OBJECTIVE: As a result of their complex aetiology and periodicity, dark circles are difficult to characterize and measure, with current assessment techniques relying on specialist equipment, image analysis or proprietary grading scales. There is therefore a need to develop and validate a photonumeric scale for assessing infraorbital dark circles, which can provide an objective and consumer relevant tool for evaluating this condition and the efficacy of treatment products and procedures. METHODS: A panel of expert clinical evaluators reviewed approximately three thousand facial photographs collected over a 5-year period and selected images representing a dynamic range of dark circles. A 10-point photonumeric scale was created, with corresponding descriptors and images for each grade of the scale. To rigorously validate the scale, linearity, sensitivity and precision were assessed by colorimetry and in-clinic evaluation. Reproducibility was assessed photographically with both experienced and inexperienced clinical evaluators, whereas intragrader repeatability was assessed live in-clinic. The scale was then employed in a split-face randomized clinical trial on 58 subjects to evaluate the efficacy of a cosmetic treatment product over 8 weeks. RESULTS: Colour analysis of the images showed the scale was linear, with statistically significant correlations observed when colour data (CIElab; Individual Typology Angle) were plotted against the corresponding grades (r > 0.9, P < 0.001). Colour difference (Delta E) was calculated between the infraorbital zone and the surrounding skin, and when data were plotted against the grades, a statistically significant correlation was observed (r = 0.99, P < 0.01). The magnitude of the Delta E suggested that changes in grade are visibly perceptible to the human eye, and therefore, the scale is sensitive and clinically relevant. Intergrader reproducibility showed strong correlation (0.96) and >90% agreement between experienced evaluators, whereas intragrader repeatability assessment showed >90% perfect agreement between grades. Use of this scale in a clinical trial demonstrated the efficacy of a cosmetic product, with a mean statistically significant (P < 0.001) decrease in grade of 0.74 compared to baseline, and 0.59 versus the untreated control, after 8 weeks of treatment. CONCLUSION: Our photonumeric scale for infraorbital dark circles is sensitive and robust and provides an objective and easy-to-use tool to evaluate dark circles and their treatment.
OBJECTIVE: As a result of their complex aetiology and periodicity, dark circles are difficult to characterize and measure, with current assessment techniques relying on specialist equipment, image analysis or proprietary grading scales. There is therefore a need to develop and validate a photonumeric scale for assessing infraorbital dark circles, which can provide an objective and consumer relevant tool for evaluating this condition and the efficacy of treatment products and procedures. METHODS: A panel of expert clinical evaluators reviewed approximately three thousand facial photographs collected over a 5-year period and selected images representing a dynamic range of dark circles. A 10-point photonumeric scale was created, with corresponding descriptors and images for each grade of the scale. To rigorously validate the scale, linearity, sensitivity and precision were assessed by colorimetry and in-clinic evaluation. Reproducibility was assessed photographically with both experienced and inexperienced clinical evaluators, whereas intragrader repeatability was assessed live in-clinic. The scale was then employed in a split-face randomized clinical trial on 58 subjects to evaluate the efficacy of a cosmetic treatment product over 8 weeks. RESULTS: Colour analysis of the images showed the scale was linear, with statistically significant correlations observed when colour data (CIElab; Individual Typology Angle) were plotted against the corresponding grades (r > 0.9, P < 0.001). Colour difference (Delta E) was calculated between the infraorbital zone and the surrounding skin, and when data were plotted against the grades, a statistically significant correlation was observed (r = 0.99, P < 0.01). The magnitude of the Delta E suggested that changes in grade are visibly perceptible to the human eye, and therefore, the scale is sensitive and clinically relevant. Intergrader reproducibility showed strong correlation (0.96) and >90% agreement between experienced evaluators, whereas intragrader repeatability assessment showed >90% perfect agreement between grades. Use of this scale in a clinical trial demonstrated the efficacy of a cosmetic product, with a mean statistically significant (P < 0.001) decrease in grade of 0.74 compared to baseline, and 0.59 versus the untreated control, after 8 weeks of treatment. CONCLUSION: Our photonumeric scale for infraorbital dark circles is sensitive and robust and provides an objective and easy-to-use tool to evaluate dark circles and their treatment.