Abstract
Purpose :
Various ocular surface conditions including dry eye disease can present with pain.However,it is challenging to establish etiology and prescribe right treatments due to discordance between patient symptoms and signs.To understand basis of such discordance,we stratified subjects with ocular surface pain based on concordance between severity of signs and symptoms and evaluated corneal structural features and tear molecular factors.We also analysed the various confocal nerve parameters along with systemic and orthoptic parameters in these patients presenting with ocular surface pain using a random forest artificial intelligence(AI) model.
Methods :
300 eyes of 151 patients underwent slit lamp examination,ocular surface disease index (OSDI) scoring,dry eye evaluation and ocular surface staining.Subjects were stratified into -group 1-with no symptoms and clinical signs;group 2-with no symptoms but with signs;group 3-with similar severity of symptoms and signs;and group 4- with symptom severity higher than that of signs.In vivo confocal imaging (IVCM)evaluation was performed in all study subjects.We also evaluated presence or absence of orthoptic issues and connective tissue disorders and in the first step the area under curve (AUC),accuracy,recall,precision and F1-score of the AI model were evaluated.Tear fluid was also collected using Schirmer’s strips and evaluated for soluble factors related to inflammation by multiplex ELISA.
Results :
The AI achieved an AUC of 0.736,accuracy of 86%,F1-score of 85.9%,precision of 85.6% and recall of 86.3%.Top 5 parameters used for classification by AI were microneuromas,immature and mature dendritic cells,presence of orthoptic issues and nerve fractal dimension parameter.Patients with higher grade of symptoms and signs showed increased corneal dendritic cells(cDC) density which was more pronounced in subjects with discordant symptoms to signs(group 4).Significantly higher proportion of microneuroma-like structures and cDC were seen in group 4.Higher levels of IL-17A were found in group with more discomfort.
Conclusions :
Correlating ocular surface pain and the disparity between signs and symptoms with IVCM and tear molecular factors can help clinicians improve diagnosis and provide targeted treatment for pain and coupled with AI can improve the diagnoses and help better customise treatment of ocular surface pain.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.