Total Tissue Factor Pathway Inhibitor and Incident Venous Thrombosis: The Longitudinal Investigation of Thromboembolism Etiology

Introduction: Tissue Factor Pathway Inhibitor (TFPI) inhibits tissue factor, a potent coagulation initiator. Limited evidence suggests that low TFPI levels are associated with increased risk of venous thromboembolism (VT).

Methods: VT was ascertained in the Longitudinal Investigation of Thromboembolism Etiology, which combines data from the Atherosclerosis Risk in Communities Study and the Cardiovascular Health Study. Total TFPI was measured in a nested case-control sample. With 12 years of follow-up, 534 cases of new VTE were age-, sex-, and race-frequency matched with 1091 controls. The odds ratio for VT was determined for quartiles of TFPI, and for the bottom 5% versus the top 95% in logistic regression models adjusting for demographics (age, race, gender, study), coagulation factors (Factors VII, VIII, IX, XI, D-dimer), and other VT risk factors. To evaluate for greater than additive interactions we calculated the percent relative excess risk of interaction between TFPI and other VT risk factors.

Results: Mean TFPI in ng/mL (standard deviation) in cases and controls was 36.4 (12.8) and 35.0 (11.1), respectively. Increasing quartiles of TFPI were associated with male gender and increasing age, body mass index, factors VII, VIII, IX, XI, and D-dimer (all p<0.05). TFPI quartiles were not related to ethnicity, factor V Leiden, or the prothrombin gene polymorphism. Compared with those in the upper 95%, those in the bottom 5% of TFPI (32 cases, 56 controls) had an age-, sex-, race-, and cohort-adjusted odds ratio of 1.35 (0.86, 2.12) for VT. Adjusting for factors VII, VIII, IX, XI, and D-dimer increased the odds ratio to 1.93 (1.05, 3.53). Further addition of BMI to the coagulation factor adjusted model slightly attenuated the odds ratio to 1.70 (0.98, 2.93). There were no associations of TFPI in quartiles or as a continuous variable with VT. TFPI in the bottom fifth percentile did not demonstrate greater than additive interaction with factor V Leiden or with higher BMI.

Conclusions: Associations of high TFPI levels with VT were masked by concurrent higher procoagulant levels. After adjusting for procoagulant factors and D-dimer, TFPI ≤ 5% was associated with a 1.9-fold increased odds of VT. These data suggest high total TFPI may be a response to procoagulant states.