advertisement
Abstract #12317 Published in IGR 7-2
Agreement between stereoscopic photographs, clinical assessment, Heidelberg retina tomograph and digital stereoscopic optic disc camera in estimating vertical cup:disc ratio
Jayasundera T; Danesh-Meyer HV; Donaldson M; Gamble GClinical and Experimental Ophthalmology 2005; 33: 259-263
BACKGROUND: The purpose of the present study was to determine the agreement between stereo photography, Heidelberg retina tomograph (HRT), digital stereoscopic optic disc camera (DISCAM) and ophthalmoscopy in estimating the vertical cup : disc ratio (VCDR). METHODS: One hundred and ninety-nine eyes of 100 subjects from glaucoma clinics were examined by a single glaucoma specialist and the VCDR on ophthalmoscopy was recorded. All eyes were classified as definite glaucoma, glaucoma suspect or no glaucoma based on the entire clinical assessment. Two glaucoma specialists viewed stereo photographs of the eyes separately and independently. A trained technician acquired HRT and DISCAM images. The agreement of VCDR among the four modalities was assessed by the method described by Bland and Altman. RESULTS: Of the 199 eyes, 28.1% were classified as definite glaucoma, 36.2% as glaucoma suspect and 35.7% as no glaucoma. The 2 SD for the agreement between the two glaucoma specialists using stereo photographs was 0.18 VCDR. The 2 SD for the agreement between stereo photographs and HRT, DISCAM and ophthalmoscopy were 0.31, 0.31 and 0.28, respectively. The 2 SD for ophthalmoscopy and HRT was 0.32, for ophthalmoscopy and DISCAM it was 0.32, and for HRT and DISCAM it was 0.22. When the eyes were divided into tertiles according disc size and VCDR, the 2 SD was higher for smaller optic discs ( < 1.5 mm vertical height) and smaller optic cups (VCDR < 0.4). CONCLUSIONS: There was good agreement between VCDR values estimated independently using stereo photographs by the two clinicians. However, when the different modalities were compared using Bland and Altman analysis, all showed poor agreement with a large 2 SD. The agreement was worse for small discs and smaller cups. The VCDR values obtained by the different methods were found not to be interchangeable in a clinic setting.
Dr. T. Jayasundera, Department of Ophthalmology, University at Auckland, Auckland, New Zealand. thiran88@hotmail.com
Classification:
6.8.2 Posterior segment (Part of: 6 Clinical examination methods > 6.8 Photography)
6.9.1 Laser scanning (Part of: 6 Clinical examination methods > 6.9 Computerized image analysis)
