Introduction Intracranial stenosis (ICS) is believed to convey a high risk of recurrent ischaemic stroke, but the SAMMPRIS trial failed to show superiority of ICS stenting over intensive medical management. However, the risk of recurrent stroke in the medical arm was low, leading to questions about generalisability. We studied the long-term risk of recurrent stroke in patients with ICS in a population-based setting.

Methods Patients with TIA or minor stroke [NIHSS £3] underwent MRA in the Oxford Vascular Study. Stroke risk was determined by face-to-face follow-up out to 7-years with a policy of intensive medical treatment and no stenting.

Results Of 1026 patients (686-TIA; 340-stroke) who underwent MRA, 210 (20.5%) had >50% ICS (TIA=18.5%; stroke=24.4%). Compared to patients with no ICS, risk of recurrent ischaemic stroke was not increased in those with recently symptomatic ICS (HR=1.36, 95%CI=0.54–3.44) or in those with any ICS (1.28, 0.70–2.35) Moreover, the absolute risk of stroke in those with recently symptomatic ICS was lower than reported in the medical arm of SAMMPRIS (2-year risks: 1.6%, 0–4.7 vs 14.1%, 10.1–19.4)

Conclusion The ‘real-world’ prognosis of patients with ICS on intensive medical treatment is more benign than supposed, indicating that the findings of SAMMPRIS are generalisable.