Epidemiological evidence indicates that the coagulation system plays an important role in the pathophysiology of atherosclerosis (1), and increased incidence of cardiovascular events has been associated with elevated circulating levels of markers of hemostatic activation (2–4). A prothrombotic state is characterized by intrinsic subclinical activation of the hemostatic system and is a well-recognized risk factor for cardiovascular events in the general population (5, 6) and in patients with hypertension (7, 8) and early renal failure (9). In hypertension, a prothrombotic state is associated with subclinical changes of the arterial tree (10–12), impaired left ventricular diastolic properties (13), and intrarenal hemodynamic changes (14) that are associated with decreased glomerular filtration rate (15). Furthermore, a link of the prothrombotic state with an activated renin-angiotensin-aldosterone system has been demonstrated in hypertension (16, 17) suggesting interaction of the hemostatic system with blood pressure regulatory mechanisms.

A prothrombotic state can be detected in many disease states including those characterized by exogenous or endogenous hypercortisolism (18). In patients with Cushing’s syndrome, elevated circulating levels of a multitude of hemostatic markers were demonstrated, together with evidence of an impaired fibrinolytic system (19–21). Interestingly, initial studies reported activation of the hemostatic system also in patients with subclinical Cushing’s syndrome (22, 23) such as those with incidentally detected adrenal masses in the absence of overt clinical features of hypercortisolism (24). Along these lines, studies conducted in other groups of patients have reported observations that might suggest the existence of a relationship between circulating glucocorticoids and different components of the hemostatic and fibrinolytic systems (25–27).

Although patients with essential hypertension have cortisol secretory rates and response to overnight dexamethasone suppression (DST) within the physiologic range, minor differences in circulating cortisol levels might have some impact on the coagulation system. To date, no information is available on the possible relationship of circulating cortisol with the hemostatic-fibrinolytic balance in patients with hypertension. This information would be relevant for better understanding of mechanisms that contribute to the development and progression of hypertensive organ damage.

The aim of this study was to test the hypothesis that cortisol secretory rates and response to DST are associated with changes in the activity of the coagulation system. We examined the relationship of cortisol daily production and response to DST with a broad panel of hemostatic markers in nondiabetic patients with hypertension who were free of major cardiovascular and renal complications.

One-hundred-forty-nine patients (age, 48±13 years; 77 males, 72 females) with essential hypertension were recruited for data analysis. Forty (27%) patients had BMI >30 of whom 21 (14%) had grade 1 (BMI 30–35) and 19 (13%) grade 2 obesity. Sixty-two (42%) patients had never been treated with antihypertensive agents and the remaining 87 (58%) who were taking an average of 1.3 antihypertensive agents (angiotensin-converting enzyme inhibitors or angiotensin receptor antagonists, 39%; calcium-channel blockers, 38%; diuretics, 19%; beta-blockers, 19%; alpha-blockers, 7%) had their drugs withdrawn for at least two weeks before the study.

For statistical reasons, patients were grouped according to tertiles of AUC-cortisol and DST-cortisol. The clinical characteristics of the study patients are summarized in Table 1 together with general biochemistries. Across AUC-cortisol and DST-cortisol tertiles, no significant differences were observed in demographic and anthropometric characteristics, systolic and diastolic blood pressure, plasma glucose and glycated hemoglobin, plasma lipids, electrolyte levels, renal function, and C-reactive protein. No differences were also observed in frequency of previous use of antihypertensive drugs. In particular, no significant differences were observed from patients previously treated with diuretics and the remaining patients. Table 2 summarizes hormonal measurements of the study patients showing that plasma cortisol levels measured at 8 AM, 3 PM, and 12 AM increased significantly across AUC-cortisol and DST-cortisol tertiles, whereas plasma ACTH, active renin and aldosterone, and daily urinary excretion of epinephrine, norepinephrine, and dopamine did not differ among groups. Table 3 shows the hemostatic variables showing that plasma D-dimer, F1 + 2, and vWF levels were significantly and progressively higher across both AUC-cortisol and DST-cortisol tertiles. No significant differences among groups were observed in fibrinogen, t-PA, PAI-1, AT-III, and protein C and protein S.

Analysis of univariate correlations ( Table 4 ) showed that D-dimer, F1 + 2, and vWF were significantly and directly correlated with patients’ age, AUC-cortisol ( Figure 1 ) and DST-cortisol ( Figure 2 ). No further significant relationship of plasma cortisol levels was observed with the other hemostatic markers. F1 + 2 was also inversely correlated with 24-hour creatinine clearance while fibrinogen, tPA and PAI-1 were directly correlated with BMI. Only fibrinogen was directly correlated with systolic blood pressure. None of the cortisol and hemostatic variables was correlated with C-reactive protein.

Multivariate regression analysis was conducted including different hemostatic markers as the dependent variables and age, sex, systolic blood pressure, creatinine clearance, and either AUC-cortisol (Model 1) or DST-cortisol (Model 2) as the independent variables, respectively (Supplementary Material). In both models, age was independently and directly related with F1 + 2 and vWF, and sex was independently related with log D-dimer. Both AUC-cortisol and DST-cortisol were significantly and independently correlated with log D-dimer (β-coefficient 0.001, P=0.029; β-coefficient 0.003, P<0.001; respectively), F1 + 2 (β-coefficient 0.267, P=0.002; β-coefficient 1.217, P<0.001; respectively), and vWF (β-coefficient 0.106, P=0.035; β-coefficient 0.877, P<0.001; respectively).

A prothrombotic state is associated with major cardiovascular events in hypertension, and many factors can contribute to hemostatic activation in hypertensive patients (37). Evidence previously obtained in patients with Cushing’s syndrome indicates that excess cortisol could contribute to a hypercoagulable state. We tested the hypothesis that even minor differences in regulation of cortisol secretion are associated with a prothrombotic state in patients with essential hypertension. Results show that plasma levels of D-dimer, F1 + 2, and vWF are progressively greater with increasing levels of plasma cortisol daily profile and response to DST. Both AUC-cortisol and DST-cortisol are significantly correlated with plasma D-dimer, F1 + 2, and vWF levels independently of age, BMI, blood pressure, and renal function. These findings indicate that differences in regulation of cortisol production within the physiologic range might contribute to a prothrombotic state in patients with hypertension.

Hypercortisolism is associated with high risk of cardiovascular morbidity and mortality that might be related to a multiplicity of factors (38) including a prothrombotic state. After initial studies that suggested presence of a hypercoagulable state in patients with Cushing’s syndrome (39), spontaneous activation of the hemostatic system was consistently shown in patients with hypercortisolism. Increased circulating levels of fibrinogen (19, 20, 40, 41), D-dimer (20, 23, 40), vWF (20, 42–44), AT-III (19–21, 44, 45), protein C-protein S complex (21, 22, 41, 45), and PAI-1 (19, 20, 41) were reported together with changes of additional coagulation factors and hemostatic tests (19–21, 41, 43, 46) in several cross-sectional comparisons of patients with Cushing’s syndrome with healthy subjects. Some studies suggested that these changes could be more relevant in patients with ACTH-producing adenomas (44) and increased levels of procoagulants and antifibrinolytics were reported also in children with Cushing’s syndrome that resolved after surgical treatment (45). Reversal of hypercoagulability was reported also in adults after surgical resolution of disease (43, 46), although this was not confirmed in other studies (40, 41). Interestingly, hemostatic changes were reported also in patients with excess plasma cortisol or incidentally detected adrenal masses in the absence of overt clinical features of hypercortisolism (22, 23), further supporting the hypothesis of a contribution of cortisol levels to regulation of coagulation and fibrinolysis (18). In our highly selected hypertensive patients who had physiologic suppression of plasma cortisol after DST, cortisol levels were significantly and independently associated with markers of intrinsic hemostatic activation. This study expands to a physiologic range of hormonal secretion the evidence of the potential role of cortisol in causing a prothrombotic state.

Due to its high prevalence in the general population, arterial hypertension is considered the leading cardiovascular risk factor. In patients with essential hypertension, we have previously demonstrated that a prothrombotic state contributes to the development of target organ damage and thereby might increase cardiovascular morbidity and mortality (7, 9–15). In fact, substantial experimental and clinical data indicate that platelets and coagulation cascade are important determinants of both atherogenesis and atherothrombosis (1). The hemostatic system exerts several actions on the vasculature that could influence the structure of the arterial wall and presumably the progression of atherosclerotic lesions. Hemostatic components have been implicated in causing disruption of endothelial lining, leukocyte recruitment, oxidative stress, vascular inflammation, migration of vascular smooth muscle cells, apoptosis, and angiogenesis (47, 48). For these reasons, identification of conditions that may contribute to a prothrombotic state in hypertension would be relevant as an issue for possible preventive interventions.

Many factors could cause activation of the hemostatic system in hypertension and the present study demonstrates that even minor differences in regulation of cortisol secretion are associated with a prothrombotic state as defined by higher levels of D-dimer, F1 + 2, and vWF. Measurements of fibrin D-dimer, the principal breakdown fragment of fibrin, and F1 + 2 that is released when coagulation factor Xa converts prothrombin to thrombin provide a reliable estimation of the overall state of activation of the coagulation pathways. On the other hand, vWF enhances adhesion of platelets to subendothelial collagen and increases the stability of coagulation factor VIII in the hemostatic cascade. Older age, obesity, and impaired renal function are associated with a prothrombotic state, and this is why statistical analysis was corrected for these variables showing that D-dimer, F1 + 2, and vWF were all independently related to both AUC-cortisol and DST-cortisol. Thus, differences in cortisol secretion within a physiological range could contribute to a prothrombotic state and thereby to hypertensive organ damage. Also, to this point it should be noticed that minor differences in regulation of cortisol secretion were previously found to be associated with impaired glucose metabolism in nondiabetic hypertensive patients (36) and greater left ventricular mass (30), suggesting additional mechanisms that might mediate detrimental cardiovascular effects of cortisol in essential hypertension.

Mechanisms that could link excess cortisol secretion to the hypercoagulable state are mostly speculative. Cortisol-induced up-regulation of gene transcription of multiple coagulation factors with intrinsic activation of the hemostatic cascade is the most likely mechanism. Activation of markers of the overall activity of the coagulation system such as D-dimer and F1 + 2 would support this possibility. On the other hand, observation of significantly increased levels of vWF might suggest that the prothrombotic state caused by increased circulating cortisol could be related to an enhanced metabolic function of endothelial cells. This was also suggested by Fatti et al. who reported increased levels of additional markers of endothelial activation (thrombin-antithrombin complex and plasmin-antiplasmin complex) in patients with Cushing’s syndrome (43). Hypercortisolism may also affect the multimeric structure of vWF causing an overexpression of abnormally high molecular weight multimers, capable of inducing spontaneous platelet aggregation (42).

Limitations to the present study need to be considered. First, the cross-sectional design limits the possibility to establish causality of the relationship between cortisol secretion and the prothrombotic state, although independence of this relationship from confounders in the multivariate analysis would suggest so. However, statistical adjustments cannot account for all biological and pathophysiological variables. Also, the possibility that plasma cortisol levels and prothrombotic markers causally affect one another cannot be excluded. Second, use of a selected clinic sample of white Caucasian hypertensive patients might limit the possibility to extend the present findings to a broader context. Third, inclusion of a significant proportion of patients who had been treated with antihypertensive drugs might have affected the results. However, it must be noticed that no differences were found in either hemostatic markers or cortisol measurements between these patients and those who were treatment naïve, nor differences were observed among patients who were treated with different categories of antihypertensive drugs. Last, due to the relatively small size of this study, the possibility that associations are merely due to chance should be considered.

This study demonstrates for the first time that even minor differences in daily cortisol secretion and cortisol levels after overnight suppression are independently associated with markers of a prothrombotic state in patients with essential hypertension who are free of major cardiovascular and renal complications. These findings expand this evidence beyond the boundary of clinical and subclinical Cushing’s syndrome. Results also provide better knowledge of factors that might contribute to develop a prothrombotic state in hypertension, thereby increasing target organ damage and, in turn, the risk of cardiovascular events. This study might have important clinical implications opening new paths to the possibility to identify those hypertensive patients at higher risk of development of subclinical organ damage. Detection of cortisol levels in the upper limit of normality in conjunction with markers of hemostatic activation could be useful to guide physicians toward more aggressive treatment and control of blood pressure and additional risk factors. Also, these findings open a window to the possibility to effectively prevent hypertensive organ damage through specific interventions on cortisol production and hemostatic system. Possible benefits of these interventions will have to be tested in future studies.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Institutional Review Board - Department of Medicine - University of Udine. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

GB: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Supervision, Writing – original draft. AD: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Supervision, Writing – original draft. LB: Data curation, Investigation, Methodology, Software, Supervision, Validation, Visualization, Writing – review & editing. AV: Data curation, Investigation, Methodology, Software, Supervision, Validation, Visualization, Writing – review & editing. NB: Data curation, Investigation, Supervision, Validation, Visualization, Writing – review & editing. CV: Data curation, Investigation, Supervision, Validation, Visualization, Writing – review & editing. CC: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing. LS: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing.