Predictive performance of corneal and lid margin sensitivity for dry eye disease: An investigator-masked, prospective, prognostic accuracy study

Abstract

Purpose

To evaluate the prognostic ability of non-contact esthesiometry corneal and lid margin sensitivity measurements in detecting symptoms and signs of dry eye disease, as defined by the global consensus TFOS DEWS II criteria.

Methods

A total of 87 community residents (58 females; mean ± SD age, 53 ± 16 years) were recruited in an investigator-masked, prospective, prognostic accuracy study. Dry eye symptomology, tear film parameters, and ocular surface characteristics were evaluated in a single clinical session, and non-contact esthesiometry corneal and lid margin sensitivity measurements performed by an independent masked assessor.

Results

Overall, 49 (56%) participants fulfilled the TFOS DEWS II criteria for dry eye disease, while 57 (66%) exhibited clinical symptoms, and 67 (77%) had positive signs. The prognostic abilities of corneal and lid margin sensitivity measurements were significantly greater than chance for the detection dry eye signs (both p ≤ 0.03), but not for symptoms or overall disease diagnosis (all p > 0.10). The Youden-optimal prognostic cut-offs for corneal and lid margin sensitivity thresholds were both ≥0.8 mbar for the detection of clinical dry eye signs. Lid margin sensitivity demonstrated marginally higher predictive performance than corneal sensitivity (C-statistic, 0.688 versus 0.658), and was significantly correlated with tear film stability, corneal, conjunctival and lid wiper staining (all p < 0.05).

Conclusions

Corneal and lid margin sensitivity demonstrated moderate prognostic utility for detecting clinical dry eye signs. Future research is warranted to investigate the utility of incorporating non-contact esthesiometry in the workup for dry eye disease and neurotrophic keratopathy.