Classification of dry eye disease subtypes

Abstract

Purpose The current subclassifications of dry eye disease (DED) are aqueous deficient (ADDE) and evaporative (EDE) forms, but there lacks consistency in the clinical characteristics used to define each of these. This study used clinical data to inform cut-off values for the subclassification of ADDE and EDE, to allow more consistent study of the epidemiology of both DED subtypes. Methods The study enrolled 261 residents from the UK, extracted from a cohort with demographics representing the population (mean 42.4 ± 18.7 years, 56 % females). The TFOS DEWS II diagnostic criteria were used to identify those with DED. Meibomian gland loss/drop-out (from meibography), lipid layer thickness (LLT − from interferometry graded on the Guillon-Keeler scale), and tear meniscus height (TMH − Keratograph 5M) along with tear evaporation (Delfin Vapometer) were used to characterise the subclassification. The Dry Eye Risk Factor Survey was used to assess risk factors associated with each DED subtype. Results Compared to individuals who were not diagnosed with DED, EDE was characterized by signs of meibomian gland loss of > 28 %, LLT grade < 3 and tear evaporation > 46 g/m2/h. In contrast, ADDE was best characterized by a reduced TMH < 0.2 mm. Based on these criteria, the prevalence of ADDE was 6.2 %, EDE was 64.2 %, and 11.1 % exhibited features of both ADDE and EDE, with 18.5 % unclassified despite having a DED diagnosis. Contact lens wear and computer use were risk factors for ADDE (p < 0.05), whereas age was a positive risk factor for EDE (p < 0.01). Meibomian gland loss (occurring in 27.9 %) was the most commonly observed sign in EDE. Conclusions Data driven-classification of DED confirms that the evaporative form is most prevalent and identified that in a generalisable UK population, ADDE alone occurs only in approximately 1 in 16 cases of DED.