Introduction

Fibrin was discovered in 1666 by Marcello Malpighi (De Polypo Cordis), who compared cardiac thrombi with blood clots formed ex vivo using a light microscope, revealing their similar structure.1 Fibrin was named in 1788 by Antoine-François de Fourcroy, whereas the term “fibrinogen,” referring to the precursor of fibrin, was introduced by Rudolf Virchow in 1847.1 Due to the structural similarity between fibrinogen and fibrin, the informal term “fibrin(ogen)” was introduced to denote indistinguishable fibrinogen and / or fibrin, as both forms occur in vivo.2

Fibrinogen, with a molecular mass of approximately 340 kDa, is synthesized in the liver. It has a half-life of about 4 days, and its normal levels range between 2 and 4 g/l, making it the most abundant coagulation protein in circulating blood. In all vertebrates, fibrinogen has multiple functions, the most essential being related to hemostasis and involving prevention of blood loss through formation of a hemostatic plug. Fibrinogen interacts with cell surface receptors, modulating platelet aggregation and linking coagulation with inflammatory pathways. Additionally, it serves as a nonspecific indicator of inflammation and can bind histones, preventing their cytotoxicity.1,3

Elevated fibrinogen levels can contribute to a hypercoagulable state by increasing blood viscosity, red blood cell red blood cell and platelet aggregation, and platelet activity, reducing blood flow in vessels, and, last but not least, affecting fibrin structure and function.4

Fibrinogen contains multiple domains with specific binding sites involved in many interactions.5,6 The fibrinogen molecule is a dimer, with each subunit containing 3 polypeptide chains (Aα, Bβ, and γ), encoded by the FGA, FBG, and FGG genes, respectively, located within a 65 kb region on the chromosome 4, and linked by numerous disulfide bonds.5,6 The length of the fibrinogen molecule is approximately 45 nm. Due to its high flexibility, analyzing 3-dimensional structure of fibrinogen was challenging for decades.7 Proteolytic cleavage of fibrinogen results in the removal of 2 αC regions, 2 terminal D fragments (Figure 1A), and 1 central E fragment, derived from analogous regions.8 Fibrinopeptides A (FpA) and B (FpB) are N-terminal amino acids of the Aα and Bβ chains, respectively, located in the central E region. Cleavage of FpA by thrombin is critical for fibrin monomer (desA fibrin) and oligomer formation, whereas the release of FpB is related to the lateral association of protofibrils and the attraction of leukocytes to the site of injury.9 Activated factor (F) XIII provides covalent cross-linking, stabilizing fibrin structure.5

Figure 1. Fibrinogen and fibrin visualized using microscopic techniques; A – single fibrinogen molecules visualized using atomic forced microscopy (scale bar, 200 nm); fibrinogen D domains, larger than the central E domain, are indicated by the arrows. B – typical network of dehydrated fibrin fibers observed using scanning electron microscopy (SEM; scale bar, 2 µm); C – intracoronary thrombus containing fibrin, erythrocytes, and platelets (SEM image; scale bar, 20 µm); D – a cross-section of the saphenous vein graft after massive thrombosis (nuclei stained blue and fibrin stained green; scale bar, 1 mm). All microphotographs represent the author’s own work and have not been published elsewhere.

Fibrin formation stimulates the conversion of plasminogen to plasmin by tissue plasminogen activator (tPA). tPA is a serine protease synthesized by endothelial cells. It has a half-life of 3 minutes, and is converted to its active form by plasmin.10 Plasmin cleaves fibrin at specific lysine / arginine sites and exposes additional binding sites for plasminogen and tPA. Plasminogen activation by urokinase plasminogen activator (uPA) is fibrin-independent but less efficient.5 uPA, also known as urokinase, was initially purified from human urine. It is secreted by renal epithelial cells and has a longer half-life (up to 16 minutes) than tPA.10 Recombinant tPA is considered the most reliable agent for intravenous thrombolysis (alteplase); however, recent data suggest that recombinant human prourokinase exerts similar effects to alteplase but demonstrates a safer bleeding profile in patients with acute ischemic stroke.11 Plasminogen activator inhibitor type 1 (PAI-1) binds to tPA, thereby limiting plasmin generation.10 Type 2 plasminogen activator inhibitor (PAI-2) primarily binds to uPA.10 Thrombin-activatable fibrinolysis inhibitor (TAFI), an antifibrinolytic protein activated by thrombin in a complex with thrombomodulin, removes lysine residues from the fibrin surface, reducing generation of plasmin.10

Fibrinogen is especially prone to post-translational modifications that affect its function, such as glycosylation (associated with aging), oxidation, and nitration (as a result of oxidative stress).7 Nonenzymatic fibrinogen glycation (associated with prolonged high blood glucose) and acetylation (in the context of aspirin use) at specific lysine residues have been observed in vitro in a concentration-dependent manner.12 Homocysteinylation or citrullination have also been shown to affect the fibrinogen molecule, influencing fibrin formation, with further implications for its architecture and stability.7

A lesson from congenital fibrinogen disorders

Afibrinogenemia is an extremely rare coagulation disorder, with scarce data estimating its incidence at 1–2 cases per 1 million in the general population.13 Afibrinogenemia is inherited in an autosomal recessive manner and affects one of the 3 genes encoding the fibrinogen chains.13 In afibrinogenemia, the level of immunoreactive fibrinogen is similar to that of functional fibrinogen, and in most patients, fibrinogen concentrations in plasma are below 0.1 g/l.13 The most frequent hemorrhages occurring in patients with afibrinogenemia are hemarthroses, muscle hematomas, gastrointestinal bleedings, epistaxis, and menorrhagia.13 However, the reported mean annual incidence of bleeding on replacement therapy is about 0.5–0.7, which is relatively low, as compared with hemophilia.13 Interestingly, thrombosis, such as ischemic lesions of the limbs related to severe stenosis of arteries, can occur in patients with fibrinogen deficiency regardless of replacement therapy.13 Such effect results from the lack of fibrin activity to neutralize thrombin (known as antithrombin I activity) and interaction of thrombin with platelets and smooth muscle cells.14 Therefore, thrombin generation potential in afibrinogenemic (or fibrinogen-depleted) plasma is increased.15 Of note, the influence of plasma dilution on thrombin generation and fibrin formation yielded unexpected results, because clots were susceptible to degradation by plasmin, but the fibrin networks presented a denser and less permeable structure.16

The gold standard to assess fibrinogen activity remains the commonly used von Clauss assay, in which very high concentration of bovine thrombin is used for the rapid conversion of fibrinogen to fibrin. The functional fibrinogen level is estimated based on the clotting time (Table 1). To distinguish quantitative and qualitative fibrinogen disorders, a ratio of functional fibrinogen and its antigen, measured by immunoassays, is used.17 Reduced functional fibrinogen levels at normal antigen concentrations are typical in patients with dysfibrinogenemia.17

Table 1. Plasma-based assays commonly used to evaluate fibrinogen and fibrin characteristics

Method / technique

Type

Approach

Detection

Measure

Advantages / limitations

Fibrinogen

von Clauss assay

Activity (functional assay)

Based on thrombin time

Mechanical or optical

Concentration

–

PT/APTT-derived

Activity (functional assay)

Based on PT or APTT

Optical

Concentration

Used on automated coagulometers

Clottable fibrinogen

Activity (functional assay)

Thrombin-activated clot formation and its dissolution in concentrated alkaline urea

Optical

Concentration

Time-consuming and requiring much experience

Precipitation

Total fibrinogen concentration

Sodium sulfite or heated precipitation and centrifugation

Nephelometry or refractometry

Concentration

–

Immunological assays

Fibrinogen antigen concentration

Specific antibodies

Immunodiffusion, immunonephelometry, immunoturbidimetry, ELISA, radial immunodiffusion, electrophoresis

Concentration

–

Fibrin network

Clot permeability

Hydraulic conductivity reflecting porosity of the clot

Thrombin- or TF-activated clot formation

Volumetry of the percolating buffer

Darcy constant (Ks)

Hands-on experience required

Clot density

Functional

Thrombin- or TF-activated clot formation

Turbidimetry

Lag time, clot rate (slope), maximum absorbance, mass-length ratio

High accuracy required

Clot compaction

Functional

Thrombin- (possibly TF-) based clot formation

Optical or volumetric assessment of supernatant evacuated after clot retraction

Percentage of retraction

–

Fibrin resistance to lysis

Functional

Thrombin- or TF-activated clot formation in the presence of different tPA concentrations

Turbidimetry

Lysis times (CLT, Lys50), lysis rate, area under the curve, overall hemostatic potential

High accuracy required

Clot strength

Functional

Thrombin-, TF-activated, or surface-dependent (FXII-based) clot formation

Viscoelastic testing, including ROTEM and TEG (mostly assessed in whole blood)

Storage/elastic moduli (G’, G’’), tanδ (G’’/G’), clotting times, clot strength (amplitude), angle of the clot formation curve

High accuracy required in manual assessment, dedicated reagents for functional fibrinogen assessment using TEG or ROTEM (specifically FIBTEM)

D-dimer release from clot

Functional

Thrombin- (possibly TF-) based clot formation and lysis in the presence of high tPA concentrations

Modified Ks-based assay with assessment of D-dimer levels in the effluent

Maximum or rate of D-dimer release

Time-consuming and

requiring much experience

Microscopic techniques

Structure

Thrombin- or TF-activated clot formation

Confocal laser, scanning electron, atomic force microscopy

Fibrin fibers diameter, porosity, branching

High accuracy required

Fibrinolytic potential (microscopic)

Functional

Thrombin- (possibly TF-) based clot formation and lysis in the presence of high plasminogen/tPA concentrations

Confocal laser microscopy

Lysis time or percentage of lysis

High accuracy required

Abbreviations: APTT, activated partial thromboplastin time; CLT, clot lysis time (eg, the time from midpoints of the clear-to-maximum-turbid transition to the maximum-turbid-to-clear transition); ELISA, enzyme-linked immunosorbent assay; FIBTEM, type of the rotational thromboelastometry test to determine functional fibrinogen; FXII, factor XII; Lys50, time to 50% lysis; TEG, thromboelastography; TF, tissue factor; PT, prothrombin time; ROTEM, rotational thromboelastometry; tPA, tissue-type plasminogen activator

Fibrinogen concentrate is widely used in the treatment of patients with hypofibrinogenemia to stop bleeding. The first fibrinogen purification by precipitation with half-saturated NaCl was performed by Olaf Hammarsten in 1879. Isolation of a stable fibrinogen fraction (the first fraction, called fraction I-0, obtained from plasma by extraction in acidic buffer with glycine and ethanol at –3 °C) for treatment purposes was developed by Margareta and Birger Blombäck in 1956.18 Due to high content of a large von Willebrand factor (vWF)-FVIII complex, fraction I-0 was used in patients with fibrinogen deficiencies as well as those with bleeding diatheses, such as hemophilia A or von Willebrand disease.18 In the early 1980s, blood products containing fibrinogen isolated from the blood of HIV-positive volunteers led to the spread of this virus.18 A pasteurization method developed later for blood concentrates was beneficial for patients with bleeding diatheses who survived this dramatic era.18

Fibrin clots and their properties

Fibrin formation and degradation occur physiologically at a low rate, as reflected by detectable fibrin degradation products, including D-dimer, in healthy individuals.19 It is considered that the properties of hydrated plasma fibrin clots formed ex vivo are similar to those of clots formed in vivo.20 Enclosed water provides specific gel properties of “sticky” fibrin.14,20 Despite the fact that thrombotic events in vivo are complex and determined by different pro- and anticoagulant factors, vessel wall properties, interaction with all cellular blood constituents, and flow conditions, the physical properties of fibrin evaluated in a controlled environment seem to be crucial for pathological sequelae.21 The plasma fibrin characteristics most frequently used based on available literature, along with supporting data from several seminal studies,22-33 are presented in Table 1. The use of a proper clotting trigger also matters and may affect the outcome.34 Similar to simple clotting time assays, recombinant tissue factor (TF) or different concentrations of exogenous thrombin are used to study specific fibrin characteristics.34 Clotting activators and plasma dilution have been shown to influence the results of commonly used fibrinolysis assays performed in plasma. This indicates that the arbitrary selection of an assay affects the extent to which specific lysis determinants, such as PAI-1 and TAFI, are linked to lysis time.35 Notably, plasma C-reactive protein was the predictor of all commonly used lysis assays after extensive adjustment.35

The contribution of environmental factors to determining fibrin porosity (Table 1) has been shown to be about 36% higher than the influence of heritability.36 Formation of a denser fibrin structure, resistant to lysis, was repeatedly reported at higher levels of fibrinogen, thrombin, C-reactive protein, or glucose (Table 2).37 Platelet-derived factors, immunoglobulins, oxidative stress, or neutrophil extracellular trap–related markers have also been associated with unfavorable fibrin alterations (Table 2).37-39 Recently, circulating levels of FXI, the extent of protein carbonylation, and lipopolysaccharide (LPS)-associated low-grade inflammation have also been shown to modulate clot phenotype (Table 2).40,41 Other factors related to impaired fibrin clot structure and function include increased growth-differentiation factor-15 (GDF-15) and histidine-rich glycoprotein (HRG); however, their impact on fibrin formation has not been fully understood.40,42 GDF-15 is considered a cytoprotective cytokine induced by stress, ischemia, and inflammation, making it a promising biomarker in cardiovascular disease (CVD).40 While it remains unclear whether upregulated GDF-15 contributes to disease progression, its positive association with enhanced thrombin generation and prolonged lysis time, as well as its negative correlation with clot porosity (reflected by the Darcy constant, Ks), suggest that this biomarker may directly reflect prothrombotic alterations.40 HRG, as a highly abundant protein in plasma and fibrin clots, exerts anticoagulant and anti-inflammatory effects.42 However, HRG elevation has been linked to the development of post-thrombotic syndrome in patients following the first-ever deep vein thrombosis (DVT) episode. Also, HRG has been shown to have negative associations with Ks and positive correlations with clot density and lysis time, which suggests that it interacts with fibrinogen and affects fibrin formation.42 Lastly, higher PAI-1 levels, which largely contribute to impaired clot susceptibility to lysis, have been observed in patients with aortic stenosis and COVID-19.43,44

Table 2. Revisited factors and related clinical conditions or habits associated with the formation of denser and less permeable fibrin clots and / or their resistance to lysis

Factor

Clinical condition / habit

↑ Fibrinogen (reviewed in37,46)

CAD, DVT, ischemic stroke, PAD, RA, COPD, smoking

↑ Thrombin (reviewed in37,46)

CAD, DVT, PAD, ischemic stroke, cancer, asthma, antithrombin deficiency, smoking

↑ C-reactive protein (reviewed in37,46)

CAD, DVT, ischemic stroke, PAD, RA, COPD, IBD, AS

↑ Glucose (reviewed in37,46)

Diabetes

↑ Homocysteine (reviewed in37,46)

CAD, PAD

↑ Lipoprotein(a) (reviewed in37,46,77)

CAD, PAD, residual vein thrombosis, AS

Coagulation factors (genes; reviewed in37,45)

Genetic variants of fibrinogen chains (FGB rs1800790), prothrombin (G20210A), FV Leiden, FXIII (F13 rs5985)

↑ FXI/FXIa (reviewed in40)

CAD, VTE, LVT, eosinophilic granulomatosis with polyangiitis

↑ Platelet activity/-derived factors (reviewed in37)

CAD, PAD, residual vein thrombosis, AF

↑ Immunoglobulins (reviewed in37,39)

Multiple myeloma, APS

↑ Oxidative stress (reviewed in37,41)

Acute MI, DVT, RA, COPD, ischemic stroke, AS, AF, menopause, smoking

↑ NETs-related markers (reviewed in38,96)

Acute PE, AF, asthma

↑ GDF-15 (reviewed in40)

AF

↑ LPS (reviewed in40,41)

Acute PE, acute MI, acute ischemic stroke, AF

↑ complement activation33,50

Diabetes, acute PE

↑ PAI-143,44

AS, COVID-19

↑ HRG42

VTE

Abbreviations: ↑, increase; AF, atrial fibrillation; APS, antiphospholipid syndrome; AS, aortic stenois; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; DVT, deep vein thrombosis; F, factor; GDF-15, growth differentiation factor-15; HRG, histidine-rich glycoprotein; IBD, inflammatory bowel disease; LPS, lipopolysaccharide; LVT, left ventricular thrombus; MI, myocardial infarction; NET, neutrophil extracellular traps; PAD, peripheral artery disease; PAI-1, plasminogen activator inhibitor type 1; PE, pulmonary embolism; RA, rheumatoid arthritis; VTE, venous thromboembolism

Genetic risk factors for venous thromboembolism (VTE), such as FV Leiden, prothrombin G20210A, or antithrombin deficiency, have been related to the formation of compact and poorly lysable plasma fibrin clots.37 For instance, the association between common fibrinogen (FGB rs1800790) and FXIII (F13 rs5985) gene variants with unfavorably altered fibrin clot characteristics, specifically lower Ks and prolonged lysis time, was strong enough to be observed in the acute phase of pulmonary embolism (PE).45

The formation of plasma fibrin clots displaying denser architecture, as reflected by scanning electron microscopy (SEM; Figure 1B and 1C) or fluorescent / confocal laser microscopy (Figure 1D), and reduced Ks, with its functional consequences, such as higher stiffness and resistance to fibrinolysis (ie, prolonged clot lysis time), have been shown in multiple diseases related to thrombotic events or inflammatory states, mostly interconjugated.46 Importantly, cardiovascular risk factors, such as smoking, hyperlipidemia, obesity, hypertension, and diabetes mellitus have also been associated with unfavorably altered fibrin clot properties.46

Evaluation of thrombi obtained by mechanical thrombectomy from patients with acute myocardial infarction (MI; Figure 1C) confirmed that reduced Ks, reflecting low clot porosity, was the strongest predictor of high fibrin content visualized ex vivo using SEM within intracoronary thrombi.47 To date, assessment of thrombus properties in vivo is elusive; however, currently used diagnostic techniques, such as high-resolution optical coherence tomography, can identify novel clinically relevant factors linking detailed arterial plaque pathology with ex vivo fibrin properties.

Mass spectrometry–based proteomic profiling provides a detailed, quantitative depiction of clot protein composition, showing subtle changes between particular clinical conditions,48 thereby uncovering novel biomarkers and mapping molecular pathways related to thrombosis. Based on quantitative liquid chromatography–mass spectrometry, nearly 500 proteins, among 3000 distinct proteins in human plasma and 20 000 in the human proteome, have been identified within plasma fibrin clots of healthy individuals.49 The most abundant were fibrinogen chains and fibronectin, followed by α2-antiplasmin, FXIII, complement component C3, and HRG.49 The clot proteome differs in patients following thromboembolism as well as in those at a risk of thrombosis.48,50 Current evidence indicates that even less abundant proteins or those not directly related to the classic Virchow triad, such as hypercoagulability, endothelial dysfunction, or restricted blood flow, are associated with key fibrin clot measures, such as porosity and susceptibility to lysis.50,51

Fibrin is characterized by unique extensibility, and its single fibers can be strained up to 3-fold without permanent deformation, and up to 500% before rupture, enabling the formation of stable clots and a scaffold for other blood components.52 Indeed, blood flow during clot formation not only provides coagulation proteins to the site of thrombosis but also aligns fibrin fibers, with further implications for clot architecture.53

In the past few years, besides typical forms of fibrin detected in vivo, such as fibrin monomers, thrombi (heterogeneous in composition), fibrin deposits, or degradation products, fibrin biofilm and microclots have also been reported (Figure 2). The former type has been shown to cover the surface of blood clots and intravascular thrombi, while in vitro fibrin film was formed at the air-blood interface via the Langmuir technique.54-56 This type of biofilm was protective against bacteria in mice with dermal infection.54 Recently, it has been reported that red blood cells can support the formation of fibrin film, which limited platelet adhesion.57 Moreover, fibrin film formation was shown to be reduced on inflamed cells, which may allow for thrombus growth in vivo.57

Figure 2. Schematic overview of fibrin formation and degradation. Fibrin polymerization occurs physiologically. Fibrin monomers polymerize via noncovalent interactions between the D and E regions. Thrombin specifically cleaves fibrinopeptide A (FpA) and B (FpB), the latter at a slower rate, leading to protofibril formation and lateral aggregation. Fibrin resistance to lysis is determined by covalent cross-linking between γ-γ, γ-α, and α-α chains, mediated by activated factor (F)XIII. FXIII is also responsible for interactions of fibrin with other proteins, such as α2-antiplasmin or plasminogen activator inhibitor type 1, the principal inhibitor of fibrinolysis. Plasminogen conversion to plasmin by tissue plasminogen activator occurs at C-terminal lysine binding sites on fibrin and leads to fibrin degradation and generation of different cleavage products of cross-linked fibrin, such as 2 covalently bound fibrin D-domains, known as D-dimer.

Cross-liked fibrin can capture thrombin, a process known as antithrombin I activity, thereby attenuating thrombin generation. Other proteins, such as antithrombin or α2-macroglobulin, also bind thrombin, diminishing thrombin burst.

Proteomic studies have revealed the presence of nearly half a thousand clot-bound proteins, including those unrelated to the coagulation system, highlighting the complex mechanisms and interactions involved in thrombosis. Atypical forms of fibrin have been discovered in vivo, such as amyloid fibrin—fibrinolysis-resistant aggregates called microthrombi—formed in response to the S1 subunit of the SARS-COv-2 spike protein, as reported in COVID-19, or lipopolysaccharide (LPS). Microclots can aggregate with platelets to form microthrombi, which can further lead to embolization in small capillaries (eg, within lung vasculature) or contribute to massive thrombosis. Another unusual form of fibrin recently described is a biofilm, which is involved in wound healing, inflammation and clot contraction, the latter being a physiological mechanism resulting in the shrinkage of clot volume.

Created in BiOrender. Hojda, A. (2025) https://BiOrender.com/g16n849

Another atypical form of fibrin reported previously is the microclot. Soluble fibrin monomers and oligomers (<⁠16 monomers) are highly reactive and may form microclots.58 Levels of soluble fibrin are markedly higher than those of D-dimer and regulated by transendothelial transfer, endocytosis, and interaction with cellular receptors, such as macrophage-1 antigen and low-density lipoprotein receptor, since desA fibrin does not stimulate plasminogen activation by tPA, in contrast to microclots.58 However, under specific clinical conditions, aggregates composed of the amyloid form of “fibrin(ogen)” (fibrinaloid) containing extensive β-sheets, with or without the involvement of other proteins, are formed.59 Amyloid microclots (50–150 nm) were stained in platelet-poor plasma with Thioflavin T and visualized by fluorescence microscopy. Such clots are resistant to fibrinolysis and therefore may not correlate with a typical D-dimer increase.60 For successful proteome analysis, a double digestion of amyloid microclots with trypsin was required,61 highlighting their potential thrombogenic nature. In most patients with low-to-intermediate risk of acute PE, a 24-hour treatment with low-molecular-weight heparin reduced fibrin aggregates; however, their detailed composition, particularly amyloid content, was not evaluated.62 Fibrinaloid is formed in the presence of serum amyloid A, lipopolysaccharide (LPS), S1 subunit of the SARS-CoV-2 spike protein, iron ions, estogens, or lipoteichoic acid.60 Additionally, extension / compression of hydrated fibrin clots results in a conversion of α-helix to β-sheet, favoring amyloid formation.63 Interestingly, fibrin clot can capture LPS, which has been documented to accumulate within murine thrombi.64 In low-grade endotoxemia detectable in patients with acute PE, higher LPS levels correlated with unfavorably altered fibrin clot properties.33 Similarly, COVID-19 was associated with the formation of fibrin clots that were denser and more resistant to lysis.65 Although the important role of fibrinaloid microclots in COVID-19 has been established,60 it remains unknown how to manage these amyloid structures in a real-life setting, and whether they can lead to severe thrombosis.

Fibrinogen as a predictor of coronary heart disease and thrombotic events

Large epidemiologic studies have shown associations between increased plasma fibrinogen levels and coronary heart disease (CHD), MI, and ischemic stroke.66-72

An increase in fibrinogen level by 1 g/l, adjusted for age and sex, doubles the risk of CHD (hazard ratio [HR], 2.42; 95% CI, 2.24–2.6) and ischemic stroke (HR, 2.08; 95% CI, 1.74–2.48).73 Fibrinogen concentration increases with age, and was shown to be higher in women and individuals of black ethnicity, explaining over 30% of the variation in fibrinogen levels.74 Modifiable risk factors, such as smoking, body-mass index, or high-density lipoprotein cholesterol level, explained 7% of the total fibrinogen variability, while inflammatory markers explained a further 10% of the variation in fibrinogen levels.74 Importantly, factors contributing to half of the total variability in fibrinogen levels remain unrecognized.74 Although risk alleles associated with higher fibrinogen levels are known (eg, the single nucleotide polymorphism rs1800789 in FGB), combined genetic variants can explain only up to 2% of the total variance in fibrinogen levels.75

A complex interaction between fibrinogen and CVD is also supported by fibrin deposition within atherosclerotic plaque or stenotic aortic valve, both associated with low-grade inflammation, lipid peroxidation, and a prothrombotic state.76,77 A combined fixed-effect meta-analysis confirmed a causal effect of fibrinogen in CHD (HR, 1.75; 95% CI, 1.22–2.51) but not in incident MI (HR, 1.45; 95% CI, 0.85–2.49; P = 0.17).78 Moreover, a meta-analysis of over 20 000 patients with coronary artery disease showed that fibrinogen can predict adverse outcomes and serve as a biomarker for risk stratification.79

A recent meta-analysis of 24 case-control and cohort studies, using a Mendelian randomization (MR) approach, confirmed that the risk of ischemic stroke was causally related to fibrinogen levels, with each increase in fibrinogen by 1 g/l doubling the risk (odds ratio [OR], 2.28; 95% CI, 1.53–3.03).80 Such an association was not observed for hemorrhagic stroke (OR, 0.88; 95% CI, 0.06–18.28; P = 0.93).81

Less evident data showed a causal association between increased fibrinogen levels and the risk of VTE. Results of the Leiden thrombophilia study supported the concept that elevated fibrinogen was more likely a cause than a consequence of VTE, as fibrinogen levels greater than or equal to 5 g/l, as compared with the reference value below 3 g/l, were shown to be associated with a 4-fold higher VTE risk after adjustment for C-reactive protein.82 Klovaite et al83 assessed 77 608 participants from the general Danish population, including individuals with a history of DVT and / or PE, and showed that those with fibrinogen levels greater than or equal to 4.6 g/l, as compared with those with levels of 3 g/l or lower, had a higher risk for PE combined with DVT (OR, 2.1; 95% CI, 1.2–3.8), but not for DVT alone. However, no causal association was shown using the MR approach.83 Payne et al,84 in a case-control study designed to identify risk factors for VTE (the GATE [Genetic Attributes and Thrombosis Epidemiology] study) among 1145 cases with a first or recurrent VTE and 1309 controls, showed that fibrinogen levels exceeding 4 g/l were associated with VTE in both black (crude OR, 1.81; 95% CI, 1.18–2.78) and white populations (crude OR, 2.85; CI, 1.89–4.31) after adjustment for age and sex. However, after extensive adjustment for other contributing factors, including FVII, FVIII, vWF and comorbidities, fibrinogen no longer remained a predictor of VTE.84 A very recent prospective population-based study identified common missense variants in the 3 fibrinogen genes, showing that FGG (rs6063) was associated with markedly increased odds of first-ever VTE (OR, 8.2; 95% CI, 1.05–63.6; adjusted for age and sex), while FGA (rs6050) predicted recurrent VTE (HR, 1.8; 95% CI, 1.1–2.8).85 Interestingly, both genetic variants had a stronger effect on VTE risk than the well-recognized FV Leiden (HR, 1.9; 95% CI, 1.2–3).85 Combined risk alleles from these factors proportionally increased the risk of unprovoked recurrent VTE up to 4-fold (HR, 4.6; 95% CI, 1.9–11.3 for ≥3 risk alleles).85 On the other hand, Maners et al86 showed a protective effect of higher γ′ fibrinogen, comprising up to 10% of total plasma fibrinogen, on the risk of VTE (OR, 0.34; 95% CI, 0.28–0.41), cardioembolic stroke (OR, 0.67; 95% CI, 0.56–0.8), and large artery stroke (OR, 0.64; 95% CI, 0.53–0.77; all P <⁠0.001).

Fibrinogen as a predictor of mortality

Data on associations between fibrinogen levels and mortality in the general population are limited. In the largest available meta-analysis involving 154 211 participants from 31 prospective studies, an increase in fibrinogen by 1 g/l markedly elevated the risk of vascular mortality not associated with CHD or stroke (HR, 2.76; 95% CI, 2.28–3.35) as well as nonvascular mortality (HR, 2.03; 95% CI, 1.9–2.18) during 1.38 million person-years.73 Higher plasma fibrinogen concentration was also an independent risk factor for all-cause mortality among 3571 men from the general population aged 71 to 93 years over a median follow-up of 4.4 years.87 After adjustment for age and confounding risk factors, the relative risk (RR) for all-cause mortality associated with an increase in fibrinogen levels by 0.64 g/l was 1.3 (95% CI, 1.2–1.4), and was similar irrespective of CVD, cancer, or other-cause mortality.87

Among 3092 men with CHD, including those with a history of MI, followed for over 3 years, the RR for all-cause mortality associated with the highest fibrinogen tertile (≥3.68 g/l), as compared with the lowest tertile (<⁠3.09 g/l) was 1.67 (95% CI, 1.16–2.41).88 A plasma fibrinogen increase by 0.75 g/l was shown to be associated with a 31% higher risk of all-cause mortality in this cohort of patients.88 Similarly, fibrinogen levels greater than or equal to 3.5 g/l, as compared with lower values, independently predicted 1-year mortality (OR, 1.69; 95% CI, 1.12–2.55) in consecutive patients with ischemic stroke who were admitted to a hospital within 24 hours of the symptom onset (n = 900).89 In patients with chronic obstructive pulmonary disease, fibrinogen levels greater than or equal to 3.5 g/l vs below 3.5 g/l were associated with about a 2-fold higher risk of death within 36 months (HR, 1.94; 95% CI, 1.62–2.31).90 A similar risk of all-cause mortality at fibrinogen levels greater than or equal to 3.5 g/l vs below 2.9 g/l (HR, 1.87; 95% CI, 1.09–3.23) was observed among patients with peripheral artery disease.91 Some studies reported a U-shaped association of fibrinogen levels with mortality, especially in women, suggesting that lower fibrinogen concentrations may also increase the mortality risk,92 as is well evidenced in trauma patients.93

Fibrin clot measures and unfavorable clinical outcomes

A growing body of evidence from epidemiological studies indicates that unfavorably altered fibrin clot properties, that is, 10%–15% reduced Ks and / or prolonged lysis time, are observed in patients with a history of both arterial and venous thrombotic events, as compared with controls, and that a more prothrombotic fibrin clot phenotype (the lowest values of Ks or the longest lysis times showing a difference of ≥10% compared with the values in the opposite quartiles) can predispose to recurrent thrombosis.46 Among 786 stable CAD patients receiving aspirin, clot area under the curve (Table 1; first vs fourth quartile, crude HR, 2.4; 95% CI, 1.2–4.6), but not maximum absorbance or lysis time, predicted the composite end point encompassing MI, ischemic stroke, or cardiovascular death during a median follow-up of 3 years.94 However, impaired susceptibility to lysis reflected by increased levels of D-dimer (>4.26 mg/l) released during tPA-induced clot degradation (Table 1) predicted cardiovascular death (HR, 5.43; 95% CI, 1.99–14.79) in patients with type 2 diabetes over a median follow-up of 72 months.95 Combined biomarkers, including prolonged clot lysis time, higher endogenous thrombin generation, and citrullinated histone H3, markedly increased the odds of PE-related death within 1 year of acute PE (OR, 9.3; 95% CI, 2.1–40.3).96 The PLATO (Platelet Inhibition and Patient Outcomes) substudy97 assessed clinical outcomes in 4354 patients who recently experienced acute coronary syndrome (ACS), randomized to clopidogrel or ticagrelor during 1-year follow-up. Ex vivo–assessed plasma clot resistance to fibrinolysis, as reflected by prolonged lysis time, was independently associated with cardiovascular death or spontaneous MI (per each 50% increase in lysis time; HR, 1.17; 95% CI, 1.05–1.31) and cardiovascular death alone (HR, 1.36; 95% CI, 1.17–1.59).97 Moreover, each 50% increase in maximum clot turbidity, reflecting higher clot density that is moderately explained by plasma fibrinogen concentrations, elevated the risk of cardiovascular death (HR, 1.24; 95% CI, 1.03–1.5).97 After extensive adjustment for inflammatory and cardiac biomarkers, only the association of lysis time with cardiovascular death remained significant (HR, 1.2; 95% CI, 1.01–1.42). Similar conclusions were drawn in the subgroup analysis involving diabetic patients extracted from the initial cohort.98 In this substudy, a 50% prolonged lysis time was associated with a 38% higher risk of cardiovascular death among the ACS patients with diabetes (HR, 1.38; 95% CI, 1.08–1.76), and remained significant after comprehensive adjustment (P = 0.034).98 However, long-term follow-up studies are required to determine the value of plasma fibrinolytic potential in predicting adverse clinical outcomes in different cohorts of patients, such as those with risk factors for VTE.

Conclusions

Elevated and paradoxically reduced levels of fibrinogen, with significant contributions from genetic variants of fibrinogen chains affecting fibrin architecture and function, are associated with an increased risk of thromboembolic complications and mortality. Specific drugs affecting fibrinogen levels, such as ancrod, which directly causes defibrinogenation, did not improve clinical outcomes in patients with acute ischemic stroke.99 However, nonspecific medications, such as fibrates or statins, which are suggested to reduce fibrinogen levels100 and reported to favorably modify fibrin clot phenotype,101 may additionally reduce cardiovascular risk beyond their lipid-lowering effects. Aside from thrombolytic therapy used in acute thrombotic states, there are no validated strategies to improve fibrinolytic potential in vivo, especially for patients at a high risk of thrombosis. Potential candidates for therapeutic applications include Affimers against α2-antiplasmin, the primary inhibitor of plasmin, and modulators of PAI-1 or other serine protease inhibitors.102-104

Clot lysis time is practically the easiest fibrin measure to assess, though the method of its evaluation requires clinical validation. The subcommittee on Factor XIII and Fibrinogen of the International Society on Thrombosis and Haemostasis is actively working on the standardization of plasma clot turbidity and lysis assays, highlighting the significant potential of clot turbidimetric assessment for clinical applications in CVD.105 Combining fibrinogen and clot lysis time with routine biomarkers may benefit high-risk patients and offer a novel approach for personalized therapy. This strategy has been well proven in patients at a risk of post-thrombotic syndrome, where implementing high fibrinogen levels as a criterion for additional ultrasound-accelerated catheter-directed thrombolysis resulted in better outcomes.106 Long-term follow-up studies are needed to confirm the effectiveness of using fibrinogen and / or clot lysis time as additional markers to guide personalized treatment.