Fibrinogen Level Predicts Prognosis of Patients with Acute Ischemic Stroke or TIA

Background: Fibrinogen is involved in acute stroke. This study aimed to investigate the association between brinogen and prognosis in patients with acute ischemic stroke or transient ischemic attack (TIA). Methods: Using data from the CNSR-(cid:0) (Third China National Stroke Registry), this sub-study included 10 518 (69%) consecutive patients who had brinogen levels measured. The primary outcome was a poor functional outcome dened as modied Rankin Scale score of 3 to 6 within 90 days. The secondary outcomes were stroke recurrence, ischemic stroke recurrence, composite vascular events, and poor functional outcome during the 1-year follow-up and a new vascular event at 90 days. Multivariate logistic regression and Cox regression analyses were used to assess the associations between brinogen and prognosis of patients. Results: In total, 1446 (13.9%) patients had a poor functional outcome at 90 days. High brinogen levels were associated with poor functional outcome (adjusted odds ratio [OR], 1.35; 95% condence interval [CI], 1.12-1.64) at 90 days after adjustment for confounding risk factors. High brinogen levels also independently predicted poor functional outcome during the 1-year follow-up. Stroke recurrence occurred in 657 (6.3%) patients at 90 days. High brinogen levels were associated with stroke recurrence, ischemic stroke recurrence, and composite vascular events in the crude model, but further adjustment eliminated these associations in the multivariate models. Conclusion: Our study showed that high brinogen level was independently associated with poor functional outcome but not with stroke recurrence in patients with acute ischemic stroke or TIA. test for the continuous variables. The brinogen levels were examined continuously and as quartiles to investigate their association with the prognosis of ischemic stroke or TIA. The association between the brinogen levels and poor functional outcome was estimated using logistic regression models, while the association between the brinogen levels and a new vascular event was estimated using Cox proportional hazards models. The potential confounders included demographic factors, prior published clinical risk factors, stroke subtypes, complications during hospitalization, and medications used during hospitalization. All signicant covariates with a P value < 0.05 in the univariate analysis were included to the multivariate model. also evaluated the association between the brinogen level and the risk of poor functional outcome using a logistic regression model with restricted cubic splines for brinogen levels (continuous measures) that was adjusted for potential covariates. The and 95th rst quartile


Background
Acute ischemic stroke and transient ischemic attack (TIA) occur frequently and are major causes of functional disability. Moreover, stroke recurrence within the 90-day follow-up period accounts for 10-15% of these patients [1,2]. Thus, identifying potential markers associated with prognosis of stroke in addition to traditional risk factors and further developing targeted therapeutics are critical.
Fibrinogen is a pleiotropic protein secreted from the liver that plays critical roles in coagulation, the proin ammatory cascade and the inhibition of tissue repair [3]. Growing data indicate brinogen mediates cardiovascular risk through coagulation and in ammatory cascades, which may promote formation of thrombin substrates, platelet aggregation, endothelial activation, and smooth muscle cell proliferation and migration [3,4].
Several large trails suggested that high brinogen level is a risk factor for initial stroke or vascular disease in the general population during long term follow-up period [5][6][7]. A higher brinogen level has been previously observed in patients with acute ischemic stroke [8,9]. However, the association between brinogen and stroke recurrence or a new vascular event is still unde ned [10][11][12]. On the contrary, although the association of high brinogen level with death after stroke has been previously expounded [13][14][15], only a few studies investigated the association between brinogen and functional outcome [16][17][18]. In addition, its predictive value for poor functional outcome after acute stroke is limited and inconsistent [16,17].
The CNSR-(Third China National Stroke Registry) is a nationwide, multicenter, prospective registry of patients with acute ischemic stroke or TIA [19]. In this subgroup study, we aimed to evaluate the association of brinogen levels with poor functional outcome and stroke recurrence.

Study Population
The CNSR-study is a nationwide, consecutive, multicenter, prospective stroke registry that enrolled 15 166 patients with acute ischemic stroke or TIA within 7 days of onset who were treated at 201 hospitals in China from August 2015 to March 2018. Acute ischemic stroke was diagnosed according to the WHO criteria [20] and con rmed by brain computed tomography or magnetic resonance imaging. A subset of 10 518 patients in the CNSR-study provided blood samples for the measurement of brinogen levels. All participating centers used consistent diagnostic criteria and a standard protocol.
Standard protocol approval, registration, and patient consent The study protocol was approved by the Central Institutional Review Board at the Beijing TianTan Hospital. All participants or their legal proxies provided written informed consent.

Biochemical Measurements
EDTA treated plasma and serum samples were obtained from the CNSR-patients at the time of enrollment and stored at -80℃. No freeze thaw cycles occurred before testing. The brinogen, D-dimer and high-sensitive C-reactive protein (hsCRP) levels were measured (from February 2018 to February 2019) in a center laboratory certi ed by the College of American Pathologists, and the laboratory personnel were blinded to the clinical data. The plasma brinogen and D-dimer levels were analyzed with an OLYMPUS AU2700 analyzer (Beckman, Japan) using an immunoturbidimetric assay (Kamiya Biomedical, Seattle, USA) [21][22][23], which is a mass-based assay with international standards. The serum hsCRP levels were measured with a Cobas c501 analyzer using a cardiac high-sensitive C-reactive protein (latex) assay (Roche, Basel, Switzerland). All measurements were performed according to the manufacturers' instructions.

Patient Follow-up and Outcome Assessment
The patients were interviewed face to face at 90 days and contacted by telephone at 1 year after enrollment by a trained research coordinator in CNSR- [19]. Information regarding any end point event, and the functional status was collected at each follow-up interview. Modi ed Rankin Scale (mRS) scores ranging from 0 to 6 points were used to assess patients' functional dependence.
The primary outcome was poor functional outcome as de ned by an mRS score of 3 to 6 at 90 days [24]. The secondary outcomes were stroke recurrence de ned as aggravated primary neurological de cit, a new neurological de cit, rehospitalization with a diagnosis of ischemic or hemorrhagic stroke (intracerebral hemorrhage and subarachnoid hemorrhage), ischemic stroke recurrence, composite vascular events (ischemic stroke, hemorrhagic stroke, myocardial infarction, or vascular death), and poor functional outcome during the 1-year follow-up and a new vascular event at 90 days [24,25].

Statistical Analysis
The baseline characteristics were compared based on the quartiles of the brinogen levels using the χ 2 test for the categorical variables and Kruskal-Wallis test for the continuous variables. The brinogen levels were examined continuously and as quartiles to investigate their association with the prognosis of ischemic stroke or TIA. The association between the brinogen levels and poor functional outcome was estimated using logistic regression models, while the association between the brinogen levels and a new vascular event was estimated using Cox proportional hazards models. The potential confounders included demographic factors, prior published clinical risk factors, stroke subtypes, complications during hospitalization, and medications used during hospitalization. All signi cant covariates with a P value < 0.05 in the univariate analysis were included to the multivariate model. We also evaluated the association between the brinogen level and the risk of poor functional outcome using a logistic regression model with restricted cubic splines for brinogen levels (continuous measures) that was adjusted for all potential covariates. The 5 knots of the spline were placed at the 5th, 25th, 50th, 75th, and 95th percentiles of the brinogen levels, and the rst quartile of the brinogen level was treated as the reference.
All statistical analyses were performed with SAS software, version 9.4 (SAS Institute Inc., Cary, NC). All P values were 2-sided, and P < 0.05 was considered indicative of statistical signi cance.

Patient Characteristics
In total, 15 166 patients with acute ischemic stroke or TIA were enrolled in the CNSR-study. Among these patients, 4648 patients who did not provide blood samples were excluded. A subset of 10 518 patients constituted the study population for this analysis (Fig. 1).
The baseline characteristics of the 10 518 patients strati ed according to the brinogen quartiles are shown in Table 1. The mean age was 62.3 ± 11.4 years, and 3283 (31.21%) patients were female. The median brinogen level was 378.0 (interquartile range [IQR], 315.0-451.5) mg/dl. The patients with higher brinogen levels were signi cantly older, more likely to be female, had a higher baseline National Institutes of Health Stroke Scale (NIHSS) score, and had a history of hypertension, diabetes mellitus, atrial brillation, and ischemic stroke (Table 1).

Fibrinogen and Poor Functional Outcome
In total, 1446 (13.9%) patients had a poor functional outcome at 90 days. In the quartile analyses, high brinogen levels were associated with an increased risk of poor functional outcome (P for trend < 0.001) at 90 days ( Table 2).  Table 3). High brinogen levels also independently predicted poor functional outcome during the 1-year follow-up. Similar results were observed when brinogen was assessed as a continuous variable (Table 3). Using a regression model with a restricted cubic spline, we observed an association between high brinogen levels and an increased risk of poor functional outcome at 90 days and 1 year (Fig. 2).

Fibrinogen and Stroke Recurrence
There were 657 (6.3%) patients with stroke recurrence within 90 days. In the quartile analyses, high brinogen levels were signi cantly associated with an increased risk of 90-day stroke recurrence, ischemic stroke recurrence, and composite vascular events (P for trend = 0.014, = 0.002, and = 0.001, respectively) ( Table 2). After adjustment for confounding risk factors, these associations were no longer signi cant (Table 3).
Furthermore, brinogen levels were not associated with a new vascular disease during the 1-year follow-up. Similar results were observed when brinogen was assessed as a continuous variable (Table 3).

Discussions
In this prospective hospital-based nationwide stroke registry in China, we found that high brinogen levels were associated with increased risks of short-term and long-term poor functional outcome, but not stroke recurrence, ischemic stroke recurrence, and composite vascular events in patients with acute ischemic stroke or TIA.
Although previous studies con rmed that a high brinogen level was associated with rst-ever cardiovascular disease [5,6], the association between brinogen and ischemic stroke is contradictory. A large meta-analysis involving 154 211 participants proved that moderately strong associations were found between plasma brinogen and the risk of rst stroke in healthy middle-aged adults [26]. However, subsequent studies investigating the association between brinogen and initial ischemic stroke showed inconsistent results [7,[27][28][29]. In contrast, only a few studies indicated that brinogen level was associated with stroke recurrence in recent TIA or minor ischemic stroke [12,30] A preliminary study showed that brinogen combined with tissue-type plasminogen activator and von Willebrand factor was associated with coronary events, but not stroke recurrence among patients with recent TIA or ischemic stroke [31]. In addition, previous studies have indicated that the association between brinogen and a new vascular event was no longer signi cant after adjustment for confounding risk factors in patients with acute stroke [10,11]. On the contrary, the association between brinogen and poor functional outcome is less established. It was previously suggested that brinogen was associated with poor functional outcome after adjusted simple risk factors, but further adjustment for potential factors eliminated this association [32]. Previous studies suggested that patients with lower initial brinogen levels had good functional outcome after acute stroke, indicating that the independent association between brinogen levels and functional outcomes need to be veri ed using a larger acute stroke dataset [16]. In the present study, the associations of brinogen with poor functional outcome and stroke recurrence were assessed base on a large-scale stroke registry study in China. Furthermore, D-dimer, hsCRP, and complications during hospitalization were considered in addition to the baseline NIHSS score and stroke subtype according to the Trial of Org 10172 in Acute Stroke Treatment.
In our analysis, high brinogen levels were associated with an increased risk of poor functional outcome. This association was independent of the stroke severity, stroke subtype, complications, D-dimer, and hsCRP. Our ndings indicate that this association might depend on other potential mechanisms of brinogen. One possible explanation is that high brinogen level may affect the structure of the brin clot. Previous studies have suggested that high brinogen levels might be related to the formation of more stable brin clots, which determine the e cacy of arterial recanalization [33]. Another potential explanation is that high brinogen levels cause high blood viscosity, which could potentially compromise the microvascular blood ow in marginally perfused brain areas [16]. An additional explanation is that brinogen plays a critical role in interfering with tissue repair [3].Blood-brain barrier permeability increases after ischemia and reperfusion [34], and brinogen enters the brain tissue, thus participating in acute tissue necrosis and causing secondary tissue damage [35,36]. Based on our ndings, high brinogen levels independently predict poor functional outcome, which suggest that brinogen might be a targeted therapeutic after acute ischemic stroke or TIA.
Consistent with previous studies, we observed no association between brinogen and a new vascular event in the multivariate model [11].
Potential explanation might be that brinogen levels were associated with other risk factors. Although brinogen is the most common hemostatic factor, its role in the development of atherosclerosis and vascular disease extends to a key proin ammatory player [3]. The Canakinumab Antiin ammatory Thrombosis Outcome Study showed that antiin ammatory therapy led to the reduction in hsCRP and brinogen, indicating that brinogen had potentially plays dual roles as both an acute-phase reactant and a mediator of coagulation [37]. Previous studies proved that there was a strong association between brinogen and hsCRP [10]. In addition, it has been suggested that hsCRP was more related to the risk of new cardiovascular events after acute ischemic stroke than brinogen [11]. Therefore, our ndings suggest that brinogen might be involved in stroke recurrence through other critical risk factors.
Our study has several limitations. First, our study only included the D-dimer and hsCRP levels in addition to the brinogen levels, while other related coagulation or in ammatory factors may also be associated with stroke outcomes. Future studies focusing on the coagulation or in ammatory pathway could help to reveal the molecular mechanism underlying stroke outcomes. Second, this study obtained one-point brinogen measurements. Since brinogen levels persistently increase from the acute phase to the stable phase in patients after stroke [8,9], dynamic change might be necessary to con rm its predictive value. Third, 4648 (30.6%) of the 15166 patients did not provide blood samples and were excluded because their brinogen levels were not measured. Although the baseline characteristics of the patients included and those excluded were well balanced, some deviation in the analysis of the associations between the brinogen levels and outcomes may exist. Finally, all enrolled patients were Chinese adults with acute ischemic stroke or TIA, and these results may not be generalizable to patients of other ethnicities.

Conclusions
In conclusion, our study show that high brinogen levels were associated with poor functional outcome, but not with stroke recurrence, ischemic stroke recurrence, or composite vascular events. Declarations YW had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Availability of data and materials The corresponding author takes full responsibility for the data, the analyses and interpretation, and the conduct of the research; that he has full access to all of the data.
Ethics approval and consent to participate The study protocol was approved by the Central Institutional Review Board at the Beijing TianTan Hospital. All participants or their legal proxies provided written informed consent.

Consent for publication
Not applicable.