Stroke is a major cause of morbidity, disability and mortality in the populations of many countries. Type 2 diabetes mellitus (T2D) increases the risk of ischemic stroke (IS) by 2–2.5 times, and the risk of death from stroke by 3 times [1, 2]. Hyperglycemia during the first hours of stroke could be the body's stress response to brain ischemia [3]. Severity of neurohormonal and metabolic disorder reflects the IS severity and contributes to the outcome of the disease [4].

High prevalence of the internal carotid artery (ICA) atherosclerosis in patients with T2D compared to non-diabetic patients has been reported. That is why atherothrombotic stroke is the most common IS pathogenetic subtype in individuals with T2D [5].

Oxidative stress is one of the key factors involved in the pathogenesis of IS. Accumulation of free radical oxidation products results in the enzyme SH group blockage and inactivation, DNA hydroxylation, DNA fragmentation, and the resulting destabilization of cell membranes [6, 7].

Low molecular weight aminothiols (LMWTs) are highly sensitive to oxidative stress. Oxidized forms of LMWTs predominate in blood plasma; redox status (RS) is the ratio between the reduced forms and the total content of each thiol. Experimental studies and clinical trials report the decrease in the levels of reduced forms during the acute period of IS, which could indicate activation of oxidative processes in case of insufficient antioxidant defence [8, 9].

Both the increase in total homocysteine content (tHcy) and the decrease in the total glutathione content (tGSH) in females were revealed in patients with T2D [10]. Other papers report the decreased glutathione redox status (GSH RS) in blood plasma [11] and whole blood [12] of patients with T2D. Negative correlation between hyperglycemia and plasma tGSH levels was observed in individuals with coronary artery disease (CAD) (r = –0.328; р = 0.011), furthermore, it was shown that T2D was a factor, having an additional negative effect on the tGSH levels [13]. Perhaps low plasma tGSH levels are associated with the increased activity of  γ-glutamylcysteine transferase in hyperglycemia [10]. The decreased GSH synthesis along with the possible increase in GSH consumption in red blood cells of patients with T2D were also reported [14].

The study was aimed to assess the impact of T2D on the total content (t),  reduced forms (r) and RS of plasma LMWTs in patients with IS.

METHODS

The study included 175 patients with IS in the ICA basin (89 males and 86 females aged 46–84 (the average age was 62 (55–69) years)), admitted to the Research Center of Neurology (Moscow; Russia) within the first 10–24 h since the onset of the neurological disorder. The index group consisted of 68 patients with IS and T2D (males made up 41.2%). The comparison group included 107 patients with IS and stress hyperglycemia (males made up 57%), and the control group consisted of 31 non-diabetic patients with chronic cerebrovascular disease (CCVD) (males made up 54.8%, the average age was 69 (60– 75) years).

Inclusion criteria: age 45–85 years; the first IS; admission within the first 6–24 h since the onset of neurological symptoms; cerebral infarction in the ICA basin confirmed by neuroimaging data; T2D or stress hyperglycemia on admission; submitted informed consent to participation in the research study.

Exclusion criteria: T1D; acute myocardial infarction; decompensated kidney disease, liver failure or respiratory failure; class III–IV heart failure.

General medical examination and neurological examination were performed in all patients.

The neurological impairment severity was assessed on admission and in the end of the IS acute period using the National Institutes of Health Stroke Scale (NIHSS) [15]. Neurological impairment was considered mild when NIHSS < 7, moderate when 7 ≤  NIHSS < 14, and severe stroke when 14 ≤ NIHSS. The activities in daily living and the ability to function independently were evaluated on admission and on day 21 of IS using the Bartel Index (BI) [16], and the functional status was assessed with the use of the modified Rankin scale (mRs). The ability to perform all the routine tasks and to be engaged in daily activities corresponded to the mRs score of 0–2 by day 21, and poor functional outcome corresponded to the score of  ≥ 3 [17].

IS was diagnosed based on the clinical features and magnetic resonance imaging data (Magneton Symphony and Magneton Avanto, 1.5 T) acquired using standard sequences (Т2, Т1, Т2-FLAIR, T2*). Cerebral arteries were examined by 3D TOF MR angiography. Duplex ultasonography of the cerebral arteries was performed using the Philips iU22 ultrasound system (Philips; Netherlands).

Pathogenetic subtype of IS was defined in accordance with the internationally accepted TOAST (Trial of ORG 10172 in Acute Stroke Treatment) criteria [18].

Fibrinogen assay was carried out with the ACL 9000 coagulation analyzer (Instrumentation Laboratory; USA).

Glucose (hexokinase method), glycated hemoglobin (НbA1c) (immunoturbidimetric assay), total cholesterol, lowdensity lipoprotein, urea, and creatinine levels were assessed with the Konelab 30i clinical chemistry analyzer (Thermo Fisher Scientific; Finland) using the reagent kits (Randox; UK).

All patients with IS underwent blood glucose monitoring on admission. The HbA1c test was performed in patients with admission blood glucose levels of 6.1 mmol/L or more. The HbA1c levels reflect the blood glucose levels over the last 3 months. The diagnosis of T2D was established based on the American Diabetes Association criteria [19].

All patients with IS received backbone therapy, which included hypotensive, antithrombotic and lipid-lowering agents. Glucose-lowering therapy was used along with the blood glucose monitoring.

The content of LMWTs was defined as described previously [8]. Venous blood was collected into tubes containing sodium citrate (0.38%) and centrifuged at 3,000 g for 3 min. Blood plasma for assessment of total LMWTs was collected and stored at –20 °C. For derivatization, 100 μL of blood plasma were added to 10 μL of 50 mМ dithioteritol and 10 μL of the internal standard (0.45 mМ penicillamine). Dithiothreitol and penicillamine solutions contained 10 мМ Na-EDTA. The mixture was incubated at 37 °C for 15 min. Subsequently, the 5,5'-dithiobisnitrobenzoic acid (600 μL, 20 mМ) dissolved in ethanol was added, and the mixture was incubated at 4 °C for 30 min. After centrifugation at 15 000 g for 5 min, supernatant was dried under a vacuum at 60 °C for 2 h. The pellet was resuspended in 30 mМ NaOH prior to analysis.

To define the levels of the LMWTs reduced forms, plasma (100 μL) was added to 25 μL of the 5-sulfosalicylic acid solution (230 g/L) immediately after plasma collection, frozen and and stored at –80 °C. The samples were centrifuged at 15,000 g for 5 min prior to derivatization. Subsequently, 40 μL of supernatant were mixed with 40 μL of 20 mМ 5,5'-dithiobisnitrobenzoic acid and 2.5 μМ penicillamine in 0.4 М sodium phosphate buffer (pH 8.0). Then 10 μL of 1 М NaOH were added, the solution was mixed for 5 s, and subsequently 12.5 μL of 1 М HCl with 20 mМ N-ethylmaleimide were added to quench the reaction.

Analysis was performed using the Waters ACQUITY UPLC system (Waters, Milford; USA), equipped with the photodiode array (λPDA) UV detector (absorbance 330 nm; resolution 10.8 nm; frequency 5 s-1) and the Poroshell 120 SBC18 column (2.8 μm, 150 μm × 2 mm) (Agilent; USA). The temperature of the column and samples of 50 and 10 °C, respectively, was maintained. The volume of the sample injection was 10 μL, and the flow rate was 0.2 mL/min. Eluent A: 0.1 М ammonium acetate with 0.12% (v/v) HCOOH; eluent B: acetonitrile. Chromatography involved gradient elution using a linear gradient of solvent B of  2.5–10% for 5 min. Regeneration was performed with the use of 70% B for 1.5 min, and equilibration was carried out using 2.5% B for 4 min.

Statistical data processing was performed with the IBM SPSS Statistics  Version 20.0 software package (IBM Corp.; USA). The Descriptive Statistics module was used to obtain the discriminative model. The quantitative characteristics were presented as median, 25th and 75th percentiles (Ме (Q₁–Q₃%)), qualitative data were presented as absolute frequencies and percentage. The Kruskal–Wallis test and Mann–Whitney U test were used to compare the groups based on their quantitative characteristics. The logistic regression procedures were used to identify the prognostic factors. The variables were selected by conditional inclusion. The quality of the logistic regression model was assessed by ROC analysis and calculation of statistical characteristics of the tests (sensitivity, specificity). To evaluate the predictive power of the model, the Area Under Curve (AUC) was assessed. To define the optimal cutoff value, the importance of maintaining the sensitivity-specificity balance was taken into account. The two-tailed critical p-value, used in all comparisons and tests, was set at 0.05.

RESULTS

The main characteristics of patients with IS and CCVD are presented in tab. 1. The groups of patients showed no differences in age, gender, and indicators of lipid and protein metabolism. The patients with IS had been constantly taking antihypertensive drugs in 46 cases (26.3%), antiplatelet drugs in 22 cases (12.6%), anticoagulants in 11 cases (6.3%), and statins in 8 cases (4.6%) before being included in the study. The group with CCVD received no preventive therapy  before inclusion in the study.

Stroke, caused by the ICA atherothrombosis, was diagnosed in 35 cases (20%), cardioembolic stroke was diagnosed in 50 cases (28.6%), and lacunar stroke due to small-artery disease was diagnosed in 90 cases (51.4%).

In patients with atherothrombotic stroke, the developing step-by-step neurological deficits and the occurrence of single large infarctions affecting the cortical and subcortical regions outside the neighboring vascular territories were observed.

The heart-related causes of the cerebral artery thrombosis were as follows: embologenic forms of CAD (paroxysmal atrial fibrillation in 29 cases (58%), permanent atrial fibrillation in 17 cases (34%), postinfarction cardiosclerosis in 4 cases (8%)). In 34 patients, atrial fibrillation was diagnosed for the first time during the acute period of IS. Sudden emergence of persistent neurological symptoms was a clinical feature typical for cardioembolic stroke. Based on the MRI data, the cortical and subcortical infarctions were localized mostly in the middle cerebral artery basin. 

IS, resulting from the hypertensive, small, deep cerebral infarctions, was characterised by the gradually increasing intensity of neurological deficits in the form of lacunar syndromes. Small, deep cerebral infarctions were localized in the basal ganglia, white matter of cerebral hemispheres, and internal capsule. Stage III hypertension was found in 71 patients (78.9%) with lacunar stroke, and stage II hypertension was observed in 19 patients (21.1%).

The baseline IS severity assessment made it possible to reveal mild neurological impairment (NIHSS < 7) in 92 cases (52.6%), moderate neurological impairment (7 ≤ NIHSS < 14) in 67 cases (38.3%), and severe neurological impairment (14 ≤ NIHSS) in 16 cases (9.1%) (see tab. 1).

The admission blood glucose level in patients with IS was 6.50 (6.22–7.72) mmol/L.

Analysis of LMWTs showed that the tGSH, rGSH, and GSH RS levels in the group with IS were significantly decreased (р < 0.0001), and the rHcy and Hcy RS values were significantly increased (р = 0,018 and р = 0.041) compared to the group with CCVD (tab. 2).

Analysis of the total sample revealed higher rHcy values in males with RS compared to females (0.17 (0.14–0.24) μМ vs. 0.15 (0.12–0.21) μМ; p = 0.026).

In the group of IS patients enrolled, T2D was diagnosed in 68 cases, and stress hyperglycemia was diagnosed in 107 cases (tab. 3). The duration of T2D of less than 5 years was observed in 22 patients (32.4%), and the disease duration exceeding 5 years was observed in 46 patients (67.6%). In the group of patients with a combination of IS and T2D, the HbA1c level was 7.8% (6.8–9.6). The levels of less than 6.5% were defined in 107 patients (61.1%), the levels of 6.5–6.9% were found in 32 patients (18.3%), 20 patients (11.4%) had the HbA1c levels of 7–7.9%, and the levels exceeding 8% were found in 16 patients (9.2%).  

Stroke, resulting from cerebral artery thrombosis with a cardiac source of embolism, was more common in patients with T2D (р < 0.0001), and atherothrombotic stroke was more common in patients with stress hyperglycemia (р < 0.0001). No intergroup differences in the prevalence of lacunar stroke were found (р = 0.162). The examined groups of patients were comparable in the neurological impairment severity and their functional status during the acute period of stroke (see tab. 3).

In the group of patients with a combination of IS and T2D, the significantly decreased tCys, tGSH, and tHcy levels were observed compared to the group of patients with IS and stress hyperglycemia (р < 0.0001). However, there were no significant differences in the rCys, rGSH, and rHcy levels between the groups of patients with T2D and stress hyperglycemia. Probably, this entails higher Cys RS, GSH RS, and Hcy RS values in the group of patients with IS and T2D (р < 0.0001) (tab. 4).

No significant correlations between the blood glucose levels and the levels of LMWTs were revealed both in the total group of patients with IS and in the distinct groups of patients (T2D and stress hyperglycemia).

When the patients with a combination of IS and T2D were divided based on the degree of functional recovery by day 21, analysis of LMWTs showed that in the group of patients with slight limitations in performing the activities of daily living  (mRS score 0–2), the Cys RS and GSH RS were significantly higher compared to patients with severe functional limitations (Cys RS  5.26 vs. 3.51 μM, р = 0.043; GSH RS 4.92 vs. 2.79 μM, р = 0.018) (tab. 5).

The logistic regression procedures were used to identify the markers of poor outcome (mRS ≥ 3) in patients with a combination of IS and T2D. The predictive value of GSH RS is presented in figure. ROC analysis of 68 patients with IS in the ICA basin and T2D showed that the threshold level GSH RS ≤ 4.06% during the first 24 h of IS is a predictor of poor IS outcome (mRS ≥ 3, 21 days after IS). Sensitivity of the model was 63.6%, specificity was 69.6%, and AUC was 0.74 ± 0.09, which corresponds to the good quality in predicting the functional recovery in patients with IS.

DISCUSSION

Chronic hyperglycemia, associated with T2D, is considered one of the risk factors for IS. Glucose is a chemical compound, which actively interacts with proteins and lipids, forming the advanced glycation end products [20, 21]. Hyperglycemia, chronic oxidative stress, and mitochondrial dysfunction, associated with T2D, result in endothelial dysfunction, impaired angiogenesis, activation of hemostasis, and the increased blood-brain barrier permeability [22]. Hyperglycemia enhances oxidative processes by lowering the levels of vitamins Е, С, and other antioxidants (uric acid) [11], it is also an additional factor that stimulates and facilitates the reactive oxygen species formation [23].   

Chronic hyperglycemia is defined based on the degree of glucose binding to hemoglobin, and the percentage of НbA1c. The higher the level of НbA1, the higher blood glucose levels were observed in the patient during the last 3 months. The normal levels do not exceed 6.5%. The increase in НbA1c by 1% increases the risk of stroke by 17% [24]. In our study, the duration of T2D was 7 years, and the level of НbA1c was 7.8 %.

In individuals with IS, the effects of hyperglycemia on the brain are mediated by both impaired cerebral microcirculation, and the toxic effects on brain tissue. Lactate accumulation, free radical formation, development of cytotoxic cerebral edema, and abnormal blood-brain barrier permeability are attributed to hyperglycemia [25, 26].

Hyperglycemia in individuals with T2D and IS results in depletion of the antioxidant system and disorders of all types of tissue metabolism. The thiol/disulfide ratio (SH/SS) is the indicator of cellular and tissue redox homeostasis, as well as of the blood plasma antioxidant capacity [11, 27].  

Glutathione is an endogenous thiol tripeptide, composed of cysteine, glutamic acid, and glycine. It is being synthesized continuously, but at a relatively low rate [28]. Glutathione is also a factor involved in regulation of glucose metabolism in patients with cardiovascular disorders [13]. Polymorphism of glutathione S-transferase, the enzyme, involved in the glutathione-mediated detoxification of xenobiotics, contributes to the early cardiovascular complications of diabetes mellitus [29].

In individuals with acute oxidative stress, the total content of glutathione (GSH) is decreased, and the levels of oxidized glutathione (GSSG) are increased, which results in the accelerated GSH/GSSG cycle [27].

Hyperglycemia during the acute period of IS is associated with poor functional outcome [4], however, the efficacy of insulin-based glucose lowering therapy and its effects on the severity and outcome of stroke have not been proven [1].

The findings show that T2D in patients with IS is associated with the rapid decrease in plasma tCys, tGSH, tHcy, rGSH, and GSH RS, along with the increase in  Cys RS and Hcy RS.

Perhaps the chronic oxidative stress results in depletion of LMWTs in blood plasma. Irreversible disposal of GSH may be associated with intense oxidative stress, when GSH is exported from cells in order to prevent the significant shift in the redox equilibrium [28]. Presumably, it is the main mechanism explaining the lack of correlations between tGSH, tCys and tHcy, as well as between tGSH and rGSH. No significant correlations between plasma glucose levels and LMWTs have been revealed. This indicates that alterations in the metabolism of LMWTs associated with T2D are mediated by non-glycemic mechanisms.

The GSH RS value of 4.06% or lower during the first 24 h of IS is a predictor of poor IS outcome (mRS ≥ 3, three weeks after IS). Therefore, the search for approaches to glutathione metabolism correction in patients with T2D may be regarded as the potential therapeutic objective during the acute period of IS.

CONCLUSIONS

T2D is a factor having a major impact on the metabolism of LMWTs in patients with IS. Regardless of the lack of correlations between the glucose levels and LMWTs, T2D was associated with reduced total content of homocysteine, cysteine and glutathione, and the glutathione redox status of 4.06% or lower during the first 24 h of IS was associated with poor functional outcome. Glutathione metabolism correction in individuals with a combination of IS and T2D may have a positive impact on the course of IS.