Hematological Biomarkers Associated with Stroke Types: A Clinical Cross-Sectional Analysis

Abstract

Background: Stroke is a major cause of morbidity and mortality worldwide, with distinct pathophysiological mechanisms between ischemic stroke (IS) and hemorrhagic stroke (HS). Hematological parameters, such as lymphocyte and erythrocyte count, have been implicated in stroke prognosis, but their predictive value remains uncertain. Objective: To evaluate the association between hematological biomarkers and stroke subtypes (ischemic stroke and hemorrhagic stroke), and transient ischemic attack. Methods: This cross-sectional study analyzed clinical, metabolic, and hematological parameters in patients with stroke. Logistic regression models, adjusted for age, gender, and ethnicity, were applied to assess the association between lymphocyte and erythrocyte counts and stroke subtypes. Results: Lymphopenia was significantly associated with higher odds of hemorrhagic stroke (HS) in both the TIA–HS (OR 1.15, 95% CI: 1.05–1.26, p = 0.004) and the IS–HS models (OR 1.11, 95% CI: 1.03–1.20, p = 0.009). Additionally, erythrocyte count was significantly associated with increased odds of conversion from IS to HS (OR 3.97, 95% CI: 1.45–10.89, p = 0.007). The lymphocyte-to-monocyte ratio (LMR) was significantly different between IS and HS (OR = 1.38, 95% CI: 1.07–1.78, p = 0.014), while no significant association was found between TIA and HS (p = 0.399). Conclusions: Hematological parameters varied among stroke subtypes, with lymphopenia associated with hemorrhagic stroke and erythrocyte count differing between IS and HS. While these findings may aid in stroke characterization, further studies are needed to confirm their clinical relevance.

1. Introduction

Stroke events are classified as non-traumatic injuries and can be defined as brain damage resulting from interrupted blood flow or blood leakage due to vessel rupture [1]. Consequently, inappropriate hemostasis leads to cell death in nervous tissue.

Stroke is divided into two types: ischemic stroke (IS), the most common type, accounting for 87% of cases, according to the American Heart Association [2]. IS can be caused by thrombosis (originating in the brain) or embolism (originating in another organ and traveling to the brain) and is classified into five subtypes: large artery thrombosis, small penetrating artery thrombosis, cardioembolic stroke, cryptogenic stroke, and other causes [2,3].

Hemorrhagic stroke (HS), representing 13% of cases [2], is classified into two main subtypes: subarachnoid hemorrhage, which accounts for 3% of HS cases (occurring in the subarachnoid space), and intracranial hemorrhage, which constitutes 10% of cases (occurring in cerebral arteries) [2,4]. These subtypes differ in causes and mechanisms of action [5,6].

Stroke is a serious disorder associated with high rates of morbidity, mortality, and disability-adjusted life years, making it a major public health issue and economic burden. It is the leading neurological cause of acquired disability in adults and one of the primary causes of death, resulting in nearly six million deaths worldwide annually [2,6].

In the Brazilian population, the annual age-adjusted incidence rate of IS varies from 62 to 92 per 100,000 inhabitants, with a mixed pattern of conventional risk factors (RFs), such as dyslipidemia, hypertension, smoking, obesity, and diabetes [3]. According to the Ministry of Health of Brazil, in the Jaú/SP region, the proportional mortality rate by age group, according to cause group—ICD-10 (International Classification of Diseases, 10th Revision), is 30.8% for circulatory system diseases, with the highest incidence in individuals over 65 years old (37.7%), of which 55.7% are cerebrovascular diseases [7].

It is known that 10% of stroke events remain unclassified, reinforcing the hypothesis of a multifactorial disease model, including environmental and genetic risk factors of small effect [8]. Currently, stroke incidence is increasing among young adults worldwide [6] and in Brazil [2]. A study in the USA revealed that between 2000 and 2010, there was a 44% increase in stroke cases among individuals aged 25 to 44 years [2]. Similarly, in Brazil, there was a significant increase between 2008 and 2012 in individuals aged 15 to 49 years [9]. This increase was attributed to IS rates but not to HS [10].

Despite decades of epidemiological and experimental studies, knowledge of RFs can account for only one-third of atherosclerotic disease cases. Detecting early markers would enable the recognition of subclinical atherosclerotic disease and allow early intervention on modifiable RFs [11].

Some studies have highlighted potential early biomarkers for circulatory system diseases [12,13,14,15]. The role of immunology in stroke is not yet fully understood, but it is known to be involved not only in post-stroke recovery but also in the pathophysiology of the disease, especially IS [16].

Among the various hematological alterations observed in stroke, lymphopenia and erythrocyte count have gained attention due to their association with systemic inflammatory responses and hemodynamic alterations [17]. Lymphopenia has been linked to immune dysregulation in patients with stroke, contributing to increased susceptibility to infections and worse clinical outcomes [18,19,20]. Similarly, alterations in erythrocyte parameters, such as red blood cell distribution width (RDW), have been associated with stroke severity and prognosis, reflecting underlying vascular and metabolic imbalances [21,22]. However, their role in differentiating ischemic and hemorrhagic stroke remains unclear, highlighting the need for further investigation. Understanding these associations could contribute to improve diagnostic accuracy and patient management in cerebrovascular diseases.

The main mechanisms involved in the establishment and progression of the systemic inflammatory response include the complement system cascade, activation of immune cells, cytokine release, and endothelial dysfunction [23,24,25,26,27]. Both components of the immune system, cellular and humoral, have been implicated in atherogenesis [21]. The interaction between the arterial wall and circulating blood cells, such as monocytes and T lymphocytes, mediated by cellular adhesion molecules, is an important step in the inflammatory process. The expression of adhesion molecules and reactive oxygen species is stimulated via RFs and is regulated by inflammatory cytokines and hemodynamic factors, making their characterization extremely important [25].

In addition to traditional risk factors, measuring plasma markers of systemic inflammation and endothelial dysfunction contributes to predicting metabolic diseases [27,28]. Thus, biomarkers can help identify the risk of cardiovascular disease (CVD) and stroke [29].

Under normal conditions, the anti-inflammatory state of the nervous system is maintained through specific interactions between neurons and microglia, as well as through cell–cell junctions. These interactions help regulate microglial activation, reducing the production of pro-inflammatory molecules.

Therefore, this study aimed to determine the clinical, metabolic, and hematological profiles of patients diagnosed with ischemic and/or hemorrhagic stroke treated in a municipality in the interior of São Paulo state, Brazil.

2. Materials and Methods

2.1. Study Design and Sample Selection

This is a clinical, cross-sectional study conducted in the emergency department of Santa Casa de Jaú—SP. The study population consisted of patients who sought emergency care and received a diagnosis of stroke in the acute phase.

Using the G*Power 3.1.9.7 software, with a significance level of 5% and a type II error of 5% (test power of 95%), it was estimated that at least 130 patients needed to be evaluated. The study group included a total of 147 men and women, meeting the following inclusion criteria: (1) diagnosis of stroke (ischemic stroke, hemorrhagic stroke, or transient ischemic attack—TIA); (2) no restriction regarding personal history of cardiovascular disease (acute myocardial infarction or prior stroke); and (3) age ≥ 18 years. Patients with autoimmune diseases, acquired immunodeficiency syndrome (AIDS), infectious contagious diseases, or other neurological diseases were excluded. Data collection was conducted between May 2019 and December 2019. This project was approved by the Ethics and Research Committee of the University of Western São Paulo—UNOESTE (Protocol number: 4895917).

2.2. Diagnosis Criteria for Stroke and Transient Ischemic Attack (TIA)

Stroke was diagnosed based on the sudden onset of focal neurological deficits lasting more than 24 h, confirmed via computed tomography (CT) or magnetic resonance imaging (MRI), identifying either ischemic or hemorrhagic lesions. transient ischemic attack (TIA) was diagnosed in cases where patients presented with sudden focal neurological symptoms lasting less than 24 h, with complete resolution and no residual deficits. The absence of ischemic lesions on diffusion-weighted MRI (DWI) was used to differentiate TIA from ischemic stroke, following the American Heart Association/American Stroke Association (AHA/ASA) guidelines [30]. Additionally, the ABCD2 score was applied to assess the risk of subsequent stroke, considering factors such as age, blood pressure, clinical features, symptom duration, and the presence of diabetes.

2.3. Investigation of Clinical and Demographic Data

Data on age, gender, ethnicity, personal history of cardiovascular disease (CVD), systemic arterial hypertension (SAH), and diabetes were collected. All patients underwent routine basic laboratory tests for standard stroke diagnosis. From these routine laboratory tests, a complete blood count (CBC) was analyzed, including red blood cells, hemoglobin, hematocrit, mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), white blood cells, band cells, segmented cells, lymphocytes, monocytes, and platelets. Additionally, coagulation parameters such as prothrombin time (PT), international normalized ratio (INR), and activated partial thromboplastin time (aPTT) were assessed. To confirm the stroke diagnosis, biochemical analyses of urea and creatinine were performed. All hematological analyses were conducted in a certified clinical laboratory following standardized protocols. Quality control measures included periodic equipment calibration, internal control samples, and adherence to guidelines for hematological testing to ensure result accuracy and reproducibility [31,32].

2.4. Statistical Analysis

Continuous variables were presented as means and standard deviations and compared between stroke subtypes using t-tests or ANOVA. Categorical variables were expressed as absolute frequencies and percentages and analyzed using chi-square tests or Fisher’s exact tests. To assess the association between hematological parameters and stroke subtypes, logistic regression models were applied, with adjustments for age, gender, and ethnicity to control for potential confounders. Odds ratios (OR) and 95% confidence intervals (CI) were derived to quantify the strength of associations. Separate models were constructed for TIA vs. HS and IS vs. HS, allowing for subgroup-specific comparisons. A significant level of p < 0.05 was applied to all statistical tests. Data analysis was performed using JAMOVI (version 2.3.28).

3. Results

The baseline characteristics of the study cohort, including 147 patients categorized by gender and diagnostic classification, are summarized in

Table 1. Distribution of clinical and demographic data.

Variables	TIA	Ischemic Stroke (IS)	Hemorrhagic Stroke (HS)
Diagnosis	15	112	20
Years, y	69 (9.7)	68.5 (12.8)	64.5 (17.4)
Gender, (f/m)	5/10	51/61	10/10
Ethnicity, (b/w)	3/12	21/91	4/16
T2DM, (y/n)	4/11	64/48	5/15
HTN, (y/n)	13/2	101/11	15/5

(f/m): female/male; (b/w): black/white; (y/n): yes/no; T2DM, Type 2 Diabetes Mellitus; HTN, systemic arterial hypertension.

. The cohort included 66 women (44.89%) and 81 men (55.11%). Regarding ethnic distribution, 28 patients (19%) were Black, and 119 patients (81%) were White. The prevalence of systemic arterial hypertension was 88.7%, diabetes mellitus 46.2%, and cardiovascular comorbidities 24.5%. Patients were further classified into three diagnostic groups: transient ischemic attack (TIA; 15 cases: 5 women and 10 men, mean age 68.7 ± 9.7 years), ischemic stroke (IS; 112 cases: 51 women and 61 men, mean age 67.8 ± 12.8 years), and hemorrhagic stroke (HS; 12 cases: 5 women and 7 men, mean age 63.1 ± 17.4 years).

The hematological and metabolic variables evaluated in this study are summarized in

Table 2. Distribution of laboratory variables according to diagnosis.

Variables	TIA	Ischemic Stroke	Hemorrhagic Stroke
Erythrocytes, M/mm3	4.30 (0.99)	4.53 (0.75)	4.35 (1.09)
Hemoglobin, g/dL	13 (3.2)	13.6 (1.95)	12.9 (2.28)
Hematocrit, %	37.40 (5.6)	40.05 (5.7)	37.6 (6.32)
MCV, fL	85 (3.5)	88.5 (7.0)	87 (5.0)
MCH, pg	30 (1.5)	30 (2.25)	30 (2.0)
MCHC, g/dL	35 (2.0)	34 (2)	34 (1.0)
Leukocytes, mm3	7100 (4100)	9300 (3425)	10,400 (6650)
Band Neutrophils, %	2 (1.5)	2 (2)	3 (2.25)
Segmented Neutrophils, %	57 (19)	64 (14)	74 (8)
Eosinophils, %	2 (1)	2 (1)	2 (1)
Lymphocytes, %	31 (17)	24 (15)	15 (7.5)
Absolute Lymphocytes, mm3	1848 (791)	2194 (1366)	1476 (959)
Monocytes, %	10.4 (4.29)	8.18 (3.88)	6.27 (3.78)
Absolute Monocytes, mm3	769 (142)	770 (150)	720 (129)
Lymphocyte/Monocyte Ratio (LMR)	2.55 (1.49)	2.99 (1.83)	2.17 (1.14)
Platelets, K/uL	185,000 (63,500)	205,500 (69,500)	177,000 (81,750)
PT, seconds	13 (2.30)	13 (1.6)	13 (1.7)
INR,	1.02 (0.11)	1.02 (0.05)	1.02 (0.14)
aPTT, seconds	32 (4.5)	32 (4.5)	32 (4.5)
Urea, mg/dL	44 (47.5)	43 (24)	39.5 (20.5)
Creatinine, mg/dL	1.2 (0.5)	1.1 (0.30)	1 (0.43)

TIA, transient ischemic attack; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; PT, prothrombin time; INR, international normalized ratio; aPTT, activated partial thromboplastin time.

.

In specific models for TIA and HS, lymphocytes showed a significant association with stroke odds (OR 1.15, 95% CI: 1.05–1.26). Similarly, in models for IS and HS, lymphocytes were also significantly associated with stroke (OR 1.11, 95% CI: 1.03–1.20). Additionally, in the IS–HS model, erythrocytes demonstrated a significant association with increased odds of stroke (OR 3.97, 95% CI: 1.45–10.89, p = 0.007), as shown in

Table 3. Associations of clinical and hematological variables with stroke types (TIA–HS and IS–HS).

Strokes	OR	CI 95%	p
TIA–HS model
Years, y	1.03	0.97–1.10	0.295
Gender, (f/m)	1.78	0.37–8.59	0.473
T2DM, (y/n)	0.82	0.15–4.53	0.816
HTN, (y/n)	3.01	0.32–28.02	0.334
Ethnicity, (b/w)	0.85	0.11–6.27	0.870
Erythrocytes, 106/µL	1.91	0.49–7.42	0.349
Monocyte, %	1.10	0.88–1.38	0.415
Lymphocytes, %	1.15	1.05–1.26	0.004
Lymphocyte/Monocyte Ratio (LMR)	1.18	0.81–1.71	0.399
IS–HS model
Years, y	1.03	0.99–1.08	0.176
Gender, (f/m)	0.93	0.30–2.91	0.900
T2DM, (y/n)	2.79	0.77–10.11	0.117
HTN, (y/n)	3.19	0.65–15.58	0.152
Ethnicity, (b/w)	1.05	0.23–4.82	0.954
Erythrocytes, M/mm3	3.97	1.45–10.89	0.007
Monocyte, %	1.03	0.85–1.25	0.765
Lymphocytes, %	1.11	1.03–1.20	0.009
Lymphocyte/Monocyte Ratio (LMR)	1.38	1.07–1.78	0.014

y, years; f/m, female/male; b/w, black/white; y/n, yes/no; T2DM, Type 2 Diabetes Mellitus; HTN, systemic arterial hypertension; OR, odds ratio; CI 95%, confidence interval.

.

The LMR was analyzed for its association with stroke subtypes. A significant difference was found between IS and HS (OR = 1.38, 95% CI: 1.07–1.78, p = 0.014), suggesting a potential role in differentiating these conditions. However, no significant association was observed for TIA vs. HS (OR = 1.18, 95% CI: 0.81–1.71, p = 0.399).

Figure 1 and Figure 2 provide a detailed visualization of significant blood count components and their association with stroke subtypes within the study cohort. Figure 1 illustrates the model-estimated probability distribution of stroke types (HS, TIA, and IS) as a function of erythrocyte levels, adjusted for covariates such as age, gender, and ethnicity. The data indicate a progressive increase in the relative likelihood of IS as erythrocyte levels rise, while the probability of HS declines. Figure 2 shows the model-estimated probabilities of stroke subtypes based on lymphocyte percentages. It demonstrates that higher lymphocyte levels are associated with an increased probability of IS, whereas lower lymphocyte percentages correspond to a higher likelihood of HS and TIA.

4. Discussion

This study described the clinical, metabolic, and hematological profiles of patients diagnosed with stroke in a city in the midwestern region of São Paulo State, Brazil. While no significant associations were identified with clinical and demographic variables, the relationships between cellular components observed in hematological tests demonstrated statistically significant associations. These results underscore patterns observed within the study cohort and highlight associations rather than predictive or generalizable conclusions about stroke risk.

A cohort study evaluating the segmented-to-lymphocyte ratio in cardiovascular events (stroke and TIA) among 841 patients found no significant independent association between hematological ratios in the multivariate analysis [33]. However, the segmented-to-lymphocyte ratio, calculated as the ratio between the absolute count of segmented cells and lymphocytes, has been proposed as a potential prognostic marker and predictor of cardiovascular events [33]. Similarly, lymphocyte count, an easily accessible parameter from routine blood tests, has been associated in previous studies with stroke prognosis and management. Our findings showed a trend of reduced lymphocyte levels in patients with stroke, which aligns with evidence suggesting a relationship between lymphopenia, systemic inflammation, and functional impairment. However, as our study is cross-sectional, we cannot infer direct prognostic implications. Additionally, an imbalance in inflammatory markers, such as a high neutrophil-to-lymphocyte ratio (NLR), has been identified as a significant predictor of stroke recurrence and adverse clinical outcomes.

These findings underscore the importance of monitoring lymphocyte counts and related inflammatory markers to guide therapeutic strategies and improve patient prognosis in the early post-stroke period [33,34].

According to Silina et al. (2021) [35], under hemorrhagic stroke conditions, segmented leukocytosis develops. The action of endotoxins and neurotoxins promotes the release of neutrophils from the bone marrow reservoir. This differentiation is proportional to the speed and severity of neural tissue deterioration [27].

Our data revealed neutrophilic leukocytosis and lymphopenia, which have been previously associated with inflammatory responses in patients with stroke. Although the segmented/lymphocyte ratio have been proposed as a potential marker in long-term cardiovascular risk assessment, our study does not allow for conclusions regarding its independent prognostic value. Further longitudinal studies are needed to explore this relationship.

Hematological analysis showed significant differences between the IS and HS groups. Patients with HS had higher percentages of segmented neutrophils (74%) compared to those with IS (64%) and TIA (57%). This finding reinforces the hypothesis that an exacerbated inflammatory response, characterized by increased neutrophil infiltration, may be associated with a worse prognosis in HS cases [36,37].

Patients with HS had a lower absolute lymphocyte count (1476/mm³) compared to those with IS (2194/mm³) and TIA (1848/mm³), suggesting that lymphopenia may be a marker of greater severity in hemorrhagic stroke [38]. Additionally, logistic regression analysis showed that lymphopenia was significantly associated with increased odds of progression to HS in the TIA–HS model (OR 1.15, 95% CI: 1.05–1.26, p = 0.004) and the IS–HS model (OR 1.11, 95% CI: 1.03–1.20, p = 0.009), indicating that reduced lymphocyte levels may represent a relevant factor in stratifying patients for worse stroke prognosis.

The lymphocyte-to-monocyte ratio (LMR) showed mean values of 2.17 in HS, 2.99 in IS, and 2.55 in TIA. A statistical comparison revealed that LMR was significantly associated with IS compared to HS (OR = 1.38, 95% CI: 1.07–1.78, p = 0.014), while no significant association was observed between TIA and HS (OR = 1.18, 95% CI: 0.81–1.71, p = 0.399). These variations in LMR between stroke types may reflect differences in systemic inflammation and immune response, with lower LMR in HS possibly indicating a more pronounced immunosuppressive state, which has been linked to poorer prognosis and increased risk of secondary complications [39,40].

The LMR measured at hospital admission can serve as a predictor of unfavorable clinical outcomes three months later in patients with ischemic stroke undergoing mechanical thrombectomy. Additionally, in patients with spontaneous intracerebral hemorrhage, LMR independently predicted the three-month prognosis [41,42]. Given the crucial role of the inflammatory response in stroke recovery, closely monitoring patients with elevated LMR levels in spontaneous intracerebral hemorrhage could be a promising approach to reducing brain damage after a stroke. Furthermore, further exploration of the Systemic Inflammatory Response Index and its influence on clinical outcomes is essential [43].

The clinical blood results indicate a decrease in protective mechanisms, suppression of immune responses, and regression in antibody formation and/or cytokine levels [44,45]. According to a previous study, the assessment of cytokine levels was limited by the reorganization of the hospital sector responsible for stroke care to accommodate patients with COVID-19 [46].

Leukocyte transformations, observed in most patients with stroke as early as the first day—particularly in hospitalized cases—reflect the systemic inflammatory response, confirmed by an increase in circulating cell counts [47]. Studies have concluded that elevated hematological cell counts on the first day of hospitalization predict a high risk of mortality in patients with stroke [48].

The statistical analysis of logistic regression models revealed a significant association between lymphocyte count and stroke outcomes. In the TIA–HS and IS–HS models, lymphopenia was a significant factor for progression to HS (OR 1.15, 95% CI: 1.05–1.26, p = 0.004 in the TIA–HS model; OR 1.11, 95% CI: 1.03–1.20, p = 0.009 in the IS–HS model). These findings reinforce the association between lymphopenia, a chronic inflammatory state, and a higher likelihood of adverse clinical outcomes [38,49].

Some studies have suggested that elevated white blood cell counts are associated with stroke severity, depending on hematoma volume and intraventricular extension [47]. These fluctuations in leukocyte dynamics may result from the inflammatory response, further confirming the role of inflammation in stroke pathophysiology [47,48]. Additionally, some studies have found that oxidative stress, mean platelet volume, erythrocyte sedimentation rate, and inflammatory profiles can act as independent risk factors for ischemic stroke [47,50,51].

Another relevant finding was the relationship between erythrocyte count and the odds of IS–HS (OR 3.97, 95% CI: 1.45–10.89, p = 0.007), suggesting that a higher concentration of erythrocytes may be associated with increased blood viscosity, consequently, a greater likelihood of hemorrhagic conversion in IS [52,53]. These findings highlight the importance of incorporating hematological monitoring into routine clinical practice to prevent potential complications in patients with stroke.

In this context, the inclusion of lymphopenia and erythrocyte count as predictive factors may aid in clinical decision-making, especially in the follow-up of patients with ischemic stroke. Tracking these biomarkers could enable early therapeutic adjustments to minimize the risk of hemorrhagic transformation, ultimately improving prognostic stratification and patient outcomes.

Due to structural and organizational limitations faced during the analysis of the patient sample studied, highly complex data were underexplored due to the lack of collection or proper recording of information by the public hospital system. A similar issue was observed regarding the absence of standardized methodologies in test requests by healthcare professionals. Thus, implementing a structured framework for the evaluation and investigation of patients with stroke is essential.

5. Conclusions

Our cross-sectional analysis identified significant associations between specific hematological parameters—lymphopenia, erythrocyte count, and LMR—and stroke subtypes. However, further longitudinal studies are required to establish causal relationships and to explore the potential clinical utility of these hematological markers in stroke management.