Search

Aims/hypothesis Individuals with type 2 diabetes mellitus and subclinical inflammation have stimulated coagulation, activated platelets and endothelial dysfunction. Recent studies with the direct factor Xa inhibitor rivaroxaban in combination with low-dose aspirin demonstrated a significant reduction of major cardiovascular events, especially in individuals with type 2 diabetes and proven cardiovascular disease. Therefore, we asked the question of whether treatment with rivaroxaban could influence endothelial function, arterial stiffness and platelet activation. Methods We conducted a multi-centre, prospective, randomised, open-label trial in 179 participants with type 2 diabetes (duration 2–20 years), subclinical inflammation (high-sensitivity C-reactive protein 2–10 mg/l) and at least two traits of the metabolic syndrome to compare the effects of the direct factor Xa inhibitor rivaroxaban (5 mg twice daily) vs aspirin (100 mg every day) on endothelial function (assessed by forearm occlusion plethysmography), skin blood flow (assessed by laser-Doppler fluxmetry), arterial stiffness (assessed by pulse wave velocity) and serum biomarkers of endothelial function and inflammation. Furthermore, we investigated phosphorylation of vasodilator-stimulated phosphoprotein (VASP) in platelets, the concentration of platelet-derived microparticles (PMPs) and the effects of isolated PMPs on HUVEC proliferation in vitro. Results Rivaroxaban treatment for 20 weeks (n = 89) resulted in a significant improvement of post-ischaemic forearm blood flow (3.6 ± 4.7 vs 1.0 ± 5.2 ml/100 ml, p = 0.004), a numerically increased skin blood flow and reduced soluble P-Selectin plasma level vs aspirin. We did not find significant differences of arterial stiffness or further biomarkers. Neither rivaroxaban nor aspirin influenced VASP phosphorylation of platelets. The number of PMPs increased significantly with both rivaroxaban (365.2 ± 372.1 vs 237.4 ± 157.1 μl−1, p = 0.005) and aspirin (266.0 ± 212.7 vs 201.7 ± 162.7 μl−1, p = 0.021). PMPs of rivaroxaban-treated participants stimulated HUVEC proliferation in vitro compared with aspirin. Rivaroxaban was associated with a higher number of bleeding events. Conclusions/interpretation Our findings indicate that the direct factor Xa inhibitor rivaroxaban improved endothelial function in participants with type 2 diabetes and subclinical inflammation but also increased the risk of bleeding. Trial registration: ClinicalTrials.gov NCT02164578. Funding The study was supported by a research grant from Bayer Vital AG, Germany. Graphical abstract

We analysed the relationship between fasting plasma glucose, carotid intima media thickness and some atherosclerosis risk factors in 307 non-diabetic individuals. Male (n = 120) and female subjects (n = 187) with a familial history of Type II diabetes mellitus and/or obesity and hyperlipoproteinaemia were examined in the age group 40–70 years. Plasma triglycerides, total and high-density-lipoprotein cholesterol, plasminogen activator inhibitor were measured by conventional methods. Specific insulin, proinsulin and C-peptide were measured by specific enzyme immunoassay. Intima media thickness increased in quintiles for fasting plasma glucose in men, but not in women. There was a rise of triglycerides, body mass index, waist to hip ratio, plasminogen activator inhibitor, true insulin, proinsulin, C-peptide and a decrease of high-density-lipoprotein cholesterol in quintiles for fasting plasma glucose. Fasting plasma glucose was found to be significantly positively correlated to intima media thickness, body mass index, waist to hip ratio, haemoglobin A1 c, insulin, C-peptide, triglycerides, plasminogen activator inhibitor and significantly negatively correlated to high density lipoprotein cholesterol. However, the correlation of fasting plasma glucose to intima media thickness was no longer significant after adjustment for age and sex. After adjustment for age and sex intima media thickness was significantly correlated to body mass index, total cholesterol, triglycerides, albuminuria and inversely correlated to high-density-lipoprotein cholesterol. In multivariate analysis age, male sex, high-density-lipoprotein cholesterol and total cholesterol were significant determinants of intima media thickness. Our data suggest that a weak association exists between fasting plasma glucose and intima media thickness, which may be mediated by a clustering of risk factors in the upper range of non-diabetic fasting plasma glucose level with a central role for dyslipidaemia. [Diabetologia (1998) 41: 706–712]