Results from the Dutch CAVA trial were published recently in The Lancet Haematology and, at first glance, this is another trial that failed to demonstrate any meaningful benefit for early thrombus removal in patients with acute iliofemoral deep vein thrombosis (DVT).
In CAVA, 184 participants with iliofemoral DVT were randomised to ultrasound-accelerated catheter-directed thrombolysis (CDT) or standard care alone. One-year rates of post-thrombotic syndrome (PTS) were 29% and 35% in the intervention and standard care groups respectively, a non-significant difference.
The researchers should be commended for their efforts in trying to answer a question of enormous clinical relevance and for their perseverance in completing a trial that took over seven years to recruit. However, there are important methodological flaws to raise. Twenty-two participants (12%) withdrew consent within two days of randomisation, 30% of thrombolysis interventions were stopped early, and only 45% (35/77) of participants starting thrombolysis underwent venous stenting, with only half of patients considered to have had successfully delivered therapy. Follow-up was limited to one year and the sample size was relatively small, meaning that CAVA was underpowered to detect anything other than a large difference.
The generalisability of CAVA will be questioned, as with ATTRACT and CaVenT. Patients enrolled into these studies only constitute a small minority of the overall population presenting with iliofemoral DVT, illustrated by the fact that less than 2.5% of the screened population in ATTRACT were randomised. Inevitably, many patients with severe symptoms will not have been included due to the lack of clinician (and/or patient) equipoise. Perhaps the bigger question relates to the technical success. In all three randomised controlled trials (RCTs), rates of technical success and venous stenting were relatively low, which could explain poor outcomes in the intervention arms of these trials. Conversely, the technical outcomes in these multicentre studies could be seen as more reflective of practice, rather than selected centres of excellence. The relationship between technical success of early thrombus removal (and durable deep venous patency) and clinical outcomes does warrant further investigation, and debate on the open vein hypothesis continues as a result.
Even with the reduced rate of postthrombotic syndrome after CDT in the CaVenT trial, and the reduced rate of moderate-severe PTS in the iliofemoral DVT subgroup in the ATTRACT trial, these benefits were underwhelming and did not match the advantages perceived by the many enthusiastic advocates of these interventions. The results of the CAVA trial compound this situation. The apparent discord between published evidence and the experience of large swathes of the clinical community creates significant challenges, particularly as the next iterations of international guidelines are likely to reflect results from published randomised trials, and are therefore unlikely to recommend widespread early thrombus removal, which is common practice in many centres.
Despite another large RCT hoping to clarify the issue, the role of early thrombus removal in the management of acute iliofemoral DVT remains as controversial as ever. Worryingly, there seem to be increasingly partisan views for and against intervention. In reality, both perspectives have merit.
Unwavering advocates of aggressive early thrombus removal should recognise that many patients with iliofemoral DVT treated with anticoagulation alone do not develop PTS (around 50% or more in CaVenT, ATTRACT and CAVA). Likewise, those sceptical of early thrombus removal strategies should accept that many patients with severe symptoms (most likely to benefit from thrombolysis) were not included, and superior technical and clinical outcomes from thrombus removal—than those reported in CaVenT, ATTRACT and CAVA—are feasible. Both sides can also probably agree that the outcome measures in RCTs remain suboptimal.
There are exciting technological advances for thrombus removal, with a range of pharmacomechanical thrombolysis and thrombectomy options widely available. Whether these advances can deliver better technical success and superior clinical outcomes remains to be seen. As with many areas of venous disease, the biggest challenge remains case selection. More work is needed to define specific subgroups of patients that may benefit most from aggressive thrombus removal. Until we make progress with better patient selection and clearly good technical results though, this controversy is likely to persist.
Manj Gohel is a consultant vascular and endovascular surgeon at Cambridge University Hospitals, Cambridge, UK.
Stephen Black is a consultant vascular surgeon at Guy’s and St Thomas’ NHS Foundation Trust, London, UK.