Reliability of a new modified tear breakup time method: dry tear breakup time

Abstract

Background: Tear breakup time (TBUT) is a useful technique for diagnosing dry eye disease (DED). The conventional method of measuring TBUT using fluorescein strips adds supplemental normal saline (NS) to the tear film (wet tear breakup time, WBUT) but does not represent the actual state of the tear film. We introduced a new TBUT method-dry tear breakup time (DBUT)-and investigated its reliability.

Methods: We included DED patients with mild symptoms and a Schirmer test of 6 mm or more. Patients using tear substitutes and/or suffering from Sjogren's syndrome were excluded. For measuring DBUT, 1 μl of 5 % fluorescein solution was instilled on the end of an applicator using a micropipette. Once the fluorescein solution was dry, the applicator was sterilized using ethylene oxide gas, and the dried fluorescein was applied to the lower palpebral conjunctiva without supplemental NS. Agreement between the two tests and their sensitivity and specificity were analyzed.

Results: The average age of the patients was 32.1 ± 9.7 years (n = 124). The DBUT was lower than the WBUT, with a mean difference of -0.63 s. Both tests had a low statistically significant correlation with the Ocular Surface Disease Index. When the basis of definite diagnosis for DED was over grade 1 on the Oxford schema of staining, the cutoff value for the WBUT was 4.48 s and the sensitivity and specificity were 0.790 and 0.548 respectively. The area under the receiver operating characteristic (AUROC) curve was 0.609 (95 % confidence interval (CI): 0.517-0.695). When the cutoff value was 3.5 s for the DBUT, the sensitivity and specificity were 0.726 and 0.694 respectively, and the AUROC curve was 0.724 (95 % CI: 0.637-0.801).

Conclusions: DBUT had a higher correlation with symptoms than the conventional WBUT. When DED was diagnosed with corneal or conjunctival staining based on the Oxford schema of staining, DBUT had a better accuracy than the conventional WBUT. DBUT may replace WBUT after confirming studies, especially when surface damage is to be detected as opposed to tear flow or symptoms.