Assaf Kratz1, Ridia Lim, Ivan Goldberg. 1. Department of Ophthalmology, Soroka University Medical Centre, Faculty of Health Sciences, Ben Gurion University of the Negev, Beer Sheva, Israel; Glaucoma Unit, Sydney Eye Hospital, Sydney, New South Wales, Australia.
Abstract
BACKGROUND: The purpose of this study was to analyse the relationship between optic nerve head (ONH) parameters measured by spectral domain optical coherence tomography and confocal scanning laser ophthalmoscope. DESIGN: Prospective, cross-sectional study. Hospital setting. PARTICIPANTS: One hundred seventy-three subjects (85 glaucoma and 88 normal subjects). METHODS: One eye from each individual was selected randomly for ONH imaging by the spectral domain Cirrus OCT and Heidelberg Retinal Tomograph 3 (HRT3). MAIN OUTCOME MEASURES: Four ONH parameters that are measured by both technologies (average cup-to-disc ratio [CDR], rim area, disc area and cup volume) were analysed and compared for differences, agreement of the categorical classification, diagnostic sensitivities and specificities and the area under the receiver operating characteristic curves (AUC). RESULTS: ONH parameters, as determined by the two technologies were significantly different but strongly correlated. Proportional bias was demonstrated for all measurements. The agreement of categorical classification was excellent for CDR (κ = 0.94) and good for rim area and cup volume (κ = 0.63 and 0.71, respectively). The highest sensitivities at fixed specificities were achieved by Cirrus OCT. AUCs for CDR, rim area, disc area and cup volume were not significantly different between the two technologies. CONCLUSIONS: The diagnostic capability of ONH measurements by both technologies is similar. Paired ONH measurements by Cirrus OCT and HRT3 are strongly correlated but significantly different and proportionally biased. The results preclude interchangeable use of the absolute values, but categorical classification of ONH parameters may be interchangeable in clinical practice.
BACKGROUND: The purpose of this study was to analyse the relationship between optic nerve head (ONH) parameters measured by spectral domain optical coherence tomography and confocal scanning laser ophthalmoscope. DESIGN: Prospective, cross-sectional study. Hospital setting. PARTICIPANTS: One hundred seventy-three subjects (85 glaucoma and 88 normal subjects). METHODS: One eye from each individual was selected randomly for ONH imaging by the spectral domain Cirrus OCT and Heidelberg Retinal Tomograph 3 (HRT3). MAIN OUTCOME MEASURES: Four ONH parameters that are measured by both technologies (average cup-to-disc ratio [CDR], rim area, disc area and cup volume) were analysed and compared for differences, agreement of the categorical classification, diagnostic sensitivities and specificities and the area under the receiver operating characteristic curves (AUC). RESULTS: ONH parameters, as determined by the two technologies were significantly different but strongly correlated. Proportional bias was demonstrated for all measurements. The agreement of categorical classification was excellent for CDR (κ = 0.94) and good for rim area and cup volume (κ = 0.63 and 0.71, respectively). The highest sensitivities at fixed specificities were achieved by Cirrus OCT. AUCs for CDR, rim area, disc area and cup volume were not significantly different between the two technologies. CONCLUSIONS: The diagnostic capability of ONH measurements by both technologies is similar. Paired ONH measurements by Cirrus OCT and HRT3 are strongly correlated but significantly different and proportionally biased. The results preclude interchangeable use of the absolute values, but categorical classification of ONH parameters may be interchangeable in clinical practice.