Abstract

In addition to the “traditional” risk factors for venous trombo-embolism (VTE), like age, trauma and immobilisation, inflammation could also be regarded a risk factor for VTE. For example, patients with acute inflammatory conditions (sepsis), but also patients with chronic inflammation, like inflammatory bowel disease (IBD) and rheumatoid arthritis (RA), have an increased risk of thrombosis.

Inflammation can lead to activation of coagulation, and vice-versa, coagulation also has considerable effects on overall inflammatory activity. First, the inflammatory cytokine network induces several pro-thrombotic conditions including insulin resistance, dyslipidaemia, endothelial dysfunction and alteration of coagulation and fibrinolysis. Second, activation of the extrinsic coagulation system and impairment of the fibrinolytic pathway may contribute to amplify and perpetuate the inflammatory response. Previous studies have reported several blood parameters that reflect a prothrombotic state in RA. These include increased levels of thrombin-antithrombin complex, prothrombin fragment F1+2, von Willebrand factor, plasmin-alpha2-antiplasmin complex and D-dimer, as well as an increased platelet count. Impaired fibrinolysis combined with increased antithrombin levels have also been reported in RA. An important mediator in the inflammatory pathway is tumor necrosis factor-α (TNF-α). In the general population, TNF-α induces a disbalance between clotting and fibrinolysis, resulting in a hypercoagulable state. Since TNF-α is the key player in RA, RA is an ideal “human model” to study the interplay between inflammation and coagulation. Hence, RA can be considered as a pro-thrombotic state, which explains partly why patients with RA are at increased risk of thrombo-embolic cardiovascular events.(1)

Only one small study suggested that TNF-inhibitors (TNFi) is accompanied with normalization of thrombotic biomarkers: an improvement of clinical and laboratory parameters as well as a reduction in the activation of coagulation and endothelial dysfunction was found in RA patients treated with a TNFi. In addition, we previously demonstrated that combination therapy with corticosteroids improves the procoagulant state that exists in early RA. (2)

Nowadays, tapering of biological therapies is becoming more and more standard of care. However, the effects on the coagulation status in RA are unknown. In light of the growing evidence of an increased cardiovascular morbidity and mortality in RA, mostly independent of traditional risk factors, treatment strategies in RA should not only aim at relieving symptoms and inhibiting joint destruction but should have a beneficial effect on the vasculature and haemostasis to reduce cardiovascular events. Although modest, there is evidence suggesting a beneficial effect of TNFi on the haemostatic status in RA. Unfavourable changes in haemostatic markers, such as TAT, F1+2, vWF, PAP, D-dimer and thrombin generation, which indicate a pro-thrombotic state, may therefore (re)occur when RA patients stop with TNFi treatment. We first assessed arterial wall inflammation with 18F-FDG PET scans in RA patients in remission under TNFi therapy or DMARD therapy versus controls. The FDG uptake in the aorta in DMARD remission patients was similar to the controls, whereas the uptake in RA patients in remission under antiTNF was significantly higher than in controls either when looking at the overall aortic uptake or the most diseased segment. Theoretically, stopping TNF blockade in these patients might lead to increased inflammation and thus coagulation activation. Therefore, we are presently investigating it and to what extent tapering/stopping TNFi therapy induces a pro-thrombotic state in RA patients.