Combining rim area to disc area asymmetry ratio and Moorfields regression analysis of confocal scanning laser ophthalmoscopy for glaucoma screening

Abstract

Objective To determine the utility of combining rim area to disc area asymmetry ratio (RADAAR) and Moorfields regression analysis (MRA) to detect primary open-angle glaucoma (POAG) using confocal scanning laser ophthalmoscopy (CSLO) (Heidelberg retina tomograph 3 (HRT3)) in high-risk populations. Design Cross-sectional study. Participants Subjects with high risk for developing POAG. Methods Subjects underwent HRT, frequency doubling technology perimetry (FDT), and complete ophthalmic examination. Based on an ophthalmologic examination and FDT results of the worse eye, subjects were classified into 4 categories: normal, possible glaucoma, probable glaucoma, and definite glaucoma. Main outcome measures included sensitivities (Se), specificities (Sp), positive and negative predictive values (PPV, NPV), and positive and negative likelihood ratios (PLR, NLR) of the MRA, RADAAR, and combined MRA- RADAAR tests. Results Of 375 subjects, 11 were classified as having definite glaucoma. Depending on the reference standard, range results for Se, Sp, PPV, NPV, PLR, and NLR were, respectively, 57.7% to 100%; 55.1% to 85.4%; 6.3% to 39%; 92.6% to 100%; 1.97 to 4.9; 0 to 0.49 for MRA; 11.5% to 27.3%; 98.1% to 98.8%; 30% to 60%; 87.4% to 97.8%; 9.29 to 14.1; and 0.74 to 0.89 for RADAAR; and 57.7% to 90.9%; 81.3% to 85.1%; 12.8% to 38.5%; 92.6% to 99.7%; 3.87 to 4.85; 0.11 to 0.50 for combination MRA-RADAAR. Discussion When using HRT in populations at high risk for glaucoma, RADAAR had a higher specificity than MRA in identifying glaucoma. When patients were classified as borderline on MRA, combining RADAAR values decreased the number of false-positives as well, and may be recommended when high specificity is required.