Optimization of the perioperative management of direct oral anticoagulants

Promoter

Prof. Jean-Michel DOGNE, UNamur, Department of Pharmacy

Jury

Profs C-M. Samama (Université Paris Descartes), B. Ickx (ULB), B. Chatelain (UCL), N. Caron (présidente, UNamur), M. Gourdin (co-promoteur, UCL), J-M. Dogné (Promoteur, UNamur)

Summary

Direct oral anticoagulants (DOACs) (i.e.  dabigatran,  rivaroxaban,  apixaban and  edoxaban) are used for the prevention and treatment of thrombotic events including the treatment and prevention   of recurrent venous thromboembolism (VTE) and stroke prevention in patients with non-valvular atrial fibrillation(NVAF). Annually, approximately 10-15 % of these patients will require DOAC interruption before an elective or invasive procedure. Laboratory testing is not recommended routinely, as these drugs have a rapid onset and offset of action, a short half-life, predictable pharmacokinetics and a large on-therapy range. However, many frail patients (i.e. low body weight, several drug interferences) who receive DOACs in real-life were initially excluded from the clinical phase III trials. In addition, a link between dabigatran or edoxaban concentration and clinical events (venous  thromboembolism or major bleeding) was demonstrated. Multicenter studies have shown a high interindividual variability with DOACs in real-life patients that cannot account only for the renal function. Furthermore, there is no agreement between the different proposals that aimed at guiding the best timing to surgery. Therefore, DOAC monitoring has become progressively a part of the perioperative management of DOACs, especially in emergencies.

Initial perioperative proposals suggested that routine coagulation assays like the activated partial thromboplastin time (aPTT) or the prothrombin time (PT) could assess residual plasma concentrations of DOACs. Nevertheless, these assays give only qualitative informations, as their responsiveness is coagulometer and reagent dependent and as they are not sensitive enough to exclude clinically relevant DOAC concentrations. Thrombin time (TT) is the most sensitive coagulation test for dabigatran. However, its methodology is not standardized between the laboratories and its performance is reagent dependent. Therefore, its use has been restricted to exclude clinically relevant concentration of dabigatran in a perioperative context. Specific coagulation assays were    proposed to allow quantitative measurements of DOACs. These include the diluted TT(dTT), the ecarin clotting time (ECT), or the ecarin chromogenic assay (ECA) for dabigatran, and chromogenic anti-FXa assays for rivaroxaban, apixaban, and edoxaban. They have been largely validated in several publications outside the perioperative context. These tests showed a limit of quantitation with standard methodology and calibrators around 30 and 50 ng/ml, which was not accurate enough for the perioperative context. Indeed, the safety threshold proposed for DOAC plasma concentrations before an invasive procedure carrying a high bleeding risk was set at 30ng/ml. Therefore, the primary objectives of this work were:

• To optimize and validate   the use of thrombin time in the perioperative management of patients on dabigatran etexilate.
• To compare the performance of coagulation tests, specifically developed for the measurement of low plasma concentrations of dabigatran and rivaroxaban with the reference LC-MS/MS method.
• To study the interference of heparin bridging on the performance of these specific coagulation assays. Our in vitro results  demonstrated  the influence  of  thrombin  concentration, thrombin  origin  and  the  typeof  coagulometer  on  the  measurement  of thrombin. To use thrombin time in a perioperative context, we optimized the thrombin concentration on two coagulometers. In addition, we illustrated the lack of stability of plasma samples and suggested measuring TT within 2 hours of sampling in plasma containing dabigatran. We confirmed  in  an ex  vivo pilot  study  performed  on  24  plasma  samples from patients treated with dabigatran etexilate, that an optimized TT may be useful  in assessing low dabigatran concentrations (<  50  ng/ml),  especially for  laboratories  that  do  not  have  access  to  specific  assays. However, clinicians need to consider the different variables affecting TT (i.e. heparin bridging, inflammatory syndrome, fibrin/fibrinogen degradation products).

In the second ex vivo study performed on 33 plasma samples from patients treated with dabigatran  etexilate,  we  showed  that  two  coagulation  tests specifically  developed  for the measurements  of low dabigatran plasma concentrations (the Hemoclot Thrombin Inhibitors®LOW (HTI LOW) from Hyphen  BioMed®,  Neuville-sur-Oise,  France  and  the  STA®-ECA  II  (ECA-II) from Diagnostica  Stago®,  Asnière-sur-Seine,  France),  performed  well with  dabigatran plasma concentrations  <50  ng/ml, and  thus  should  be preferred to the standard procedure of HTI in this range of concentrations. When  TT  is  available  before  a dabigatran  level is  requested,  then  it  can guide  the  laboratory in  choosing  the more  accurate coagulation tests  (i.e. HTI  or  HTI  LOW/ECA-II)  in  order  to  avoid  unnecessary  costs.  The added value of this strategy was illustrated in a case-series of patients monitored in a perioperative context. In  the  last ex  vivo study  performed on 79 patients,  we demonstrated  that specific procedures with adapted methodology and calibrators (i.e. Biophen®DiXaI  LOW  and  STA®-Liquid  Anti-Xa) should  be  preferred  for the estimation of rivaroxaban concentrations below 50 ng/ml. These methods are sensitive to low molecular weight heparins (LMWHs) and may consequently be influenced with residual LMWH found even 24 h after the last administration.

The   development of a chromogenic anti-Xa assay accurate for low rivaroxaban concentrations and insensitive to heparins is required to optimize the perioperative management of patients at high risk of thrombosis that receive a bridging therapy with heparin before an elective procedure. In conclusion, DOACs monitoring may be necessary to manage some patients in a perioperative or emergent context. Accurate specific assays with an adapted methodology and calibrators for low DOACs concentrations may guide the clinicians for the optimal management of these patients. Despite attractive pharmacokinetic properties, the high interindividual variability of DOACs plasma concentration support that their pre-procedural interruption should not be based only on their  respective half-life, but also on residual drug concentrations. Further perioperative studies using DOAC monitoring with accurate laboratory tests are needed to validate a unique and safe peri-procedural management.