ESOC 2026 | 24-month results of EXCOA-CVT: extending oral anticoagulation treatment after CVT

The EXCOA-CVT trial was a cluster randomized trial dedicated to the question of what is the duration of anticoagulation after cerebral venous thrombosis, what should be in patients that have no indication for permanent anticoagulation. So, of course, patients that have recurrent events, high thrombophilic risk, high risk of recurrence, as understood by the physicians, they will receive long-term anticoagulation. But then for all the patients for which we have transient causes or causes that are not associated with this very high risk, which are actually the majority of the patients, we had no evidence on what would be the best approach in terms of duration of anticoagulation. Basically, it was based, and in the previous European guidelines and American guidelines, this was based on the evidence from other types of venous thrombosis. But we know that the recurrence rates and the pattern of the risk profile of patients with CVT is different from other types of venous thrombosis. So there was really a lack of evidence in this field. This is the first interventional study dedicated to this. And this study also had a parallel study, which was a prospective observational study exactly with the same outcomes, with the same design, except for the fact that it was not randomized in terms of clusters. But still, hospitals in the prospective study would decide whether they would prefer a policy of a short-term or a long-term exactly as in the trial.

So overall, we have a large cohort of patients that were in centers that were either doing a short-term policy defined as three to six months or a long-term policy defined as 12 months and then a second year of follow-up after this in which patients were expected not to receive anticoagulation. So overall and the cluster randomized trial, which is of course the higher quality evidence from in terms of design, we could at the 24 months see a slight divergence of the risk in terms of venous thrombotic events, a bit numerically higher in the patients on the short-term policy, but still there was no significant difference between the groups, and overall the number of events was lower than expected.

Still, in the subgroup analysis, we find some interesting results. There was a predefined subgroup analysis for some specific groups that observational evidence, previous observational evidence suggested could be interfering or modulating the risk in these patients. And indeed, we could see that sex was there was a significant interaction in terms of the benefit of long-term anticoagulation. So men had a higher risk of thrombosis in the short-term arm and therefore there was an interaction with sex with male having more benefit from long-term anticoagulation, but of course this is a subgroup analysis and we will of course look into that in future studies. But still overall the conclusion is that there was no difference no statistically significant difference between the groups in terms of short or long-term policy.

Also, in terms of safety, there was no difference in terms of bleeding events, but still there was higher mortality rates numerically in the cluster randomized study and significantly higher in the observational study for patients on the long-term policy. So overall, this suggests that a standardized policy of prolonging anticoagulation up to 12 months in patients with CVT is not evidence-based. And so our practice currently is mostly to discuss individually by case by case and overall preferring a policy that is more towards the six months, which was the short term policy in this study. So these are the conclusions and we were happy to be able to also follow these patients for this additional period so that we could understand whether there was a divergence of the risk after this period, because we know the thrombotic risk is often delayed, is not immediate after stopping anticoagulation. And also we could see that we were just postponing the risk if we were doing a bit longer policy of anticoagulation, but that was not the case. So indeed we found no difference in the risk in these patients.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.