Antithrombotic treatment of cervical arterial dissection: Big data needed to inform long-term management

Cervical artery dissection (CAD) is a leading cause of stroke in young people and a significant concern in all age groups. ¹ The recently published long-term follow-up data of the TREAT-CAD trial of antithrombotic treatment for symptomatic CAD (with or without stroke) showed that between 3 and 6 months from the detection of CAD, among the 122 participants included in the trial, only four outcome events occurred that were all hemorrhagic and asymptomatic. ² This trial randomized patients to a 300 mg daily dose of acetylsalicylic acid or vitamin K antagonists – warfarin, acenocoumarol, or phenprocoumon, at the discretion of the treating physician – with a target INR value of 2.0–3.0. Notably, direct oral anticoagulants were not tested.

It could be surprising to find that a study with only four outcome events can impact our understanding of the treatment of CAD. Nevertheless, it is the first trial to provide 6-month follow-up data on those patients. The study showed that the long-term ischemic risk is very low in patients with CAD and should be compared with the hemorrhagic risk of antithrombotic drugs, particularly anticoagulants. On the other hand, the TREAT-CAD did not have a placebo group and consequently we cannot conclude that antithrombotic medications are ineffective in the long term.

The two available trials of antithrombotic treatment for CAD – TREAT-CAD and CADISS – do not solve all the problems related to this clinical issue. First, CAD can present in different arteries and arterial traits – either intra- or extracranial – which need to be considered for therapeutic decisions. Second, the outcome of CAD can vary according to vessel recanalization status after the event. Third, patients might have different clinical presentations – either with or without ischemic stroke – and different vascular risk profiles. Fourth, CAD can be treated with many antithrombotic strategies, including single antiplatelets, dual antipletelets, vitamin K antagonists, direct oral anticoagulants, and parenteral anticoagulants, that would all need a robust assessment; the future use of Factor XI inhibitors might add a further treatment option. Fifth, a 6-month follow-up cannot be properly considered “long-term” for young populations such as those with CAD. Given the relative rarity of CAD, it is difficult to set up conventional trials, and even guidelines mostly rely on non-randomized evidence. ³ Robust trials on antithrombotic treatments for CAD would need to be conducted worldwide, with very large sample sizes. Observational data can provide evidence about the treatment of CAD only if data are well collected, controlled for accuracy, and correct for potential sources of bias. To date, the CADISP collaboration is the only example of wide collaboration on patients with CAD. ⁴ The creation of big datasets from electronic medical records, digital twins, and virtual patients might expand data collection and allow unprecedented inferences on the use of antithrombotic medication for patients with CAD.