RESUMO
PURPOSE: To develop a standardized and universally reproducible grading scale for artistic outcome of a custom ocular prosthesis. METHODS: A retrospective review of photographs of patients with custom ocular prosthesis. From the ocularistry database of a tertiary eye care hospital, photographs of patients with custom ocular prosthesis (30 random photos with additional 18 repetitions, total n = 50) were selected. Two independent oculoplastic surgeons, on the basis of characteristics of limbus, vascularity, and pigmentation, graded these photos. The photos were adjusted for confounders like magnification and illumination, and both graders used the same display system. The surgeons were blinded to each other. The characteristics were graded on a scale of 0 to 2 and total score was calculated for each photo. Inter-observer and intra-observer agreement was calculated using the kappa statistic. RESULTS: The inter-observer kappa coefficient for the three characteristics (95% confidence interval) were as follows: limbus 0.7 (0.55-0.85), vascularity 0.76 (0.58-0.94), and pigmentation 0.6 (0.34-0.85). The intra-observer kappa coefficient for the first surgeon was as follows: limbus 0.88 (0.7-0.95), vascularity 0.88 (0.73-0.95), and pigmentation 0.82 (0.54-0.93). The intra-observer kappa coefficient for the second surgeon was as follows: limbus 0.8 (0.44-0.93), vascularity 0.55 (0.13-0.8), and pigmentation 0.76 (0.6-0.84). CONCLUSIONS: A good intra- and inter-observer agreement was demonstrated. In view of no existing grading for prosthesis, this is a unique system to objectively grade the outcome of custom ocular prosthesis and can be incorporated in ocularistry teaching modules for standardization of outcomes.
