Studies in the 1990s demonstrated that misery perfusion is a predictor of subsequent stroke in medically treated patients with symptomatic major cerebral artery disease. A recent randomized controlled trial demonstrated no benefit of bypass surgery for such patients. In this light, outcome in patients with misery perfusion has regained interest. The purpose of this study was to determine whether misery perfusion is still a predictor of subsequent stroke despite recent improvements in medical treatment for secondary prevention of stroke, and if so, whether the predictive value of misery perfusion has changed in recent years. We prospectively studied 165 non-disabled patients with symptomatic atherosclerotic internal carotid artery or middle cerebral artery occlusive diseases who underwent positron emission tomography from 1999 to 2008. Misery perfusion was defined as decreased cerebral blood flow, increased oxygen extraction fraction and decreased ratio of cerebral blood flow to blood volume in the hemisphere supplied by the diseased artery. All patients were followed up for 2 years until stroke recurrence or death. Bypass surgery was performed in 19 of 35 patients with and 16 of 130 patients without misery perfusion. The 2-year incidence of ipsilateral ischaemic stroke was six and four patients with and without misery perfusion, including two and one after surgery, respectively (P < 0.002). Total strokes occurred in nine patients with misery perfusion and 12 patients without (P < 0.01). The relative risk conferred by misery perfusion in whole sample was 6.3 (95% confidence interval 1.7-22.4, P < 0.005) for ipsilateral ischaemic stroke and 3.5 (95% confidence interval 1.4-8.9, P < 0.01) for all strokes, while the respective values in medically treated patients were 12.6 (95% confidence interval 2.7-57.8, P < 0.005) and 4.7 (95% confidence interval 1.3-16.3, P < 0.02). The all-stroke incidence in patients entering the study from 2004 to 2008 (4/72) was significantly lower than in those entering from 1999 to 2003 (17/93; P < 0.02), although the prevalence of misery perfusion or bypass surgery did not differ. Between these periods, patients without misery perfusion demonstrated a decrease in stroke rate (from 16.2% to 0%), but patients with misery perfusion did not (26.3 and 25.0%). In symptomatic major cerebral artery disease, misery perfusion remains a predictor of subsequent stroke, although the recurrence rate was lower than the previous study. In patients without misery perfusion, the risk of stroke was reduced over time. Thus, identification and stricter management of patients with misery perfusion are essential to further improve prognosis.