Is pulmonary embolism recurrence linked with the severity of the first event? A French retrospective cohort study.

Objectives Severity of a first pulmonary embolism (PE) is sometimes proposed as a criterion for prolonging anticoagulant treatment. However, little evidence supports this idea. We attempted to determine the connection between severity of first PE and the risk of recurrence. Participants Patients admitted with PE between 2012 and 2018 and for whom anticoagulant treatment had been discontinued were followed. PEs were classified according to the severity into the following two groups: those with associated cardiac involvement (increased cardiac biomarker(s) and/or echocardiographic right ventricular dysfunction) and those with no cardiac involvement which were classified as non-severe. Recurrence-free survivals were estimated using the Kaplan-Meier method and compared using the log-rank test. Results 417 patients with PEs (186 with cardiac involvement) were followed for at least 1 year after discontinuation of treatment with a mean follow-up of: 3.5±1.9 years. 72 patients (17.3%) experienced venous thromboembolism recurrence: 24 (5.8%), 44 (12 %) and 72 (28.3 %) respectively, at 1, 2 and 5 years. In 63 patients (88%), recurrence was a PE. Mean time to onset of recurrence was 24.9±19.9 months. At 5 years, the recurrence rate is higher when the first PE was associated with cardiac involvement p=0.043. In contrast, in patients with provoked PE, the recurrence rate is higher when the first PE event was associated with cardiac involvement: p=0.032. Multivariate analysis demonstrates that PE severity is an independent factor of recurrence (HR 1.634 (1.015–2.632), p=0.043). Conclusion We report for the first time a possible link between a higher recurrence rate and the severity of the first PE. This result which must be confirmed in a dedicated prospective trial could become an important criterion for the duration of anticoagulant therapy after a PE. Trial registration number NCT04980924.