Objective: To describe the severity and early neurological deterioration (END) in patients with symptomatic carotid stenosis and to analyse the influence of related factors. Methods: Observational cohort study of patients with ischaemic stroke, ipsilateral carotid stenosis and without evidence of cardiac sources of embolism prospectively recorded since January 2003 to January 2012. Initial severity was categorised as mild (NIH stroke scale (NIHSS) ≤7), moderate (NIHSS 8-14) or high (NIHSS >14). Logistic ordinal and regression analyses were performed for stroke severity and END risk. Results: Of 2332 ischaemic strokes attended, 338 patients were included. Stroke severity was mild in 254 (75.1%) cases, moderate in 53 (15.7%) and severe in 31 (9.2%). Adjusted ORs (95% CI) for stroke severity were: degree of carotid stenosis, 2.20 (1.55 to 3.11, p<0.001); intracranial disease, 1.93 (1.18 to 3.17, p=0.009); plasma glucose, 1.01 (1.003 to 1.02, p<0.001); and previous transient ischaemic attack (TIA), 0.37 (0.17 to 0.82, p=0.014). 78 patients (23.1%) had END. Multivariate analysis showed independent association between END and degree of carotid stenosis (OR 1.64, 1.14 to 2.34, p=0.007), previous TIA (OR 2.40, 1.25 to 4.57, p=0.008) and mean arterial pressure (OR 1.02, 1.01 to 1.04, p=0.003). Conclusions: Strokes due to large vessel disease in the carotid artery are in general of mild severity and have a high rate of END. The degree of stenosis has a clear association with higher severity and END risk.