Abstract
Purpose: :
Myopia is much more prevalent in Japanese than in Caucasian. Uchida et al developed a new formula to diagnose glaucoma for myopic eyes with refractive errors < -3.0 diopter. The aim of this study is to compare the sensitivity and specificity of Moorfield’s Regression Analysis (MRA) and the Myopic Eye HRT Classification (MEHC, Uchida et al. ARVO 2007) which incorporates the ratio of disc diameter along the long axis to the short axis using Heidelberg Retina Tomograph II (HRT II, Heidelberg Instruments, Heidelberg, Germany) data since the optic discs in myopic eyes are often deformed.
Methods: :
Two hundred and fifty-one subjects (502 eyes) who were suspected to have glaucoma in community health screening were examined by slit lamp biomicroscopic examination, Goldmann applanation tonometry, Humphrey Automated Perimetry Central 30-2 program, gonioscopy and optic nerve head evaluation. Only reliable visual fields and acceptable images (topographic standard deviation ≤ 50 µm) were included and consequently we analyzed 327 eyes of all eyes examined.
Results: :
For eyes with refractive errors ≥ -3.0 D (262 eyes), the sensitivity and specificity using MRA were 51.1 % and 82.1 %, respectively, when treating the ‘borderline’ outcomes as test negative, and 71.1 % and 65.3 %, respectively, when treating the ‘borderline’ outcomes as test positive. For eyes with refractive errors < -3.0 D (65 eyes), the sensitivity and specificity using MRA were 62.5 % and 81.6 %, respectively, when treating the ‘borderline’ outcomes as test negative, and 75.0 % and 65.3 %, respectively, when treating the ‘borderline’ outcomes as test positive. For eyes with refractive errors < -3.0 D (65 eyes), the sensitivity and specificity using MEHC were 81.3 % and 63.3 %, respectively.
Conclusions: :
MEHC may be more useful for detecting glaucoma than MRA in myopic eyes in glaucoma screening.
Keywords: optic disc • imaging/image analysis: clinical • visual fields