Safety and efficacy of anticoagulation for atrial fibrillation in sepsis: a systematic review and meta-analysis
S Faizollah Zadeh Ardebili, L Galvao da Silva De Paula, M Angel Samaniego Laguna, T Shah, L Melfior, N De Paula PortilhoAbstract
Background
Atrial fibrillation (AF) in the setting of sepsis is associated with an increased risk of ischemic stroke during hospitalisation. Whether the use of anticoagulation (AC) in this population can improve outcomes remains uncertain.
Purpose
To evaluate the safety and efficacy of AC (compared to control) in the management of patients with AF and sepsis through a systematic review and meta-analysis following the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) and Cochrane guidelines.
Methods
We systematically searched PubMed, Embase, and Cochrane databases until May 2025 for eligible studies. When available, data from propensity score matched (PSM) studies were prioritised to minimise confounding. Studies assessing in-hospital outcomes of AC use were analysed separately from out-patient studies. Major bleeding was defined as a Bleeding Academic Research Consortium (BARC) type ≥ 3 event. We pooled risk ratios (RRs) and mean differences (MDs) with their corresponding 95% confidence intervals (CIs) using a random effects model in R studio. Heterogeneity was assessed with I2 statistics.
Results
After screening 1,467 articles, we included six retrospective studies comprising 29,485 patients (49.9% receiving AC). Four studies (3 PSM) assessed in-hospital outcomes, while two studies were out-patient designs. Compared to control, in-hospital AC was associated with a significant reduction in mortality (RR 0.62; 95% CI 0.40-0.94; p = 0.026; I2 = 91.6%; Fig. 1A), while increasing the risk of bleeding (RR 1.19; 95% CI 1.10-1.29; p < 0.001; I2 = 0%; Fig. 1B) and non-significantly increasing the risk of major bleeding according to BARC criteria (RR 1.49; 95% CI 0.99-2.26; p = 0.058; I2 = 0%; Fig. 1C). No differences were observed between groups for in-hospital ischemic stroke (RR 0.94; 95% CI 0.77-1.15; p = 0.561; I2 = 0%; Fig. 1D), or ICU and hospital lengths of stay. In unadjusted analyses of two studies, out-patient AC was associated with a significant increase in ischemic stroke (RR 1.58; 95% CI 1.02-2.44; I2 = 0%; p = 0.039), but similar bleeding events to control (RR 1.17; 95% CI 0.88-1.55; I2 = 0%; p = 0.286).
Conclusion
In this observational meta-analysis of 29,485 patients, in-hospital AC in patients with sepsis and AF was associated with a significant reduction in mortality - at the cost of increased bleeding - but not ischemic stroke. The paradoxical increase in stroke risk with out-patient AC prescription may be due to confounding by indication. Until prospective randomised trials are available, AC use in this setting should be individualised and based on estimated net clinical benefit.Figure 1.Forest plots (in-patient AC use)Graphical Abstract