Restart TICrH: An Adaptive Randomized Trial of Time Intervals to Restart Direct Oral Anticoagulants after Traumatic Intracranial Hemorrhage

Abstract

Anticoagulants prevent thrombosis and death in patients with atrial fibrillation and venous thromboembolism (VTE) but also increase bleeding risk. The benefit/risk ratio favors anticoagulation in most of these patients. However, some will have a bleeding complication, such as the common trip-and-fall brain injury in elderly patients that results in traumatic intracranial hemorrhage. Clinicians must then make the difficult decision about when to restart the anticoagulant. Restarting too early risks making the bleeding worse. Restarting too late risks thrombotic events such as ischemic stroke and VTE, the indications for anticoagulation in the first place. There are more data on restarting patients with spontaneous intracranial hemorrhage, which is very different than traumatic intracranial hemorrhage. Spontaneous intracranial hemorrhage increases the risk of rebleeding because intrinsic vascular changes are widespread and irreversible. In contrast, traumatic cases are caused by a blow to the head, usually an isolated event portending less future risk. Clinicians generally agree that anticoagulation should be restarted but disagree about when. This uncertainty leads to long restart delays causing a large, potentially preventable burden of strokes and VTE, which has been unaddressed because of the absence of high quality evidence. Restart Traumatic Intracranial Hemorrhage (the "r" distinguished intracranial from intracerebral) (TICrH) is a prospective randomized open label blinded end-point response-adaptive clinical trial that will evaluate the impact of delays to restarting direct oral anticoagulation (1, 2, or 4 weeks) on the composite of thrombotic events and bleeding in patients presenting after traumatic intracranial hemorrhage.

Keywords: adult brain injury; clinical management of central nervous system injury; clinical trial; head trauma; traumatic brain injury.

Conflict of interest statement

No competing financial interests exist.

Figures

FIG. 1.

Restart Traumatic Intracranial Hemorrhage (the “r” distinguished intracranial from intracerebral) (TICrH) is a prospective randomized open label blinded end-point (PROBE) response adaptive clinical trial of different time intervals to restart anticoagulation after traumatic intracranial hemorrhage (1). Arm restart days (7, 14, and 28) are illustrated here by the change from blue to red in the horizontal bars, a transition from thrombotic risk to bleeding risk (2), which are the opposing risks combined in the primary composite end-point. The composite end-point of major bleeding and thrombotic events (3) is illustrated by the red and blue vertical bars, which reflect the proportions of bleeding and thrombosis and predicted combined rate, along with the expected rates used for trial power in the next column. It is the differences in these composite end-points by arm that drives randomization adaptation (4). The algorithm decreases the probability of randomization to underperforming arms; that is, those with higher event rates. A second adaptative mechanism for the inclusion criteria is driven by total end-point event rate in all arms (and therefore does not require unblinding) (5). If total event rates are low, the trial will have difficulty detecting a difference between arms; therefore, the algorithm increases the subject end-point risk by increasing the CHA2DS2-VASc score inclusion criteria, enriching the sample for events. This makes detecting a difference between groups easier and prevents the trial from stopping for futility. The PROBE component of the trial is achieved with central blinded assessment of subjects for potential end-point events (by videoconference) with a blinded expert study end-point adjudication board determination of whether a valid end-point event has occurred.

FIG. 2.

Trends of new prescriptions for anticoagulants 2011–2017.

FIG. 3.

This figure is adapted from Xu and coworkers, a global survey of experts across multiple specialties for restarting anticoagulation recommendations across 11 scenarios. For TICrH, 98% agreed that these patients should be restarted, but the timing of restart varied widely.

FIG. 4.

The trial end-point of 60 days was chosen because all arm event rate curves are expected to become parallel 30 days after the last arm is anticoagulated. This is the known time that it takes for most clots to be replaced with connective tissue and resorbed into the vessel wall, and for the wall to re-endothelialize. Thrombi that developed before anticoagulation are still at risk of embolization during this period. We do not expect any change in differences in event rates between arms after that period.

FIG. 5.

Secondary analysis removes deep vein thrombosis (DVT) from the composite, as it has lower mortality than the other elements.

FIG. 6.

Secondary analysis evaluates if the bleeding and thrombotic components are equal or if one actually causes more disability than the other.

FIG. 7.

Trauma Quality Improvement Program (TQIP) closed loop translation with enforcement mechanisms already in place ensures that trial findings are implemented in the larger trauma network and that reductions in thrombotic events and attendant healthcare costs, morbidity, and mortality are realized. For a trial with a total cost, direct and indirect, of $12,000,000, the 1st year return on investment is 1000 thrombotic events prevented at 850 trauma centers at an acute care cost of $35,000,000.