Ischaemic Stroke, Thromboembolism and Clot Structure

Ischaemic stroke is a major cause of morbidity and mortality worldwide. Blood clotting and thromboembolism playacentralroleinthepathogenesisofischaemicstroke.Anincreasingnumberofrecentstudiesindicatechanges in blood clotstructureand compositionin patientswithischaemic stroke.Inthis review,weaim to summarise and discuss clot structure, function and composition in ischaemic stroke, including its relationships with clinical diagnosis and treatment options such as thrombolysis and thrombectomy. Studies are summarised in which clot structure and composition is analysed both in vitro from patients ’ plasma samples and ex vivo in thrombi obtained through interventional catheter ‐ mediated thrombectomy. Mechanisms that drive clot composition and architecture such as neutrophil extracellular traps and clot contraction are also discussed. We ﬁ nd that, while in vitro clot structure in plasma samples from ischaemic stroke patients are consistently altered, showing denser clots that are more resistant to ﬁ brinolysis, current data on the composition and architecture of ex vivo clots obtained by thrombectomy are more variable. With the potential of advances in technologies underpinning both the imaging and retrieving of clots, we expect that future studies in this area will generate new data that is of interest for the diagnosis, optimal treatment strategies and clinical management of patients with ischaemic stroke.


Introduction
Blood clotting is well known to play a key role in the pathogenesis of ischaemic stroke.Known risk factors for ischaemic stroke include hypertension, atrial fibrillation, hypercholesterolaemia, smoking, diabetes and heart disease.Blood clots that cause stroke may be generated in the heart, e.g.due to atrial fibrillation or heart disease, and cause cardioembolic stroke, or they may be generated due to atherosclerosis in the carotid artery, leading to plaque rupture and subsequent thrombus formation.However, the identification of where the original blood clot that caused the stroke came from is not always easy to establish clinically.Furthermore, more often than not the blood clot that causes the ischaemic stroke is embolic in nature (i.e. the site of occlusion is distal from the original site of thrombosis), whether they originate from the heart or the carotid artery.These different levels of intricacies of the origin and role of blood clots in stroke complicate the diagnosis and treatment of the disease, including secondary prevention of recurrence of ischaemic events.
Recent evidence supports an important contribution of blood clot structure to the risk of thromboembolism and ischaemic events.Earlier studies discussed below have investigated in vitro clot structure using plasma samples from patients with ischaemic stroke.With the advent of rapidly developing and improving options for treating ischaemic stroke with intravascular removal of thrombi using cathetermediated clot retrievers or thrombectomy, new opportunities have arisen to also study the ex vivo structure and composition of thrombi obtained from patients with stroke as further discussed below.Through these studies, we are developing an increased understanding of the heterogeneity of blood clots causing stroke.We are also gaining insight into how blood clots in stroke may behave differently due to their composition and structure, and thus may respond differently to therapeutic approaches of thrombolysis and mechanical clot removal.This review aims to summarise and discuss the fundamental new developments in our knowledge of clot structure and composition in stroke, including exploring its relevance to treatment with thrombolysis or thrombectomy, and examining the mechanistic roles of neutrophil extracellular traps (NETs) and clot contraction (Fig. 1).

Clot structure and stroke
Fibrin clot structure and function is a significant risk factor for cardiovascular disease and diseases associated with thrombosis (Undas and Ariens, 2011).The structure and properties of the fibrin clot are significantly altered in patients with stroke, involving changes in fibrin fibre diameter, fibre branching and clot network density.Moreover, compositional alterations in clots and thrombi such as platelet, erythrocyte and white blood cell content are also relevant for patients with stroke.
Several papers have demonstrated that patients with acute ischaemic stroke, on average, form denser, less permeable clots, and at a faster rateresulting in reduced susceptibility to fibrinolysis (Undas et al., 2009;Undas et al., 2010;Rooth et al., 2011;Pera et al., 2012;Stanford et al., 2015); while one study indicated that the fibrin fibres themselves may also be thicker and more numerous (Undas et al., 2009).It has been suggested that the combination of these structural changesparticularly that of clot density and permeabilitymay contribute to reduced fibrinolysis, limiting clot degradation.The effects of denser clots on fibrinolysis are likely due to reduced permeation of lytic enzymes such as tissue plasminogen activator (tPA) into the clot, reduced plasmin generation, increased incorporation of inhibitors of fibrinolysis, or a combination of these mechanisms.
The significance of these structural clot changes is further highlighted by findings that reduced clot permeability (more compact fibrin network) was associated with an increased stroke risk amongst patients with atrial fibrillation (AF) taking vitamin K antagonists (Drabik et al., 2017).Patients with reduced clot permeability showed a six times greater incidence of transient ischaemic attacks and stroke, with a hazard ratio of 6.55 during the 3.7-4.8years of follow-up.Associations have also been made between fibrin clot properties and stroke severity, with reports showing that the extent of clot compaction correlated with neurological deficit on admission and discharge (Undas et al., 2010), however, no association with mortality was found in long-term follow-up.Other studies found no link between fibrin network permeability and stroke severity (graded based on the NIHSS score) or outcome (Rooth et al., 2011), but a smaller study size and the loss of some patients prior to followup, alongside the ongoing treatment of patients and differing methods (use of re-calcified EDTA-plasma instead of citrate) may account for this.Pro-thrombotic clot structure alterations were shown to be persistent, with patients with acute ischaemic stroke showing comparatively denser clots and reduced susceptibility of clots to fibrinolysis (Fig. 1) during both the acute (onset of symptoms) and convalescent (60 days after) stages of strokealthough the rate of/capacity for fibrin formation did decrease (Rooth et al., 2011).
In addition to studies on the structure of in vitro clots made with plasma samples from patients with stroke, the recent developments in thrombectomy options for patients with stroke allow for the structural and compositional analysis of in situ thrombi obtained from patients.Reports have shown that clot composition differs depending on stroke aetiology, with one retrospective study finding clots in large artery stroke showing a higher red blood cell (RBC) but lower platelet density compared to cardio-embolic clots (Brinjikji et al., 2021).A subsequent paper also showed a higher percentage of white blood cells (WBC) (Boeckh-Behrens et al., 2016).Another study showed similar results, with RBC count being highest in atherogenic thrombi (56.9 ± 12.2%), while cardiogenic thrombi were more fibrin-rich (39.5 ± 13.5%), with only a minor (4.9 ± 3.0%) WBC component (Ahn et al., 2016).The authors also described the arrangement of these componentsindicating that atherogenic clots showed central RBC masses surrounded by a thin outer shell of platelets and fibrin, while cardiogenic clots contained dispersed areas of platelet aggregates.These findings may indicate that the structural distribution of components may be more significant for aetiology than a breakdown of the composition of the clot.Nevertheless, similar prior studies have failed to find a relationship between stroke mechanism and clot composition, with one investigating thrombi retrieved during endovascular surgery from the middle cerebral artery and intracranial carotid artery finding no distinct differences in histological characteristics between those of cardioembolic and atherosclerotic origin (Marder et al., 2006).Another study that utilised H&E and immunostaining of tissue specimens also found no recurrent patterns in clot composition within cardioembolic clots (Simons et al., 2015).However, interestingly one paper found that platelet-rich, 'white clots' were typically of cardioembolic or cryptogenic origin, and implicated in more distal occlusions, chiefly of the middle cerebral artery, showing reduced hyper-density and smaller Fig. 1.Role of NETs, clot composition and clot contraction in thrombolysis and thrombectomy.Presence of NETs within fibrin clots prevents or limits thrombolysis with tPA or its derivatives, and some studies indicate it also impedes thrombectomyresulting in worse functional outcomes.Clot composition has a variable effect on thrombolysis, with a denser fibrin network slowing thrombolysis, whereas more permeable, porous clots are more susceptible.Clots with a greater red blood cell content may improve thrombectomy.Clot contraction is associated with the inhibition of external lysis (thrombolysis), while internal lysis is accelerated.Greater clot contraction may result in more stable clots that could be removed in fewer passes, with reduced risk of generating micro-thrombi.Red arrowsinhibition.Orangemodulation.Green -promotion.(Figure created using Biorender and Microsoft PowerPoint).
size.Conversely, fibrin and RBC rich 'red clots', which were instead linked to LAA etiology and found to cause more occlusions of the internal carotid artery (Mereuta et al., 2021).The potential reasons behind these differences in composition remain poorly understood.
Overall, clot characteristics are significantly, unfavourably altered in stroke, with clot permeability and density being increasingly shown as potential markers of both risk and prognosis or clinical outcome.It remains unclear, however, how in vitro clot structure, or the propensity to form certain clot structures, relates to thrombus composition in situ.Furthermore, there is still a lack of consensus on whether stroke subtype can be determined based on clot composition or structure, which appears heterogeneous in nature.Future studies based on new methods or larger groups of patients may be able to shed light on this.

Stroke treatment and clot structure
Treatment of stroke can be approached with strategies of systemic thrombolysis or intravascular thrombectomy, the latter representing an area undergoing rapid recent developments.Several studies have suggested that both treatment modalities are associated with improved long-term neurological/functional outcomes and survival (Emberson et al., 2014;Goyal et al., 2015;Saver et al., 2016).It is possible that the effectiveness and benefits of these treatments may be influenced by clot structure.Thrombolysis involves administering 'clot busting' drugs to dissolve the occlusion causing the strokewith tPA and its derivative thrombolytic agents (e.g.alteplase) showing particular promise due to their fibrin specificity, rendering these drugs less likely to cause intracranial haemorrhage than streptokinase and urokinase, which are not fibrin-specific (Zivin et al., 1985; The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group, 1995;Sun et al., 2020).Recent clinical trials indicate that a genetically modified version of rtPA (Tenecteplase) may be superior for ischemic stroke treatment, usurping Alteplase due to its higher fibrin affinity, longer circulating half-life and simpler administration (Bivard et al., 2022;Mahmood and Muir, 2022;Menon et al., 2022;Murphy et al., 2023).However, data remain limited due to small sample sizes, indicating the need for studies with larger samples (Huang et al., 2015;Kvistad et al., 2022).Hence we have focussed on trials using Alteplase in this review.
Thrombolysis has a time-critical treatment windowwith intravenous administration within 1.5 hrs post onset showing almost double the odds of a 3-month favourable outcome (2.8) as compared to those where treatment was administered after 1.5-3 hrs (1.55 odds ratio), indicating that earlier intervention is associated with improved efficacy and outcome (Hacke et al., 2004).Administration of alteplase 3-4.5 hrs after onset continued to convey some benefit, with 35.3% of patients showing positive outcomeswith an odds ratio of 1.26 (Emberson et al., 2014).However, proposals that this window could be extended up to 6 hrs are less well-evidenced (Clark et al., 2000;Sandercock et al., 2012).
Alternative methods of administration have also been explored, with intra-arterial fibrinolytics providing a potential mechanism for reaching persisting, more distal (micro) clots following thrombectomy (Desilles et al., 2015); These are typically delivered via the same catheter used for the stent retriever.Principally, the CHOICE trial showed improvement in neurological outcomes at 90 days post-stroke for participants receiving intra-arterial alteplase post-thrombectomy (Renú et al., 2022).This aligns with other studies indicating intra-arterial alteplase is an effective rescue therapy, resulting in faster recanalization and higher reperfusion rates, while causing either no significant increase or a reduction in the risk of intracranial haemorrhage and mortality (Berkhemer et al., 2015;Heiferman et al., 2017;Yi et al., 2018;Anadani et al., 2019;Kaesmacher et al., 2019;Zaidi et al., 2019).Therefore, while further replication of findings, better-standardized, precise reporting of patterns between studies and more large-scale trials are needed to confirm (Kaesmacher et al., 2021a;2021b), the con-cept of adjunct therapies may be a credible new avenue for continuing to improve long-term stroke outcomes.
Clot location and size (del Zoppo et al., 1992;Saqqur et al., 2007;Kamalian et al., 2013) have been indicated as factors dictating thrombolysis efficacy.Previous studies have highlighted the importance of clot structure, particularly focusing on fibre arrangement, with denser, tighter networks demonstrating slower lysis (Fig. 1) (Collet et al., 2000).Another study indicated that an outer layer of thrombi, which is organised as a continuous outer shell and comprised of compacted platelets, fibrin fibres and vWF, acts as a 'shield' that is resistant to tPA protecting the thrombus core (Di Meglio et al., 2019).Fibrin fibre diameters are also significant, with densely packed, thinner fibre networks displaying slower lysis in numerous in vitro studies (Gabriel et al, 1992;Carr and Alving, 1995;Collet et al., 2000).However, some studies suggest that thinner fibres are lysed more rapidly (Bucay et al., 2015), and Bembenek et al. (2017) showed that slower forming, thinner fibre clots lysed faster in a study of clot characteristics in patients 24hrs post-thrombolysis.The latter clots were more loosely packedindicating that arrangement of the fibre network rather than fibre diameter may be more decisive for fibrinolysis susceptibility/efficacy.Some of the controversies may arise due to in vitro studies typically delivering fibrinolytic agents to the outside of a pre-formed clot, whereas in vivo, polymerization and fibrinolysis occur concurrently, allowing agents to permeate more extensively (Hudson, 2017).Histological composition may also hold significance, with RBC-rich clots generally being looser and thus more susceptible to thrombolysis, whereas platelet (Jang et al., 1989;Tomkins et al., 2015) and fibrinrich clots were more resistant to thrombolysis (Choi et al., 2018), although research into this area remains limited and the mechanisms behind how clot composition influences t-PA mediated fibrinolysis, unclear (Brouwer et al., 2018).
Thrombectomy may be offered alongside or instead of thrombolysis, particularly in patients with proximal vessel occlusion (of the internal carotid and middle cerebral artery) and larger thrombi/small infarct cores, (NHS England, 2018).Goyal et al. (2015) found that administering tPA in combination with endovascular intervention versus tPA alone resulted in a lower % 90-day mortality (10.4% vs 19% in the tPA alone) and improved functional outcomes.Thrombectomy involves the use of Magnetic resonance imaging, CT or CTA to identify thrombus location and thus eligible patients.A clot retrieval device can then be inserted via catheter, with stent retrievers, the first of which was the Merci Retriever System; this ranks among the most researched devices (Marder et al., 2006), though since then, many more devices have been developed, with their impact on clinical outcome being explored (Muller-Hulsbeck et al., 2001;Mokin et al., 2015;Yoo and Andersson, 2017).Key factors affecting thrombectomy efficacy include time between stroke onset and successful recanalization/ reperfusion as well as the number of recanalization attempts.Generally, 6 hrs is considered to be a critical treatment window (Khatri et al., 2009;Fransen et al., 2016), with the HERMES study showing that for each hour delay the chance of a patient being functionally independent 90 days post-thrombectomy decreased by 3.4% (Saver et al., 2016).However, there is also evidence that clinical benefit may still be obtained up to 24 hrs post-onset (Yoshimoto et al., 2021).
The histological composition of the clot could be an important factor in thrombectomy outcome.Several studies have examined the role of erythrocyte composition, noting that clots with a higher RBC content tended to be denser (Maekawa et al., 2018) and were associated with higher rates of successful and faster reperfusion (Hashimoto et al., 2016;Shin et al., 2018).It is postulated that the characteristics exhibited by clots with greater RBC composition may be responsible for this link, with in vitro studies suggesting that RBCs alter fibrin network structure causing increased viscosity and deformability (Gersh et al., 2009) -resulting in erythrocyte rich clots being comparatively less stiff (Chueh et al., 2011).Contrastingly, a higher platelet/fibrin content is correlated with increased thrombus stiffness (Boodt et al., 2021), the impact of which has been observed in clinical studies demonstrating fibrin-rich clots generally required more recanalization attempts and had a longer procedure time (Fig. 1) (Maekawa et al., 2018), contributing to more failed/poor reperfusions.Potential mechanisms underpinning this could include clots having greater friction, hindering their removal (Gunning et al., 2018), and their stiffness causing impeded penetration and reduced engagement with stent retrievers (Weafer et al., 2019).Meanwhile, platelet-rich thrombi similarly showed increased stiffness and were linked to worse revascularisation outcomes in thrombectomy (Douglas et al., 2020), also proving more difficult to remove via stent retriever than RBC-rich clots (Maekawa et al., 2018).Although, interestingly, one study suggested 'white clots' (those containing higher levels of platelets, vWF and calcifications) were less susceptible to aspiration, with their stiffness instead rendering them easier to remove via stent-retriever as a 'rescue-therapy', while 'red clots' (fibrin and red blood cell rich) typically only required aspiration (Mereuata et al., 2021).
Further to this, novel targeted treatments have focussed on other thrombi components that are pathologically altered in stroke, such as vWF and platelet glycoprotein expression.Multiple papers have demonstrated that higher plasma vWF levels are associated with increased stroke risk (Bongers et al., 2006;Wieberdink et al., 2010;Hanson et al., 2011;Williams et al., 2017) and greater stroke severity, with enduringly elevated levels post-fibrinolysis linked to poorer functional outcomes and survival (Bath et al., 1998;Carter et al., 2007;Samai et al., 2014;Tóth et al., 2018).This may occur due to higher levels of vWF promoting greater platelet packing and extracellular DNA accumulation, thus denser fibrin clot formation, with a higher platelet content, that is more resistant to fibrinolysis (Staessens et al., 2019).In light of this, in vitro and in vivo trials have explored the efficacy of selective vWF inhibitors, with aptamers such as BT200 (Kovacevic et al., 2021), ARC15105 (Siller-Matula et al., 2012), ARC1779 (Diener et al., 2009;Markus et al., 2011) and DTRI-031 (Nimjee et al., 2019), all showing promise in reducing platelet aggregation and thromboembolism, predictors of stroke.Moreover, the creation of antidotes for these aptamers would allow for rapid reversal of their effects and thus any associated bleeding (Nimjee et al., 2019).
Platelet glycoproteins are a further potential target, in particular, GpIb and GpVI receptors have been most explored, as these are implicated in enabling platelet adherence to the endothelium through binding with vWF and collagen respectively (Kehrel, 1995;Kleinschnitz et al., 2007), thus initiating thrombus formation and growth.Studies into the inhibition of these receptors have yielded some success, notably, the GpVI inhibitor, Revacept.During extensive pre-clinical trials, Revacept showed the ability to decrease thrombus formation, cerebral infarct size and resulting cerebral oedema, while improving functional outcomes without increasing risk of intracranial haemorrhage in mice.Interestingly, combination therapy of Revacept and rtPA resulted in the greatest increase in reperfusion, implying its use as an adjunct therapy may also be beneficial (Goebel et al., 2013;Reimann et al., 2016).Following this, Revacept as a monotherapy has since undergone its first-in-human trial, producing similarly positive findings (Ungerer et al., 2011), with the results of a larger-scale clinical trial now awaited (Schüpke et al., 2019).Meanwhile, the effect of GpIbblocking monoclonal antibodies has been investigated, showing an anti-thrombotic effect and thrombus size reduction in animal studies, similarly without increased bleeding complications (Cauwenberghs et al., 2000;Kraft et al., 2015;Schuhmann et al., 2017).More recently, an initial randomised clinical trial has been conducted demonstrating the safety and efficacy of GpIbα antagonist Anfibatide in humans (Li et al., 2021), suggesting the potential of these agents to be used as adjuncts alongside thrombolysis or thrombectomy to maximise patient outcomes and prevent stroke recurrence.
In the future, an improved understanding of clot composition may become more important for making informed treatment decisions in ischaemic stroke.For thrombolysis, fibrin network arrangement is a potential prognostic marker, and patients with denser clots could possibly require earlier administration of tPA due to the slower lysis rate exhibited.However, this would need rapid assessment of clot porosity which with current laboratory measurements is not possible.Thrombus stiffness, regulated by the relative contents of RBCs, platelets and fibrin, may indicate the patients in whom thrombectomy is most indicated or where thrombolysis may need to be given in conjunction.Further research will be needed to help create mechanical reperfusion devices that are better suited to heterogeneous thrombus structure/composition and expand on how it influences outcomes in studies using more standardized methods (Yoo and Andersson, 2017;Staessens et al., 2020).Meanwhile, the development of novel targeted treatments offers an exciting new approach to stroke management that may help overcome its current limitations, carrying a reduced risk of adverse events such as intracranial haemorrhage and neurological deficit despite successful recanalization.

Neutrophil extracellular traps and clot structure
Neutrophil Extracellular Traps (NETs) are an important factor that may influence clot structure and treatment efficacy in stroke.NETs are web-like histone DNA complexes that are ejected from neutrophils, typically as part of the innate immune response to invasive pathogens/inflammation, although their additional role in the coagulation pathway and as a prothrombotic agent, especially when in excess, has been demonstrated (Fuchs et al., 2010).NETs provide a DNA scaffold for platelets, RBCs and pro-coagulant plasma proteins, such as vWF and fibronectin, to accumulate and adhere to during blood clot formation, acting independently of fibrin (Fuchs et al., 2010;Darbousset et al., 2012;Thålin et al., 2019).Their significance in stroke is underpinned by ex vivo studies that have found NETs to be present in all stroke thrombi, regardless of aetiology, though particularly abundant in those of cardio-embolic origin (Laridan et al., 2017;Genchi et al., 2021).There is also evidence that biomarkers of circulating NETs, such as citrullinated histone H3 (H3Cit), are important for predicting the severity and outcome for acute ischaemic stroke (Vallés et al., 2017;Novotny et al., 2020;Zhang et al., 2021).
There are important associations between NETs and clot structure and function.Longstaff et al. (2013) showed that addition of histones resulted in thicker fibrin fibres and an overall more rigid, stable clot with a slower lysis time.On the other hand, the addition of DNA created clots that were weaker, looser, and softer, with a marginally slower lysis rate than fibrin only clots.However, the combination of DNA and histones resulted in the greatest increase in median fibre diameter, creating a synergistic effect on lysis time, clot stability and clot strength.The authors hypothesised that this could be due to both DNA and histones binding to fibrin degradation products, thus preventing clot dissolution rather than inhibiting lysis onset.A subsequent study also observed similar effects, with networks containing DNA and histones showing a greater median fibre diameter compared to those without DNA or histones as well as reduced permeability, thus slowing tPA-mediated fibrinolysis (Varjú et al., 2015).
Alterations in fibrin clot properties have been further validated by confocal microscopy, showing that clot areas surrounding NETs have increased density while those without NETs have larger pores and reduced stability (Shi et al., 2021).Interestingly, this effect differed from that observed in neutrophil-induced clots, which instead showed altered permeability, not densityresulting in weak, easily ruptured clots -mediated through FXI and FXII activation.The inclusion of both NETs and neutrophils in plasma clots significantly slowed lysis time compared with thrombin-only controls.Further studies have implicated that changes in clot structure related to NETosis may reduce success of Thrombectomy, with a greater NET content associated with longer procedure length, a greater number of passes and, ultimately, poorer outcomes (Fig. 1) (Ducroux et al., 2018;Novotny et al., 2020).
Thus, the evaluation of NETs components within a clot, e.g.via the use of biomarkers such as H3Cit (indicative of NETosis) or through immunohistochemical staining, may be an important future prognostic tool to inform stroke outcomes and treatment.It has been shown that clots resistant to tPA demonstrate an enduring NET-induced DNA scaffold, that continues to hold clot components together, independent of fibrin (Fuchs et al., 2010).This scaffold is also associated with higher vWF, PAI-1 and microcalcifications contributions, which are known to increase lysis resistance (Fig. 1) (Staessens et al., 2021;Zhang et al., 2021).DNase 1 and Heparin have both shown promise in breaking down clots (Longstaff et al., 2013) thus further research into the role of NETs may indicate that for some patients, treatment with a combination of tPA and Dnase/Heparin may improve lysis efficacy and outcome (Ducroux et al., 2018).

Clot contraction, clot structure and stroke
Clot contraction is an important event occurring after initial clot formation during haemostasis.This process is driven by the pulling of activated platelets on the fibrin fibres, which helps to stabilize the clot at the site of injury to allow subsequent tissue repair.Clot contraction also occurs during thrombosis; its role remains less well understood, but reduction in the size of the blood clot is thought to improve vessel patency, thus blood flow, past thrombi (Cines et al., 2014).The molecular machinery involved in clot contraction involves cytoplasmic motility proteins within the thrombin-activated platelets as the initiators of clot retraction, key among them being (non-muscle) myosin IIA, which generates an ATP-mediated contractile force by pulling on actin filaments (Niederman Pollard, 1975;Litvinov and Weisel, 2022).This is conducted throughout the clot via platelet filopodia and interactions with fibrin mediated by interaction with integrin αIIbβ3, resulting in the expulsion of serum and water, rearrangement of platelets and fibrin, as well as erythrocyte compaction, concurrently with, or shortly after, the formation of the network itself (Fong et al., 2021;Litvinov and Weisel, 2022).In pro-thrombotic diseases such as stroke, this process becomes impaired due to deranged platelet function associated with exhaustion of the activation mechanisms, which impacts the extent of clot contraction achieved and the overall strength and size of the clot.(Jurk et al., 2004;Tutwiler et al., 2016).
Several studies have analysed the role of clot contraction in ischaemic stroke pathogenesis and its influence on clinical outcome.In the first systematic assessment of this, Tutwiler et al. (2017) utilised automated optical tracking to calculate the contraction rate, extent and initiation time of clots contracting in vitro.They found that on average, samples from patients with acute ischaemic stroke displayed 60% reduction in clot contraction compared to healthy controls.The reduced clot contraction led to larger size of the clots, which correlated with a reduced number and functioning of platelets, as well as raised fibrinogen and haematocrit levels.Interestingly, however, samples from patients with severe stroke showed relatively greater contraction than those with mild-moderate stroke, as did samples from patients with strokes of atherosclerotic aetiology in comparison to cardioembolic.These differences may be attributed, in part, to more extensive brain ischaemia in severe strokes causing increased tissue factor exposure, heightened coagulation activation and greater clot contraction (Vecht et al., 1975), while aetiology-based differences may be due to contrasting clot composition and pathogenesis.Further studies have shown similarly reduced contraction in patients with ischaemic stroke, associated with weaker thrombi that are more prone to rupture and embolization (Peshkova et al., 2018;Fong et al., 2021), thus more likely to cause a thromboembolic event (Fig. 1) (Carroll et al., 1981).
Treatment efficacy in stroke has also been associated with the extent of clot contraction, with studies showing that external fibrinolysis is inhibited by clot compactionas thrombolytic agents, such as tPA, are unable to penetrate the tightly packed fibrin-platelet shell of the newly rearranged clot structure (Blinc et al., 1992;Kunitada et al., 1992;Samson et al., 2017;Staessens et al., 2020).A 3D mathematical modelling and experimental validation study identified peripheral fibrin densification as the individual structural change during clot contraction that had the greatest effect on tPA diffusion (Risman et al., 2022).The authors' findings suggest the existence of a fibrin-density boundary, beyond which entry is critically impeded, with densely packed fibrin resulting in peripheral tPA binding that slows deeper progression into the clot.In contrast, internal fibrinolysis is accelerated in contracted clots, with an in-vitro study recording an approximate 2-fold increase in the average rate of internal lysis after contraction (Tutwiler et al., 2019) (Fig. 1).It is hypothesized that this may be due to the fibrin fibres being closer together, allowing plasmin and other fibrinolytic agents to move between them more efficiently, while the reduction in clot volume may also increase the speed at which plasmin is generated (Carr, 2003).Therefore, the generally less contracted nature of stroke thrombi may contribute to their persistence and stability by making them resistant to the body's natural clot degradation methods (Litvinov and Weisel, 2022).
Clot contraction remains an understudied mechanism, despite its importance in shaping clot properties/architecture, such as fibrin density, which are known to be significant in clinical outcomes.Further in vivo research, examining the effect of contraction on all clot components, is needed to help determine whether it is ultimately a protective or pathogenic process in strokeas contraction may reduce vessel occlusion but also create a denser thrombus, that is more resistant to thrombolysis.An improved understanding could help ascertain which patients are less likely to benefit from therapeutic external fibrinolysis, while the finding that the ratio of polyhedrocytes to normal RBCs within clots is exponentially related to contraction indicates that a potential marker for this may already exist (Khismatullin et al., 2022).Moreover, further appreciation of the impact of changes in blood constituents, particularly platelets, on clot contraction may suggest future possibilities of clot contraction assays that could act as an early warning of an increased thrombotic risk (Tutwiler et al., 2016;Litvinov and Weisel, 2022).

Conclusion
Our understanding and knowledge of clot architecture and composition in ischaemic stroke is rapidly expanding as illustrated in this review.In vitro clot structure in plasma samples from patients with stroke consistently shows dense fibrin clot networks that are resistant to fibrinolysis.The increasing understanding of the composition of thrombi obtained from patients through thrombectomy is still, however, rudimentary.Some thrombi may be more RBC-rich and others platelet-rich, with a variable content of fibrin and other components.There are frequent reports of the contribution of NETs in stroke thrombi, and we are beginning to understand how clot contraction (also called clot retraction) impacts on thrombosis risk and on clot removal by thrombolysis and thrombectomy.However, current available data still only provides a heterogeneous picture, and clear correlations with clinical diagnosis and treatment options are often lacking.That said, it will be a matter of time until future studies are able to deliver consistent data that may be useful for clinical management of patients with ischaemic stroke.Moreover, future advances in methodologies, e.g. in devices for thrombectomy, methods for ex vivo analysis of clots and in vivo imaging of thrombus structure and composition, can be expected to deliver clear future patient and healthcare benefits based on an increased understanding of the nature of the individual blood clots that cause ischaemic stroke.