Relative importance of tear homeostatic signs for the diagnosis of dry eye disease

Abstract

Aim Disease misdiagnosis is more likely if standardised diagnostic criteria are not used. This study systematically examined the effect on diagnosing dry eye disease (DED), when tests for evaluating tear film homeostasis were included/excluded from a multi-test protocol. Method For 1427 participants across five sites, data for the full suite of diagnostic tests defined in the Tear Film and Ocular Surface Society Dry Eye Workshop II (TFOS DEWS II) Diagnostic Methodology report algorithm were evaluated; diagnostic sensitivity was calculated when individual signs were removed, and when different combinations of signs were required. Results Evaluating just one of the three TFOS DEWS II homeostatic signs resulted in between 12.3% and 36.2% of patients who met the DED diagnostic criteria not being assigned this diagnosis. While comprehensive ocular surface staining evaluation, including corneal, conjunctival and lid margin staining, in combination with symptoms had the highest sensitivity (87.7%) of the three markers, the sensitivity dropped to 44.6% if only corneal staining was evaluated. Omitting either non-invasive tear breakup time or osmolarity each dropped the sensitivity by <5%. The prevalence of DED was substantially reduced if a diagnosis required symptoms and two of the three signs to be present (by 43.7% to 61.2%) and by 65.9% if all three signs indicating a loss of tear film homeostasis were required. The analysis did not change significantly across differing severities of DED symptoms. Conclusions The TFOS DEWS II diagnostic algorithm of symptoms plus assessing for a tear film or ocular surface sign can be considered a robust and appropriate approach for DED diagnosis.