Oral anticoagulant reversal and mortality in trauma patients: a multicentre propensity score-matched cohort study

Abstract

Background: Oral anticoagulant (OAC) therapy increases bleeding risk but its impact on trauma outcomes and the benefit of reversal remains uncertain. This study aimed to evaluate 1/the effect of preinjury OAC therapy on trauma mortality and 2/the protective role of OAC reversal and its associated thrombotic risk.

Methods: We conducted an observational study using a prospective multicenter trauma registry between January 2012 and December 2023. OAC-treated patients were matched with non-OAC-treated patients using a propensity score. Univariable and multivariable logistic regressions assessed associations between OAC therapy and day 1 and day 7 mortality. The effect of guideline-concordant OAC reversal was evaluated. Thrombotic complications were recorded.

Findings: Of the 27,426 trauma patients, 3% were OAC-treated. They were older, had more comorbidities, and experienced higher mortality. After matching (n = 2196), OAC therapy remained independently associated with increased mortality (day 1: OR 2·21, 95% CI [1·41-3·43]; day 7: OR 2·06, [1·41-3·00]), with greater risk from vitamin K antagonists (VKA) than direct oral anticoagulants (DOAC). Guideline-concordant OAC reversal, achieved only in 21% of cases, independently reduced mortality at day 1 (OR 0·10, 95% CI [0·03-0·31], p < 0·01) and day 7 (OR 0·51, 95% CI [0·22-0·97], p < 0·01). No significant association was found between reversal and thrombotic complications.

Interpretation: Preinjury OAC therapy substantially increased trauma mortality, particularly with VKA. Guideline-concordant reversal was associated with reduced mortality in both VKA- and DOAC-treated patients without excess thrombotic risk but remains underused. These findings emphasise the urgent need for systematic implementation of reversal strategies in OAC-treated trauma patients.

Funding: The Traumabase registry is funded by several Regional Health Agencies (Agences Régionales de Santé, ARS): ARS Île-de-France, ARS Occitanie, ARS Grand Est, ARS Hauts-de-France, and ARS Auvergne-Rhône-Alpes. The registry is also funded by the French Road Safety Observatory-Road Safety Delegation Service (Observatoire National Interministériel de la Sécurité Routière-Délégation à la Sécurité Routière).

Keywords: Bleeding; Mortality; Oral anticoagulant; Reversal; Thrombosis; Trauma.

Fig. 1

Flowchart of the study.

Fig. 2

Rates (%) of OAC reversal in OAC-treated patients according to the anticoagulant and trauma. For example, the bottom band indicates that among OAC-treated patients undergoing neurosurgery 64·8% had guideline-concordant reversal and 12·4% had no reversal. Anti-Xa DOAC, Direct Oral Anticoagulant targeting Factor Xa; ISS, Injury Severity Score; VKA, Vitamin K Antagonist.

Fig. 3

Multivariate logistic regression analysis of 1-day mortality in OAC-treated patients. GCS, Glasgow Coma Scale; DOAC, Direct Oral Anticoagulant; ISS, Injury Severity Score; OAC, Oral Anticoagulant; VKA, Vitamin K Antagonist.

Fig. 4

A, B, C:Association between quality of reversal and mortality according to anticoagulant agent using survival analysis with Cox regression in the ARIANE cohort. OAC, Oral Anticoagulant; DOAC, Direct Oral Anticoagulant; VKA, Vitamin K Antagonist.

Fig. 5

Thrombotic Events in ICU according to anticoagulant reversal in trauma patients with preinjury OAC therapy. Among the 257 pre-injury OAC-treated patients who were hospitalised in the ICU for more than 48 h, 12 (4·7%) patients experienced a thrombotic event in the ICU. The figure described these events: the type of pre-injury anticoagulant agent is listed in brackets after the type of thrombotic event. The light gray portion of the bar indicates the time from hospital admission to the thrombotic event, the dark gray portion indicates the time in the ICU after the event and the black portion indicates death. Red diamonds indicate the initiation of therapeutic anticoagulation. PE, pulmonary embolism; DVP, deep vein thrombosis.