Cover Stories | Jul 2005
Automated Dissection of the Donor in DLEK
Does this procedure improve visual outcomes?
Kenneth Mark Goins, MD
In the first US clinical series using deep lamellar endothelial keratoplasty (DLEK), Mark A. Terry, MD, of Portland, Oregon, reported that three of eight eyes (38%) had a BCVA of 20/40 or better (range, 20/30 to 20/70), with three of four eyes being 20/40 or better at 12 months (range, 20/30 to 20/60).1 Unfortunately, none of the eyes in this clinical series achieved a final visual acuity of 20/20. One may conclude that the final visual acuity after DLEK may be reduced compared with conventional penetrating keratoplasty (PKP), despite the presence of lower astigmatism. The etiology of the decline in vision is unclear. However, several investigators have determined that higher-order wavefront aberrations do not play a significant role, as measured with the 3-D Wave (Marco, Jacksonville, FL) by skiascopy,2 the Ladarwave Customcornea Wavefront System (Alcon Laboratories, Inc., Fort Worth, TX) through the use of the Hartmann-Shack method,3 or the Tracey iScan (Tracey Technologies, Bellaire, TX) by means of ray tracing.4
In terms of the different modalities of wavefront analysis, skiascopy and ray tracing seem to achieve more reproducible results following DLEK, although Hartmann-Shack instruments, which rely on an eye's outgoing wavefront information, have more difficulty providing useful measurements. Therefore, DLEK may have a significant effect on light rays exiting the visual system. This finding is substantiated by Baratz et al,5 who reported a statistically significant increase in reflected light in the anterior and posterior cornea by scatterometry in DLEK corneas as compared with PKP corneas (P=.04 anteriorly and P=.005 posteriorly). Current evidence suggests that improving the quality of the interface is important in DLEK surgery. Automated creation of the endothelial lenticules may be the answer for providing a consistently smooth interface in DLEK, just as automated microkeratomes have improved the quality of LASIK. My colleagues and I hope that automated dissection of the donor tissue may reduce reflected light at the interface after DLEK.
MICROKERATOMES FOR DONOR
Whenever possible, my colleagues and I use the Moria Automated Lamellar Therapeutic Keratectomy microkeratome system (ALTK; Moria, Antony, France). It consists of the LSK One Microkeratome (Moria), which is manually driven by the surgeon. Multiple microkeratome heads may be used to achieve dissections of various thicknesses, ranging from 130 to 350µm (130, 150, 250, 300, and 350µm). The Moria ALTK artificial anterior chamber does require a donor scleral rim that is symmetrically greater than 16mm in diameter to provide proper vacuum during the microkeratome pass. When a small, asymmetrically cut donor scleral rim is present, a manual dissection system such as the Bausch & Lomb (Rochester, NY) artificial anterior chamber may be used (Figure 1).
In our clinical series, we used the 350-µm head of the Moria ALTK system most often to perform DLEK. We achieved a range of anterior corneal dissection of between 350 and 440µm. Although surgical time is markedly reduced as compared with manual dissection techniques in DLEK, inadvertent microkeratome-related complications may arise when using the Moria ALTK. This problem may occur secondary to a loss of vacuum during the microkeratome's pass or due to a thin preoperative pachymetry measurement (Figure 2). This observation re-emphasizes the importance of conducting intraoperative pachymetry during DLEK and of having access to multiple microkeratome heads in order to alter the microkeratome's depth to accommodate the corneal thickness. To avoid these potential complications, some surgeons place the donor rim into the artificial anterior chamber with the endothelial side up, thereby creating thinner donor endothelial lenticules. However, the long-term effect on endothelial cell counts has not been reported.
Initial findings from the study conducted at our institution suggest that automated dissection of the donor in DLEK does not result in a statistically significant difference in (1) BCVA, (2) corneal astigmatism as measured with the Humphrey Atlas Topography System (Carl Zeiss Meditec Inc., Dublin, CA), (3) measurement of higher-order aberrations with the Tracey iScan, or (4) measurement of the interface's opacity with the Pentacam (Oculus, Inc., Lynnwood, WA) when compared with manual dissection (Table 1).6 Our previous experience in lamellar keratoplasty may account for the lack of statistical significance (ie, surgeons more experienced in lamellar keratoplasty may not necessarily benefit from the use of automated microkeratomes in DLEK). Conversely, lamellar keratoplasty surgeons who are just beginning their careers may find the use of automated microkeratomes more beneficial than reported in our clinical series.
Automated dissection of the donor cornea may not enhance visual outcomes after DLEK as desired, but it does reduce surgical time and stress in most cases. Despite these positive attributes of automated dissection, the adequate preoperative screening of donor corneas (ie, pachymetry, scleral-rim diameter, and past ocular history) is imperative to avoid microkeratome complications. It has been suggested that eye-bank–donor screening using corneal topography may improve the detection of unsuitable corneas for keratoplasty.7 Such screening may become increasingly important if there is a shift toward eye banks' providing endothelial lenticules for DLEK surgeons.
Kenneth Mark Goins, MD, is Associate Professor of Ophthalmology at the University of Iowa Hospitals in Iowa City. He states that he holds no financial interest in any of the products or companies mentioned herein.
Dr. Goins may be reached at (319) 356-2861;
1. Terry, MA, Ousley, PJ. Replacing the endothelium without corneal surface incisions or sutures. The first United States clinical series using the deep lamellar endothelial keratoplasty procedure. Ophthalmology. 2003;110:755-764.
2. John T. Wavefront analysis of penetrating keratoplasty vs. deep lamellar endothelial keratoplasty. Poster presented at: The AAO Annual Meeting; October 25, 2004; New Orleans, LA.
3. Armour, RL, Wertheim, MS, Mathers WD, et al. Deep lamellar endothelial keratoplasty (DLEK) reduces total and higher order aberrations. Invest Ophthalmol Vis Sci. 2005;46:E-Abstract 2708.
4. Coombs JM, Goins KM, Sutphin JE. Characterization of wavefront aberrations in patients who underwent deep lamellar endothelial keratoplasty (DLEK). Invest Ophthalmol Vis Sci. 2005;46:E-Abstract 2714.
5. Baratz KH, Nau CB, Hodge DO, Bourne WM. A prospective, randomized study of deep lamellar endothelial keratoplasty versus penetrating keratoplasty: early results. Invest Ophthalmol Vis Sci. 2005;46:E-Abstract 2703.
6. Goins KM, Sjoberg SA, Gonzales M, Sutphin JE. Comparison of the visual outcome after manual and automated donor keratectomy in deep lamellar endothelial keratoplasty (DLEK). Invest Ophthalmol Vis Sci. 2005;46:E-Abstract 2691.
7. Ousley PJ, Terry MA. Use of a portable topography machine for screening donor tissue for prior refractive surgery. Invest Ophthalmol Vis Sci. 2002;43:E-Abstract 150.