Disagreement in contour placement can produce inconsistent tomograph parameters

When contouring optic disc images, disagreement in contour placement between independent observers can result in disproportionate effects on some calculated Heidelberg Retina Tomograph parameters, independent of image quality and lens opacity, according to a new analysis of data from the Bridlington Eye Assessment Project.

"[Rim-to-disc] area ratio is the HRT parameter least affected and may be more useful in comparing normative databases and diagnostic studies between centers," the study authors said.

Matthew J. Hawker, BMedSci(Hons), MRCOphth, and colleagues assessed the agreement in optic nerve head parameters observed between two investigators while independently contouring Heidelberg Retina Tomograph (HRT) II images of 550 randomly selected optic discs of 550 normal patients. These patients averaged 73.1 years of age and had normal IOPs and visual fields. Subsequently, the investigators calculated coefficients of variation for interobserver and intraobserver variability.

Disc area averaged 1.97 mm² in the population, the authors noted.

The investigators found no relationship between magnitude of disagreement and the average disc area. However, there was a significant bias between investigators with an average difference in disc area of –0.06 mm², according to the study.

"The 95% limits of agreement of disc area were –0.59 mm² to +0.47 mm²," they said.

The 95% coefficients of agreement were similar for global and sectoral disc areas (26.7% to 32.3%). However, coefficients for rim area (35.2%), cup area (48.5%) and rim volume (76.5%) were higher.

Conversely, the rim-to-disc area ratio was 12.7%, the authors noted.

The researchers found no significant correlation between disagreement and grade of cataract or HRT image quality, according to the study, published in the June/July issue of Journal of Glaucoma.