Myocardial infarction as a transient risk factor for incident venous thromboembolism: Results from a population-based case-crossover study
File(s) with restricted access are under embargo until 2020-06-16
AuthorSejrup, Joakim Knutsen; Børvik, Trond; Grimnes, Gro; Isaksen, Trond; Hindberg, Kristian; Hansen, John-Bjarne; Morelli, Vania Maris; Brækkan, Sigrid Kufaas
Patients with myocardial infarction (MI) are at increased short-term risk of venous thromboembolism (VTE). The mechanisms behind this association are unclear. We aimed to investigate the impact of acute MI as a transient risk factor for incident VTE while taking other concomitant VTE risk factors into account. We conducted a case–crossover study of VTE patients (n = 707) recruited from the fourth survey of the Tromsø Study. VTE risk factors and hospitalizations were registered during the 90-day period preceding the VTE diagnosis (hazard period) and in four 90-day control periods. Conditional logistic regression was used to calculate odds ratios (ORs) with 95% confidence intervals (CIs) for VTE according to acute MI and after adjustment for other risk factors. Additionally, we applied a mediation analysis to quantify how much the known transient risk factors account for the observed effect of MI on VTE risk. MI was recorded in 13 (1.8%) of the hazard periods and in 6 (0.2%) of the control periods, which yielded a crude OR of 11.9 (95% CI: 3.9–36.7). Adjustment for immobilization and infection yielded an OR of 2.7 (95% CI: 0.6–11.2). The OR was attenuated to 2.6 (95% CI: 0.6–11.9) after further adjustment for major surgery, trauma, red blood cell transfusion, and central venous catheterization. Approximately 60% of the association between MI and VTE was mediated through infection and immobilization. In conclusion, our findings suggest that the increased VTE risk after MI may to a large extent be explained by concomitant conditions related to MI, particularly infections and immobilization.