Key result
Forty-seven phenotypic exposures, including body mass index (Log RR 0.40) and varicose veins (Log RR 1.90), were found to be causally associated with deep vein thrombosis in a hypothesis-free Mendelian randomization analysis.
Why the study?
Despite the prevalence and associated morbidity of deep vein thrombosis, its underlying causes are not well understood.
What are the causal risk factors for deep vein thrombosis using a phenome-wide Mendelian randomization approach?
Observational (n=337,159)
Mendelian randomization
What are the causal risk factors for deep vein thrombosis using a phenome-wide Mendelian randomization approach?
Effect estimate: Log RR 0.40 (95% CI 0.32-0.47)
p-value: p=1.60E-22
Hyperthyroidism, varicose veins, COPD, and BMI are putative causal risk factors for DVT, with the circulating protein PAI-1 mediating the BMI-DVT relationship.
Summary Background Deep vein thrombosis (DVT) is the formation of a blood clot in a deep vein. DVT can lead to a venous thromboembolism (VTE), the combined term for DVT and pulmonary embolism, a leading cause of death and disability worldwide. Despite the prevalence and associated morbidity of DVT, the underlying causes are not well understood. Objective To leverage publicly available genetic summary association statistics to identify causal risk factors for DVT. Methods & Results We conducted a Mendelian randomization phenome-wide association study (MR-PheWAS) using genetic summary association statistics for 973 exposures and DVT (6,767 cases and 330,392 controls in UK Biobank). There was evidence for a causal effect of 57 exposures on DVT risk, including previously reported risk factors (e.g. body mass index - BMI and height) and novel risk factors (e.g. hyperthyroidism, chronic obstructive pulmonary disease (COPD) and varicose veins). As the majority of identified risk factors were adiposity-related, we explored the molecular link with DVT by undertaking a two-sample MR mediation analysis of BMI-associated circulating proteins on DVT risk. Our results indicate that circulating neurogenic locus notch homolog protein 1 (NOTCH1), inhibin beta C chain (INHBC) and plasminogen activator inhibitor 1 (PAI-1) influence DVT risk, with PAI-1 mediating the BMI-DVT relationship. Conclusion Using a phenome-wide approach, we provide putative causal evidence that hyperthyroidism, varicose veins, COPD and BMI enhance the risk of DVT. The circulating protein PAI-1 has furthermore a causal role in DVT aetiology and is involved in mediating the BMI-DVT relationship.
No takes yet. Share an insight, caveat, or question.
Constantinescu et al. (2018) conducted an observational in Deep vein thrombosis (n=337,159). 973 phenotypic exposures (e.g., body mass index, varicose veins) vs. Unexposed/reference was evaluated on Causal association with deep vein thrombosis (Log RR 0.40, 95% CI 0.32-0.47, p=1.60E-22). Forty-seven phenotypic exposures, including body mass index (Log RR 0.40) and varicose veins (Log RR 1.90), were found to be causally associated with deep vein thrombosis in a hypothesis-free Mendelian randomization analysis.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: