Magnetic Resonance Imaging White Matter Hyperintensity as a Predictor of Stroke Recurrence in Patients with Embolic Stroke of Undetermined Source

Background and Purpose

This study aimed to investigate factors associated with stroke recurrence and new-onset atrial fibrillation (AF) in patients with embolic stroke of undetermined source (ESUS).

Methods

Between January 2005 and March 2012, a total of 1514 consecutive patients were admitted to our hospital with a diagnosis of acute ischemic stroke within 7 days after onset. Of these, 236 patients met the diagnostic criteria for ESUS. A retrospective analysis was performed of the following characteristics of these ESUS patients: medical history, comorbidities, and magnetic resonance imaging (MRI) findings including the deep and subcortical white matter hyperintensity (DSWMH) grade, drugs used for preventing recurrence, and clinical information such as NIH Stroke Scale scores on admission and outcomes evaluated by modified Rankin scale. The rate of recurrence after discharge and the frequency of new-onset AF were also determined.

Results

There were 236 ESUS patients (141 men, mean age 70.2 ± 12.1 years). During the follow-up period, ranging from 7 days to 12.9 years (median 54.3 months), 32 (13.6%) and 44 (18.6%) of these patients had a recurrent ischemic stroke and new-onset AF, respectively. The most prevalent subtype of recurrent ischemic stroke was ESUS itself; this type of stroke occurred in 19 (59.3%) patients. AF was observed at stroke recurrence, but only 2 patients were diagnosed with cardioembolism. Multivariate analysis with a Cox proportional hazard model demonstrated that DSWMH grade greater than or equal to 3 was significantly associated with both recurrent ischemic stroke (hazard ratio 3.66, 95% confidence interval 1.69-7.92, P = .001) and new-onset AF (2.00, 1.03-3.90, .04).

Conclusions

MRI classification of white matter hyperintensity could be effectively used as a predictor for recurrent ischemic stroke and new-onset AF in patients with ESUS.