Relationship between retinal layer thicknesses of the macula and depression

Abstract

Purpose : To study the relationship between depression and thickness of macular layers.

Methods : From the population-based, age- and sex-stratified LIFE-Adult-Study, reliable (quality ≥20 dB) macular spectral-domain optical coherence tomography (OCT) volume scans (97 B-scans at 512 A-scans) from participants with healthy eyes were selected and segmented into ten retinal layers (Fig. 1A). Each of the 49,664 A-scans of each layer was statistically associated with two measures of depression: First, a lifetime diagnosis of depression as reported by participants in the anamnestic interview (logistic regression; outcome: depression diagnosis, regressors: layer thickness with age as covariate); second, depressive symptoms as evaluated using the Center for Epidemiologic Studies Depression Scale (CES-D), a questionnaire assessing depressive symptoms during the past week (partial Pearson correlation between thickness and CES-D score adjusted for age). P-values were adjusted for multiple comparisons by the false-discovery method.

Results : 11,124 eyes of 6,471 participants were included (mean age: 55.5 years, 53.1% female). Participants who reported physician-diagnosed depression (n=705) had a significantly higher CES-D score (t-test, p<10^-15, see Fig. 1B). Concerning physician-diagnosed depression, retinal thickness of the ten layers was not significantly related to existing depression diagnosis for any of the examined retinal locations. Regarding current depressive symptoms, the heat maps in Fig. 2 show the partial (age-adjusted) correlations between CES-D and thickness for each of the ten layers. There were significantly correlated locations on each retinal layer, except for IZ and RPE layers, ranging from less than 1% of the retinal area for OPL to over 60% for EZ+OS. While for EZ+OS and the ONL, higher depression scores were associated with a thinner retina over widespread areas, for other layers, particularly GCL and the IPL, specific, localized spatial patterns of associations were found with more pronounced thinning at a ring around the fovea.

Conclusions : In 8 out of 10 retinal layers we found significant correlations between macular layer thickness and depressive symptoms, assessed using the CES-D. Anamnestic physician-diagnosed depression was not significantly associated with thickness.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.

 

Fig.1

 

Fig.2