Central thickness measurement is a vital tool for diagnosing and monitoring corneal diseases, avoiding complications of corneal refractive procedures such as laser in situ keratomileusis (LASIK), and for treating glaucoma. Such measurements are also essential in keratoplasties, surgical procedures in which an opaque, opalescent, or damaged portion of the cornea is removed and replaced with a donor corneal graft.
The human cornea consists of five layers: the outer epithelium, Bowmann's membrane, the middle stromal (supportive tissue) layer, Descemet's membrane, and the inner endothelium (EC), which is made up of a single layer of cells attached to Descemet's membrane. The function of the EC is to maintain corneal hydration at ∼76%. If the EC stops working, the cornea hydrates and becomes cloudy, causing blurry vision. A new procedure known as Descemet's stripping automated endothelial keratoplasty (DSAEK) remedies this situation by using a thin button of donor corneal tissue with an intact EC layer to replace the back surface of the patient's cornea. Donor graft thickness and its uniformity are important factors for postoperative outcome. Consequently, donor tissue is always measured prior to DSAEK to determine the appropriate microkeratome (a precision surgical instrument) cutting head or the femtosecond cutting depth needed to achieve the desired tissue thickness. Furthermore, the residual corneal stromal bed is also measured to ensure that this thickness has been achieved.
The gold standard for measuring corneal thickness is ultrasound pachymetry, which provides central thickness measurement at one location. Because it is a contact technique requiring tissue indentation, pachymetry may result in errors of thickness measurement, epithelial damage, and spread of infection. Moreover, ultrasound pachymetry fails to recognize abnormal uniformities in corneal thickness because it only provides a measurement in one location. These drawbacks led us to develop an optical coherence tomography (OCT)-based system for evaluating donor corneas using a mechanical adaptor that accepts standard glass vials or plastic viewing chambers designed for plug-and-play use in a Zeiss Visante OCT1, 2 (see Figure 1). Typically operating in the near-IR, OCT performs imaging by measuring the magnitude and echo time delay of backscattered light.
The Florida Eye Bank tissue adaptor does not require modification of optical coherence tomography and can be easily plugged in and out without compromising the sterility of donor corneal tissue.1, 2
To validate the use of this technique against the current ultrasound standard, six donor corneas were prepared by an expert eye bank operator using the standard method for obtaining DSAEK precut tissues. Each corneal lenticule (i.e., slice) was prepared from the donor cornea using a 350μm-head microkeratome (Moria One, Moria S.A., Antony, France) and an artificial anterior chamber (AAC, Moria S.A., Antony, France) cushioned with Eusol-C. The chamber is used to support donor tissue and maintain adequate pressure for measurement and sectioning. We created each lenticule by making a full pass of the microkeratome blade, which resulted in a posterior lamellar donor and a free anterior cap. The cap was removed for use in lamellar keratoplasty (corneal graft), and the posterior lamellar tissue was stored over a period of 48h at 4°C in Eusol-C between measurements. We measured central thickness of the donor cornea before dissection and of the lenticule immediately and after 4, 24, and 48h post-cut, first by OCT (Visante, Carl Zeiss Meditec, Dublin, CA) and then by ultrasound pachymetry using a 20MHz probe (Hiscan, Optikon 2000 S.p.A., Rome, Italy) capable of measuring thickness values lower than 100μm. Measurements were performed five times by the same operator to assess repeatability of both methods.2
The mean central thickness before dissection was 647±35μm and 660±38μm (p=0.001), as measured by OCT and ultrasound pachymetry, respectively. Immediately after dissection, posterior lenticule central thickness values were 235±43μm and 248±44μm (p<0.001). No statistically significant changes in lenticule central thickness values (one-way analysis of variance, p>0.05) were assessed during the post-cut time course, as measured by both devices.2 Despite these subtle differences, the two systems provide tomography measurements that are highly correlated, supporting the adoption of OCT as an alternative to ultrasound pachymetry.
Non-contact OCT donor corneal tissue assessment is in clinical use at the Bascom Palmer Eye Institute and elsewhere in the world. Apart from circumventing the problems associated with ultrasound pachymetry, the same OCT is used to follow corneal transplant patients, eliminating conversion errors and concentrating donor and patient data in the same format, a must for the Electronic Medical Record system required by the US government. We have shown that OCT is a reliable alternative to ultrasound pachymetry for measuring posterior donor lenticule central thickness. As a next step, we plan to use the 3D-imaging capabilities offered by OCT to enhance donor tissue-screening procedures, with the ultimate goal of improving patient outcomes.
Derek Nankivil, Sonia Yoo, Jean-Marie Parel
Bascom Palmer Eye Institute
University of Miami
Miller School of Medicine
Derek Nankivil is the senior biomedical engineer at the Ophthalmic Biophysics Center of the Bascom Palmer Eye Institute.
San Giovanni-Addolorata-Britannico Hospital
1. S. Yoo, C. De Freitas, B. Lee, I. Nose, J. Pena, C. Triglia, S. Uhlhorn, S. Dubovy, W. Culbertson, J.-M. Parel, Anterior chamber OCT adaptor to assess intact and precut eye bank corneal buttons for PK and DSEAK, Assoc. Res. Vis. Opthalmol., 2009. E-Abstract 6211
2. D. Amato et al., Evaluation of a new method for the measurement of corneal thickness in eye bank posterior corneal lenticules using Anterior Segment Optical Coherence Tomography, Br. J. Ophthalmol. 95, pp. 580-584, 2011.