Effective hemostasis and lower mortality with DOAC reversal agent, but inappropriate usage occurs
Performance of idarucizumab as antidote of dabigatran in daily clinical practice
Introduction and methods
International guidelines recommend usage of idarucizumab, a monoclonal antibody fragment that binds to the direct oral anticoagulant (DOAC) dabigatran, for urgent dabigatran reversal in the presence of life-threatening bleeding or urgent surgery associated with high risk of bleeding [1,2]. However, practice-based data on the usage of idarucizumab for urgent dabigatran reversal are scarce. Therefore, this study aimed to determine the appropriateness of idarucizumab usage as well as its hemostatic effectiveness and clinical outcomes in daily practice.
This Dutch observational, multicenter cohort study included consecutive patients who were treated with idarucizumab between 2016 and 2018 (n=88) and followed them for 90 days. Data on idarucizumab usage and outcomes were collected by investigating medical records, including medical notes, laboratory results, radiology reports, and other relevant details.
The primary outcome was the appropriateness of idarucizumab usage, based on criteria proposed in an expert consensus of the ISTH on reversal of DOACs. Uncontrollable bleeding was defined as meeting one or more of the following criteria: symptomatic intracranial bleeding, a reduction in hemoglobin (Hb) of ≥5g/dL, transfusion of ≥4 units of blood or packed cells, bleeding requiring use of intravenous inotropic agents, or necessitating surgical intervention (one that could not be delayed for ≥8 hours). An additional criterion for appropriateness in this study was indicators for the presence of dabigatran plasma levels. Secondary outcomes were hemostatic effectiveness after administration for urgent reversal in bleeding events and incidence of thromboembolism, (re-)bleeding and death after 90 days. 53 (60%) Patients who received idarucizumab presented with bleeding, and 35 (40%) required urgent intervention.
Appropriateness of idarucizumab usage
- Inappropriate usage of idarucizumab was observed in 25 patients (28%). 14 of whom required intervention (40% of that group) and 11 presented with bleeding (21% of bleeders).
- The reasons for inappropriate usage were that all 14 interventions could have been delayed for ≥8 hours and 8 of the 11 (72%) bleeding complications were not considered uncontrollable.
- In three bleeding patients (5.7%) no dabigatran was detected in plasma; two had a last intake >72 hours and normalized aPTT levels, and one patient used rivaroxaban instead of dabigatran.
- Nearly all bleeding events in which idarucizumab was considered inappropriate were located in the gastrointestinal tract (73%).
- Effective treatment with idarucizumab was seen in 32 of 48 (67%) bleeding patients in whom assessment was possible.
- The effectiveness did not differ significantly between intracranial and extracranial bleedings (RR: 1.2, 95%CI: 0.53–2.7), nor between traumatic and non-traumatic bleeding (RR: 1.5, 95%CI: 0.40– 6.1).
- Seven of 16 bleeding patients (44%, 5 intracranial) in whom no effective hemostasis was achieved, died, compared with 2 of 31 patients (6.5%) with effective hemostasis (RR: 7.0, 95%CI: 1.6–30).
- During follow-up 4 thrombotic complications were observed: 2 ischemic strokes on day 1 (before DOAC resumption) and day 41 (after DOAC resumption), and 2 pulmonary embolisms (1 fatal), on day 5 (before DOAC resumption) and day 21 (after DOAC resumption).
- Four (re-)bleeding complications were observed: a major pericardial re-bleeding 6 days after restart of anticoagulation in a 65-year old patient who developed ischemic stroke on the first day after idarucizumab administration, a fatal pericardial bleeding (after DOAC resumption) and two minor bleedings (before DOAC resumption), all occurring within ten days and at the same anatomical location of the index presentation.
- All thrombotic and bleeding complications occurred in patients who initially presented with bleeding.
- During follow-up period, 17 patients died (19%); 10 (11%) within five days due to sepsis (3 patients), post-operative shock (3, 1 possibly bleeding-related), intracranial bleeding (2), pericardial bleeding (1), and lung bleeding (1). Of these 17 patients, 12 had presented with bleeding (6 intracranial) and five underwent urgent intervention.
- In 60 of 88 patients (68%) antithrombotic therapy was restarted; after a median of three days (IQR: 1-5) in 31 of 35 patients requiring intervention (89%), and after a median of six days (IQR: 3-11)in 30 of 53 patients (57%) presenting with bleeding.
In this observational study idarucizumab usage was considered inappropriate in 28% of patients, mostly because interventions could have been delayed and gastrointestinal bleeding complications might have responded to supportive measures alone. However, it should be noted that the criteria applied to judge appropriateness have not been tested in clinical trials and may not fully reflect daily clinical care. When effective hemostasis was achieved (in two-thirds) in bleeding patients, this was associated with a lower mortality risk as compared with ineffective hemostasis.