Epithelial mapping

Epithelial thickness has high diagnostic importance in detecting corneal abnormalities

Roibeard O’hEineachain

Posted: Thursday, March 1, 2018

Dan Z Reinstein, MD,

Mapping the corneal epithelium with layered tomographic imaging can be useful in determining whether refractive surgery candidates are at risk of postoperative ectasia, reports Dan Z Reinstein, MD, FRCSC, FRCOphth, FEBO, London, UK.

There are now currently several tomography devices available that can distinguish between the corneal epithelium and the underlying stroma with considerable accuracy. These include optical coherence tomography (OCT) and very high-frequency (VHF) digital ultrasound, Dr Reinstein told the XXXV Congress of the ESCRS in Lisbon, Portugal.

He noted that because ultrasound requires immersion of the eye and examination time takes a little longer, it is less popular than OCT. However, ultrasound has an inherent advantage over OCT in terms of accuracy.

“Even though the wavelength is longer than light, VHF digital ultrasound has higher accuracy than OCT for measuring the epithelium, and the reason for this is that the acoustic impedance differences at the level of epithelium and stroma are much higher than they are for light.”

In a study comparing the Artemis Insight 100 (ArcScan) VHF digital ultrasound device with the RTVue OCT (Optovue) device they found that there is close relation between the measurements of the two machines for measuring the central epithelium. However, the measurements diverged progressively as they extended more peripherally. He pointed out that although the difference was only a few microns, it was still enough to have clinical importance from a diagnostic perspective.

In another study using VHF digital ultrasound in 110 eyes with a mean central thickness of 53.4µm, the epithelial thickness was found to be not uniform throughout, but on average was 5.7µm thicker inferiorly than superiorly, and 1.2µm thicker nasally than temporally.

Dr Reinstein said this could be explained as a result of the eyelid’s tarsus exerting greater frictional force on the superior and temporal region of the cornea than the inferior nasal region when the eye blinks, which occurs 10,000 times a day. The eye balances out the chafing of the epithelium by stimulating its outward growth, thereby maintaining the cornea’s curvature.

Following laser refractive surgery, epithelial growth will partially compensate for the change in curvature by becoming thicker in the centre and thinner paracentrally following myopic ablations, and by thinning in the centre and thickening paracentrally following hyperopic ablations. The bulk of the compensatory epithelial changes occur within the first postoperative day and continue at a slower rate for three months, after which the corneal curvature remains stable.

In eyes with keratoconus, the same principle applies, in that the epithelium gets thinner at the cone but thickens around the cone. The result is that epithelial mapping will show a donut pattern that is highly characteristic of the condition. Epithelial mapping can therefore be useful in screening laser refractive surgery patients for subclinical keratoconus.

For example, in an eye with early keratoconus, epithelial mapping can detect whether a cone not detected by front surface topography is being masked by the epithelium. Conversely, in a topographically suspicious but actually normal cornea, epithelial mapping will show a normal epithelial pachymetry.

Dr Reinstein noted that he and his associates have developed an automated algorithm to distinguish between eyes with keratoconus and those without the condition, based on epithelial and stromal pachymetric profiles. They studied a wide range of variables and found six variables that could detect keratoconus with 99% specificity and 95% sensitivity without additional data.

In a study involving 136 forme fruste keratoconus suspects identified among 1,532 myopic LASIK candidates by tomography and topography, epithelial mapping with VHF digital ultrasound showed that only 22 (16%) were true keratoconus. The remaining 114 eyes underwent LASIK and none have subsequently developed ectasia.

The contribution of epithelial remodelling after corneal refractive surgery to the cornea’s final refraction can also lead to undercorrection in eyes that have undergone myopic LASIK and overcorrection in post-hyperopic eyes. The integration of epithelial thickness profiles into IOL power calculation formulae could further improve their accuracy, Dr Reinstein said.

Dan Z Reinstein: