109 FRAIL-AF RANDOMISED CONTROLLED TRIAL: RESULTS particular treatment strategy (switch to a NOAC or continue with a VKA) was clearly indicated or clearly contraindicated in terms of a net difference in the primary outcome (i.e. a difference of at least 3 standard deviations; P value around 0.002). Following observations in the trial, an interim analysis was planned after having observed at least 160 primary outcome events, at which time point the DSMB could advise the trial steering committee to halt the trial for futility if, at that stage, the hazard ratio (HR) for the primary outcome of the intervention arm versus the control arm exceeded 0.9925. The last author (GJG) vouches for the accuracy and completeness of the data and for the fidelity of the trial to the protocol. Patients To be eligible, patients needed to meet all the following criteria: age ≥75 years; currently managed on INR guided VKA treatment for AF by one of the eight participating Dutch thrombosis services; a Groningen Frailty Indicator (GFI) ≥3; and willingness to switch from VKA management to a NOAC based treatment strategy. The GFI is a validated questionnaire that assesses frailty from a functional perspective on several domains (see Supplementary file S1).13 A potential subject who met any of the following criteria was excluded from randomisation: valvular AF (this is AF in the presence of a mechanical heart valve or severe mitral valve stenosis); an estimated glomerular filtration rate (eGFR) <30 mL/min/1.73 m2; taking part in another medical scientific research program; and unwilling or unable to provide written informed consent. Randomisation, procedures, and follow-up Patients were randomly assigned to either the intervention arm (i.e. switch to a NOAC based treatment strategy: stop the VKA and start a NOAC if the INR is <1.3), or to the control arm (i.e. continue with INR guided VKA management: either 1 mg acenocoumarol or 3 mg phenprocoumon with targeting INR levels between 2.0 and 3.0). Computerised block randomisation was used, stratified by thrombosis service and renal function at baseline (with two strata: an eGFR of 30–50 mL/min/1.73 m2; an eGFR ≥50 mL/min/1.73 m2). Patients initially randomly assigned to a NOAC based treatment strategy started NOAC therapy when the INR was <2.0 after stopping VKA therapy. However, shortly after the trial was initiated, the DSMB observed a tendency of more bleeding during the switching period. As a result, in July 2019, after having included 102 patients in the intervention arm, an INR level <1.3 was used to prevent too high anticoagulation during the switching period. The decision on the type of NOAC was at the discretion of the treating physician, if needed, in collaboration with the study team. The study team had no preference for one NOAC or the other. Yet, when asked to help making a NOAC choice, they aimed 7
RkJQdWJsaXNoZXIy MTk4NDMw