Regional Relationship between Glaucomatous Optic Disc Changes and Visual Field Findings

Abstract

Abstract: : Purpose: Moorfields regression analysis (MRA) of the confocal laser scanning ophthalmoscope (Heidelberg Retina Tomograph [HRT]) has been shown to be a sensitive and specific method for identification of glaucomatous optic disc changes. This study aimed to: 1) Investigate the relationship between pointwise visual field (VF) threshold values and corresponding segmental optic nerve head (ONH) MRA findings; and 2) Evaluate a new parameter, optic disc rim area deviation (RAD), for quantifying ONH abnormality, as determined by MRA. Methods: 46 normal eyes (26 subjects) and 39 glaucomatous eyes (35 subjects) were recruited for the study. All glaucomatous eyes had reproducible VF defects confined to one hemifield. VF pattern deviation threshold levels were localized to the ONH based on the Garway–Heath, et al. anatomical map (Ophthalmol 2000; 1809–1815). RAD was defined as: [(Measured segmental rim area – Predicted 50 percentile rim area)/Predicted 50 percentile rim area]. For all analyses, borderline MRA results were categorized as normal. Results: The sensitivity and specificity of the MRA was 64% and 98%, respectively. Sixteen of 39 glaucomatous eyes had localized VF defects with corresponding HRT abnormalities. Fourteen of the remaining 23 eyes had VF defects with no MRA abnormality. Only one eye had MRA abnormality in an area where the corresponding VF was normal. Segmental VF threshold levels were significantly associated with the corresponding segmental MRA findings (p<0.0001, Fisher exact test). Significant association was found between VF threshold values and MRA findings in nasal inferior, temporal, temporal inferior, and temporal superior sectors (p<0.05, mixed effects model). In contrast, average VF threshold values and optic disc RAD values were significantly associated in the temporal inferior and temporal superior sectors only (p<0.0001, mixed effects model). Conclusions: HRT MRA showed good association with corresponding visual field locations when borderline MRA findings were considered normal. Optic disc RAD did not provide additional information when compared to the currently employed MRA–defined binary categorization of normal or abnormal. Longitudinal studies are needed to assess the usefulness of this method of quantification for monitoring glaucoma progression.

Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • visual fields • optic disc