To compare the meibographies and dry eye parameters of paretic vs non-paretic sides of patients with a facial palsy diagnosis. Twenty patients with unilateral facial palsy were recruited and the severity of the disease was staged using the House-Brackmann scale. A comprehensive dry eye evaluation was performed using the Oculus 5M Keratograph. A Pearson correlation coefficient was performed to determine correlation strength between House-Brackmann score and Meibomian gland atrophy. Meibographies were analyzed via ImageJ software to determine the affected area, and they were compared to the observer manual score. Cohen's Kappa coefficient was calculated to compare agreement between manual and ImageJ meibography scoring. Tear breakup time was reduced in the affected side (p = 0.21), tear meniscus height was much greater in the non-affected side (p = 0.02). Finally, Meibomian gland alterations were more evident in the affected side, with upper Meibomian glands having a loss of 29.55 ± 13.31% (p = 0.03) and lower glands presenting a loss of 44.44 ± 16.9% (p =< 0.001). Pearson correlation coefficient between House-Brackmann score and Meibomian gland atrophy was 0.841 (p < 0.001 [95% CI: 0.64-0.94]). Cohen's kappa coefficient was 0.643 (p < 0.001). A clear difference in Meibomian gland and tear film dynamics can be observed in paretic vs non-paretic sides. A greater House-Brackmann score was correlated with a greater Meibomian gland atrophy area. A strong positive correlation is seen between the House-Brackmann score and Meibomian gland atrophy. Software-based analysis also showed a greater glandular area loss when compared to clinician's analysis. The level of agreement was moderate, so disparities are observed, especially in grade 2 Meibomian gland dropout where the least level of agreement was seen in cross tabulation. This study further incentivizes multimodal patient evaluation, which has been a growing area of interest in healthcare.
Read full abstract