Longitudinal Evaluation Of Optic Disc Measurement Variability With Optical Coherence Tomography And Confocal Scanning Laser Ophthalmoscopy

Background: Structural assessment of the optic nerve is an integral part of ophthalmic examination. It provides the most important method for the diagnosis and the evaluation of glaucoma. Before the development of advanced technologies, careful comparison of simultaneous stereo-optic nerve photographs was, perhaps, the clinician's best tool for detecting progressive optic nerve damage. The photographs can be stored for subsequent study and review. However, this technique is qualitative and subjective because it involves the interpretation of subtle findings by the clinician. Optical coherence tomography (OCT) and confocal scanning laser ophthalmoscopy (CSLO) are commercially available imaging instruments that provide quantitative assessment of the optic disc. Although high sensitivity for glaucoma detection has been demonstrated with these instruments, their relative reliabilities during longitudinal assessment are less well-defined.Purpose: To evaluate and compare the longitudinal variability of optical coherence tomography (OCT) and confocal scanning laser ophthalmoscopy (CSLO) optic disc measurements.Methods: A total of 25 normal and 50 glaucomatous eyes from 75 subjects were included in the analysis. The optic disc was measured by OCT and CSLO. Three separate measurements collected over an average period of 8.5+/‐0.9 months were used to evaluate reproducibility. Univariate and multivariate regression analyses were performed to evaluate the associations between age, refraction, diagnosis (glaucoma or normal), visual field mean deviation, optic disc area, signal strength variance (OCT), optic disc area variance (OCT), image quality standard deviation (CSLO), reference height variance (CSLO) and rim area variability.Results: The intraclass correlation coefficient (ICC) of optic disc measurements (except for optic disc area) ranged between 0.86 and 0.95 for OCT, and between 0.89 and 0.96 for CLSO. The ICC for rim area measurement was significantly higher in CSLO (0.95) than that of OCT (0.86, P<0.001). After adjustment for other predictors, optic disc area variance and reference height variance were the most important factors responsible for rim area variability in OCT and CSLO, respectively. Conclusions: While both OCT and CSLO have relatively low variability for optic disc measurements, CSLO demonstrates higher measurement reproducibility for rim area compared with OCT. Variations of disc area in OCT and reference height in CSLO constituted a significant proportion of the rim area variability during longitudinal assessment.