Purchase this article with an account.
N. Wang, C. Qiao, X. Liu, J. Ge, X. Sun, W. Guo, Y. Jiang, Z. Luo, J. Zhao; Glaucoma Detection Using GDx VCC With Enhanced Variable Corneal Compensation, a Chinese Multi–Center Study . Invest. Ophthalmol. Vis. Sci. 2005;46(13):2525.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To evaluate the discriminating power of GDx VCC device with enhanced variable corneal compensation (ECC) between normal and glaucomatous eyes in a Chinese population. Methods: GDx VCC system (Laser Diagnostic Technologies, Inc., San Diego, California) employs scanning laser polarimetry technology for retinal nerve fiber layer (RNFL) assessment. A new software algorithm, ECC, was recently developed for the GDx VCC system. In ECC, individualized corneal compensation is achieved through a software method which mathematically removes the contribution from the anterior segment in each RNFL image. In this multi–center study, clinical data were collected from 5 sites in China under a single study protocol. RNFL image of 256 normal eyes and 134 glaucoma eyes were acquired with GDx VCC systems using the new ECC software. A Chinese normative database and a neural network classifier (NFI) were developed to objectively assess the RNFL status. Results: The median visual field mean deviation for the glaucoma patient group was –5.54 DB. Area under the ROC curve for various RNFL measurement parameters are: TSNIT average: 0.871, superior average: 0.850, inferior average: 0.885, TSNIT standard deviation: 0.870, NFI: 0.914. For the deviation map, at specificity of 90%, sensitivity is71.6% for super–pixels with p<5%, 76.1% for super–pixels with p<1% , and 78.4% for super–pixels with p<0.5%, respectively. Glaucomatous RNFL damage, both focal defect and diffuse defect, is readily identified in the retardation images, parameters, TSNIT curves, and deviation maps. RNFL assessment by GDx VCC agrees well with clinical diagnosis. RNFL damage prior to visual field loss was observed in eyes diagnosed with early stage glaucoma. Significant correlation was observed between RNFL measurement and visual field mean deviation in the study population. Conclusions: Glaucomatous RNFL damage can be detected effectively with GDx VCC technology. The new Chinese Normative database and neural network classifier may further improve GDx VCC’s ability to discrimination normal and glaucomatous eyes in Chinese population.
This PDF is available to Subscribers Only