Abstract
Purpose :
To correlate 3 different Ocular Surface Staining (OSS) scores: Oxford, NEI-CLEK and SICCA OSS and analyze the relationship between them and with Schirmer's test, tear break-up time (TBUT) and tear osmolarity (Osm) in patients with primary and secondary Sjögren’s syndrome (SS).
Methods :
This is a prospective, consecutive, transversal case series of patients with SS confirmed with salivary gland biopsy, serology and/or SICCA OSS according to 2012 classification criteria for SS. The dry eye evaluations included TBUT, Schirmer’s test without anesthesia, Osm, and the 3 different OSS scores (Oxford, NEI-CLEK and SICCA OSS). The analysis was performed with Spearman's correlation test, Kappa agreement test and Bland-Altman diagrams.
Results :
We evaluated 189 eyes of 95 patients: Mean age ± standard deviation (SD) was 53.38 ± 12.24 years. The OSS scores showed a significant very strong positive correlation and a fair agreement between them: Oxford vs NEI-CLEK: rho: 0.819, p<0.001, kappa:0.496. Oxford vs SICA OSS: rho: 0.862, p<0.001, kappa: 0.507. NEI CLEK vs SICCA OSS: rho: 0.847, p<0.001, kappa: 0.366. Bland-Altman scatterplots showed: Oxford vs NEI-CLEK: difference between tests was not related to the severity of the disease, but a mild tendency to increase differences in more severe cases. Oxford vs SICA OSS: a discrete increase in differences between tests with increasing severity of the staining. NEI CLEK vs SICCA OSS: a discrete tendency to increase differences between tests in moderated cases and reduced differences in mild and severe cases.
All OSS correlated significantly with most variables, Schirmer's test showed moderate correlation with SICCA OSS and Oxford. Osm showed moderate correlation with SICCA OSS and weak correlation with another OSS.
Conclusions :
The Spearman’s correlation test showed very strong significant positive correlation between all OSS scores. The Kappa analysis showed a fair agreement between all OSS scores. The Bland-Altman diagrams showed a valid limit of agreement between all OSS, the differences between tests were mostly random with a mild tendency to increase differences in more severe cases. The three staining scores are adequate to evaluate the ocular surface and its severity; we suggest further analysis of sensitivity and specificity
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.