Skip to main content
  • Neuro-Ophthalmology/Orbit, Retina/Vitreous

    This population-based study investigated the association between inner retinal layer thickness and dementia among Dutch adults.

    Study design

    Using longitudinal data from the prospective population-based Dutch Rotterdam Study, investigators assessed retinal nerve fiber layer (RNFL) and ganglion cell-inner plexiform layer (GC-IPL) thicknesses to see if these parameters could predict the development of dementia. Baseline OCT measurements were obtained between September 2007 to June 2012, and individuals were followed until January 2015 for the onset of dementia.

    Outcomes

    The analysis included 3,289 individuals (57% female) with a mean age of 68.9. The 41 participants (1.2%) who had dementia at the start of the study had a thinner GC-IPL than those without dementia (OR 1.37), but there was no difference in RNFL thickness between groups.

    Eighty-six individuals (2.6%) developed dementia during the follow-up period. A thinner RNFL was associated with a higher risk of developing dementia (HR 1.51), but there was no association between GC-IPL thickness and incident dementia.

    Limitations

    It remains unclear why thinning of the RNFL predicted incident dementia, while GC-IPL predicted prevalent dementia, but not vice versa. In addition, there is a lot of overlap between normal patients and dementia and therefore OCT alone will unlikely be able to distinguish patients destined to develop dementia.

    Clinical significance

    This was the first prospective longitudinal study to assess the association between dementia and retinal thinning using OCT imaging. Prior cross-sectional studies have showed that patients with Alzheimer’s disease have a thinner RNFL and GC-IPL than age-matched controls, but those studies did not address whether retinal thinning precedes the occurrence of dementia.

    This study suggests that RNFL thinning may precede dementia and therefore could be used as a biomarker for dementia. Retinal neurodegeneration could potentially be used to monitor disease progression or evaluate treatments.