Plasma Adiponectin Levels and Risk of Future Venous Thromboembolism

Abstract

Background: Obesity is a major modifiable risk factor for venous thromboembolism (VTE). Adiponectin is a hormone-like protein primarily secreted by adipocytes and reported to have anti-thrombotic properties by regulation of key haemostatic factors in vitro. However, the association between adiponectin and VTE remains unclear.

Aims: To investigate the association between plasma adiponectin levels and risk of future incident VTE, and to assess whether adiponectin mediates the association between elevated body mass index (BMI) and VTE.

Methods: We performed a nested case-control study including 381 incident VTE cases and 772 age-and sex-matched controls from the Tromsø4 survey (1994/95-2007). Adiponectin and high-sensitivity C-reactive protein (hsCRP) were measured in plasma samples from cohort baseline. Logistic regression was used to calculate odds ratios (ORs) with 95% confidence intervals (CIs) for VTE across quartiles (Q) of adiponectin, established in controls using sex-specific quartile cut-offs, and adjusted for age, sex, BMI and hsCRP. To assess potential short-term associations, time-restricted analysis was performed only including VTE cases from ≤5 years after inclusion. The Karlson-Holm-Breen-method was used to estimate the mediating effect of adiponectin.

Results: High adiponectin levels were associated with decreased VTE risk, with OR for VTE of 0.76 (95%CI 0.52-1.10) for Q4 vs. Q1 (reference) in the full study period. The association between adiponectin and VTE was stronger with analysis restricted to the first 5 years of follow-up (OR 0.42 [95%CI 0.21-0.83] for Q4 vs. Q1). Low plasma adiponectin levels were estimated to mediate 12.8% (95%CI -42.1 to 102.8%) and 7.4% (95%CI -1.3 to 23.5%) of the VTE risk in overweight (BMI 25-30kg/m2) and obesity (BMI ≥30kg/m2), respectively.

Conclusion: High levels of adiponectin in plasma were associated with decreased risk of future VTE, especially in the short-term. Adiponectin might be a mediator of the association between elevated BMI and VTE, but the estimates were uncertain.