Activated factor XI is associated with recurrent venous thromboembolism in patients following pulmonary embolism: a cohort study
P S Michel, T M Kuncewicz, E Paszek, A UndasAbstract
Background
Venous thromboembolism (VTE) carries a substantial risk of recurrence, estimated at 1-5% of patients within one year despite guideline-recommended management. Detectable activated factor XI (FXIa) in circulating blood, as demonstrated previously1, predisposes to atherothromboembolism and venous thrombosis; however, the prognostic value of FXIa in patients following pulmonary embolism (PE) is unknown.
Purpose
We investigated whether detectable circulating FXIa is associated with VTE recurrence in patients following PE.
Methods
This prospective cohort study included 198 white noncancer patients aged 18 to 64 years who were receiving anticoagulation for at least 3 months following documented PE (with or without concomitant deep vein thrombosis). Baseline FXIa was quantified by measuring clotting time in the presence of the inhibitory monoclonal antibody anti-FXIa2. Symptomatic recurrent VTE and major bleeding were recorded during a median follow-up of 83 (IQR 76-90) months.
Results
FXIa was detectable in 41 (20.7%) patients at enrollment and did not correlate with demographic or clinical parameters. During follow-up, VTE recurred in 34 PE patients (17.2%; 2.5% per year), including 9 individuals who stopped anticoagulation. Patients with and without recurrent VTE were similar with regard to demographic, clinical, and laboratory variables, including anticoagulant type and cessation (Table 1). Detectable FXIa was 11.7-fold more prevalent in PE patients with recurrent VTE compared with the remainder (85.3% vs 7.3%, p<0.001; Table 1, Figure 1). At a time horizon (τ) of 96 months, equivalent to the 90th centile of follow-up duration3, the difference in restricted mean survival time was 34.1 months (bootstrapped 95% confidence interval [CI]: 24.1-43.6, p<0.001), reflecting 1.6-fold longer event-free survival in patients without circulating FXIa. After adjustment for age, unprovoked VTE, and number of prior VTE events, FXIa was independently associated with VTE recurrence (hazard ratio [HR]: 38.0, 95% CI: 15.3-120.2, p<0.001). The incidence of major bleeding during follow-up was not affected by detectable FXIa (Table 1).
Conclusion
We have shown that FXIa can help identify PE patients at risk of recurrent VTE, irrespective of anticoagulation status or type. Detectable circulating FXIa may encapsulate residual prothrombotic risk in this patient group. Given promising outcomes of FXI/FXIa inhibitors in primary prevention of VTE, it is tempting to suggest that these agents could be used in PE patients with activated FXI.For image description, please refer to the figure legend and surrounding text.For image description, please refer to the figure legend and surrounding text.