Abstract
Purpose :
To investigate inter-eye agreement in dry eye signs in patients with dry eye disease (DED) from the Dry Eye Assessment and Management (DREAM) study.
Methods :
Secondary analysis of data from the DREAM study, a multicenter randomized clinical trial evaluating the effect of omega-3 fatty acid supplementation on 535 participants with moderate-to-severe DED. DED signs including tear break-up time (TBUT) (seconds), anesthetized Schirmer’s test (mm/5 minutes), conjunctival staining, corneal staining, meibomian gland dysfunction (MGD), and tear osmolarity (mOsm/l) were measured at baseline, 3-, 6- and 12-months follow-up. Intraclass correlation coefficient (ICC), weighted kappa, mean absolute difference, and percent of patients with absolute inter-eye difference (AID) exceeding a threshold were used to assess inter-eye agreement. The worse eye (left or right) for each DED sign was determined at each visit to evaluate the consistency in membership of the worse eye overtime.
Results :
At baseline (Table 1), DED signs had moderate inter-eye agreement with ICC ranging from 0.50 (tear osmolarity) to 0.81 (corneal staining and Schirmer test scores) and weighted kappa of 0.58 for plugging and 0.62 for lid secretion. An AID of 2 points or greater was observed in 20% participants for conjunctival staining score, 31% for corneal staining score and 11% for MGD. An AID of 3 points or greater in TBUT was observed in 11% participants; AID greater than 5 points in Schirmer test was observed in 16% participants, and AID greater than 12 mOsms/L in tear osmolarity was observed in 31% participants. The eye with worse signs was not the same eye over 12 months in 63% (MGD) to 86% (corneal staining score) of participants (Table 2). Conjunctival staining and corneal staining scores consistently showed significant correlation in inter-eye difference with Pearson correlation coefficient of 0.21 at baseline and 0.24 at 6 and 12 months (all p<0.0001).
Conclusions :
In the DREAM study, participants demonstrated moderate inter-eye agreement with a substantial portion showing clinically significant inter-eye differences in each sign. The eye with worse DED signs was not the same eye in the majority of participants during follow-up. These findings suggest that it is valuable to consider both eyes in the design and statistical analysis for dry eye clinical trials.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.