Abstract
Purpose :
When measuring macular thickness using optical coherence tomography (OCT), axial length (AL) is known to affect the size of the scanning area due to transverse magnification effects (TME). However, most studies do not adjust for TME when reporting or analysing macular thickness. In this study, we compared macular thickness with and without TME correction in a large cohort of young adults.
Methods :
In a sample of 953 healthy adults in a general population (18-22years; 22% myopes, AL range: 20–28mm), the full retinal thickness at the 9 Early Treatment of Diabetic Retinopathy Study (ETDRS) macular regions (central macula, and the superior, inferior, temporal, and nasal quadrants of the inner and outer macular rings) were obtained using spectral domain OCT. Images were exported into a custom MATLAB program that corrected for TME. Paired t-tests were used to compare TME-corrected and uncorrected measurements. Statistical associations between the TME-corrected or uncorrected measurements with AL and other known predictors of macular thickness (sex, ethnicity, choroidal thickness) were explored using generalized estimating equations.
Results :
In the full study sample, failure to account for TME resulted in under-estimation of macular thickness at the central 0.5mm and over-estimation at other regions by 0.1–3.4µm (all p<0.001). Conversely in myopes, uncorrected measurements over-estimated thickness at the central macula and under-estimated at other regions (all p<0.001; difference=1.6–5.9µm). At the central macula, AL was not statistically associated with the uncorrected measurements (p=0.77) but had an inverse relationship with the TME-corrected thickness (p<0.001). At all other macular regions, there were significant associations between AL and uncorrected macular thickness, but not TME-corrected thickness. Sex, ethnicity, or choroidal thickness associations with macular thickness did not differ between the corrected and uncorrected measurements.
Conclusions :
In a general population where the majority are non-myopes, the macular thickness measurement error produced by TME is small and may not be clinically significant. However, as the global population becomes more myopic, this measurement error will increase. It is thus important to use TME-corrected macular thickness when exploring associations with AL or when study samples have high rates of ametropia.
This is a 2021 ARVO Annual Meeting abstract.