Background: Carotid endarterectomy (CEA) reduces the risk of stroke in patients with internal carotid stenosis of 50-99 per cent. This study assessed national surgical practice through audit of CEA procedures and outcomes. Methods: This was a prospective cohort study of UK surgeons performing CEA, using clinical audit data collected continuously and reported in two rounds, covering operations from December 2005 to December 2007, and January 2008 to September 2009. Results: Some 352 (92·6 per cent) of 380 eligible surgeons contributed data. Of 19 935 CEAs recorded by Hospital Episode Statistics, 12 496 (62·7 per cent) were submitted to the audit. A total of 10 452 operations (83·6 per cent) were performed for symptomatic carotid stenosis; among these patients, the presenting symptoms were transient ischaemic attack in 4507 (43·1 per cent), stroke in 3572 (34·2 per cent) and amaurosis fugax in 1965 (18·8 per cent). The 30-day mortality rate was 1·0 per cent (48 of 4944) in round 1 and 0·8 per cent (50 of 6151) in round 2; the most common cause of death was stroke, followed by myocardial infarction. The rate of death or stroke within 30 days of surgery was 2·5 per cent (124 of 4918) in round 1 and 1·8 per cent (112 of 6135) in round 2. Conclusion: CEA is performed less commonly in the UK than in other European countries and probably remains underutilized in the prevention of stroke. Increasing the number of CEAs done in the UK, together with reducing surgical waiting times, could prevent more strokes.