Purchase this article with an account.
G. Lindhorst, W.E. Sponsel, M. Sinai, Y. Trigo; Asymmetry Analysis Using the Heidelberg Retina Tomograph Scanning Laser in a Predominantly Mexican–American Clinical Population . Invest. Ophthalmol. Vis. Sci. 2006;47(13):3406.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare the positive and negative predicative values of the Heidelberg Retina Tomograph II (HRT II) Moorfields regression analysis algorithm, compared to values obtained using the asymmetry algorithm (Pearson Product Coefficient) available in the HRT III forthcoming version, among a predominantly Mexican–American clinical population.
44 clinic patients, with an age range of 24 to 80 years, were referred to the glaucoma service after complete eye exam by comprehensive ophthalmologists on the basis of abnormal optic nerve appearance and/or ocular hypertension suggestive of glaucoma. Each subject underwent bilateral Humphrey Visual Field perimetry (HVF 24–2), and each field was classified in masked fashion as normal, early, moderate, or severe glaucoma according to Hodapp, Parrish, Anderson (HPA) criteria. Paired visual fields were further categorized into five groups; I: both eyes normal, II: both eyes symmetrically abnormal, and III–V: asymmetric HVFs differing by 1, 2, or 3 HPA category levels, respectively. HRT II scanning laser disc analysis was performed on both eyes of all subjects, providing least common denominator Moorfield nerve fiber layer gradings of abnormal, borderline or normal for each eye pair. These same HRT images were then evaluated using HRT III software incorporating an asymmetry algorithm. Subjects were classified as normal or abnormal based on the degree of asymmetry between the two disc images.
Among the 44 subjects, HVF yielded 14, 6, 10, 11, and 3 individuals in categories I–V respectively. Moorfields analysis yielded a sensitivity of 50%, specificity of 36%, positive predicative value (PPV) of 63%, and negative predicative value (NPV) of 25%. When using the asymmetry algorithm sensitivity improved to 60%, specificity to 93%, PPV to 95% and NPV 52%.
Optic nerve asymmetry on clinical exam is a well established risk factor for glaucoma. Moorfields regression analysis is a very useful tool when applied to individuals of northern European heritage. However, among our predominantly Mexican American population, the new HRT III asymmetry algorithm is a far better predictor of glaucomatous disease, effective in predicting HVF normality or glaucomatous field loss in the presence or absence of visual field asymmetry.
This PDF is available to Subscribers Only