Circadian variation of fibrinolytic activity in blood.

Chronobiology International Vol. 8, No. 5 , pp. 336-351 0 1991 International Society of Chronobiology

Circadian Variation of Fibrinolytic Activity in Blood

Chronobiol Int Downloaded from informahealthcare.com by University of Sussex Library on 01/23/13 For personal use only.

Felicita Andreotti and *Cornelis Kluft Cardiovascular Research Unit, Department of Medicine, Royal Postgraduate Medical School, Hammersmith Hospital, London, U.K.; and *Gaubius Laboralory, IVVO- TNO, Leiden, The Netherlands

Summary: Approximately 35 years ago, it was discovered that spontaneous fibrinolytic activity in blood showed a sinusoidal variation with a period of 24 h; it increased severalfold during the day, reaching a peak at 6:OO p.m. and then dropped to trough levels at 3:OO-4:00 a.m. The range of the fluctuation and the 24-h mean levels were highly reproducible within an individual; moreover, the timing of the oscillation was remarkably consistent among individuals, with a fixed phase relationship to external clock time. The biorhythm could not be accounted for simply by variations in physical activity, body posture, or sleepfwake schedule. Gender, ethnic origin, meals, or resting levels of blood fibrinolytic activity also did not influence the basic features of the rhythm. Older subjects, compared to younger ones, showed a blunted diurnal increase in fibrinolytic activity in blood. Recent studies have established that, of the known components of the fibrinolytic system, only tissue-type plasminogen activator (tPA) and its fast-acting inhibitor, plasminogen activator inhibitor- 1 (PAL l), show a marked circadian variation in plasma. In contrast, levels of plasminogen, a,-antiplasmin, urinarytype plasminogen activator, and a reversible tPA inhibitor vary little or none during the 24 h. Quenching antibodies to tPA have shown that the circadian rhythm of fibrinolytic activity in blood is due exclusively to changes in tPA activity. However, the 24-h fluctuation of plasma tPA activity is phase shifted in relation to the rhythm of immunoreactive tPA, but shows a precise phase inversion with respect to the 24-h variation of PAL 1 activity and antigen. Therefore, plasma tPA activity, as currently measured in vitro, is tightly and inversely related to the levels of PAL 1 throughout the 24-h cycle. The factors controlling the rhythmicity of plasma PAI-1 are not fully elucidated but probably involve a humoral mechanism; changes in endothelial function, circulating platelet release. products, corticosteroids, catecholamines, insulin, activated protein C, or hepatic clearance do not appear to be responsible. Shift workers on weekly shift rotations show a disrupted 24-h rhythm of plasma tPA and PAL 1. In acute and chronic diseases, the circadian rhythmicity of fibrinolytic activity may show a variety of alterations, affecting the 24-h mean, the amplitude, or the timing of the fluctuation. It is advisable, therefore, to define the 24-h pattern of plasma tPA and PAI- 1 in patient groups, before levels based on a single blood sampling time are compared to those ~

~

~

~

Received May 13, 199 I;accepted with revisions May 2 I , 199 1 . Address correspondence and reprint requests to Dr. F. Andreotti, CardiovascularUnit, RPMS, Hammersmith Hospital, Ducane Road, London W 12 ONN, U.K.

336

CIRCADIAN RHYTHM OF FIBRINOLYSIS

33 7


of a control population. In normal conditions, the 24-h variation of plasma tPA and PAI- 1 is not associated with parallel circadian changes in effective fibrinolysis, assessed as plasma D-dimer concentrations,presumably because fibrin generation in the circulation is low. In diseases in which fibrin formation is increased, however, the physiological drop of fibrinolytic activity in the morning hours may favour thrombus development at this time of day, in agreement with the reported higher morning frequency of acute thrombotic events. Key Words: CircadianDiurnal-Fibrinolysis-Tissue-type plasminogen activator-Plasminogen activator inhibitor-1-Urinary-type plasminogen activator.

Astrup first postulated that fibrin formation, through coagulation, is balanced by fibrin removal, through fibrinolysis ( 1). Alterations of the fibrinolytic mechanism, therefore, may promote either a thrombotic or a haemorrhagic tendency. The potential importance of changes in fibrinolytic activity has stimulated a continuous interest in defining the behaviour of fibrinolytic factors, their normal range of variation, and their impairment in relation to disease. Over the last few decades, great advances have been made in identifyingthe components ofthe fibrinolyticsystem. It is now known that plasmin, the proteolytic enzyme that degrades fibrin, is converted from the inactive precursor, plasminogen, via plasminogen activators. Two major activators of plasminogen have been identified: the endothelial-derived tissue-type plasminogen activator (tPA), present in active form, and the urinary-type plasminogen activator, urokinase (uPA), present in proenzyme form (2). The breakdown of fibrin is temporarily prevented by natural inhibitors of fibrinolysis, mainly a,-antiplasmin, which neutralizes freshly formed plasmin, and the irreversible plasminogen activator inhibitor type 1 (PAL 1), which rapidly inactivates both tPA and uPA (3). In normal conditions, free active plasmin and urokinase are not detectable in the circulation, so that spontaneous fibrinolytic activity in blood is mainly determined by tPA, which, in turn, is closely regulated by PAI-1.

METHODS USED TO MEASURE FIBRINOLYSIS The study of fibrinolysis has been considerably hindered by problems of methodology. Only in the 1950s was it found that fibrinolytic activity in fluid blood was labile at room temperature, but could be preserved by cooling the blood immediately after collection. It was then recognised that normal blood had spontaneous fibrinolytic activity; however, special methods had to be used for its demonstration, aimed at reducing or removing inhibitory factors that normally conceal its presence (4). Broadly, these methods can be divided into two classes: overall measurements of fibrinolytic activity and measurements of individual components of the fibrinolytic system. An overall measurement, mostly used in the past, is the dilute blood clot lysis time (DBCLT), which is sufficiently sensitive to measure fibrinolytic activity of a physiological degree. The dilution of blood causes preferential reduction in inhibition compared with fibrinolytic activity (4). The time to lyse thrombin- or calcium-induced clots is normally of several hours. Other overall measurements with a greater specificity for plasminogen activators include the euglobulin-clot lysis time (ELT) and the

Chronobiol Int, Vol. 8, No. 5 , 1991


338

F. ANDREOTTI AND C. KLUFT

fibrin plate method, both still amply used. For the ELT, the euglobulin fraction of plasma is obtained after dilution and precipitation by acidification; most of the antiplasmins and part of the antiactivators remain in the supernatant; the principle is similar to that of the DBCLT, but the lysis time of euglobulin clots is considerably shorter, as the inhibitors are removed rather than just diluted (4). With the fibrin plate method, the substance to be tested (e.g., plasma euglobulin fractions) is placed on a fibrin-coated Petri dish; lysis is manifest as a clear zone whose area gives a measure of the fibrinolytic activity of the sample. In the 1980s, the development of synthetic substrates, purified proteins, and specific antibodies has led to a proliferation of methods for the measurement of virtually all of the known components of the fibrinolytic system. These methods have a high degree of sensitivity and specificity; they include functional and immunological assays; the results are usually quantified by means of chromogenic substrates.

CIRCADIAN VARIATION OF FIBRINOLYTIC ACTIVITY IN BLOOD Early Studies Fearnley et al. in 1957 first recorded in two subjects a marked circadian fluctuation of fibrinolytic activity in peripheral venous blood (5). Using the DBCLT method, they detected a greater than fivefold change over the 24-h period, with the shortest lysis time (i.e., greatest fibrinolysis) in blood taken at 6:OO p.m., followed by a steady prolongation of lysis time during the evening and night, reaching minimal fibrinolysis at 4:OO a.m. The times of peak and trough activity were identical in the two subjects. One subject, reinvestigated while remaining in bed all day, showed similar results to those obtained when up and about, suggestingthat the rhythm overrode the known stimulatory effect of physical activity, and any effect of posture, on fibrinolysis. Moreover, fibrinolytic activity at 4:OO p.m. in I5 active day nurses and in 15 sleeping night nurses was consistently greater than at 4:OO a.m., when the day nurses slept and the night nurses worked. Thus, the existence of a circadian rhythm of natural fibrinolysis in blood, basically unaffected by sleep, posture, or physical activity, was first established. The fluctuating fibrinolytic component was postulated to be a kinase that activated plasminogen to plasmin (5). In many subsequent studies, the period of investigation has been limited to daytime hours; therefore, the term “diurnal increase” of fibrinolytic activity has been introduced, with reference to the rising activity between early morning and afternoon. It should be kept in mind, however, that after 6:OO p.m., blood fibrinolytic activity normally starts to decline rapidly. In 1959, Buckell et al. showed that, during 3 h in the morning, the increase in plasma clot lysis measured after 6 h of incubation was greater in 17 young subjects (270%increase) than in 10 older subjects (140%increase), suggestingthat the physiological diurnal rise in fibrinolytic activity was blunted by age; this difference, however, was not apparent with the ELT method ( 6 ) . That same year, Billimoria et al. reported that the diurnal increase in fibrinolytic activity, measured by a modified dilute blood clot lysis method, was present in ambulant subjects (n = 28) but not in bedridden ones (n = 25), again raising the question that the diurnal changes might at least partly depend on physical activity (7). The period of investigation, however, was only 2 h. Moreover, many of these subjects (36 of 5 3 ) were in some way diseased.

Chronobiol Int. Vol. 8, No. 5, 1991



339

In the 1960s, Cepelak et al. in Czechoslovakia recorded the ELT every 6 h for 24 h in 28 subjects (8). They found that clot lysis was 70% faster at 4:OO p.m., when it was maximal, compared to 4:OO a.m., when it was minimal, thereby reproducing a similar circadian rhythm to the one detected years before by Fearnley. In a separate study, Cepelak et al. reported the circadian variation of ELT in four monozygotic and four dizygotic twins; a greater intrapair concordance was found in the monozygotic pairs than in the dizygotic ones, indicating a substantial participation of genetic factors in the determination of this biological rhythm (9). The effect of age and gender on the diurnal variation of fibrinolytic activity was investigated in 1967 by Mann in 80 subjects divided into four groups of 10 men and 10women; each group belonged to the third, fourth, fifth, and sixth decades. Approximately a 20%decrease in DBCLT between 1O:OO a.m. and 4:OO p.m. was found, with no difference related to age or gender (10). Thus, by the end of the 1960s, it was generally accepted that fibrinolytic activity in blood was lower in the morning and increased during the day, but questions remained concerning the effect of age and physical activity. In the early 1970s, Rosing et al. made a series of clarifying observations (1 1,12). Using the method of the euglobulin fibrin plate lysis area, they confirmed the presence of a diurnal increase in fibrinolytic activity between 8:OO a.m. and 3:OO p.m. in 14 bedridden subjects. Furthermore, they found no relationship between the resting levels of fibrinolytic activity and peak fibrinolysis attained during the day. The diurnal changes of five fasting subjects were similar to those observed in nine subjects eating regular meals. The responsiveness of fibrinolytic activity to stimulation by standardised exercise, studied in two subjects, showed a diurnal variation, the fibrinolytic response evoked by maximal and 40% maximal exercise being greater at 4:OO p.m. than at 8:OO a.m. Measurements of fibrinolytic activity over 12 h repeated in 19 subjects up to 13 months later showed a good intraindividual reproducibility. Moreover, a greater diurnal increase in fibrinolytic activity was found from 8:OO a.m. to 8:OO p.m. in a group of 19 young subjects compared to 24 older subjects (mean age of 20 and 44 years, respectively). In 1973, Korsan-Bengtsen et al. reported, in 76 middle-aged men, a lack of relationship between usual physical activity during work or leisure time and the diurnal changes in fibrinolytic activity, measured by the fibrin plate method (1 3). By the mid- 1970s, therefore, it was established that fibrinolytic activity in blood had a definite circadian variation, with a stable phase relationship to the external clock time. The rhythm was thought to be mainly due to plasminogen activator activity. The timing of peaks and troughs was remarkably consistent among individuals and the range of fluctuation showed good intraindividual reproducibility. If recorded over an interval of several hours, the variation was detectable at rest, although physical activity appeared to enhance the natural daytime increase and to elicit a greater fibrinolytic response in the evening than in the morning. The rhythm was basically unaffected by geographical provenance, gender, resting levels of fibrinolytic activity, meals, posture, or degree of usual physical exercise. A blunted diurnal increase in fibrinolytic activity with age was apparent when methods with greater specificity for plasminogen activators were used. Furthermore, genetic factors contributed to the definition of this biological rhythm.

Chronohiol In!. Vol. 8. No. 5. 1991

340



Recent Studies In 1978, Kluft et al. described a specific method to measure plasminogen activator activity in plasma euglobulin fractions ( 14). Using quenching antibodies, they later demonstrated that this activity originated exclusively from tPA ( 15).When measured in 12 subjects between 9:OO a.m. and 4:OO p.m., a marked increase of approximately 200% was recorded in its plasma activity. In 1983, the fast-acting PAI-1 was discovered (16); at the same time, assays to measure functional and immunoreactive plasma levels of tPA, PAI- 1, and uPA were being developed. Thus, in 1985, Kluft et al. measured the plasma concentration of immunoreactive tPA, or tPA antigen, as well as the activity of tPA and its irreversible inhibitor PAI- 1 in 10 young healthy subjects between 9:OO a.m. and 3:OO p.m. (1 7). Surprisingly, as tPA activity increased by 200% during the day, tPA antigen fell by 20%, while PAI-I activity declined by 50%. It became clear, therefore, that the diurnal rise in fibrinolytic activity was determined not by an increasing concentration of tPA, which instead dropped during the day, but rather by a sharp fall in PAL1 resulting in a net rise in tPA activity (1 7). These findings led to a reinterpretation of the changes affecting blood fibrinolytic activity, since variations in PAI- 1 appeared to have an overriding effect on changes in tPA antigen. In 1987, Neerstrand et al. reproduced the diurnal changes in plasma tPA antigen and PAI- 1 activity in six young subjects between 7:OO a.m. and 3:OO p.m. The plasma levels of plasminogen and a,-antiplasmin were also recorded and found to change little or none ( 18). Recently, we defined the entire 24-h variation of plasma tPA, PAI- 1, and uPA in six healthy active young subjects (19,20). The median values of these factors, measured by functional and immunological assay every 3 h for 24 h, are represented in Fig. 1. As shown in the upper panel of Fig. 1, tPA activity reached the highest values at 6:OO p.m. and then fell dramatically to barely detectable levels at 3:OO a.m. The times of peak and trough tPA activity are in agreement with those of fibrinolytic activity assessed by overall methods (DBCLT or ELT), suggestingthat tPA activity is largely responsible for the circadian changes in blood fibrinolysis. The variation of tPA antigen over the 24 h was less striking, exhibiting a 47% drop from its peak value at 9:OO a.m. to its lowest value at midnight. Moreover, the direction of change of tPA antigen and tPA activity showed only partial concordance during the 24-h cycle. In contrast, the fluctuations of PAI-1 activity and PAL1 antigen were closely linked throughout the 24 h (middle panel, Fig. 1). Between 3:OO a.m. and 6:OO p.m., PAI-1 activity and antigen levels both fell by approximately 70%.The lowest inhibitor levels at 6:OO p.m. coincided precisely with the time of highest tPA activity and, conversely, the highest inhibition at 3:OO a.m. was associated with minimal tPA activity. Thus, the fluctuation of PAL1 and tPA activity showed an inverted phase relationship; the rhythm of tPA antigen, instead, lagged by 6 h compared to that of PAI- 1. Total uPA and prourokinase (pro-uPA) underwent smaller circadian changes (lower panel, Fig. 1): uPA fell by 27% between 3:OO p.m. and 6:OO a.m., whereas pro-uPA dropped by 18% between 9:OO a.m. and midnight. A comparison between values at peak and trough times yielded a highly significant difference for tPA and PAI- 1 (p < 0.0 I), but not for uPA and pro-uPA.

Chronobiol Int, Vol. 8, No. 5, 1991


341

800


a 200 4e. 9 12 3

am

pm

9 0

6

3

6 9

am

- 20

f-

FIG. 1. Median plasma levels of fibrinolytic factors in six healthy, active, young subjects during a 24-h period. tPA = tissue-type plasminogen activator. Ag = antigen. PAL I = plasminogen activator inhibitor type I . uPA = urinary-type plasminogen activator. pro-uPA = proenzyme form of uPA.

2

12-

-15

Y

.-?? .B

H r

a' a

c

8

w

-

4I

1

1

1

1

I

I

I

9 1 2 3 6 9 0 3 6 am

pm

I

9

P %

-5

a n

'0 r 4

3-

w

a

2

am

41

ii

-10

r

0'

A

fm

2:

v pro-u-PA

1-

01

0 6 9

9 1 2 3 am

0

pm

3 6

9

am

Time of day

Concomitant studies in other laboratories have further clarified the mechanisms underlying the circadian variation of fibrinolytic activity. Kluft et al. recorded the plasma activity of a reversible inhibitor of tPA in 10 young subjects over a 6-h interval and found that it did not change (21). Grimaudo et al., during a 10-h diurnal study in eight young subjects, demonstrated that the daytime increase in fibrinolysis assessed by ELT and by the fibrin plate method was exclusively due to changes in tPA activity (22), in agreement with previous findings ( 15,20). By electrophoretic-zymographic analysis of the euglobulin fractions, they revealed the highest concentrations of free tPA and lowest concentrations of tPA/PAI-1 complexes in the afternoon. Moreover, the diurnal variation of PAI- 1 antigen was not associated with changes in

Chronobiol Int. Vol. 8, No. 5, 1991


342


plasma P-thromboglobulin, an index of platelet activation; therefore, a contribution of platelet-contained PAI- 1 to the diurnal changes could be excluded (22). Recently, Angleton et al. measured plasma tPA activity after 10 min of brachial venous occlusion in 48 subjects at 8:OO a.m. and 8:OO p.m. and found no significant difference in postocclusion tPA activity at the two times of day (23). Very recently (this issue), Johansen et al. followed the plasma concentrations of D-dimer, a specific fibrin degradation product, every 4 h for 24 h, in 20 healthy men; no circadian variation in D-dimer was detected, suggesting that in health the circadian changes of fibrinolytic factors do not influence the degree of effective fibrinolysis (24). In summary, these recent studies have led to the identification of tPA and PAI-1 as the individual components responsible for the circadian variation of fibrinolyticactivity in blood. In further references, we will therefore concentrate on these two specific parameters. The results based on current assays indicate that changes in PAI- 1 dominate over those of tPA throughout the 24-h period, such that residual tPA activity is tightly and inversely related to the levels of PAI- 1. It is not clear yet to what extent this dependence of tPA activity on PAL 1 observed in vitro also holds true for the in vivo situation. The changes in PAL1 are not platelet dependent. Furthermore, the response of tPA activity to local venous stasis does not appear to have a diurnal variation. In health, the circadian fluctuation of tPA and PAI-1 is not associated with changes of effective fibrinolysis, as assessed by plasma D-dimer.

CIRCADIAN VARIATION OF FIBRINOLYTIC ACTIVITY IN SPECIAL SUBGROUPS Ethnic Groups The circadian pattern of fibrinolytic activity in blood detected by Fearnley et al. in England (5) was later reproduced by Cepelak et al. in Czechoslovakia (8) and by us again in England for plasma tPA activity (19), suggesting that geographical provenance has little effect on the timing and basic features of this biorhythm. Takada et al. also reported a diurnal shortening of the ELT, attributable mainly to changes in PAI-1, in 25 healthy Japanese subjects (25). Recently, Johansen et al. measured the plasma levels of tPA and PAL 1 activity and antigen, every 4 h for 24 h, in 10 Eskimos and in 10 Caucasians residing in Greenland; the two groups led distinct lifestyles and had different dietary habits (24). No significant differences were found between groups with reference to the timing and amplitude of the fluctuations, area under the curves, and median tPA and PAL1 values during the 24 h, indicating that ethnic origin, lifestyle, and dietary habits do not significantly affect the circadian rhythm of fibrinolytic factors.

Shift Workers

If the timing of synchronizing factors-e.g., light/dark cycle, sleep/wake schedule, meals, or social cues-is abruptly shifted, a circadian rhythm usually requires many periods to regain its normal phase relationship to the synchronizing factor (26). When Fearnley et al. measured blood fibrinolytic activity at 4:OO p.m. and 4:OO a.m. in day and night nurses, a smaller variation was found in the night nurses; the authors




343

suggested that the effect was due to enhancement of fibrinolysis by physical activity running counter to the circadian rhythm in the night workers (5). Although it is clear, in that study, that the rhythm in the night nurses did not undergo a complete phase inversion, it is not possible to conclude, due to the lack of more frequent sampling, whether the range of fluctuation was reduced, or whether the rhythm was phase shifted, or both. Peternel et al. recently measured the plasma activity and antigen levels of tPA and PAI-1, every 4 h for 24 h, in 10 shift workers rotating weekly between a morning, evening, and night shift (27). An altered rhythm was found during both day and night shifts, characterised by a reduced amplitude in the fluctuation of tPA and by a phase advancement in the rhythm of tPA and PAI-1, compared to 10 matched control subjects. These findings are in agreement with previous reports indicating that frequently recumng alterations in sleeping patterns and environmental time cues lead to a state of desynchrony (28). The meaning of a disrupted rhythm of fibrinolytic factors, however, is not clear.

Effects of Drugs At least three classes of drugs have been investigated for an effect on the circadian variation of fibnnolytic factors: p-blockers, heparin, and gonadal steroids. Harenberg et al. measured, in 15 healthy subjects, the diurnal change of the ELT over a 2.5-h period after acute p- and a-receptor blockade, achieved by intravenous propranolol alone or in combination with phentolamine (29). No changes were observed compared to placebo treatment, suggesting that adrenergic-receptor stimulation is not responsible for the diurnal increase in fibrinolytic activity. We have recently investigated the effects of “chronic” p-blockade on the circadian variation of plasma tPA and PAI-1 activity and antigen levels in seven healthy subjects taking 160 mg daily of long-acting propranolol for 14 days (30). This regimen presumably achieved blockade of the p-adrenergic receptors in the pineal gland mediating the synthesis and secretion of melatonin (31), a biochemical messenger involved in the regulation of circadian biorhythms (32). No significant difference in the timing and range of fluctuations or in the 24-h median values were found in the treated group, compared to six control subjects (30). Therefore, none of the multiple effects of &blockade, including those in the central nervous system, bradycardia, reduction in blood pressure and cardiac output, or inhibition of lipolysis and glycogenolysis, seems to interfere with the endogenous pacemaker that controls the rhythm of tPA and PAL 1. Neerstrand et al. followed the diurnal changes of tPA antigen and PAL 1 activity over 8 h in six healthy subjects receiving subcutaneous low-molecular-weight heparin or conventional heparin; no differences were found compared to placebo treatment ( 18), indicating that, in normal conditions, inhibition of coagulation does not affect the rhythm and levels of tPA and PAI- 1. Similar results were obtained by Eriksson et al. (33). The effects of oral contraceptives and of the anabolic steroid, stanozolol, on the plasma levels of tPA and PAL 1 have been measured in morning blood samples; both drugs reduce the levels of PAI-1 activity (34,35); stanozolol also reduces tPA antigen


344



concentrations, but net tPA activity is increased, presumably because of a sharper reduction in PAL1 (35). Jespersen reported that the pattern of diurnal increase in euglobulin fibrinolytic activity in women on oral contraceptives was similar to that of women on no drugs (36). We recently followed the plasma levels of tPA antigen over 24 h in a young woman taking oral contraceptives; a marked reduction in the 24-h averaged value was found, with persistence of the normal range and phase of fluctuation, compared to control subjects (37). These data suggest that gonadal steroids depress the circulating levels of tPA antigen and PAI-I, but probably do not alter their 24-h rhythm. Patients With Acute or Chronic Diseases Acute Thrombosis Boyles in 1972 measured the ELT and a “fibrinolysin” inhibitor, by caseinolytic assay, every 4 h for 24 h in 16 patients with acute myocardial infarction, pulmonary embolism, or deep venous thrombosis (38). A circadian variation of both ELT and the inhibitor was found, with a pattern similar to that of matched control subjects; in particular, peak levels of the inhibitor were recorded at 4:OO a.m. Compared to controls, however, fibrinolytic activity in the patients was reduced, while the inhibitor was high throughout the 24-h period. Huber et al. measured the plasma levels of PAL 1 activity and tPA antigen every 6 h for 48 h in patients with unstable angina (n = 27) or acute myocardial infarction (n = 36) (39). The phase and range of the circadian variation of tPA and PAL1 was similar to that of matched controls, but PAL1 activity in the patients was significantly higher, especially in those with infarction. We recently recorded daily plasma PAI-1 activity levels from admission up to 3 days later in 24 patients with acute myocardial infarction, treated with recombinant tPA within 6 h of the onset of symptoms (40). On admission, PAL1 activity was significantly higher in patients hospitalized during the night, compared to the inhibitor activity of patients admitted during the day. After 24 h, however, the inhibitor levels increased significantly and failed to show a relationship with the time of day, suggesting that the underlying circadian rhythm was concealed by an acute-phase reactant behaviour of PAI- 1 to myocardial necrosis (4 1). Our findings are slightly at variance with those of the previous investigators; it is possible that the thrombolytic treatment in our study hindered the detection of the circadian rhythm of PAL1 through effects on the assay system or through in vivo induction of PAL1 by tPA (vide infra). Six et al. also reported the persistence of the circadian variation of plasma PAI-I activity in patients with unstable angina (n = 18)and acute myocardial infarction (n = 20) at the time of hospital admission (42). Chronic Diseases Hajjar et al. in 1961 reported the presence of a physiological diurnal fluctuation of the ELT over a 5-h interval in five patients with controlled diabetes and in five patients with hepatic cirrhosis; in the latter, mean fibrinolytic activity was enhanced (43). Cepelak et al. in 1967 measured the ELT over a 24-h period in patients with

Chronobiol Int, Vol. 8, No. S. 1991



345

coronary artery disease and found a reduction in fibrinolytic activity during both day and night, compared to healthy subjects (44). In contrast, patients with recurrent venous thrombosis showed a wider range of fluctuation of fibrinolytic activity over the 24 h compared to normals (44). In 1973, Rosing et al. measured the euglobulin fibrin plate lysis area, every 3 h from 8:OO a.m. to 8:OO p.m., in 20 patients with type IV hyperlipoproteinaemia and in 16 normolipidaemic patients with coronary artery disease (12). A defective diurnal rise was observed in the two groups, with only a 52 and 93% increase, respectively, compared to a 184%rise recorded in 24 age-matched controls. This blunted diurnal increase in fibrinolytic activity, observed using overall methods, might be explained by a smaller drop in PAL 1 in the late afternoon or by a shift in the rhythm of PAI- 1 resulting in a delayed time of trough; both of these mechanisms have been recently reported in patients with diffuse atherosclerosis (vide infra). Simpson et al. in 1983 recorded fibrinolytic activity using the DBCLT, every 2 to 4 h for 24 h, in 18 patients with severe hypertriglyceridaemia and in 11 age-matched controls (45). Fibrinolytic activity, averaged over the 24 h, was significantly lower in the patient group; moreover, the range of fluctuation was reduced, and the time of trough activity was shifted from 3:OO to 9:30 a.m. After triglyceride reduction, from 6.8 to 3. I mmol/L, the mean 24-h fibrinolytic activity in the patients rose significantly by 579’0,the range of variation increased, but trough activity remained shifted to 9:30 a.m. In 1985, Keegan et al. measured fibrinolytic activity by the fibrin plate lysis area between 9:OO a.m. and 2:OO p.m. in 25 patients with systemic sclerosis;the physiological diurnal increase was absent in the patient group, compared to 16 active or 10 resting control subjects (46). Recently, we recorded the plasma levels of tPA antigen and tPA and PAI- 1 activity every 3 h for 24 h in six patients with severe, diffuse atherosclerosis(37). For tPA and PAL1 activity, a significant fluctuation was found in the patient group; the 24-h averaged values did not differ significantly from those of young healthy subjects, although PAL 1 activity in the late afternoon tended to remain high; moreover, the timing of peak and trough tPA and PAI-1 activity was phase delayed by 3 h. In the patient group, tPA antigen was strikingly higher, suggesting that a high proportion of tPA was complexed to inhibitors, and did not vary significantly over the 24-h period (37). Angleton et al. recently reported significantly lower plasma tPA activity levels after brachial venous occlusion in patients with coronary artery disease, compared to a younger control group, in the evening but not in the morning (23); the latter may be a reflection of the defective diurnal rise in fibrinolytic activity, previously reported in patients with coronary artery disease (12). These findings in various patient groups suggest that the circadian variation of plasma PAL 1 is preserved in the very early stage of thrombotic disorders. In chronic diseases, the circadian rhythm of fibrinolytic activity may remain unchanged, as in controlled diabetes, or may undergo a variety of alterations; it may show changes in the 24-h mean towards a rise (cirrhosis) or a fall, (hypertriglyceridaemia, coronary artery disease); the phase of the oscillation may be shifted, as in patients with hypertriglyceridaemia or diffuse atherosclerosis; the range of the fluctuation may increase (deep venous thrombosis) or fall (hypertriglyceridaemia, coronary artery disease);

Chronobiol Int. Vol. 8, No. 5. 1991

346



finally, the rhythm may disappear, as in patients with systemic sclerosis. The exact significance of any one of these alterations is not clear. However, given the frequency and variety of changes, if plasma levels of tPA and PAI- 1 are to be used to characterise disease, it is of obvious importance to define the 24-h pattern in patient groups before selecting a single blood sampling time. As pointed out by Kluft et al., the most accurate detection of changes in PAI- 1 will occur at different times of day, depending on whether an alteration in the 24-h mean, amplitude, or phase of the circadian fluctuation has occurred (47).

CONTROL OF THE CIRCADIAN VARIATION OF FIBRINOLYTIC ACTIVITY Since changes in plasma PAI- 1 ultimately determine the circadian rhythm of spontaneous fibnnolytic activity in blood, as recorded in vitro, we will concentrate mainly on the possible mechanisms regulating a circadian variation in the synthesis, secretion, and clearance of this protein. PAI-1 is produced constitutively by a number of cell types, including endothelium, hepatocytes, and vascular smooth muscle, and is contained in platelets (3). It is largely cleared by the liver and has a circulating half-life on the order of tens of minutes. The origin of plasma PAI-1 has not been definitely established (3). Circadian changes in overall endothelial function do not appear to be responsible for the fluctuation of PAI-1, since the plasma concentration of von Willebrand factor, a marker of endothelial function, does not vary over the 24 h (48). Likewise, a circadian variation in platelet-derived PAI-1 also appears to be unlikely, as the plasma levels of other platelet release products, such as P-thromboglobulin or platelet factor 4, remain stable during the 24-h cycle (22,48). Simultaneous measurements in plasma of fibrinolytic activity and hormone levels at different times of day have shown a lack of concordance between the diurnal changes in fibrinolysis and changes in corticosteroids (49,50), catecholamines (50), and insulin (50). These observations have led to the conclusion that no causal relationship exists between such hormones and the circadian rhythm of fibrinolysis. When discussing the involvement of humoral factors, however, it should be kept in mind that full induction of PAI- 1 synthesis requires at least 4-6 h between the time of stimulation and an effect on protein secretion (5 1,52). Moreover, a rhythm of secretion may be generated even in response to a constant stimulatory input, provided the target organ has a circadian rhythm of responsiveness (26). Evidence against an adrenergic mechanism regulating the circadian variation of fibrinolytic activity is provided by the fact that acute p- and a-blockade (29) and chronic &blockade (30) have no effect on the circadian changes of the ELT or of plasma tPA and PAI-1 activity and antigen levels. Activated protein C can inactivate and degrade PAI- 1 (53) and might therefore influence the circadian rhythm of this inhibitor. Okajima et al. recently administered highly purified human activated protein C to healthy subjects and found no effect on the diurnal variation of PAI- 1 (54). Although exercise can stimulate fibrinolytic activity, several studies have demonstrated that the circadian rhythm of blood fibrinolytic activity, and presumably of plasma PAI- 1, is preserved in the absence of physical exertion ( 5 1 1,12). Body pos-

Chronohiol Int, Vol. 8, No. 5, 1991



34 7

ture might influence fibrinolytic activity through its effect on regional blood flow distribution; in particular, in the supine position, hepatic blood flow is increased ( 5 5 ) and consequently the hepatic clearance of fibrinolytic factors is expected to rise. Such a mechanism might explain the small nocturnal fall in plasma uPA and pro-uPA, but is not sufficientto account for the larger changes in tPA and PAL 1. Furthermore, the nocturnal rise in plasma tPA and PAI- 1 antigen would invoke a reduction rather than an increase in liver blood flow at night. Changes in cardiac output may influence the rhythm of both tPA and PAI-1, as suggested by Peternel et al.; in 10 patients with fixed heart rate due to artificial pacemaker implantation, they reported a reduced fluctuation of plasma tPA antigen and PAI- 1 activity compared to 10 matched controls (56). The circadian rhythm of plasma PAL1 may be coordinated to that of tPA antigen, despite the different timing of the two fluctuations. This concept is supported by the relationship frequently observed between the plasma levels of tPA antigen and PAI- 1 activity or antigen (2 1,57-59); moreover, the genomic code of both proteins shows structural homology (60). Recently, induction of PAL1 by tPA has been reported in human hepatic and endothelial cells (52). If a common mediator is involved in the regulation of tPA and PAI-1, the shifted phase relationship could be explained by differences in the kinetics of induction. Body temperature has a distinct circadian variation, reaching a peak between 6:OO and 1O:OO p.m., and then dropping to a minimum around 3:OO a.m. (61). If the macrophage product, endogenous pyrogen/interleukin- 1, plays a role in the physiological variation of body temperature, it is tempting to hypothesise a rhythmical PAL 1 induction by interleukin- 1, consistent with in vitro studies demonstrating increased PAL1 expression by this cytokine ( 5 1). Other powerful stimuli for PAI-1 production in vitro and in vivo include several growth factors (62,63). It is conceivable that cyclical variations in the plasma concentration or effect of these factors may regulate the circadian rhythm of PAI- 1. Thus, the mechanisms regulating the circadian rhythm of fibrinolytic activity in blood, and in particular of plasma PAL 1, are not fully understood. Variations in platelet release products, endothelial function, and adrenergic stimulation are unlikely to be involved. Changes in haemodynamic parameters alone appear to be insufficient to explain the broad range of fluctuation. A humoral mechanism would appear to be the most likely.

CONCLUSION Plasminogen activation is involved in many diverse physiological processes in addition to fibrin degradation, including tissue repair, nidation, embryogenesis, cell migration, and angiogenesis (3). The normally occurring circadian fluctuation of tPA and PAI- 1 may represent an index of the functional capacity of the fibrinolytic/proteolytic system (12) and should be regarded as a complementary feature of homeostasis. With reference to the circulation, some possible clinical implications of the circadian rhythm of fibrinolytic activity can be derived from Fig. 2. Under normal conditions, tPA and PAI- 1 largely represent a potential fibrinolyticcapacity, such that their

Chronobiol Inl, Vol. 8. No. 5. 1991


348

B

11.11.111.1.111111


A t

9 am

I

9 Pm Time of day

I am

FIG. 2. Schematic interaction between fibrin generation and fibrinolytic activity in blood during a 24-h period. In normal conditions, fibrin formation is constantly low (line A) and fibrinolytic capacity, represented by the area under the sinusoidal curve, is efficient in degrading it throughout the day (area below line A = efficient fibrinolysis). In thrombotic disorders, fibrin generation (line B) may exceed fibrinolytic capacity at certain times of day (hatched area) or may overwhelm the fibrinolytic capacity of blood during the entire 24 h (line C).

circadian variation is not associated with changes in effective fibrinolysis (Fig. 2, line A). In thrombotic diseases, the haemostatic balance is shifted towards greater fibrin production, so that the circadian fluctuation of fibrinolytic activity may result in times of day when fibrinolysis is incompetent to counterbalance the higher level of fibrin generation (Fig. 2, line B). In this circumstance, the physiological drop of fibrinolytic activity in the morning hours may favour thrombus development at this time of day. Such a mechanism is consistent with the reported higher morning incidence of acute thrombotic events (64-67). If the stimulus to thrombosis is so high to overwhelm the fibrinolytic capacity in blood (Fig. 2, line C), the latter may become insufficient to neutralize fibrin production over the entire 24-h period. Such a mechanism could explain the large proportion of acute thrombotic episodes that occur with equal distribution throughout the day.

REFERENCES I . Astrup T. The biological significance of fibrinolysis. Lancet 1956;2:565-8. 2. Bachmann F. Fibrinolysis. In: Verstraete M, Vermylen J, Leijnen L, Arnout J, eds. Thrombosis and huemostusis. Leuven: Leuven University Press, I987:227-65. 3. Sprengers ED, Kluft C. Plasminogen activator inhibitors. Blood 1987;69:381-7. 4. Fearnley GR. Fibrinolysis. London: Edward Arnold Ltd., 1965:28-44. 5. Fearnley GR, Balmforth G, Fearnley E. Evidence of a diurnal fibrinolytic rhythm; with a simple method of measuring natural fibrinolysis. Clin Sci 1957; 16:645-50. 6 . Buckell M, Elliott FA. Diurnal fluctuation of plasma fibrinolytic activity in normal males. Lancet 1959;1:660-2. 7. Billimoria JD, Drysdale J, James DCO, Maclagan NF. Determination of fibrinolytic activity ofwhole blood with special reference to the effects of exercise and fat feeding. Lancet 1959;11:471-5. 8. Cepelak V, Barcal R, Lang N, Cepelakova H. Zum Tag- und Nachtrhythmus der Fibrinolyse. Z Inner Med 1966;2 1:202-4. 9. Cepelak V, Barcal R. Fibrinolysis in twins and its variation in circadian rhythm. Plzen Lek Sborn 1959;129:22. 10. Mann RD. Effect of age, sex, and diurnal variation on the human fibrinolytic system. J Clin Puthol 1967;20:223-6. 1 I . Rosing DR, Brakman P, Redwood DR, et al. Blood fibrinolytic activity in man: diurnal variation and the response to varying intensities of exercise. Circ Res 1970;27: 17 1-84. 12. Rosing DR, Redwood DR, Brakman P, Astrup T, Epstein SE. Impairment of the diurnal fibrinolytic response in man: effects of aging, type IV hyperlipoproteinemia and coronary artery disease. Circ Res 1973;32:752-7.

Chronobiol Inl, Vol. 8, No. 5. 1991



349

13. Korsan-Bengtsen K, Wilhelmsen L, Tibblin G. Blood coagulation and fibrinolysis in relation to degree of physical activity during work and leisure time. Acta Med Scand 1973;193:73-7. 14. Kluft C. C 1-inactivator-resistant fibrinolytic activity in plasma euglobulin fractions: its relation to vascular activator in blood and its role in euglobulin fibrinolysis. Thromb Res 1978;13: 135-5 1. 15. Kluft C, Jie AFH, Allen AR. Behaviour and quantitation ofextrinsic (tissue-type) plasminogen activator in human blood. Thromb Haemost 1983;50518-23. 16. KruithofEKO, Ransijn A, Bachmann F. Inhibition of tissue plasminogen activator by human plasma. In: Davidson JF, Bachmann F, Bouvier CA, Kruithof EKO, eds. Progress injibrinolysis, Vol. 4. Edinburgh: Churchill Livingstone, 1983:362-6. 17. Kluft C, Verheijen JH, Rijken DC, Chang GTG, Jie AFH, Onkelinx C. Diurnal fluctuations in the activity of the fast-acting t-PA inhibitor. In: Davidson JF, Donati MB, Coccheri S, eds. Progress in jibrinolysis, Vol. 7. Edinburgh: Churchill Livingstone, 1985:1 17-9. 18. Neerstrand H, Ostergaard P, Bergqvist D, Matzsch T, Hedner MU. tPA inhibitor, tPAAg, plasminogen and alpha-2-antiplasmin after low molecular weight heparin or standard heparin. Fibrinolysis 1987;1:39-43. 19. Andreotti F, Davies GJ, Hackett DR, et al. Major circadian fluctuations in fibrinolytic factors and possible relevance to time of onset of myocardial infarction, sudden cardiac death and stroke. Am J Cardiol 1988;62:635-7. 20. Andreotti F, Davies GJ, Hackett D, et al. Circadian variation of fibrinolytic factors in normal human plasma. Fibrinolysis 1988;290-2. 2 1. Kluft C, Jie AFH, Rijken DC, Verheijen JH. Daytime fluctuations in blood oftissue-type plasminogen activator (t-PA) and its fast-acting inhibitor (PAI-1). Thromb Haemost 1988;59:329-32. 22. Grimaudo V, Hauert J, Bachmann F, Kruithof EKO. Diurnal variation of the fibrinolytic system. Thromb Haemost 1988;59:495-9. 23. Angleton P, Chandler WL, Schmer G. Diurnal variation of tissue-type plasminogen activator and its rapid inhibitor (PAL 1) Circulation 1989;79:10 1-6. 24. Johansen LG, Gram J, Kluft C, Jespersen J. Chronobiology of coronary risk markers in Greenland Eskimos-a comparative study with Caucasians residing in the same Arctic area. Chronobiol Znt 1991;8:352-60. 25. Takada A, Takada Y, Urano T, Sakakibara K, Rydzewski A. Fluctuations of euglobulin lysis time, tissue plasminogen activator, and free and total plasminogen activator inhibitor levels in plasma in daytime. Thromb Res 199057:13-20. 26. Aschoff J. Circadian rhythms: general features and endocrinological aspects. In: Martini L, Krieger DT, eds. Comprehensive endocrinology series. Endocrine rhythms. New York: Raven Press, 1979; 161. 27. Peternel P, Stegnar U, Salobir B, Salobir D, Keber D, Vene N. Shift work and circadian rhythm of blood fibrinolytic parameters. Fibrinolysis 1990;4:1 13-5. 28. Moore-Ede MC, Czeisler CA, Richardson GS. Circadian timekeeping in health and disease. Part 2. Clinical implications of circadian rhythmicity. N Engl J Med 1983;309:530-6. 29. Harenberg J, Weber E, Spohr U, Marl H. Is the diurnal increase of fibrinolytic activity influenced by alpha- or beta-adrenergic blockade? Hut 1980;4 1:455-8. 30. Andreotti F, Kluft C, Davies GJ, Huisman LGM, de Bart ACW, Maseri A. Effect of propranolol (long-acting) on the circadian fluctuation of tissue-plasminogen activator and plasminogen activator inhibitor-1. A m J Cardiol (in press). 3 1. Ganong WF. Review of medical physiology, 9th edition. Los Altos: Lange Medical Publications, 1979:360. 32. Palms L, Blennow G, Wetterberg L. Correction of non-24-hour sleeplwake cycle by melatonin in a blind retarded boy. Ann Neurol 1991;29:336-9. 33. Eriksson E, Wollter IM, Christenson B, Stigendal L, Risberg B. Heparin and fibrinolysis-comparison of subcutaneous administration of unfractionated and low molecular weight heparin. Thromb Haemost 1988;59:284-8. 34. Jespersen J, Kluft C. Inhibition of tissue-type plasminogen activator in women using oral contraceptives and in normal women during a menstrual cycle. Thromb Haemost 1986;55:388-9. 35. Verheijen JH, Rijken DC,Chang GTG, Preston FE, Kluft C. Modulation of rapid plasminogen activator inhibitor in plasma by stanozolol. Thromb Haemost 1984;s 1:396-7. 36. Jespersen J. The diurnal increase in euglobulin fibrinolytic activity in women using oral contraceptives and in normal women, and the generation of extrinsic fibrinolytic activity. Thromb Haemost 1986;56:183-8.

Chronobiol Int, Vol. 8, No. 5. 1991


350


37. Andreotti F, Davies GJ, Maseri A, Kluft C. Inhibition of fibrinolysis in blood: circadian fluctuation and possible relevance to coronary artery disease. In: von Arnim T, Maseri A, eds. Predisposing conditions for acute ischaemic syndromes. Darmstadt: Steinkopff Verlag, 1989:20-32. 38. Boyles PW. Fibrinolytic activity and thromboembolism. Clin Pharrnacol Ther 1972;I3:534-8. 39. Huber K, Rosc D, Resch 1, et al. Circadian fluctuations of plasminogen activator inhibitor and tissue plasminogen activator levels in plasma of patients with unstable coronary artery disease and acute myocardial infarction. Thromb Huemost 1988;60:372-6. 40. Andreotti F, Roncaglioni MC, Hackett DR, et al. Early coronary reperfusion blunts the procoagulant response of plasminogen activator inhibitor- I and von Willebrand factor in acute myocardial infarction. J A m CON Curdiol 1990;16:1553-60, 41. Kluft C, Verheijen JH, Jie AFH, et al. The postoperative fibrinolytic shutdown: a rapidly reverting acute phase pattern for the fast-acting inhibitor of tissue-type plasminogen activator after trauma. Scund J Clin Lab Invest 1985;45:605-10. 42. Six AJ, Liem MK, Dooijewaard G, et al. The acute-phase reaction of the fibrinolytic system in acute myocardial infarction and unstable angina. Coronary Artery Dis (in press). 43. Hajjar GC, Whissen NC, Moser KM. Diurnal variation in plasma euglobulin activity and fibrinogen levels. Angiology 1961;12:160-4. 44. Barcal R, Cepelak V, Cepelakova H, Lang N. The role of diurnal cycle in the blood clotting and fibrinolysis. Plzen Lek Sborn 1967; I8:7 1. 45. Simpson HCR, Mann JI, Meade TW, Chakrabarti R, Stirling Y, Woolf L. Hypertriglyceridaemia and hypercoagulability. Lancet 1983;I:786-90. 46. Keegan AL, Holland CD, Jayson MIV. Loss of diurnal variation in fibrinolytic activity in systemic sclerosis. Br J Rheumatol 1985;24:2 I I . 47. Kluft C, Andreotti F. Consequences of the circadian fluctuation in plasminogen activator inhibitor 1 (PAI-I) for studies on blood fibrinolysis. Fihrinolysis I988;2:93-5. 48. Johansen LG, Gram J, Kluft C, Jespersen J. Circadian variation of extrinsic fibrinolytic components in blood from Eskimos. Fibrinolysis 1990;4:35-9. 49. Menon SI, Smith PA, White RWB, Dewar HA. Diurnal variations of fibrinolytic activity and plasma1 I-hydroxycorticosteroid levels. Lancet 1967;11:531-3. 50. Chandler WL, Mornin D, Whitten RO, et al. Insulin, cortisol and catecholamines do not regulate circadian variations in fibrinolytic activity. Thromb Res 1990;58:1- 12. 5 I . Schleef RR, Bevilacqua MP, Sawdey M, Gimbrone MA, Loskutoff DJ. Cytokine activation of vascular endothelium. Effects on tissue-type plasrninogen activator and type 1 plasminogen activator inhibitor. J Biol Chem 1988;263:5797-803. 52. Fujii S, Lucore CL, Hopkins WE, Billadello JJ, Sobel BE. Induction of synthesis of plasminogen activator inhibitor type- I and tissue-type plasminogen activator in human hepatic and endothelial cells. Thromb Haemost 199064:412-9. 53. de Fouw NJ, van Hinsbergh VWM, de Jong YF, Haverkate F, Bertina RM. The interaction of activated protein C and thrombin with the plasminogen activator inhibitor released from human endothelial cells. Thromb Huemost 1987;57: 176-82. 54. Okajima K, Koga S, Kaji M, et al. Effect of protein C and activated protein C on coagulation and fibrinolysis in normal human subjects. Thromb Huemost 1990;63:48-53. 55. de Boer A, Kluft C, Cohen AF. Life style and blood fibrinolytic activity: a comment on clearance of plasminogen activators. Frbrinolysis 1990;4:6 1-3. 56. Peternel P, Mozina H, Stegnar M, Keber D, Vene N. Reduced diurnal rhythm of fibrinolytic activity in patients with pace-maker. Fibrinolysis 1988;2(suppl l):7. 57. Paramo JA, Colucci M, Collen D. Plasminogen activator inhibitor in the blood of patients with coronary artery disease. Br Med J 1985;291573-4. 58. Olofsson BO, Dahlen G, Nilsson TK. Evidence for increased levels of plasminogen activator inhibitor and tissue plasminogen activator in plasma of patients with angiographically verified coronary artery disease. Euu Heart J 1989; 1077-82. 59. Juhan-Vague I, Alessi MC, Joly P, et al. Plasma plasminogen activator inhibitor-I in angina pectoris. Influence of plasma insulin and acute-phase response. Arteriosclerosis 1989;9:362-7. 60. Bosma PJ, van den Berg EA, Kooistra T, Siemieniak DR, Slightom JL. Human plasminogen activator inhibitor-I gene. Promoter and structural gene nucleotide sequences. JBiol Chem 1988;263:9129-41. 6 1. Petersdorf RG, Root RK. Disturbances of heat regulation. In: Braunwald E, Isselbacher KJ, Petersdorf

Chronohrol Int, Vol. 8. No. 5 . 1 YY1


62. 63. 64. 65.


66. 67.

351

RG, Wilson JD, Martin JB, Fauci AS, eds. Harrison’.vprinciples of internal medicine, I Ith edition. New York: McGraw-Hill Book Company, 1987:44. Laiho M, Saksela 0,Andreasen PA, Keski-Oja J . Enhanced production and extracellular deposition of the endothelial-type plasminogen activator inhibitor in cultured human lung fibroblasts by transforming growth factor-beta. J Cell Biol 1986;103:2403- 10. Fujii S, Sobel BE. Induction of plasminogen activator inhibitor by products released from platelets. Circulation 199082:1485-93. WHO International Study Group. Myocardial infarction community registers: public health in Europe, Vol. 5. Copenhagen: Regional Office for Europe (WHO), 1976:188-91. MullerJE, Ludrner PL, Willich SN, et al. Circadian variation in the frequency ofsudden cardiacdeath. Circulation 1987;75:131-8. Behar S, Reicher-Reiss H, Goldbourt U, Kaplinski E, and Principal Investigators of SPRINT, Israel. Circadian variation in pain onset in unstable angina. Am J Cardiol 1991;67:91-3. Marshall J. Diurnal variation in occurrence of strokes. Stroke 1977;88:230-1.

Chronobiol Int, Vol. 8, No. 5 , 1991

Circadian variation of blood-pressure.

[The intraindividual variation of fibrinolytic activity in volunteers (author's transl)].

Circadian variation of blood pressure in patients with diabetic nephropathy.

Circadian blood pressure variation in normotensive patients with panic disorder.

Circadian variations of platelet aggregability and fibrinolytic activity in patients with ischemic stroke.

Circadian variations of platelet aggregability and fibrinolytic activity in healthy subjects.

Insulin, cortisol and catecholamines do not regulate circadian variations in fibrinolytic activity.

Distribution and variation of fibrinolytic activity in the walls of human arteries and veins.

Fibrinolytic activity in vascular grafts.

Letter: Fibrinolytic activity in treatment of diabetes.

Fibrinolytic activity in malignant disease.

Physical activity and the circadian rhythm of blood pressure.

Fibrinolytic activity in women on oral contraceptive pills: variation due to haemoglobin genotype.

Fibrinolytic Activity of Blood and its Determinants in Healthy Medical Students.

Menstruation: extravascular fibrinolytic activity and reduced fibrinolytic capacity.

Circadian variation of gentamicin toxicity in rats.

Antifibrinolytic therapy in ruptured intracranial aneurysm through repeated monitoring of fibrinolytic activity of blood.

Fibrinolytic activity in health and vascular disease.

Decreased fibrinolytic activity in juvenile chronic arthritis.

Low vascular fibrinolytic activity in obesity.

Circadian variation and cardiovascular disease.

In vitro whole blood clot lysis for fibrinolytic activity study using d-dimer and confocal microscopy.

Circadian variation in unexpected postoperative death.

Fibrinolytic activity in patients with acute tonsillitis.