Introduction

A Global Burden of Disease study showed that stroke is the second leading cause of death and the third leading cause of disability worldwide1. In 2019, there were 12.2 million new strokes and 6.55 million deaths from stroke. The overall lifetime risk of stroke in China is 39.9%, which is the highest in the world, meaning that approximately two out of every five Chinese people will suffer a stroke. In addition, stroke is the number one cause of disease-induced years of life lost in China2. AIS caused by cerebral artery blockage accounts for approximately 87% of all strokes3,4. Most patients with cerebral infarction have serious complications, causing heavy social and family burdens1,5. Therefore, there is an urgent need to find novel biological markers to predict disease progression and functional prognosis to improve the current status of acute stroke patient diagnosis and treatment.

AIS patients can be treated with tissue fibrinogen activator, intravenous thrombolytic agents, or mechanical thrombectomy (using a stent retriever or endovascular therapy) to reduce brain tissue damage through reperfusion within the therapeutic time window6,7. Due to the strict time constraints and range of intraoperative and postoperative complications, only a small proportion of patients benefit from mechanical thrombectomy and intravenous thrombolysis8. Based on their clinical status, most clinical patients still choose conservative medical treatment. To improve the poor prognosis of AIS patients, targeting neuroinflammation to alleviate this poor prognosis has become a research hotspot in recent years.

Ischaemic stroke is characterised by multiple highly correlated neuropathological processes in which an intense and persistent inflammatory response plays a crucial role in worsening brain damage7,9. A large body of evidence suggests that postischemic inflammatory responses are associated with acute blood‒brain barrier (BBB) disruption, vasogenic oedema, haemorrhagic transformation, and severe neurological consequences10. Cell counts and their composite inflammatory indices, such as the NLR11, NHR12, PLR13, MLR14, SII15, and SIRI16, are regarded as valuable biological predictors of prognosis in AIS patients. These cells are readily available in peripheral blood and are considered classical haematological markers of systemic inflammation. Furthermore, all six composite inflammatory markers can predict the prognosis of AIS patients. No studies have compared the precision of several composite inflammation indicators in predicting the prognosis of stroke.

We collected clinical data from 516 patients with acute cerebral infarction and statistically analysed the clinical value of the above six composite inflammatory indicators for predicting the short-term prognosis of AIS patients. The experimental design is shown in Fig. 1.

Figure 1
figure 1

Study flow chart. AIS: acute ischaemic stroke; mRS: modified Rankin scale; NLR: neutrophil/lymphocyte ratio; NHR: neutrophil/high-density lipoprotein cholesterol ratio; PLR: platelet-to-lymphocyte ratio; MLR: monocyte-to-lymphocyte ratio; SII: systemic immune-inflammatory index; SIRI: systemic inflammatory response index.

Full size image

Methods

Subject population

This study included a total of 516 patients with AIS who were hospitalised at Nantong Third People's Hospital between September 2019 and february 2024.

Inclusion criteria:

  1. 1.

    Age 18 years or older.

  2. 2.

    First stroke, with a diagnosis of AIS within 24 h of onset (the diagnosis of ischaemic stroke was confirmed by the use of noncontrast computed tomography and computed tomography angiography).

  3. 3.

    Head MRI completed within 48 h.

  4. 4.

    Standardised treatment for acute and secondary prevention of ischaemic stroke.

Exclusion criteria:

  1. 1.

    Previous history of AIS or cerebral haemorrhage;

  2. 2.

    Early selection for reperfusion therapy in the IS;

  3. 3.

    Pregnancy or breastfeeding;

  4. 4.

    Severe cardiac, pulmonary, hepatic, renal, neoplastic or autoimmune disease or combined severe infection;

  5. 5.

    Automatic withdrawal from the study or loss to follow-up.

The study strictly adhered to the Declaration of Helsinki and was reviewed and approved by the Ethics Committee of the Third People's Hospital of Nantong (Ethics No: YJ2022034). All patients voluntarily enrolled in this study and signed an informed consent form. For privacy purposes, the subjects' personally identifiable information was anonymized.

Data collection

A complete history, neurological examination and baseline data (basic information (age, sex and body mass index), vascular risk factors (hypertension, diabetes mellitus, dyslipidaemia, atrial fibrillation, history of smoking and alcohol consumption, history of previous stroke, peripheral arterial disease and coronary arterial disease), and history of medication use (antiplatelets, anticoagulants, statins, etc.) were documented by a specialised neurologist for all the patients within four hours of admission to hospital. Electrocardiography, chest X-ray, brain CT or MRI and carotid ultrasound were completed in all patients within 24 h of admission.

The mRS score was used to define short-term outcomes at 30 days, an mRS score > 2 defining adverse outcomes.

Blood sampling and calculation of complex inflammatory indicators

Blood routine values, including high-density lipoprotein (HDL), low-density lipoprotein (LDL), C-reactive protein (CRP), total cholesterol, monocyte, neutrophil, and lymphocyte data, were collected within one hour of admission. Peripheral blood cells were counted using an automated analyser (XT-1800i, Sysmex, Kobe, Japan) and used to calculate the compound inflammation ratio.

NLR = neutrophil/lymphocyte ratio; NHR = neutrophil count/high-density lipoprotein cholesterol ratio; PLR = platelet count/lymphocyte ratio; MLR = monocyte/lymphocyte ratio; SII = platelet count × neutrophil count/lymphocyte count; SIRI = monocyte × neutrophil/lymphocyte.

Statistical analysis

The Kolmogorov‒Smirnov test was used to test the normality of the distribution of the data. The mean ± standard deviation was used to describe continuous variables with a normal distribution, and the median (quartiles [Q25, Q75]) was used to describe continuous variables with a nonnormal distribution. The independent-sample t tests and Mann‒Whitney U tests were used to analyse between-group differences in continuous variables. Categorical variables are presented as n (%), and differences between the groups were analysed with the chi-squared test.

A subject operating characteristic (ROC) curve was drawn by plotting the sensitivity versus 1-specificity and calculating the area under the curve (AUC). Optimal thresholds were calculated through Youden's J test. Binary logistic regression was used to test the independent effects of the 6 composite inflammation ratios on the short-term prognosis (30 days) of AIS patients. The data were visualised with SPSS Statistics 25.0 software (IBM Analytics) and GraphPad (Prism 8). p < 0.05 was the threshold for statistical significance.

Results

Baseline data

A total of 516 patients with acute stroke were included in this study. To eliminate data bias, after excluding endovascular therapy and thrombolytic therapy, standard medical treatment remained in 306 patients. As shown in Table 1, 205 (67%) of the 306 patients with AIS had a good prognosis, and 101 (33%) had a poor prognosis. The percentage of males, proportion of cardiac embolism type (CE, TOAST classification), age, percentage of patients with coronary artery disease, percentage of patients with atrial fibrillation, white blood cell count, percentage of neutrophils, apolipoprotein A1 (ApoA1), acid glycoprotein, PLR, NHR, NLR, MLR, SII, and SIRI were significantly greater in the poor prognosis group than in the good prognosis group (P < 0.05). ApoB/ApoA1 and the lymphocyte percentage in the poor prognosis subgroup were significantly lower than those in the good prognosis subgroup. The remaining baseline information was not significantly different.

Table 1 Comparison of general information and clinical characteristics of patients with AIS with good and poor prognosis.
Full size table

Correlation analysis of the complex inflammation ratio and short-term prognosis

Pearson's correlation coefficients are shown in Fig. 2. All six inflammation indicators, PLR, NHR, NLR, MLR, SII, and SIRI, were positively correlated with the mRS score. The correlation coefficients of the SIRI, PLR, NHR, NLR, MLR, SII, and MLR were 0.304, 0.262, 0.222, 0.302 , 0.308 and 0.277, respectively (P < 0.0001).

Figure 2
figure 2

Correlation analysis. AIS: acute ischaemic stroke; mRS: modified Rankin scale; NLR: neutrophil/lymphocyte ratio; NHR: neutrophil/high-density lipoprotein cholesterol ratio; PLR: platelet-to-lymphocyte ratio; SII: systemic immune-inflammatory index; SIRI: systemic inflammatory response index.

Full size image

Multifactor analysis of AIS patients with short-term prognosis

After correcting for confounders, binary logistic regression analysis revealed that SIRI (OR 5.912, 95% CI 1.653 to 21.14, P = 0.006), NLR (OR 1.334, 95% CI 1.038 to 1.715, P = 0.025) and PLR (OR 1.02, 95% CI 1.006 to 1. 034, P = 0.005) were independent risk factors for poor short-term prognosis (30 days) in patients with AIS (Table 2, Fig. 3).

Table 2 Multinomial logistic regression models for the severity of AIS.
Full size table
Figure 3
figure 3

Tree diagram: correlations between the PLR, NLR, and SIRI and AIS prognosis. NLR: neutrophil-to-lymphocyte ratio; PLR: platelet-to-lymphocyte ratio; SIRI: systemic inflammatory response index.

Full size image

Clinical value of SIRI, NLR, and PLR in predicting poor outcomes in AIS patients

The subject ROC curves (Fig. 4) were drawn based on sensitivity versus 1-specificity and area under the curve, and the AUC of SIRI for predicting poor prognosis in patients with acute ischaemic stroke was 0.748 (95% CI: 0.688 to 0.807), with a sensitivity of 63.4% and specificity of 82%. The positive predictive value of SIRI was 81.95%(168/205), and the negative predictive value was 62.38%(63/101). SIRI was at the optimal cut-off value of 1.778 × 109/L. NLR predicted poor prognosis in patients with acute ischaemic stroke. The AUC for NLR for predicting poor prognosis in patients with acute ischaemic stroke was 0.745(95% CI = 0.686–0.804), and when NLR was at the optimal cut-off value of 2.689 × 109/L, the sensitivity was 83.2% and the specificity was 60.5%. The positive predictive value of NLR was 87.94%(124/141), and the negative predictive value was 50.91%(84/165). Similarly, the area under the curve (AUC) for PLR for predicting poor prognosis in patients with acute ischaemic stroke was 0.676(95% CI = 0.613–0.739); when PLR was at the optimal cut-off value of 134.149 × 109/L, the sensitivity was 63.4% and the specificity was 69.3%. The positive predictive value of PLR was 79.12% (144/182), and the negative predictive value was 50.81% (63/124).

Figure 4
figure 4

Correlations between the PLR or NLR and between SIRI and AIS disease severity. NLR: neutrophil-to-lymphocyte ratio; PLR: platelet-to-lymphocyte ratio; SIRI: systemic inflammatory response index.

Full size image

Discussion

In this study, for the first time, we evaluated the clinical value of six composite inflammation ratios (NLR, NHR, SII, SIRI, MLR, and PLR) at admission for predicting 30-day prognosis in patients with AIS. Our results showed that all six inflammatory indicators were associated with poor short-term prognosis in patients with AIS. Among them, the NLR, SIRI and PLR were more advantageous for predicting AIS prognosis. Our study provides basic clinical evidence for the relationship between inflammation and AIS. Most importantly, our study suggests a potentially simple method to predict and improve unfavourable functional outcomes in AIS patients through the use of peripheral-blood inflammatory markers.

Stroke is caused by an obstruction of a cerebral artery or its branches, and the formation of an embolism reduces the blood supply to the brain, damaging neurons. Inflammation, oxidative stress, immune cell recruitment and microglial activation contribute to the development of stroke, causing irreversible damage to brain tissue17. Among these factors, inflammation is considered a key factor in all stages of stroke development, from occlusion in the acute phase to postischemic repair18. Under normal conditions, inflammation is a defence response that removes toxic substances and limits their harmful effects6. A large body of evidence suggests that some inflammatory responses play an unfavourable role in the development of stroke19. BBB destruction is an important pathophysiological process in AIS that leads to severe malignant brain oedema and haemorrhagic transformation. Rapid activation of immune cells plays a key role in BBB destruction after ischaemic stroke. Immune cells (e.g., neutrophils, monocytes, and lymphocytes) that infiltrate brain tissue can increase BBB permeability by secreting inflammatory factors and affecting microvessels. Curiously, they play opposite roles in the later stages of ischaemic stroke, when they can promote BBB repair and angiogenesis20,21.

Given that inflammatory cells play an important role in the development of IS, researchers have begun to explore ways to improve the prognosis of stroke patients by targeting inflammatory cells to alleviate brain tissue damage. In their mouse model of a tethered middle cerebral artery, Hu et al. found that suppressing the number of neutrophils in the ischaemic region of acute ischaemic stroke minimised cell death and promoted functional recovery of the nervous system22; Huang et al. found that microglial IL-1RA ameliorates brain injury after ischaemic stroke by inhibiting astrocytic CXCL1-mediated neutrophil recruitment23. Al Ruwaili et al. demonstrated that inhibition of phosphodiesterase 5 (PDE5) promotes neurological recovery in AIS patients by modulating the brain cyclic adenosine monophosphate (cAMP)/cyclic guanosine monophosphate (cGMP)/nitric oxide (NO) pathways24. Studies on reducing the severity of AIS and promoting neurological recovery by alleviating neuroinflammation have been carried out in mouse models, and there is an urgent need for additional researchers to increase their efforts.

Our findings showed that all six inflammatory indicators tested were negatively associated with short-term prognosis in patients with AIS, providing valuable clinical evidence for further basic experiments. Our study does have some limitations. First, it was a retrospective study conducted at a single centre and therefore suffered from selection bias. Second, the sample was small, and the findings need to be validated in other larger populations. Third, we analysed only peripheral-blood data at admission and should have tracked the dynamic changes in these values. A multicentre prospective study will be considered in the future to further explore the relationship between the compound inflammation ratio and AIS incidence.

Conclusion

SIRI, PLR and NLR are important predictors of early functional outcomes in AIS patients and can provide a reference for early diagnosis and treatment by clinicians. Further studies dynamically monitoring SIRI, PLR and NLR are needed to better understand the predictive value of these parameters in larger cohorts. Therefore, AIS patients with elevated SIRI, PLR, and NLR should be closely monitored, as these values may suggest new therapeutic strategies to improve poor outcomes in AIS patients.