Perception of color mismatch or conspicuous marginal adaptation in extraoral prostheses with eye-tracking

• Published in: The Journal of prosthetic dentistry 2024-02, Vol.131 (2), p.332-339
• Main Authors: Özcivelek, Tuğgen, Basmacı, Fulya, Turgut, Berna, Akbulut, Kuddusi, Kılıçarslan, Mehmet Ali
• Format: Article
• Language: English
• Subjects: Dental Implants; Eye-Tracking Technology; Face; Female; Humans; Male; Maxillofacial Prosthesis; Perception
• ISSN: 0022-3913; 1097-6841
• DOI: 10.1016/j.prosdent.2023.11.021

Color matching and marginal integrity are major challenges when providing extraoral maxillofacial prostheses. Which of the color and marginal harmony features are more important for the extraoral prostheses to be inconspicuous is unclear. Studies on the perception of these prostheses with objective evaluation criteria are lacking. The purpose of this observational study was to investigate the significance of color mismatch and conspicuous marginal adaptation in the perception of extraoral maxillofacial prostheses using eye-tracking technology. The secondary aim was to evaluate the perception of extraoral maxillofacial prostheses with regard to the observers’ sex. Twenty-seven face images, in 3 groups, representing well-fitting orbital prostheses with a color mismatch (IC), prostheses with a good color match but distinct marginal adaptation (IM), and symmetrical face images, were viewed for 5 seconds by 52 laypeople. Time to first fixation (TFF), fixation duration (FD), and fixation count (FC) at defined areas of interest were recorded and analyzed by an eye-tracking device. Because of the nested structure of data, a sex- and age-adjusted random intercept linear mixed effects model was used to assess the difference between IC, IM, and SI. Bonferroni corrected P values were used for pairwise comparisons. The difference between observers’ sex was evaluated with random intercept mixed model by adjusting for age for each image. For repeated measurement analysis, the lm4, lmerTest, and emmeans libraries in R version 4.3.1 (R Foundation for Statistical Computing) were used (α=.05 for all tests). Significant differences were found between the symmetrical image group and other study groups at the facial prosthesis region in all parameters (each P