Fibrinogen as a predictor of intracranial hemorrhage after intravenous thrombolysis in ischemic stroke

Overview

Background and objective: Although intravenous thrombolysis (IVT) is a recommended recanalization therapy for ischemic stroke, it is associated with a risk of symptomatic intracranial hemorrhage. This academic clinical study aims to evaluate the relationship between fibrinogen levels and their drop measured at 6 and 24 h after IVT and the risk of intracranial hemorrhage, specifically parenchymal hematoma, which is the most clinically significant type of intracranial hemorrhage. Methods: Seven primary stroke centers and comprehensive stroke centers in the Czech Republic will be involved in the multicenter retrospective study. The estimated total number of patients enrolled in the observation and control groups is 300, with defined entry parameters maintained. The parameters monitored include personal and pharmacological history, laboratory parameters, neurological findings, and imaging results. Results: The study will evaluate the degree of association between fibrinogen levels and their drop measured at 6 and 24 h after IVT and the risk of parenchymal hematoma. Conclusion: Evidence of an association between the development of parenchymal hematoma and a decrease in fibrinogen levels in patients with ischemic stroke treated with IVT may be used to design an interventional study with fibrinogen replacement to prevent bleeding.

Keywords:

stroke – ischemia – intravenous thrombolysis – intracranial hemorrhage – fi brinogen – parenchymal hematoma

This is an unauthorised machine translation into English made using the DeepL Translate Pro translator. The editors do not guarantee that the content of the article corresponds fully to the original language version

Introduction

Intravenous thrombolysis (IVT) is the recommended recanalization therapy for ischemic stroke (iCMP). The benefit of IVT treatment for the patient is clear. The number needed to treat (NNT) to achieve a score of 0–1 on the modified Rankin Scale (mRS) of 0–1 (i.e., functional independence) varies for the 0–3-hour treatment window (according to various publications, the NNT ranges from 8 to 10) and for the 3–4.5-hour treatment window (NNT 12–19) [1,2], and therefore efforts are made to initiate IVT within the shortest possible time frame to increase the likelihood of recanalization and reduce the risk of intracranial hemorrhage, permanent sequelae, and mortality. Every 15-minute delay reduces the chance of achieving an mRS score of 0–1 by 16% [3]. Although the benefits of this treatment are clear, one of its serious complications is the development of symptomatic intracranial hemorrhage, with an incidence of 6–7% [4] worldwide and approximately 3–12% in the Czech Republic (Table 1).

Randomized clinical trials have shown that the recombinant tissue plasminogen activator –⁠ alteplase (Actilyse^®, Boehringer Ingelheim International GmbH, Ingelheim am Rhein, Germany) –⁠ improves the likelihood of achieving a good clinical outcome when administered within 4.5 hours of the onset of iCMP symptoms. At the same time, however, relevant data are available demonstrating a benefit within the 4.5–9-hour time window, as well as in cases with an unknown onset time, in patients selected based on multimodal CT examination [5]. However, alteplase also increases the risk of developing intracranial hemorrhage [3]. The severity of bleeding varies considerably from case to case, and its clinical manifestation can influence the patient’s long-term clinical outcome (assessed using the mRS scale). In recent years, an increasing number of publications have pointed to a lack of evidence regarding the risks of intracranial hemorrhage following alteplase administration in various patient groups, particularly those with stroke symptom onset exceeding 4.5 hours [6,7].

Intracranial hemorrhage can be divided into intracerebral hemorrhage (ICH) and bleeding outside the brain tissue (subarachnoid hemorrhage, intraventricular hemorrhage, subdural hemorrhage). ICH is further classified according to the European Cooperative Acute Stroke Study (ECASS) into two main categories—hemorrhagic infarction and parenchymal hematoma (PH) [8]; in addition, ICH localized outside the area of ischemia may occur. A meta-analysis of randomized clinical trials evaluating the prevalence of ICH following IVT compared to a control cohort (placebo or no treatment) includes data from 6,756 patients. Of this number, type 2 PH (a hematoma affecting >30% of the ischemic zone with marked expansive behavior) was diagnosed in 231 (6.8%) of the 3,391 patients treated with alteplase, compared with the control group, where type 2 PH occurred in 44 (1.3%) of 3,365 subjects. This corresponds to an almost sixfold increased risk of this bleeding (odds ratio [OR] = 5.55; 95% confidence interval [CI] 4.01–7.70). Among treated patients, this type of hemorrhage was associated with a sudden clinical deterioration of at least 4 points on the National Institutes of Health Stroke Scale (NIHSS) in 53% of cases. Fatal ICH occurred in 91 (2.7%) vs. 13 (0.4%) patients in the control group [9].

The risk of intracranial hemorrhage is predicted by several factors, such as advanced age, uncontrolled or chronic arterial hypertension, heart disease, hyperglycemia, low/high body weight, previous stroke, a longer time window for performing IVT, or a higher dose of thrombolytic agent. However, according to some publications, the still-debated variability in blood pressure following IVT in acute stroke does not predict the risk of intracranial hemorrhage, but rather the patient’s poor clinical outcome [10].

Following thrombolytic therapy with alteplase, fibrinogen levels decrease [11]. During this decrease, bleeding complications may occur as a result of systemic fibrinolysis. Low fibrinogen levels, as one of the potential risk factors for intracranial hemorrhage, are thus relatively easy to modulate therapeutically compared to other factors. If fibrinogen levels fall below the reference range, fibrinogen can be replaced with Haemocomplettan (CSL Behring GmbH, Marburg, Germany) or Fibryga (Octapharma, Anderlecht, Belgium), which are intended for intravenous use. Another treatment option is the administration of fresh frozen plasma (FFP) [12]. Fibrinogen replacement is available in hospitals but is associated with higher financial costs.

Although the relationship between IVT and the risk of developing intracranial hemorrhage has long been known, there is a lack of sufficiently robust data to clearly demonstrate whether and to what extent a decrease in fibrinogen levels following alteplase administration contributes to the development of clinically significant intracranial hemorrhage. Current guidelines from the Cerebrovascular Section of the Czech Neurological Society of the Czech Medical Association (ČLS JEP) for the treatment of IVT include blood sampling with testing of coagulation parameters, including fibrinogen levels, prior to IVT and 6 and 24 hours after IVT, as well as a follow-up brain CT scan 22–36 hours after IVT [5] unless the patient’s clinical condition requires this examination on an individual (earlier) schedule.

The aim of the study is to evaluate the relationship between fibrinogen levels and their decline 6 and 24 hours after IVT and the risk of developing PH as the clinically most severe form of ICH.

 

Methodology

Study Design and Participating Centers

The following institutions will participate in the retrospective multicenter study: University Hospital Hradec Králové, St. Anne’s University Hospital in Brno, Brno University Hospital, Pardubice Hospital, General University Hospital in Prague, AGEL Hospital Ostrava-Vítkovice, and the Central Military Hospital –⁠ Military University Hospital Prague.

The study design was conceived as a comparative analysis of a group of patients with intracranial hemorrhagic complications resulting from IVT (observation group) and a control group, while maintaining the same or similar defined baseline parameters (age, baseline neurological deficit assessed using the NIHSS scale, comorbidities). To ensure homogeneity of the study population, centers with a similar percentage (i.e., 8–12%) of hemorrhagic complications in patients following IVT will be included in the study (Table 1).

 

Study Population

The study will include patients who have experienced an iCMP and were treated with IVT between 2019 and 2021, without the use of other recanalization therapies. These patients will be identified based on records in hospital information systems.

Inclusion criteria:

                iCMP treated with IVT;

    age between 18 and 90 years;

    NIHSS ≥ 2.

 

Exclusion criteria:

concomitant endovascular treatment of iCMP;

concomitant surgical treatment of iCMP;

    fibrinogen replacement with Haemocomplettan, Fibryga, or FFP between IVT and the follow-up CT scan.

 

Patient data will be retrospectively entered into the REDCap database system by trained study nurses from the STROCZECH network or other trained members of the study team. Database records will be maintained in accordance with the GDPR. All personal data that could be used to identify a patient will be replaced with an identification code (so-called pseudonymization).

 

Monitored Parameters

The following parameters will be monitored: coagulation parameters, information on CT scans, comorbidities, and treatment, which are collected and recorded in hospital records as part of standard procedures in accordance with current guidelines.

Anamnestic data from the admission report:

• age;
• gender;
• hypertension (assessed as “yes” if the diagnosis is listed in the medical history or if the patient’s chronic medication regimen includes at least one antihypertensive drug other than a diuretic);
• diabetes mellitus;
• antiplatelet therapy (specifically acetylsalicylic acid, clopidogrel, other);
• anticoagulant therapy (specifically warfarin, dabigatran, rivaroxaban, apixaban, edoxaban, unfractionated heparin, low-molecular-weight heparin at a prophylactic dose, low-molecular-weight heparin at a therapeutic dose).

 

Physical examination:

value of the first blood pressure measurement in the hospital (both systolic and diastolic);

severity of baseline neurological deficit prior to IVT, as measured by the NIHSS scale.

CT scan findings:

• Alberta Stroke Program Early CT Score (ASPECTS) prior to IVT;
• presence of intracranial hemorrhage on a scan performed within 36 hours after IVT (ICH is specified as “hemorrhagic infarction 1,” “hemorrhagic infarction 2,” “parenchymal hematoma 1,” or “parenchymal hematoma 2,” or “bleeding outside the ischemic area,” which also includes subarachnoid, intraventricular, or subdural hemorrhage).

 

Laboratory values:

fibrinogen at baseline, 6 hours after IVT, and 24 hours after IVT (if unavailable, indicate “not available”);

baseline International Normalized Ratio (INR) (a value between 0.8 and 1.2 may be classified as “normal”; otherwise, state the specific value);

baseline activated partial thromboplastin time (aPTT) (a value between 0.8 and 1.2 may be classified as “normal”; otherwise, specify the exact value);

    renal insufficiency –⁠ urea, creatinine, estimated glomerular filtration rate;

    platelet count –⁠ before IVT, first value after IVT.

 

Statistical analysis

Statistical analysis will be performed using R software (version 4.3.1, The R Foundation for Statistical Computing, Vienna, Austria).

Descriptive statistics will be reported as the median and interquartile range (IQR) for continuous variables and as frequencies and percentages for categorical variables. The normality of the data distribution will be assessed using the Shapiro-Wilk test and visually based on histograms and Q-Q plots.

For comparing groups (patients with vs. without intracranial hemorrhage, particularly PH), the following will be used:

            the Mann-Whitney U test for continuous variables with a non-parametric distribution,

    Student’s t-test in the case of an approximately normal distribution,

    cthe chi-square test or Fisher’s exact test for categorical variables.

 

Changes in fibrinogen levels between individual time points (baseline, 6 h, and 24 h after IVT) will be analyzed using the Friedman test for repeated measures, with post hoc comparisons adjusted by the Bonferroni method for multiple testing. A multivariate logistic regression model will be used to evaluate the relationship between the decrease in fibrinogen levels after IVT and the occurrence of intracranial hemorrhage, where the occurrence of PH will be the dependent variable (outcome). The model will include clinically relevant variables (age, sex, alteplase dose, baseline NIHSS score, baseline blood glucose, blood pressure, and fibrinogen levels) or those that were significant at the p < 0.10 level in univariate analysis.

Results will be presented as odds ratios (ORs) with 95% confidence intervals (95% CI). Statistical significance will be assessed at a p-value < 0.05 (two-sided).

The analysis will be conducted according to the intention-to-treat principle, with missing data handled using multiple imputation if they exceed 5% of the observed values.

 

Discussion

This study aims to demonstrate an association between a decrease in fibrinogen levels below the reference range and the development of clinically significant intracranial hemorrhage as a consequence of IVT therapy for iCMP. If this association is demonstrated, this finding would support an active therapeutic approach involving the use of replacement therapy to address low fibrinogen levels as a strategy to reduce the risk of hemorrhagic complications following IVT treatment. In such a case, it would be desirable to verify in a follow-up study whether fibrinogen replacement actually reduces the risk of developing intracranial hemorrhage.

If no clear association is found, it would be possible to reevaluate fibrinogen replacement as a therapeutic approach, as well as to modify the current clinical guidelines, which recommend monitoring fibrinogen levels at specific intervals. This could have clear clinical implications: an active approach aimed at fibrinogen replacement would be deemed ineffective, and monitoring of fibrinogen levels could be redefined in the current guidelines of the Cerebrovascular Section of the Czech Neurological Society of the Czech Medical Association of J. E. Purkyně as a parameter that does not need to be routinely monitored following IVT. This approach aligns with the recommendations of the American Heart Association/American Stroke Association (AHA/ASA), which do not recommend routine fibrinogen testing, but only in cases of symptomatic intracranial hemorrhage occurring within the first 24 hours after IVT treatment [13].

However, several limitations of this study should be emphasized. The retrospective design carries a risk of selection bias and heterogeneity among individual centers, particularly regarding the methodology used for laboratory determination of fibrinogen and the timing of blood draws. Another limitation may be incomplete data for some patients and the absence of a uniform definition of a clinically significant decrease in fibrinogen levels across the literature. These factors may, to some extent, affect the accuracy and interpretability of the results. A further limitation of our study is the fact that the follow-up CT scan was performed according to current recommendations 22–36 hours after IVT [5]. It is therefore not possible to determine with certainty whether intracranial hemorrhage had already developed at the time of fibrinogen sampling 6 hours after IVT or only subsequently. This uncertainty limits the ability to precisely establish the temporal relationship between the decrease in fibrinogen levels and the development of hemorrhage. From a clinical perspective, the moment when fibrinogen concentration falls below a critical level is of fundamental importance, as is the question of whether its timely replacement could prevent the development of hemorrhage. For the practical application of these findings, the correct timing of both the sample collection and any potential intervention is therefore crucial. In a previous study of patients treated with alteplase for myocardial infarction, fibrinogen reached its lowest levels between 90 minutes and 3 hours after IVT [11]. This interval could represent the optimal time window for targeted measurement of fibrinogen levels and for its potential replacement.

Future research should aim for a prospective multicenter study that would confirm or refute the predictive role of fibrinogen levels and their temporal evolution in the context of IVT treatment. The goal of such a study should therefore be to closely monitor the dynamics of the decline in fibrinogen levels while simultaneously performing follow-up CT scans. Such a design would allow for a better understanding of the pathophysiological relationship between the decline in fibrinogen levels and the development of PH following IVT, while simultaneously providing a basis for testing the safety and efficacy of early replacement therapy. At the same time, it would be advisable to expand the range of monitored hemostatic parameters, e.g., to include levels of D-dimers, fibrin degradation products, or plasminogen, and to monitor their dynamic changes over time. The data obtained in this way could contribute to the creation of a comprehensive predictive model for the risk of hemorrhagic complications following IVT and, subsequently, to the revision of current clinical recommendations.

 

Conclusion

This study aims to evaluate the relationship between fibrinogen levels and their decline following IVT and the risk of developing PH as the most clinically significant intracranial hemorrhage. The results may provide new insights into the mechanisms underlying PH, particularly regarding predictive factors and the temporal course of this complication, thereby enabling more accurate identification of patients at increased risk of hemorrhagic complications following IVT. The data obtained may also contribute to the validation of current guidelines or provide a basis for their optimization in light of current clinical and laboratory data.

Ethical Principles

This study is conducted in accordance with the 1975 Declaration of Helsinki and its revisions from 2004, 2008, and 2012. Given the retrospective nature of the study, patient consent to the processing of information regarding their health status will not be required. The study protocol was approved by the Multicenter Ethics Committee of the University Hospital in Hradec Králové on October 20, 2022, under reference number 202210 J02. The database will be maintained using anonymized patient identifiers in accordance with the GDPR.

Funding

Supported by Charles University (Cooperatio program, NEUR research area), the Ministry of Health of the Czech Republic –⁠ RVO (FNHK, 00179906), and the STROCZECH network within the CZECRIN research infrastructure (project no. LM2023049) funded by the state budget of the Czech Republic.

Acknowledgments

The authors would like to thank all colleagues and collaborators for their valuable advice and comments, which contributed not only to the development of the protocol but also to this entire academic clinical study. Special thanks go to the centers participating in this study.

Conflict of Interest

The authors declare that they have no conflict of interest in connection with the subject of this work.

 

Table 1. Number of patients treated with intravenous thrombolysis (without the use of other recanalization therapies) and incidence of intracranial hemorrhage in 2019–2021 at hospitals in the STROCZECH network.

Hospital Name	Number of IVT	Number of IVT cases with intracranial hemorrhage	Proportion of IVT with intracranial hemorrhage
Olomouc University Hospital	328	11	3.3%
Ostrava City Hospital	254	11	4.3%
Vyškov Hospital	161	7	4.3%
Písek Hospital	241	11	4.6%
Karviná Mining Hospital	215	10	4.7%
Ostrava University Hospital	270	13	4.8%
České Budějovice Hospital	628	32	5.1%
Královské Vinohrady University Hospital, Prague	252	14	5.6%
Thomayer University Hospital, Prague	139	8	5.8%
Na Homolce Hospital, Prague	219	13	5.9%
Liberec Regional Hospital	505	34	6.7%
Mladá Boleslav Regional Hospital –⁠ Klaudiánova Hospital	241	17	7.1%
Kladno Regional Hospital	250	18	7.2%
T. Bata Regional Hospital, Zlín	390	30	7.7%
Jihlava Hospital	278	24	8.6%
Příbram Regional Hospital	152	13	8.6%
AGEL Hospital Ostrava–Vítkovice	229	20	8.7%
Pardubice Hospital	257	23	8.9%
Motol University Hospital, Prague	378	35	9.3%
Brno University Hospital	301	29	9.6%
General University Hospital in Prague	246	24	9.8%
University Hospital Hradec Králové	215	22	10.2%
Central Military Hospital –⁠ Military University Hospital Prague	278	30	10.8%
St. Anne’s University Hospital in Brno	249	30	12.0%
Total	6,676	479	7.2%

IVT –⁠ intravenous thrombolysis