Coagulation and Fibrinolytic Responses to Manual versus Automated Snow Removal


Medicine & Science in Sports & Exercise: October 2003 - Volume 35 - Issue 10 - pp 1755-1759
APPLIED SCIENCES: Physical Fitness and Performance

WOMACK, C. J., C. M. PATON, A. M. COUGHLIN, P. R. NAGELKIRK, A. T. DEJONG, J. ANDERSON, and B. A. FRANKLIN. Coagulation and Fibrinolytic Responses to Manual versus Automated Snow Removal. Med. Sci. Sports Exerc., Vol. 35, No. 10, pp. 1755–1759, 2003.

Purpose: The purpose of this study was to assess coagulation and fibrinolytic responses to snow removal.

Methods: Thirteen healthy male subjects (age = 26 ± 5 yr, height = 179.0 ± 7.0 cm, weight = 78.7 ± 16.1 kg, V̇O2max = 54.7 ± 8.9 mL·kg−1·min−1) underwent maximal treadmill stress testing (TM), 10 min of snow shoveling (SS), and 10 min of snow removal using an automated snow thrower (ST). Blood was collected immediately before and after each test and analyzed for von Willebrand Factor antigen (vWF:ag), tissue plasminogen activator (tPA) antigen, and plasminogen activator inhibitor-1 (PAI-1) activity. Data were analyzed using a two-factor repeated-measures analysis of variance.

Results: vWF:ag significantly increased during TM (84.7 ± 21.7% normal preexercise, 149.0 ± 45.6% normal postexercise) but not SS or ST. TM resulted in significant increases in tPA antigen (6.54 ± 2.76 ng·mL−1 preexercise, 21.39 ± 10.56 ng·mL−1 postexercise) and both SS and TM caused significant reductions in PAI-1 activity (SS = 15.1 ± 3.8 AU·mL−1 preexercise, 13.2 ± 4.3 AU·mL−1 postexercise; TM = 15.3 ± 3.6 AU·mL−1 preexercise, 10.5 ± 5.3 AU·mL−1 postexercise). Postexercise PAI-1 activity was significantly lower for TM versus SS. tPA antigen was unchanged after SS and ST, and PAI-1 activity was unchanged after ST.

Conclusion: vWF:ag is unchanged after self-paced snow shoveling and automated snow removal in young, healthy males. Snow shoveling acutely increases fibrinolytic potential in this population, although not to the degree observed after maximal treadmill exercise.

Typically, hemorrhage or ulceration of atherosclerotic plaque initiates an acute cardiovascular event. Thrombus formation inside the ulcerated plaque or in the atherosclerotic vessel ultimately occludes the lumen. However, plaque ruptures have been observed in autopsy studies of individuals without a history of cardiovascular events (5,19), suggesting occlusive thrombus formation does not always occur with plaque rupture. Therefore, blood coagulation potential and/or the potential to lyse clot (fibrinolysis) are likely important determinants in this cascade of acute cardiovascular events (and their sequelae). Accordingly, increased coagulation potential and decreased fibrinolysis are associated with acute cardiovascular events (4,12). Furthermore, potential for coagulation is increased (20,23,26) and fibrinolysis decreased (10,13,16,21) in patients with cardiovascular disease (CVD).

Vigorous exercise transiently increases the risk of acute cardiovascular events, especially in habitually sedentary persons with documented or occult CVD (22,25). Coronary artery disease (CAD) is the leading cause of death during exercise in adults (22), and physical activity has been noted as a trigger of infarction in up to one third of patients able to identify a possible trigger (25). The transient increase in cardiovascular events during exercise is likely due, at least in part, to an altered coagulation and/or fibrinolytic exercise response, as myocardial infarctions triggered by physical stress are more likely to occur due to an occlusive thrombus (9). Furthermore, acute exercise results in a larger increase in coagulation potential in patients with CVD (20).

The rate of death from cardiovascular events is higher in the winter months (1,3), which may be partially due to snow shoveling (11). The hemodynamic responses to heavy snow shoveling are similar to those seen during maximal treadmill testing (8), suggesting disproportionate and potentially threatening myocardial demands. The high degree of muscular effort during snow shoveling may predispose at-risk individuals to exertion-related myocardial infarction as the majority of these events occur during or immediately after physical tasks that require significant lifting or isometric contraction (9). It is not known whether this increased risk is related to altered coagulation and/or fibrinolytic responses, although El-Sayed (6) reported robust increases in both factor VIII, a marker of coagulation potential, and fibrinolysis after acute strength training. If snow shoveling causes similar hemostatic responses, this could explain the high rate of cardiovascular events that are triggered by this activity.

Although the hemostatic response to automated snow removal has not been evaluated, use of an automated snow thrower has been shown to result in an attenuated hemodynamic and metabolic response as compared with snow shoveling (8,24). The purpose of the proposed study was to determine changes in thrombotic and fibrinolytic potentials during snow shoveling versus maximal treadmill exercise and automated snow removal. We hypothesize that snow shoveling elicits coagulation and fibrinolytic responses similar to those observed after maximal treadmill exercise and that automated snow removal decreases these hemostatic responses.

1Human Energy Research Laboratory, Department of Kinesiology, Michigan State University, East Lansing, MI; and

2Division of Cardiology, Cardiac Rehabilitation and Exercise Laboratories, William Beaumont Hospital, Royal Oak, MI

Address for correspondence: Christopher J. Womack, Ph.D., 3 IM Sports Circle, Michigan State University, East Lansing, MI 48824; E-mail:

Submitted for publication December 2002.

Accepted for publication June 2003.

Article Outline
Back to Top | Article Outline


Approach to the problem and experimental design.

This study utilized a repeated measures design with mode of physical stress (maximal treadmill exercise, snow shoveling, and automated snow removal) as the independent variable and von Willebrand Factor antigen (vWF:ag), tissue plasminogen activator (tPA), and plasminogen activator inhibitor-1 (PAI-1) as the dependent variables. The dependent variables were selected because they are commonly used markers of thrombotic potential (vWF:ag) and fibrinolytic potential (tPA and PAI-1). It is unknown whether snow shoveling results in hemostatic changes that favor coagulation, which could increase risk for an acute cardiovascular event. Therefore, thrombotic and fibrinolytic responses to snow shoveling were compared with maximal treadmill exercise, a mode known to elicit large increases in both vWF:ag and fibrinolysis (30). If snow shoveling causes acute hemostatic responses that favor coagulation, alternative modes of snow removal that decrease the coagulation response would have important clinical significance. Thus, we also evaluated the use of an automated snow thrower on our selected dependent variables.

Back to Top | Article Outline


Thirteen young, healthy male subjects (age = 26 ± 5 yr, height = 179.0 ± 7.0 cm, weight = 78.7 ± 16.1 kg, V̇O2max = 54.7 ± 8.9 mL·kg−1·min−1) were studied. To eliminate potential confounding variables that could influence coagulation or fibrinolysis, subjects were free of known cardiovascular or metabolic disease, did not use tobacco, and were not on any medications at the time of this study. Written informed consent was obtained from each subject and the protocol was approved by the Michigan State University Committee on Research in Human Subjects.

Back to Top | Article Outline


Each subject completed the following tests on three separate occasions at the same time of day. Immediately before each test, subjects assumed a seated resting position for 15 min before the initial blood collection. Blood was collected immediately after the tests in the same resting position. Subjects fasted for 12 h before each test and all tests were performed between 8:00 and 11:00 a.m.

Back to Top | Article Outline

Maximal treadmill test (TM).

Subjects performed a ramped treadmill exercise test starting at a speed of 2.5 mph and 0.0% elevation. The speed continuously increased at a rate of 0.5 mph per minute until 6.0 mph was achieved, at which point grade was increased by 3% per minute until volitional exhaustion. During the test, oxygen consumption (V̇O2) was continuously monitored using a SensorMedics 2900 Metabolic Measurement Cart, which was calibrated before each test with gases of known concentration. Heart rate (HR) was assessed using a Polar® HR monitor. All subjects achieved a minimum of two of the following criteria: 1) attainment of ≥ 90% of age-predicted maximal HR and 2) respiratory exchange ratio (RER) ≥ 1.15 and maximal blood lactate concentration ≥ 8.0 mmol·L−1.

Back to Top | Article Outline

Snow-shoveling test.

The snow-shoveling tests (SS) were conducted outdoors in an average temperature of 6.5 ± 2.8°C. To standardize the density of the snow, ice shavings from a Zamboni machine were used. The snow was spread to an approximate depth of 25 cm. Subjects were instructed to clear the snow from a driveway at their own pace for 10 min. During each minute of snow shoveling, HR was obtained using a Polar® HR monitor.

Back to Top | Article Outline

Automated snow removal test (ST).

Testing conditions were identical to the SS test, except subjects used a commercially available (Toro Model CCR 3650) automated snow thrower to clear the snow. During each minute of testing, HR was obtained using a Polar® HR monitor. Average temperature during the ST test was 12.4 ± 6.2°C.

Back to Top | Article Outline

Blood samples.

Fasting venous blood samples were drawn into a tube prepared with an acidified citrate solution (Biopool International, Ventura, CA). Samples were immediately spun at 10,000 rpm for 20 min at 4°C to obtain platelet-poor plasma that was stored at −80°C until assayed. vWF:ag was assayed using the Laurell rocket technique for gel electrophoresis (14). tPA antigen was assayed using an enzyme-linked immunoadsorbent assay (American Diagnostica, Inc; Greenwich, CT). PAI-1 activity was assayed using an amidolytic activity assay (15). Approximately 0.5 mL of each whole blood sample was used for determination of blood lactate concentration using a YSI 2300 Stat Plus analyzer and hematocrit using the microhematocrit technique. Hematocrits were used to estimate changes in plasma volume (27), and all postexercise values were corrected for plasma volume changes.

Back to Top | Article Outline

Statistical analyses.

The effects of exercise mode (SS, ST, and TM) and exercise time (pre, post) on vWF:ag, tPA, PAI-1, and blood lactate concentration were assessed using a two-factor repeated measures analysis of variance. A one-way repeated measures ANOVA was used to compare peak HR obtained during the SS, ST, and TM tests. Post hoc means comparisons were performed using Tukey’s HSD tests. A priori statistical significance was set at P < 0.05. Effect sizes (0.75 for vWF:ag, 1.03 for tPA antigen, and 1.21 for PAI-1 activity) for statistical power were calculated based on previous, unpublished data from our laboratory evaluating vWF:ag, tPA, and PAI-1 responses to high-intensity versus low-intensity exercise. This was based on our a priori assumption that differences in vWF:ag or fibrinolytic variables between modes would occur due to differing levels of exertion. For 14 subjects, statistical power was determined to be 0.74 for vWF:ag, 0.94 for tPA antigen, and 0.98 for PAI-1 activity.

Back to Top | Article Outline


Average maximal HR from the TM test (187 ± 7 beats·min−1) was significantly higher than the peak HR achieved during the SS (144 ± 25 beats·min−1) or ST (117 ± 15 beats·min−1) tests (P < 0.05). Peak HR during the SS test was significantly higher than the ST test (P < 0.05). Peak HR during SS and ST averaged 77% and 62% of the maximum values achieved during the TM test, respectively.

Figure 1 displays average pre- and postexercise blood lactate concentrations for the three tests. There were significant increases in blood lactate concentration during the TM (preexercise = 0.77 ± 0.34 mmol·L−1, postexercise = 9.15 ± 1.56 mmol·L−1) and SS (preexercise = 0.72 ± 0.28 mmol·L−1, postexercise = 4.79 ± 2.23 mmol·L−1) tests but not the ST (preexercise = 0.72 ± 0.36 mmol·L−1, postexercise = 1.81 ± 0.88 mmol·L−1) test. Postexercise blood lactate concentration was significantly higher for the TM test than the SS, and ST tests and significantly higher for the SS versus the ST test.

Figure 2 displays mean vWF:ag, which significantly increased during the TM test (preexercise = 84.7 ± 21.7% normal, postexercise = 149.0 ± 45.6% normal) but not during the SS test or the ST test. tPA antigen (Fig. 3) significantly (P < 0.05) increased during the TM (preexercise = 6.54 ± 2.76 ng·mL−1, postexercise = 21.39 ± 10.56 ng·mL−1) test but not the SS (preexercise = 7.04 ± 2.57 ng·mL−1, postexercise = 11.61 ± 6.57 ng·mL−1) or ST (preexercise = 7.79 ± 3.62 ng·mL−1, postexercise = 7.59 ± 3.09 ng·mL−1) tests. As shown in Figure 4, there were significant reductions in PAI-1 activity for both SS (preexercise = 15.1 ± 3.8 AU·mL−1, postexercise = 13.2 ± 4.3 AU·mL−1) and TM (preexercise = 15.3 ± 3.6 AU·mL−1, postexercise = 10.5 ± 5.3 AU·mL−1). Postexercise PAI-1 activity was significantly lower for TM than SS. No significant change in plasma PAI-1 was observed for ST.

Back to Top | Article Outline


The major finding of this study is that snow shoveling does not increase vWF:ag but increases fibrinolytic potential in young healthy males. Snow shoveling is an activity that results in significant muscular strain and isometric contraction. Furthermore, the majority of exertion-related cardiovascular events occur during activities that require considerable muscular strain (9). Based on prior data from El-Sayed (6) that demonstrated increases in factor VIII and fibrinolysis in response to heavy lifting, we hypothesized that snow shoveling would elicit similar or greater alterations in hemostatic variables as compared with maximal treadmill exercise. However, maximal treadmill exercise was the only mode that evoked significant elevations in vWF:ag in our study. Furthermore, maximal treadmill exercise resulted in greater elevations in fibrinolytic potential than snow shoveling.

The lack of change in vWF:ag during snow shoveling is likely due to a lower relative exertion than anticipated in these subjects. Franklin et al. (8) reported HR responses corresponding to approximately 98% of maximal HR during heavy snow shoveling of similar duration, which is noticeably higher than the average of 77% max HR observed in the present study. Although the subjects in the former study were similar with respect to age and training status, their V̇O2max was approximately 40% lower than the subjects in the present study. Furthermore, Franklin et al. (8) observed a significant correlation between V̇O2max and the HR response to snow shoveling, suggesting that our unintentional selection of subjects with higher aerobic capacity caused a lower hemodynamic response to the snow shoveling trial.

What is perhaps more intriguing is that vWF:ag did not significantly increase during snow shoveling despite an almost sevenfold increase in blood lactate concentration (Fig. 1). Previous data have shown that factor VIII and vWF:ag significantly increase during exercise at a submaximal intensity close to the lactate threshold (LT) (2). This association between blood lactate concentration and factor VIII is likely due to plasma epinephrine, which has been reported to increase in a threshold-type pattern close to LT (17,29). Beta-adrenergic stimulation and blockade respectively stimulates and inhibits the release of vWF from endothelial cells (28), suggesting that sympathetic stimulation is also related to the vWF:ag response to snow shoveling. Therefore, snow shoveling may not result in large enough increases in sympathetic nervous system activity to significantly affect vWF:ag in this population, despite the significant increase in blood lactate concentration.

As vWF:ag was the only marker of coagulation potential evaluated in the present study, it is possible that other markers such as factor VIII activity or thrombin-antithrombin complex are elevated during snow shoveling. However, previous data have shown that vWF:ag increases similarly to factor VIII activity during acute exercise (30). Furthermore, because thrombin antithrombin is an indicator of thrombin formation, and vWF:ag an indicator of coagulation potential, it is likely that the marker used in the present study is more sensitive to any acute change in coagulation potential. The lack of change in vWF:ag does suggest that if there was any increase in coagulation potential during snow shoveling, it was, at best, a modest increase.

Both coagulation and fibrinolysis typically increase during acute exercise. Interestingly, we found that snow shoveling significantly increased fibrinolytic potential, as reflected by a decreased PAI-1 activity, without increases in vWF:ag. Furthermore, there was a nonsignificant 64% increase in tPA antigen during snow shoveling. This suggests a cardioprotective effect against persistent thrombus formation in this population during snow shoveling. This finding is clinically relevant, in that myocardial infarctions triggered by strenuous physical exertion are more likely to occur due to an occlusive thrombus (9). Paradoxically, snow shoveling may acutely decrease the risk of occlusive thrombus formation in younger, healthy subjects.

Caution should obviously be taken in extrapolating these findings to older populations with functional limitations, concomitant CVD, or both. Sheldahl et al. (24) observed that both age and presence of CVD increase the relative metabolic demand of self-paced snow shoveling. Furthermore, greater increases in coagulation potential during acute exercise have been reported for patients with CVD versus age-matched healthy individuals (20). It may be that the higher relative metabolic cost of snow shoveling in older, less-fit individuals with CVD would result in substantially increased coagulation potential during snow shoveling. Future research should evaluate this possibility.

Use of an automated snow thrower resulted in a lower blood lactate accumulation and HR as compared with snow shoveling. In addition, coagulation and fibrinolysis were unaffected during automated snow removal. The lower HR observed in our population supports previous observations that use of an automated snow thrower significantly reduces HR relative to manual snow shoveling (8,24). However, Sheldahl et al. (24) did not observe significant differences in HR between snow shoveling and automated snow removal in older subjects with CAD. Therefore, although our results suggest that there are no hemostatic perturbations during automated snow removal in younger subjects, it remains unclear whether the higher relative metabolic cost of automated snow removal in older subjects with CAD would result in significantly elevated coagulation and/or fibrinolysis. Recently, acute myocardial infarction was reported in two previously sedentary men (58 and 64 yr of age) who engaged in automated snow removal during the early morning hours (7).

The mean average ambient temperature was 5.9°C higher during automated snow removal compared with snow shoveling, which could potentially alter factors related to coagulation and fibrinolysis (18). However, as higher temperatures would theoretically decrease coagulation potential and/or increase fibrinolytic potential, we would have expected higher resting tPA and/or lower PAI-1 during automated snow removal if the temperature had a significant impact. We found no significant differences in these variables and furthermore found that vWF:ag, tPA, and PAI-1 were unaffected after the bouts of automated snow removal, suggesting that the difference in ambient temperature did not impact our findings.

In summary, results from the present study suggest that self-paced snow shoveling results in a positive fibrinolytic response without increases in vWF:ag in young, healthy males. In addition, use of an automated snow thrower does not increase vWF:ag or fibrinolysis and significantly reduces the HR and blood lactate responses as compared with snow shoveling. Future research should evaluate coagulation and fibrinolytic responses to snow shoveling and automated snow removal in older individuals and patients with CVD.

This research was supported by a grant from the Toro Corporation. The results of this study do not constitute endorsement of the product by the authors or ACSM.

Back to Top | Article Outline


1. Anderson, T. W., and C. Rochard. Cold snaps, snowfall and sudden death from ischemic heart disease. Can. Med. Assoc. J. 121: 1580–1583, 1979.
2. Andrew, M., C. Carter, H. O’Brodovich, and G. Heigenhauser. Increases in factor VIII complex and fibrinolytic activity are dependent on exercise intensity. J. Appl. Physiol. 60: 1917–1922, 1986.
3. Baker-Blocker, A. Winter weather and cardiovascular mortality in Minneapolis-St. Paul. Am. J. Public Health 72: 261–265, 1982.
4. Catto, A. J., A. M. Carter, J. H. Barrett, J. Bamford, P. J. Rice, and P. J. Grant. von Willebrand factor and factor VIII: C in acute cerebrovascular disease: relationship to stroke subtype and mortality. Thromb. Haemost. 77: 1104–1108, 1997.
5. Davies, M. J., J. M. Bland, J. R. Hangartner, A. Angelini, and A. C. Thomas. Factors influencing the presence or absence of acute coronary artery thrombi in sudden ischaemic death. Eur. Heart J. 10: 203–208, 1989.
6. El-Sayed, M. S. Fibrinolytic and hemostatic parameter response after resistance exercise. Med. Sci. Sports Exerc. 25: 597–602, 1993.
7. Franklin, B. A., P. George, R. Henry, S. Gordon, G. C. Timmis, and W. W. O’Neill. Acute myocardial infarction after manual or automated snow removal. Am. J. Cardiol. 87: 1282–1283, 2001.
8. Franklin, B. A., P. Hogan, K. Bonzheim, et al. Cardiac demands of heavy snow shoveling. JAMA 273: 880–882, 1995.
9. Giri, S., P. D. Thompson, F. J. Kiernan, et al. Clinical and angiographic characteristics of exertion-related acute myocardial infarction. JAMA 282: 1731–1736, 1999.
10. Grzywacz, A., W. Elikowski, P. Psuja, M. Zozulinska, and K. Zawilska. Impairment of plasma fibrinolysis in young survivors of myocardial infarction with silent ischaemia. Blood Coagul. Fibrinolysis 9: 245–249, 1998.
11. Heppell, R., S. K. Hawley, and K. S. Channer. Snow shoveller’s infarction. Br. Med. J. 302: 469–470, 1991.
12. Jansson, J. H., T. K. Nilsson, and O. Johnson. von Willebrand factor in plasma: a novel risk factor for recurrent myocardial infarction and death. Br. Heart J. 66: 351–355, 1991.
13. Killewich, L. A., A. W. Gardner, R. F. Macko, et al. Progressive intermittent claudication is associated with impaired fibrinolysis. J. Vasc. Surg. 27: 645–650, 1998.
14. Laurell, C. B. Quantitative estimation of proteins by electrophoresis in agarose gel containing antibodies. Anal. Biochem. 15: 45–52, 1966.
15. Leprince, P., B. Rogister, and G. Moonen. A colorimetric assay for the simultaneous measurement of plasminogen activators and plasminogen activator inhibitors in serum-free conditioned media from cultured cells. Anal. Biochem. 177: 341–346, 1989.
16. Macko, R. F., S. F. Ameriso, A. Gruber, et al. Impairments of the protein C system and fibrinolysis in infection-associated stroke. Stroke 27: 2005–2011, 1996.
17. Mazzeo, R. S., and P. Marshall. Influence of plasma catecholamines on the lactate threshold during graded exercise. J. Appl. Physiol. 67: 1319–1322, 1989.
18. Mercer, J. B., B. Osterud, and T. Tveita. The effect of short-term cold exposure on risk factors for cardiovascular disease. Thromb. Res. 95: 93–104, 1999.
19. Muller, J. E., P. G. Kaufmann, R. V. Luepker, M. L. Weisfeldt, P. C. Deedwania, and J. T. Willerson. Mechanisms precipitating acute cardiac events: review and recommendations of an NHLBI workshop. National Heart, Lung, and Blood Institute. Mechanisms Precipitating Acute Cardiac Events Participants. Circulation 96: 3233–3239, 1997.
20. Mustonen, P., M. Lepantalo, and R. Lassila. Physical exertion induces thrombin formation and fibrin degradation in patients with peripheral atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 18: 244–249, 1998.
21. Paramo, J. A., I. Olavide, J. Barba, et al. Long-term cardiac rehabilitation program favorably influences fibrinolysis and lipid concentrations in acute myocardial infarction. Haematologica 83: 519–524, 1998.
22. Ragosta, M., J. Crabtree, W. Q. Sturner, and P. D. Thompson. Death during recreational exercise in the State of Rhode Island. Med. Sci. Sports Exerc. 16: 339–342, 1984.
23. Rice, G. I., and P. J. Grant. FVIII coagulant activity and antigen in subjects with ischaemic heart disease. Thromb. Haemost. 80: 757–762, 1998.
24. Sheldahl, L. M., N. A. Wilke, S. Dougherty, and F. E. Tristani. Snow blowing and shoveling in normal and asymptomatic coronary artery diseased men. Int. J. Cardiol. 43: 233–238, 1994.
25. Tofler, G. H., P. H. Stone, M. Maclure, et al. Analysis of possible triggers of acute myocardial infarction (the MILIS study). Am. J. Cardiol. 66: 22–27, 1990.
26. Tracy, R. P., A. M. Arnold, W. Ettinger, L. Fried, E. Meilahn, and P. Savage. The relationship of fibrinogen and factors VII and VIII to incident cardiovascular disease and death in the elderly: results from the cardiovascular health study. Arterioscler. Thromb. Vasc. Biol. 19: 1776–1783, 1999.
27. Van Beaumont, W., J. E. Greenleaf, and L. Juhos. Disproportional changes in hematocrit, plasma volume, and proteins during exercise and bed rest. J. Appl. Physiol. 33: 55–61, 1972.
28. von Kanel, R., J. E. Dimsdale, K. A. Adler, E. Dillon, C. J. Perez, and P. J. Mills. Effects of nonspecific b-adrenergic stimulation and blockade on blood coagulation in hypertension. J. Appl. Physiol. 94: 1455–1459, 2003.
29. Weltman, A., C. M. Wood, C. J. Womack, et al. Catecholamine and blood lactate responses to incremental rowing and running exercise. J. Appl. Physiol. 76: 1144–1149, 1994.
30. Wheeler, M. E., G. L. Davis, W. J. Gillespie, and M. M. Bern. Physiological changes in hemostasis associated with acute exercise. J. Appl. Physiol. 60: 986–990, 1986.


©2003The American College of Sports Medicine