Essential hypertension (EH) is one of the most common modifiable risk factors for cardiovascular diseases and death. According to some estimates, more than 1 billion adults suffer from EH. Their number is expected to have reached 1.5 billion in 2025 [1]. It was long thought that end-organ damage resulted from the increased hydrostatic pressure against blood vessel walls. However, recently a few research studies have been conducted [2–10] into the cytokine pathways in EH complications, including myocardial infarction (MI), acute cerebrovascular events (ACVE), and transient ischemic attacks (TIA), that pose a significant risk of death and morbidity to the working-age population. It is hardly disputable that end-organ damage is the ultimate sequela of chronic inflammation in the presence of elevated blood pressure and atherosclerosis. However, it is still unclear whether certain cytokines have a prognostic potential in the assessment of long-term risks for cardiovascular disorders in patients with stage II EH. The aim of this study was to investigate a correlation between peripheral blood concentrations of some cytokines in patients with stage II EH and the rate of cardiovascular complications (MI, ACVE, TIA) developed during the 5-year observation period.

METHODS

The study was carried out at the facilities of the Regional Vascular Center of Mordovian Republican Clinical Hospital No. 4, Mordovian Republican Clinical Hospital No. 3 and the Department of Immunology, Microbiology and Virology of Ogariov Mordovian State University in 2013 through 2018. The study recruited 490 participants with stage II EH. Of them, 200 patients were prescribed comparable antihypertensive therapies and then followed up for 5 years. Among them were 100 females and 100 males with the mean age of 57.5 ± 1.2 years at the beginning of the study.

The following inclusion criteria were applied: stage II EH lasting for 10–14 years; working age (females under 60, males under 65); comparable treatment regimens prescribed (monotherapy with ACE inhibitors or ACEIs+ diuretics); informed consent to participate. Patients with comorbidities, type I diabetes mellitus, metabolic syndrome, allergies, autoimmune conditions, secondary hypertension, infections or mental health problems 1 month before the study, alcoholism/drug abuse, and those who refused to participate were excluded from the study. The following cytokines were measured in the peripheral blood samples of 100 female and 100 male patients with stage II EH in order to assess their potential in predicting long-term cardiovascular complications (MI and ACVE): IL1β, IL1α, IL1ra, IL18, IL18BP, IL37, IL6, sIL6r, LIF, sLIFr, IGF-1, IGFBP-1, TNFα, sTNF-RI, sVCAM-1, IL17, IL2, IL4, IL10, TGF-β1, IL8, CX3CL1, CXCL10, INFγ, M-CSF, IL34, VEGF-A, erythropoietin, and a group of vasoactive peptides including NО, iNOS, eNOS, ADMA, SDMA, Nt-proСNP, and Nt-proBNP. These proteins were selected because they are either synthetized by vascular cells or the end organs have receptors recognizing the listed cytokines. Their concentrations were determined using immunoassays. In the 5-year follow-up period, the patients were surveyed on the phone using the original questionnaire; complications (MY, ACVE, TIA) were noted down and later correlated to the patients’ medical records.

The obtained data were processed in Statisticа 8.0 (StatSoft; ver 8.0). Normality of data distribution was tested using the one-sample Kolmogorov-Smirnoff test. In this article, the data are presented as an arithmetic mean (М), a standard deviation (SD) and 95% confidence interval for the mean (95% CI) for normal distribution. Student’s t-test was used to compare the main groups. Absolute and relative risks of end-organ damage (MI, ACVE, TIA) were computed, as well as 95% CI for sensitivity (Se) and specificity (Sp). Uni- and multivariate Cox regressions were calculated.

RESULTS

We correlated the levels of IL1β, IL1α, IL1ra, IL37, IL18, IL18BP, IL6, sIL6r, LIF, sLIFr, TNFα, sTNF-RI, sVCAM, IL2, IL8, IL4, IL10, IFNγ, IGF-1, IGFBP-1, M-CSF, IL34, VEGF-A, CX3CL1, CXCL10, TGFβ1, IL17A, and erythropoietin in the peripheral blood of 200 patients in treatment for stage II EH who had been suffering from the condition for 10-14 years to the development of cardiovascular complications during the 5-year follow-up period. The retrospective analysis revealed a decline in IL1ra (р < 0.001) and IL10 (р < 0.001) concentrations and a rise in IL1β (р < 0.001), TNFα (р < 0.001), and M-CSF (р < 0.001) levels in the group of 47 patients who developed cardiovascular complications in the follow-up period, including MI (20 patients), ACVE (14 patients) and TIA (13 patients), as compared to the group of patients who did no develop any long-term complications during the study (tab. 1). No significant differences were observed in the average values of other parameters between the groups of patients with and without complications (р > 0.05).

Further analysis of interquartile ranges for the patients with EH who developed complications after 6–10 years of observation demonstrated (:tab_2) that at IL1ra < 513 pg/ml, the rate of complications over those 5 years increased 2.24 times (р < 0.05) reaching 45% (Sp was 83%, Se was 41.7%), χ2 = 5.25 (р < 0.05), С = 0,33 (a moderate correlation); at IL10 < 26.3 pg/ml the risk grew 1.94 times (р < 0.05), reaching 31% (Sp 54.9%, Se 66%), χ2 = 6.26 (р < 0.05), С = 0.17 (a weak correlation); at IL1β > 18.7 pg/ml the risk increased 2.37 times (р < 0.05) reaching 38% (Sp 59%, Se 80%), χ2 = 7.59 (р < 0.05), С = 0.22 (a moderate correlation); at IL6 > 23.8 the risk grew 1.76 times (р < 0.05) reaching 30% (Sp 54%, Se 63.8%), χ2 = 6.45 (р < 0.05), С = 0.17 (a weak correlation). At M-CSF concentrations in the blood serum ranging from 389 to 453 pg/ml the risk of cardiovascular complications in year 5 and in the next 5 years grew 3.87-fold (р < 0.001), as compared to M-CSF concentrations < 389 pg/ml and > 453 pg/ml, reaching 62% (Sp 81.6%, Se 66%), χ2 = 32.5 (р < 0.001), С = 0.6 (a relatively strong correlation). M-CSF was the most significant risk predictor in comparison with other cytokines. The analysis of interquartile ranges for the patients who developed cardiovascular complications in years 6–10 of the observation did not reveal any significant correlations between TNFα concentrations measured in year 5 and the analyzed risks.

The univariate Cox regression analysis confirmed a reliable correlation between the changes in the peripheral blood cytokine concentrations in patients with stage II EH (IL1β > 18.8 pg/ml; IL1ra < 511 pg/ml; IL6 > 23.8 pg/ml; IL10 < 26.3 pg/ml; 389 pg/ml < M-CSF < 453 pg/ml) and the risk of cardiovascular disorders during the observation period (tab. 3).
The analysis of relationships between all studied parameters and vasoactive peptides (tab. 4, tab. 5) demonstrated that cytokines that could be regarded as potential long-term predictors of cardiovascular complications in patients with stage II EH, namely IL1ra, IL10, IL1β, TNFα, and M-CSF, correlate with the serum levels of asymmetric dimethylarginine (ADMA), a vasoactive peptide. Therefore, when building the pathogenic model, we added ADMA to the list of variables subjected to the multivariate regression analysis to assess the independence of the revealed risk factors. This has a pathogenic significance and affects the diagnostic and clinical value of the recorded changes. M-CSF concentrations in the blood serum of patients with stage II EH have the strongest correlation with ADMA. The model subjected to the multivariate analysis also included classic risk factors as recommended by the international and Russian guidelines for the risk assessment of complications in patients with hypertension. Based on the results yielded by the multivariate analysis, increased rates of end-organ damage in patients with stage II EH reported during the 5-year observation period correlated with M-CSF levels in the range between 389 and 453 pg/ml (р < 0.001) regardless of patients’ sex, as well as with the classic risk factor: LDP > 3.0 mmol/L (р < 0.01) (tab. 6).

DISCUSSION

Significant correlations have been reported between EH complications, including ACVE and MI, and the peripheral blood concentrations of IL17, IFNγ, TNFα, IL6 [:lit_2, 11;], sTNF-RI [9], IL1 [12, 13], CXC chemokines [6], LIF [14], IL12 [15], and other cytokines. However, there have been few longitudinal studies of this problem, and only a small range of cytokines has been analyzed. Considering that the literature on the role of cytokines in the pathogenesis of EH is scarce, it is important to study the dynamics of these immunoregulatory factors in patients who develop EH complications and to identify factors that maintain their statistical and pathogenic significance when other cytokines or classic risk factors are introduced into the pathogenic model. We have found that changes in IL1ra, IL10, IL1β, IL6, and M-CSF levels correlate with the development of complications (MI, ACVE, TIA) during a 5-year period. In our study, the increase in the concentrations of anti-inflammatory IL1ra and IL10 correlated with a lower number of complications. Correlations between other analyzed cytokines and the risk for EH complications were not observed. Out findings are not fully consistent with the literature reporting correlations between LIF, IL1α, etc. and the risk for developing EH complications [10, 14]. This could be due to different inclusion criteria applied in different studies (the sex ratio, age, antihypertensive treatment, normalized blood pressure, etc) and indicates the high significance of the observational group homogeneity, which affects the pathogenic significance of the obtained data. Importantly, our study shows that only cytokines correlating with ADMA levels can be regarded as potential predictors of end-organ damage in patients undergoing treatment against stage II EH and suffering from the pathology for 10–14 years. ADMA is a methylated analog of L-arginine (a substrate for NO synthesis) that competitively inhibits the functional activity of eNOS [16], curbing the NO synthesis and leading to its poor availability for vasorelaxation and vasoprotection [17]. This pathogenic pathway is important: the role of ADMA and SDMA in pathology is currently in the focus of scientific research. This study established that M-CSF was the only independent criterium (from the entire spectrum of the analyzed parameters) that had a high predictive value (even when compared to ADMA and classic risk factors) in the assessment of risks for developing ACVE, MI and TIA in patients undergoing treatment for stage II EH who had been suffering from this condition for 10 to 14 years. This confirms the priority of the cytokine in the M-CSF–ADMА correlational pathogenic model with a subsequent cascade of reactions causing progression of the pathology. Earlier, we published an article demonstrating a direct correlation between M-CSF concentrations > 453 pg/ml and the levels of VEGF-A in the peripheral blood, which was consistent with a significant increase in the myocardial collateral blood flow (coronary angiography) and might explain a low rate of MI in the studied cohort of patients [18], affecting the total risk for cardiovascular complications. M-CSF can activate MAP-kinases via the M-CSFR-1 receptor; the kinases play a key role in the production of VEGF-А by activating ERK, increasing the p38 and JNK promoter activity and stabilizing VEGF-А mRNA in a dose-dependent pattern [19].

CONCLUSIONS

The data yielded by this study prove that changes in the cytokine concentrations (IL1β > 18.8 pg/ml, IL1ra < 511 pg/ml, IL6 > 23.8 pg/ml, IL10 < 26.3 pg/ml) measured in the peripheral blood of patients suffering from stage II EH for 10 to 14 years and undergoing antihypertensive treatment correlate with a 5-year rate of cardiovascular complications (MI, ACVE, TIA). Only M-CSF at concentrations between 389 and 453 pg/ml can be regarded as a predictor of cardiac and cerebrovascular complications. Although the obtained data have a theoretical significance, M-CSF concentrations in the range from 389 to 453 pg/ml are a highly specific (81%) but lowly sensitive (66%) parameter in terms of predicting MI, ACVE and TIA in the studied cohort of patients. This means that an additional criterium should be added to the model to improve the diagnostic (prognostic) value of this cytokine. The role of patients’ individual characteristics, such as genetic components including CSF1R ТС/СА rs386693509: ТС/СА variants in the established correlations should be further studied in patients with the progredient disease.