Study of the Correlation between Average Values of Optic Disc Parameters and Their Measurement Variability using Stereovideographic Digital Analysis

Abstract

Purpose of our study was to find whether there is a correlation between average values of optic nerve head parameters and their coefficient of variation for repeated measurements. Disc stereo video images were taken with Topcon Image-net (IS100) and analyzed with the Optic Disc Analysis(c) program. Eleven normal and 9 glaucomatous eyes were selected. A significant correlation was found for horizontal cup/disc ratio (r =––0.037, p = 0.037) and disc area (r = 0.776, p = 0.027) in intra-observer intra-image repeated measurements. Vertical, horizontal and average cup/disc ratios (r=––0.81, r = ––0.74, r=––0.78 p < 0.05) showed a significant correlation for intra-observer inter-image repeated measurements. For inter-observer intra-image measurements, a significant correlation was present for average cup/disc ratio (r = ––0.64 p = 0.043). Our data show that bidimensional semiquantitative parameters in use for optic nerve head assessment, analyzed by computerized stereovideography, have an intrinsic measurement variability. The variability is significantly related to the value of the measured parameter only when their reference points are determined manually by the operator.

Comparison of optic area measurement using fundus photography and Optical Coherence Tomography between optic nerve head drusen and control subjects

“… Difference of image capture : While fundus photography measurement is performed on something “visible” (2D image), an OCT measurement is a tomographic measurement made by the examiner on something ‘invisible’. Subjectivity of measurement : Complexity and interobserver variability exists when determining the exact limit of the optic disc both in the image from fundus photography (more so in hidden ONHD due to its increased appearance) and when manually determining the optic disc limit in OCT Different optic disc limits according to the instruments used : The beginning and end of the optic disc in planimetry is on the inner edge of the Elschnig ring; in TD‐OCT in the retinal pigment epithelium layer; and in SD‐OCT in Bruch's membrane. Effects of peripapillary atrophy and shadows caused by the presence of blood vessels may hinder the measurement …”

Optic disc surface smoothness and visual field indices

“…The latter is a good HRT parameter to distinguish normal ONH from glaucomatous ones (as already shown by Brigatti and Caprioli [3], Iester et al [12,13] and Mikelberg and colleagues [20], who used this parameter in a discriminant analysis formula in which it had the highest power to differentiate normal from glaucomatous optic discs). Although those two instruments were completely different regarding technique, both could be considered semicomputerized because they need observer input to draw the outline of the ONH by using a line [8,19,24], or by positioning at least four points in the four extreme positions (top, right, bottom, left) [11,25,26,29].…”

Correlation between perimetric indices and optic disc surface smoothness

“…The analogue signals from each camera travel to the red and green analogue-to-digital converters (Rolando et a!. 1994;lester et a!. 1995;Varma 1993).…”