The coagulation and fibrinolysis systems are pivotal in the pathophysiology of stroke. Detecting related factors is vital for the early diagnosis and treatment of stroke. Biomarkers such as thrombin-antithrombin complex (TAT), tissue plasminogen activator inhibitor complex (t-PAIC), activated partial thromboplastin time (APTT), prothrombin time (PT), fibrinogen (FIB), D-dimer, and fibrin degradation products (FDP) reflect dynamic changes in thrombosis and fibrinolytic activity, aiding clinical decision-making. Elevated D-dimer levels indicate active fibrinolysis, commonly used to assess thrombus burden and prognosis in stroke patients. Ohara et al. highlighted that serum D-dimer assists in diagnosing cryptogenic stroke and secondary prevention, with continuous monitoring enhancing the efficacy of antithrombotic treatment in cryptogenic stroke (18). Lee et al., through proteomics, identified four candidate biomarkers—prothrombin, plasminogen, fibrinogen alpha chain, and histidine-rich glycoprotein—with AUC values over 0.9, confirming their diagnostic value related to coagulation mechanisms (19). Misra et al. used SWATH-MS-based proteomics to identify ceruloplasmin, α-1-antitrypsin (SERPINA1), von Willebrand factor (vWF), and coagulation factor XIII B chain (F13B) as effective biomarkers for distinguishing total stroke, ischemic stroke, and intracerebral hemorrhage (ICH) from healthy controls (20). Bioinformatics suggested common pathways in stroke cases, including complement and coagulation cascades, platelet degranulation, immune processes, and acute phase reactions (21).

Endothelial dysfunction can lead to dysregulation of vascular endothelial cell function, further exacerbating cerebral blood flow reduction and tissue damage.

C-reactive protein (CRP), as an acute-phase protein, serves not only as an indicator of systemic inflammation but also as a crucial biomarker for evaluating prognosis post-stroke. Elevated CRP levels are associated with poor outcomes in stroke patients and reflect the systemic inflammatory state (35). IL-6 plays a key role in immune regulation within the central nervous system, while TNF-α exacerbates brain tissue damage by inducing apoptosis and promoting inflammatory responses (36).

Lasek-Bal et al. reported that IL-6 levels on the first day after stroke could predict acute neurological and functional status, while increased CRP and leukocyte counts were associated with worse acute stroke prognosis (37). Reiche et al. studied two composite indices reflecting inflammation levels—INFLAM Index 1 (comprising the z-scores of hsCRP, IL-6, ferritin, ESR, and WBC) and INFLAM Index 2 (derived by subtracting the z-score of 25(OH)D from INFLAM Index 1 and adding the z-scores of iron and TSP). These indices demonstrated significant predictive value for AIS in both healthy volunteers and AIS patients, with AUC values of 0.851 and 0.870, respectively. They also identified redox imbalance related to IL-6 signaling as a potential target for preventing short-term mortality in AIS (38). Ma et al. developed diagnostic and prognostic models for ischemic stroke using inflammatory markers such as hs-CRP, IL-6, and TNF-α, which were validated in additional cohorts (39).

Matrix metalloproteinase-9 (MMP-9) plays a crucial role in degrading components of the extracellular matrix, activating pro-inflammatory cytokines, and compromising the integrity of the blood–brain barrier (BBB). The activation of M1-polarized microglia has been shown to upregulate MMP-9 expression, leading to BBB disruption and ischemic brain injury (40). Abdelnaseer et al. reported that serum MMP-9 levels within 24 h of stroke onset were significantly correlated with clinical stroke severity (41). Similarly, Weekman et al. demonstrated a positive association between MMP-9 levels and infarct volume, with the strongest correlation observed within the first 6 hours post-stroke. Notably, MMP-9 is considered the only biomarker capable of precisely predicting the final infarct volume, where higher MMP-9 expression is linked to larger infarct areas (42). Further research by Zhong et al. revealed that elevated serum MMP-9 levels in the acute phase of ischemic stroke were positively correlated with mortality and severe disability within 3 months post-stroke (43). These findings suggest that targeted inhibition of MMP-9 activity may serve as a promising therapeutic strategy to mitigate brain injury and improve stroke prognosis (44).

High mobility group box 1 (HMGB1) has been identified as a potential diagnostic and prognostic biomarker for ischemic stroke (45). A study by Tsukagawa demonstrated that serum and plasma HMGB1 levels were significantly elevated in patients with ischemic stroke (46). Moreover, Sapojnikova et al. reported a strong correlation between MMP-9 and HMGB1 levels in stroke patients, with both biomarkers closely associated with poor prognosis (47). Similarly, Shen et al. found that elevated serum HMGB1 levels served as a reliable predictor of AIS recurrence (48). However, it is important to note that HMGB1 exhibits a complex biphasic role in the pathogenesis and progression of ischemic stroke. In the hyperacute and acute phases (within 4–5 days post-stroke), HMGB1 functions as a pro-inflammatory mediator, exacerbating neuronal death and blood–brain barrier disruption. Conversely, in the late acute, subacute, and chronic phases (>3 weeks post-stroke), HMGB1 contributes to vascular remodeling and neurofunctional recovery (49).

Ischemic stroke leads to increased blood–brain barrier permeability and the release of neuronal and axonal injury biomarkers, such as GFAP, neurofilament light chain protein (NFL), and S100 proteins. These markers are rapidly released into the blood following brain tissue injury, with their levels accurately reflecting the extent of brain damage and providing crucial prognostic information. Luger et al. conducted a study to assess the diagnostic accuracy of serum GFAP and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1) concentrations, measured using ELISA, in differentiating between acute cerebral hemorrhage and ischemic stroke. The results indicated that the area under the curve (AUC) for GFAP was 0.866, surpassing the 0.590 AUC for UCH-L1 (21). Bustamante et al. validated a panel of blood biomarkers, including RBP-4, NT-proBNP, and GFAP, which distinguished IS from ICH with moderate accuracy at 100% specificity (50). Kalra et al. tested the diagnostic accuracy of GFAP in a prospective cohort of stroke patients in India. Using the highly sensitive SIMOA technology, GFAP concentrations were measured within 12 h of admission in acute stroke patients. ROC analysis identified an optimal GFAP threshold of 0.57 μg/L for distinguishing intracerebral hemorrhage from ischemic stroke and stroke mimic conditions (AUC 0.871 [95% CI 0.810–0.933], p < 0.001) (51). Recently, a systematic review confirmed the high diagnostic accuracy of blood GFAP levels as a discriminative test for cerebral hemorrhage and ischemic stroke (52). Additionally, Lee et al. developed a time-resolved fluorescence lateral flow immunoassay (TRF-LFIA) utilizing europium nanoparticle (EuNP)-conjugated specific monoclonal antibodies targeting NT-proBNP and GFAP for simultaneous quantification. The combination of GFAP and NT-proBNP was determined to be the most effective biomarker pair for differentiating IS from HS based on an algorithm (53). In summary, the use of blood biomarkers, particularly GFAP, holds promise for the diagnosis and differentiation of ischemic stroke and cerebral hemorrhage. Further research is necessary to validate these findings and establish standardized protocols for measuring acute stroke biomarkers.

Neurofilament light chain (NfL) levels have been shown to be a robust biomarker in cerebrospinal fluid (CSF) for neuronal damage and neurodegeneration. Several studies have demonstrated the potential of NfL as a blood biomarker for ischemic stroke. Sanchez et al. conducted a systematic review and meta-analysis, including 19 studies that reported serum/plasma NfL values from a total of 4,237 different stroke patients, to evaluate the utility of blood NfL as a diagnostic, prognostic, and monitoring biomarker for stroke. They found that blood NfL levels varied significantly across three different time periods: acute (0–7 days), subacute (9–90 days), and chronic (>90 days) phases of stroke, with a sharp peak observed in the early subacute phase, 14 to 21 days post-stroke. Additionally, blood NfL can serve as a diagnostic biomarker for differentiating AIS from transient ischemic attacks and other cerebrovascular subtypes (54). Pedersen et al. included 595 ischemic stroke cases in their study to investigate the correlation between serum NfL concentrations at different time points post-stroke. They found that NfL could predict neurological and functional outcomes in both the acute phase (range 1–14 days, median 4 days) and the long-term (cases followed up after 3 months) (55). Uphaus et al. also highlighted NfL as a biomarker for predicting cerebrovascular function 90 days post-ischemic stroke (56). Barba et al. explored the relationship between serum NfL concentrations and clinical outcomes in patients with AIS, finding that patients with higher NfL levels showed less clinical improvement post-treatment. In patients with moderate to severe AIS, serum NfL levels were correlated with clinical and radiological scores at different time points and were predictive of short-term and intermediate-term clinical outcomes (57).

Myelin basic protein (MBP) is a membrane protein synthesized by oligodendrocytes that plays a crucial role in stabilizing myelin structure and is highly specific to neural tissue. In cases of brain injury, myelin damage leads to the release of MBP into the bloodstream, resulting in elevated serum levels. The concentration of MBP in the blood serves as an indicator of central nervous system injury. Studies have shown that serum MBP levels in patients with AIS are positively correlated with infarct volume, suggesting its potential as a biomarker for brain injury assessment (58).

Exosomes are nanoscale extracellular vesicles secreted by most cells, capable of crossing the blood–brain barrier and transferring various bioactive molecules between cells. They facilitate intercellular communication and are closely associated with the occurrence and progression of various diseases. Exosomes, particularly the functional substances they carry, play a crucial role in the pathogenesis and recovery process of ischemic stroke by affecting the neurovascular unit. Following an ischemic stroke event, various types of cells, including peripheral blood cells, endothelial cells, and brain cells, release exosomes. These exosomes can traverse the blood–brain barrier and be detected in cerebrospinal fluid and peripheral blood. Consequently, exosomes are increasingly recognized as potential biomarkers for the early diagnosis and prognosis of IS.

Xu et al. demonstrated that long non-coding RNAs (lncRNAs) such as lnc-CRKL-2, lnc-NTRK3-4, RPS6KA2-AS1, and lnc-CALM1-7, isolated from the serum of acute stroke patients, are significantly elevated (59). Bai et al. isolated exosomes from serum samples of IS patients and normal controls, finding elevated expression of circFUNDC1 in exosomes derived from IS patients’ serum. Receiver operating characteristic (ROC) analysis revealed an area under the curve (AUC) of 0.882 for circFUNDC1, indicating its high sensitivity and specificity as a diagnostic biomarker for IS (60). Xiao et al., through exosome circular RNA sequencing, large-sample validation, and diagnostic model construction, identified exosomal circ-0043837 and circ-0001801 as independent predictors of large-artery atherosclerosis (LAA) stroke. These circular RNAs showed significantly higher expression levels compared to controls, with diagnostic accuracies of AUC = 0.89 and AUC = 0.91, respectively, surpassing the diagnostic performance of plasma circular RNAs (61). Zhou et al. found that levels of miR-134 and miR-223 in exosomes from IS patients were significantly higher than those in non-ischemic stroke patients. Additionally, these levels correlated positively with NIHSS scores (r = 0.65, p < 0.01) and infarct volume (r = 0.68, p < 0.01), suggesting that miR-134 and miR-223 have potential diagnostic value for assessing the occurrence and severity of IS (62).

Exosomes are not only useful for the early diagnosis of IS but also help distinguish between different stages of the disease. Wang et al. found that, compared to controls, plasma exosomes in subacute and recovery-phase stroke patients showed significantly elevated levels of miRNA-21-5p, while exosomal miR-30a-5p was significantly higher in ultra-early stroke patients but lower than in controls during the acute phase. Furthermore, early diagnosis of large-artery atherosclerosis (LAA), which is associated with the worst prognosis, is particularly crucial (63). Zhang et al. observed significant increases in exosomal lnc_000048, lnc_001350, and lnc_016442 in LAA patients, with levels rising with stroke severity and showing better predictive capability for prognosis than NIHSS scores (64).

In recent years, multiple studies have demonstrated significant alterations in the expression of specific exosomal miRNAs in patients with AIS. Several miRNAs exhibit upregulated expression in AIS, including exosomal miR-212/132, miR-21, miR-9, miR-124, miR-134, and miR-223. In contrast, the expression level of exosomal miR-126 is downregulated in AIS patients (62, 65–68). These distinct miRNA expression patterns not only provide insights into the pathophysiological mechanisms of AIS but also hold potential clinical value as biomarkers for early diagnosis and disease assessment.

NETs are web-like structures released by neutrophils, primarily composed of free DNA, nucleosomes, and citrullinated histone H3 (citH3). NETs play a crucial role in the onset and progression of AIS. Studies have shown that NETs exacerbate early blood–brain barrier (BBB) disruption in AIS, increasing its permeability and potentially facilitating inflammatory cell infiltration, thereby aggravating neuronal damage (69). Research by Vallés et al. revealed that plasma NET levels in AIS patients were significantly higher than those in healthy individuals (70). Similarly, Lim et al. reported a marked increase in NET levels in AIS patients, and receiver operating characteristic (ROC) curve analysis demonstrated that the area under the curve (AUC) for double-stranded DNA (dsDNA) in early AIS diagnosis reached 0.859, suggesting its potential as an early diagnostic biomarker for AIS (17). Beyond its diagnostic implications, NETs are also closely associated with AIS severity and prognosis. Studies indicate that NET levels in thrombi and peripheral blood can reflect stroke severity and effectively predict short-term patient outcomes (71). Additionally, NETs may influence treatment responses in AIS. Evidence suggests that NETs could serve as prognostic biomarkers for futile recanalization following intravenous thrombolysis or mechanical thrombectomy. Arnaud et al. found that NETs were universally present in AIS thrombi, with all patients exhibiting NET-containing clots. High NET content in thrombi was correlated with failed recanalization, prolonged procedure time, and poorer stroke outcomes as assessed by the National Institutes of Health Stroke Scale (NIHSS) and modified Rankin Scale (mRS) scores (72). Moreover, analysis by Chen et al. demonstrated that the enrichment of NETs affects thrombus mechanical properties, potentially influencing the success rate of mechanical thrombectomy. Lower NET levels were significantly associated with higher rates of initial vascular recanalization (73).

To provide a more intuitive overview of these biomarkers and their roles, we present a summary table outlining their specific functions and recent research progress in ischemic stroke (Table 1).

ELISA is a widely used method for detecting protein levels in body fluids, characterized by high sensitivity and specificity. It utilizes antigen–antibody reactions, with enzyme-labeled antibodies to detect target proteins. ELISA is suitable for detecting various stroke-related biomarkers, such as CRP (74–76), IL-6 (43, 74, 75), TNF-α (75, 76), ICAM-1, MMP-9 (43, 75), VCAM-1 (77, 78), GFAP⁵³ (79, 80), and S100⁵³ (81). The advantages of ELISA include simplicity of operation, good reproducibility, and the ability to detect multiple samples simultaneously. However, its disadvantages are longer detection time and higher costs.

qPCR is a technique used to quantify RNA levels by labeling PCR amplification products with fluorescent dyes and monitoring the fluorescence signal changes in real-time. It is suitable for detecting genetic biomarkers like miRNA and lncRNA. The advantages of qPCR include high sensitivity, strong specificity, and accurate quantification, capable of detecting low-abundance RNA molecules (82–84). However, qPCR requires high-quality RNA samples, is complex to operate, and has high costs.

LC–MS/MS combines the separation capabilities of liquid chromatography with the high sensitivity of mass spectrometry. It is used to detect metabolites such as fatty acid derivatives, amino acids, and oxidative stress markers like 3-nitrotyrosine nitrated fibrinogen (85), 8-iso-PGF2α (86), and MDA (87). The advantages of LC–MS/MS are high sensitivity, strong specificity, and the ability to detect multiple metabolites simultaneously, making it a key tool in metabolomics research. However, LC–MS/MS equipment is expensive, the operation is complex, and it requires professional technicians.

Immunoturbidimetry measures the turbidity changes in a solution due to antigen–antibody complex formation, quantifying protein biomarkers in blood. It is suitable for detecting coagulation and fibrinolysis system-related factors such as D-dimer, FIB, TAT, t-PAIC, APTT, PT, FDP, and SAA (88, 89). The advantages of immunoturbidimetry include rapid detection, ease of operation, and lower costs, but its sensitivity and specificity are relatively lower.

NGS is a high-throughput DNA sequencing technology used for comprehensive analysis of genomes, transcriptomes, and epigenomes. It can detect DNA methylation states and RNA expression profiles, suitable for studying complex genetic biomarkers (90, 91). The advantages of NGS include high throughput, strong sensitivity, and specificity, and the ability to detect thousands of genes and their regulatory elements simultaneously. However, NGS equipment and operating costs are high, data analysis is complex, and it requires professional bioinformatics support.

Simoa is a breakthrough in biomarker detection due to its extremely high sensitivity and specificity. This technology combines single-molecule immunocapture with fluorescence detection, capable of detecting extremely low concentrations of biomarkers, making it advantageous in stroke biomarker detection. Simoa’s core lies in its ability to capture and detect single target molecules in each reaction well, significantly enhancing detection sensitivity. Compared to traditional ELISA, Simoa can detect biomarkers at femtomolar levels. Simoa has been used to detect neuronal injury markers such as GFAP, NFL, and S100, which are rapidly released into the blood after brain tissue injury, providing crucial information for prognosis assessment (92–94).

Recent research has expanded the application of Simoa in stroke biomarker detection, developing multiplex detection platforms for simultaneous detection of multiple inflammatory and neuronal injury markers, improving detection efficiency and data reliability. Onatsu et al. utilized Simoa to analyze the NfL levels in 136 patients with AIS, discovering that the presence and extent of axonal injury estimated by NfL were correlated with the final infarct volume (95). Mattila et al. measured the GFAP levels and release rates in patients with acute cerebral ischemia using the Simoa method, finding that for patients with acute cerebral ischemia, prehospital sampling within 3 h combined with a specific rule (prehospital GFAP >410 pg./mL or prehospital GFAP 90–410 pg./mL combined with GFAP release >0.6 pg./mL/min) exhibited high specificity (NPV 98.4%) in 68% of acute cerebral ischemia patients (96).

Recent years have seen significant progress in the application of biosensing technologies for stroke biomarker detection, showing great potential and prospects. Biosensors combine biological recognition elements with physical sensors, converting biological molecules into detectable signals, enabling high sensitivity, rapid, and on-site detection, playing a key role in early diagnosis, condition monitoring, and personalized treatment of stroke.