The relationship between inflammatory markers and the severity and prognosis of neurological dysfunction in stroke patients Authors: Xile Zhang, Luyu li, Qing Lin, Yu Xiao, Yani Dong, Xiaoxiao Tao, Yong Wang Published: 2026 ABSTRACT AbstractBackgroundThe aim of this study is to explore the relationship of inflammatory indicators with the severity and prognosis of neurological dysfunction in stroke patients, and to evaluate the prognostic value of inflammatory indicators for patients suffering from acute ischemic stroke (AIS).MethodsThe study recruited 225 AIS patients and 46 healthy individuals, and collected their clinical data and hematological indicators. Calculate six inflammatory markers and evaluate the National Institutes of Health Stroke Scale (NIHSS) and Modified Rankin Scale (mRS) scores at admission, discharge, 1 months of onset and 3 months of onset. Correlation analysis, multivariate logistic regression analyses, the receiver operating characteristic (ROC) curve and the area under the curve (AUC) were used to analyze the relationship of inflammatory markers with the severity and prognosis of stroke.ResultsElevated levels of neutrophil to lymphocyte ratio (NLR), systemic immune-inflammation index (SII), systemic inflammatory response index (SIRI), and pan immune-inflammation value (PIV) were observed in patients at admission (p < 0.05). Most inflammatory markers are positively correlated with the severity of neurological deficits during the acute phase, especially upon admission and discharge. Higher admission lymphocyte to monocyte ratio (LMR) levels are independently associated with larger infarct volume and adverse functional outcomes. In contrast, indicators like NLR, SIRI and PIV showed associations in the opposite direction (odds ratios < 1), suggesting complex interactions that require further investigation. For different time points, NLR showed the greatest prognostic value at admission (AUC = 0.654) and PIV at discharge (AUC = 0.725).ConclusionsThese inflammatory markers exhibit characteristic dynamic changes after AIS and have close relationship with the severity and prognosis of neurological deficits. It may contribute to early risk stratification and personalized treatment soon after.Keywords: Acute ischemic stroke (AIS), Inflammatory indicators, Lymphocyte ratio (NLR), Lymphocyte to monocyte ratio (LMR), Systemic inflammatory response index (SIRI) Introduction Acute ischemic stroke (AIS) is one of the salient causes of high mortality and long-term disability in the world [1]. Despite significant progress having been made in reperfusion therapy (such as intravenous thrombolysis and endovascular thrombectomy), the severity and final prognosis of post-stroke neurological deficits still vary greatly among individuals [2]. Therefore, the search for biomarkers that can assess the severity of AIS and the prognosis in the early stage is of great significance, which can guide personalized treatment, optimize medical resource allocation, and improving the quality of life for patients. Recently, increasing evidence has showed that the inflammatory response plays a critical role in the occurrence, development, and recovery of AIS [3–5]. Cerebral ischemic injury triggers a complex inflammatory cascade that activates the systemic immune system [6]. This reaction is a double-edged sword: moderate inflammatory response helps the clear of necrotic tissue [7], but excessive or persistent inflammatory response can exacerbate the damage to the blood-brain barrier, trigger secondary injury, and worsen neurological outcomes [8]. Consequently, the systemic inflammatory status is a key determinant of prognosis in AIS. Traditional inflammatory indicators, such as tumor necrosis factor - α, white blood cell count, C-reactive protein and so on, have been proven to be associated with the prognosis, but their specificity is limited [9]. Consequently, researchers have turned their attention to novel composite inflammatory markers derived from peripheral blood cell counts. These biomarkers can more comprehensively reflect the immune-inflammation state by collecting and organizing information from different white blood cell subsets, platelets and other factors [10, 11]. Among them, neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR) and lymphocyte to monocyte ratio (LMR) have attracted more attention due to easy to calculate and reproducibility. Furthermore, more complex composite indicators, like systemic immune-inflammation index (SII), systemic inflammatory response index (SIRI) and pan immune-inflammation value (PIV), have been developed to more accurately determine the systemic inflammation status of the host [12]. Numerous researches had preliminarily revealed the association between these biomarkers and dieases such as myocardial infarction, tumor and so on [13–15]. In stroke, previous work have revealed that there may be a correlation between higher level of NLR and SII and more severe neurological deficits (such as higher National Institutes of Health Stroke Scale (NIHSS) scores), larger infarct volumes, poorer neurological performance (such as higher modified Rankin scale (mRS) scores) [16, 17]. Besides, the elevation of systemic inflammatory markers (such as NLR, PIV, SIRI) observed in this study not only reflect the systemic inflammatory status, but also indirectly reflect neuroinflammation within the central nervous system. This neuroinflammation directly affects the recovery of cognitive domains (such as attention, executive function, memory) and behaviors (such as apathy, depression). However, existing research has mostly focused on a single or a few biomarkers, there is still insufficient research on updated indicators such as SIRI and PIV in stroke. Meanwhile, there are also few studies observing the dynamic changes and correlations after the onset of AIS. Based on the lack of the above research, our current work is eager to systematically explore the association between six markers and the degree of neurological damage and the prognosis in patients who have just experienced AIS attacks. Ultimately, our efforts hope to explore the potential value of these easily accessible inflammatory markers as stroke prognostic tools, providing new evidence-based medicine for risk stratification and early intervention. Materials and methods Study design This was a prospective, single-center, observational cohort study conducted at Taizhou First People’s Hospital. Patients The clinical study has been approved by the ethical review board of Taizhou First People’s Hospital, Zhejiang, China (No.2020-KY022-01). All processes are strictly carried out in accordance with the ethical standards of the Helsinki Declaration. We enrolled 225 AIS patients and they were admitted to Taizhou First People’s Hospital in Zhejiang from January 2021 to January 2022. The target sample size of approximately 225 AIS patients was determined based on a pragmatic consideration of achieving stable estimates for correlation and regression analyses, aiming for a minimum of 15–20 events per variable in our planned multivariate models. While not derived from a formal power calculation for a specific primary outcome, this sample size is consistent with similar exploratory biomarker studies in stroke cohorts and was deemed sufficient for identifying clinically relevant associations in the acute phase. All enrolled AIS patients underwent imaging examinations and met the following inclusion criteria: (1) Patients were over 18 years old; (2) Acute onset accompanied with focal neurological deficits; (3) Exclude cerebral hemorrhage, intracranial space occupying lesion and intracranial infections through CT or MRI; (4) NIHSS score < 20; (5) The onset time is less than 7 days; (6) Obtain the consent of the patient or their relatives. Those who met the following criteria will be excluded: (1) Patients with coexisting serious diseases such as heart and kidney dysfunction, malignant tumors and unstable vital signs; (2) Having hemorrhagic or ischemic cerebrovascular disease in the past 3 months; (4) Patients who have suffered from infectious diseases (acute or chronic) within the past three months; (5) Fever (> 38 ℃, symptoms quickly recover after cooling), hypoglycemia (< 2.8mmol/L, symptoms quickly recover after supplementing blood sugar), hysteria (rapid recovery of symptoms after implying treatment) and other types of mimic stroke. 46 healthy adults were recruited in the control group, with age and gender ratios comparable to AIS patients, and no history of stroke. All participants in this study were voluntary and signed informed consent forms. Data collection At admission, we collected the clinical information and characteristics of all participants, including sex, age, vascular risk factors (hyperlipidemia, hypertension, hyperhomocysteinemia, smoking, drinking, atrial fibrillation and diabetes) and laboratory results (counts of red blood cells, white blood cells, platelets, hemoglobin, neutrophils, lymphocytes, monocytes, basophils and eosinophils). The information on reperfusion therapies and exact time from symptom onset to blood sampling was not systematically collected, and thus these variables were not controlled for in the analyses. NLR is the value obtained by dividing the neutrophil count by the lymphocyte count; PLR is the value obtained by dividing platelet count by lymphocyte count; LMR is the value obtained by dividing the lymphocyte count by the monocyte count; SII is calculated by platelet * neutrophil / lymphocyte; SIRI is obtained by multiplying the neutrophil count by the monocyte count and then dividing by the lymphocyte count; PIV is obtained by multiplying platelet count by monocyte count by neutrophils, and then dividing by lymphocytes. Two researchers respectively evaluated mRS and NIHSS scores at four time points: admission, discharge, 1 month and 3 months of onset. We defined the adverse neurological function at admission as NIHSS ≥ 5 or mRS ≥ 3. The value we used for the definition is based on the clinical consensus and the standards which were commonly used in the field of stroke. Defining NIHSS ≥ 5 as moderate or above neurological impairment is a widely accepted standard in clinical research [18]. This value can better distinguish patients with neurological impairment of clinically significant, who typically have greater prognostic variability and require more active medical intervention and prognostic evaluation. MRS ≥ 3 is usually defined as adverse functional outcome or disability [19]. Specifically, patients with mRS = 3 do not require constant supervision from others, but their daily activities are restricted (such as walking outside requires assistance), and their quality of life has been significantly affected. Therefore, using this value as a dividing point has clear clinical significance and can effectively identify patients who have not achieved functional independence. If there is any disagreement between the two researchers, the corresponding author will make the final decision. Scan and measure the infarct volume were through MRI. Statistical analysis All continuous data were tested for normality using the Shapiro–Wilk test. Normally distributed data are presented as mean ± standard deviation (SD) and were compared using the independent two-sample t-test. Non-normally distributed data are presented as median with interquartile range (IQR) and were compared using the Mann–Whitney U test for pairwise comparisons (e.g., between patient time points and the control group). For comparisons among multiple time points within the AIS group, the Kruskal–Wallis test followed by Dunn’s post-hoc test was applied. Spearman’s correlation analysis was used to assess relationships between inflammatory markers and clinical scores. Multivariate logistic regression was performed to identify independent predictors of outcomes. To account for potential confounding effects, different sets of covariates were adjusted for in models predicting different outcomes, and results were expressed as adjusted odds ratios (OR) with 95% confidence intervals (CI). For the outcome of large infarct volume (defined as infarction volume of three or more), models were adjusted for: age, sex, history of hypertension, diabetes, atrial fibrillation, smoking, alcohol consumption, and administration of intravenous thrombolysis. For the outcome of severe neurological deficit at admission (NIHSS ≥ 5), models were adjusted for: infarct volume, age, admission mRS score, and history of atrial fibrillation. For the outcome of poor functional status at admission (mRS ≥ 3), models were adjusted for: sex, age, history of hypertension, diabetes, smoking, alcohol consumption, and atrial fibrillation. And the selection of covariates for adjustment was based on established clinical relevance and prior literature regarding stroke prognosis. Receiver operating characteristic (ROC) curves were plotted to evaluate the discriminative ability of inflammatory markers, and the area under the curve (AUC) was calculated. All analyses were performed using GraphPad Prism (version X), SPSS (version 25.0), and R (version 3.5.2). A p-value < 0.05 was considered statistically significant. Results Basic characteristics of all participants This study included 225 patients with AIS and 46 healthy individuals. The basic characteristics between the two groups are shown in Table 1. And obviously, there was no significant statistical difference between the two groups in terms of basic characteristics such as age and gender. Although the prevalence of diabetes, hypertension and atrial fibrillation in AIS patients is higher, there is no significant statistical difference between the two groups (p > 0.05). Meanwhile, counts of neutrophils, leucocytes, monocytes, NLR, SIRI and PIV were higher in AIS patients (p < 0.05, Table 1), while the count of hemoglobin, platelets, lymphocytes was lower in patients suffered from AIS(p < 0.05, Table 1). Table 1Baseline characteristics of individualsBaseline characteristicspatientshealtht/z/x2 P Sex, n (%)male138(61.33%)23(50%)2.03430.1538female87(38.67%)23(50%)Age, years< 45 years old15(6.67%)10(21.74%)1.1870.23145–60 years old53(23.56%)25(54.35%)> 60 years old157(69.77%)11(23.91%)Hypertension, n (%)Without70(31.1%)19(41.3%)1.79920.1798With155(68.9%)27(58.7%)Diabetes, n (%)Without153(68%)34(74%)0.6530.419With72(32%)12(26%)Atrial fibrillation, n (%)Without203(90.22%)43(93.48%)0.48350.4868With22(9.78%)3(6.52%)Leucocytes (×109/L)6.86(5.88,8.65)6.2 ± 1.35-3.4680.0005Erythrocytes (×1012/L)4.36(4,4.82)4.67 ± 0.451.92870.0538Hemoglobin (g/L)133(124,146)145.7 ± 13.24.0763< 0.001Platelets (×109/L)206(182,250)235.4 ± 43.82.91240.0036Lymphocytes (×109/L)1.71(1.29,2.18)1.95 ± 0.483.13580.0017Neutrophils (×109/L)4.64(3.56,6.18)3.79 ± 0.98-3.61890.0003Monocytes (×109/L)0.49(0.35,0.66)0.24 ± 0.09-8.7379< 0.001Eosinophils (×109/L)0.11(0.05,0.23)0.14(0.08,0.21)0.29220.7701Basophils (×109/L)0.01(0.01,0.02)0.02(0.02,0.03)5.3993< 0.001NLR2.91(1.82,4.79)1.90(1.57,2.29)-2.55170.0107PLR124.3(90.7,168.2)119.4(96.99,142.92)-1.21570.2241LMR3.57(2.52,4.85)7.97(6.33,10.71)9.0405< 0.001SII614.5(345.6,1097)449.32(331.74,592.61)-0.62550.5316SIRI1.3(0.66,2.57)0.46(0.33,0.75)-8.7147< 0.001PIV308.45(159.82,716.45)110.92(76.9,182.58)-7.1662< 0.001Abbreviations: NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio, LMR Lymphocyte to monocyte ratio, SII Systemic immune-inflammation index, SIRI Systemic inflammatory response index, PIV Pan immune-inflammation value Baseline characteristics of individuals Abbreviations: NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio, LMR Lymphocyte to monocyte ratio, SII Systemic immune-inflammation index, SIRI Systemic inflammatory response index, PIV Pan immune-inflammation value The change of six inflammatory markers in AIS patients Track and record information of all patients at four key time points (admission, discharge, one month and three months after onset). Due to failed follow-up and the loss of part information, there is only 186, 79, 43, and 29 patients’ information were used for later analysis. We found that most inflammatory indicators have a dynamic trend of change after onset. As is shown in the research, it revealed a dramatic increase of NLR, PIV, SII and SIRI in AIS patients (p<0.05, Fig. 1) when compared with control’s level, and reached peak at different time points. The NLR index reaches its peak at the onset of AIS and gradually decreases as the disease progresses. The levels of PIV and SIRI gradually increase after onset, reaching their peak at discharge and one month after onset, and then gradually decrease. SII has been fluctuating since the onset and still remains at a high level three months after onset. Strangely, as shown in the figure, the patient's LMR level is lower than that of healthy individuals, and the level fluctuates little after onset. Meanwhile, as for PLR, no statistical difference can be observed between the two groups. And we found that there was no noticeable dynamic change of PLR in patients after onset of the disease. Violin plots with median, 25 and 75% quartiles of the percentage of NLR, PLR, LMR, SII, SIRI and PIV of Controls (grey, n = 46) and different time points after acute ischemic stroke (Admission: blue, n = 186; Discharge: orange, n = 79; 1 month: red, n = 43; 3 months: purple, n = 29). Mann-Whitney test was employed to examine the difference between Controls and each time point. Decimals represent p-value. **: p < 0.01, ***: p < 0.001, ****: p < 0.0001, ns: no significance. NLR, neutrophil to lymphocyte ratio; PLR, platelet to lymphocyte ratio; LMR, lymphocyte to monocyte ratio; SII, systemic immune-inflammation index; SIRI, systemic inflammatory response index; PIV, pan immune-inflammation value Correlation of inflammatory markers at different stages with the degree of neurological dysfunction In order to explore the underlying association of inflammatory markers at different stages with the degree of neurological dysfunction, linear regression analysis was conducted to investigate whether there is a correlation between the levels of NLR, PLR, LMR, SII, SIRI, and PIV at four time points and key indicators of neurological dysfunction. The results were presented in a heatmap (Fig. 2). Correlation analysis between inflammatory markers and clinical outcomes. Heatmap depicting Spearman’s correlation coefficients (r) between the six inflammatory indicators (rows) measured at four time points (Adm: Admission, Dis: Discharge, 1 m: 1 month, 3 m: 3 months) and infarct volume, NIHSS, and mRS scores (columns) assessed at admission, discharge, and 3 months. The color scale represents the strength and direction of the correlation, with red indicating positive correlations and blue indicating negative correlations. Statistical significance is denoted by asterisks: *p < 0.05, **p < 0.01, ***p < 0.001. The heatmap was generated using the R package pheatmap, with columns clustered using Euclidean distance and complete linkage method From the graph, except for LMR, the levels of all other inflammatory indicators at admission are positively correlated with the severity of neurological deficits, especially NIHSS or mRS at admission and discharge. This indicates that lower levels of inflammatory markers clearly reflect milder neurological loss at admission and better neurological recovery at discharge. At one month, the levels of PIV (r=0.411, p<0.05) and SIRI (r=0.377, p<0.05) were also positively correlated with mRS scores at discharge. Meanwhile, we also found that a positive correlation between PIV (r=0.219, p<0.05), NLR (r=0.369, p<0.01), PLR (r=0.319, p<0.05) levels three months after onset and the severity of neurological deficits at discharge. The results indicated that not only there lies a correlation between the level of inflammatory markers at the onset of AIS and the severity of acute neurological deficits, but in the long term, high levels of inflammatory markers may indicate poor prognosis. As for the infarct volume, we can observe that there is a correlation between some inflammatory factors and infarct size at different time points, although no statistical significance was found, which may require a larger sample size for verification. In order to improve the quality of life of AIS patients after onset, we eager to explore a potential biomarker which may can be associated with the severity of stroke. Therefore, we used multiple logistic regression analysis to analyze the relationship between inflammatory indicators at admission and large infarct volume (defined as infarction volume of three or more), poor neurological performance (NIHSS≥5 or mRS≥3) (Table 2), which is in order to explore the potential of inflammatory factors in assessing the severity of stroke. And the results are presented as adjusted odds ratios (OR) with 95% confidence intervals (CI). After adjusting for potential confounders, multiple logistic regression analysis showed that a higher level of LMR upon admission is independently associated with large infarct volume (OR=1.347) and adverse functional outcomes upon admission (mRS≥3, OR=1.295), suggesting that LMR may be a risk factor for the severity of ischemic stroke. It is worth noting that most inflammatory indicators such as NLR, PLR, SII show an opposite trend with an OR value of less than 1 in the direction of association with neurological deficits (NIHSS≥5) and adverse functional outcomes (mRS≥3) at admission, which is different from conventional cognition. This suggests that there may be complex confounding factor interactions that need to be further validated in larger samples. Table 2Logistic regression of acute stage characteristicsAdmissionInfarct volume ≥ 3 cm^3NIHSS ≥ 5mRS ≥ 3OR(95% CI) P OR(95% CI) P OR(95% CI) P NLR1.077(0.791–1.467)0.63620.945(0.897–0.995)0.02550.662(0.492–0.891)0.0065PLR0.998(0.993–1.003)0.40830.993(0.988–0.999)0.01920.993(0.987–0.998)0.0066LMR1.347(1.029–1.762)0.02991.284(0.994–1.658)0.05551.295(1.035–1.621)0.0239SII1(0.999–1.001)0.90390.998(0.997-1)0.02460.998(0.997–0.999)0.0064SIRI0.648(0.404–1.038)0.07140.724(0.439–1.195)0.20690.514(0.322–0.819)0.0051PIV0.998(0.997-1)0.08210.999(0.997-1)0.13310.998(0.996–0.999)0.0109Abbreviations: OR Odds rate, CI Confidence interval, NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio, LMR Lymphocyte to monocyte ratio, SII Systemic immune-inflammation index, SIRI Systemic inflammatory response index, PIV Pan immune-inflammation value, NIHSS National Institutes of Health Stroke Scale, mRS Modified Rankin Scale Logistic regression of acute stage characteristics Abbreviations: OR Odds rate, CI Confidence interval, NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio, LMR Lymphocyte to monocyte ratio, SII Systemic immune-inflammation index, SIRI Systemic inflammatory response index, PIV Pan immune-inflammation value, NIHSS National Institutes of Health Stroke Scale, mRS Modified Rankin Scale Besides, we want to investigate the prognostic value of inflammatory factors, thus the logistic regression model was made in the study. Based on the multivariable logistic regression models that included the inflammatory markers and adjusted for relevant clinical covariates, we generated predicted probabilities and plotted ROC curves (Fig. 3). We conclude that for different time points, NLR has the highest discriminative ability at admission (AUC=0.654), while PIV has the highest discriminative ability at discharge (AUC=0.725). ROC curve was used to evaluate the accuracy of NLR, PLR, LMR, SII, SIRI and PIV to predict functional outcomes in AIS patients. ROC, receiver operator characteristic, NLR, neutrophil to lymphocyte ratio; PLR, platelet to lymphocyte ratio; LMR, lymphocyte to monocyte ratio; SII, systemic immune-inflammation index; SIRI, systemic inflammatory response index; PIV, pan immune-inflammation value; AIS, acute ischemic stroke Relationships between inflammatory markers and the prognosis after AIS Exploring potential factors which may affect the prognosis in patients suffered from AIS, we defined NIHSS ≥ 5 or mRS ≥ 3 at discharge, as well as NIHSS ≥ 2 or mRS ≥ 2 after three months of onset, as poor prognosis. At the same time, multivariate logistic regression analyses were conducted and summarized in Table 3 to evaluate the correlation between inflammatory factors and the recovery of neurological function at discharge and three months after onset. After adjusting for potential confounders, multiple logistic regression analysis showed that higher levels of LMR upon admission are significantly correlated with adverse neurological function (NIHSS≥5, OR=1.631) and adverse functional outcomes (mRS≥3, OR=1.341) upon discharge, demonstrating its strong ability to predict short-term prognosis. Meanwhile, indicators such as NLR, SIRI, and PIV showed a correlation with reduced risk of adverse outcomes (OR<1). All inflammatory indicators did not achieve statistical significance in predicting 3-month long-term prognosis, which may be limited by lost follow-up data and reduced sample size. These findings suggest that early inflammatory states mainly affect the recent recovery trajectory. Table 3Logistic regression of discharge and 3 months characteristicsAdmissionDischarge3 monthsNIHSS ≥ 5mRS ≥ 3NIHSS ≥ 2mRS ≥ 2OR(95% CI) P OR(95% CI) P OR(95% CI) P OR(95% CI) P NLR0.715(0.515–0.991)0.04420.689(0.515–0.923)0.01261.002(0.648–1.552)0.99111.04(0.808–1.339)0.7593PLR0.993(0.987–0.998)0.01190.994(0.989–0.999)0.0130.997(0.99–1.005)0.4931.001(0.996–1.005)0.7764LMR1.631(1.155–2.304)0.00541.341(1.045–1.72)0.02121.271(0.822–1.967)0.28080.998(0.849–1.172)0.9772SII0.998(0.997–0.999)0.00710.998(0.997-1)0.01351(0.998–1.002)0.82081(0.999–1.001)0.7445SIRI0.423(0.247–0.724)0.00170.45(0.277–0.73)0.00120.987(0.47–2.072)0.97210.975(0.647–1.47)0.905PIV0.997(0.995–0.999)0.00130.997(0.996–0.999)0.00341(0.997–1.003)0.99981(0.998–1.001)0.9314Abbreviations: OR Odds rate, CI Confidence interval, NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio, LMR Lymphocyte to monocyte ratio, SII Systemic immune-inflammation index, SIRI Systemic inflammatory response index, PIV Pan immune-inflammation value, NIHSS National Institutes of Health Stroke Scale, mRS Modified Rankin Scale Logistic regression of discharge and 3 months characteristics 0.689 (0.515–0.923) 0.994 (0.989–0.999) 1.341 (1.045–1.72) 0.998 (0.997-1) 0.45 (0.277–0.73) 0.997 (0.996–0.999) Abbreviations: OR Odds rate, CI Confidence interval, NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio, LMR Lymphocyte to monocyte ratio, SII Systemic immune-inflammation index, SIRI Systemic inflammatory response index, PIV Pan immune-inflammation value, NIHSS National Institutes of Health Stroke Scale, mRS Modified Rankin Scale The predictive ability of risk factors for the prognosis of AIS patients In stroke patients, whether there are vascular risk factors, including age, gender, hypertension, diabetes and so on, will affect the prognosis of patients to a certain extent. Meanwhile, to a certain extent, the level of inflammatory factors can also be influenced by certain diseases or lifestyle habits. Thus, the relationships between the inflammatory factors and vascular risk factors needs to be further investigated. Therefore, we conducted a correlation analysis between the two and reported the results in Table 4. After stroke, in terms of age, we found that except for LMR, all other inflammatory indicators at admission are positively correlated. For gender, except for PLR and LMR, there is also a positive correlation in the levels of the other four inflammatory factors at admission. Meanwhile, we were surprised to find that hypertensive patients had higher levels of SII, NLR and SIRI, although there was no significant statistical difference. There is a weak relationship between inflammatory indicators and clinical risks including smoking, drinking, diabetes and atrial fibrillation. Most only one or two inflammatory factors have a positive correlation and there is no significant statistical difference. For hyperlipidemia and hyperhomocysteinemia, the levels of most inflammatory factors at admission showed a positive correlation, although only some showed statistical differences. In summary, it revealed that the level of inflammatory indicators at admission is influenced to varying degrees by vascular risk factors, which in turn affected the progression and prognosis of AIS patients. Therefore, whether to combine with vascular risk factors may to some extent predict the prognosis of stroke. Table 4Relationships between inflammatory markers and clinical risksAdmissionYearSexHypertensionDiabetesAtrial fibrillationSmokedrinkHyperlipidemiaHyperhomocysteinemiat/z/x2 P t/z/x2 P t/z/x2 P t/z/x2 P t/z/x2 P t/z/x2 P t/z/x2 P t/z/x2 P t/z/x2 P NLR0.205760.00170.091370.16730.055730.4002-0.04290.51830.041330.5775-0.016980.8191-0.019070.7973-0.087040.240.074740.259PLR0.17020.0094-0.017160.7949-0.001080.9869-0.10640.1068-0.038610.6008-0.050790.4911-0.088240.2310.061860.40160.022020.7387LMR-0.39831< 0.001-0.154230.019-0.070680.2847-0.047090.4773-0.078550.28790.039470.59370.034720.639-0.056230.4471-0.30664< 0.001SII0.08370.2060.090370.172-0.020360.7588-0.146690.0264-0.05890.4271-0.013660.854-0.082180.26740.060660.41340.03890.5572SIRI0.32517< 0.0010.232480.00040.057610.38550.020650.75650.051060.49240.007470.9201-0.029180.6950.056690.44590.30792< 0.001PIV0.234730.00030.207010.00160.03150.6354-0.065980.3212-0.031260.6744-0.012360.8681-0.0620.40440.118360.11050.25806< 0.001Abbreviations: NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio, LMR Lymphocyte to monocyte ratio, SII Systemic immune-inflammation index, SIRI Systemic inflammatory response index, PIV Pan immune-inflammation value Relationships between inflammatory markers and clinical risks Abbreviations: NLR Neutrophil to lymphocyte ratio, PLR Platelet to lymphocyte ratio, LMR Lymphocyte to monocyte ratio, SII Systemic immune-inflammation index, SIRI Systemic inflammatory response index, PIV Pan immune-inflammation value Discussion The pathological and physiological mechanisms underlying the occurrence and development of AIS are complex, involving multiple stages such as energy metabolism disorders, cell apoptosis, excitotoxicity, oxidative stress, and inflammatory response [20]. Inflammatory response is not only a pivotal component in the pathophysiological cascade of AIS, but also a key determinant of clinical prognosis, influencing both acute injury severity and long-term functional recovery. Besides, it is an important factor affecting the prognosis of AIS patients [21]. Therefore, anti-inflammatory therapy has become a clear target for AIS treatment [22], and it is necessary to monitor the levels of inflammatory factors acutely and chronically. Moreover, after stroke, due to the interruption of blood flow, white blood cells in the blood vessels are activated, necrotic cells and damaged tissues release pro-inflammatory mediators, which leads to an inflammatory cascade reaction and promots secondary brain damage [23, 24]. Therefore, obtaining a biomarker with high sensitivity and specificity that may be associated with the prognosis of stroke patients is of great significance for early clinical intervention and may improve the clinical prognosis of patients to a certain extent. Firstly, we found that inflammatory markers such as neutrophils, monocytes, NLR, SIRI and PIV in AIS patients were significantly higher than those in the control group upon admission, while lymphocyte, platelet, and hemoglobin levels were significantly reduced, indicating a significant systemic inflammatory response and immune imbalance in the acute phase. This is consistent with previous studies that ischemic brain injury can rapidly activate the peripheral immune system, promote the infiltration of inflammatory cells such as neutrophils and monocytes into the brain, and release pro-inflammatory cytokines (such as Interleukin-1β, Tumor Necrosis Factor-α), thereby exacerbating blood-brain barrier disruption and secondary nerve damage [25]. The systemic inflammatory markers such as NLR, SIRI and PIV significantly increased in AIS patients during the acute phase observed in this study, while lymphocyte counts decreased, which is a direct manifestation of this pathological process in peripheral blood. Notably, the dynamic changes in several markers during disease progression are particularly significant. NLR reaches its peak in the early stages of the disease and gradually decreases with the course of the disease, which may reflect the early neutrophil dominated inflammatory response under acute stress. PIV and SIRI peak at discharge and 1 month after onset, suggesting that monocytes and platelets may play important roles in the regulation of inflammation in the subacute phase. It is worth noting that SII continues to fluctuate after onset and remains at a high level at 3 months, suggesting that it may serve as an indicator of persistent inflammatory status. In contrast, LMR levels were generally lower and fluctuated less among patients, while PLR showed no significant difference between groups and lacked dynamic changes, indicating that different inflammatory indicators have different temporal expression characteristics during stroke. The inflammatory response after stroke is not only related to the severity of acute phase injury, but its chronicization process is also a key pathological mechanism affecting the long-term recovery of neurological function [26].Univariate correlation analysis revealed that, with the exception of LMR, most inflammatory markers upon admission exhibited positive correlations with the severity of neurological deficits (as assessed by NIHSS and mRS scores) both at admission and at discharge. This preliminary finding suggests a link between the intensity of the acute inflammatory response and initial injury severity. At the same time, univariate regression analysis indicates that the PIV, NLR, and PLR at 3 months of onset are still correlated with the degree of neurological deficit at discharge, suggesting that persistent high inflammation may be associated with poor prognosis. A study had shown that after acute brain injury, inflammatory cells are continuously activated, transitioning from acute protective function to chronic destructive state, and proposed that chronic neuroinflammation is the core sequelae after nerve injury [27]. The study conducted by Iadecora et al. also clearly indicated that immune responses can last for weeks to months and are associated with adverse recovery outcomes such as increased risk of cognitive impairment, depression, and dementia [28]. A large number of high-quality literature published in authoritative journals have confirmed that various inflammatory factors such as C-reactive protein, Interleukin-6, tumor necrosis factor - α and others are effective biomarkers for predicting stroke outcomes [29, 30]. Therefore, in this article, we explore the prognostic value of inflammatory markers through multiple logistic regression analysis. However, after adjusting for confounding variables such as age and vascular risk factors, the multiple logistic regression analysis presented more complex results. A higher admission LMR level is independently associated with larger infarct volume and adverse functional outcomes (OR>1), supporting its potential as a risk marker for AIS severity. This finding is consistent with recent reports by Wang et al. in the elderly AIS population [16]. On the contrary, indicators such as NLR, SIRI, and PIV showed a trend of reducing the risk of adverse outcomes (OR<1), which contradicts the common belief that elevated inflammatory markers indicate poor prognosis [27]. Firstly, post-stroke inflammation is a double-edged sword. Moderate early inflammatory response is crucial for clearing necrotic tissue and initiating repair processes [7]. Therefore, the elevation of certain inflammatory indicators may partially reflect the body's effective immune response rather than pure pathological damage. The association of OR<1 in this research model may capture the inflammatory response dimension that has potential protective effects. Secondly, NLR, SIRI, PIV and other indicators share neutrophil and lymphocyte counts in their calculations, indicating a high degree of collinearity. In multivariate models, this collinearity may lead to instability and mutual cancellation of effect estimates. In addition, adjusting for strong prognostic factors such as age and underlying diseases may partially offset the direct effect of inflammation on prognosis, especially when these factors are closely related to inflammation levels. Finally, this study is an observational single center design. Although multiple confounding factors have been adjusted, residual confounding factors (such as unrecorded details of reperfusion treatment, specific time from onset to blood collection, infarct site, etc.) may affect the results. There may be essential differences in the immune inflammatory response patterns of stroke with different causes (such as cardiogenic embolism and atherosclerosis) or infarct sites. Future research needs more detailed subgroup analysis. Even though this study established an association between systemic inflammatory markers and comprehensive neurological prognosis (evaluated through NIHSS and mRS), whether inflammatory response has specific effects on specific regions needs more investigation. The assessment tool used in this study is mainly used for evaluating motor function and overall disability level, with limited ability to assess cognition (such as memory, executive function) or emotion (such as depression, apathy). Meanwhile, existing literatures had strongly suggested that inflammatory response may have a significant impact on cognitive and emotional regions. For example, pro-inflammatory cytokines have been shown to disrupt hippocampal synaptic plasticity and affect the function of the prefrontal cortex, leading to the cognitive impairment and fatigue after AIS [31]. Therefore, we speculate that the association between the observed inflammatory markers and poor prognosis in this study may be caused by cognitive and emotional disorders driven by inflammatory response. The verification of this inference requires future research to combine such blood biomarkers with detailed neuropsychological tests (such as montreal cognitive assessment, trail making test, etc.) to elucidate the specific role of inflammatory response in different functional recovery after stroke, and provide precise biomarkers for anti-inflammatory treatment strategies targeting specific functional domains (such as cognition). ROC curve and the AUC can further evaluate the discriminative ability of these indicators from a clinical perspective [32, 33]. NLR showed moderate discriminative ability upon admission (AUC=0.654), which is consistent with the conclusions of many published literature [34], making it a tool of early risk assessment accessiblely and effectively. PIV showed higher discriminative ability at discharge (AUC=0.725), a comprehensive indicators, which integrate neutrophil, platelet, lymphocyte, and monocyte information, may have better performance in evaluating subacute prognosis, as emphasized in recent studies by Feng et al. [35]. The outcome of stroke is of highly heterogeneous and not a single disease process. Its functional recovery, mortality rate, recurrence rate and long-term disability level are influenced by a series of immutable (such as age, gender) [36] and mutable (such as comorbidities, acute phase management) factors. The findings of this study are consistent with this consensus. In particular, the association between the level of inflammation we observed and vascular risk factors (such as hypertension, diabetes) suggests that variable factors may indirectly affect the prognosis by regulating the inflammatory state of the body. This mechanism is consistent with recent research, which directly demonstrates the decisive role of specific variable clinical factors such as atrial fibrillation and reperfusion therapy strategies in prognosis [37, 38]. Modern stroke medical even integrates these factors through prognostic prediction models to quantify individual risk and take effect in clinical communication and decision-making [39]. Future research should focus on this direction, combining the easily accessible inflammatory biomarkers (such as NLR, PIV) explored in this study with strong clinical predictive factors (such as specific etiology, treatment parameters, etc.) that have been validated through multicenter studies [37, 38], to develop more accurate and personalized comprehensive predictive tools. This can not only improve the accuracy of prognosis judgment, but also potentially identify differentiated treatment targets and rehabilitation strategies for patients with different inflammatory phenotypes. In our study, we observed the correlation between inflammatory markers and a variety of vascular risk factors (such as age, hypertension, diabetes and so on). We found a positive correlation between age and most inflammatory markers, suggesting that as age increases, the level of inflammation when stroke happened is higher. As mentioned in the literature, age is one of the strongest predictors of stroke prognosis [40]. When stroke happened, brain will be hypoxic and necrotic, which will release a large number of "danger signals". These signals immediately trigger a strong inflammatory cascade reaction. In the context of aging, this reaction will become exceptionally intense and out of control. Similarly, we also found a similar correlation in age, female patients have higher levels of inflammation. Perhaps due to the unique immune inflammatory environment shaped by sex hormones [41]. In addition, in the correlation analysis of various complications, we also found a positive correlation with some inflammatory factors. These results indicate that underlying diseases and patient characteristics may indirectly affect the progression and prognosis of stroke by regulating the inflammatory state of the body. This finding is consistent with the clinical consensus that stroke prognosis is influenced by multiple factors [42], and suggests that in future prediction models, combining inflammatory markers with traditional vascular risk factors may significantly improve prediction accuracy. The value of these composite inflammatory markers lies in the integration of the information from different immune pathways. The elevation of NLR reflects an imbalance between the exacerbation of the pro-inflammatory response dominated by neutrophil and the decline of the immune regulation dominated by lymphocyte [26]. PIV, as a more comprehensive indicator that encompasses neutrophils, platelets, lymphocytes and monocytes, may comprehensively reflect the complex interactions of pro-inflammatory state, thrombotic inflammation and immune exhaustion [15]. SIRI also emphasizes the central role of neutrophils and monocytes in expanding the inflammatory cascade after acute ischemic stroke. Therefore, these indicators are not just numbers, they are quantitative manifestations of systemic immune inflammatory response after stroke, and the high levels of the indicators may indicate more severe damage and more unfavorable immune microenvironment for repairment. In addition, there maybe lay neuroimmune mechanisms between inflammatory markers and stroke prognosis. Firstly, systemic inflammatory response can disrupt the blood-brain barrier [43], making it easier for peripheral inflammatory mediators to enter the central nervous system and exacerbate the neuroinflammation response. In the central nervous system, microglia are overactivated and polarized towards the pro-inflammatory M1 phenotype, releasing a large number of cytokines including Tumor Necrosis Factor-α and Interleukin-1β [28]. These inflammatory factors have been shown to directly inhibit synaptic plasticity and long-term potentiation in key brain regions such as the hippocampus, thereby impairing cognitive functions such as learning and memory [32]. More importantly, as we observed, some biomarkers are still associated with the prognosis at 3 months, and this chronic, low-grade inflammatory state is considered the core mechanism that triggers secondary neurodegeneration. It can lead to progressive damage to distant brain regions related to cognition, ultimately increasing the risk of post-stroke dementia [44]. Therefore, we speculate that these easily accessible systemic inflammatory markers may be effective indicators reflecting the activity level of harmful neuroimmune processes mentioned above, providing important clues for predicting the neuropsychological recovery of stroke patients. Compared with some previous studies on the relationship between inflammatory markers and stroke prognosis, our study is more innovative. Firstly, most previous studies have only focused on one or two indicators. We systematically compared six inflammatory biomarkers (including some less studied indicators, like PIV, SIRI) in parallel within the same patient cohort for the first time. It can help clinicians to get the indicator with the most prognostic value at a specific time point (such as admission or discharge) and provide them the direct evidence for selecting the best biomarker. Secondly, we not only observed the level of inflammatory markers at admission, but also tracked the dynamic changes of these indicators at discharge, one month and three months. This reveals the dynamic state of different indicators at different stages, and may provide more ideas about the inflammatory state during the subacute and recovery phases of stroke. Last but not least, indicators such as SIRI and PIV that have been poorly studied before, our study provided independent validation data from a Chinese cohort and found that PIV has excellent performance in predicting discharge prognosis, which points the way for future research. Meanwhile, there still lies some limitations. Firstly, Although multiple important confounding factors have been corrected, there may still be unmeasured residual confounding factors (such as acute phase and secondary preventive drugs, infarct site, time from symptom onset to blood sampling etc). Due to sample size limitations and data structure issues, not all potential confounding factors such as infarct site were included in the final model for analysis, which may have some impact on the results. This suggests that future studies with larger sample sizes and prospective designs are needed to further validate and accurately quantify the independent predictive value of inflammatory markers. Secondly, as a single-center study, our patient cohort may be subject to selection bias due to regional healthcare practices and demographic characteristics. However, the baseline characteristics of our patients, such as age, sex and comorbidities, are largely consistent with those reported in previous multi-center studies [45, 46]. It suggested that our sample is reasonably representative. To generalize our findings, future prospective, multi-center studies with larger sample sizes are warranted to externally validate the prognostic value of these inflammatory markers across diverse populations and clinical settings. Thirdly, there is a significant data loss at long-term follow-up points (such as three months), which may weaken our ability to comprehensively assess the relationship between inflammatory markers and long-term prognosis, resulting in some associations not reaching statistical significance. Therefore, the results of the time point of 3-month should be defined as preliminary investigation results, and relevant findings need to be further validated in large sample prospective studies. Future multicenter prospective studies with larger sample sizes are needed to validate our findings. It should be emphasized that this observational study means that the associations we reported cannot be directly explained as causal relationships. These inflammatory markers are powerful prognostic markers, but whether they play a direct causal role in stroke pathophysiology or only reflect underlying, complex pathological states remains to be elucidated. The findings from this study pave the way for several promising research directions. Firstly, future studies should aim to integrate these systemic inflammatory markers with multimodal neuroimaging biomarkers. Correlating markers like PIV and NLR with MRI-derived measures of infarct characteristics, blood-brain barrier integrity, white matter hyperintensity load, and structural connectivity (via diffusion tensor imaging) could elucidate the central nervous system structural substrates underlying the inflammation-poor prognosis relationship, enabling a more comprehensive damage assessment. Secondly, moving beyond cellular counts to multiplex cytokine profiling is crucial. Combining NLR or PIV with measurements of specific cytokines (Interleukin-6, Interleukin-1β, Tumor Necrosis Factor-α) would help identify the key inflammatory pathways driving poor outcomes and could potentially identify patient subsets for targeted immunomodulatory therapies. Finally, prospective studies incorporating detailed domain-specific neuropsychological assessments are needed to determine whether these inflammatory markers can differentially predict recovery trajectories in motor, cognitive, and affective functions, ultimately guiding personalized rehabilitation strategies. Conclusions This study suggests that systemic inflammatory response plays an important role in the occurrence, development, and prognosis of AIS. Different inflammatory indicators show dynamic changes during the course of the disease and are closely related to the degree of neurological dysfunction and short-term prognosis. Although some results differ from traditional cognition and have limited long-term predictive power, inflammatory markers still have the potential to serve as auxiliary biomarkers, especially when combined with clinical risk factors, which may provide new ideas for individualized assessment and management of stroke patients. These indicators, which are easy to calculate from routine blood tests, are expected to become practical tools for clinical doctors to evaluate the condition and prognosis of AIS patients, providing reference for early intervention and personalized treatment. Acknowledgements We thank all the patients who participated in this study. Abbreviations Acute ischemic stroke Neutrophil to lymphocyte ratio Platelet to lymphocyte ratio Lymphocyte to monocyte ratio Systemic immune-inflammation index Systemic inflammatory response index Pan immune-inflammation value National Institutes of Health Stroke Scale Modified Rankin scale Standard deviation Receiver operating characteristic Area under the curve Odds rate Confidence interval Authors’ contributions XI-LE ZHANG designed the study. QING LIN designed the clinical information of patients. YU XIAO, LU-YU LI, YONG WANG, XIAOXIAO TAO performed the experiments. XI-LE ZHANG, LU-YU LI, YA-NI DONG, YONG WANG collected the clinical information of patients. XI-LE ZHANG, XIAOXIAO TAO, YONG WANG analyzed the data. XI-LE ZHANG, YONG WANG and XIAOXIAO TAO wrote the manuscript. YONG WANG critically edited the manuscript. All authors reviewed the manuscript. Funding This project was supported by the China Postdoctoral Science Foundation (No. 2024M750535); the National Nature Science Foundation of China (grant nos. 82501570); Zhejiang Provincial Department of Health Project 2021 (2021KY1211); Zhejiang Province Traditional Chinese Medicine Project 2024 (2024ZL197); Zhejiang Provincial Department of Health Project 2025 (2025KY1855). Data availability The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Data sharing will be conducted under a formal data sharing agreement that ensures compliance with institutional policies and Chinese data protection regulations. All shared data will be de-identified following established anonymization protocols, including removal of direct identifiers (names, identification numbers, dates of birth) and quasi-identifiers that could potentially lead to re-identification. Requests for data access should be directed to the corresponding author and will be evaluated on a case-by-case basis for legitimate research purposes. Declarations Ethics approval and consent to participate Studies involving human patients were reviewed and approved by the ethical review board of Taizhou First People’s Hospital has approved our study (No.2020-KY022-01). Written informed consent to participate was obtained from all individual participants included in the study. Especially for adults with cognitive decline, we also obtain signed informed consent from legal guardians or appropriate representatives to participate on their behalf. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Xiaoxiao Tao, Email: taizhoutaotao@163.com. Yong Wang, Email: yong_wang@fudan.edu.cn. References Associated Data Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Data sharing will be conducted under a formal data sharing agreement that ensures compliance with institutional policies and Chinese data protection regulations. All shared data will be de-identified following established anonymization protocols, including removal of direct identifiers (names, identification numbers, dates of birth) and quasi-identifiers that could potentially lead to re-identification. Requests for data access should be directed to the corresponding author and will be evaluated on a case-by-case basis for legitimate research purposes.