Secondary Logo

Journal Logo

Evaluation of a Microkeratome-based Limbal Harvester Device for Limbal Stem Cell Transplantation

Behrens, Ashley M.D.; Shah, Samir B. M.D.; Li, Li M.D.; Côté, Mary A. M.D.; Liaw, Leacky L.-H. M.S.; Sweet, Paula M. M.T.; McDonnell, Peter J. M.D.; Chuck, Roy S. M.D., Ph.D.

Basic Investigations

Purpose. To assess the cut quality and reproducibility using a novel microkeratome-based limbal harvester.

Methods. An enlarged microkeratome head and stainless steel blades were coupled with a nitrogen gas–driven turbine (15,000 blade oscillations/min) of a microkeratome. A large, 16-mm-diameter suction ring was attached to the globe. A lamellar sclerokeratectomy using head depths of 170 and 200 μm was performed in human donor research eyes. Obtained lenticule thickness was measured by ultrasound pachymetry and the bed size by planimetry. Histologic and scanning electron microscopy (SEM) analyses of the samples were performed.

Results. Central lenticule thickness was 294 μm (standard deviation [SD] 37) for the 170 head and 277 μm (SD 91) for the 200 head (p = 0.720). Lenticule diameter was larger in the horizontal meridian using the 170 head (12.8 mm [SD 0.8] vs. 11.9 mm [SD 0.7], p = 0.028), but similar in vertical meridian (12.0 [SD 0.6] versus 11.4 mm [SD 0.6], p = 0.093). Histology showed a multilayer epithelial cell pattern at the lenticule periphery. The SEM showed a smooth cut surface in both the stromal bed and the lenticule.

Conclusion. Cut reproducibility and quality are similar to those found using standard microkeratomes for corneal lamellar cuts. This system ensures, in a straightforward way, the presence of epithelial cells in the edges of a mechanical sclerokeratectomy for limbal stem cell transplantation.

From the Department of Ophthalmology (A.B., S.B.S., L.L., M.A.C., P.M.S., P.J.M, R.S.C.) and the Beckman Laser Institute and Medical Clinic (L. L.-H. L.), University of California, Irvine, California, U.S.A.; and the Department of Ophthalmology, Doheny Eye Institute, University of Southern California, Los Angeles, California, U.S.A. (A.B., P.J.M.).

Submitted April 18, 2001.

Revision received September 12, 2001.

Accepted September 14, 2001.

Address correspondence and reprint requests to Dr. R. S. Chuck, Department of Ophthalmology, University of California Irvine, 118 Med Surge I, Irvine, CA 92697, U.S.A.

The first studies proposing a pericorneal network responsible for corneal epithelial cell maintenance were published three decades ago. 1 Several authors have supported the hypothesis of a limbal stem cell system, based on the indirect evidence obtained from clinical experience and experimental work. 2–6 Epithelial cell behavior also has been extensively studied to evaluate the dynamics of mitosis and migration. 7,8

Limbal stem cell transplantation was first reported in 1989 by Kenyon and Tseng. 9 Results for the autologous transplantation of the limbal area in severe ocular surface disorders have been promising. 10,11 In addition, allograft transplantation of these cells has been attempted, especially when a healthy donor tissue is not available in the contralateral eye. 12–15 Although a high rate of success has been achieved in these difficult cases, one of the major problems encountered is graft rejection. Some authors have proposed the use of donor tissue from family members to increase graft–host compatibility. 16

The surgical technique for limbal stem cell transplantation, however, is time consuming. The common procedure is to obtain the donor tissue from fresh whole donor globes. Other authors have designed instruments to obtain the limbal stem cells from corneoscleral discs. 17 In both modalities, the freehand donor harvest of the limbal area is laborious and requires some experience to preserve the stem cells.

We have developed a mechanical system to perform a superficial cut of the sclerocorneal surface to obtain a partial-thickness lenticule containing a rim of sclera. This may ensure the presence of sufficient limbal area over 360° to obtain the stem cells required. The device uses a standard microkeratome turbine to perform the cut.

The purpose of this study is to assess the cut reproducibility with this novel system using fresh human donor globes.

Back to Top | Article Outline


Donor Tissue

After approval by the Institutional Review Board of the University of California, Irvine, 26 fresh human donor globes not suitable for corneal transplantation were obtained from the Donor Network of Arizona. The eyes were received in a cool, moist chamber from the eye bank. As exclusion criteria we considered a central corneal thickness of ≥1,020 μm, corneas with evident stromal or limbal scars, or a marked surface irregularity. Sixteen eyes therefore were included in the study and divided in 2 groups to accommodate 2 different microkeratome head thicknesses (170 and 200 μm, n = 8 each). Globes were randomly assigned to either group on availability.

Back to Top | Article Outline

Mechanical Device

This microkeratome system allows partial-thickness corneoscleral lenticules to be cut from whole globes. The instrument consists of an enlarged microkeratome head coupled to the nitrogen gas–driven turbine of a commercially available microkeratome (LSK One; Moria/Microtech, Doylestown, PA, U.S.A.). The turbine action moves the blade inside the head at approximately 15,000 oscillations per minute. The head is designed for a manual translation across the cornea. It fits into parallel guiding tracks of the suction ring to maintain a constant height and centration of the head during the pass. The suction ring is connected to the pump of the microkeratome, but is larger than standard suction rings. Its central opening has a diameter of 16 mm, allowing inclusion of the entire corneal and some of the scleral surface of the donor globe in the pass.

Two different heads are available: 170 and 200 μm. This number is assigned according to the measured distance (in micrometers) of the blade's cutting edge to the applanation plate built into the head. A similar cut depth is expected, including the presence of basal epithelial cells of the limbal area in the lenticule. The head also uses custom-made, 16-mm stainless steel blades to match the width of the head.

Back to Top | Article Outline

Surgical Procedure

The globes were pressurized by inserting a 25-gauge needle connected to a bottle of saline through the optic nerve. Care was taken to maintain an intraocular pressure within 15–25 mm Hg. The intraocular pressure was measured before application of the suction ring with a pneumatonometer (Modular One; Mentor O&O, Norwell, MA, U.S.A.), and the infusion bottle height was set accordingly. The same surgeon operated on all eyes in a right-to-left head translation modality to avoid bias related to hand dominance and experience. The suction ring was applied to the globe surface and centered, taking the limbus as reference. The vacuum was activated and manual pressure was transferred to the ring until adequate suction was achieved. A surgical microscope (Ophthalmic 900S; Moeller-Wedel Microsurgical, Mason, OH, U.S.A.) was used to monitor the pass. Proparacaine hydrochloride 0.5% (Alcaine; Alcon, Ft. Worth, TX, U.S.A.) was dropped onto the exposed corneoscleral surface (to avoid saline), and the activated turbine with its head was passed along the suction ring without stopping. The corneoscleral lenticule was obtained under microscopic visual control during the pass (Fig. 1).

FIG. 1.

FIG. 1.

Back to Top | Article Outline

Lenticule Dimensions

To assess the cut reproducibility, physical dimension measurements were recorded. The lenticule thickness was determined in four quadrants and in the center using an ultrasound pachymeter (Ophthasonic A Scan-Pachometer III; Mentor O&O), with an arbitrary vertical and horizontal line intersecting at the center of the cornea as a reference. The horizontal line corresponded to the direction of the microkeratome pass. The quadrants measured on this line were at the right side of the center (beginning of the microkeratome pass) or at the left side of the center (end of the microkeratome pass). The superior and inferior quadrants were measured on the vertical line above or below the horizontal line, respectively. Central epithelial removal in an approximately 9-mm area was performed to avoid bias caused by irregular postmortem epithelial thickness. The thickness was recorded before and after the cut, and the lenticule thickness calculated by subtraction.

For lenticule diameter measurements, digital macrophotography (Olympus 3030; Olympus, Tokyo, Japan) of the residual bed in the whole eye at high-resolution settings (2,048 × 1,536 pixels) were taken. The images were uploaded into digital imaging software (Scion Image; Scion Corp., Frederick, MD, U.S.A.) and, after calibration, horizontal and vertical meridian diameters were recorded.

Back to Top | Article Outline

Tissue Preservation, Histology, and Scanning Electron Microscopy

To detect lenticule thickness changes after immersion in Optisol (Bausch & Lomb Surgical, Irvine, CA, U.S.A.), the lenticules were preserved in vials containing Optisol for 4 days at 4°C. Thickness measurements were performed by placing the lenticules in a plastic dome with a known thickness. The domes were made from the cut end of a Falcon test tube. Calculations were made by subtraction, as previously described.

For histologic determinations, lenticules were fixed in 10% buffered paraformaldehyde, paraffin embedded, and cut into 5-μm sections for hematoxylin and eosin staining. For scanning electron microscopy (SEM) analysis, specimens were immersed in osmium tetroxide, dehydrated with graded alcohols, and dried using increasing concentrations of hexamethyldisilazane. Samples were gold sputtered and examined under a scanning electron microscope (Philips XL 30; Philips, Limeil-Brevannes, France).

Back to Top | Article Outline

Statistical Analysis

We used StatsDirect 1.7.4 statistical software (CamCode, Ashwell, U.K.) for analysis. Descriptive statistics (mean, standard deviation [SD], minimum and maximum value) were performed for continuous variables. Comparisons between groups were performed using nonparametric tests (Mann-Whitney U for unpaired samples, Wilcoxon's signed ranks test for paired samples). A p value ≤0.05 was considered statistically significant.

Back to Top | Article Outline


The demographic and morphometric data obtained from the donor globes are shown in Table 1.



The instrument was easy to use. No problems were found during the surgery. The total procedure time was approximately 30 seconds, 15 seconds of which were related to the microkeratome pass (Fig. 1). A slight additional force had to be imparted to the suction ring held with the left hand (vertical direction) and to the microkeratome head with the right hand (horizontal direction) compared with a standard microkeratome cut. This is because of the greater size of the lenticule, and therefore a larger applanation area and a higher resistance to the cut was noticed.

Loss of suction during the pass was observed in three cases (38%) using the 170 head, and in four cases (50%) using the 200 head. However, no deviation of the thickness pattern was noted in these cases. The loss of suction was compensated for by a higher manual force transferred to the ring. No central buttonholes were observed in our series.

The results of lenticule dimension measurements are shown in Table 2. Comparing lenticule sizes using the two heads, the horizontal (in the same direction of the pass) diameter was significantly larger using the 170 head (p = 0.028). However, in the vertical diameter, the measurements were somewhat similar (p = 0.093). Regarding thickness, similar values were obtained using both heads at the center (p = 0.720), at the beginning of the pass (p = 0.943), at the end of the pass (p = 0.243), superiorly (p = 0.075), or inferiorly (p = 0.180). Moreover, no significant differences between or within groups were observed comparing the central thickness with the peripheral quadrants.



After Optisol preservation, the lenticules tend to thin. A mean central thickness of 215 μm (SD 42) was obtained for the 170 head and 182 μm (SD 55) for the 200 head. These differences were significant compared with the values obtained immediately after the surgery (p = 0.016 and p = 0.031, respectively).

The histologic results showed the presence of a multilayer epithelial cell pattern at the lenticule periphery, at the limbal area. This may indirectly reflect the presence of intact stem cells in the basal region. The results were similar in both vertical and horizontal meridians (Fig. 2). SEM showed a smooth cut surface in both the stromal bed and the lenticule. A transitional zone in the corneoscleral area was observed, with an evident change of pattern in the orientation of the collagen fibers revealing the cut of the scleral skirt (Fig. 3).

FIG. 2.

FIG. 2.

FIG. 3.

FIG. 3.

Back to Top | Article Outline


Lamellar corneal surgery advanced exponentially with the invention of the microkeratome by Barraquer. 18 This device has been shown to be an accurate tool in obtaining lamellar cuts of the cornea compared with the freehand technique. 19–22 Several changes in design have been seen in the past few years, especially with the advancement of the laser-assisted in situ keratomileusis techniques. 23–25

With the development of new strategies of immunosuppression and the refinement of donor tissue compatibility screening, allograft procedures for limbal stem cell transplantation are achieving greater success rates. 26–30 One of the technical concerns in the surgical procedure is the limbal tissue harvest. This is a time-consuming process and certainly inaccurate, because the cut depth can be estimated only visually during manual harvest. We have designed a new tool, taking advantage of new microkeratome technology, to perform a mechanical limbal lamellar dissection. 31 This ensures a reproducible cut in every donor eye. According to our results, the variability of the lenticule thickness is very similar to that previously reported in corneal lamellar cuts for refractive surgery. 19–21,23,32 Furthermore, after short term-storage in Optisol, the corneoscleral lamellar lenticules tend to become slightly thinner, making them potentially useful for lamellar corneal transplantation.

The lenticule size was slightly larger in the horizontal meridian using the 170 head. No differences were observed in the vertical meridian. The smallest vertical diameter obtained was 10.8 mm in one eye, which is likely the minimum diameter planned for this type of transplant (11–13 mm). 33,34 However, in the 170 head group, the smallest diameter obtained, 11.1 mm, was again in the vertical meridian. The means of both groups show values significantly above 11.4 mm, which guarantees a size large enough to include the limbus in average eyes.

Regarding the lenticule thickness, we observed very similar results using both trial microkeratome heads. The mean thickness was close to one half of the total corneal thickness, which is desirable for ocular surface reconstruction because thin lenticules are preferred for this type of surgery. The beveled edges of the limbal cut also might facilitate the apposition of the transplant over the recipient surface as opposed to those obtained with manual dissection.

Loss of suction was frequently observed in our preliminary series. However, it was easily compensated for by exerting additional pressure to the suction ring on the globe, without producing significant changes in lenticule thickness or diameter. A short learning curve is required to become familiar with the cutting procedure techniques using this modified microkeratome.

The histologic slides demonstrated the presence of epithelial cells in the area where limbal stem cells are thought to be present. Although we did not measure the viability of these cells because of the prolonged postmortem time of the donors, they appear intact. The thermomechanical damage induced by the blade oscillation is negligible. However, further studies are being conducted with vital stains to show the degree of deterioration, if any, to the limbal epithelial area. Additional culturing of the limbal epithelial cells obtained from the lenticules are required to determine cell phenotype and clone formation activity.

The epithelial cells appeared reduced in number in the limbal area. We believe that the reduced epithelial adherence and cell loss are caused by the long preservation time of the globes (5–6 days) in a moist chamber. Over time, epithelial cells are more easily detachable, and the total population may be affected. In a clinical setting, the globes should be processed earlier to avoid this problem. We were able to observe the corneoscleral transitional area by SEM, which demonstrated the change in collagen fiber orientation when the cut travels into the limbal area. The quality of the dissection was very good up to the edges of the cut.

More recently, total keratolimbal transplantations to correct ocular surface disorders have been proposed. 33–38 Although the results are encouraging, some improvement is needed in the control of immune rejection. Once this is accomplished, this instrument potentially can be a major aid in the efficient procurement of regular and reproducible lenticules for such transplantations.

The system evaluated here represents a novel instrument for surface reconstruction. Further applications of this enlarged keratectomy are forthcoming. Specialized training other than in basic microkeratome use is not required. Compared with traditional hand dissecting techniques, lamellar dissection of the corneoscleral lenticule with this instrument was performed easily and quickly, in a procedure lasting no longer than 30 seconds. The cut surface was very smooth and the edges beveled, allowing better apposition of the graft surface with the diseased recipient tissue. We believe these advantages surpass the acquisition costs of the system. Clinical studies are underway to assess the relevance of the use of this device in patients with severe ocular surface disease.

Back to Top | Article Outline


1. Davanger M, Evensen A. Role of the pericorneal papillary structure in renewal of corneal epithelium. Nature 1971; 229: 560–1.
2. Cotsarelis G, Cheng SZ, Dong G, et al. Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells. Cell 1989; 57: 201–9.
3. Lehrer MS, Sun TT, Lavker RM. Strategies of epithelial repair: modulation of stem cell and transit amplifying cell proliferation. J Cell Sci 1998; 111: 2867–75.
4. Chen JJ, Tseng SC. Corneal epithelial wound healing in partial limbal deficiency. Invest Ophthalmol Vis Sci 1990; 31: 1301–14.
5. Dua HS, Forrester JV. The corneoscleral limbus in human corneal epithelial wound healing. Am J Ophthalmol 1990; 110: 646–56.
6. Dua HS, Azuara-Blanco A. Limbal stem cells of the corneal epithelium. Surv Ophthalmol 2000; 44: 415–25.
7. Thoft RA, Friend J. The X, Y, Z of corneal epithelial maintenance. Invest Ophthalmol Vis Sci 1983; 24: 1442—3.
8. Beebe DC, Masters BR. Cell lineage and the differentiation of corneal epithelial cells. Invest Ophthalmol Vis Sci 1996; 37: 1815–25.
9. Kenyon KR, Tseng SC. Limbal autograft transplantation for ocular surface disorders. Ophthalmology 1989; 96: 709–22.
10. Basti S, Rao SK. Current status of limbal conjunctival autograft. Curr Opin Ophthalmol 2000; 11: 224–32.
11. Tsai RJ, Li LM, Chen JK. Reconstruction of damaged corneas by transplantation of autologous limbal epithelial cells. N Engl J Med 2000; 343: 86–93.
12. Tsai RJ, Tseng SC. Human allograft limbal transplantation for corneal surface reconstruction. Cornea 1994; 13: 389–400.
13. Coster DJ, Aggarwal RK, Williams KA. Surgical management of ocular surface disorders using conjunctival and stem cell allografts. Br J Ophthalmol 1995; 79: 977–82.
14. Tsubota K, Satake Y, Kaido M, et al. Treatment of severe ocular-surface disorders with corneal epithelial stem-cell transplantation. N Engl J Med 1999; 340: 1697–1703.
15. Dua HS, Azuara-Blanco A. Allo-limbal transplantation in patients with limbal stem cell deficiency. Br J Ophthalmol 1999; 83: 414–9.
16. Rao SK, Rajagopal R, Sitalakshmi G, et al. Limbal allografting from related live donors for corneal surface reconstruction. Ophthalmology 1999; 106: 822–8.
17. Mannis MJ, McCarthy M, Izquierdo Jr. L Technique for harvesting keratolimbal allografts from corneoscleral buttons. Am J Ophthalmol 1999; 128: 237–8.
18. Barraquer JI. Queratomileusis para la corrección de la miopía. Arch Soc Am Oftamol Optom 1964; 5: 27–48.
19. Behrens A, Seitz B, Langenbucher A, et al. Evaluation of corneal flap dimensions and cut quality using a manually guided microkeratome. J Refract Surg 1999; 15: 118–23.
20. Behrens A, Seitz B, Langenbucher A, et al. Evaluation of corneal flap dimensions and cut quality using an automated microkeratome. J Refract Surg 2000; 16: 83–9.
21. Kim YH, Choi JS, Chun HJ, et al. Effect of resection velocity and suction ring on corneal flap formation in laser in situ keratomileusis. J Cataract Refract Surg 1999; 25: 1448–55.
22. Rasheed K, Rabinowitz YS. Superficial lamellar keratectomy using an automated microkeratome to excise corneal scarring caused by photorefractive keratectomy. J Cataract Refract Surg 1999; 25: 1184–7.
23. Behrens A, Langenbucher A, Kus MM, et al. Experimental evaluation of two current-generation microkeratomes: Hansatome® and Supratome®. Am J Ophthalmol 2000; 129: 59–67.
24. Walker MB, Wilson SE. Lower intraoperative flap complication rate with the Hansatome microkeratome compared with the Automated Corneal Shaper. J Refract Surg 2000; 16: 79–82.
25. McDonnell PJ. Emergence of refractive surgery. Arch Ophthalmol 2000; 118: 1119–20.
26. Terry MA. The evolution of lamellar grafting techniques over twenty-five years. Cornea 2000; 19: 611–6.
27. Shimazaki J. The evolution of lamellar keratoplasty. Curr Opin Ophthalmol 2000; 11: 217–23.
28. Shimazaki J, Kaido M, Shinozaki N, et al. Evidence of long-term survival of donor-derived cells after limbal allograft transplantation. Invest Ophthalmol Vis Sci 1999; 40: 1664–8.
29. Tsubota K. Ocular surface management in corneal transplantation: a review. Jpn J Ophthalmol 1999; 43: 502–8.
30. Tsai RJ, Tseng SC. Human allograft limbal transplantation for corneal surface reconstruction. Cornea 1994; 13: 389–400.
31. Chuck RS, Behrens A, McDonnell PJ. Microkeratome-based limbal harvester for limbal stem cell transplantation. Am J Ophthalmol 2001; 131: 377–8.
32. Yi WM, Joo CK. Corneal flap thickness in laser in situ keratomileusis using an SCMD manual microkeratome. J Cataract Refract Surg 1999; 25: 1087–92.
33. Vajpayee RB, Thomas S, Sharma N, et al. Large-diameter lamellar keratoplasty in severe ocular alkali burns: technique of stem cell transplantation. Ophthalmology 2000; 107: 1765–8.
34. Holland EJ. Epithelial transplantation for the management of severe ocular surface disease. Trans Am Ophthalmol Soc 1996; 94: 677–743.
35. Shimmura S, Ando M, Shimazaki J, et al. Complications with one-piece lamellar keratolimbal grafts for simultaneous limbal and corneal pathologies. Cornea 2000; 19: 439–42.
36. Daya SM, Bell RWD, Habib NE, et al. Clinical and pathologic findings in human keratolimbal allograft rejection. Cornea 2000; 19: 443–50.
37. Sundmacher R, Reinhard T. Homologe lamelläre zentrale Limbokeratoplastik bei schwerer Limbusstämmzellinsuffizienz. Klin Monatsbl Augenheilkd 1998; 213: 254–5.
38. Sundmacher R, Reinhard T. Central corneolimbal transplantation under systemic cyclosporin A cover for severe limbal stem cell insufficiency. Graefes Arch Clin Exp Ophthalmol 1996; 234 (Suppl 1): S122–5.

Limbal stem cells; Microkeratome; Stem cell harvester; Limbal transplantation; Allograft

© 2002 Lippincott Williams & Wilkins, Inc.