Jean-Claude Mwanza1, Donald L Budenz2, David G Godfrey3, Arvind Neelakantan3, Fouad E Sayyad4, Robert T Chang5, Richard K Lee4. 1. Department of Ophthalmology, University of North Carolina School of Medicine, Chapel Hill, North Carolina. 2. Department of Ophthalmology, University of North Carolina School of Medicine, Chapel Hill, North Carolina. Electronic address: donald_budenz@med.unc.edu. 3. Glaucoma Associates of Texas, Dallas, Texas. 4. Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida. 5. Department of Ophthalmology, Stanford University School of Medicine, Palo Alto, California.
Abstract
PURPOSE: To evaluate the glaucoma diagnostic performance of ganglion cell inner-plexiform layer (GCIPL) parameters used individually and in combination with retinal nerve fiber layer (RNFL) or optic nerve head (ONH) parameters measured with Cirrus HD-OCT (Carl Zeiss Meditec, Inc, Dublin, CA). DESIGN: Prospective cross-sectional study. PARTICIPANTS: Fifty patients with early perimetric glaucoma and 49 age-matched healthy subjects. METHODS: Three peripapillary RNFL and 3 macular GCIPL scans were obtained in 1 eye of each participant. A patient was considered glaucomatous if at least 2 of the 3 RNFL or GCIPL scans had the average or at least 1 sector measurement flagged at 1% to 5% or less than 1%. The diagnostic performance was determined for each GCIPL, RNFL, and ONH parameter as well as for binary or-logic and and-logic combinations of GCIPL with RNFL or ONH parameters. MAIN OUTCOME MEASURES: Sensitivity, specificity, positive likelihood ratio (PLR), and negative likelihood ratio (NLR). RESULTS: Among GCIPL parameters, the minimum had the best diagnostic performance (sensitivity, 82.0%; specificity, 87.8%; PLR, 6.69; and NLR, 0.21). Inferior quadrant was the best RNFL parameter (sensitivity, 74%; specificity, 95.9%; PLR, 18.13; and NLR, 0.27), as was rim area (sensitivity, 68%; specificity, 98%; PLR, 33.3; and NLR, 0.33) among ONH parameters. The or-logic combination of minimum GCIPL and average RNFL provided the overall best diagnostic performance (sensitivity, 94%; specificity, 85.7%; PRL, 6.58; and NLR, 0.07) as compared with the best RNFL, best ONH, and best and-logic combination (minimum GCIPL and inferior quadrant RNFL; sensitivity, 64%; specificity, 100%; PLR, infinity; and NPR, 0.36). CONCLUSIONS: The binary or-logic combination of minimum GCIPL and average RNFL or rim area provides better diagnostic performances than those of and-logic combinations or best single GCIPL, RNFL, or ONH parameters. This finding may be clinically valuable for the diagnosis of early glaucoma.
PURPOSE: To evaluate the glaucoma diagnostic performance of ganglion cell inner-plexiform layer (GCIPL) parameters used individually and in combination with retinal nerve fiber layer (RNFL) or optic nerve head (ONH) parameters measured with Cirrus HD-OCT (Carl Zeiss Meditec, Inc, Dublin, CA). DESIGN: Prospective cross-sectional study. PARTICIPANTS: Fifty patients with early perimetric glaucoma and 49 age-matched healthy subjects. METHODS: Three peripapillary RNFL and 3 macular GCIPL scans were obtained in 1 eye of each participant. A patient was considered glaucomatous if at least 2 of the 3 RNFL or GCIPL scans had the average or at least 1 sector measurement flagged at 1% to 5% or less than 1%. The diagnostic performance was determined for each GCIPL, RNFL, and ONH parameter as well as for binary or-logic and and-logic combinations of GCIPL with RNFL or ONH parameters. MAIN OUTCOME MEASURES: Sensitivity, specificity, positive likelihood ratio (PLR), and negative likelihood ratio (NLR). RESULTS: Among GCIPL parameters, the minimum had the best diagnostic performance (sensitivity, 82.0%; specificity, 87.8%; PLR, 6.69; and NLR, 0.21). Inferior quadrant was the best RNFL parameter (sensitivity, 74%; specificity, 95.9%; PLR, 18.13; and NLR, 0.27), as was rim area (sensitivity, 68%; specificity, 98%; PLR, 33.3; and NLR, 0.33) among ONH parameters. The or-logic combination of minimum GCIPL and average RNFL provided the overall best diagnostic performance (sensitivity, 94%; specificity, 85.7%; PRL, 6.58; and NLR, 0.07) as compared with the best RNFL, best ONH, and best and-logic combination (minimum GCIPL and inferior quadrant RNFL; sensitivity, 64%; specificity, 100%; PLR, infinity; and NPR, 0.36). CONCLUSIONS: The binary or-logic combination of minimum GCIPL and average RNFL or rim area provides better diagnostic performances than those of and-logic combinations or best single GCIPL, RNFL, or ONH parameters. This finding may be clinically valuable for the diagnosis of early glaucoma.
Authors: Hrvoje Bogunović; Young H Kwon; Adnan Rashid; Kyungmoo Lee; Douglas B Critser; Mona K Garvin; Milan Sonka; Michael D Abràmoff Journal: Invest Ophthalmol Vis Sci Date: 2014-12-09 Impact factor: 4.799
Authors: Hema L Ramkumar; Brian Nguyen; Dirk-Uwe Bartsch; Luke J Saunders; Ilkay Kilic Muftuoglu; Qisheng You; William R Freeman Journal: Retina Date: 2018-11 Impact factor: 4.256