Background: Both prophylaxis and treatment of venous thromboembolism (VTE: deep venous thrombosis (DVT) and pulmonary emboli (PE)) with anticoagulants are associated with significant risks of major and fatal hemorrhage. Anticoagulation treatment of VTE has been the standard of care in the USA since before 1962 when the U.S. Food and Drug Administration began requiring randomized controlled clinical trials (RCTs) showing efficacy, so efficacy trials were never required for FDA approval. In clinical trials of 'high VTE risk' surgical patients before the 1980s, anticoagulant prophylaxis was clearly beneficial (fatal pulmonary emboli (FPE) without anticoagulants = 0.99%, FPE with anticoagulants = 0.31%). However, observational studies and RCTs of 'high VTE risk' surgical patients from the 1980s until 2010 show that FPE deaths without anticoagulants are about one-fourth the rate that occurs during prophylaxis with anticoagulants (FPE without anticoagulants = 0.023%, FPE while receiving anticoagulant prophylaxis = 0.10%). Additionally, an FPE rate of about 0.012% (35/28,400) in patients receiving prophylactic anticoagulants can be attributed to 'rebound hypercoagulation' in the two months after stopping anticoagulants. Alternatives to anticoagulant prophylaxis should be explored.
Methods and findings: The literature concerning dietary influences on VTE incidence was reviewed. Hypotheses concerning the etiology of VTE were critiqued in relationship to the rationale for dietary versus anticoagulant approaches to prophylaxis and treatment.Epidemiological evidence suggests that a diet with ample fruits and vegetables and little meat may substantially reduce the risk of VTE; vegetarian, vegan, or Mediterranean diets favorably affect serum markers of hemostasis and inflammation. The valve cusp hypoxia hypothesis of DVT/VTE etiology is consistent with the development of VTE being affected directly or indirectly by diet. However, it is less consistent with the rationale of using anticoagulants as VTE prophylaxis. For both prophylaxis and treatment of VTE, we propose RCTs comparing standard anticoagulation with low VTE risk diets, and we discuss the statistical considerations for an example of such a trial.
Conclusions: Because of (a) the risks of biochemical anticoagulation as anti-VTE prophylaxis or treatment, (b) the lack of placebo-controlled efficacy data supporting anticoagulant treatment of VTE, (c) dramatically reduced hospital-acquired FPE incidence in surgical patients without anticoagulant prophylaxis from 1980 - 2010 relative to the 1960s and 1970s, and (d) evidence that VTE incidence and outcomes may be influenced by diet, randomized controlled non-inferiority clinical trials are proposed to compare standard anticoagulant treatment with potentially low VTE risk diets. We call upon the U. S. National Institutes of Health and the U.K. National Institute for Health and Clinical Excellence to design and fund those trials.