Incidence of venous thromboembolism in persons with cancer – a population-based cohort study in Sweden

Background Venous thromboembolism (VTE) is a well-recognised complication of cancer, but reported incidence rates vary due to differences in population characteristics and study design. Objective To estimate incidence rates of VTE among persons with cancer and to assess the association between cancer and first-ever VTE in a population-based Swedish cohort. Methods This cohort study was based on participants in the Västerbotten Intervention Programme who underwent health examinations between 1985 and 2014. Individuals with prior cancer or VTE were excluded. Cancer and VTE diagnoses were identified through Swedish national registers, and all potential VTE events were manually validated. Participants were followed prospectively from health examination until incident VTE, death, emigration, or end of follow-up. Incidence rates were calculated by time since diagnosis. Cox proportional hazards models with cancer as a time-varying exposure were used to estimate HRs for first-ever VTE. Results Among 105,997 participants, 9,643 developed cancer during follow-up, and 677 experienced a first-ever VTE after cancer diagnosis. The VTE incidence in individuals with cancer was 13.2 per 1000 person-years (95% CI 12.3–14.3). Incidence was highest during the first six months after diagnosis (60.2 per 1000 person-years). The incidence varied across cancer types (4.9–201.1 per 1000 person-years). Cancer was significantly associated with VTE, with an adjusted hazard ratio of 8.3 (95% CI 7.5–9.2). Conclusion In a population-based cohort we found a high incidence of VTE after cancer diagnosis. The incidence declined rapidly over time and varied across cancer types. Our study confirms the association between cancer and VTE.