Comparison of Moorfields Regression Analysis and Glaucoma Probability Score Using Heidelberg Retina Tomograph in Glaucoma and Healthy Subjects

Abstract

Purpose:: To evaluate the Moorfields Regression Analysis (MRA) and Glaucoma Probability Score (GPS) classifications provided by the Heidelberg Retina Tomograph (HRT3).

Methods:: Glaucoma patients and healthy volunteers were enrolled in this study. The glaucoma subjects were included if they had glaucomatous optic nerve head alterations with correspondent visual field defects in the standard automated perimetry (Humphrey 24-2). The subjects were classified as healthy if they had normal-appearing optic nerve head, automated standard perimetry without visual field defect and intraocular pressure lower than 22 mmHg. The subjects were excluded if they had significant ocular disease other than glaucoma. The participants underwent confocal scanning laser ophthalmoscopy (HRT 3) in a dim light room without pupil dilation.

Results:: A total of 58 glaucoma patients (108 eyes) and 50 healthy subjects (98 eyes) were included in this study. The sensitivity of MRA and GPS were 71% (95% confidence interval [CI], 61%-79%) and 78% (95% CI, 69%-85%), respectively. There was no statistically significant difference in the sensitivity between the two classifications (P=0.21). The agreement between MRA and GPS in glaucoma patients was 65% (weighted kappa = 0.16). The specificity of MRA and GPS were 91% (95% CI, 84%-95%) and 87% (95% CI, 77%-93%), respectively. There was no statistically signifincat difference in the specificity between the two classifications (P=0.46). The agreement between MRA and GPS in healthy subjects was 56% (weighted kappa = 0.22).

Conclusions:: MRA and GPS classifications show similar diagnostic accuracy with moderate sensitivity and good specificity. The GPS classification has the advantage in comparison with MRA of identifying the optic nerve head contour without the intervention of the operator.

Keywords: optic disc • imaging/image analysis: clinical • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)