The Detection of Keratoconus using Anterior Segment OCT

Abstract

Purpose : This pilot study aims to determine whether corneal arc length and central cross-sectional area measured using ocular coherence tomography (OCT) allows for the detection of keratoconus (KCN). Earlier detection of KCN would allow for earlier treatment, slowing the progression of corneal degradation. Earlier detection would also allow for improved screening for refractive surgery, and ultimately decrease the prevalence of end-stage KCN.

Methods : 18 eyes of healthy patients and 12 eyes of keratoconic patients between the ages of 20 and 80 years old were imaged using anterior segment OCT. Pregnant, nursing, and patients with other anterior segment pathology were excluded. The image slice through which the fixation light beam appeared was selected for analysis. Using imageJ, posterior corneal arc lengths and anterior corneal arc lengths were measured over a fixed 6 mm baseline (Figure 1). Both arc lengths were also modeled with 4^th degree polynomial regressions and the difference between their area under the curves was calculated, corresponding to the cross-sectional area of that particular cross section of cornea (Figure 2). In addition, orthogonal linear thicknesses were measured at the subjective thinnest point in the cornea and peripherally at the 6 mm outer bounds.

Results : Posterior corneal arc length was 6.23 +/- 0.03 mm and 6.34 +/- 0.11 mm for healthy and KCN eyes, respectively (p = 0.004). Anterior posterior corneal arc length was 6.16 +/- 0.02 mm and 6.22 +/- 0.06 mm (p = 0.005), respectively. The cross-sectional area was 3.25 +/- 0.21 mm² in healthy eyes and 2.91 +/- 0.40 mm² in KCN eyes (p = 0.01). Comparing the thinnest sections resulted in a thickness of 512 +/- 36 μm in healthy eyes and 417 μm +/- 97 um in KCN eyes (p = 0.01).

Conclusions : There are statistically significant differences noted between both posterior and anterior corneal arc lengths as well as corneal cross-sectional areas between healthy and KCN eyes. These results correlate with simple corneal thickness measurements. Further analysis is needed to determine whether these novel metrics (corneal arc lengths and corneal cross-sectional area) can detect KCN earlier than standard videokeratography.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.

 

Figure 1. ImageJ analysis with measurements superimposed.

 

Figure 2. Polynomial model for corneal arc curvature.