Haemostasis in hypoxia and exercise: an antioxidant intervention

Introduction: Patients with occlusive arterial disease typically undergo focal hypoxia in the affected limbs and exercise training is utilised to improve regional perfusion in the ischaemic tissues. Hypoxia and exercise can independently and synergistically increase oxidative stress (Bailey et al., 2009); and oxidative stress has been implicated as a modulator of haemostasis (Görlach, 2005), and as a catalyst for thrombin generation (Wang et al., 2009). We therefore designed a novel proof-of-concept study among young, healthy individuals to assess the impact of high-dose antioxidant prophylaxis on the haemostatic system and markers of thrombin generation. We hypothesised that since oxidative stress can increase global thrombogenicity, the intervention would attenuate changes in haemostasis previously shown by our laboratory (Fall et al., 2011). Methods: Following pre-intervention blood sampling, 40 male subjects were randomly assigned to either a placebo or intervention group. The intervention group were administered 1000 mg per day of vitamin C with 900 IU per day of vitamin E and the placebo group were given tablets of identical appearance, but no nutritional value for 8 weeks and returned to the laboratory after an overnight fast. Bloods were taken after 10 minutes supine rest in normoxia, after 6 hours of passive exposure to hypoxia (12% inspired oxygen) in an environmental chamber and after a cycling challenge to volitional exhaustion in hypoxia. Bloods were sampled into sodium citrate vacutainers, centrifuged at 600 g at 4 °C and stored at -80 °C until analysis. Bloods were batch-analysed for plasma levels of thrombin-antithrombin complex (T-AT), prothrombin fragments 1and2 (PF1+2), activated partial thromboplastin time (aPTT), prothrombin time (PT), thrombin time (TT), fibrinogen (FB) and d-dimer (DD). Results: Eight weeks intervention significantly increased levels of T-AT and PF1+2 compared to controls (Pandlt;0.05) but this difference was abrogated after hypoxia and remained so after exercise. There were no differences between groups in aPTT, PT, TT, FB or DD in any time point. There were group effects in several markers; aPTT shortened with exercise, as did PT. TT elongated with hypoxia and remained so post-exercise; there was a reduction in FB with hypoxia which returned to baseline following exercise. D-dimer remained unchanged. Discussion and conclusions: Increases in T-AT and PF1+2 with intervention suggests an increase in thrombin generation and factor Xa activity (Mannucci, 1994) in the supplemented group. Interestingly, this difference does not translate into the more routine haemostasis analysis that would be performed in the patient population. The group data suggest that hypoxia alone does not affect haemostasis; this is contrary to existing research (Wang et al., 2009). The addition of exercise hastens both the intrinsic and extrinsic pathways of coagulation in line previous research (Fall et al., 2011), but this is not translated into down-stream markers, most likely due to the activation of antithrombin III. The T-AT and PF1+2 data suggest that free radicals in fact limit thrombin generation at rest. Hypoxia reverses this phenomenon, normalising thrombin generation. We are currently focusing on understanding whether the haemostatic:fibrinolytic balance is disturbed.