Author: Yifan Chen (United Kingdom)
Co-authors: Zhuoting Zhu, Mingguang He
Purpose
To investigate associations of retinal age gap with arterial stiffness and incident cardiovascular disease (CVD).
Setting/Venue
The current study operates in accordance with the principles of the Declaration of Helsinki, with written informed consent from all participants, under the UK Biobank application number 62489.
Methods
A total number of 80,170 fundus images from 46,970 participants in the UK Biobank had reasonable image quality and were included. A deep learning (DL) model based on 19,200 fundus images of 11,052 disease-free participants was used to predict the retinal age via 5-fold cross-validation. Retinal age gap (retinal age predicted based on the fundus image from the right eye minus chronological age) was generated by the trained DL model for the remaining 35,917 participants. Arterial stiffness index (ASI) was derived from the analysis of digital volume pulse. Incident CVD was ascertained by self-reported questionnaire, hospital admission records or death records, whichever the earliest. Linear regression models were used to assess the association between retinal age gap and ASI. Cox proportional hazards regression models were used to explore the relationship between retinal age gap and incident CVD in the population free of CVD at the baseline.
Results
In the 35,541 participants with available ASI data, we found each one-year increase in the retinal age gap was significantly associated with ASI (β = 0.0016, 95% confidence interval [CI]: 0.001- 0.003, P<0.001). After a median follow-up of 5.83 years (interquartile range [IQR]: 5.73-5.97), 675 (2.00%) out of 33,817 participants who were deemed free of CVD had incident CVD. In the fully adjusted model, each one-year increase in retinal age gap was associated with a 3% increase in the risk of incident CVD (hazard ratio [HR]=1.03, 95% CI: 1.01-1.06, P=0.012). This association remained significant after further adjustment for ASI.
Conlusions
We found that retinal age gap was significantly associated with ASI and incident CVD, implying the potential of this biomarker in identifying individuals at high risks of CVD.
Financial Disclosure
none
Comments
-
