Thromboprophylaxis in Patients With Fontan Circulation

Abstract

Background: The optimal strategy for thromboprophylaxis in patients with a Fontan circulation is unknown.

Objectives: The aim of this study was to compare the efficacy and safety of aspirin, warfarin, and nonvitamin K oral anticoagulants (NOACs) in a network meta-analysis.

Methods: Relevant studies published by February 2022 were included. The primary efficacy outcome was thromboembolic events; major bleeding was a secondary safety outcome. Frequentist network meta-analyses were conducted to estimate the incidence rate ratios (IRRs) of both outcomes. Ranking of treatments was performed based on probability (P) score.

Results: A total of 21 studies were included (26,546 patient-years). When compared with no thromboprophylaxis, NOAC (IRR: 0.11; 95% CI: 0.03-0.40), warfarin (IRR: 0.23; 95% CI: 0.14-0.37), and aspirin (IRR: 0.24; 95% CI: 0.15-0.39) were all associated with significantly lower rates of thromboembolic events. However, the network meta-analysis revealed no significant differences in the rates of major bleeding (NOAC: IRR: 1.45 [95% CI: 0.28-7.43]; warfarin: IRR: 1.38 [95% CI: 0.41-4.69]; and aspirin: IRR: 0.72 [95% CI: 0.20-2.58]). Rankings, which simultaneously analyze competing interventions, suggested that NOACs have the highest P score to prevent thromboembolic events (P score 0.921), followed by warfarin (P score 0.582), aspirin (P score 0.498), and no thromboprophylaxis (P score 0.001). Aspirin tended to have the most favorable overall profile.

Conclusions: Aspirin, warfarin, and NOAC are associated with lower risk of thromboembolic events. Recognizing the limited number of patients and heterogeneity of studies using NOACs, the results support the safety and efficacy of NOACs in patients with a Fontan circulation.

Keywords: aspirin; nonvitamin K oral anticoagulants; single ventricle; warfarin.

FIGURE 1. PRISMA Flow Diagram of Studies Included in Data Search

PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram for the meta-analysis detailing identification of the included studies.

FIGURE 2. Pooled Incidence Rates of Thromboembolic Events for Each Treatment Group

The top represents no treatment, followed by aspirin, warfarin, and nonvitamin K oral anticoagulant (NOAC).

FIGURE 3. Pooled Incidence Rates of Major Bleeding for Each Treatment Group

The top represents no treatment, followed by aspirin, warfarin and nonvitamin K oral anticoagulant (NOAC).

FIGURE 4. Results of the Random-Effects Network Meta-Analysis

(A and B) Network plots showing the strategies that have been tested for their efficacy in preventing thromboembolic events (A) and safety in terms of major bleeding (B). Lines are weighted according to the number of studies comparing the 2 connected strategies. (C and D) League plots representing the results of the network meta-analyses comparing the effect of all strategies. The estimates (incidence rate ratio [IRR], 95% CI) are presented for each comparison; IRR >1 favors the row-defining treatment, and IRR <1 favors the column-defining treatment. (E and F) Forest plots summarizing the network estimates for each thromboprophylaxis strategy compared with no thromboprophylaxis. (G and H). All strategies have a lower risk for thrombosis compared with no treatment and no increased risk of bleeding. Funnel plots for thromboembolic events (G) and major bleeding (H). The results resemble a symmetrical inverted funnel suggesting low between study heterogeneity. NOAC = nonvitamin K oral anticoagulant.

FIGURE 4. Results of the Random-Effects Network Meta-Analysis

(A and B) Network plots showing the strategies that have been tested for their efficacy in preventing thromboembolic events (A) and safety in terms of major bleeding (B). Lines are weighted according to the number of studies comparing the 2 connected strategies. (C and D) League plots representing the results of the network meta-analyses comparing the effect of all strategies. The estimates (incidence rate ratio [IRR], 95% CI) are presented for each comparison; IRR >1 favors the row-defining treatment, and IRR <1 favors the column-defining treatment. (E and F) Forest plots summarizing the network estimates for each thromboprophylaxis strategy compared with no thromboprophylaxis. (G and H). All strategies have a lower risk for thrombosis compared with no treatment and no increased risk of bleeding. Funnel plots for thromboembolic events (G) and major bleeding (H). The results resemble a symmetrical inverted funnel suggesting low between study heterogeneity. NOAC = nonvitamin K oral anticoagulant.

CENTRAL ILLUSTRATION. Treatment Effects of Different Antithrombotic Strategies in the Fontan Circulation

Treatment effects of aspirin, warfarin, and nonvitamin K oral anticoagulants (NOACs) were compared using no thromboprophylaxis as a reference in a bivariate analysis. Incidence rate ratios (IRRs) of thromboembolic events (TEs) are reported on the y-axis and IRRs of major bleeding (MB) on x-axis. No thromboprophylaxis as a reference has a neutral effect on TE and MB, which equals 1 on the x- and y-axes. NOAC were associated with the greatest reduction in thromboembolic events, yet the greatest risk of major bleeding. Aspirin tended to have the most favorable overall profile, with both a significantly reduced risk of thromboembolic events and a tendency toward lower risk of MB.