Causal effect of serum matrix metalloproteinase levels on venous thromboembolism: a Mendelian randomization study

Abstract

Objectives: Serum matrix metalloproteinase (MMP) levels are associated with cardiovascular diseases. However, the causal associations between serum levels of specific MMPs and venous thromboembolism (VTE) remain unclear. The present study sought to explore the causal relationship between serum MMP levels and VTE by using the Mendelian randomization (MR) method.

Methods: In this study 2-sample MR study, the exposure data on serum MMP levels were derived from genome-wide association studies involving 21,758 individuals from 13 cohorts of European descent. The outcome data on VTE, including deep vein thrombosis and pulmonary embolism, were derived from the FinnGen research project. The primary method used was the inverse-variance weighting method. The MR-Egger intercept test and the Cochran Q test were used to evaluate pleiotropy and heterogeneity.

Results: Using the inverse-variance weighting method, higher serum MMP-12 levels were found to be associated with an increased risk of VTE (odds ratio, 1.04; 95% confidence interval, 1.01 to 1.07; p=0.001). Moreover, there was a weak association between the levels of certain MMPs and VTE. Sensitivity analyses revealed no significant heterogeneity and pleiotropy in our study, and the Steiger directionality test did not reveal a significant reverse causation association.

Conclusions: There is a causal association between MMP-12 levels and VTE, which may have substantial implications for the diagnostic and therapeutic strategies used for VTE.

Keywords: Matrix metalloproteinases; Mendelian randomization analysis; Pulmonary embolism; Venous thromboembolism; Venous thrombosis.

Figure 1.

Three core assumptions of the 2-sample Mendelian randomization design in this study. ① The instrumental variable is strongly associated with the serum MMP levels. ② The instrumental variable is independent of confounders. ③ The instrumental variable affects the VTE (including DVT and PE) solely via the serum MMP levels. MMP, matrix metalloproteinase; VTE, venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism.

Figure 2.

The causal association between MMP and VTE (including DVT and PE) using the inverse-variance weighted Mendelian randomization method. A p-value <0.0033 was considered statistically significant. A p-value ranging from 0.0033 to 0.0500 was regarded as suggestive evidence. OR, odds ratio; CI, confidence interval; MMP, matrix metalloproteinase; VTE, venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism.

Figure 3.

Trends in the causal associations between (A) the MMP-1 level and PE, (B) the MMP-3 level and PE, (C) the MMP-3 level and DVT, (D) the MMP-7 level and PE, (E) the MMP-10 level and VTE, (F) the MMP-12 level and VTE, and (G) the MMP-12 level and PE. MMP, matrix metalloproteinase; VTE, venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism; MR, Mendelian randomization.