Prediction Factors of Recurrent Stroke among Chinese Adults Using Retinal Vasculature Characteristics

Yuanyuan Zhuo1, Haibo Yu1, Zhuoxin Yang1

  • 1Shenzhen Traditional Chinese Medicine Hospital, Shenzhen, China.

Insights

Retinal vessel characteristics, like caliber and angles, are linked to recurrent stroke risk. Specific retinal measurements can help predict future strokes, especially when combined with clinical factors.

Area of Science:

  • Ophthalmology
  • Neurology
  • Cardiology

Background:

  • Recurrent stroke poses a significant health risk.
  • Identifying predictive markers for recurrent stroke is crucial for timely intervention.
  • Retinal vasculature may offer insights into cerebrovascular health.

Purpose of the Study:

  • To identify retinal vessel characteristics associated with recurrent cerebral infarction.
  • To explore the predictive potential of these retinal markers for stroke recurrence.

Main Methods:

  • Analysis of retinal vessel characteristics in 196 patients (137 new onset, 59 recurrent cerebral infarction) using an automated retinal image analysis system.
  • Comparison of retinal and clinical characteristics between new onset and recurrent stroke groups.
  • Multivariate logistic regression to identify risk factors for recurrent ischemic stroke within one year post-initial stroke.

Main Results:

  • Increased central retinal artery and vein equivalent caliber correlated with higher stroke recurrence risk in coronary heart disease patients.
  • Elevated arterial and venous angles increased stroke recurrence risk in atrial fibrillation patients.
  • Venous asymmetry index showed a decreased risk of recurrence when interacting with atrial fibrillation, coronary heart disease, and arteriole bifurcation coefficient.

Conclusions:

  • Retinal vessel characteristics interact with each other and with clinical risk factors.
  • These interactions significantly influence the risk of stroke recurrence.
  • Retinal imaging may serve as a valuable tool in assessing stroke recurrence risk.
Abstract