Background: Endovascular treatment by transluminal balloon angioplasty or stent insertion may be a useful alternative to carotid endarterectomy.
Objectives: To assess the benefits and risks of endovascular treatment compared with carotid endarterectomy or medical therapy.
Search strategy: We searched the Cochrane Stroke Group trials register (last searched 14 March 2007) and the following bibliographic databases: Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, Issue 1, 2007), MEDLINE (1950 to March 2007), EMBASE (1980 to March 2007) and Science Citation Index (1945 to March 2007). We also contacted researchers in the field.
Selection criteria: We selected randomised trials of endovascular treatment compared with endarterectomy or medical therapy for carotid artery stenosis.
Data collection and analysis: One review author independently applied the inclusion criteria, extracted data and assessed trial quality. Search results were validated by a second review author.
Main results: Data were available from 12 trials (3227 patients) but not all contributed to each analysis. The primary outcome comparison of any stroke or death within 30 days of treatment favoured surgery (odds ratio (OR) 1.39, P = 0.02, not significant (NS) in the random-effects model). The following outcome comparisons favoured endovascular treatment over surgery: cranial neuropathy (OR 0.07, P < 0.01); 30 day neurological complication or death (OR 0.62, P = 0.004, NS in the random-effects model, with significant heterogeneity). The following outcome comparisons showed little difference between endovascular treatment and surgery: 30 day stroke, myocardial infarction or death (OR 1.11, P = 0.57 with significant heterogeneity); stroke during long-term follow up (OR 1.00). Comparison between endovascular treatment with or without protection device showed no significant difference in 30 day stroke or death (OR 0.77, P = 0.42 with significant heterogeneity). Analysis of stroke or death within 30 days of the procedure in asymptomatic carotid stenosis showed no difference (OR 1.06, P = 0.96). In patients not suitable for surgery, there was no significant difference in 30 day stroke or death (OR 0.39, P = 0.09 with significant heterogeneity).
Authors' conclusions: The data are difficult to interpret because the trials are heterogeneous (different patients, endovascular procedures, and duration of follow up) and five trials were stopped early, perhaps leading to an over-estimate of the risks of endovascular treatment. The pattern of effects on different outcomes does not support a change in clinical practice away from recommending carotid endarterectomy as the treatment of choice for suitable carotid artery stenosis.