How I interpreted the randomised trials of carotid angioplasty/stenting versus endarterectomy

Carotid endarterectomy (CEA) for carotid stenosis is effective in preventing ipsilateral carotid territory ischaemic stroke. Paradoxically however, it causes a stroke (the event it is trying to prevent) in about 5% or more of cases. If carotid angioplasty/stenting (CAS) is to have a place in the management of patients with carotid stenosis (beyond those who are not suitable for CEA), it has to demonstrate that it is also effective and safe. Limited data from 12 randomised trials comparing CAS with CEA (the current "gold standard") in a total of 3227 patients with carotid stenosis (90% symptomatic) question the safety of CAS and suggest that it may cause more non-fatal, procedural strokes than CEA despite similar mortality rates and a much lower immediate local complication rate (eg cranial neuropathy). However, the published trials are rather heterogeneous (clinically and methodologically), none is large enough to provide robust and convincing data and long-term follow-up is very limited. Accordingly, it remains unknown whether CAS is effective in preventing recurrent stroke among patients with carotid stenosis, or whether it is safe. More data (from at least another 3,000 patients) are needed from the ongoing randomised trials before it can reliably be concluded whether CAS is inferior to, non-inferior to, or more effective than, CEA. More importantly, it will be possible to determine which patients should be treated preferentially with CAS, which patients with CEA, and which patients should not undergo either revascularisation procedure.