Introduction

Acute cerebral infarction (ACI) accounts for approximately 75% of cerebrovascular diseases, causing serious social problems and heavy economic burden1,2. Early neurological deterioration (END), a common adverse event after ACI, is very frequently associated with risk of poor prognosis in such patients3,4. Recently, there has been a growing interest in discriminating new biological markers for ACI to aid in predicting prognosis, assessing disease severity and exploring pathophysiological process5,6.Platelet-derived microparticles (PDMPs) are small particles that are shed from the plasma membrane and enter the peripheral circulation after platelet activation7. PDMPs play important roles in many physiological and pathological processes, such as deep vein thrombosis, stroke, and acute myocardial infarction8,9,10. Reportedly, anti-platelet therapy can prevent END and improve 6-month outcome in patients with acute stroke, indicating that END could be related to platelet activation11. Compelling data have shown that PDMPs may boast critical effects in ACI. Chiva-Blanch et al.12 found that the concentration of PDMPs was significantly increased in the ischemic stroke patients compared to the controls. Schrick et al.13 reported that some platelet derived microparticles may monitor the response to anti-platelet therapy and predict the next vascular event in stroke patients. Chiva-Blanch et al. found that the concentration of PDMPs was significantly increased in the ischemic stroke patients compared to the controls. Thus, it is hypothesized that PDMPs may be potential prognostic predictors of END and neurological outcome following ACI. Here, serum PDMPs levels were quantified in a cohort of patients with ACI, in order to reveal the relationship between serum PDMPs levels, END and neurological outcome of ACI.

Methods

Study design, patient and control enrollments, and ethical approval

In this prospective observational cohort study, we consecutively enrolled patients with ACI, who were admitted at the Wenzhou Central Hospital (Wenzhou, China) from January 2018 to January 2022. Inclusion criteria were as follows: (1) age of 18 years or greater; (2) hospital admission within 12 h after onset of stroke symptom; (3) the clinical diagnoses were confirmed by computerized tomography (CT) or Magnetic Resonance Imaging (MRI) according to the Chinese Stroke Association 2023 guidelines14. Exclusion criteria were in the following: (1) previous thrombolysis or endovascular therapy; (2) coexistence with other neurological diseases, such as severe traumatic brain injury, intracranial tumors and infections; (3) severe diseases in other organs, such as liver, heart, lung and kidney failure, ascites and cirrhosis; (4) refusal to participation, unqualified samples, incomplete clinical data or loss to follow-up. A group of healthy individuals were recruited as controls. This study complied with the Declaration of Helsinki and its later amends15 and its protocol was approved by the Ethics Committee at the Wenzhou Central Hospital (No. L2024-03-020). All patients’ proxies or controls themselves submitted written informed consent to participate in this study.

Data collections, END definition and outcome assessment

Some relevant data were collected, including demographic variables, medication use, medical history, cigarette smoking and alcohol drinking. The severity of the neurological deficit was assessed at admission by use of the National Institutes of Health Stroke Scale (NIHSS) score16. END was defined as poststroke 24-hour NIHSS total score increasing by ≥ 4 or the NIHSS individual score increasing by ≥ 2 compared with baseline17. According to the definition, we divided the patients into END and non-END groups. In addition, we assessed the functional outcome based on the modified Rankin score (mRS) at 6 months after onset of stroke. Patients with an mRS > 2 were considered to have a poor outcome, and ≤ 2 predicted a favorable outcome. The patients were classified into different groups by at least two experienced senior neurologists.

Laboratory examination

Peripheral venous blood samples from all patients were at admission taken and controls’ samples at entry into study. All blood samples were drawn into vacutainers containing ethylenediaminetetra acetic acid and acid citrate dextrose using 21-gauge needles to minimize platelet activation. The samples were gently mixed and then kept at room temperature for the maximum period of 1 h. Immediately after centrifugation at 8000 g for 5 min, 200 µL was collected from the upper-layer supernatant of the samples to avoid contamination by platelets. The collected samples were stored at − 80℃ until analysis. PDMPs levels were determined in a blinded fashion using the PDMPs ELISA kit (Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) according to the manufacturer’s instructions18. The coefficients of variation is below 10%. The kit used two monoclonal antibodies against glycoproteins CD42a and CD42b to measure PDMPs levels. Each sample was measured twice, and mean values were recorded.

In addition, other laboratory indicators, such as triglycerides, total cholesterol, high-density-lipoprotein cholesterol (HDL-c), low-density-lipoprotein cholesterol (LDL-c), glucose, platelet count, activated partial thromboplastin time (APTT), thrombin time (TT) and fibrinogen levels, were measured and recorded at the same time.

Statistical analysis

Data analyses were performed using Statistical Package for Social Sciences Version 23.0. (IBM Corp., Armonk, NY, USA) and R software (version 4.3.2; R Foundation for Statistical Computing). We plotted the figures via GraphPad Prism version 8.0 (GraphPad Software Inc., La Jolla, CA, USA). The Kolmogorov-Smirnov test was utilized for estimating the compatibility of normally distributed data. For continuous data, homogeneous data were expressed as the mean ± standard deviation, while non-homogeneous data were presented as median and 25th–75th percentiles.All normally-distributed data were compared using the independent-sample Student’s t-test. Data shown to be non-normally distributed were analyzed using the Mann-Whitney U test. Logistic regression analysis was used to determine the relationships between serum PDMPs levels and END as well as poor outcome at 6 months in patients with ACI. In the model, we entered the variables with a p-value of < 0.05 on the univariate analyses, and the odds ratio (OR) value and 95% confidence interval (CI) were calculated. Furthermore, we conducted subgroup analyses to assess whether the potential covariables (age, gender, hypertension, diabetes mellitus, coronary heart disease, dyslipidemia, smoker, alcohol, anti-platelet therapy and thrombolysis) modified the relationships between serum PDMPs levels and END as well as poor outcome at 6 months. The interactions between serum PDMPs levels and END as well as poor outcome at 6 months were tested using the likelihood ratio test of models with interaction terms. Additionally, we applied restricted cubic spline regression models to explore the shapes of the associations between serum PDMPs levels and END as well as poor outcome at 6 months in patients with ACI. The receiver operating characteristic (ROC) curve was constructed to investigate the predictive value of serum PDMPs levels for END and 6-month poor outcomes in ACI patients, and area under curve (AUC) and 95% CI values were calculated. The difference of P < 0.05 was defined as statistical significance.

Results

Participant characteristics

Initially, 762 patients with ACI were assessed. According to exclusion criteria, 141 patients were excluded because of the reasons outlined in Fig. 1. Finally, a total of 621 patients were under assessment. Additionally, 125 healthy individuals were recruited as controls. In Table 1, there were no significant differences between the patients and the healthy controls in age (P = 0.167), sex percentage (P = 0.266), BMI(P = 0.103), smoking (P = 0.469), and drinking (P = 0.123), while percentages of diabetes mellitus (P < 0.001) and hypertension (P < 0.001) were substantially higher in the patients than in the controls. Other baseline demographical, clinical, radiological and biochemical characteristics of patients are displayed in Table 2.

Fig. 1
figure 1

Flowing chart for screening eligible patients with acute cerebral infarction.

Full size image
Table 1 Comparisons of characteristics between controls and patients with acute cerebral infarction.
Full size table
Table 2 Baseline characteristics in relation to poor prognosis six months after acute cerebral infarction.
Full size table

Correlation of serum PDMPs levels with disease severity

In Fig. 2, serum PDMPs levels in patients with ACI were significantly higher than those in healthy controls (P < 0.001). Figure 3 shows that serum PDMPs levels were positively correlated with infarction volume (P < 0.001) and NIHSS scores at admission (P < 0.001). Also, Infarction volume was divided into three groups: small infarct (< 10 cm3), medium infarct (10–100 cm3), and large infarct (> 100 cm3). There was a significant difference in serum PDMPs levels among the three infarction volumes, with the highest levels detected in patients with large infarct (P < 0.05; Fig. 4). The patients were also divided into three groups according to NIHSS score: NIHSS score below 6, NIHSS score 6–10, and NIHSS score above 10. Figure 4 shows that serum PDMPs levels were markedly highest in patients with NIHSS score above 10 (P < 0.05).

Fig. 2
figure 2

Comparison of serum platelet-derived microparticles levels between healthy controls and acute cerebral infarction patients. PDMPs indicates platelet-derived microparticles.

Full size image
Fig. 3
figure 3

Correlations of serum platelet-derived microparticles levels with infarction volume and National Institutes of Health Stroke Scale score in patients with acute cerebral infarction. PDMPs indicates platelet-derived microparticles; NIHSS, National Institutes of Health Stroke Scale.

Full size image
Fig. 4
figure 4

Comparison of serum platelet-derived microparticles levels in the light of different infarction volumes and National Institutes of Health Stroke Scale score in patients with acute cerebral infarction. PDMPs means platelet-derived microparticles; NIHSS, National Institutes of Health Stroke Scale. *P < 0.05.

Full size image

Serum PDMPs levels and poor prognosis

Figure 5 shows that serum PDMPs levels had a significant upward trend when patients’ mRS scores increased (P < 0.001). Also, serum PDMPs levels were significantly different among subgroups with different mRS scores (P < 0.001; Fig. 5). An aggregated of 186 (30.0%) patients experienced poor prognosis. As shown in Table 1, as opposed to patients with good prognosis, those with poor prognosis had significantly elevated percentage of diabetes mellitus (P < 0.05), and displayed substantially increased age (P < 0.01), body mass index (P < 0.05), NIHSS scores (P < 0.01), infarction volume (P < 0.05), serum glucose levels (P < 0.05), platelet count (P < 0.01) and PDMPs levels (P < 0.01). In addition, the ROC curve displays that serum PDMPs levels significantly predicted poor prognosis after ACI (Fig. 6, P < 0.001), and serum PDMPs levels > 22.50 ng/L predicted poor prognosis, with medium-high sensitivity and specificity values (Maximum Youden index J = 0.316; Fig. 6). Using the binary logistic regression analysis, age, NIHSS score and serum PDMPs levels were independently associated with END (all P < 0.05; Table 3). Furthermore, under restricted cubic spline, there was a linear relationship between serum PDMPs levels and poor prognosis risk in patients with ACI (P > 0.05; Fig. 7). In Fig. 8, serum PDMPs levels significantly interacted with age (P < 0.01), but not with gender, hypertension, diabetes mellitus, coronary heart disease, dyslipidemia, smoker, alcohol, anti-platelet therapy and thrombolysis (all P > 0.05).

Fig. 5
figure 5

Receiver operating characteristic curve analysis using serum platelet-derived microparticles level as a predictor of poor prognosis six months after acute cerebral infarction. AUC, area under curve; 95% CI, 95% confidence interval.

Full size image
Fig. 6
figure 6

Linear relationship between serum platelet-derived microparticles levels and risk of poor prognosis in patients with acute cerebral infarction. PDMPs denotes platelet-derived microparticles; 95% CI, 95% confidence interval.

Full size image
Table 3 Multivariate regression analysis of related factors of poor prognosis six months after acute cerebral infarction.
Full size table
Fig. 7
figure 7

Subgroup analyses of interactions between serum platelet-derived microparticles levels and poor prognosis in patients with acute cerebral infarction. OR denotes odds ratio; 95% CI, 95% confidence interval.

Full size image
Fig. 8
figure 8

Association of serum platelet-derived microparticles levels with modified Rankin score in patients with acute cerebral infarction. PDMPs means platelet-derived microparticles; mRS, modified Rankin score.

Full size image

Association of serum PDMPs levels with END after stroke

A total of 171 (27.5%) patients suffered from END after stroke. As shown in Table 4, compared with patients without END, those with END had substantially higher body mass index (P < 0.05), NIHSS score (P < 0.05), infarction volume (P < 0.05), proportion of diabetes mellitus (P < 0.05), serum glucose levels (P < 0.01), serum PDMPs levels (P < 0.01) levels and serum fibrinogen levels (P < 0.05). Multivariate analysis demonstrated that infarction volume, NIHSS scores and serum PDMPs levels were the independent predictors of END after ACI (All P < 0.05; Table 5). According to the ROC curve in Fig. 9, serum PDMPs levels significantly predicted the development of END after ACI (P < 0.001) and using the Youden method, serum PDMPs levels > 24.00 ng/L distinguished patients at risk of END with medium-high sensitivity and specificity values. Further restricted cubic spline shows the linear relationships between serum PDMPs levels and risk of END in patients with ACI (P > 0.05, Fig. 10). The subgroup analyses demonstrated that serum PDMPs levels were significantly associated with END in most categories, stratified by age, gender, hypertension, diabetes mellitus, coronary heart disease, dyslipidemia, smoker, alcohol, anti-platelet therapy and thrombolysis (all P < 0.05). Except coronary heart disease (P < 0.05), other variables had no significant interactions with serum PDMPs levels (all P > 0.05; Fig. 11).

Table 4 Baseline characteristics related to early neurological deterioration after acute cerebral infarction.
Full size table
Table 5 Multivariate regression analysis of related factors of early neurological deterioration after acute cerebral infarction.
Full size table
Fig. 9
figure 9

Receiver operating characteristic curve analysis using serum platelet-derived microparticles level as a predictor of early neurological deterioration after acute cerebral infarction. AUC denotes area under curve; 95% CI, 95% confidence interval.

Full size image
Fig. 10
figure 10

Linear relationship between serum platelet-derived microparticles levels and risk of early neurological deterioration in patients with acute cerebral infarction. PDMPs denotes platelet-derived microparticles; 95% CI, 95% confidence interval.

Full size image
Fig. 11
figure 11

Subgroup analyses of interactions between serum platelet-derived microparticles levels and early neurological deterioration in patients with acute cerebral infarction. OR denotes odds ratio; 95% CI, 95% confidence interval.

Full size image

Discussion

The current study investigated relationships between serum PDMPs concentrations and END in addition to clinical outcomes in ACI patients. Our study confirmed that (1) as compared with controls, PDMPs levels in serum of patients with ACI were increased significantly; (2) serum PDMPs levels of ACI patients were highly correlated with disease severity indicated by infarction volume and NIHSS score; and were also with 6-month mRS score; (3) serum PDMPs levels were independently associated with END and poor outcome at 6 months in patients with ACI and moreover, similar findings were observed in the subgroup analysis, further emphasizing the robustness of these associations. Hence, serum PDMPs may be a potential prognostic biomarker of ACI.

Previous studies showed that the incidence of END after ACI was 8.7–28.1%19,20. The incidence of END in our study was 27.5%. Currently, the underlying mechanisms of END are not fully understood. Several studies have revealed that high platelet reactivity is associated with increased risk of END in patients with ischemic stroke21 and anti-platelet therapy could prevent END and improves prognosis in these patients22, suggesting that platelet activation maybe an important factor in END in patients with ACI.

Microparticles may be released from cells upon cell activation or cell injury, or following cell activation-independent processes, including senescence and apoptosis23. They interact with neighboring cells through direct extracellular interactions and have also been shown to deliver their cytosolic contents to target cells as a means of altering cellular function24. In 1967, Wolf et al.25 showed that platelet shedding after activation results in “platelet dust”, which is now called PDMPs. PDMPs are generated during platelet activation or might result from platelet apoptosis. They are the common indicators reflecting platelet activation, which harbor approximately 100 times more prothrombotic properties than platelets. Thus, detection of PDMPs can indirectly reflect the functional status of platelets26.

Previous studies have indicated that PDMPs might play a role in the autoimmune diseases27, metabolic dysfunction and inflammation28, and contributes to tumor progression in several tumor cells29. Expression of PDMPs may be involved in various atherosclerosis and thrombotic diseases, including the development process in venous thromboembolism and myocardial infarction. Specifically, the levels of PDMPs showed significant elevation in both ST-elevation myocardial infarction and deep vein thrombosis30,31, and may reflect the severity of acute myocardial infarction as well as prothrombotic state32. Geng et al.33 reported that PDMPs levels were increased in myocardial infarction rats, and could be used as a tool to assess infarction area.

In cerebrovascular diseases, PDMPs have been found to induce cell proliferation, neurogenesis, and angiogenesis at the cerebral infarct boundary zone in a stroke model in vivo34. Additionally, PDMPs might exert a remote protective effect against cerebral ischemic–reperfusion injury9. These data indicate that PDMPs might exert a protective effect on ischemic injury. Another clinical study involving 57 ischemic stroke patients showed that circulating levels of platelet derived CD41 + microparticles were significantly related to reperfusion and recanalization, and high levels of peripheral CD41 + microparticles concentrations were detected in patients with partial or complete recanalization35. Our study showed that serum PDMPs levels were significantly increased in patients with END, and independently associated with END in patients with ACI. The subgroup analysis also confirmed that serum PDMPs levels was positively associated with END in patients with coronary heart disease, highlighting the importance of reducing serum PDMPs levels in patients with premorbid coronary heart disease. Kanazawa et al.36 found that baseline platelet count in venous blood at admission may affect the prognosis of cerebral infarction and END, which supported our results and suggested that platelet status and function could be closely related to END.

Recently, some studies has reported the role of new biomarkers in predicting the outcomes in ACI patients. Zhang et al.37 showed that serum Cav-1 combined with NGB may be helpful to predict the risk of END and 3-month poor prognosis after r intravenous thrombolysis in patients with acute ischemic stroke. Another meta-analysis found that the C-Reactive Protein/Albumin ratio was significantly associated with unfavorable outcomes38. These data indicated the intrinsic relationship of neuroinflammation with unfavorable outcomes in patients with ACI. As an inflammatory biomarker, PDMPs has been observed an significant increase in stroke patients28,39, suggesting the potential role of PDMPs as a prognostic biomarker for ACI. However, these studies were limited by relatively small sample sizes, the absence of detailed data on END, or the variability in follow-up durations (ranging from 1.5 to 90 days), which may lead to diverse impacts on outcomes. In addition, due to adoption of published literature, the meta-analysis was subject to a degree of heterogeneity and publication bias. The present study involved a relatively large sample of patients (621 patients with ACI), followed up for 6 months, and enriched the limited current evidence on the association between PDMPs and END and extends this relationship specifically to patients with ACI. The identification and quantification of PDMPs may thus be useful for the prediction of END after ACI.

A growing body of data have pointed to the prognostic role of circulating PDMPs concentrations in various kinds of cancers40,41. However, it is unclear whether there may be a correlation between serum PDMPs levels and severity plus clinical outcomes in patients with ACI. Our study not only demonstrated that serum PDMPs levels were highly associated with the severity of the neurological deficit estimated by infarction volume and NIHSS score at admission, but also showed that serum PDMPs levels were highly correlated with the functional outcome based on mRS score at 6 months after onset of stroke. Furthermore, we performed ROC analysis to reveal that serum PDMPs levels displayed a good prognostic predictive ability. Overall, it is plausible that increased serum PDMPs levels should be associated with 6-month poor prognosis of ACI, substantializing serum PDMPs as a potential prognostic biomarker of ACI.

Although a series of statistical methods were applied here to reveal some interesting results that serum PDMPs may be a potential prognostic biomarker of ACI, the current study has several limitations. First, this is a single-center study, which is characterized by a relatively small sample size, and therefore, a further cohort study with a larger sample size is required to prove our conclusions. Second, although there may be coexpression of some specific antigens between megakaryocyte-derived microparticles and PDMPs in healthy individuals, the PDMPs ELISA kit in the study used two monoclonal antibodies against glycoproteins CD42a and CD42b, which have not been detected in megakaryocyte-derived microparticles42. Thus, the kit had good specificity. Functional differences between the two microparticles would be the direction of future researches. Third, some relevant variables (such as age, diabetes and hypertension) were unbalanced between ACI patients and health controls. The further matching cohorts were needed to determine the difference in serum PDMPs levels. Finally, serum PDMPs levels were measured in a time point after ACI in this study and dynamic change of serum PDMPs levels is not detected. Investigating serial change of its level in future may be of clinical significance. Whereas, the current study was a pilot and exploratory study, which could provide some new information for the further studies. Thus, the independent and multicenter studies measuring serial PDMP levels were needed to validate the findings.

Conclusions

This study confirmed that increased serum PDMPs levels at admission are strongly relevant to the severity determined by infarction volume and admission NIHSS scores. Moreover, elevated serum PDMPs levels could assist in predicting the risk of END occurrence and poor prognosis at 6 months after ACI in patients, indicating that PDMPs in serum might have the potential to be a useful prognostic biomarker in patients with ACI.