The Degree of Posttraumatic Enophthalmos Detectable by Lay Observers

Abstract

AbstractThis study was to determine the degree of posttraumatic enophthalmos that is detectable by lay observers. In this case-control study, 60 photographs (frontal and oblique angles) of patients from an oculoplastics registry at the authors' institution were reviewed by lay observers. Patients were included if they had between 1 and 5 mm of unilateral enophthalmos, at least 6 months following trauma. They were divided into five groups of 10 and each group was separated by increments of 1 mm. Controls were selected from the same database and had 0 mm difference between each eye. Patients were excluded if they had any medical condition that might cause enophthalmos (including use of prostaglandin analogues). Participants were also excluded if they had any medical or surgical condition affecting the contralateral side. Lay observers were randomly selected from a pool of volunteers in nonmedical programs at our affiliated university. The primary outcome measure was the asymmetry cut-off (in millimeters) in which more than 50% of the cases (chance) were identified correctly. The generalized mixed model predicts that 66% of lay observers would correctly identify a difference between both eyes at a rate greater than would be predicted by chance, if at least a 5 mm difference existed between them (95% confidence interval [CI] 55.5–75%). This study shows that lay observers only reliably detect unilateral enophthalmos when there is at least a 5 mm difference between the eyes. This study used two-dimensional (2D) photographs and, therefore, results with three-dimensional (3D) observation may be different. In this study of 2D photographs, unilateral posttraumatic enophthalmos is only reliably detected by lay observers when a 5 mm difference exists between the eyes.