Early Fibrinogen Degradation Coagulopathy Is Predictive of Parenchymal Hematomas in Cerebral rt-PA Thrombolysis
A Study of 157 Cases
Background— Little is known about the coagulation factors as predictors of cerebral bleeding in rt-PA thrombolysis. The aim of this study was to determine what early coagulation parameters could predict early hemorrhagic lesions.
Methods— Consecutive patients were included in the Lyon rt-PA protocol. Early hematomas (within 24 hours), diagnosed on an anatomoradiological basis (symptomatic and not symptomatic) were considered for the study. Fibrinogen and fibrin(ogen) degradation products (FDP) were assessed at entry and at 2 and 24 hours after the beginning of thrombolysis.
Results— Of 157 patients, 11 had early parenchymal hematomas (7%), 31 had early hemorrhagic infarcts (19.7%), and 115 had no bleeding (73.2%). In logistic regression, FDP at 2 hours was the single predictor of parenchymal hematomas (OR: 2.5; CI: 1.09 to 5.8), whereas an increase of FDP >200 mg/L multiplied the odds of parenchymal hematoma by 4.95 (IC: 1.09 to 22.4). Early parenchymal hematomas were indicative of a poor prognosis at 3 months (P=0.001).
Conclusions— Early parenchymal hematomas appear as both “malignant” and exclusively related to an explosive increase of FDP at 2 hours, ie, an early fibrinogen degradation coagulopathy (EFDC). All patients scheduled to rt-PA thrombolysis should have an assay of FDP 2 hours after the beginning of thrombolysis: patients with an established EFDC (FDP >200 mg/L) should be monitored specifically, with no antithrombotic drug during the first 72 hours. Patients with FDP >100 mg should share the same monitoring.
Intracranial hemorrhage after cerebral thrombolysis is the major complication of this type of treatment, especially with intravenous streptokinase,1 but also with intraarterial prourokinase2 and intravenous rt-PA.3 In the double-blind studies with intravenous rt-PA (NINDS, ECASS I and II, ATLANTIS),3–6 predictive factors of these bleedings included severity at entry,7,8 brain edema,7 early ischemic changes on CT,8–12 age, and aspirin.9 Phase IV studies mentioned the role of protocol violations13 and hyperglycemia.14 However, coagulation factors have not been taken into account in all these trials.
In 2000, after a prospective study of coagulation parameters in the Lyon intravenous rt-PA trial,15,16 we described a biological syndrome predictive of cerebral bleeding related to an early coagulopathy and characterized by an increase of fibrin(ogen) degradation products (FDP) 2 hours after the beginning of thrombolysis.17 In the present study of 157 patients, we show that this early coagulopathy is exclusively predictive of early parenchymal hematomas (<24 hours) and is the major and single factor of their occurrence.
Patients and Methods
Beginning January 1, 1994, we proposed an open protocol of intravenous rt-PA to all patients with acute internal carotid artery territory stroke, regardless of severity and early CT symptoms.15,16 Of the 360 variables of the files of the registry, 53 basic variables concerning clinical, etiological, radiological (baseline CT features), and biological data were selected for this study.
An anatomic definition of the hemorrhagic lesion was used, regardless of deterioration. CT scans were systematically performed at entry and at 24±6 hours. Parenchymal hematomas (PH) were defined according the anatomic definition of ECASS (PH1 and PH2)18 as hemorrhagic infarcts (HI; HI1 and HI2). Only early bleeding lesions, which occurred <24 hours after thrombolysis, were studied to better explore the relationship of the lesions with thrombolysis and the early biological events induced by thrombolysis. Only hematomas were considered in this report.
Coagulation tests were obtained before treatment, and at 2 and 24 hours after the initiation of thrombolysis in all patients. Fibrinogen levels were determined by the Clauss method on fresh plasma obtained from blood collected into 5-mL vacuum tubes containing 0.5 mL sodium citrate 0.129 mol/L (Vacutainer; Becton Dickinson). For FDP, 2 mL aliquots of blood were collected into vacuum tubes containing thrombin and soybean trypsin inhibitor (Becton Dickinson). When necessary, protamine sulfate (50 μL) was added to neutralize heparin. The blood was allowed to clot for 30 minutes at 37°C and the serum harvested by centrifugation. FDP were then measured in the serum by latex agglutination (SpliPrest; Diagnostica Stago). Baseline platelet count and international normalized ratio (INR) were also studied.
Three subgroups were studied: (1) early (<24 hours) PHs; (2) early (<24 hours) HIs; and (3) patients with no bleeding lesions. Each hemorrhagic subgroup was compared, using univariate analysis, to the nonbleeding subgroup (group 3) for the 53 variables. A logarithmic transformation was also used for FDP and fibrinogen values. The Mann–Whitney test and the Wilcoxon Rank sum test were used for quantitative data and the exact Fisher test was used for qualitative data. A logistic regression analysis was performed, with models for the different values of FDP and fibrinogen.
Multivariate analysis included logistic regression analysis by stepwise backward elimination method to select minimum sets of variables, which allowed a distinction between hematomas and the nonbleeding subgroup. Based on this regression model, the variables were selected and then used to analyze and describe the additional risk of early PH. SPSS Windows 95 (version 7.5) was used for calculations.
Data Concerning the Whole Cohort
Between January 1, 1994 and January 5, 2001, 157 patients were included in the Lyon rt-PA trial. The clinical results and the distribution of the cumulated hemorrhagic lesions are given in Table 1. Within 24 hours, 11 patients had early PHs (7%), 31 had early HIs (19.7%), and 115 had no bleeding (73.3%). The evolution of the values of fibrinogen and FDP at entry, 2 hours after thrombolysis, and at 24 hours are shown in Table 2. The decrease in fibrinogen (Figure 1) and the increase in FDP 2 hours after the beginning of thrombolysis were statistically significant (P<0.0001) (Table 2). Figure 2 shows the evolution of the FDP values at entry and at 2 hours (after logarithmic transformation) in the whole cohort. A significant correlation could be established between the increase in FDP and the decrease in fibrinogen between time 0 and 2 hours (Spearman Rho −0.28, P<0.002). The presence of FDP >100 mg/L at 2 hours was correlated with a poor prognosis (P=0.03).
Comparison of Early PHs With Nonbleeding Patients
In univariate analysis, no baseline clinical, causative, or radiological factor could be characterized as significant; glycemia, platelet count, or history of antiplatelet or anticoagulant treatments were not factors. The only significant factor was an increase of FDP at 2 hours after thrombolysis (median: 60 mg/L; 95% CI: 10 to 500 versus 20 mg/L; 95% CI: 10 to 30; P<0.04) (Table 3). Logistic regression demonstrated that FDP at 2 hours was a powerful independent predictor of early PHs (OR: 2.5; CI: 1.09 to 5.8; P=0.03), whereas an increase in FDP >200 mg/L multiplied the odds of PH by 4.95 (IC: 1.09 to 22.4; P=0.03). For values >100 mg/L, there was a statistical trend, with an OR of 3.04, that did not reach significance, given the size of the series. When the cutoff point was set at an FDP of 100 mg/L, the sensitivity was 36.4%, the specificity was 84.2%, the positive predictive value was 18.2%, and the negative predictive value was 93.2%. When the cutoff point was set at an FDP of 200 mg/L, the sensitivity was 27.3%, the specificity was 93%, the positive predictive value was 27.3%, and the negative predictive value was 93%.
Models of logistic regression with platelet count and glycemia confirmed that FDP at 2 hours was the single independent factor.
Descriptive Analysis of FDP and Fibrinogen Values in Patients With PH, HI, and No Bleeding
Figure 3 represents the box plots of the logarithmic values of FDP at 2 hours in the 3 subgroups, showing that the distribution of FDP values at 2 hours is very different in early hematomas, whereas early HIs and nonbleeding patients have similar patterns.
Prognosis at 3 Months
The occurrence of early PH was significantly related to a poor outcome at 3 months (mean modified Rankin score: 4.7 versus 2.6; P<0.0001) and to a poor outcome as soon as day 1 (P=0.005).
Early Coagulopathy With Increased FDP Values Is Predictive of Early PHs
In this single-center study concerning 157 patients undergoing a fibrinogenolysis exploration, we demonstrate that an early coagulation parameter, a considerable increase of FDP 2 hours after the beginning of thrombolysis, appears as the powerful and single predictive factor for early PHs; no other clinical or radiological feature was apparent. These hematomas, which represent 78.5% of the PHs observed within 7 days, appear to be exclusively linked to a coagulopathy induced by the thrombolytic process itself. Although the 200 mg/L value seems to be the definite statistically significant limit, the value of 100 mg/L represents a potential limit according to the statistical trends.
Such a coagulopathy, with early increase of FDP (at 5 hours), has been described in intravenous rt-PA thrombolysis of myocardial infarction (MI) and has been shown to be predictive of general bleeding in the TIMI Trial Phase I: values >100 mg/L were associated with a likelihood ratio for general bleeding of 2.25,19 whereas >400 mg/L multiplied the risk of intracranial hemorrhage by 319 and 4.1.20 In the first 3 trials of the European Cooperative Study Group for rt-PA,21 patients with an FDP value >85 mg/L at 90 minutes after the start of thrombolysis also had twice the risk of a general bleeding complication risk. Moreover, a specific cerebral bleeding hazard, in relation to increased FDP value determined at the time of the complication, was demonstrated in MI thrombolysis with duteplase.22 It has been shown that ICH in MI thrombolysis is basically an early complication, with a mean occurrence time within 24 hours.23,24
Thus, in cardiac thrombolysis, as in cerebral rt-PA thrombolysis, the levels of 85 to 100 mg/L appear to be the actual limit of risk for general or cerebral early bleeding.
We have found a relationship between the FDP increase and fibrinogen decrease, a possible indication that the degradation products assayed in our trial are primarily FDPs, as in MI rt-PA thrombolysis.25
Thus, the coagulopathy described here might involve, basically, an excessive fibrinogenolysis appearing early after thrombolysis, with release of fragments X and Y26 acting as antithrombins.26,27 In addition, FDPs inhibit fibrin polymerization, resulting in defective fibrin structure, and also impair platelet function.27 However, it remains to be explained why there is no significant concomitant decrease of plasma fibrinogen in this study as in the MI ones.19–21 Complex hemostasis interactions might occur, because rt-PA in acute ischemic stroke also provokes an early elevation of fibrin monomer and d-dimer.28 Other factors that might explain this excessive fibrinogenolysis could involve specific qualitative features of fibrinogen responsible for its explosive cleavage.
The early fibrinogen degradation coagulopathy might also be specific of the type of thrombolytic molecule: in the TIMI Trial Phase I, FDP were found significantly higher in patients treated with streptokinase than in those treated by rtPA,19 a fact that might contribute to the particularly high frequency of the PH rate in cerebral thrombolysis with streptokinase.1 Thus, early FDP values might then be an important parameter in security procedures of phase I and II studies of new thrombolytic molecules by which to identify a high risk for PHs. An early assay of FDP should also be obtained in animals when new thrombolytic molecules are tested.
However, only 40% of acute infarct patients with hematomas have FDP >100 mg/L, which indicates that other factors might play a role. We found no other factor, but methods like pretherapeutic stroke MRI might provide interesting data on the hemodynamic status of the ischemic zones that will provoke the hematoma. Another hypothesis is that despite FDP levels <100 mg/L, a qualitative coagulopathy does take place with high levels of fragments X and Y. This hypothesis should be studied.
Conversely, in patients with FDP >100 mg/L with no bleeding at all or only HI, factors that provide resistance to malignant bleeding may be hypothesized. Another hypothesis might be a special hemodynamic status or the lack of significant “qualitative coagulopathy” in these cases, with low levels of fragments X and Y.
Clinical Importance of Early High FDP for Prevention of Early Hematomas
Early PHs represent an immediate deterioration factor (54.5% were symptomatic) and a significant factor for a poor outcome at 3 months. This characteristic of “malignant” bleeding has been documented.18 The predictive value of the FDP increase represents a direct tool by which to detect the patients at risk for hematoma and organize its prevention. Given the data in MI thrombolysis and the statistical trends concerning the value of 100 mg/L, FDP at 2 hours >100 mg should be the relevant limit of the prehematoma early fibrinogen degradation coagulopathy.
Thus, in all stroke centers performing thrombolysis, intraarterially or intravenously, an FDP assay 2 hours after the beginning of thrombolysis should be standard practice to detect the patients with FDP >100 mg at risk for hematoma and who should be monitored with a specific attention and surveyed every hour.
Given the role of heparin (reflected by activated partial thromboplastin time values) in the occurrence of hematomas in MI thrombolysis29 and the particular frequency of hematomas in the PROACT II cerebral thrombolysis study,2 in which intravenous heparin was added to intraarterial prourokinase, it seems reasonable to avoid heparin in this type of patient, not only during the first 24 hours but also during the 48 subsequent hours. Other antithrombotic drugs, such as aspirin, low–molecular-weight heparin, or others, should also be avoided.
Immobilization and a particularly stringent surveillance of hypertension and glycemia should also be followed. Moreover, in cases with FDP >200 mg, the use antifibrinolytic agents such as tranexamic acid or other protease inhibitors could be considered. Further studies are necessary to test the safety of this possibility.
Based on this single-center study with homogenous clinical, radiological, and fibrinogen and FDP assessment in 157 patients, the predictive factors of cerebral bleeding in rt-PA thrombolysis have been clarified. Early PH appear as both “malignant” and exclusively related to an early fibrinogen degradation coagulopathy caused by rt-PA itself, which may be recognized early by the assay of FDP at 2 hours and indicate the need for specific monitoring of patients at risk (those with FDP >100 mg/L). Given the presence of a prehemorrhagic antithrombotic state in these patients, it seems logical to avoid new antithrombotic drugs for 72 hours.
This study was supported by a grant from Association pour la Recherche Neurovasculaire (ARNV).
- Received June 30, 2003.
- Revision received January 15, 2004.
- Accepted February 3, 2004.
Hacke W, Kaste, Fieschi C, Toni D, Lesaffre E, Von Kummer R, Boysen G, Bluhmki E, Höxter G, Mahagne MH, Hennerici M, for the ECASS Study Group. Intravenous thrombolysis with tissue plasminogen activator for acute hemispheric stroke. The European Cooperative Acute Stroke Study (ECASS). JAMA. 1995; 274: 1017–1025.
Hacke W, Kaste, Fieschi C, Von Kummer R, Davalos A, Meier D, Larrue V, Bluhmki E, Davis S, Donnan G, Schneider D, Diez-Tejedor E, Trouillas P, for the Second European-Australasian Acute Stroke Study Investigators. Randomised double-blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II). Lancet. 1998; 352: 1245–1250.
Clark WM, Wissman S, Albers GW, Jhamandas JH, Madden KP, Hamilton S. Recombinant tissue-type plasminogen activator (Alteplase) for ischemic stroke 3 to 5 hours after symptom onset. The ATLANTIS Study: a randomized controlled trial. Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke. JAMA. 1999 Dec 1; 282 (21): 2019–26.
The NINDS t-PA Stroke Study Group. Intracerebral hemorrhage after intravenous t-PA for ischemic stroke. Stroke. 1997; 28: 2109–2118.
Larrue V, Von Kummer R, Del Zoppo G, Bluhmki E. Hemorrhagic transformation in acute ischemic stroke. Potential contributing factors in the European Cooperative Acute Stroke Study. Stroke. 1997; 28: 586–590.
Larrue V, Von Kummer R, Müller A, Bluhmki E. Risk factors for severe hemorrhagic transformation in ischemic stroke patients treated with recombinant tissue plasminogen activator. A secondary analysis of the European-Australasian Acute Stroke Study (ECASS II). Stroke. 2001; 32: 438–441.
Berger C, Fiorelli M, Steiner T, Schäbitz WR, Bozzao L, Bluhmki E, Hacke, Von Kummer R. Hemorrhagic transformation of ischemic brain tissue. Asymptomatic or Symptomatic? Stroke. 2001; 32: 1330–1335.
Kase CS, Furlan AJ, Wechsler LR, Higashida RT, Rowley HA, Hart RG, Molinari GF, Frederick LS, Roberts HC, Gebel JM, Sila CA, Schulz GA, Roberts RS, Gent M, and the PROACT II investigators. Cerebral hemorrhage after intra-arterial thrombolysis for ischemic stroke: the PROACT II trial. Neurology. 2001; 57: 1603–1610.
Jaillard A, Cornu C, Durieux A, Moulin T, Boutitie F, Kennedy R, Hommel M. Hemorrhagic transformation in acute ischemic stroke. The MAST-E study. Stroke. 1999; 30: 1326–1332.
Lopez-Yunez AM, Bruno A, Williams LS, Yilmaz E, Zurru C, Biller J. Protocol violations in community-based stroke treatment are associated with symptomatic intracerebral hemorrhage. Stroke. 2001; 32: 12–16.
Demchuk AM, Morgenstern LB, Krieger DW, Chi LT, Hu W, Wein TH, Hardy RJ, Grotta JC, Buchan AM. Serum glucose level and diabetes predict tissue plasminogen activator-related intracerebral hemorrhage in acute ischemic stroke. Stroke. 1999; 30: 34–39.
Trouillas P, Nighoghossian N, Getenet JC, Riche G, Honnoral J, Neuschwander P, Froment JC, Turjman F, Jin JX, Malicier D, Fournier G, Gabry AL, Ledoux X, Derex L, Berthezène Y, Adeleine P, Xie J, Ffrench P, Dechavanne M. Open trial of intravenous tissue plasminogen activator in acute carotid territory stroke. Correlation of outcome with clinical and radiological data. Stroke. 1996; 27: 889–890.
Trouillas P, Nighoghossian N, Derex L, Adeleine P, Honnoral J, Neuschwander P, Riche G, Getenet JC, Weil L, Froment JC, Turjman F, Malicier D, Fournier G, Gabry AL, Ledoux X, Berthezène Y, Ffrench P, Dechavanne M. Thrombolysis with intravenous rtPA in a series of 100 cases of acute carotid territory stroke. Stroke. 1998; 29: 2529–2540.
Trouillas P, Philippeau F, Nighoghossian N, Derex L, Honnorat J, Dechavanne MA. Biological syndrome predictive of cerebral bleeding in rtPA thrombolysis (data from the Lyon Thrombolysis Registry) (9th European Stroke Conference, Vienna, 24–27 May 2000). Cerebrovasc Dis. 2000; 10: 71.
Fiorelli M, Bastianello S, Von Kummer R, Del Zoppo G, Larrue V, Lesaffre E, Ringleb AP, Lorenzano S, Manelfe C, Bozzao L, for the ECASS I Study Group. Hemorrhagic transformation within 36 hours of a cerebral infarct. Relationship with early clinical deterioration and 3-month outcome in the European Cooperative Acute Stroke Study I (ECASS I) cohort. Stroke. 1999; 30: 2280–2284.
Rao AK, Pratt C, Breke A, Jaffe A, Ockene I, Schreiber TL, Bell WR, Knatterud G, Robertson TL, Terrin ML, for the TIMI investigators. Thrombolysis In Myocardial Infarction (TIMI) trial-Phase I. hemorrhagic manifestations and changes in plasma fibrinogen and the fibrinolytic system in patients treated with recombinant tissue plasminogen activator and streptokinase. J Am Coll Cardiol. 1988; 11: 1–11.
Arnold AE, Brower RW, Collen D, Van Es GA, Lubsen J, Serruys PW, Simoons ML, Verstraete M, for the European Cooperative Study Group for rt-PA. Increased serum levels of fibrinogen degradation products due to treatment with recombinant tissue-type plasminogen activator for acute myocardial infarction are related to bleeding complications, but not to coronary patency. J Am Coll Cardiol. 1989; 14: 581–588.
Kase CK, Pessin MS, Zivin JA, Del Zoppo GJ, Furlan AJ, Buckley JW, Snipes RJ, Littlejohn JK. Intracranial hemorrhage after coronary thrombolysis with tissue plasminogen activator. JAMA. 1992; 92: 384–390.
Gore JM, Sloan M, Price TR, Randall AMY, Bovill E, Collen D, Forman S, Knatterud GL, Soppko G, Terrin ML, and the TIMI Investigators. Intracerebral hemorrhage, cerebral infarction, and subdural hematoma after acute myocardial infarction and thrombolytic therapy in the thrombolysis in myocardial infarction study. Thrombolysis in Myocardial Infarction, phase II pilot and clinical trial. Circulation. 21991; 83: 448–459.
Gebel JM, Sila CA, Sloan MA, Granger CB, Mahaffey KW, Weisenberger J, Green CL, White HD, Gore JM, Weaver WD, Califf RM, Topol EJ, for the GUSTO-1 Investigators. Thrombolysis-related intracranial hemorrhage. A radiographic analysis of 244 cases from the GUSTO-1 trial with clinical correlation. Stroke. 1998; 29: 563–569.
Declerck P, Mombaerts P, Holvoet P, Collen D. Plasma levels of fragment D-dimer of cross-linked fibrin during thrombolytic therapy with recombinant tissue-type plasminogen activator. Thromb Hemost. 1987: 58; 231. Abstract.
Collen D, Bounameaux H, De Cock F, Lijnen HR, Verstraete M. Analysis of coagulation and fibrinolysis during intravenous infusion of recombinant human tissue-type plasminogen activator in patients with acute myocardial infarction. Circulation. 1986; 73: 511–517.
Wintrobe MM, Lee GR, Bogs DR, et al. Clinical Hematology. 8th ed. Philadelphia, Pa: Lea and Febiger; 1981: 434.
Fassbender k, Dempfle CE, Mielke O, Schwartz A, Daffertshofer M, Eschenfelder C, Dollman M, Hennerici M. Changes in coagulation and fibrinolysis markers in acute ischemic stroke treated with recombinant plasminogen activator. Stroke. 1999; 30: 2101–2104.
Granger CB, Hirsch J, Califf RM, Col J, White HD, Bertriu A, Woodlief LH, Lee KL, Bovill EG, Simes RJ, Topol EJ, for the GUSTO-I investigators. Activated partial thromboplastin time and outcome after thrombolytic therapy for acute myocardial infarct. Results from the GUSTO-I trial. Circulation. 1996; 93: 870–878.