Stroke is one of the most common cerebrovascular diseases. It is a sudden onset of focal or generalized brain dysfunction that lasts more than 24 hours and is caused by vascular disorders related to cerebral blood flow. Oxygen and glucose deprivation in the brain leads to decreased ATP synthesis and impaired synaptic conduction, causing neuronal malfunction and subsequent apoptosis or necrosis (1). The most common stroke causes include atherosclerotic lesions in extracranial and intracranial vessels, cardioembolism, carotid and vertebral artery dissections, hypercoagulable syndromes, and multiple systemic diseases. There are two types of strokes: ischemic (80%) and hemorrhagic (20%). Ischemic stroke is caused by closure/constriction of intracerebral vessels and hemodynamic abnormalities resulting in slowed cerebral flow, whereas hemorrhagic stroke occurs due to blood extravasation within the brain tissue (2). It is estimated that one person dies every 6 seconds due to stroke, which annually accounts for more than 5-6 million people worldwide. Thus, stroke is the third cause of death after heart disease and cancer (3). It is also the most common reason for permanent disability in people over 40 years old, with severe clinical, social, and economic consequences (4).

Many studies have shown that early revascularization treatment significantly improves patient prognosis (5–10). Although imaging studies such as CT, MRI, ultrasonography, arteriography, and echocardiography are the mainstay of the diagnosis, they may not be sufficient in some groups of stroke patients. Indeed, many other diseases have similar symptoms, making the differential diagnosis of stroke include brain tumors, migraine, epileptic seizure, hypoglycemia, hyperglycemia, hyponatremia, hypertensive encephalopathy, and hepatic encephalopathy (11–14). Not all patients are also treated at specialized stroke/neurology centers. Therefore, it is not surprising that rapid and non-invasive biomarkers of the disease are still being sought (15). Their source may be saliva containing numerous substances that pass from the brain into the blood (16–20). It is well known that the blood-brain barrier (BBB) is disrupted in stroke pathophysiology. The main factors damaging the BBB are mechanical failures or hypoxia damaging the cerebrovascular endothelium. These also include increased activity of matrix metalloproteinases (MMPs), enhanced secretion of cytokines, chemokines, and growth factors, and overproduction of reactive oxygen (ROS)/nitrogen (RNS) species by neuronal, glial, and immune cells. The damaged BBB becomes permeable to leukocytes inducing inflammatory processes and promoting the release of brain biomolecules into the blood (2, 21–29). Subsequently, circulating biomarkers can pass into saliva by passive, facilitated, or active diffusion, making saliva highly attractive in laboratory diagnostics. Saliva’s advantages are also evidenced by its easy availability, non-invasive and painless collection, and relatively high durability compared to blood and cerebrospinal fluid (16–18, 20, 30–35).

Recent studies suggest the potential use of saliva in ischemic stroke diagnosing or assessing the stress severity in stroke patients (36–38). However, there is a lack of non-invasive biomarkers to differentiate between different types of strokes. Given the critical role of xanthine oxidase (XO) in ischemic stroke pathomechanism (22, 24, 29, 39–41), we decided to compare the XO activity in saliva of patients with ischemic and hemorrhagic stroke. Ischemia-reperfusion injury is an inflammatory process with characteristic cellular changes leading to microvascular disruption. In ischemic stroke, ROS overproduction by XO hyperactivation is a major pathogenic factor contributing to brain dysfunction. XO catalyzes the conversion of hypoxanthine to xanthine and xanthine to uric acid (UA), generating high amounts of superoxide radicals, hydrogen peroxide (H₂O₂), and peroxynitrite (Figure 1). Higher formation of ROS/RNS in stroke brain results in neuronal oxidative stress responsible for altering the fluidity of biological membranes, modifying enzyme activity, uncoupling membrane transport, or deregulating membrane potential (24, 26, 29, 42–44). Although previous studies have not confirmed the clinical usefulness of serum/plasma XO in stroke diagnostics (45), evaluation of enzyme activity in saliva may bring new light to this issue. The results of many studies showed that saliva has the highest correlation with the brain, not the blood (46–48). When the BBB is breached, several glial, neuronal, and pericyte biomarkers appear in the blood, which then passes into saliva (22, 49–51). This is caused by the robust vascularization of the salivary glands allowing the efficient exchange of blood-derived components (52, 53). Since saliva can be collected non-invasively multiple times a day, it is a particularly attractive material for diagnosing cerebrovascular diseases (18, 32). Therefore, the present study aimed to compare XO activity and the enzyme products (e.g., hydrogen peroxide and UA) in non-stimulated and stimulated saliva of patients with hemorrhagic and ischemic stroke.

Vascular brain diseases are one of the most common causes of death and disability worldwide. Despite significant progress in understanding stroke pathogenesis, the number of cases is still increasing (1, 66, 67). Therefore, laboratory biomarkers of stroke are sought to allow rapid and non-invasive diagnostics (15). Although diagnosis is based primarily on clinical examination and CT scans, it is not always possible to rule out conditions mimicking strokes, such as subdural or epidural hematoma, brain tumors, craniocerebral or cervical spine injury, infections (meningitis, encephalitis, brain abscess), seizures, migraine complications, and metabolic disturbances. Moreover, the sensitivity of CT in newly diagnosed ischemic stroke is less than 30-35%, which, in the absence of widespread availability of MRI and CT perfusion, indicates the need to search for new diagnostic strategies (12, 14). The ideal stroke biomarker should differentiate between ischemic and hemorrhagic stroke, have high sensitivity and specificity (at least 75-85%), capture disease dynamics/treatment effectiveness, and be easily quantifiable (68). In addition, the collection of biological material should be non-invasive, uncomplicated, and inexpensive. Therefore, saliva has become of increasing interest in the diagnosis of neurovascular diseases (69). Saliva collection is stress-free, requires no specialized medical personnel, and allows continuous monitoring of the patient’s condition through non-invasive sampling multiple times per day (18, 32).

The main factors leading to ischemic brain injury are inhibition of ATP production, excitotoxicity, inflammation, and cerebral oxidative/nitrosative stress. Indeed, lack of energy substrates disrupts depolarization of the neuronal membrane and increases intracellular levels of Na⁺, Cl^- and Ca²⁺ ions. This leads to activation of Ca²⁺-dependent enzymes such as protein kinase C (PKC), phospholipase A2, and other cellular proteases initiating neuronal apoptosis and necrosis. Simultaneously, there is a conversion of xanthine dehydrogenase to XO, which is crucial in post-stroke complications. The substrate for the enzyme is hypoxanthine (a breakdown product of ATP), which accumulates in the ischemic brain. When O₂ is delivered under reperfusion, XO causes the conversion of hypoxanthine to xanthine accompanied by the release of superoxide radicals(O₂-• ) via reduction of molecular oxygen. Subsequently O₂-• induces the formation of more toxic ROS (e.g., H₂O₂) and stimulates the production of inflammatory mediators (24, 70, 71). However, XO also catalyzes the conversion of xanthine to UA (Figure 1). In ischemic stroke animal models, cerebral XO activity correlates with infarct volume and severity of neurological complications, thus postulating the use of XO in the identification/differentiation of the disease. Because assessment of XO activity in the blood is not diagnostically relevant (72), we were the first to investigate the usefulness of salivary XO in stroke patients.

We demonstrated that XO activity is significantly higher in NWS of ischemic stroke patients compared to hemorrhagic stroke and healthy subjects. However, statistically important differences were not observed in SWS. Indeed, the composition of stimulated saliva depends not only on the salivary gland but also on the environmental stimuli. Under resting conditions, 2/3 of the saliva is produced by the submandibular glands, whose filtrate generally reflects the composition of blood plasma. However, this ratio shifts during food/smell stimulation in favor of the parotid glands (52, 53, 73). These glands are also the primary source of antioxidants in the oral cavity, making SWS more protective against salivary oxidative stress (17, 74). The results of our previous studies support this. In SWS of stroke patients, we have shown higher activity/concentration of enzymatic and non-enzymatic antioxidants, which is an adaptive response to ROS overproduction in the parotid glands. These facts may explain the lack of differences in XO activity in SWS, especially if the concentration of enzyme substrates (xanthine and hypoxanthine) is very low in saliva samples. SWS is also much more dilute than NWS, making it less useful in laboratory medicine (33). Indeed, the main limitation of diagnostic applicability of saliva is the low biomarker concentration/activity compared to other bioliquids and tissues (75). Also in our study, salivary XO activity assessed by a standard enzyme assay was below the detection level in both NWS and SWS. However, using the Amplex Red xanthine/XO fluorometric method, XO can be detected at levels as low as 0.1 mU/mL. The use of a commercial kit has particular advantages in diagnostics, as it allows the comparison of results obtained in different laboratories.

The biomarkers found in saliva can be divided into compounds produced in the salivary glands and those outside of them (33, 74, 76). The ability to pass into saliva depends on the mechanism of transport and occurs via intracellular (diffusion, filtration, facilitated transport, active transport) or extracellular routes. Since most salivary proteins have a molecular weight between 20 and 60 kDa (low molecular weight proteins) (77), XO can migrate into saliva through ultrafiltration or damaged cellular membranes (78). XO is a homodimer composed of two subunits with an approximate molecular weight of 150-155 kDa (79). During BBB injury, XO can penetrate the brain into the circulation and then into the oral cavity. The salivary glands are very well vascularized resulted in a very efficient secretion of many substances into saliva (52). This may explain the higher correlation of many biomarkers between brain and NWS compared to blood (33). It may also result from a common developmental origin, or it may simply reflect the more significant variability of XO in the saliva (46). It should not be forgotten that XO can also be produced in the salivary glands (79). However, in some stroke patients, salivary gland dysfunction manifests as decreased salivary secretion (hyposalivation) or subjective dryness of the oral mucosa (xerostomia) (62, 80). Since the activity/concentration of salivary biomarkers is highly dependent on salivary gland function, we standardized XO measurement on total protein content and salivary flow rate. Our study indicates that XO activity should be standardized to total protein concentration, as with other salivary enzymes. Specific XO activity (nU/mg protein) better differentiates ischemic from hemorrhagic stroke and from healthy subjects.

The brain is particularly vulnerable to oxidation because it uses more than 20% of O₂ supplied to the body (81). Neurons also show a very unfavorable surface-to-volume ratio, a high content of unsaturated fatty acids, and low efficiency of enzymatic antioxidant systems (82). Therefore, the main cause of brain damage during ischemia and reperfusion is the increased formation of ROS by XO (42). The superoxide radical generated by XO is enzymatically reduced to H₂O₂, with the simultaneous conversion of xanthine to uric acid. Although UA has strong antioxidant properties, in high concentrations, it also has a robust pro-oxidant effect. UA can generate free radicals by reaction with peroxynitrite or alkylate cellular biomolecules disrupting their structure and function (40). Thus, increased synthesis/release of proinflammatory cytokines, chemokines, and growth factors promotes neuronal apoptosis and necrosis under XO overexpression (83). In the acute phase of stroke, the BBB is disrupted, causing many biomarkers to infiltrate the blood and saliva (50). In our study, UA and H₂O₂ levels were significantly higher in stroke patients’ saliva than controls, whereas they did not differentiate between stroke types. Although XO activity correlated strongly with uric acid and hydrogen peroxide levels in stroke cases, their source in saliva may also be diffusion from blood plasma (UA) or exposure to environmental factors (84). Indeed, the oral cavity is the only place in the body exposed to many pro-oxidant agents such as diet, xenobiotics, oral microbiota, dental procedures, and materials (85–87).

An essential part of our study was also to evaluate the saliva’s usefulness for assessing XO activity in stroke patients. For this purpose, we used ROC curves, which are a graphical relationship between a test’s sensitivity and specificity. Of all the parameters evaluated, the specific activity of XO has the best clinical utility. We have shown that this biomarker differentiates ischemic stroke from hemorrhagic stroke (AUC: 0.764) and controls (AUC: 0.973) with very high sensitivity (IS vs. HS: 66.67%; IS vs. C: 93.33%) and specificity (IS vs. HS: 66.67%; IS vs. C: 90%). It is also noteworthy that XO specific activity correlates negatively with cognitive impairment according to ACE III scale, postural balance in BBS scale and performance in living activities using BI scale. Importantly, these relationships were observed only in the NWS of stroke patients. Unfortunately, we do not have clinical data on stroke location, brain infarct volume, and severity of neurological symptoms, making further clinical studies necessary. Although this undoubtedly represents limitations of the study, we are the first to demonstrate the utility of salivary XO in the differential diagnosis of stroke and for assessing patients’ functional status.

Stroke is the most common cause of cognitive impairment in people over 65 years old. Cognitive deficits include all areas of daily functioning affecting treatment and rehabilitation outcomes and quality of patient’s life (88). In our study, XO specific activity correlated highly negatively with cognitive abilities in the ACE III score. In detail, this biomarker was adversely associated with attention and orientation (r = -0.57, p = 0.001), memory (r = -0.63, p = 0.0002), visual perception (r = -0.54, p = 0.002), language (r = -0.56, p = 0.001) and visuospatial functions (r = -0.53, p = 0.003). ACE III is a comprehensive screening tool for cognitive function assessment, useful for early detection of cognitive impairment, differential diagnosis of dementia, and monitoring of the disease progression (54). Therefore, in the next step, we divided stroke patients into three subgroups based on the severity of cognitive decline: normal cognition (100-89 in the ACE III), mild cognitive impairment (88-61 in the ACE III), and moderate cognitive dysfunction (< 61 in the ACE III) (89, 90). We have demonstrated that XO specific activity differentiates with high accuracy (100%) and specificity (93.75%) between stroke patients with mild to moderate cognitive decline (AUC = 0.988). Thus, salivary XO assessment may be a potential screening tool for a comprehensive neuropsychological evaluation. Unfortunately, this parameter did not distinguish cognitively mild patients from those without cognitive impairment, which may be due to the small number of patients.

The ideal diagnostic biomarker should be detectable at an early stage of the disease. Our study involves patients in the acute phase of stroke (30-35 days after the incident), and therefore, it is essential to evaluate salivary XO activity in newly diagnosed cases. Since ischemia-reperfusion is most severe at this time, assessment of XO activity may have even greater diagnostic value (91).

In conclusion, our study demonstrates the utility of salivary xanthine oxidase in the differential diagnosis of stroke. The biological material for assessing XO activity should be non-stimulated saliva, and the results must be standardized to total protein content. We have shown that XO specific activity distinguishes, with very high sensitivity and specificity, between ischemic and hemorrhagic strokes and controls, as well as patients with mild and moderate cognitive impairment. Currently, there is no generally available and sensitive diagnostic test for stroke, which is a major limitation in its early diagnosis as well as its treatment. Salivary XO may be the first potential biomarker used in the differential diagnosis of stroke and to assess the functional status of patients.

Our study also has some limitations. Although appropriate statistical tests determined the sample size calculation, further studies on a larger patient population are needed. It is also necessary to evaluate XO activity in cases with periodontal and other oral and systemic diseases, which may affect the parameters assessed in saliva. It would also be interesting to perform a more extended follow-up of the patients, to investigate the possible correlation between evolution of XO salivary levels and the improvement of cognitive and physical functions, during the recovery after the acute phase of the stroke. To assess biomarker specificity, evaluation of salivary XO activity in patients with other neurological diseases [e.g., small vessel disease and transient ischemic attack (TIA)] is also essential.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by Local Ethics Committee at the Poznan University of Medical Sciences. The patients/participants provided their written informed consent to participate in this study.

MM: conceptualized, did laboratory determinations, performed statistical analysis, interpreted data, did performance of the graphic part of the manuscript, wrote the manuscript, reviewed the article, final approval of the version to be published, coordination of the project. MN: did laboratory determinations, interpreted data. AZ: conceptualized, did laboratory determinations, reviewed the article, final approval of the version to be published. GB: did laboratory determinations. PG: performed clinical examination, qualified patients to the study, collected material. KH: performed clinical examination, qualified patients to the study, collected material. KG: conceptualized, did laboratory determinations, reviewed the article, final approval of the version to be published. All authors contributed to the article and approved the submitted version.

This work was supported by the Medical University of Bialystok, Poland (Grant No. SUB/1/DN/22/002/3330).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.