Purpose: To assess the correlation of upper eyelid retraction and the disease natural course of untreated thyroid eye disease patients, using multiple radial midpupil lid distance, for individualized eyelid morphological treatment.Methods: Semiautomatic photographic analysis was performed using ImageJ. Temporal/nasal multiple radial midpupil lid distance ratios of the same angles with respect to the midline was calculated. The influence of several variable factors on the asymmetrical upper eyelid retraction was evaluated.Results: One hundred and fifty-one eyes of 85 untreated thyroid eye disease patients were included. The mean age of patients was 45.4 +/- 14.7 years. Multiple radial midpupil lid distance ratios increased during the first 12 months, 105 degrees/75 degrees (r = 0.196, p = 0.087), 120 degrees/60 degrees (r = 0.250, p = 0.028), 135 degrees/45 degrees (r = 0.309, p = 0.006), and 150 degrees/30 degrees (r = 0.275, p = 0.015). The ratios then slowly decreased, 120 degrees/60 degrees (r = -0.332, p = 0.006), 135 degrees/45 degrees (r = -0.297, p = 0.014), and 150 degrees/30 degrees (r = -0.254, p = 0.037). The ratios plateaued at 13 months but were still greater than normal range. No correlation found between smoking history (p = 0.230), family history (p = 0.382), exophthalmometry (p = 0.597), and natural course. Women had a higher likelihood of upper eyelid retraction asymmetry, 120 degrees/60 degrees (p = 0.041) and 135 degrees/45 degrees (p = 0.048). The absence of systemic disease was associated with a lower likelihood of upper eyelid retraction asymmetry in 120 degrees/60 degrees (p = 0.036). No significant difference was found in the contralateral eye of unilateral patients.Conclusion: Multiple radial midpupil lid distance is a valuable method for measuring asymmetrical upper eyelid retraction, enabling improved understanding in upper eyelid contour with the duration of untreated thyroid eye disease. Allowing a better decision on individualized treatment.