Physical activity before venous thromboembolism and risk of recurrence in a population-based inception cohort.

Introduction: The association between moderate-to-vigorous physical activity (MVPA) and recurrent venous thromboembolism (VTE) is unclear, but an improved understanding could inform behavioral health recommendations.

Methods: The Heart and Vascular Health study, set in a large integrated healthcare system, identified adults with a validated incident VTE between January 2002 and December 2010. An inception cohort was formed from these cases and followed for a first recurrent VTE through December 2014. Usual MVPA pre-incident VTE was self-reported by 1381 adults via telephone interview, and MVPA amount was calculated in metabolic equivalent of task (MET) hours (h) per week. Multivariable-adjusted Cox proportional hazards models estimated adjusted hazard ratios (HR_adj) for any MVPA versus none and MVPA amount, continuously and in quartiles, in MET-h/week among participants reporting any MVPA. Secondary analyses separately evaluated MET-h/week, by intensity.

Results: During follow-up (median = 5.23 years), 288 (20.9%) individuals developed a recurrent VTE. There was no evidence of an association between any MVPA versus none and VTE recurrence (HR_adj=1.24, [95% confidence interval [CI]: 0.80, 1.91]). Among participants with any MVPA, there was no evidence of an association between MVPA in MET-h/week (HR_adj per 7.5 MET-h/week = 1.00, [95% CI: 0.98, 1.03]), nor quartiles of MVPA (p-trend = 0.62) with VTE recurrence risk. In secondary analyses there was no evidence of an association of MET-h/week of moderate or vigorous physical activity (PA) with VTE recurrence.

Conclusions: In this cohort of adults who experienced incident VTE, there was no evidence of an association between self-reported MVPA pre-incident VTE and VTE recurrence risk.