Fibrin-Platelet Clots in Acute Ischemic Stroke. Predictors and Clinical Significance in a Mechanical Thrombectomy Series

Introduction: The histological composition of the clot influences its mechanical properties, affects the efficacy of endovascular treatment (EVT), and could determine the clinical outcome of patients with acute ischemic stroke (AIS). Insights into clot composition may guide therapeutic decision-making prior to EVT and facilitate revascularization therapies. Material and Methods: Consecutive patients with AIS recorded in a prospective single-center reperfusion registry from December 2015 to December 2019 and treated with EVT were included. Baseline, laboratory [including post-procedural C-reactive protein (CRP)], radiological, and angiographic variables were analyzed. We aimed to study the relationship between histological composition of the clot with basal neuroimaging, laboratory markers, and recanalization technique. The secondary outcome was to analyze the correlation between clot composition and functional outcome at 3 months assessed by the modified Rankin scale (mRS). Results: From the study period, 360 AIS patients treated with EVT were included, of whom 189 (53%) fulfilled the inclusion criteria. One hundred (53%) cases of fibrin-predominant clot (FPC) were recorded. Full recanalization in FPC cases was achieved with higher probability when stent retrievers (SR) were selected as the first-line device (68.2%, p = 0.039). Patients with FPC had higher levels of CRP (p = 0.02), lower frequency of the hyperdense middle cerebral artery (HMCA) in baseline imaging (p = 0.039), and higher rates of mortality (p = 0.012). The multivariate analysis showed that the absence of HMCA (OR = 0.420; 95% CI 0.197–0.898; p = 0.025) and higher levels of CRP (OR = 1.01; 95% CI 1.003–1.019; p = 0.008) were predictors of FPC. Leukocytes and platelet counts were not associated with clot histology. Conclusions: The absence of HMCA and higher levels of CRP were markers of FPC. In patients with FPC, complete recanalization was most likely to be achieved when a SR was selected as first line of treatment. Mortality was higher in patients within this histologic group.


Introduction:
The histological composition of the clot influences its mechanical properties, affects the efficacy of endovascular treatment (EVT), and could determine the clinical outcome of patients with acute ischemic stroke (AIS). Insights into clot composition may guide therapeutic decision-making prior to EVT and facilitate revascularization therapies.
Material and Methods: Consecutive patients with AIS recorded in a prospective single-center reperfusion registry from December 2015 to December 2019 and treated with EVT were included. Baseline, laboratory [including post-procedural C-reactive protein (CRP)], radiological, and angiographic variables were analyzed. We aimed to study the relationship between histological composition of the clot with basal neuroimaging, laboratory markers, and recanalization technique. The secondary outcome was to analyze the correlation between clot composition and functional outcome at 3 months assessed by the modified Rankin scale (mRS).
Results: From the study period, 360 AIS patients treated with EVT were included, of whom 189 (53%) fulfilled the inclusion criteria. One hundred (53%) cases of fibrin-predominant clot (FPC) were recorded. Full recanalization in FPC cases was achieved with higher probability when stent retrievers (SR) were selected as the first-line device (68.2%, p = 0.039). Patients with FPC had higher levels of CRP (p = 0.02), lower frequency of the hyperdense middle cerebral artery (HMCA) in baseline imaging (p = 0.039), and higher rates of mortality (p = 0.012). The multivariate analysis showed that the absence of HMCA (OR = 0.420; 95% CI 0.197-0.898; p = 0.025) and higher levels of CRP (OR = 1.01; 95% CI 1.003-1.019; p = 0.008) were predictors of FPC. Leukocytes and platelet counts were not associated with clot histology.

INTRODUCTION
Endovascular treatment (EVT) has become the standard of care in patients with acute ischemic stroke (AIS) and largevessel occlusion (1). Complete recanalization and first-pass recanalization have been shown to be independent predictors of good functional outcome (2)(3)(4). However, these endpoints are currently achieved in <50% of patients, probably due to inadequate patient selection or factors related to the procedure (5).
One of the main reasons for the lack of efficacy of EVT is the scant knowledge of the mechanical properties of the clot (6)(7)(8) and its histological composition. Both factors are relevant to attainment of successful reperfusion (9) and could aid in the development of endovascular thrombectomy devices.
Based on the predominant cell type, clots are usually divided into red clots (RPC), white or fibrin-rich clots (FPC), and mixed clots (MC) (10). In addition, the distribution of the fibrin network (6) and platelet clumps can be highly variable (11). Therefore, some FPC are more resistant to revascularization than others due to their higher friction coefficient (7,8,12). On the other hand, RPC demonstrate a lower density and are more easily removed as a result. A better understanding of the histological composition of the clot could facilitate better clinical and angiographic outcomes.
We aimed to study the relationship between histological composition of the clot with basal neuroimaging, laboratory markers, and recanalization strategies including intravenous thrombolysis and clot retrieval devices. The secondary outcome was to analyze the correlation between clot composition and functional outcome at 3 months as assessed by the modified Rankin scale (mRS).

Study Design and Patient Selection
We performed a retrospective, observational, unicentric study based on a prospective registry of all consecutive AIS patients treated with reperfusion therapies in a tertiary stroke center, between 15 December 2014 and 15 December 2019. All patients included were clinically managed according to our institutional protocols, which are based on current international stroke guidelines.
The study was approved by the local Clinical Research Ethics Committee with reference number 2019/03/031. Written informed consent was obtained from all included patients or their relatives permitting entry of their information into our reperfusion registry and subsequent use of the data for scientific purposes, in accordance with the Spanish Personal Data Protection law.
Patients treated with EVT in our center were selected for this study if they fulfilled the following criteria: (1) substantial neurological deficit and ischemic stroke with demonstrated vessel occlusion; (2) time from symptom onset to groin puncture <24 h, including wake-up strokes and those of unknown onset; (3) no intracranial hemorrhage at baseline cranial tomography (CT); (4) no extensive early ischemic signs as defined by ASPECTS (Alberta Stroke Program Early CT Score) > 5; (5) in patients with time from symptom onset to EVT > 6 h, presence of target mismatch profile on CT perfusion; (6) occlusion in anterior circulation and absence of multiple-territory involvement; (7) effective recanalization (TICI ≥2B) and thrombus for histological analysis; and (8) absence of previous relevant disability as defined by the modified Rankin scale (mRs) (pre-stroke mRs score ≤2). If no contraindication existed, previous treatment with intravenous recombinant tissue-type plasminogen activator (IV r-tPA) was administered before EVT (Figure 1). etiology of the stroke based on the TOAST (trial of ORG 10172 in acute stroke treatment) classification (13) were recorded. Information regarding clinical stroke severity was assessed by the NIHSS (National Institute of Health Stroke Scale) on admission and during hospitalization. Urgent (blood count, biochemistry, and coagulation) and in-hospital laboratory tests were performed after EVT [biochemistry including thyroid hormones, lipid profile, and C-reactive protein (CRP)]. The white cell count was detailed, and the ratio of neutrophil and lymphocyte numbers was calculated.
Times from symptom onset to the CT, onset to groin puncture and procedural times (time from groin puncture to the end of the endovascular procedure), and type of device used (SR, aspiration, or combination) were registered. Number of passes and degree of recanalization by TICI (Thrombolysis in Cerebral Infarction) scale were obtained; TICI ≥ 2B was considered as effective recanalization. Prognosis at 3 months was evaluated by mRS considering good prognosis as mRS ≤ 2 and mortality mRS = 6.

Neuroimaging Protocol
A basal cranial non-contrast CT was performed before EVT. Data on early ischemia was evaluated on basal CT by ASPECTS. Arterial occlusion was assessed on angio-CT scan and mismatch by CT perfusion.
All CT imaging was performed using a Philips Brilliance CT, 64-slice (Koninklijke Philips Electronics N.V., Amsterdam, Netherlands), and images were analyzed by a neuroradiologist.
At 24 h from the procedure or in the case of neurological deterioration, a control CT was performed to evaluate the presence of established infarct or hemorrhagic transformation. Symptomatic intracranial hemorrhage (PH-2) was defined according to Heidelberg criteria (14) as any intraparenchymal hemorrhage associated with an increase in the NIHSS scale score ≥ 4 points or death.
All baseline CTs were reviewed retrospectively on a blind basis by two independent neuroradiologists. The presence or absence of the hyperdense middle cerebral artery (HMCA) sign was classified by consensus based on visual inspection and according to the following criteria: (15) (1) spontaneous visibility of the MCA, (2) attenuation of the MCA greater than in the surrounding parenchyma, (3) disappearance in the bone window, (4) unilaterality, and (5) absence of subarachnoid bleeding. In case of disagreement, the case was discussed until a consensus was reached. Hounsfield units were also measured, and the ratio of healthy to pathological radiodensities was also calculated. The final value was considered as the arithmetic mean of the values obtained by each observer.

Endovascular Procedures
The endovascular procedures were performed using a singleplane angiography, Innova GE model (General Electric Company, Schenectady, NY, US) or an Azurion 7 B20/15 biplane model (Koninklijke Philips Electronics N.V., Amsterdam, Netherlands) by one board-certified neurologist and a neuroradiologist. General anesthesia was used systematically. or after SR failure following two passes, an ACE64 or ACE68 (Penumbra Inc, Alameda, CA, USA) aspiration catheter is used, connected to an automatic pump for 3 min. Removal of the device and simultaneous manual aspiration with a 60-cc. syringe. In case the outcome was not satisfactory after two aspirations, combined use of SR and local aspiration is undertaken. After each pass, the device, intermediate catheter if withdrawn, and aspiration syringe were inspected for the presence of thrombus fragments. The device was gently washed with heparinized saline to remove thrombus fragments. Aspirated material was gently flushed with saline to identify any smaller fragments.

5) Control angiographic series in anteroposterior and lateral
projections. Determination of the TICI scale.

Histopathological Analysis
The clot was stored in 4% formaldehyde and sent to the Pathology Department, where it was embedded in paraffin, microtomed into 4-µ sections, and prepared using the established histological protocol, with hematoxylin-eosin, Gram, and Gomori's Trichromic staining. The histopathological study was performed by an expert neuropathologist, who analyzed the macro and microscopic characteristics of the clot, including red cell, white cell, and fibrin distribution. The semiquantitative percentage of these components was estimated, considering a clot either FPC or RPC when the percentage of these components was >60% (16)(17)(18). The rest of the clots, with

Primary and Secondary Outcomes
The main outcome was to study the relationship between histological composition of the clot with basal neuroimaging, laboratory markers, and recanalization strategies including intravenous thrombolysis and clot retrieval devices. The secondary outcome was to analyze the correlation between clot composition and functional outcome and mortality at 3 months assessed by the modified Rankin scale (mRS).

Statistical Analysis
A descriptive study of the variables was performed using central and scattering tendency measures for quantitative variables and the exact calculation and percentage for qualitative variables.  ROC (receiver operating characteristic) curve was performed to analyze the cutoff point with the best discriminatory value of the quantitative variables. A p < 0.05 was considered significant for all analyses. All the results were analyzed with the statistical software IBM SPSS Statistics R Version 26 (SPSS, Chicago, IL, USA).

Primary Outcome
Bivariate Analysis of FPC FPC patients were more frequently under oral anticoagulation (p = 0.042) and had higher levels of CRP (p = 0.02). On the contrary, they received less IV thrombolysis (p = 0.047) compared with the rest of the cohort and had lower frequency of HMCA in baseline CT (p = 0.039). FPC patients had higher rates of mortality (p = 0.012) ( Table 1).

Bivariate Analysis of Patients Treated With IV Thrombolysis
Seventy-three (39%) of the patients received IV thrombolysis bridging EVT. Thirty-two (44%) had FPC. In a bivariate analysis ( Table 2), patients with FPC treated with IV thrombolysis obtained higher rates of complete recanalization compared to the rest of the patients (p = 0.005).

First-Line Devices in Patients With Fibrin-Predominant Clot (FPC)
For patients with FPC, SR were used as the first-line option in 85 cases (85%), whereas direct aspiration and combined use of both devices were used in 13 (13%) and 2 (2%), respectively. First-pass recanalization was achieved in 50 patients (58.8%) treated with SR as first-line device, six cases with direct aspiration (46.2%), and one case with the combination (50%). No significant differences were found regarding first-pass recanalization. However, SR as the device of choice were more successful in achieving full recanalization (68.2%, p = 0.039) compared to direct aspiration (46.2%) or a combination of both (0%) ( Table 3).
The ROC curve did not disclose any significant association between levels of CRP after the EVT and the histological composition of the clots (area under the curve 0.6, CI 0.5-0.7, p = 0.18).

Bivariate Analysis of Functional Outcome and Mortality at 3 Months
In bivariate analysis, higher baseline NIHSS and lower ASPECTS on admission were related to poor functional outcome. Higher levels of baseline glucose and CRP, tandem occlusion, TICI 2B recanalization, and PH-2 were also associated with mRS > 2, as shown in Table 5.
The absolute number of neutrophils, lymphocytes, and their ratio (neutrophils/lymphocytes) were significantly associated with prognosis at 3 months, while a significant relationship was detected between the absolute number of lymphocytes and mortality ( Table 5). None of the cell series analyzed were significantly related to the predominant histological composition ( Table 1).

DISCUSSION
Our series provided a substantial amount of histopathological data about heterogeneous clot samples, as well as clinical, laboratory, neuroimaging, and angiographic information, which allowed us to perform several comparisons between histological subtypes of the thrombus with clinical, laboratory, and radiological variables and recanalization strategies. Our main finding was the association between patients with FPC and absence of HMCA in their baseline CT scan. In line with our results, a systematic review grouping neuroimaging studies by histological characteristics demonstrated that patients with signs of HCMA on the baseline CT were more likely to have RPC (OR 9.0, 95% CI 2.6-31.2, p < 0.01) (17), but not FPC.
We also detected a significant association between levels of CRP after EVT and the presence of FPC. Other studies looking at coronary artery disease (19) and deep venous thrombosis (20) have shown that CRP-level increases were associated with FPC formation. A study performed in patients with acute coronary syndrome (21) has shown a strong association between CRP levels and resistance to FPC lysis, so acute-phase reactants could act as platelet aggregation and clot hardness markers. With regard to the existing association with CRP levels, it must be clarified that in our series these analyses were obtained immediately after the procedures, so it cannot be considered as a baseline biomarker of the patients' condition. No significant association was found between leukocyte count and FPC. Consistent with previous studies (22), higher neutrophil counts and relative neutrophilia on admission were associated with bad functional outcome, while relative lymphocytosis corresponded with higher mortality.
To evaluate the effect of IV thrombolysis, we performed a subanalysis in patients with FPC and we found that these patients had higher complete recanalization rates. These data suggest a synergistic effect of the IV thrombolysis bridging EVT in patients with FPC and reinforce the indication of current guidelines giving IV thrombolysis if no contraindication exists (23).
In our study, FPC patients had higher mortality rates compared to the rest of the cohort; these patients were more frequently under oral anticoagulation and received less intravenous thrombolysis as expected. Regarding thrombectomy devices, patients with FPC achieved full recanalization (TICI   Frontiers in Neurology | www.frontiersin.org 3) with higher probability when SR were selected as firstline treatment.
There are limited studies that have taken into consideration thrombus composition when comparatively analyzing the efficacy of the main recanalization strategies (stent retrieval or aspiration). In particular, the higher coefficient of friction of FPC may increase their resistance to direct suction and may require a higher number of contact points for complete extraction. In line with our results, an experimental study comparing direct aspiration to SR (24), white clots were retrieved by the SR and balloon guide catheter with fewer distal emboli.
The study of the histopathological and mechanical properties of the clot should be part of the strategy for the development of new EVT endovascular devices. Behavior of clots is currently analyzed using experimental models in vitro (8,25), thus trying to obtain better recanalization rates (12). In the clinical setting, the potential to establish predictors of the histological composition of the clots could optimize the selection of reperfusion strategies, as well as extend the use of endovascular devices with a higher efficacy in case of resistant clots.
The fibrin structure present in a clot is a critical element that modifies its mechanical behavior, as it gives it properties of hardness (26), impenetrability (7), and friction, which involve greater technical difficulty (12), lower vascular recanalization rates, and treatment delay (18,27).
Despite this evidence, uncertainty persists regarding the relationship between clot composition, angiographic outcome, and stroke etiology (17), so the practical value of obtaining the histopathological characteristics of the clot in acute stroke remains unclear.
On the other hand, the analysis of the neuroimaging prior to reperfusion treatment allows the identification of recanalization standards, by means of qualitative signs such as HMCA on the baseline CT (28), or quantitative signs, such as clot permeability obtained by HU measurement (29). However, it has not been established if there are neuroimaging models which allow us to infer the histological composition of the clot, using CT (6,9), and with magnetic resonance imaging (MRI) (30). As in previous studies (10), we have not been able to establish a direct association between HU determination and the predominant fibrin composition.
This study has some limitations. First, it is a single-center, retrospective series, so the results could be affected by various biases. The determination of a cutoff point of 60% to discriminate the predominance of a histologic subtype by a semiquantitative method makes it difficult to neatly identify subgroups and may alter the analysis. Although we have used several observers to minimize variability in the analysis of images, precision could have been greater if we had recorded the kappa coefficient. CRP was measured only in 78% of the patients and once the EVT was completed, so it cannot be categorized as a baseline biomarker. The ultrasensitive CRP systematic analysis before reperfusion treatment might contribute toward the elucidation of this. We have only included patients with single anterior circulation occlusion and with high recanalization rates, so the sample may have been excessively selected, losing sensitivity in some of the known prognostic variables (as first-pass effect). The higher percentage of anticoagulated patients existing in the FPC group could condition the administration of IV thrombolysis, limiting the scope of the results.

CONCLUSIONS
The absence of HMCA on baseline CT and CRP levels appear to be a reliable marker of FPC, which could allow an individualized treatment approach such as prior administration of intravenous thrombolysis and the use of specific thrombectomy devices. In our series, the choice of a SR as first-line treatment in FPC cases was more effective in achieving complete recanalization than aspiration devices. Histopathological study and identification of treatment-refractory clots improve patient characterization and could guide therapeutic strategies for recanalization.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by CEIm of University Hospital of Albacete. The patients/participants provided their written informed consent to participate in this study.