Intra- and interobserver reliability of a modified distraction test based on digital images to assess lower eyelid horizontal tension

• Published in: Arquivos brasileiros de oftalmologia Vol. 83; no. 2; pp. 127 - 131
• Main Authors: Stuchi, Daniella de Paiva Almeida, Rossato, Juliana, Bocaccio, Francisco José de Lima, Procianoy, Fernando
• Format: Journal Article
• Language: English
• Published: Brazil Conselho Brasileiro de Oftalmologia 01.03.2020
• Subjects: Adult; Eyelid Diseases - diagnostic imaging; Eyelid Diseases - physiopathology; Eyelids - diagnostic imaging; Eyelids - physiopathology; Humans; Image Interpretation, Computer-Assisted - methods; Image Processing, Computer-Assisted - methods; Image Processing, Computer-Assisted - standards; Middle Aged; Observer Variation; OPHTHALMOLOGY; Photography - methods; Prospective Studies; Reference Values; Reproducibility of Results; Software; Eyelid diseases; Eyelids; Doenças palpebrais; Ectropion; Pálpebras; Digital image; Ectrópio; Entrópio; Imagem digital; Entropion
• ISSN: 0004-2749; 1678-2925; 1678-2925
• DOI: 10.5935/0004-2749.20200018

Inferior eyelid laxity is classically evaluated using "snap-back" and "distraction" tests. This study aimed to assess the reproducibility of the technique used to indirectly quantify the horizontal tension in the lower eyelids using digital image processing. This longitudinal study was conducted to assess the reproducibility of a new technique that quantifies the horizontal tension in the lower eyelid. The study was conducted at the Hospital das Clínicas of Porto Alegre. The protocol was established by two trained ophthalmologist examiners, allowing intra- and interobserver agreement analyses. Image acquisition was done in two stages: the first image was captured with the eyelid in primary gaze position and the second with the eyelid in traction position. All images and measurements were processed using Image J 1.33m software from the National Institute of Health. The Bland-Altman method, intraclass correlation coefficients, concordance correlation coefficients, and technical measurement error were used to evaluate reproducibility. The study participants comprised healthy individuals with no ophthalmologic pathologies. The measurements obtained in the neutral position showed a slightly higher agreement than those obtained in the traction position. The mean difference between the measurements performed in the traction position was 0.028 ± 0.7 mm and 0.014 ± 0.9 mm in the intra- and interobserver analyses, respectively. The Bland-Altman method demonstrated adequate confidence limits for both measurements. Correlation coefficients for measurements varied between 0.87 [95% confidence interval (CI) 0.68-0.95] and 0.91 (95% CI 0.77-0.97) in the neutral position and between 0.72 (95% CI 0.37-0.89) and 0.76 (95% CI 0.4-0.91) in the traction position. A high intra- and interobserver concordance was observed in the studied method to quantify lower eyelid tension. The proposed method is simple and easily reproducible, and to the best our knowledge, this is the first method that quantifies lower eyelid horizontal tension on the basis of digital image processing. This modified distraction test might be useful in studies quantifying lower eyelid horizontal tension.