Clinical Biochemistry 47 (2014) 570–573
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
D-dimer
predicts major bleeding, cardiovascular events and all-cause mortality during warfarin treatment Marcus Lind a,⁎, Kurt Boman a, Lars Johansson a, Torbjörn K. Nilsson b, Lisbeth Slunga Järvholm a, Jan-Håkan Jansson a a b
Department of Public Health and Clinical Medicine, Umeå University, Umeå, Sweden Department of Clinical Chemistry, Örebro University Hospital, Örebro, Sweden
a r t i c l e
i n f o
Article history: Received 15 October 2013 Received in revised form 3 February 2014 Accepted 9 March 2014 Available online 15 March 2014 Keywords: D-dimer Bleeding Warfarin Mortality PAI-1 Cardiovascular events
a b s t r a c t Objectives: Previous studies have shown that biomarkers in blood plasma can predict bleeding complications during anticoagulant treatment as well as thromboembolic events and may improve existing risk stratification schemes in patients on or considered for oral anticoagulant treatment. The aim of this study was to investigate if levels of D-dimer, tissue plasminogen activator (tPA) and its complex with plasminogen inhibitor type 1 (tPA/PAI-1 complex) can predict major bleedings, cardiovascular events and all-cause mortality in patients with warfarin treatment. Design and methods: In a longitudinal cohort study, 719 patients on oral anticoagulant treatment were followed for a total of 3001 treatment years. Major bleeding, stroke, arterial emboli, myocardial infarction and death were recorded and classified. Blood samples collected at baseline were analyzed for D-dimer, tPA, and tPA/PAI-1 complex. Results: In multivariate Cox regression analysis, high levels of D-dimer were associated with major bleeding (HR 1.27 per SD; 95% CI: 1.01–1.60), cardiovascular events (HR 1.23 per SD; 95% CI: 1.05–1.45) and all-cause mortality (HR 1.25 per SD; 95% CI: 1.06–1.47). Neither tPA nor the tPA/PAI-1 complex was associated with major bleeding, cardiovascular events or all-cause mortality. Conclusion: We conclude that high levels of D-dimer predict major bleeding, cardiovascular events and all-cause mortality during warfarin treatment. © 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction Treatment with an oral anticoagulant is effective in preventing thromboembolic events but carries an increased risk of bleeding complications [1]. To maximize the benefit of anticoagulant treatment in atrial fibrillation several risk scores like the CHADS2 [2] have been developed to help predict which patients have the highest risk of future thromboembolic events. Unfortunately risk factors for thromboembolic events are also associated with a higher risk of serious bleedings [3] and higher CHADS2 scores are associated with an increased risk of bleeding complications [4]. We have previously shown that biomarkers in blood plasma can predict bleeding complications during anticoagulant treatment [5] as well as thromboembolic complications [6]. It is possible that biomarkers could improve existing risk stratification schemes in patients on or considered for oral anticoagulant treatment.
⁎ Corresponding author at: Department of Medicine, Skellefteå County Hospital, 931 86 Skellefteå, Sweden. Fax: +46 910771657. E-mail address: [email protected] (M. Lind).
From previous studies we know that tissue plasminogen activator (tPA) and the complex between tPA and its inhibitor plasminogen activator inhibitor (tPA/PAI-1 complex) are related to cardiovascular diseases [7,8]. In patients on warfarin treatment tPA has been associated with cardiovascular events [9]. Lack of fibrinolytic inhibitors causes bleeding symptoms of varying severity [10,11]. D-dimer, a marker of fibrin degradation, is also considered to be associated with cardiovascular disease [12,7,13]. Studies on warfarin treated patients have shown diverging results for the association between D-dimer and cardiovascular events. Two studies have shown a significant association between high levels of D-dimer and cardiovascular events in patients with atrial fibrillation [8,13] and also that high levels of D-dimer could be associated with bleeding complications [13]. However, one recent study failed to shown any association between D-dimer and mortality or cardiovascular events [14]. It is reasonable to assume that levels of fibrinolytic factors could be associated with risk of bleeding complications as well as thromboembolic events during anticoagulant treatment. The aim of this study was to investigate if levels of D-dimer, tPA and tPA/PAI-1 complex can predict major bleedings, cardiovascular events and all-cause mortality in patients with warfarin treatment.
http://dx.doi.org/10.1016/j.clinbiochem.2014.03.003 0009-9120/© 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
M. Lind et al. / Clinical Biochemistry 47 (2014) 570–573
Methods Patients The study population has been described earlier [5]. In short patients were simultaneously recruited from the warfarin clinics at Skellefteå County Hospital and at Umeå University Hospital in 1996. All patients with a planned treatment of more than 3 months were eligible for the study. Consent forms were sent to all 957 eligible patients, and 847 accepted to participate in the study. Before inclusion, 128 patients were excluded due to treatment stoppage, death or missing blood samples. In the end, 719 patients (356 from Skellefteå and 363 from Umeå warfarin clinics) were included in the study cohort. From the Skellefteå warfarin clinic data about diabetes, prior peptic ulcer, prior bleeding peptic ulcer, diabetes, weight and height were obtained through questionnaires. The study was approved by the Research Ethics Committee of Umeå University and all patients gave informed consent. Blood sampling and laboratory procedures All patients were on stable treatment with oral anticoagulants for at least 2 months prior to blood sampling. Venous blood samples were drawn with a minimum of stasis and collected in siliconized routine citrated plasma tubes containing 0.13 mol/L sodium citrate. After centrifugation the plasma samples were frozen and stored at − 70 °C until analyzed. Both study cohorts were analyzed at the same time and location. The laboratory staff had no knowledge of event status. tPA and tPA/PAI-1 complex antigen were determined with enzyme-linked immunosorbent assay (ELISA). The reagent kits, Imulyse tPA and TintElize tPA/PAI-1 were purchased from Biopool AB, (Umeå, Sweden). For tPA, coefficient of variation (CV) was 14.1% at 6.6 ng/mL in our hands. For tPA/PAI-1, CV was 10.7% at 2.4 ng/mL in our hands. D-dimer was measured with an Immuno-turbidimetric Assay from Diagnostica Stago, Inc. The interassay CV for D-dimer at 1.7 mg/mL was 5.0% in our hands. C-reactive protein (CRP) was determined with an automated CRP method, (IMMULITE Diagnostic Products Corporation, USA). The interassay co-efficient of variation was b6% in our hands. Creatinine was analyzed on a Hitachii 911 multianalyzer (Roche, Mannheim, Germany) with kits from Roche/Boehringer (Crea plus, enzymatic method). International normalized ratio (INR) was determined at each hospital laboratory. The laboratories used the same Owren method with the reagent SPA 50 (Diagnostica Stago, Inc.) and INR calibration. Follow-up study protocol The date of inclusion was set to the date of blood sampling with the earliest inclusion date being June 1st 1996. All patients were followed longitudinally until death, cessation of warfarin treatment or until January 1st 2002. In an effort to identify all bleeding and cardiovascular events causing hospital admission or death, all medical records from the departments of medicine, cardiology, surgery, ear nose and throat, ophthalmology, urology, neurology, neurosurgery, and orthopedic surgery were studied from June 1, 1996 up to January 1, 2002. One patient moved out of the region during the study period and was followed up to the date of migration. Major bleeding, myocardial infarction (MI), ischemic stroke (IS) and peripheral arterial emboli (AE) causing admission to hospital or death were recorded and classified by a panel of three researchers (ML, JHJ, LJ). Major bleedings were defined according to the International Society of Thrombosis and Haemostasis [15] as fatal bleeding and/or symptomatic bleeding in a critical area or organ and/or bleeding causing a fall in a hemoglobin level of 20 g L−1 or more or leading to transfusion of two or more units of whole blood. All other bleedings were classified as minor and were excluded. Cardiovascular events included all MI, IS and AE with and without fatal outcome. The cause of death was registered
571
and classified according to the death certificate and in all but one the cause of death could be classified. Ascertainment bias was avoided by investigators classifying events being blinded to the biochemical results. Statistical analysis Analyses of D-dimer, tPA and tPA/PAI-1 complex were performed as continuous variables and categorized into tertiles, with the lowest tertile as the reference group. Univariate Cox regression analysis was performed on each of the variables to estimate the hazard ratio (HR) and 95% confidence interval (CI). Multivariate Cox regression analysis was performed to estimate the effects on different determinants when controlling for other factors. Factors were excluded from the multivariate model if the univariate Cox regression resulted in a clearly non-significant HR defined as P N 0.2. Both creatinine and CRP were considered to be potential confounders and were included in the multivariate analysis. The distributions of tPA, tPA/PAI-1 complex, D-dimer, creatinine, and CRP were skewed and therefore transformed using the natural logarithm (ln). Data on hypertension, diabetes, BMI, and previous peptic ulcer were available from the Skellefteå warfarin clinic and were tested for association with major bleeding, cardiovascular events and all-cause mortality with Cox regression analysis. Direct age adjustment was performed in ten year intervals. A P-value b 0.05 (two-sided) was considered statistically significant. Individuals with missing values were excluded from the statistical analyses (32 for D-dimer and creatinine, 2 for hsCRP and none for tPA and tPA/PAI-1 complex). SPSS version 15.0 was used for all statistical analyses. Results The mean age of participants was 70 years and 37% were female. Patients were followed for a total of 3001 treatment years with a mean of 4.2 years and a maximum follow-up time of 5.6 years. The most common indication for warfarin treatment was prosthetic heart valve, followed by atrial fibrillation; see Table 1. INR values at the time of sampling were available from the Skellefteå warfarin clinic (9% of Table 1 Baseline characteristics of the study population. Study cohort (n = 719) Mean age at inclusion, years (SD) Female (%) Mean follow-up time, years (SD) D-dimer, μg/L, median (IQR) Lowest tertile Intermediate tertile Highest tertile tPA, ng/mL, median (IQR) Lowest tertile Intermediate tertile Highest tertile tPA/PAI-1 complex, ng/mL, median (IQR) Lowest tertile Intermediate tertile Highest tertile CRP, mg/L, median (IQR) Creatinine, μmol/L, median (IQR) Indications for warfarin treatment Prosthetic heart valve (%) Atrial fibrillation (%) Venous thromboembolism (%) Ischemic stroke (%) Peripheral arterial thromboembolism Miscellaneous (%) Not defined (%)
70 (11) 268 (37) 4.2 (1.8) 0.13 (0.07–0.24) 0–0.09 0.10–0.18 0.19–∞ 11.6 (8.6–14.8) 0–9.49 9.50–13.32 13.33–∞ 11.4 (5.3–11.2) 0–6.01 6.02–9.88 9.89–∞ 3.3 (1.62–6.62) 79.6 (68.0–91.2) 248 (35) 228 (32) 83 (11) 73 (10) 40 (6) 40 (6) 7 (1)
Abbreviations: SD — standard deviation, IQR — interquartile range, tPA/PAI-1 complex — the complex between tissue plasminogen activator and plasminogen activator inhibitor-1, tPA — tissue plasminogen activator, CRP — C-reactive protein.
M. Lind et al. / Clinical Biochemistry 47 (2014) 570–573
the patients had a PK-INR value below 2.0, 5% had a PK-INR value above 3.5 and 86% were within the therapeutic range used at the time, PK-INR 2.0–3.5). Time in therapeutic range was 73.5% for the Skellefteå warfarin clinic and 67.3% for the Umeå warfarin clinic. Major bleeding A total of 73 patients experienced a major bleeding event during follow-up; the annual risk of major bleeding was 2.4% per year. The most common site of bleeding was gastrointestinal, while 17 intracranial and 11 fatal bleedings occurred. D-dimer showed a significant positive association with major bleeding events, and this association remained significant for the continuous variable after adjustment for age in the multivariate model (Table 2). Age was significantly associated with major bleeding. To further test the association of D-dimer with major bleeding, a Cox regression analysis on the Skellefteå cohort was performed. When adjusting for age, diabetes, hypertension and BMI in a multivariate model, D-dimer as a continuous variable remained significantly associated with major bleeding complications (HR 1.43; 95% CI: 1.03–1.98, per SD). CRP, creatinine, tPA, and tPA/PAI-1 complex, both as continuous variables and as tertiles, were tested in univariate Cox regression analysis and no significant association with major bleedings was found. Cardiovascular events During follow-up 47 myocardial infarctions, 42 ischemic strokes, 7 peripheral arterial embolism and 72 cardiovascular deaths comprised a total of 168 cardiovascular events. In univariate analysis CRP, creatinine, D-dimer, tPA, and tPA-PAI-1 complex all were associated with cardiovascular events. After adjustment in multivariate models only age and D-dimer continued to show a significant association with cardiovascular events; see Table 2. All-cause mortality A total of 161 patients died while on warfarin treatment. Univariate analysis demonstrated that CRP, age, D-dimer, and creatinine as continuous variables all were significantly associated with mortality and all these associations remained significant in the multivariate analysis. The age adjusted incidence of major bleeding, cardiovascular events and all-cause mortality for different tertiles of D-dimer is shown in Fig. 1.
8
Incidence per 100 treatment years
572
7 6
Major bleeding Cardiovascular events All-cause mortality
5 4 3 2 1 0
Low
Intermediate
High
Fig. 1. Age-adjusted incidence of major bleeding, cardiovascular event and all-cause mortality for tertiles of D-dimer.
Discussion The main finding of this longitudinal cohort study is the association of with major bleeding, cardiovascular events as well as all-cause mortality. Our findings regarding D-dimer strengthen the association between D-dimer and cardiovascular events and mortality shown in previous studies [8,13,16,7]. In our study, with 73 major bleeding events, we extend earlier findings of a possible association between high levels of D-dimer and major bleeding complications [13]. The mechanism for the dual association of D-dimer is not clear. A high D-dimer is an established biomarker of increased fibrinolysis; an underlying condition leading to increased fibrinolysis could increase the risks of both cardiovascular events and bleeding complications. It is also known that therapeutic treatment with oral anticoagulants decreases D-dimer levels and can even normalize them [17], while a fixed low dose warfarin does not [18,19]. It is possible that increased level of D-dimer in patients on oral anticoagulant treatment reflects inadequate treatment, or poor compliance leading to an increased risk of both bleedings and thromboembolic events. When evaluating D-dimer repeatedly during anticoagulant treatment, a rise in D-dimer precedes cardiovascular events [20], but it is not clear to what extent this might reflect an activation of vascular processes linked to thrombosis formation. It has been shown that levels of D-dimer seem to be dependent on calculated glomerular filtration D-dimer
Table 2 Univariate and multivariate Cox regression analysis of risk for cardiovascular events for continuous variables and according to tertiles of the biomarker concentrations. Results are presented as hazard ratios with 95% confidence intervals per increment in standard deviation for continuous variables. Major bleeding n = 73
Age, 10 year interval D-dimer, per 1 SD Lowest tertile Intermediate tertile Highest tertile tPA, per 1 SD Lowest tertile Intermediate tertile Highest tertile tPA/PAI-1 complex, per 1 SD Lowest tertile Intermediate tertile Highest tertile CRP, per 1 SD Creatinine, per 1 SD
Cardiovascular events n = 168
All-cause mortality n = 161
Univariate
Multivariate
Univariate
Multivariate
Univariate
Multivariate
1.50 (1.17–1.93) 1.33 (1.06–1.67) 1 1.32 (0.72–2.42) 1.83 (1.04–3.24) 0.97 (0.77–1.22) 1 0.81 (0.46–1.44) 0.94 (0.55–1.64) 1.05 (0.83–1.33) 1 1.03 (0.59–1.79) 0.97 (0.55–1.70) 1.02 (0.80–1.28) 0.90 (0.68–1.19)
1.43 (1.12–1.84) 1.27 (1.01–1.60) 1 1.29 (0.70–2.36) 1.67 (0.94–2.97) – – – – – – – – – –
1.55 (1.32–1.83) 1.34 (1.15–1.58) 1 1.30 (0.87–1.95) 1.83 (1.25–2.68) 1.18 (1.01–1.37) 1 1.34 (0.91–1.95) 1.30 (0.89–1.91) 1.22 (1.04–1.42) 1 1.54 (1.05–2.27) 1.58 (1.07–2.33) 1.20 (1.04–1.20) 1.24 (1.07–1.44)
1.54 (1.29–1.83) 1.23 (1.05–1.45) 1 1.20 (0.80–1.80) 1.51 (1.02–2.22) 1.08 (0.91–1.28) 1 1.19 (0.80–1.78) 1.09 (0.69–1.56) 1.12 (0.94–1.33) 1 1.20 (0.79–1.82) 1.29 (0.85–1.97) 1.15 (0.99–1.34) 1.15 (0.99–1.33)
1.76 (1.47–2.10) 1.38 (1.18–1.62) 1 1.41 (0.92–2.18) 2.29 (1.54–3.42) 1.12 (0.96–1.31) 1 1.02 (0.69–1.50) 1.19 (0.82–1.74) 1.37 (0.97–1.33) 1 1.39 (0.95–2.03) 1.22 (0.83–1.82) 1.37 (1.18–1.60) 1.23 (1.01–1.50)
1.66 (1.38–2.0) 1.25 (1.06–1.47) 1 1.23 (0.79–1.91) 1.72 (1.14–2.59) 1.01 (0.85–1.19) 1 0.85 (0.57–1.26) 0.87 (0.59–1.28) 1.01 (0.98–1.04) 1 1.00 (0.68–1.49) 0.93 (0.62–1.40) 1.23 (1.06–1.43) 1.20 (1.03–1.38)
Abbreviations: tPA/PAI-1 complex — the complex between tissue plasminogen activator and plasminogen activator inhibitor-1, tPA — tissue plasminogen activator, CRP — C-reactive protein.
M. Lind et al. / Clinical Biochemistry 47 (2014) 570–573
rate [21], but the ability of D-dimer to predict all-cause mortality was not affected by adjustment for age and creatinine in our study population. If D-dimer can help identify patients with an increased risk of cardiovascular events despite being on oral anticoagulant treatment measurement of D-dimer could contribute to a more individualized anticoagulant treatment. The complex between tPA and PAI-1 and tPA were not related to mortality, cardiovascular events or bleeding complications. These results are in contrast to earlier results linking tPA to stroke and cardiovascular events which to some extent could be attributed to the fact that previous studies have been smaller, on different populations, and without adjustment for potential confounders (12). There are limitations to this study. Blood sampling was made only at baseline on patients with long-term, ongoing warfarin treatment. The risk of bleeding complications is higher during initiation of warfarin treatment and it is possible that patients prone to bleeding have been filtered out before inclusion in our study leading to an underestimation of bleeding frequency. Fibrinolytic variables have been shown to vary with season [22], and fibrinolytic activity with time of day [23], which have not been adjusted for in our study. The major advantage of our study is the prospective cohort design with blood samples collected before the events. Furthermore, the study included a broad cross-section of the warfarin treated population irrespective of age and treatment indications. We conclude that high levels of D-dimer predict cardiovascular events and all-cause mortality during warfarin treatment. We also found an association between high levels of D-dimer and an increased risk of bleeding complications. Future studies should focus on the predictive value of D-dimer as a risk marker in relation to the established risk stratification schemes, especially in patients treated with the new oral anticoagulants. References [1] Mant J, Hobbs FD, Fletcher K, Roalfe A, Fitzmaurice D, Lip GY, et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet 2007;370:493–503. [2] Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the national registry of atrial fibrillation. JAMA 2001;285:2864–70. [3] Schulman S, Beyth RJ, Kearon C, Levine MN. Hemorrhagic complications of anticoagulant and thrombolytic treatment: American college of chest physicians evidencebased clinical practice guidelines (8th edition). Chest 2008;133:257S–98S. [4] Hylek EM, Evans-Molina C, Shea C, Henault LE, Regan S. Major hemorrhage and tolerability of warfarin in the first year of therapy among elderly patients with atrial fibrillation. Circulation 2007;115:2689–96.
573
[5] Lind M, Boman K, Johansson L, Nilsson TK, Ohlin AK, Birgander LS, et al. Thrombomodulin as a marker for bleeding complications during warfarin treatment. Arch Intern Med 2009;169:1210–5. [6] Lind M, Boman K, Johansson L, Nilsson TK, Jarvholm LS, Jansson JH. Von Willebrand factor predicts major bleeding and mortality during oral anticoagulant treatment. J Intern Med Mar 2012;271(3):239–46. [7] Wennberg P, Wensley F, Di Angelantonio E, Johansson L, Boman K, Rumley A, et al. Haemostatic and inflammatory markers are independently associated with myocardial infarction in men and women. Thromb Res Jan 2012;129(1):68–73. [8] Vene N, Mavri A, Kosmelj K, Stegnar M. High d-dimer levels predict cardiovascular events in patients with chronic atrial fibrillation during oral anticoagulant therapy. Thromb Haemost 2003;90:1163–72. [9] Brannstrom M, Jansson JH, Boman K, Nilsson TK. Endothelial haemostatic factors may be associated with mortality in patients on long-term anticoagulant treatment. Thromb Haemost 1995;74:612–5. [10] Aoki N, Saito H, Kamiya T, Koie K, Sakata Y, Kobakura M. Congenital deficiency of alpha 2-plasmin inhibitor associated with severe hemorrhagic tendency. J Clin Invest 1979;63:877–84. [11] Schleef RR, Higgins DL, Pillemer E, Levitt LJ. Bleeding diathesis due to decreased functional activity of type 1 plasminogen activator inhibitor. J Clin Invest 1989;83:1747–52. [12] Smith A, Patterson C, Yarnell J, Rumley A, Ben-Shlomo Y, Lowe G. Which hemostatic markers add to the predictive value of conventional risk factors for coronary heart disease and ischemic stroke? The Caerphilly study. Circulation 2005;112:3080–7. [13] Sadanaga T, Sadanaga M, Ogawa S. Evidence that d-dimer levels predict subsequent thromboembolic and cardiovascular events in patients with atrial fibrillation during oral anticoagulant therapy. J Am Coll Cardiol May 18 2010;55(20):2225–31. [14] Roldan V, Marin F, Muina B, Torregrosa JM, Hernandez-Romero D, Valdes M, et al. Plasma Von Willebrand factor levels are an independent risk factor for adverse events including mortality and major bleeding in anticoagulated atrial fibrillation patients. J Am Coll Cardiol Jun 21 2011;57(25):2496–504. [15] Schulman S, Kearon C. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005;3:692–4. [16] Folsom AR, Delaney JA, Lutsey PL, Zakai NA, Jenny NS, Polak JF, et al. Associations of factor viiic, d-dimer, and plasmin-antiplasmin with incident cardiovascular disease and all-cause mortality. Am J Hematol 2009;84:349–53. [17] Lip GY, Lowe GD, Rumley A, Dunn FG. Increased markers of thrombogenesis in chronic atrial fibrillation: effects of warfarin treatment. Br Heart J 1995;73:527–33. [18] Li-Saw-Hee FL, Blann AD, Lip GY. Effects of fixed low-dose warfarin, aspirin-warfarin combination therapy, and dose-adjusted warfarin on thrombogenesis in chronic atrial fibrillation. Stroke 2000;31:828–33. [19] Koefoed BG, Feddersen C, Gullov AL, Petersen P. Effect of fixed minidose warfarin, conventional dose warfarin and aspirin on INR and prothrombin fragment 1 + 2 in patients with atrial fibrillation. Thromb Haemost 1997;77:845–8. [20] Mahe I, Bergmann JF, Chassany O, Dit-Sollier CB, Simoneau G, Drouet L. A multicentric prospective study in usual care: D-dimer and cardiovascular events in patients with atrial fibrillation. Thromb Res Jun 2012;129(6):693–9. [21] Tanaka H, Sonoda M, Kashima K, Tanaka Y, Nakamura K, Nuruki N, et al. Impact of decreased renal function on coagulation and fibrinolysis in patients with nonvalvular atrial fibrillation. Circ J 2009;73:846–50. [22] Hypponen E, Berry D, Cortina-Borja M, Power C. 25-hydroxyvitamin d and pre-clinical alterations in inflammatory and hemostatic markers: A cross sectional analysis in the 1958 british birth cohort. PLoS One May 24 2010;5(5):e10801. [23] Fearnley GR, Balmforth G, Fearnley E. Evidence of a diurnal fibrinolytic rhythm; with a simple method of measuring natural fibrinolysis. Clin Sci (Lond) 1957;16:645–50.
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
D-dimer
predicts major bleeding, cardiovascular events and all-cause mortality during warfarin treatment Marcus Lind a,⁎, Kurt Boman a, Lars Johansson a, Torbjörn K. Nilsson b, Lisbeth Slunga Järvholm a, Jan-Håkan Jansson a a b
Department of Public Health and Clinical Medicine, Umeå University, Umeå, Sweden Department of Clinical Chemistry, Örebro University Hospital, Örebro, Sweden
a r t i c l e
i n f o
Article history: Received 15 October 2013 Received in revised form 3 February 2014 Accepted 9 March 2014 Available online 15 March 2014 Keywords: D-dimer Bleeding Warfarin Mortality PAI-1 Cardiovascular events
a b s t r a c t Objectives: Previous studies have shown that biomarkers in blood plasma can predict bleeding complications during anticoagulant treatment as well as thromboembolic events and may improve existing risk stratification schemes in patients on or considered for oral anticoagulant treatment. The aim of this study was to investigate if levels of D-dimer, tissue plasminogen activator (tPA) and its complex with plasminogen inhibitor type 1 (tPA/PAI-1 complex) can predict major bleedings, cardiovascular events and all-cause mortality in patients with warfarin treatment. Design and methods: In a longitudinal cohort study, 719 patients on oral anticoagulant treatment were followed for a total of 3001 treatment years. Major bleeding, stroke, arterial emboli, myocardial infarction and death were recorded and classified. Blood samples collected at baseline were analyzed for D-dimer, tPA, and tPA/PAI-1 complex. Results: In multivariate Cox regression analysis, high levels of D-dimer were associated with major bleeding (HR 1.27 per SD; 95% CI: 1.01–1.60), cardiovascular events (HR 1.23 per SD; 95% CI: 1.05–1.45) and all-cause mortality (HR 1.25 per SD; 95% CI: 1.06–1.47). Neither tPA nor the tPA/PAI-1 complex was associated with major bleeding, cardiovascular events or all-cause mortality. Conclusion: We conclude that high levels of D-dimer predict major bleeding, cardiovascular events and all-cause mortality during warfarin treatment. © 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction Treatment with an oral anticoagulant is effective in preventing thromboembolic events but carries an increased risk of bleeding complications [1]. To maximize the benefit of anticoagulant treatment in atrial fibrillation several risk scores like the CHADS2 [2] have been developed to help predict which patients have the highest risk of future thromboembolic events. Unfortunately risk factors for thromboembolic events are also associated with a higher risk of serious bleedings [3] and higher CHADS2 scores are associated with an increased risk of bleeding complications [4]. We have previously shown that biomarkers in blood plasma can predict bleeding complications during anticoagulant treatment [5] as well as thromboembolic complications [6]. It is possible that biomarkers could improve existing risk stratification schemes in patients on or considered for oral anticoagulant treatment.
⁎ Corresponding author at: Department of Medicine, Skellefteå County Hospital, 931 86 Skellefteå, Sweden. Fax: +46 910771657. E-mail address: [email protected] (M. Lind).
From previous studies we know that tissue plasminogen activator (tPA) and the complex between tPA and its inhibitor plasminogen activator inhibitor (tPA/PAI-1 complex) are related to cardiovascular diseases [7,8]. In patients on warfarin treatment tPA has been associated with cardiovascular events [9]. Lack of fibrinolytic inhibitors causes bleeding symptoms of varying severity [10,11]. D-dimer, a marker of fibrin degradation, is also considered to be associated with cardiovascular disease [12,7,13]. Studies on warfarin treated patients have shown diverging results for the association between D-dimer and cardiovascular events. Two studies have shown a significant association between high levels of D-dimer and cardiovascular events in patients with atrial fibrillation [8,13] and also that high levels of D-dimer could be associated with bleeding complications [13]. However, one recent study failed to shown any association between D-dimer and mortality or cardiovascular events [14]. It is reasonable to assume that levels of fibrinolytic factors could be associated with risk of bleeding complications as well as thromboembolic events during anticoagulant treatment. The aim of this study was to investigate if levels of D-dimer, tPA and tPA/PAI-1 complex can predict major bleedings, cardiovascular events and all-cause mortality in patients with warfarin treatment.
http://dx.doi.org/10.1016/j.clinbiochem.2014.03.003 0009-9120/© 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
M. Lind et al. / Clinical Biochemistry 47 (2014) 570–573
Methods Patients The study population has been described earlier [5]. In short patients were simultaneously recruited from the warfarin clinics at Skellefteå County Hospital and at Umeå University Hospital in 1996. All patients with a planned treatment of more than 3 months were eligible for the study. Consent forms were sent to all 957 eligible patients, and 847 accepted to participate in the study. Before inclusion, 128 patients were excluded due to treatment stoppage, death or missing blood samples. In the end, 719 patients (356 from Skellefteå and 363 from Umeå warfarin clinics) were included in the study cohort. From the Skellefteå warfarin clinic data about diabetes, prior peptic ulcer, prior bleeding peptic ulcer, diabetes, weight and height were obtained through questionnaires. The study was approved by the Research Ethics Committee of Umeå University and all patients gave informed consent. Blood sampling and laboratory procedures All patients were on stable treatment with oral anticoagulants for at least 2 months prior to blood sampling. Venous blood samples were drawn with a minimum of stasis and collected in siliconized routine citrated plasma tubes containing 0.13 mol/L sodium citrate. After centrifugation the plasma samples were frozen and stored at − 70 °C until analyzed. Both study cohorts were analyzed at the same time and location. The laboratory staff had no knowledge of event status. tPA and tPA/PAI-1 complex antigen were determined with enzyme-linked immunosorbent assay (ELISA). The reagent kits, Imulyse tPA and TintElize tPA/PAI-1 were purchased from Biopool AB, (Umeå, Sweden). For tPA, coefficient of variation (CV) was 14.1% at 6.6 ng/mL in our hands. For tPA/PAI-1, CV was 10.7% at 2.4 ng/mL in our hands. D-dimer was measured with an Immuno-turbidimetric Assay from Diagnostica Stago, Inc. The interassay CV for D-dimer at 1.7 mg/mL was 5.0% in our hands. C-reactive protein (CRP) was determined with an automated CRP method, (IMMULITE Diagnostic Products Corporation, USA). The interassay co-efficient of variation was b6% in our hands. Creatinine was analyzed on a Hitachii 911 multianalyzer (Roche, Mannheim, Germany) with kits from Roche/Boehringer (Crea plus, enzymatic method). International normalized ratio (INR) was determined at each hospital laboratory. The laboratories used the same Owren method with the reagent SPA 50 (Diagnostica Stago, Inc.) and INR calibration. Follow-up study protocol The date of inclusion was set to the date of blood sampling with the earliest inclusion date being June 1st 1996. All patients were followed longitudinally until death, cessation of warfarin treatment or until January 1st 2002. In an effort to identify all bleeding and cardiovascular events causing hospital admission or death, all medical records from the departments of medicine, cardiology, surgery, ear nose and throat, ophthalmology, urology, neurology, neurosurgery, and orthopedic surgery were studied from June 1, 1996 up to January 1, 2002. One patient moved out of the region during the study period and was followed up to the date of migration. Major bleeding, myocardial infarction (MI), ischemic stroke (IS) and peripheral arterial emboli (AE) causing admission to hospital or death were recorded and classified by a panel of three researchers (ML, JHJ, LJ). Major bleedings were defined according to the International Society of Thrombosis and Haemostasis [15] as fatal bleeding and/or symptomatic bleeding in a critical area or organ and/or bleeding causing a fall in a hemoglobin level of 20 g L−1 or more or leading to transfusion of two or more units of whole blood. All other bleedings were classified as minor and were excluded. Cardiovascular events included all MI, IS and AE with and without fatal outcome. The cause of death was registered
571
and classified according to the death certificate and in all but one the cause of death could be classified. Ascertainment bias was avoided by investigators classifying events being blinded to the biochemical results. Statistical analysis Analyses of D-dimer, tPA and tPA/PAI-1 complex were performed as continuous variables and categorized into tertiles, with the lowest tertile as the reference group. Univariate Cox regression analysis was performed on each of the variables to estimate the hazard ratio (HR) and 95% confidence interval (CI). Multivariate Cox regression analysis was performed to estimate the effects on different determinants when controlling for other factors. Factors were excluded from the multivariate model if the univariate Cox regression resulted in a clearly non-significant HR defined as P N 0.2. Both creatinine and CRP were considered to be potential confounders and were included in the multivariate analysis. The distributions of tPA, tPA/PAI-1 complex, D-dimer, creatinine, and CRP were skewed and therefore transformed using the natural logarithm (ln). Data on hypertension, diabetes, BMI, and previous peptic ulcer were available from the Skellefteå warfarin clinic and were tested for association with major bleeding, cardiovascular events and all-cause mortality with Cox regression analysis. Direct age adjustment was performed in ten year intervals. A P-value b 0.05 (two-sided) was considered statistically significant. Individuals with missing values were excluded from the statistical analyses (32 for D-dimer and creatinine, 2 for hsCRP and none for tPA and tPA/PAI-1 complex). SPSS version 15.0 was used for all statistical analyses. Results The mean age of participants was 70 years and 37% were female. Patients were followed for a total of 3001 treatment years with a mean of 4.2 years and a maximum follow-up time of 5.6 years. The most common indication for warfarin treatment was prosthetic heart valve, followed by atrial fibrillation; see Table 1. INR values at the time of sampling were available from the Skellefteå warfarin clinic (9% of Table 1 Baseline characteristics of the study population. Study cohort (n = 719) Mean age at inclusion, years (SD) Female (%) Mean follow-up time, years (SD) D-dimer, μg/L, median (IQR) Lowest tertile Intermediate tertile Highest tertile tPA, ng/mL, median (IQR) Lowest tertile Intermediate tertile Highest tertile tPA/PAI-1 complex, ng/mL, median (IQR) Lowest tertile Intermediate tertile Highest tertile CRP, mg/L, median (IQR) Creatinine, μmol/L, median (IQR) Indications for warfarin treatment Prosthetic heart valve (%) Atrial fibrillation (%) Venous thromboembolism (%) Ischemic stroke (%) Peripheral arterial thromboembolism Miscellaneous (%) Not defined (%)
70 (11) 268 (37) 4.2 (1.8) 0.13 (0.07–0.24) 0–0.09 0.10–0.18 0.19–∞ 11.6 (8.6–14.8) 0–9.49 9.50–13.32 13.33–∞ 11.4 (5.3–11.2) 0–6.01 6.02–9.88 9.89–∞ 3.3 (1.62–6.62) 79.6 (68.0–91.2) 248 (35) 228 (32) 83 (11) 73 (10) 40 (6) 40 (6) 7 (1)
Abbreviations: SD — standard deviation, IQR — interquartile range, tPA/PAI-1 complex — the complex between tissue plasminogen activator and plasminogen activator inhibitor-1, tPA — tissue plasminogen activator, CRP — C-reactive protein.
M. Lind et al. / Clinical Biochemistry 47 (2014) 570–573
the patients had a PK-INR value below 2.0, 5% had a PK-INR value above 3.5 and 86% were within the therapeutic range used at the time, PK-INR 2.0–3.5). Time in therapeutic range was 73.5% for the Skellefteå warfarin clinic and 67.3% for the Umeå warfarin clinic. Major bleeding A total of 73 patients experienced a major bleeding event during follow-up; the annual risk of major bleeding was 2.4% per year. The most common site of bleeding was gastrointestinal, while 17 intracranial and 11 fatal bleedings occurred. D-dimer showed a significant positive association with major bleeding events, and this association remained significant for the continuous variable after adjustment for age in the multivariate model (Table 2). Age was significantly associated with major bleeding. To further test the association of D-dimer with major bleeding, a Cox regression analysis on the Skellefteå cohort was performed. When adjusting for age, diabetes, hypertension and BMI in a multivariate model, D-dimer as a continuous variable remained significantly associated with major bleeding complications (HR 1.43; 95% CI: 1.03–1.98, per SD). CRP, creatinine, tPA, and tPA/PAI-1 complex, both as continuous variables and as tertiles, were tested in univariate Cox regression analysis and no significant association with major bleedings was found. Cardiovascular events During follow-up 47 myocardial infarctions, 42 ischemic strokes, 7 peripheral arterial embolism and 72 cardiovascular deaths comprised a total of 168 cardiovascular events. In univariate analysis CRP, creatinine, D-dimer, tPA, and tPA-PAI-1 complex all were associated with cardiovascular events. After adjustment in multivariate models only age and D-dimer continued to show a significant association with cardiovascular events; see Table 2. All-cause mortality A total of 161 patients died while on warfarin treatment. Univariate analysis demonstrated that CRP, age, D-dimer, and creatinine as continuous variables all were significantly associated with mortality and all these associations remained significant in the multivariate analysis. The age adjusted incidence of major bleeding, cardiovascular events and all-cause mortality for different tertiles of D-dimer is shown in Fig. 1.
8
Incidence per 100 treatment years
572
7 6
Major bleeding Cardiovascular events All-cause mortality
5 4 3 2 1 0
Low
Intermediate
High
Fig. 1. Age-adjusted incidence of major bleeding, cardiovascular event and all-cause mortality for tertiles of D-dimer.
Discussion The main finding of this longitudinal cohort study is the association of with major bleeding, cardiovascular events as well as all-cause mortality. Our findings regarding D-dimer strengthen the association between D-dimer and cardiovascular events and mortality shown in previous studies [8,13,16,7]. In our study, with 73 major bleeding events, we extend earlier findings of a possible association between high levels of D-dimer and major bleeding complications [13]. The mechanism for the dual association of D-dimer is not clear. A high D-dimer is an established biomarker of increased fibrinolysis; an underlying condition leading to increased fibrinolysis could increase the risks of both cardiovascular events and bleeding complications. It is also known that therapeutic treatment with oral anticoagulants decreases D-dimer levels and can even normalize them [17], while a fixed low dose warfarin does not [18,19]. It is possible that increased level of D-dimer in patients on oral anticoagulant treatment reflects inadequate treatment, or poor compliance leading to an increased risk of both bleedings and thromboembolic events. When evaluating D-dimer repeatedly during anticoagulant treatment, a rise in D-dimer precedes cardiovascular events [20], but it is not clear to what extent this might reflect an activation of vascular processes linked to thrombosis formation. It has been shown that levels of D-dimer seem to be dependent on calculated glomerular filtration D-dimer
Table 2 Univariate and multivariate Cox regression analysis of risk for cardiovascular events for continuous variables and according to tertiles of the biomarker concentrations. Results are presented as hazard ratios with 95% confidence intervals per increment in standard deviation for continuous variables. Major bleeding n = 73
Age, 10 year interval D-dimer, per 1 SD Lowest tertile Intermediate tertile Highest tertile tPA, per 1 SD Lowest tertile Intermediate tertile Highest tertile tPA/PAI-1 complex, per 1 SD Lowest tertile Intermediate tertile Highest tertile CRP, per 1 SD Creatinine, per 1 SD
Cardiovascular events n = 168
All-cause mortality n = 161
Univariate
Multivariate
Univariate
Multivariate
Univariate
Multivariate
1.50 (1.17–1.93) 1.33 (1.06–1.67) 1 1.32 (0.72–2.42) 1.83 (1.04–3.24) 0.97 (0.77–1.22) 1 0.81 (0.46–1.44) 0.94 (0.55–1.64) 1.05 (0.83–1.33) 1 1.03 (0.59–1.79) 0.97 (0.55–1.70) 1.02 (0.80–1.28) 0.90 (0.68–1.19)
1.43 (1.12–1.84) 1.27 (1.01–1.60) 1 1.29 (0.70–2.36) 1.67 (0.94–2.97) – – – – – – – – – –
1.55 (1.32–1.83) 1.34 (1.15–1.58) 1 1.30 (0.87–1.95) 1.83 (1.25–2.68) 1.18 (1.01–1.37) 1 1.34 (0.91–1.95) 1.30 (0.89–1.91) 1.22 (1.04–1.42) 1 1.54 (1.05–2.27) 1.58 (1.07–2.33) 1.20 (1.04–1.20) 1.24 (1.07–1.44)
1.54 (1.29–1.83) 1.23 (1.05–1.45) 1 1.20 (0.80–1.80) 1.51 (1.02–2.22) 1.08 (0.91–1.28) 1 1.19 (0.80–1.78) 1.09 (0.69–1.56) 1.12 (0.94–1.33) 1 1.20 (0.79–1.82) 1.29 (0.85–1.97) 1.15 (0.99–1.34) 1.15 (0.99–1.33)
1.76 (1.47–2.10) 1.38 (1.18–1.62) 1 1.41 (0.92–2.18) 2.29 (1.54–3.42) 1.12 (0.96–1.31) 1 1.02 (0.69–1.50) 1.19 (0.82–1.74) 1.37 (0.97–1.33) 1 1.39 (0.95–2.03) 1.22 (0.83–1.82) 1.37 (1.18–1.60) 1.23 (1.01–1.50)
1.66 (1.38–2.0) 1.25 (1.06–1.47) 1 1.23 (0.79–1.91) 1.72 (1.14–2.59) 1.01 (0.85–1.19) 1 0.85 (0.57–1.26) 0.87 (0.59–1.28) 1.01 (0.98–1.04) 1 1.00 (0.68–1.49) 0.93 (0.62–1.40) 1.23 (1.06–1.43) 1.20 (1.03–1.38)
Abbreviations: tPA/PAI-1 complex — the complex between tissue plasminogen activator and plasminogen activator inhibitor-1, tPA — tissue plasminogen activator, CRP — C-reactive protein.
M. Lind et al. / Clinical Biochemistry 47 (2014) 570–573
rate [21], but the ability of D-dimer to predict all-cause mortality was not affected by adjustment for age and creatinine in our study population. If D-dimer can help identify patients with an increased risk of cardiovascular events despite being on oral anticoagulant treatment measurement of D-dimer could contribute to a more individualized anticoagulant treatment. The complex between tPA and PAI-1 and tPA were not related to mortality, cardiovascular events or bleeding complications. These results are in contrast to earlier results linking tPA to stroke and cardiovascular events which to some extent could be attributed to the fact that previous studies have been smaller, on different populations, and without adjustment for potential confounders (12). There are limitations to this study. Blood sampling was made only at baseline on patients with long-term, ongoing warfarin treatment. The risk of bleeding complications is higher during initiation of warfarin treatment and it is possible that patients prone to bleeding have been filtered out before inclusion in our study leading to an underestimation of bleeding frequency. Fibrinolytic variables have been shown to vary with season [22], and fibrinolytic activity with time of day [23], which have not been adjusted for in our study. The major advantage of our study is the prospective cohort design with blood samples collected before the events. Furthermore, the study included a broad cross-section of the warfarin treated population irrespective of age and treatment indications. We conclude that high levels of D-dimer predict cardiovascular events and all-cause mortality during warfarin treatment. We also found an association between high levels of D-dimer and an increased risk of bleeding complications. Future studies should focus on the predictive value of D-dimer as a risk marker in relation to the established risk stratification schemes, especially in patients treated with the new oral anticoagulants. References [1] Mant J, Hobbs FD, Fletcher K, Roalfe A, Fitzmaurice D, Lip GY, et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet 2007;370:493–503. [2] Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the national registry of atrial fibrillation. JAMA 2001;285:2864–70. [3] Schulman S, Beyth RJ, Kearon C, Levine MN. Hemorrhagic complications of anticoagulant and thrombolytic treatment: American college of chest physicians evidencebased clinical practice guidelines (8th edition). Chest 2008;133:257S–98S. [4] Hylek EM, Evans-Molina C, Shea C, Henault LE, Regan S. Major hemorrhage and tolerability of warfarin in the first year of therapy among elderly patients with atrial fibrillation. Circulation 2007;115:2689–96.
573
[5] Lind M, Boman K, Johansson L, Nilsson TK, Ohlin AK, Birgander LS, et al. Thrombomodulin as a marker for bleeding complications during warfarin treatment. Arch Intern Med 2009;169:1210–5. [6] Lind M, Boman K, Johansson L, Nilsson TK, Jarvholm LS, Jansson JH. Von Willebrand factor predicts major bleeding and mortality during oral anticoagulant treatment. J Intern Med Mar 2012;271(3):239–46. [7] Wennberg P, Wensley F, Di Angelantonio E, Johansson L, Boman K, Rumley A, et al. Haemostatic and inflammatory markers are independently associated with myocardial infarction in men and women. Thromb Res Jan 2012;129(1):68–73. [8] Vene N, Mavri A, Kosmelj K, Stegnar M. High d-dimer levels predict cardiovascular events in patients with chronic atrial fibrillation during oral anticoagulant therapy. Thromb Haemost 2003;90:1163–72. [9] Brannstrom M, Jansson JH, Boman K, Nilsson TK. Endothelial haemostatic factors may be associated with mortality in patients on long-term anticoagulant treatment. Thromb Haemost 1995;74:612–5. [10] Aoki N, Saito H, Kamiya T, Koie K, Sakata Y, Kobakura M. Congenital deficiency of alpha 2-plasmin inhibitor associated with severe hemorrhagic tendency. J Clin Invest 1979;63:877–84. [11] Schleef RR, Higgins DL, Pillemer E, Levitt LJ. Bleeding diathesis due to decreased functional activity of type 1 plasminogen activator inhibitor. J Clin Invest 1989;83:1747–52. [12] Smith A, Patterson C, Yarnell J, Rumley A, Ben-Shlomo Y, Lowe G. Which hemostatic markers add to the predictive value of conventional risk factors for coronary heart disease and ischemic stroke? The Caerphilly study. Circulation 2005;112:3080–7. [13] Sadanaga T, Sadanaga M, Ogawa S. Evidence that d-dimer levels predict subsequent thromboembolic and cardiovascular events in patients with atrial fibrillation during oral anticoagulant therapy. J Am Coll Cardiol May 18 2010;55(20):2225–31. [14] Roldan V, Marin F, Muina B, Torregrosa JM, Hernandez-Romero D, Valdes M, et al. Plasma Von Willebrand factor levels are an independent risk factor for adverse events including mortality and major bleeding in anticoagulated atrial fibrillation patients. J Am Coll Cardiol Jun 21 2011;57(25):2496–504. [15] Schulman S, Kearon C. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005;3:692–4. [16] Folsom AR, Delaney JA, Lutsey PL, Zakai NA, Jenny NS, Polak JF, et al. Associations of factor viiic, d-dimer, and plasmin-antiplasmin with incident cardiovascular disease and all-cause mortality. Am J Hematol 2009;84:349–53. [17] Lip GY, Lowe GD, Rumley A, Dunn FG. Increased markers of thrombogenesis in chronic atrial fibrillation: effects of warfarin treatment. Br Heart J 1995;73:527–33. [18] Li-Saw-Hee FL, Blann AD, Lip GY. Effects of fixed low-dose warfarin, aspirin-warfarin combination therapy, and dose-adjusted warfarin on thrombogenesis in chronic atrial fibrillation. Stroke 2000;31:828–33. [19] Koefoed BG, Feddersen C, Gullov AL, Petersen P. Effect of fixed minidose warfarin, conventional dose warfarin and aspirin on INR and prothrombin fragment 1 + 2 in patients with atrial fibrillation. Thromb Haemost 1997;77:845–8. [20] Mahe I, Bergmann JF, Chassany O, Dit-Sollier CB, Simoneau G, Drouet L. A multicentric prospective study in usual care: D-dimer and cardiovascular events in patients with atrial fibrillation. Thromb Res Jun 2012;129(6):693–9. [21] Tanaka H, Sonoda M, Kashima K, Tanaka Y, Nakamura K, Nuruki N, et al. Impact of decreased renal function on coagulation and fibrinolysis in patients with nonvalvular atrial fibrillation. Circ J 2009;73:846–50. [22] Hypponen E, Berry D, Cortina-Borja M, Power C. 25-hydroxyvitamin d and pre-clinical alterations in inflammatory and hemostatic markers: A cross sectional analysis in the 1958 british birth cohort. PLoS One May 24 2010;5(5):e10801. [23] Fearnley GR, Balmforth G, Fearnley E. Evidence of a diurnal fibrinolytic rhythm; with a simple method of measuring natural fibrinolysis. Clin Sci (Lond) 1957;16:645–50.
