|| Checking for direct PDF access through Ovid
Introduction: The long-term cerebrovascular consequences of hypertensive encephalopathy (HE) are poorly understood. Therefore, we aimed to measure the risk of stroke following HE.Methods: We identified all adult patients discharged from nonfederal acute care hospitals between 2005 and 2013 in New York with a primary ICD-9-CM discharge diagnosis of HE (437.2). Only patients who underwent magnetic resonance imaging were included to reduce the likelihood of misclassification error. Patients with all other forms of hypertension (401-405), without concomitant codes for HE or cerebrovascular disease (430-438), served as controls. The primary outcome was incident stroke (431, 433.x1, 434.x1, or 436 in the absence of hemorrhage, trauma, or rehabilitation codes). Kaplan-Meier survival analysis was used to calculate cumulative rates of incident stroke and Cox proportional hazards analysis was used to determine the association between HE and incident stroke while adjusting for demographics, vascular risk factors, and the Elixhauser comorbidity index.Results: We identified 1,386 patients with HE and 2,869,873 with hypertension. Over a mean follow-up period of 3.3 (+/-1.8) years, we identified 66,594 ischemic and 12,343 hemorrhagic strokes. After 5 years, the cumulative rate of stroke was 7.8% (95% CI, 6.2-9.9%) in patients with HE and 3.2% (95% CI, 3.2-3.2%; P<0.001 for the log-rank test) in patients with any other hypertensive disease. After adjusting for demographics, vascular risk factors, and the Elixhauser comorbidity index, HE was independently associated with incident stroke (hazard ratio, 1.9; 95% CI, 1.5-2.4) as compared to controls. This association was similar when considering ischemic and hemorrhagic stroke separately.Conclusions: Patients discharged after HE face a higher long-term risk of subsequent stroke than patients without prior neurological complications of hypertension.