Cerebrovascular diseases (CVD) are major causes of disability and mortality worldwide; two major types are ischemic stroke (IS) and cerebral venous thrombosis (CVT). Although both are characterized by vascular occlusion (blood vessel blockage), their pathological mechanisms, risk factors, and clinical management have significant difference.

IS is caused mainly by the rupture of atherosclerotic plaques or cardiac embolism leading to cerebral blood flow interruption. It is the second leading cause of death and disability worldwide, with an annual incidence of 200 per 100,000, of which more than 75 per cent occurs in people over 60 years of age (1). Recent studies have revealed that ischemia–reperfusion injury activates microglia through the TLR4/NF-κB pathway, releasing proinflammatory factors such as IL-1β and TNF-α, which drive microglia toward proinflammatory (M1-type) polarization, resulting in an “inflammation-thrombosis vicious cycle” (2, 3). Although antithrombotic therapy reduces the acute-phase mortality rate, survivors still have a risk of recurrence, suggesting that the chronic inflammation-driven pathological process requires clinical intervention.

Unlike IS, CVT is a rare cerebrovascular disease of the intracranial venous system (annual incidence of about 1.3/100,000 people), which affects mainly young women (the male-to-female ratio is 1:3) and is characterized by endothelial damage to the venous sinuses, hypercoagulability, and stagnant blood flow (4, 5). The thrombi in CVT are fibrin-rich and their formation is closely related to the activation of the coagulation cascade driven by the monocyte-tissue factor (TF) axis and the inflammatory response mediated by the interleukin (IL)-6/STAT3 pathway (6). Although anticoagulation significantly affects prognosis, many patients may experience chronic symptoms such as cognitive impairment or headache after treatment, possibly related to a persistent low inflammatory response (e.g., mildly elevated IL-6 and CRP) (5).

Although both IS and CVT involve “inflammation-thrombosis interaction” pathways, their specific mechanisms are significantly different (Table 1). Commonalities: Systemic inflammatory responses drive endothelial injury and thrombus expansion, forming a vicious cycle. Differences: In IS, arterial thrombosis with inflammation concentrated in the ischaemic region. In CVT, venous thrombosis is associated with more a more extensive inflammatory response and a high risk of chronicity.

Currently, in clinical practice, the diagnosis of these two diseases relies mainly on imaging evidence. However, it often faces issues such as insufficient sensitivity or delayed identification. Therefore, clarifying the differences in specific inflammatory markers between the two types of diseases is valuable for early differential diagnosis, targeted anti-inflammatory therapy and prognostic stratification. By systematically comparing their inflammatory marker profiles, this study aims to: reveal the differences in inflammatory marker profiles between IS and CVT, screen for disease-specific markers, and provide a low-cost diagnostic tool for resource-limited areas.

Currently, the exploration of the associations between inflammatory biomarkers and CVD continues to deepen. This study included 18 studies and aimed to explore the differences in inflammatory biomarkers between CVT and IS through a systematic review and network meta-analysis. This study transcended the limitations of traditional single-disease research. To our knowledge, this is the first NMA to compare the inflammatory markers of CVT and IS to reveal their common features and specificity in the “inflammation-thrombosis” interactions. However, the inevitable heterogeneity, unclear methodologies, and data transformations require cautious interpretation.

Our analysis revealed significantly elevated levels of IL-6, CRP, hs-CRP, SII, and PLR in the peripheral blood of CVT patients during acute phase. These findings indicate two key pathophysiological processes: (1) a robust systemic inflammatory response, and (2) amplified cross-talk between inflammation and coagulation cascades. This finding is in great agreement with earlier studies: the levels of hs-CRP and IL-6 in the acute stage of CVT are much higher than those in the chronic stage; SII is the strongest independent predictor for CVT, and its ability to predict CVT is better than NLR and PLR (25, 26). The results of this study further support the value of SII in evaluating the thrombo-inflammatory microenvironment. Therefore, for patients with unexplained headache, MRI/MRV is recommended as a priority to screen for CVT when SII is too high (26). As a pro-inflammatory cytokine, IL-6 induces hepatic synthesis of CRP and hs-CRP via the JAK–STAT pathway (33). These acute-phase proteins further to promote tissue factor expression and initiate the extrinsic coagulation pathway to facilitate thrombus formation (18). The significant elevation of SII and PLR reflects neutrophil and platelet overactivation and lymphocyte immunosuppression in the patients. Neutrophils activate endothelial cells and platelets by releasing neutrophil extracellular traps (NETs), reactive oxygen species (ROS), and proinflammatory factors (e.g., IL-6 and TNF-α) to promote coagulation (34, 35). Platelets are not only coagulation elements but also an important sources of inflammatory mediators and a key hubs linking the inflammatory response to thrombosis (5). An increased PLR is independently associated with the risk of recurrent venous thrombosis through a mechanism involving platelet-activated TF expression and increased fibrin production (36). In contrast, although monocytes and MHR showed statistical significance (p < 0.05), their effect sizes (MD = 0.1 and 0.09) were lower than that those of IL-6 and CRP. These findings suggest that monocyte-related inflammation may serve as a secondary driver in CVT. However, owing to the statistical significance of the sample size, its clinical value needs to be verified in a larger cohort.

In IS patients, elevated neutrophil levels signify their recruitment to arterial plaque rupture sites. At these sites, neutrophils release myeloperoxidase (MPO) and matrix metalloproteinases (MMPs), which exacerbate endothelial damage and degrade the collagen matrix of the plaque fibrous cap. This process leads to plaque structural instability and promotes ischemic reperfusion injury (37). This process leads to plaque structural instability and promotes ischaemia-reperfusion injury. Hs-CRP is strongly associated with plaque vulnerability, and its mild to moderate elevation indicates chronic inflammation within the plaque (38). Increases in SII and PLR signify synergistic neutrophil–platelet activation: SII indicates an immunocoagulation interaction in thrombus formation, and PLR denotes platelet hyperreactivity, both of which are involved in thromboembolism after plaque rupture (39). An elevated MHR highlights the important role of dysregulated lipid metabolism-inflammation axis interactions in driving atherosclerotic plaque instability. Nonetheless, the analysis of MHR is also limited by sample size, and IS subtype stratification (e.g., large-artery atherosclerosis, cardioembolic stroke, and small-vessel disease) was not performed. This methodological gap likely contributed to heterogeneous inflammatory marker profiles. Future investigations should combine advanced imaging (e.g., plaque component analysis) and molecular approaches (e.g., single-cell RNA sequencing) to dissect subtype-dependent inflammatory pathways, informing the development of targeted antithrombotic and anti-inflammatory strategies.

Compared with the largely focal arterial injury in IS, CVT induces venous congestion-driven endothelial hypoxia and diffuse BBB perturbation, amplifying systemic IL-6/CRP signals. Our network meta-analysis showed that CRP and IL-6 levels are significantly higher in CVT than in IS. This finding highlights a key difference in inflammatory pathways: in CVT, venous endothelial injury triggers a strong systemic acute-phase response, leading to marked elevations in CRP and IL-6. Several studies have shown that serum IL-6 and CRP levels are significantly higher in CVT patients than in healthy controls and are correlated with disease severity (40). Pathophysiologically, venous thrombosis induces endothelial hypoxia and venous congestion, which together provoke a robust systemic acute-phase response. Venous congestion due to impaired venous drainage further exacerbates endothelial injury, leading to a more pronounced local inflammatory response. Endothelial hypoxia in the setting of impaired venous drainage stimulates local IL-6 production and endothelial activation; IL-6 subsequently activates hepatic STAT3 signaling, markedly increasing CRP synthesis and creating an injury-cytokine → acute-phase protein cascade that also upregulates TNF-α and IL-1β (6, 33, 41). In addition, venous stasis and endothelial activation promote neutrophil recruitment and the formation of neutrophil extracellular traps (NETs), a thrombo-inflammatory process that amplifies both thrombus growth and local cytokine release and may further increase systemic IL-6 levels (34, 39). Venous congestion in CVT may produce a relatively more diffuse pattern of blood–brain barrier (BBB) perturbation than the typically focal BBB disruption seen in many cases of arterial ischemia (42); such diffuse BBB disturbance could facilitate translocation of brain-derived cytokines and inflammatory mediators into the systemic circulation and thereby magnify peripheral IL-6 and CRP signals in some patients. In IS, IL-6 is primarily secreted by activated microglia, astrocytes, and cells within arterial plaques in ischemic brain tissue (43). However, the selective permeability of the BBB restricts the diffusion of cytokines such as IL-6 into the systemic circulation, resulting in a significantly less pronounced increase in systemic IL-6 concentrations in IS patients compared to those with CVT (40, 43). This divergence stems from a fundamental difference in injury mechanisms. In CVT, venous thrombosis causes systemic endothelial hypoxia through impaired venous drainage. Conversely, IS results from localized arterial damage due to plaque rupture in cerebral arteries. These distinct pathological processes establish contrasting inflammatory profiles—CVT elicits a strong systemic acute-phase response, whereas IS is characterized by local inflammatory propagation. In IS patients, the mild elevation of CRP and IL-6, likely reflects a postischemic dynamic balance between central and peripheral inflammation: the activation of microglia in the brain results in the release of cytokines, including IL-6, which mediate the dual regulation of neurorestoration and inflammatory injury; concurrently, peripheral monocyte migration into the ischemic penumbra intensifies local oxidative stress, facilitating neuroimmune crosstalk (44). Although this chronic inflammatory state fails to initiate a strong systemic acute-phase response, sustained elevation of CRP is associated with plaque progression and long-term recurrence risk. This provides evidence for statin therapy to improve prognosis through regulating lipid metabolism and exerting anti-inflammatory effects (45). This pathophysiology explains why CRP/IL-6—but not NLR/PLR/SII—best discriminate CVT from IS in our analysis. Given the central role of IL-6 in driving thrombus formation, anti-IL-6 therapies may provide clinical benefits by blocking the IL-6 receptor to reduce inflammation and lower thrombotic risk (46). Recent CVT-specific clinical data show that short-term corticosteroid regimens combined with anticoagulation significantly reduce serum and CSF IL-6 and hs-CRP and are associated with improved functional outcomes, providing indirect clinical support for IL-6–mediated pathology in CVT (47). However, no IL-6–targeted agents have yet been evaluated in patients with CVT, therefore, future prospective randomized trials are warranted to determine the safety, optimal timing, and patient selection for potential anti-IL-6 or other anti-inflammatory interventions.

No significant intergroup differences were observed in other inflammatory markers, likely reflecting stress-induced inflammation inherent to CVD rather than thrombosis-type specific changes. This finding implies a shared core mechanism of systemic inflammation activation across arterial and venous thrombotic disorders. For example, neutrophils are associated with chemokine-mediated recruitment following venous endothelial injury in CVT, whereas in IS, they arise from ischaemic reperfusion injury. Despite distinct pathogenic pathways, neutrophils exhibit similar elevations in both diseases. The levels of novel inflammatory biomarkers, such as NLR, PLR, and SII are markedly increased in both CVT and IS, demonstrating their potential as markers to distinguish healthy individuals from CVD patients. High SII values can independently predict poor prognosis and monitor therapeutic efficacy in these patients (28, 41). These markers (SII, NLR and PLR) are recognized as critical indicators of cerebral arterial ischemia, potentially offering more accurate thrombosis diagnoses than immune cell counts alone (48). Our NMA findings further revealed that these novel inflammatory markers remain prominently elevated in affected patients. Importantly, neuroimaging remains the diagnostic gold standard for both IS and CVT, and the biomarkers identified in our analysis should be regarded as exploratory adjuncts that may complement but cannot substitute for imaging.

The inflammatory signatures of CVT and IS demonstrate both convergence and divergence. Shared mechanisms include activation of neutrophil–platelet crosstalk and systemic immunocoagulative responses, reflected by elevated SII, NLR, and PLR in both conditions. In contrast, CVT exhibits a more pronounced systemic acute-phase reaction, dominated by IL-6–driven CRP elevation and diffuse endothelial activation secondary to venous congestion. Conversely, IS is characterized by localized arterial inflammation, in which monocyte–macrophage and lipid-metabolism–related inflammation, reflected by indices such as MHR and markers linked to plaque vulnerability. Many other markers remain non-specific and overlapping; their elevation likely indexes a shared host response to vascular injury rather than pathognomonic disease-specific biology.

Notably, both CVT and IS groups presented decreased lymphocyte counts. This shared feature may be associated with systemic inflammatory stress responses and immunological homeostasis disruption mediated by vascular injury. Vascular damage under both conditions triggers the innate immune response, leading to the release of proinflammatory cytokines such as IL-6 and TNF-α (49). These cytokines induce lymphocyte apoptosis and promote lymphocyte migration to thrombotic or ischemic foci, leading to a decrease in the number of circulating lymphocytes. IS is associated with more profound lymphopenia. This may be because arterial thrombosis-associated inflammation activates the hypothalamic–pituitary–adrenal (HPA) axis to augment cortisol-mediated immunosuppression. HPA axis activation not only directly inhibits B lymphocyte production but also elevates cortisol levels, which further suppresses lymphocyte proliferation and accelerates apoptosis, forming a “stress–immunosuppression” vicious cycle (50). Additionally, in the ischemic penumbra, bursts of oxygen free radicals, excessive microglial activation, and neutrophil infiltration trigger the release of chemokines (e.g., IL-6, TNF-α and CXCL10), facilitating lymphocyte migration into the brain and subsequent depletion (51). Thus, lymphopenia in CVT and IS likely reflects a combination of vascular injury, inflammatory signaling, and neuroendocrine regulation, with mechanistic disparities tied to thrombus etiology, inflammatory magnitude.

Research indicates that reduced peripheral blood lymphocyte counts in IS patients are significantly associated with the severity of neurological deficits and negatively correlated with infarct volume (52). Clinically, lymphopenia in IS has also been linked to worse neurological outcomes and higher rates of post-stroke infections (53, 54). This association may arise from increased lymphocyte apoptosis due to poststroke immunosuppression and impaired adaptive immunity (53, 54). In CVT patients, although the link between lymphocyte reduction and thrombus burden awaits confirmation, early evidence suggests that lymphopenia may signify a more aggressive thrombotic inflammatory phenotype (55). Abnormalities in lymphocyte count and derived indices such as NLR and PLR indicate a systemic “pro-inflammatory-anti-inflammatory” imbalance—for example, an elevated NLR combined with marked lymphocyte reduction suggests a paradoxical state of coexisting intense immune activation and suppression (41). When lymphopenia co-occurs with elevated IL-6 and CRP, the constellation may represent dysregulated immunity that could both exacerbate tissue injury and increase susceptibility to secondary infections such as pneumonia or urinary tract infection. Future cohorts should implement time-locked immune monitoring (including lymphocyte subsets) and capture secondary infections to delineate the clinical impact of immunosuppression following acute cerebrovascular events.

This study has several limitations. First, the limited number of studies included in the analysis (such as MHR and SII) precludes robust conclusions from the network meta-analysis. As a result, paired comparisons yielded imprecise effect size estimates, which should be interpreted as exploratory evidence rather than definitive findings in the current context. Second, although sensitivity analyses were conducted, inherent bias arose from heterogeneous study designs and variable data collection timepoints across the included studies, which may influence our conclusions. Third, conversion of medians to means for skewed data may introduce additional bias. Additionally, NMA is highly dependent on model specifications, which can increase variance in effect estimates, widen confidence intervals, and potentially compromise conclusion reliability. Finally, owing primarily to the insufficient sample size, subgroup analyses based on CVT and IS subtypes were not conducted. This omission led to excessive heterogeneity in some inflammatory markers. Moreover, geographical publication bias across the included studies and inconsistent measurement methods for certain markers also affected the accuracy and generalizability of the results. These constraints indicate that the present results are hypothesis-generating and require prospective, standardized validation before clinical application.

This study revealed differences and shared characteristics in the inflammatory marker profiles between CVT and IS. CVT is characterized by systemic inflammatory activation marked by elevated IL-6 and CRP, whereas IS involves localized arterial inflammation (monocytes, MHR) linked to atherosclerotic plaque instability. Network meta-analysis further confirmed that CRP and IL-6 can serve as specific biomarkers to distinguish between these two diseases. Markers such as SII, NLR, and PLR (due to their significant elevation) may have potential as exploratory screening indicators across CVD. The observation of lymphopenia in both CVT and IS patients suggests stress-induced immunosuppression, although the specific mechanisms and clinical implications require further exploration. Given the limitations of this study, these findings should be considered hypothesis-generating and require validation in prospective cohorts before clinical translation.