Three randomised controlled trials (RCTs) comparing outcomes after carotid artery stenting (CAS) with carotid endarterectomy (CEA) have recently been published. Recent systematic reviews have recommended that CAS is no longer justified for patients suitable for CEA. Indeed, in many centres, pooled data of RCTs show higher peri-operative risk of performing CAS vs. CEA with comparable long-term efficacy. Due to limitations in SPACE, EVA3S and ICSS study design and conduct, the inferiority of CAS to CEA as a method remains inconclusive. The goal of this review is not to discredit these trials but to develop a more differentiated and critical interpretation of the data and to create more discussion. It will discuss the necessity of RCTs for Interventional Neuroradiology in general and particular problems in study design (non-inferiority design and interpretation of results, clinical equipoise, study endpoints), practical study conduct difficulties (operator and centre experience, antiaggregation, timing of treatment) and the interpretation of the results (relation of internal and external validity, procedural complexity, the 68-year surprise, longer-term outcome). A premature rejection of CAS based on the data from these studies could harm future patients who would have had benefited from this procedure. For the time being, there is no reason why centres with good and independently controlled track records should stop performing CAS. Designing a single cooperative European trial that incorporates the lessons learned would be major step forward.
Three randomised controlled trials (RCTs) comparing outcomes after carotid artery stenting (CAS) with carotid endarterectomy (CEA) have recently been published. Recent systematic reviews have recommended that CAS is no longer justified for patients suitable for CEA. Indeed, in many centres, pooled data of RCTs show higher peri-operative risk of performing CAS vs. CEA with comparable long-term efficacy. Due to limitations in SPACE, EVA3S and ICSS study design and conduct, the inferiority of CAS to CEA as a method remains inconclusive. The goal of this review is not to discredit these trials but to develop a more differentiated and critical interpretation of the data and to create more discussion. It will discuss the necessity of RCTs for Interventional Neuroradiology in general and particular problems in study design (non-inferiority design and interpretation of results, clinical equipoise, study endpoints), practical study conduct difficulties (operator and centre experience, antiaggregation, timing of treatment) and the interpretation of the results (relation of internal and external validity, procedural complexity, the 68-year surprise, longer-term outcome). A premature rejection of CAS based on the data from these studies could harm future patients who would have had benefited from this procedure. For the time being, there is no reason why centres with good and independently controlled track records should stop performing CAS. Designing a single cooperative European trial that incorporates the lessons learned would be major step forward.