Some 21 randomized trials involving 19 279 patients have evaluated the roles of carotid endarterectomy (CEA), best medical therapy (BMT) and carotid artery stenting (CAS) in patients with symptomatic and asymptomatic carotid disease. Before 2011, the data were generally interpreted as CEA being preferable in ‘average-risk’ symptomatic patients with 50–99 per cent stenoses and in ‘average-risk’ asymptomatic patients with 60–99 per cent stenoses. CAS was preferred in symptomatic patients deemed ‘high risk’ for CEA, and BMT thought safer in asymptomatic patients considered ‘high risk’ for CEA. This consensus was, however, somewhat fragile as there was an uneasy relationship between surgeons who wanted guidelines to remain unchanged, radiological interventionists who argued that the benefits of CAS were being withheld, and purchasers of healthcare, wary of changes in practice being driven by the ‘medical industrial complex’ rather than evidence. Liberalizing CAS indications might also have financial implications for surgeons and interventionists in privately funded healthcare systems. This consensus effectively collapsed following the 2011 US Food and Drug Administration approval of the RX Acculink Carotid Stent system (Abbott Laboratories, Abbott Park, Illinois, USA) in average-risk patients1. Along with 2011 guidelines from the American Heart Association (AHA)2,3, CAS was declared comparable with CEA in the treatment of average-risk symptomatic and asymptomatic patients. Other countries have advocated a more restricted expansion of CAS. Guidelines from Australia andNewZealand concluded that, although CAS was appropriate in symptomatic patients aged less than 70 years and in ‘high-risk for CEA’ symptomatic patients, they did not recommend CAS in average-risk, asymptomatic patients4. The 2011 National Institute for Health and Clinical Excellence guidelines advised that the evidence supported CAS in recently symptomatic patients (provided that arrangements were made for clinical governance, audit and research), but that CAS should be used only in asymptomatic patients in randomized trials or by ‘special arrangement’5,6. Given past experience, however, it seems likely that the rest of the world will adopt guidelines similar to those of the AHA. Those familiar with the literature will be well versed in what the randomized trials have told us, but many may be less familiar with what they have not told us. Current evidence offers ten key observations7–10: