September 04, 2019
5 min read

Measuring densitometry in PDEK eyes

The PDEK densitometry values were comparable to the corneal quality values seen in DMEK.

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

The cornea is a transparent ocular structure. The transmission and refraction of light through the cornea depend primarily on a highly specialized ultrastructure.

The corneal stroma is carefully structured to minimize light scatter with important contributions from the extracellular matrix with its proteoglycans, carefully spaced collagen fibrils and crystallin-expressing keratocytes. Along with ultrastructure, the corneal transparency is maintained by active endothelial pumps. Tear film, epithelium and ocular surface also contribute to the transparent nature of the cornea. Abnormalities in transparency can lead to edema, corneal haze and opacity. The clinical significance of loss of transparency is the reduction in visual acuity. Therefore, the ideal objective of any treatment of the cornea is to maintain its transparency.

Endothelial keratoplasty has evolved over the last two decades. Starting from posterior lamellar keratoplasty, newer techniques have been performed, such as Descemet’s stripping endothelial keratoplasty to Descemet’s membrane endothelial keratoplasty. Pre-Descemet’s endothelial keratoplasty is a recent addition to the list of procedures. Although the clinical effects in terms of visual acuity and complications have been widely reported, corneal quality after PDEK has not been widely discussed. In this column, we will discuss the surgical outcomes in terms of corneal quality after PDEK.

Pre-Descemet’s endothelial keratoplasty

PDEK was first introduced by Agarwal and colleagues in 2014 when the pre-Descemet’s layer along with Descemet’s membrane and endothelium was transplanted in a decompensated cornea. Our initial reports have been widely described in literature. The technique of graft harvesting by pneumatic dissection and the use of young donors have also been detailed in our previous reports. Under peribulbar anesthesia, the host Descemet’s endothelial complex is removed using a reverse Sinskey hook. The anterior chamber can be maintained by a trocar anterior chamber maintainer. Donor cornea can be harvested by pneumatic dissection using a 30-gauge needle. The graft is loaded into an IOL injector and injected into the anterior chamber. This is followed by unrolling the graft using saline or ophthalmic viscoelastic device. Once the graft is unrolled with correct side upward, an air bubble is injected below it to produce pneumatic adhesion to the overlying stroma. The patient is kept in the supine position for a short period before leaving the operating room. The usual postoperative medicines include antibiotic-steroid combination drops and lubricants in the first month. Steroids are required for at least 6 months on a tapering dose.


Corneal clarity and assessment

Corneal clarity is usually evaluated using slit lamp microscopy. Clinical evaluation is typically done by grading the cornea in relation to the visibility of underlying structures with good illumination. The clinical grading of corneal clarity is grade 0, clear; grade 1, trace haze, not obscuring the iris; grade 2, corneal opacity mildly obscuring the iris; grade 3, corneal opacity moderately obscuring the iris; and grade 4, leucomatous opacity with no view of the iris. However, the limitation of clinical grading is that it is subjective and subject to variations. There have been devices and methods that can be used to document or quantify corneal clarity. Anterior segment OCT, Scheimpflug imaging and even ultrasound biomicroscopy can be able to visualize corneal status.

Densitometry and Scheimpflug imaging

Densitometry is the quantitative measurement of optical density in light-sensitive materials. Density is usually measured by the decrease in the amount of light that shines through a transparent film; it is also called absorptiometry. The reader should understand that a densitometer does not measure the density; instead, it measures the reflectance. The reflectance is the measure of the amount of light reflected from a surface or film. Using the Scheimpflug imaging principle with the Pentacam (Oculus), we have measured or quantified the reflectance or backscatter of the cornea after PDEK. With the patient seated with a chin rest, 50 images were taken, centered on the corneal vertex. The working algorithm provided in the machine’s software has coded instructions to subdivide the cornea into four concentric zones: central zone from 0 mm to 2 mm; second zone extending from 2 mm to 6 mm; third zone or mid-periphery from 6 mm to 10 mm; and fourth or periphery from 10 mm to 12 mm. Corneal depth is divided as anterior layer (120 µm), central layer, posterior layer (60 µm) and total depth. Densitometry is expressed in grayscale units (GSU) (Figure 1).

linical picture and corresponding corneal densitometry
Figure 1. Clinical picture comparing post-PDEK eye (left eye) (b) and the fellow normal eye (right eye) (a) with the corresponding corneal densitometry values of fellow eye (c) and post-PDEK eye (d).

Source: Dhivya Ashok Kumar MD, FRCS, FICO, and Amar Agarwal MS, FRCS, FRCOphth

Backscattering in PDEK eyes

Overall, 21 eyes that underwent PDEK were examined with the Pentacam imaging system. When possible, the densitometry values were compared between the PDEK eyes and normal fellow eyes (control). The mean patient age was 56.4 ± 6 years; 10 patients were men and 10 were women. The mean time from the PDEK surgery was 17.7 ± 17 months (range, 6 to 56 months). The mean graft size was 7.6 ± 0.64 mm (range, 6.5 mm to 9 mm). The clinical slit lamp grading of the cornea in PDEK eyes was grade 4 in 19 eyes (90.4%) and grade 3 in two eyes (9.5%).


In total, the anterior cornea (40.7 ± 7.6 GSU) showed higher total densitometry values compared with the posterior cornea (20.9 ± 3.4 GSU) and the central corneal layers (25.3 ± 4.8 GSU). The least corneal backscattering was recorded in the posterior corneal layer in the 2- to 6-mm zone, about 17.1 ± 3.5 GSU. On interzone comparison between the central and peripheral cornea, there was a statistically significant difference between the 0- to 2-mm and 6- to 12-mm zones (P < .05). There was no correlation for endothelial cell density and central corneal thickness with densitometry. However, there was a significant correlation of best corrected visual acuity with corneal densitometry in the 0- to 2-mm (P = .016) and 6- to 12-mm (P = .009) zones. In eyes with BCVA of 20/20, there was no significant difference between PDEK and control eyes (Figure 2) in the total corneal densitometry in the entire depth of the cornea. There was a significant difference in densitometry values between inside the graft and outside the graft.

Clinical picture comparing post-PDEK eye and the fellow normal eye
Figure 2. Clinical picture comparing post-PDEK eye (right eye) (a) and the fellow normal eye (left eye) (b).


The PDEK eyes with vision of 20/20 showed no significant difference from the fellow normal eyes with clear corneas. The PDEK corneal densitometry in the central 0- to 2-mm and 2- to 6-mm zones was lower in comparison to the peripheral annulus. The PDEK densitometry values were comparable to the corneal quality values seen in DMEK. Nevertheless, large population studies with long follow-up may be required to quantify long-term effects on densitometry.

Disclosures: The authors report no relevant financial disclosures.