Monica Augsburg Project (16) showed that the unadjusted relative risk of an IHD event (first ever fatal or nonfatal myocardial infarction) among middle-aged men associated with a 1-SD increase in plasma viscosity (0.07 mPa·s1) was 1.6 (95% CI 1.25–2.03). Plasma viscosity had been reported to be inversely associated with vigorous physical activity in both sexes within a representative subsample of the Augsburg cohort (13). The above study showed a significant mean difference (0.01 mPa·s1) between nonactive middle-age men and those reporting > 2 h·wk−1 physical activity, following adjustment for age, smoking status, BMI, total cholesterol, high density lipoprotein (HDL), alcohol, and years of education.

The present study has shown that predicted V̇O2max is inversely associated with plasma viscosity levels within a cohort of employed nonsmoking middle-aged men. Mean age-adjusted plasma viscosity levels were also inversely associated with predicted V̇O2max. This inverse association remained highly significant following adjustment for age, BMI, total cholesterol, log-triglycerides, diastolic BP, glucose, and leukocytes. The relative difference in adjusted mean plasma viscosity levels between the lowest and upper quartiles of predicted V̇O2max scores within this study was highly significant and substantial at 0.05 mPa·s1. Significant inverse relationships between plasma viscosity and maximum physical work capacity have been reported among sedentary adults and elite athletes (10,17,38).

Several prospective studies have reported positive associations between plasma fibrinogen and IHD risk (5). Plasma viscosity may be an important mechanism by which plasma fibrinogen (its major protein determinant) promotes IHD (37). The findings of the present study relating PAI and V̇O2max categories to plasma viscosity are comparable with those previously reported for plasma fibrinogen. Within representative studies of middle-aged men the significant inverse associations between physical activity and fibrinogen have been shown to be attenuated following adjustment for confounding IHD variables (19,25), whereas the association with directly determined or predicted V̇O2max and fibrinogen has been sustained (25,27). Other investigators have also suggested that the effect of LTPA on plasma fibrinogen was attributable to increased V̇O2max (18).

Hct concentration is the strongest determinant of whole blood viscosity, contributing up to 50% of inter-individual variation (24). This study showed a significant inverse relationship between Hct concentration and physical activity among middle-aged nonsmoking men. Higher levels of overall physical activity were associated with significant reductions in Hct concentration following adjustment for all confounding variables. Physical activity showed a significant inverse relationship with Hct concentration among middle-aged men within the representative British Regional Heart Study (40). The association between physical activity and Hct in the above study appeared to be independent of other lifestyle and physiological factors known to be associated with both variables (40). The present study also demonstrated a significant inverse association between predicted V̇O2max (mL·kg1·min1) and Hct concentration. Our findings are in contrast to several previous investigations reporting no significant differences in Hct concentrations between endurance-trained and sedentary control subjects (28). Increased erythropoiesis associated with long-term regular exercise training may account for these findings (28).

The beneficial effects of regular physical activity and higher predicted V̇O2max on the primary components of blood viscosity may result from a chronic expansion of plasma volume (13). Plasma volume changes have been demonstrated using indicator techniques following endurance training (4), causing hemodilution of the plasma (10). The reported effects of chronic physical exercise on red cell deformability and/or aggregability have been inconsistent, although this may be associated with measurement difficulties (28).

The present cross-sectional study has also confirmed associations of plasma viscosity and Hct concentration with other conventional IHD risk markers. Total cholesterol and triglyceride concentrations have been associated with increases in both Hct and plasma viscosity in several (7,20,21,24,40) but not all studies (15). In the present study age-adjusted total cholesterol and log-triglycerides were significantly associated with plasma viscosity. The association of plasma viscosity with these variables may reflect the direct effects of lipoproteins on serum viscosity (20,21,33). In contrast to others (15), the present study was not able to confirm a significant inverse association between plasma viscosity and serum HDL cholesterol. The association of Hct and both total cholesterol and triglycerides appears to be independent of lifestyle and of physiological/biochemical factors (40). In the present study a significant age-adjusted partial correlation was found between plasma viscosity and glucose, whereas most previous investigations have reported no significant association with plasma fibrinogen (27,29,30). No association of Hct with blood glucose was observed in earlier investigations (7,40). Obesity (as determined by BMI) was strongly associated with plasma viscosity and Hct concentration in the present and other investigations (7,24,40). The influence of BMI on Hct concentration appears to be independent of blood pressure and blood lipids (40).

Chronic arterial hypertension results from increased total peripheral resistance (TPR), of which plasma viscosity may be a component (21). We found both age-adjusted systolic and diastolic BP were significantly associated with plasma viscosity. Several studies have also shown significant associations between both systolic and diastolic BP with whole blood and plasma viscosity in middle-aged men, independent of confounding variables (14,21,36). The positive association between Hct concentration and blood pressure is also well documented (36,40). In the present study as in others (40), Hct appears to be associated more strongly with diastolic rather than systolic BP.

A high leukocyte count is an independent predictor of IHD, as is either plasma fibrinogen or viscosity (42). Our findings of a significant age-adjusted partial correlation between plasma viscosity and leukocyte count among nonsmokers are consistent with earlier observations for both plasma viscosity and fibrinogen (39,42). Published results from prospective studies appear remarkably consistent in associating systemic “inflammatory markers” with IHD (5). Cigarette smoking has been reported to be related to a “complex hemorheological deficit” (9), with increased Hct concentrations and blood viscosity compared with nonsmokers (21,24).

In summary, the present cross-sectional study, based on a subsample of nonsmoking middle-aged men of high socioeconomic status, showed an inverse relationship of physical activity and/or predicted V̇O2max with the primary determinants of blood viscosity.


1. Åstrand, P.-O. and K. Rodahl. Textbook of Work Physiology: Physiological Basis of Exercise, 3rd Ed. New York: McGraw-Hill, 1986, pp. 383–384.
2. Åstrand, P.-O. and I. Rhyming. Nomogram for calculation of aerobic capacity (physical fitness) from pulse rates during submaximal work. J. Appl. Physiol. 7:218–221, 1954.
3. Baker, I. A., P. M. Sweetnam, J. W. G. Yarnell, D. Bainton, and P. C. Elwood. Haemostatic and other risk factors for ischaemic heart disease and social class: evidence from the Caerphilly and Speedwell studies. Int. J. Epidemiol. 17:759–765, 1988.
4. Convertino, V. A. Blood volume: its adaptation to endurance training. Med. Sci. Sports Exerc. 23:1338–1348, 1991.
5. Danesh, J., R. Collins, P. Appelby, and R. Peto. Associations of fibrinogen, c-reactive protein, albumin, or leukocyte count with coronary heart disease (CHD): meta-analysis of prospective studies. JAMA 279:1477–1482, 1998.
6. Davey-Smith, G. and J. N. Morris. Assessment of physical activity and physical fitness in population surveys. J. Epidemiol. Community Health 46:89–91, 1992.
7. de Simone, G., R. B. Devereux, and S. Chien. Relation of blood viscosity to demographic and physiologic variables and to cardiovascular risk factors in apparently normal adults. Circulation 81:107–117, 1990.
8. Elwood, P. C., J. W. G. Yarnell, J. Pickering, et al. Exercise, fibrinogen and other risk factors for ischaemic heart disease: Caerphilly Prospective Heart Disease Study. Br. Heart J. 69:183–187, 1993.
9. Ernst, E. Review: haemorheological consequences of chronic cigarette smoking. J. Cardiovasc. Risk 2:435–439, 1995.
10. Ernst, E. Influence of regular physical activity on blood rheology. Eur. Heart J. 8(Suppl. 6):59–62, 1987.
11. Harkness, J. A new method for the measurement of plasma viscosity. Lancet ii:280–281, 1963.
12. Jacobs, D. R., Jr., B. E. Ainsworth, T. J. Hartman, and A. S. Leon. A simultaneous evaluation of 10 commonly used physical activity questionnaires. Med. Sci. Sports Exerc. 25:81–91, 1993.
13. Koenig, W., M. Sund, A. Doring, and E. Ernst. Leisure-time physical activity but not work-related physical activity is associated with decreased plasma viscosity: results from a large population sample. Circulation 95:335–341, 1997.
14. Koenig, W., M. Sund, E. Ernst, U. Keil, J. Rosenthal, and V. Hombach. Association between plasma viscosity and blood pressure: results from the MONICA project Augsburg. Am. J. Hypertens. 4:529–536, 1991.
15. Koenig, W., M. Sund, E. Ernst, W. Mraz, V. Hombach, and U. Keil. Association between rheology and components of lipoproteins in human blood: results from the MONICA project Augsburg. Circulation 85:2197–2204, 1992.
16. Koenig, W., M. Sund, B. Filipiak, A. Doring, H. Lowel, and E. Ernst. Plasma viscosity and the risk of CHD: results from the MONICA Augsburg Cohort Study, 1984–1992. Arterioscler. Thromb. 18:768–772, 1998.
17. Kovacic, J. and R. Telford. Review: blood viscosity in athletes. Excel 8:73–84, 1992.
18. Lakka, T. A. and J. T. Salonen. Moderate to high intensity conditioning leisure time physical activity and high cardiorespiratory fitness are associated with reduced plasma fibrinogen in eastern Finnish men. J. Clin. Epidemiol. 46:1119–1127, 1993.
19. Lee, A. J., W. C. S. Smith, G. D. O. Lowe, and H. Tunstall-Pedoe. Plasma fibrinogen and coronary risk factors: the Scottish Heart Health Study. J. Clin. Epidemiol. 43:913–919, 1990.
20. Lowe, G. D. O. Editorial comment: blood viscosity, lipoproteins, and cardiovascular risk. Circulation 85:2329–2331, 1992.
21. Lowe, G. D. O. Invited review article: blood viscosity and cardiovascular disease. Thromb. Haemost. 67:494–498, 1992.
22. Lowe, G. D. O., A. J. Lee, A. Rumley, J. F. Price, and F. G. R. Fowkes. Blood viscosity and risk of cardiovascular events: the Edinburgh Artery Study. Br. J. Haematol. 96:168–173, 1997.
23. Lowe, G. D. O., D. A. Wood, J. T. Douglas, et al. Relationships of plasma viscosity, coagulation, and fibrinolysis to coronary risk factors and angina. Thromb. Haemost. 65:339–343, 1991.
24. Lowe, G. D. O., W. C. S. Smith, H. D. Tunstall-Pedoe, et al. Cardiovascular risk and haemorheology: results from the Scottish Heart Health Study and the MONICA Project, Glasgow. Clin. Hemorheol. 8:518–524, 1988.
25. MacAuley, D., E. E. McCrum, G. Scott, et al. Physical activity, physical fitness, blood pressure, and fibrinogen in the Northern Ireland Health and Activity Survey. J. Epidemiol. Community Health 50:258–263, 1996.
26. McMurray R. G., B. E. Ainsworth, J. S. Harrell, T. R. Griggs, and O. Dale Williams. Is physical activity or aerobic power more influential on reducing cardiovascular disease risk factors? Med. Sci. Sports Exerc. 30:1521–1529, 1998.
27. Moller, L. and T. S. Kristensen. Plasma fibrinogen and IHD risk factors. Arterioscler. Thromb. 11:344–350, 1991.
28. Neuhaus, D. and G. Gaehtgens. Haemorrheology and long term exercise. Sports Med. 18:10–21, 1994.
29. Rankinen, T., R. Rauramaa, S. Vaisanen, I. M. Penttila, and M. Uusitupa. Relation of habitual diet and cardiorespiratory fitness to blood coagulation and fibrinolytic factors. Thromb. Haemost. 71:180–183, 1994.
30. Rauramaa, R., S. Vaisanen, I. Penttila, and T. Rankinen. Inverse relationship between dietary starch intake and plasma fibrinogen in middle-aged men. Nutr. Metab. Cardiovasc. Dis. 4:192–196, 1994.
31. Rosenson, R. S., A. McCormick, and E. F. Uretz. Distribution of blood viscosity values and biochemical correlates in healthy adults. Clin. Chem. 42:1189–1195, 1996.
32. Sallis, J. F., W. L. Haskell, and P. D. Wood. Physical activity assessment methodology in the Five-City Project. Am. J. Epidemiol. 121:91–106, 1985.
33. Seplowitz, A. H., S. Chien, and F. R. Smith. Effects of lipoproteins on plasma viscosity. Atherosclerosis 38:89–95, 1981.
34. Shaper, A. G. and G. Wannamethee. Physical activity and ischaemic heart disease in middle-aged British men. Br. Heart J. 66:384–394, 1991.
35. Siconolfi, S. F., E. M. Cullaine, R. A. Carleton, and P. D. Thompson. Assessing V̇O2max in epidemiological studies: a modification of the Åstrand Rhyming test. Med. Sci. Sports Exerc. 14:335–338, 1982.
36. Smith, W. C. S., G. D. O. Lowe, A. J. Lee, and H. Tunstall-Pedoe. Rheological determinants of blood pressure in a Scottish adult population. J. Hypertens. 10:467–472, 1992.
37. Sweetnam, P. M., H. F. Thomas, J. W. G. Yarnell, A. D. Beswick, I. A. Baker, and P. C. Elwood. Fibrinogen, viscosity, and the 10-year incidence of ischaemic heart disease: the Caerphilly and Speedwell studies. Eur. Heart J. 17:1814–1820, 1996.
38. Telford, R. D., J. C. Kovacic, S. L. Skinner, J. B. Hobbs, A. G. Hahn, and R. B. Cunningham. Resting whole blood viscosity of elite rowers is related to performance. Eur. J. Appl. Physiol. 68:470–476, 1994.
39. Vaisanen, S., R. Rauramaa, I. Penttila, et al. Variation in plasma fibrinogen over one year: relationships with genetic polymorphisms and nongenetic factors. Thromb. Haemost. 77:884–889, 1997.
40. Wannamethee, G. and A. G. Shaper. Haematocrit: relationships with blood lipids, blood pressure, and other cardiovascular risk factors. Thromb. Haemost. 72:58–64, 1994.
41. Wannamethee, G., A. G. Shaper, and P. H. Whincup. Ischaemic heart disease: association with haematocrit in the British Regional Heart Study. J. Epidemiol. Comm. Health. 48:112–118, 1994.
42. Yarnell, J. W. G., I. A. Baker, P. M. Sweetnam, et al. Fibrinogen, viscosity, and white blood cells are major risk factors for ischaemic heart disease: the Caerphilly and Speedwell Collaborative Heart Disease Studies. Circulation 83:836–844, 1991.


©2000The American College of Sports Medicine