Abstract
Purpose: :
To describe age-related changes of the human macula in vivo using optical coherence tomography (OCT).
Methods: :
Altogether 55 eyes of 55 healthy volunteers (age range 20-88 years, 16 males, 39 females) were involved in the study. Inclusion criteria were: best corrected visual acuity of 1.0 and absence of any ocular pathologies except for uneventful cataract surgery in the history. Exclusion criteria were: diabetes mellitus, uncontrolled hypertension or the presence of any other general disease. Standard macular mapping was performed in all selected eyes by StratusOCT (Carl Zeiss Meditec, Dublin, CA), the raw scan data were exported and analyzed using Retinal Image Analysis software (OCTRIMA). Average thickness was extracted for all retinal layers. Sigmoid and linear functions were fitted using the least squares regression technique to decide which model describes better the kinetics of retinal layer thickness changes occurring with age. The models were compared using the residual standard errors (RSE), the Akaike Information Criterion (AIC) and the F test on residual sum of squares.
Results: :
A decrease in thickness of RNFL was observed with age which was better described by a sigmoid model, however F-test was not significant (RSE 2.508 vs. 2.556, AIC 265.05 vs. 265.26, sigmoid vs. linear, respectively, F-test p=0.14). In the case of GCL+IPL a sigmoidal trend gave a significantly better fit for the observed thickness decrease with age (RSE 4.703 vs. 4.968, AIC 330.27 vs. 338.37, F-test p=0.023). The thickness of the RPE layer showed a linear increase related to age, as only a linear model could be fitted (linear correlation r=0.36, p=0.006). All other layers (INL, OPL and ONL) did not show any significant age-related thickness changes.
Conclusions: :
Our results suggest that age-related cellular changes of the macula may occur with different kinetics. The known decrease in ganglion cell number (and thus GCL+IPL thickness) follows a sigmoidal trend, supporting the theory of apoptosis. This is paralleled by a decrease in the nerve fiber layer following the same, although less evident trend. The linear thickening of the RPE might be in correlation with described age-related histological changes involving the Bruch membrane.
Keywords: aging • retina • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)