Risk factors for the development of keratoconus in young adults

Abstract

Purpose : To report the incidence and investigate risk factors for the development of keratoconus in a young adult cohort with consecutive Scheimpflug imaging over an 8-year period.

Methods : Participants of the Raine Study generation 2 (Gen2) birth cohort underwent an eye examination at the 20-year (age range 19-22 years) and 28-year follow-ups. Scheimpflug imaging (Pentacam, Oculus, Wetzlar, Germany) was performed at each follow-up. Keratoconus was defined as a Belin/Ambrósio Enhanced Ectasia Display D (BAD-D) score >2.6 in either eye. Data from the worst eye was used for analysis, except for conjunctival ultraviolet autofluorescence (CUVAF, a marker of ocular sun exposure) data where the sum area of both eyes was used. Participants were excluded if they had keratoconus at the 20-year follow-up, had a history of significant eye trauma, wore orthokeratology lenses or did not have Scheimpflug imaging at both follow-ups. Logistic regression was used to investigate potential risk factors for keratoconus incidence over 8 years of follow-up.

Results : Of the 1295 participants who had Scheimpflug imaging at the 20-year follow-up, 685 (53%) had Scheimpflug imaging at the 28-year follow-up and 677 (52%) were eligible for analysis (52% female). There were 20 (3.0%) new cases of keratoconus. The mean (range) worst eye BAD-D score in the incident keratoconus group was 1.70 (0.20 to 2.57) and 3.45 (2.63 to 6.15) at the 20- and 28-year follow-ups, respectively. Keratoconus incidence was not associated with spherical refraction, visual acuity or self-reported history of asthma, eczema or asthma (p>0.05) on univariable analysis. Greater astigmatism, steeper corneal curvature and a thinner central corneal at the 20-year follow-up were associated with higher risk of keratoconus, but we focused on the BAD-D score as a derivative of these variables. In a multivariable model including sex, CUVAF area and BAD-D score at the 20-year follow-up, male sex (odds ratio [OR]=3.84, 95% confidence interval [95%CI]: 1.30, 1.1.29) and higher BAD-D score (OR=14.7, 95%CI: 5.93, 36.6), but not CUVAF area (OR=0.98, 95%CI: 0.96, 1.00), were significantly associated with higher risk of keratoconus.

Conclusions : The incidence of keratoconus was 3% over 8 years in this, predominantly Caucasian, young adult cohort in Perth, Australia. Sex and BAD-D score at the 20-year follow-up were significantly associated with greater risk of keratoconus.

This is a 2021 ARVO Annual Meeting abstract.