Inter-observer variation in clinical optic disc biometry

Optic Disc Measurements with the 78 Diopter Lens, Zeiss 4-Mirror Contact Lens and Computerized Image Analysing System

We evaluated the 78 diopter (D) lens and Zeiss 4-mirror lens for the measurement of vertical and horizontal optic disc diameters during slit-lamp biomicroscopy and compared the results with the measurements made with the computerized image analysing system (IMAGEnet 640, Topcon, Japan) in 30 eyes of 30 patients. The 78D lens and Zeiss 4-mirror lens measurements both correlated well with the computerized image analysis measurements (r=0.881 for vertical and r=0.895 for horizontal disc diameter measurements with the 78D lens and r=0.883 for vertical and r=0.891 for horizontal disc diameter measurements with the Zeiss 4-mirror lens). The 78D lens measurements overestimated vertical disc diameter by 5.3% and horizontal disc diameter by 4.4%. The Zeiss 4-mirror lens underestimated the vertical disc diameter by 2.4% and horizontal disc diameter by 2.2%. Thus either lens can be used for a quick estimation of the optic disc size, since the results correlate well with the computerized image analysis measurements.

Repeatability and reproducibility of optic disc measurement with the Zeiss 4-mirror contact lens

BACKGROUND

A simple clinical quantitative method of optic disc diameter measurement using a 4-mirror gonioscope contact lens has been described recently.

METHODS

Intraobserver and interobserver variations are assessed on 77 eyes of 41 patients. Of these eyes, 63 had refractive errors within 3 diopters of emmetropia. The disc was measured by projecting a slit beam of known height onto the disc.

RESULTS

Intraobserver variation: the coefficient of variation was 3.04% for observer 1 (experienced) and 4.76% for observer 2 (novice). The mean of the three measurements by the two observers was not significantly different (P=0.2318). For observer 1, the mean variance of the eyes with higher refractive errors did not differ significantly from the eyes with lower refractive errors. Interobserver variation: the coefficient of variation was 4.52%. The measurements for the novice were slightly larger the smallest discs and slightly smaller for the largest discs (regression analysis P=0.0209).

CONCLUSION

Repeatability and reproducibility for this method are comparable with other methods of measurement. Until experienced with the technique, the authors recommend that the mean of three readings is used.

Optic disc measurement: a comparison of indirect ophthalmoscopic methods

AIMS

Two methods of indirect ophthalmoscopic estimation of optic disc size, the 78 dioptre lens and optic disc biometer were evaluated.

METHODS

Twenty nine eyes of 29 patients were measured by both methods and compared with optic disc size calculated using the three planimetric corrections described by Bengtsson and Krakau.

RESULTS

The closest agreement with the clinical measurements was found using correction 3. There was a significant difference between both the 78 D lens (p < 0.0001) and the biometer (p = 0.0027) and the planimetric results. There was also a significant difference between the two clinical methods (p < 0.0001). Both methods showed acceptable intraobserver variation (CoV 2.45% and 3.13% respectively).

CONCLUSION

Overall, both methods give larger measurements than planimetry; the 78 D lens by 0.41 mm and the biometer by 0.15 mm. Neither method gives a satisfactory estimation of optic disc size when compared with planimetry.

Optic disc measurement with the Zeiss four mirror contact lens

A knowledge of the optic disc size may be of value when assessing the glaucoma suspect. The vertical diameter of the optic disc was measured using a Zeiss four mirror gonioscope and a 900 Haag-Streit slit-lamp in one eye of 39 patients, 32 with refractive errors within 3 dioptres of emmetropia. The disc was measured by projecting a slit beam of known height onto the image of the disc. A magnification factor for the contact lens was calculated from first principles and disc height recalculated. These measurements were compared with those obtained by photographic methods using the corrections suggested by Bengtsson and Krakau. In the analysis on the 32 eyes within 3 dioptres of emmetropia the best correlation with clinical measurements was obtained with correction 3 using spectacle refraction and keratometry (r = 0.8614). The contact lens measurement was within plus or minus 0.1 mm of the photographic measurement in 67% of cases and plus or minus 0.2 mm in all cases. This simple method is advocated for the routine assessment of optic disc size.