Regular Leisure Time Physical Activity Predicts High Activity of Tissue Plasminogen Activator: The Northern Sweden MONICA Study

Transcription

1 International Journal of Epidemiology International Epidemiological Association 1996 Vol. 25, No. 6 Printed in Great Britain Regular Leisure Time Physical Activity Predicts High Activity of Tissue Plasminogen Activator: The Northern Sweden ONICA Study ATS ELIASSON,*" KJELL ASPLUND" AND PER-ERIC EVRIN* Eliasson (Department of edicine, Lulea Hospital, S Lulea, Sweden), Asplund K and Evrin P-E. Regular leisure time physical activity predicts high activity of tissue plasminogen activator: The Northern Sweden ONICA Study. International Journal of Epidemiology 1996; 25: Background. Leisure time physical activity protects against the development of cardiovascular disease, partly by lowering blood pressure, cholesterol and body weight. Little is known about the effects of regular exercise on fibrinolytic variables in the population. ethods. In a population sample of 733 men and 774 women aged years physical activity during leisure time was related to fibrinolytic variables. Results. The activity of tissue plasminogen activator (tpa) increased linearly with greater physical activity, the difference between sedentary and most active subjects being 28.9% in men and 11.6% in women. The tpa mass concentrations decreased by 27.4% and 28.0% in men and women respectively, as did also plasminogen activator inhibitor (PAI-1) activity; 38.5% and 30.6%. Tests for trend were significant (P < 0.001) for all but tpa activity in women. Adjusting for age, body mass index and waisthip ratio only slightly decreased these relationships. hen taking triglyceride into account, tpa activity and PAI-1 activity (in men) were no longer significantly related to physical activity level but lower tpa mass concentrations and PAI-1 activity (in women) were still found in those who exercised regularly. Further adjustment for insulin levels abolished all differences, except for PAI-1 activity in women. Conclusion. Greater leisure time physical activity is associated with an increased fibrinolytic activity. This may contribute to less cardiovascular disease in subjects who exercise regularly. Keywords: physical activity, exercise, fibrinogen, tissue plasminogen activator, plasminogen activator inhibitor type 1, ONICA The risk of cardiovascular disease may depend more on lifestyle factors than has hitherto been realized. In men, a sedentary lifestyle is associated with increased morbidity from ischaemic heart disease 1 " 3 and stroke, 4 ' 5 to a varying degree explained by effects on body mass, serum cholesterol or blood pressure. Few studies include women, but low physical activity has been shown to be related to an increased risk of both myocardial infarction and stroke in middle-aged women, 6-7 although the effect on stroke was not independent of other risk factors in the Framingham Study. 8 In prospective studies, impaired fibrinolysis has been implicated in the pathogenesis of myocardial * Department of edicine, Lulea Hospital, S Lulea, Sweden. * Department of edicine, UmeS University Hospital, Sweden. * Department of Clinical Chemistry, Boden Hospital, Boden, Sweden. infarction 9 " and ischaemic stroke. 12 The activity of tissue plasminogen activator (tpa) and plasminogen activator inhibitor type 1 (PAI-1) is strongly related to cardiovascular risk factors, established or proposed, such as hypertension, high triglyceride levels, obesity 13 and fasting insulin, 14 factors that are also influenced by regular physical activity. 3 ' 6 Adverse changes in fibrinolysis may thus be a link between a sedentary lifestyle and athero-thrombotic disease. The acute effect of exercise is an increased fibrinolytic activity Less is known about the influence of regular leisure time physical activity in the general population although an inverse relationship between tpa mass concentrations and physical fitness was found in middle-aged men. 17 Our aim was to study fibrinolytic variables in relation to physical activity in a randomly selected population sample. This study is unique in that it includes a 1182

2 PHYSICAL ACTIVITY, FIBRINOGEN AND FIBRINOLYSIS 1183 large population-based sample of both men and women, with a broad age range and includes measurements of specific components of the fibrinolytic system. ATERIAL AND ETHODS The study was performed within the framework of the Northern Sweden ONICA Project which is part of the HO ONICA (onitoring of Trends and Determinants in Cardiovascular Disease) Project. In 1990, a population was screened for cardiovascular risk factors. A total of 2000 individuals in the year range were invited to participate. ithin each age group (25-34, 35-44, 45-54, years) 250 men and 250 women were randomly selected from continuously updated population registers in Norrbotten and Vasterbotten, the two northernmost provinces of Sweden. If they did not attend the examination, a reminder with a new appointment was sent. People who still did not come were contacted by telephone to ascertain reasons for not attending and to obtain basic information on their social background and risk factors. 18 In all, 1583 subjects participated in the study (79% of those invited). e excluded 22 who had had a myocardial infarction and 19 who had suffered a stroke because of the possible influence on physical activity. Complete datasets regarding tpa activity, tpa mass concentrations and PAI-1 activity were available on 733 men and 774 women. The Northern Sweden ONICA Study has been approved by the Research Ethics Committee of Umea University. Blood Sampling Participants were instructed not to use tobacco during the hours preceding the examination which took place between 7 a.m. and 3 p.m. No restrictions were given with regard to physical activity. Sampling and examination of the participants have been reported. 13 ' 19 Serum triglycerides (from 423 men and 469 women) and insulin (from 400 men and 439 women) were only sampled in subjects who had fasted for 12 hours. Insulin measurements from known diabetics were not included. The tpa activity and PAI-1 activity were determined by chromogenic assays; Spectrolyse/fibrin and Spectrolyse/pl kit (Biopool AB, UmeS, Sweden) respectively. The tpa mass concentrations were measured by an ELISA method (TintElize tpa, Biopool AB, Umea, Sweden). Details regarding assay imprecision and analytical sensitivity have been published. 19 Serum triglyceride levels were determined by an enzymatic method (Boehringer annheim GmbH, annheim, Germany) and serum insulin by a radioimmunoassay (Phadeseph Insulin R1A, Pharmacia Diagnostics AB, Uppsala, Sweden). Classification of Physical Activity Using a questionnaire, self-reported physical activity during leisure time was coded to a four-point scale: the first group took no or only light physical activity <2 h per week (sedentary), the second group engaged in light physical activity (walking, bicycle riding) 2=2 h per week (light activity). The third group was moderately physically active (playing tennis, jogging, swimming) <3 h per week (moderate activity) and those in the last group were either moderately active >3 h per week, did strenuous exercise (cross-country skiing, soccer or corresponding) several times a week or were regular competitors in endurance sports (strenuous activity). Statistics ean values and 95% confidence intervals (CI) for risk factors across groups of increasing physical activity are given and significance testing for differences between groups was done by analysis of variance (ANOVA). No correction for multiple comparisons was applied. As the distributions for tpa activity, tpa mass concentrations and PAI-1 activity, as well as insulin and triglyceride levels were skewed, we report means for these variables logarithmically transformed (i.e. geometric means) which also improved homogeneity of variance. Analysis of covariance was used to stepwise adjust for previously known predictors of fibrinolytic variables. hen comparing levels of fibrinolytic variables across groups of physical activity, a test for trend was carried out by means of a linear regression analysis. Two-tailed significance tests were used. The computer program SPSS for indows, 6.0, was used. RESULTS omen reported less physical activity than men (Table I). A greater proportion of women (77.2%) took no or only light activity, compared to 65.6% of men. hile 17.1% of men reported regular strenuous exercise, only 5.2% of women participated in such activities. ore active subjects were younger and had lower body mass index (BI) and waist:hip ratio. Triglyceride and fasting insulin levels decreased with higher levels of activity, as did diastolic blood pressure. Forty subjects had diabetes and 111 subjects were treated for hypertension. Thirty-five women used oestrogen replacement therapy and 48 used contraceptive pills. In both men and women, tpa activity increased with more intense physical activity, although only significantly so in men (Table 2). The increase from the most sedentary to the most active subjects amounted to 0.20 IU/ml (95% CI : 0.10, 0.29) or 28.9% in men and 0.10 IU/ml (95% CI : -0.07, 0.24) or 11.6% in women.

3 1184 INTERNATIONAL JOURNAL OF EPIDEIOLOGY TABLE 1 ean risk factor levels (and 95% confidence intervals) by leisure time physical activity in men and women aged 25 to 64 years Gender Light activity oderate activiy Strenuous activity P-value" n Age (years) ; ; ; ; ; ; ; ; Body mass index (kg/m 2 ) ; ; ; ; ; ; ; ; aistihip ratio ; ; ; ; ; ; ; ; Triglycerides (mmol/l) b ; ; ; ; ; ; ; ; Insulin (mu/l) c ; ; ; ; ; ; ; ; Diastolic blood pressure (mmhg) ; ; ; ; ; ; ; ; " Significance testing for differences between groups by ANOVA. b Triglyceride values from 423 men and 469 women. c Serum insulin values from 400 men and 439 women. Adjustment for age, BI and waist:hip ratio slightly attenuated this relationship but when triglyceride levels were also taken into account, the test for trend was no longer significant in men. Exchanging triglycerides for insulin gave the same result (data not shown). The tpa mass concentrations decreased linearly and significantly with each level of higher physical activity (Table 3). subjects had higher levels than strenuously active participants; in men the difference was 2.0 ug/1 (95% CI: 1.1, 2.9) or 27.4% and in women 1.6 ug/1 (95% CI: 0.6, 2.8) or 28.0%. The trend decreased only slightly if age, BI and waist:hip ratio were adjusted for. After also taking triglyceride levels into account, the trend for lower tpa mass concentrations with increasing activity remained significant, but adjusting further for fasting insulin levels abolished this (data not shown). In both genders, PAI-1 activity decreased with increasing levels of physical activity (Table 4). The PAI-1 activity was 3.0 U/ml (95% CI: 1.5, 4.8) lower or 38.5% in the most active men compared to the most sedentary men (Table 4). In women the decrease was 1.9 U/ml (95% CI: 0.5, 3.8), 30.6%. In men, adjusting for anthropometry and age made little difference for the association between activity and PAI-1, while also taking triglycerides or serum insulin levels into consideration rendered the trend insignificant. In women, the relationship was more robust and still significant after adjusting for age, BI, waistrhip ratio, triglyceride and insulin levels (/ = 0.03, test for trend). DISCUSSION This is the first study showing that self-reported leisure time physical activity is a predictor of fibrinolytic activity in a randomly selected large population comprising both men and women, with a high incidence of cardiovascular disease. 20 The tpa activity, was greater

4 PHYSICAL ACTIVITY, FIBRJNOGEN AND FIBRINOLYSIS 1185 TABLE 2 Tissue plasminogen activator (tpa) activity according to leisure time physical activity. Geometric means (95% confidence intervals). The number of subjects in each group is similar to Table I Light activity oderate activity Strenuous activity P-value* en tpa activity (IU/ml) BI b, HR C BI, HR, triglycerides 0.69 (0.62; 0.76) (0.74; 0.84) (0.73; 0.88) (0.81; 0.97) omen tpa activity (IU/ml) BI, HR BI, HR, triglycerides 0.86 (; 0.94) (0.85; 0.93) (0.86; 1.09) (0.86; 1.09) "Test for trends. b Body mass index. c aist:hip ratio. TABLE 3 Tissue plasminogen activator (tpa) mass concentrations according to leisure time physical activity. Geometric means (95% confidence intervals). The number of subjects in each group is similar to Table I Light activity oderate activity Strenuous activity />-value a en tpa mass concentrations (Ug/1) BI b, HR C BI, HR, triglycerides 7.3 (6.7; 8.0) (; 7.2) (5.2; 6.3) 6.4 (4.9; 5.8) omen tpa mass concentrations < ig/l) BI, HR BI, HR, triglycerides 5.7 (5.2; 6.3) 5.9 (5.2; 5.7) (4.1; 5.0) (3.5; 4.8) "Test for trends. b Body mass index. c aist:hip ratio. with increasing physical activity, although not significantly so in women. The tpa mass concentration ('tpa antigen') decreased with higher activity levels. This immunological method mainly measures the inactive tpa-pai-1 complex and is only weakly and inversely related to tpa activity. 19 ' 21 The main regulator of tpa activity, PAI-1 activity was strongly and inversely related to physical activity levels in both genders. A drawback of our study is the crude classification of activity. This possible imprecision and the pitfalls of self-reported activity level make random misclassification possible. But as this would only dilute the strength of the relationship it does not invalidate our results. Non-random misclassification is possible with moderately active subjects estimating their activity as strenuous, which would bias the data towards the

5 1186 INTERNATIONAL JOURNAL OF EPIDEIOLOGY TABLE 4 Plasminogen acitvalor inhibitor type I (PAI-1) activity according to leisure time physical activity. Geometric means (95% confidence intervals). The number of subjects in each group is similar to Table I Light activity oderate activity Strenuous activity P-value a en PAI-1 activity(u/ml) BI, b HR C BI, HR, triglycerides 7.8 (6.7; 9.2) (; ) (4.4; 5.8) (4.2; ) omen PAI-1 activity (U/ml) BI, HR BI, HR, triglycerides 6.2 (; 7.0) (5.1; 5.9) (4.2; 5.4) (3.6; 5.2) a Test for trends. b Body mass index. c aist:hip ratio. null. and less active participants claiming moderate activity could enhance the difference between the most active subjects and the moderately active ones. Recent validation studies of (more complex) questionnaires show them to be highly reliable and accurate, especially when assessing heavy intensity activities. 22 Individuals who exercise less tend to drink less alcohol which could influence the interpretation of our findings. e could not adjust for alcohol consumption, but as fibrinolytic activity was increased in drinkers compared to non-drinkers in the Northwick Park Study 23 this would lead to an underestimate regarding tpa activity. On the other hand, tpa concentrations increase with moderate alcohol intake. 24 Thus, we may overestimate the effect on physical activity on tpa mass concentrations. Clues to causal mechanisms can be found by stepwise adjustment for factors related to both fibrinolysis and physical activity. A minor part of the relationship was associated with age and obesity but the major links seemed to be the lower triglyceride and serum insulin levels seen in subjects with higher activity levels. The results in the group of men practising strenuous activity differ somewhat in the greater magnitude of change in tpa activity and PAI-1 activity with adjustment for triglycerides. The impact of exercise and/or triglycerides on fibrinolysis in this group may differ from that found in women and less active men. Causal interpretations could not be made from our data. Some support can be drawn from previous findings that the triad of obesity, high insulin and triglyceride levels is the major predictor of fibrinolytic variables in epidemiological studies. 13 ' 14 In cell cultures, triglyceride-rich very low density lipoprotein particles, 25 insulin 26 and proinsulin and its split products 27 stimulate the synthesis and release of PAI-1. e have been unable to find any population studies on physical exercise and fibrinolysis using assays for specific components, such as tpa activity or PAI-1 activity. Recently, a cross-sectional study in 111 middleaged men, showed lower tpa mass concentrations with increasing cardiorespiratory fitness, but the association vanished after adjustment for insulin and triglyceride levels. 17 In male offspring of patients with coronary artery disease, physical activity was a predictor of tpa and PAI-1 activity, independent of triglycerides and insulin levels. 28 Exhaustive physical exercise has been used as a test of tpa releasing capacity. 15 Increased adrenalin concentrations in blood may be responsible for greater tpa activity during exercise by stimulating endothelial release of tpa, but also by reducing hepatic blood flow which leads to decreased clearance of tpa. 16 Values return to baseline within an hour after exercise stops. 29 Interventional studies have shown that 2 weeks of heavy endurance activity led to a decrease in tpa mass concentrations and PAI-1 activity, partly attributable to lower body fat and insulin levels, but tpa activity did

6 PHYSICAL ACTIVITY, FIBRINOGEN AND FIBRINOLYSIS 1187 not change. 30 Six months of intensive training did not influence fibrinolysis in young men, but in older men a large increase in tpa activity and decrease in PAI-1 activity was found. 21 In sedentary young men who ran 5 km every other day over 5 months, PAI-1 activity decreased in parallel with weight loss. 31 Three months after they had ceased exercising the values returned to baseline. Twenty sedentary males and 15 females participated in a strenuous endurance training programme during 9 months after which PAI-1 activity decreased by approximately 75%, while tpa mass concentrations did not differ. 32 None of these studies took changes in triglyceride or insulin levels into consideration. In a clinical trial, middle-aged men were randomized to different levels of unsupervized aerobic exercise over 6 months. 33 No improvement was seen in the levels of tpa or PAI-1, in spite of a slight weight loss but this may be explained by the lack of improvement in maximal oxygen uptake. e conclude that regular leisure time physical activity is associated with increased fibrinolysis in a dose-dependent manner. Fibrinolytic activity may be an important component of the linkage between physical activity on one hand and cardiovascular disease on the other. ACKNOLEDGEENTS This study was supported by grants from the Swedish edical Research Council (27X-07192), Norrbotten and Vasterbotten County Councils, the Swedish Public Health Institute, the Heart and Chest Fund, the 1987 Stroke Fund and the Joint Committee of the Northern Sweden Health Care Region. REFERENCES 1 Berlin J A, Colditz G A. A meta-analysis of physical activity in the prevention of coronary heart disease. Am J Epidemiol 1990; 132: Lakka T A, Venb'lbinen J H, Rauramaa R, Salonen R, Tuomilehto J, Salonen J T. Relation of leisure-time physical activity and cardiorespiratory fitness to the risk of myocardial infarction in men. N Engl J ed 1994; 330: Rodriguez B L, Curb J D, Burchfiel C et al. Physical activity and 23-year incidence of coronary heart disease morbidity and mortality among middle-aged men. The Honolulu Heart Program. Circulation 1994; 889: annamethee G, Shaper A G. Physical activity and stroke in British middle aged men. Br ed J 1992; 304: Abbot R D, Rodriguez B L, Burchfiel C, Curb J D. Physical activity in older middle-aged men and reduced risk of stroke: the Honolulu Heart Program. Am J Epidemiol 1994; 139: Salonen J T, Puska P, Tuomilehto J. Physical activity and the risk of myocardial infarction, cerebral stroke and death: a longitudinal study in eastern Finland. Am J Epidemiol 1988; 127: Lapidus L, Bengtsson C. Socioeconomic factors and physical activity in relation to cardiovascular disease and death. A 12-year follow-up of participants in a population study of women in Gothenburg, Sweden. Br Heart J 1986; 55: Kiely D K, olf P A, Cupples L A, Beiser A S, Kannel B. Physical activity and stroke risk: the Framingham study. Am J Epidemiol 1994; 140: eade T, Ruddock V, Stirling Y, Chakrabarti R, iller G J. Fibrinolytic activity, clotting factors, and long-term incidence of isehaemic heart disease in the Northwick Park Heart Study. Lancet 1993; 342: Ridker P, Vaughan D E, Stampfer J, anson J E, Hennekens C H. Endogenous tissue-type plasminogen activator and risk of myocardial infarction. Lancet 1993; 341: Thompson G S, Kienast A J, Pyke S D, Haverkate F, van der Loo J C. for The European Concerted Action on Thrombosis and Disabilities Angina Pectoris Study Group. Hemostatic factors and the risk of myocardial infarction or sudden death in patients with angina pectoris. N EnglJ ed 1995; 332: Ridker P, Hennekens C H, Stampfer J, anson J, Vaughan D E. Prospective study of endogenous tissue plasminogen activator and risk of stroke. Lancet 1994; 343: l3 Eliasson, Evrin P-E, Lundblad D. Fibrinogen and fibrinolytic variables in relation to anthropometry, lipids and blood pressure. The Northern Sweden ONICA study. J Clin Epidemiol 1994; 47: l4 Eliasson, Asplund K, Evrin P-E, Lindahl B, Lundblad D. Hyperinsulinemia predicts low tissue plasminogen activator activity in a healthy population. The Northern Sweden ONICA Study. etabolism 1994; 43: Jem C, Eriksson E, Tengborn L, Risberg B, adenvik H, Jem S. Changes of plasma coagulation and fibrinolysis in response to mental stress. Thromb Haemost 1989; 62: Chandler L, Levy C, Veith R C, Stratton J R. A kinetic model of the circulatory regulation of tissue plasminogen activator during exercise, epinephrine infusion and endurance training. Blood 1993; 81: Rankinen T, Rauramaa R, Vaisenen S, Penttila I, Uusitupa. Relation of habitual diet and cardiorespiratory fitness to blood coagulation and fibrinolytic factors. Thromb Haemost 1994; 71: Huhtasaari F, Asplund K, Lundberg V, Stegmayr B, ester P O. Trends in cardiovascular risk factors in the Northern Sweden ONICA Study: ho are the winners? Cardiovasc Risk Fact 1993; 3: "Eliasson, Evrin P-E, Lundblad D, Asplund K, Ranby. Influence of gender, age and sampling time on plasma fibrinolytic variables and fibrinogen. A population study. Fibrinolysis 1993; 7: Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, Rajakangas A-, Pajak A. yocardial infarction and coronary deaths in the orld Health Organization ONICA project. Registration procedures, event rates, and casefatality rates in 38 populations from 21 countries in four continents. Circulation 1994; 90: Stratton J R, Chandler L, Schwartz RSci al. Effects of physical conditioning on fibrinolytic variables and fibrinogen in young and old healthy adults. Circulation 1991; 83:

7 1188 INTERNATIONAL JOURNAL OF EPIDEIOLOGY 22 Richardson T, Ainsworth B E, u H-C, Jacobs Jr D R, Leon A S. Ability of the Atherosclerosis Risk in Communities (ARIC)/Baecke questionnaire to assess leisure-time physical activity. Int J Epidemiol 1995; 24: eade T, Chakrabarti R, Haines A P, North R S, Stirling Y. Characteristics affecting fibrinolytic activity and plasma fibrinogen concentrations. Br ed J 1979; i: Ridker P, Vaughan D E, Stampfer J, Glynn R J, Hennekens C H. Association of moderate alcohol consumption and plasma concentration of endogenous tissue-type plasminogen activator. JAA 1994; 272: Stiko-Rahm A, iman B, Hamsten A, Nilsson J. Secretion of plasminogen activator inhibitor-1 from cultured human umbilical vein endothelial cells is induced by very low density lipoprotein. Arteriosclerosis 1990; 10: Alessi C, Juhan-Vague I, Kooistra T, Declerck P J, Collen D. Insulin stimulates the synthesis of plasminogen activator inhibitor 1 by human hepatocellular cell line Hep G2. Thromb Haemost 1988; 60: Nordt T K, Schneider D J, Sobel B E. Augmentation of the synthesis of plasminogen activator inhibitor type-1 by precursors of insulin. Circulation 1994; 89: Shaukat N, Douglas J T, Bennet J L, de Bono D P. Can physical activity explain the differences in insulin levels and fibrinolytic activity between young Indo-origin and European relatives of patients with coronary artery disease? Fibrinolysis 1995; 9: Speiser, Langer, Pschaick A et al. Increased blood fibrinolytic activity after physical exercise: comparative study in individuals with different sporting activities and in patients after myocardial infarction taking part in a rehabilitation sport program. Thromb Res 1988; 51: Boman K, Hellsten G, Bruce A, Hallmans G, Nilsson T K. Endurance physical activity, diet and fibrinolysis. Atherosclerosis 1994; 106: Gris J C, Schved J F, Aguilar-artinez P, Arnaud A, Sanchez N. Impact of physical training on plasminogen activator inhibitor activity in sedentary men. Fibrinolysis 1990; 4 (Suppl. 2): Ponjee G A E, Janssen G E, van ersch J J. Effect of longterm endurance exercise on fibrinolytic markers. Fibrinolysis 1993; 7: Rankinen T, Rauramaa R, Vaisanen S, Halonen P, Penttila I. Blood coagulation and fibrinolytic factors are unchanged by aerobic exercise or fat modified diet. Randomized clinical trial in middle-aged men. Fibrinolysis 1994; 8: (Revised version received February 1996)