The Fibrinolytic System Is Not Impaired in Older Men With Hypertension
Abstract The fibrinolytic system is thought to be impaired in older hypertensive adults, thus contributing to the elevated risk of atherothrombosis, stroke, and acute myocardial infarction in this population. However, studies that have examined the fibrinolytic system in hypertensive individuals have failed to control for the confounding effects of other metabolic risk factors, making it difficult for one to determine the independent effect of hypertension on the fibrinolytic system. The purpose of the present study was to test the hypothesis that the fibrinolytic system is not impaired in older sedentary hypertensive men when the confounding effects of cardiovascular disease, diabetes, and dyslipidemia are controlled. Plasma concentrations of tissue-type plasminogen activator antigen and activity as well as plasminogen activator inhibitor–1 antigen and activity were measured under resting conditions in 12 hypertensive (69.4±1.4 years) and 11 normotensive (65.2±1.3 years) older men. The hypertensive and normotensive subjects had similar anthropometric and metabolic characteristics. There were no significant differences between the hypertensive and normotensive men in tissue-type plasminogen antigen (7.3±0.5 versus 6.1±0.6 ng/mL) and activity (1.8±0.3 versus 1.7±0.2 IU/mL) or plasminogen activator inhibitor–1 antigen (14.1±2.3 versus 10.8±2.2 ng/mL) and activity (17.4±1.2 versus 17.5±1.8 arbitrary units [AU]/mL) levels. In addition, the molar concentration ratio of active tissue-type plasminogen activator to active plasminogen activator inhibitor–1 did not differ between the hypertensive (1:9.7±2.3 mmol/L) and normotensive (1:10.5±2.2 mmol/L) subjects, indicative of no impairment in fibrinolytic potential in either group. These results support the hypothesis that hypertension does not directly result in impaired fibrinolytic function in older adults. Furthermore, our findings suggest that abnormalities in fibrinolytic function in older hypertensive men are likely due to the primary effects of other metabolic disorders that usually accompany hypertension, such as hyperinsulinemia and dyslipidemia.
The fibrinolytic system is an important protective mechanism against potentially fatal thrombus formation. This enzymatic pathway results in the proteolytic degradation of fibrin within the developing thrombus, thus preventing arterial occlusion and interruption of blood flow.1 Reduced fibrinolytic activity primarily due to elevated PAI-1 levels, the fast-acting primary inhibitor of TPA, is now recognized to play an important role in the pathogenesis of atherothrombotic events.2 In addition, endogenous fibrinolysis is thought to be impaired in older hypertensive adults,3 4 5 which may contribute to the elevated risk of atherothrombosis, stroke, and acute myocardial infarction in this population.6 However, the link between hypertension and impaired fibrinolytic function is not well defined. Previous investigations that have examined the fibrinolytic system in older adults with hypertension failed to control for the confounding effects of other metabolic risk factors, such as hyperinsulinemia and dyslipidemia, that have also been linked to hypofibrinolysis. For example, higher resting PAI-1 activity and lower TPA activity levels were recently observed in a group of older hypertensive adults; however, a strong positive correlation was also observed between PAI-1 activity and plasma lipid and lipoprotein levels, suggesting that the elevation in PAI-1 activity may have been attributable more to cholesterol and triglyceride concentrations than to hypertension.3 Thus, the independent effect of hypertension per se cannot be discerned from available data.
The purpose of the present study was to test the hypothesis that the fibrinolytic system is not impaired in older sedentary hypertensive men after carefully controlling for the confounding effects of cardiovascular disease, diabetes, and dyslipidemia. To accomplish this, we used a cross-sectional model in which resting plasma levels of TPA antigen, TPA activity, PAI-1 antigen, and PAI-1 activity were measured in older, sedentary hypertensive and normotensive men of similar age, body composition, and metabolic profile.
Twelve hypertensive and 11 healthy normotensive sedentary men ranging in age from 56 to 76 years volunteered to participate in this study. All subjects had the research study and potential risks and benefits explained fully before providing written informed consent according to the guidelines of both the University of Maryland at Baltimore and University of Maryland at College Park Institutional Review Boards.
Subjects were screened before participation with medical history questionnaires, a physical examination, and a fasting blood chemistry profile. Subjects were excluded from the study if they presented a history or evidence of hepatic, renal, or hematologic disease; peripheral vascular disease; stroke; diabetes (fasting plasma glucose >7.8 mmol/L)7 ; dyslipoproteinemia (cholesterol ≥6.2 mmol/L, triglycerides ≥4.5 mmol/L); severe hypertension (BP >180/110 mm Hg); body mass index less than 25 or more than 40 kg/m2; or use of prescription or over-the-counter medication that may affect blood coagulation. Two hypertensive subjects who were being treated with antihypertensive medications were gradually tapered off their medication and studied after a minimum of 3 weeks of no drug therapy. Subjects had resting supine and upright 12-lead electrocardiogram and BP measurements before undergoing a graded exercise treadmill test8 to a minimum of 85% of age-predicted maximal heart rate (220−age). Subjects with ischemic electrocardiographic responses (>1-mm ST segment depression) or limited by signs or symptoms of cardiovascular decompensation during the exercise test were excluded from the study. All subjects were nonsmokers and had not participated in a regular aerobic exercise program for at least 6 months before the start of the study.
Before having their BP measured, the subjects rested quietly in a seated position for 15 minutes. Systolic and diastolic BPs were measured with an automated Dinamap (486SX, Critikon) BP monitor with the appropriately sized cuff. Triplicate measurements were made 2 minutes apart and averaged on 3 separate days over 2 weeks. Hypertension was defined as mean systolic BP of 140 mm Hg or greater and/or mean diastolic BP of 90 mm Hg or greater.9 Mean arterial pressure was calculated as two thirds diastolic BP plus one third systolic BP.
Body Composition Measurements
Body weight was measured before all testing to the nearest 0.1 kg with a medical beam balance (Detecto), and height was measured to the nearest 0.5 cm. The waist-to-hip ratio was calculated as the ratio of the minimal waist circumference to the circumference of the maximal gluteal protuberance. Body mass index was calculated as weight (kilograms) divided by height (meters) squared. Percent body fat was determined by dual-energy x-ray absorptiometry (DXA, model DPX-L, Lunar Radiation Corp), and fat-free mass was calculated as body weight minus fat mass.
Lipid and Lipoprotein Measurements
Blood samples for the determination of fasting plasma lipid and lipoprotein levels were collected into tubes containing EDTA (1 mg/mL blood) on 2 separate days after a 12-hour overnight fast. Plasma triglyceride and total cholesterol levels were measured enzymatically with a Hitachi 717 analyzer (Boehringer Mannheim).10 11 Because no subject had a plasma triglyceride level greater than 4.5 mmol/L, high-density lipoprotein cholesterol (HDL-C) was measured in the supernatant after precipitation of apoprotein B–containing lipoproteins with dextran sulfate.12 Low-density lipoprotein cholesterol (LDL-C) was calculated as LDL-C=Total Cholesterol−(TG/5+HDL-C), where TG is triglycerides.13 A second precipitation with high-molecular-weight dextran sulfate was performed on the supernatant HDL-C to separate HDL2-C and HDL3-C subclasses.14
Oral Glucose Tolerance Test
OGTTs were performed to screen for clinical diabetes. Subjects were instructed by a registered dietitian to consume at least 200 g carbohydrate per day for 3 days before the OGTT and to keep a detailed record of their food intake. The food records were analyzed with a computer-based system (Nutritionist III, N-Squared Computing). All OGTTs were performed in the morning after a 12-hour overnight fast. Thirty minutes after the insertion of a polyethylene catheter into an antecubital vein, two baseline blood samples (5 mL each) were drawn 15 minutes apart for measurement of plasma glucose and insulin concentrations. After ingestion of a solution containing 75 g glucose, blood samples (5 mL) were drawn every 30 minutes for 3 hours. Plasma glucose was measured in duplicate by a glucose oxidase method (Beckman Instruments). Aliquots of plasma for insulin determinations were separated and frozen at −80°C for analysis at the end of the study in the same assay to control for interassay variability. Plasma insulin was measured by radioimmunoassay.15 Total areas under the glucose and insulin concentration curves above baseline were calculated by computer with a trapezoidal model. The intra-assay coefficient of variation for the insulin assay was 7.4%.
To avoid diurnal variation in fibrinolytic variables, we drew blood samples for TPA and PAI-1 determinations between 8 am and noon after subjects had fasted overnight. A standardized questionnaire designed to detect and document recent infection and/or inflammation (<2 weeks) was administered before the phlebotomies. Subjects with a history of recent infection and/or inflammation did not receive phlebotomy to avoid confounding effects from potential infection- and/or inflammation-associated hemostatic changes.16
Blood Sampling and Preparation
All phlebotomies were performed with minimal venostasis. The first 2 to 3 mL of blood was discarded, and samples were used only if venous return was prompt throughout. Blood (9 mL) for determination of TPA antigen and TPA activity was collected in a 10-mL syringe containing 1.0 mL of 130 mmol/L sodium citrate (final dilution volume, 1:10). For prevention of in vitro inactivation of TPA by ongoing complex formation with PAI-1, 0.75 mL of citrate anticoagulated whole blood was acidified within 1 minute of phlebotomy by addition of 0.37 mL of 0.5 mmol/L sodium acetate, pH 4.2.17 Blood samples (4 mL) for measurement of PAI-1 antigen and PAI-1 activity were collected in a 5-mL syringe containing modified Files solution (1 mL acid citrate dextrose solution, 80 μL acetylsalicylic acid solution, 10 μL prostaglandin E1 solution)18 to minimize in vitro platelet activation (final dilution volume, 1:5). Within 30 minutes of phlebotomy, all samples were centrifuged for 20 minutes at 6000g at 4°C. Platelet-poor plasma was aliquoted and stored at −80°C until assayed at the end of the study. All TPA and PAI-1 assays were performed in duplicate with a maximum of one freeze-thaw cycle. Intra-assay variability was calculated from duplicate samples, and internal controls were used for determination of interassay variability for all fibrinolytic assays.
Assay for TPA and PAI-1 Antigens
Total TPA antigen and total PAI-1 antigen were determined by an enzyme-linked immunosorbent assay.17 TPA antigen and PAI-1 antigen levels (in nanograms per milliliter) were determined against standard curves constructed from respective standard solutions (American Bioproducts). The intra-assay and interassay coefficients of variation for TPA antigen were 6.5% and 6.1% and for PAI-1 antigen were 9.1% and 9.7%, respectively.
Assay for TPA and PAI-1 Activities
TPA activity and PAI-1 activity were measured by an amidolytic method.19 Levels of TPA activity and PAI-1 activity were determined against standard curves constructed from respective standard solutions (Chromogenix). TPA activity is expressed in international units assessed against the Second International Standard for TPA from the National Institute for Biological Standards and Control.20 The intra-assay and interassay coefficients of variation for the TPA activity assay were 6.8% and 3.1%, respectively. The molar concentration of active TPA was determined by dividing the TPA activity (in international units per liter) by the specific molar activity of TPA (4.48×1013 IU/mol).21
PAI-1 activity is expressed in arbitrary units (AU). One AU of inhibitor is defined as the amount that inhibits 1 IU of TPA per milliliter of plasma.22 The intra-assay and interassay coefficients of variation for the PAI-1 activity assay were 4.5% and 3.0%, respectively. The molar concentration of active PAI-1 was determined by dividing the PAI-1 activity (in AU per liter) by the specific molar activity of PAI-1 (4.48×1013 AU/mol).21
All data are presented as mean±SE. Differences between the normotensive and hypertensive groups for all anthropometric, hemodynamic, and metabolic data were tested with Student’s unpaired t test. Simple and forward stepwise multiple regression analyses and partial correlation coefficients were calculated to determine relationships between specific fibrinolytic variables and BP, body composition, and measures of glucose and lipid metabolism. Data were analyzed with SYSTAT 5 (Systat Inc) and StatView (Abacus Concepts) software packages. The level of significance was set a priori at a value of P<.05.
Height, body mass, body mass index, percent body fat, fat mass, fat-free mass, and waist-to-hip ratio did not differ significantly between the two groups (Table 1⇓). However, the mean age of the hypertensive group was slightly but significantly greater than that of the normotensive group (69.4±1.4 versus 65.2±1.3 years, P=.03). By design, the hypertensive subjects had significantly higher systolic (153±2 versus 120±3 mm Hg, P=.0001), diastolic (81±1 versus 71±2 mm Hg, P=.0005), and mean arterial (105±1 versus 88±2 mm Hg, P=.0001) BPs than their normotensive peers.
Fasting total cholesterol, HDL-C, HDL2-C, HDL3-C, LDL-C, and triglyceride levels did not differ significantly between the two groups (Table 2⇓). In addition, the hypertensive and normotensive subjects had similar fasting plasma glucose (5.6±0.2 versus 5.7±0.2 mmol/L) and insulin (66.0±4.5 versus 56.7±6.1 pmol/L) levels.
Glucose (7.9±0.5 versus 7.4±0.6 mmol/L) and insulin (562.5±76.8 versus 415.3±86.4 pmol/L) levels at 120 minutes of the OGTT did not differ significantly between the hypertensive and normotensive groups. Moreover, glucose (388±43 versus 313±61 mmol/L · 180 min) and insulin (64 498±7059 versus 46 429±6720 pmol/L · 180 min) areas did not differ significantly between the hypertensive and normotensive groups. Although there were no statistically significant differences between the two groups in response to an OGTT, the hypertensive group tended to have higher plasma insulin levels at 120 minutes as well as higher glucose and insulin areas.
TPA antigen (7.3±0.5 versus 6.1±0.6 ng/mL), TPA activity (1.8±0.3 versus 1.7±0.2 IU/mL), PAI-1 antigen (14.1±2.3 versus 10.8±2.2 ng/mL), and PAI-1 activity (17.4±1.2 versus 17.5±1.8 AU/mL) did not differ significantly between the hypertensive and normotensive groups (Figs 1⇓ and 2⇓). Also, the molar concentration ratio of active TPA to active PAI-1 did not differ significantly between the hypertensive (1:9.7±2.3 mmol/L) and normotensive (1:10.5±2.2 mmol/L) subjects.
Univariate analysis on the pooled data revealed a significant positive correlation between TPA antigen and diastolic BP (r=.48, P=.02), mean arterial BP (r=.44, P=.04), waist-to-hip ratio (r=.56, P=.005), plasma glucose at 120 minutes (r=.43, P=.04), and glucose area during an OGTT (r=.54, P=.007). In addition, PAI-1 antigen was positively correlated with waist-to-hip ratio (r=.492, P=.01), fasting plasma insulin (r=.51, P=.01), and insulin area during an OGTT (r=.51, P=.01). However, neither TPA activity nor PAI-1 activity correlated with any anthropometric, hemodynamic, or metabolic data. When multiple regression was applied for assessment of the independence of the observed relationships, only the glucose area under the OGTT curve was significantly associated with TPA antigen, whereas only the insulin area was significantly associated with PAI-1 antigen (Fig 3⇓).
The primary finding of this study is that older, sedentary men with hypertension who are free of other metabolic abnormalities commonly associated with high BP, such as hyperinsulinemia and hyperlipidemia,23 do not have an impaired fibrinolytic system compared with their normotensive counterparts. This finding is not consistent with previous studies that have reported impaired fibrinolytic function in hypertensive subjects due to elevated PAI-1 activity.3 4 5 However, these prior studies reported the coexistence of higher plasma insulin, cholesterol, or triglyceride levels in the study population. In contrast, this is the first investigation to characterize the fibrinolytic system in older men with hypertension who were free of confounding metabolic abnormalities such as diabetes, hyperinsulinemia, and hyperlipidemia.
The first study to demonstrate an association between hypertension and elevated PAI-1 activity was conducted by Landin et al,3 who observed higher resting PAI-1 activity in 11 nonobese hypertensive men compared with a normotensive control group. Both groups were well matched for age and body composition; however, the hypertensive subjects had significantly higher fasting plasma insulin, triglyceride, and cholesterol levels. Although PAI-1 activity was significantly correlated with diastolic BP (r=.46, P<.001) and mean arterial BP (r=.64, P<.001), the authors reported a stronger correlation between PAI-1 activity and plasma insulin (r=.80, P<.001). Furthermore, when the data were not pooled and the groups were analyzed individually, a significant correlation was observed between PAI-1 activity and fasting plasma insulin levels (r=.76, P<.001) in the hypertensive men only, suggesting that the elevations in PAI-1 activity were attributable to increased plasma insulin levels rather than hypertension. In a similar study, Jansson and associates4 reported that subjects (45 men, 39 women) who suffered from both hypertension and hypercholesterolemia also had impaired fibrinolytic function as evidenced by significantly lower TPA activity and significantly higher PAI-1 activity. Although multivariate analyses supported the authors’ hypothesis that both hypertension and hyperlipidemia contributed to the observed disturbances in the fibrinolytic system, the analyses also suggested that higher cholesterol and triglyceride levels, but not BP, play an independent role in the impairment of the fibrinolytic system.4
In the present study, the hypertensive and normotensive men did not differ significantly in terms of body composition, lipid and lipoprotein levels, fasting insulin, and glucose metabolism, which may explain in part why the measured fibrinolytic variables did not differ between the two groups. It is plausible that metabolic abnormalities such as hyperinsulinemia and/or hypercholesterolemia must accompany hypertension to impair fibrinolysis.3 4 In addition, the degree of hypertension in the present study (isolated systolic stage 1 hypertension9 ) combined with the absence of these metabolic disorders, which are known to contribute to both atherosclerosis24 25 and endothelial injury,26 may have resulted in only minor endothelial damage not severe enough to impair endogenous fibrinolysis. In vitro, endothelial cell injury or damage can disrupt the regulation and control of fibrinolysis by downregulating TPA release and enhancing PAI-1 release, thus promoting thrombotic manifestations such as acute myocardial infarction. Furthermore, increased PAI-1 gene expression, localization, and production have been reported to occur in the intima of injured atherosclerotic human arteries.26 27 However, the mechanism or mechanisms by which endothelial cell injury or damage triggers the release of PAI-1 and limits the release of TPA and whether the degree of hypertension may potentiate this phenomenon are not clear.
The finding of no difference between the hypertensive and normotensive groups in the molar concentration ratio of active TPA to active PAI-1 further suggests that the subjects with hypertension did not have an impaired fibrinolytic system. Chandler and associates21 have suggested that the ratio of active TPA to active PAI-1 may be a better index of overall fibrinolytic potential, the ability to respond to a stimulus and lyse a thrombus, than the plasma levels of either TPA or PAI-1 alone. The ratio of active TPA to active PAI-1 has been reported to be approximately 1:8 in healthy male subjects (mean age, 51±17 years) compared with 1:50 in men with thrombotic disease (mean age, 56±11 years).21 The higher the molar ratio the more TPA needed to neutralize the large excess of PAI-1, which in turn reduces TPA activity and the subsequent activation of plasmin required for fibrin degradation. In the present study, the molar concentration ratio of active TPA to active PAI-1 was similar in both the hypertensive (1:9.7±2.3) and normotensive (1:10.5±2.2) groups and comparable to the ratio previously observed in healthy men, suggesting no impairment in fibrinolytic potential in either the hypertensive or normotensive subjects.21
In contrast to previous reports,3 4 28 29 no significant relationships were detected between either TPA activity or PAI-1 activity and measures of body composition; BP; or plasma lipid, glucose, and insulin levels in our subjects. This may be related to the overall state of “good health” of our subjects who, aside from being mildly obese and hypertensive, were free of cardiovascular disease, diabetes, hyperinsulinemia, and dyslipidemia. Elevated plasma PAI-1 activity has been shown to be associated with these metabolic disorders, and although the etiology of the association is not clear, multiple regression analysis has revealed that in most cases the relationship between PAI-1 activity and obesity, dyslipidemia, and hypertension is secondary to the relationship between PAI-1 activity and insulin. Moreover, a significant inverse correlation has been reported between insulin sensitivity as measured by the hyperinsulinemic euglycemic clamp technique3 30 and plasma PAI-1 antigen and activity levels, thus suggesting that insulin resistance per se may be the common etiologic feature.31 Although none of our subjects were hyperinsulinemic (fasting insulin ≥90 pmol/L),32 the observed positive relationship between TPA antigen and glucose area and PAI-1 antigen and insulin area supports the hypothesis that impaired glucose tolerance and elevated plasma insulin concentrations, possibly due to increased insulin resistance, could have a negative effect on the fibrinolytic system by increasing TPA and PAI-1 antigen levels, which could ultimately result in elevated PAI-1 activity.
During the final preparation of this article, a study by Wall et al33 appeared in this journal suggesting that impaired fibrinolysis is a primary effect of early BP elevation and not a secondary result of atherosclerosis or metabolic abnormalities. This suggestion was based on the finding of higher resting levels of TPA antigen in young adults with high normal systolic BP compared with age-matched normotensive control subjects. However, close inspection of the data does not reveal an impairment in the fibrinolytic system in these subjects. The TPA antigen levels reported in the high normal BP group are similar to levels previously reported in young, healthy normotensive subjects.34 In addition, TPA activity and PAI-1 antigen levels did not differ between the two groups. Therefore, these results do not contradict the findings of the present study.
In conclusion, our results indicate that the fibrinolytic system is not impaired in older, sedentary hypertensive men who are free of other metabolic disorders such as diabetes, hyperinsulinemia, dyslipidemia, and glucose intolerance. Thus, it appears that abnormalities in fibrinolytic function and their cardiovascular consequences in older hypertensive adults are likely due to the primary effects of other metabolic risk factors commonly associated with hypertension, such as hyperinsulinemia and dyslipidemia.23 30 Hypertension independent of these metabolic disorders may not have a prothrombotic effect on the fibrinolytic system. However, future studies are needed to determine whether the relative rate of development and progression of vascular disease associated with hypertension differs under conditions of normal versus impaired fibrinolysis.
Selected Abbreviations and Acronyms
|OGTT||=||oral glucose tolerance test|
|PAI-1||=||plasminogen activator inhibitor–1|
|TPA||=||tissue-type plasminogen activator|
This research was supported by an American College of Sports Medicine (ACSM) Foundation Research Grant for Doctoral Students (C.A.D.); a College of Health and Human Performance Research Committee grant (M.A.R. and C.A.D.); a Department of Veterans Affairs Research Advisor Group Study grant (R.F.M.); and a National Institutes of Health Training Grant in Exercise Physiology (T-32 AG00291 to Andrew P. Golberg, MD). Our sincere appreciation to all the subjects who volunteered; Dr Goldberg and the clinical staff of the Division of Gerontology, Geriatric, Research, Education, and Clinical Center at the Baltimore Veterans Affairs Medical Center for their support and assistance; Jana Dengel, RD, for her dietary instruction and evaluations; and Marilyn Lumpkin for her technical assistance.
- Received November 21, 1995.
- Revision received December 18, 1995.
- Accepted January 22, 1996.
Ouimet H, Loscalzo J. Fibrinolysis. In: Loscalzo J, Schafer AI, eds. Thrombosis and Hemorrhage. Boston, Mass: Blackwell Scientific Publications; 1994:127-143.
National High Blood Pressure Educational Program Working Group. National High Blood Pressure Education Program Working Group report on hypertension in the elderly. Hypertension. 1994;23:275-285.
National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes. 1979;28:1039-1059.
Allian CC, Poon L, Chan SG, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem. 1974;20:470-475.
Simpson EJ, Demers LM, Frieg AF. Faster enzymatic procedure for serum triglyceride. Clin Chem. 1975;21:1983-1985.
Warnick R, Benderson JM, Albers JJ. Quantitation of high-density-lipoprotein subclasses after separation by dextran sulfate and Mg2+ precipitation. Clin Chem. 1982;28:1379-1388.
Lipid Research Clinic Programs. Manual of Laboratory Operations: Lipid and Lipoprotein Analysis. Bethesda, Md: National Institutes of Health; 1974. National Heart and Lung Institute, US Dept of Health, Education, and Welfare Publication NIH 75-628.
Gidez LI, Miller GJ, Burstein M, Slagle S, Eder HA. Separation and quantification of subclasses of human plasma high density lipoproteins by a simple precipitation procedure. J Lipid Res. 1982;23:1206-1223.
Zaharko DS, Beck LV. Studies of a simplified plasma insulin immunoassay using cellulose powder. Diabetes. 1968;17:444-447.
Ameriso SF, Wong VLY, Quislorio FP, Fisher M. Immunohematologic characteristics of infection-associated cerebral infarction. Stroke. 1991;22:1004-1009.
Files JC, Malpass TW, Yee EK, Ritchie JL, Harker LA. Studies of human platelet alpha-granule release in vivo. Blood. 1981;58:607-618.
Chandler WL, Loo S-C, Nguyen SV, Schmier G, Stratton JR. Standardization of methods for measuring plasminogen activator inhibitor activity in human plasma. Clin Chem. 1989;35:787-793.
Chandler WL, Trimble SL, Loo SC, Mornin G. Effect of PAI-1 levels on the molar concentrations of active tissue plasminogen activator (t-PA) and t-PA/PAI-1 complex in plasma. Blood. 1990;76:930-937.
Ip JH, Fuster V, Isreal D, Chesebro JH. Evolution, progression, and clinical manifestation of atherosclerosis: the role of antithrombotic therapy. In: Colman RW, Hirsh J, Marder VJ, Salzman EW, eds. Haemostasis and Thrombosis: Basic Principles and Clinical Practice. 3rd ed. Philadelphia, Pa: JB Lippincott Co; 1994:1379-1395.
Lupu F, Bergonzelli GE, Heim DA, Cousin E, Genton CY, Bachmann F, Kruithof EKO. Localization and production of plasminogen activator inhibitor–1 in human healthy and atherosclerotic arteries. Arterioscler Thromb. 1993;13:1090-1100.
Schneider DJ, Nordt TK, Sobel BE. Stimulation by proinsulin of expression of plasminogen activator inhibitor type 1 in endothelial cells. Diabetes. 1992;41:890-895.
Juhan-Vague I, Vague P. Hyperinsulinemia and its effects on coagulation and fibrinolysis in cardiovascular disease. In: Francis FB Jr, ed. Atherosclerotic Cardiovascular Disease, Hemostasis and Endothelial Function. New York, NY: Marcel Dekker, Inc; 1992:141-182.
Wall U, Jern C, Bergbrant A, Jern S. Enhanced levels of tissue-type plasminogen activator in borderline hypertension. Hypertension. 1995;26:796-800.
Stratton JR, Chandler WL, Schwartz RS, Cerqueira MD, Levy WC, Kahn SE, Larson VG, Cain KC, Beard JC, Abrass IB. Effects of physical conditioning on fibrinolytic variables and fibrinogen in young and old healthy adults. Circulation. 1991;83:1692-1697.