Correlation between Humphrey Field Analyser perimetric outcomes and macular thickness in diabetic retinopathy patients, diabetic individuals without retinopathy, and control subjects in Trinidad and Tobago

TL;DR

This study explores how visual field performance and macular thickness are related in diabetic retinopathy patients, diabetic individuals without retinopathy, and controls in Trinidad and Tobago.

Contribution

It identifies age as a major confounder and reveals specific correlations between visual field indices and macular thickness in diabetic retinopathy.

Findings

Age adjustment reduced between-group differences, highlighting age as a major confounder.

In diabetic retinopathy, lower visual field indices correlated with increased macular thickness in specific regions.

Macular total volume showed limited relevance to functional measures.

Abstract

This study examined the relationship between visual field performance measured with the Humphrey Field Analyzer and macular thickness measured using optical coherence tomography across three groups: individuals without diabetes, individuals with diabetes but no diabetic retinopathy, and individuals with diabetic retinopathy. The aim was to determine whether structural and functional measures show meaningful associations within and across disease stages in a Trinidad and Tobago population.

In unadjusted comparisons, macular and visual-field measures differed across groups (all p < 0.001). After age adjustment, most between-group differences were no longer significant, and age independently predicted VFI, confirming age as a major confounder. Within-group, age-adjusted associations persisted in DR (OD): lower VFI correlated with greater superior (partial r = − 0.519, p = 0.027) and temporal ETDRS thickness (partial r = − 0.585, p = 0.011), and higher PSD correlated with greater superior (partial r = + 0.603, p = 0.013) and temporal thickness (partial r = + 0.592, p = 0.016); all four remained significant after FDR correction. Macular total volume showed limited functional relevance. In DM_no_DR (exploratory), small-sample regressions suggested similar direction, but estimates were imprecise. MD was not analyzed via ANCOVA due to incomplete data.

The online version contains supplementary material available at 10.1186/s13104-026-07638-4.