Secondary Logo

Journal Logo

Original Article

Early Results of Penetrating Keratoplasty following Limbal Stem Cell Transplantation

Sangwan, Virender S MS; Fernandes, Merle MS; Bansal, Aashish K MS; Vemuganti, Geeta K MD; Rao, Gullapalli N MD

Author Information
Indian Journal of Ophthalmology: Jan–Mar 2005 - Volume 53 - Issue 1 - p 31-35
doi: 10.4103/0301-4738.15282
  • Open


Currently limbal transplantation (LT) with concurrent immunosuppression for limbal allografts is the only option for bilateral total limbal stem cell deficiency.1 The aim of surgery is not to restore vision but to replenish limbal stem cells and maintain a stable ocular surface. Visual acuity may be restored by a simultaneous or sequential penetrating keratoplasty (PKP). 1 A one-stage procedure combining PKP and LT utilising a single donor may be advantageous since theoretically the antigenic load to the recipient is lower.23 Recent evidence, however, shows that following simultaneous PKP and LT, there is a progressive decline in ambulatory vision, limbal and corneal graft survival.45 A gap of at least three months between LT and PKP, has been recommended until the ocular surface is stable.6 Since not much data is available on the outcome of corneal transplantation performed after successful ocular surface reconstruction by limbal allografts, we report our results of PKP in four eyes of four patients who had undergone successful limbal transplantation between 3 and 4.5 months earlier.

Materials and Methods

We reviewed the medical records of four cases of bilateral limbal stem cell deficiency that underwent cadaveric keratolimbal or living related conjunctival limbal allograft followed by PKP.

The following details were noted: demographics, primary aetiology, previous ocular surgery, type of limbal transplant (cadaveric or living related), vision before and after limbal transplantation, ocular surface stability after Letter to Editor and PKP, immunosuppression, outcome of PKP and postoperative complications.

Surgical Technique

Keratolimbal allograft

The keratolimbal allograft was harvested as described earlier.17 Two donor tissues were used to provide lenticules that would cover 360 degrees. The recipient bed was prepared first. A 360-degree peritomy was done 3 mm from the limbus. The pannus was resected off the cornea by blunt dissection. The corneo-scleral rim was dissected with a crescent knife ensuring that minimal anterior stroma was included. Amniotic membrane was spread on the cornea and adjacent conjunctiva, and anchored at the limbus with 10-0 monofilament nylon interrupted sutures. The lenticules were then sutured in the correct anatomical position with 10-0 monofilament nylon interrupted sutures. A bandage contact lens was used at the end of surgery.

Living related conjunctival limbal allograft

Conjunctival limbal allografts were harvested at the same sitting, from the right eyes of both parents of the affected patient. They were operated under local anaesthesia in the operating room immediately before the recipient′s surgery. Dissection commenced 3 mm beyond the limbus, continued 1mm into the cornea, and spanned 3 clock hours superiorly and inferiorly. Care was taken not to dissect too deeply into the corneal stroma and to avoid harvesting the Tenon′s capsule along with the conjunctiva. The four donor lenticules were stored in Ringer′s lactate solution. The recipient bed was prepared as described above. The donor lenticules were sutured in their correct anatomical position, with 10-0 monofilament nylon interrupted sutures ensuring complete limbal coverage. The amniotic membrane and donor conjunctiva was sutured to the surrounding conjunctiva with interrupted 8-0 vicryl sutures.


Systemic immunosuppressants were administered to all patients, after adequately counselling them about adverse reactions. Baseline haematological investigations, hepatic and renal parameters were obtained. These parameters were re-assessed every 4-6 weeks. Immunosuppressants were continued with in all patients.

Three doses of intravenous methyl prednisolone were given on 3 consecutive days from the first postoperative day followed by oral prednisolone (1mg/kg/body wt) which was tapered gradually. Oral cyclosporine A (10mg/kg/body wt in divided doses) was given 48 hours preoperatively and then tapered to 3-5mg/kg/body wt as a maintenance dose. Diltiazem (90mg) was added as an adjunct to Cyclosporine A to increase serum levels of Cyclosporine A thereby reducing the daily dosage and thus the cost of Cyclosporine A (vide infra). Topical prednisolone acetate 1% was started postoperatively 8 times daily and tapered gradually based on the existing inflammation. Gentamicin sulphate 0.3% eyedrops were prescribed 6 hourly in the first week postoperatively. The patients were examined every six weeks.

The demographic data, aetiology and duration of limbal stem cell deficiency, previous surgeries, best-corrected visual acuity and surgical details of all four cases are shown in Table 1. A representative case has been described below.

Table 1
Table 1:
Demographic, pre- and postoperative data of the 4 cases who received PKP after successful LT

Case report

A 29-year-old man had bilateral total stem cell deficiency following chemical injury 13 months before presenting at our hospital. His visual acuity was counting fingers at 1m in the right eye and perception of light with accurate projection of rays in the left eye. Living-related conjunctival limbal allograft with pannus resection was done in the right eye. A similar procedure was done for the left eye. The ocular surface was stable and there was no evidence of conjunctivalisation or epithelial defects (Figure 1a). A corneal transplant with cataract extraction and intraocular lens implantation was done 4.5 months later. A graft-host disparity of 0.25mm was maintained. Intraoperatively, the recipient bed was extremely thinned out. Difficulties were encountered while suturing and in maintaining a deep anterior chamber. Over 2 months, his visual acuity improved to 6/36 and the graft was clear (Figure 1b). Due to peripheral anterior synechiae in two quadrants, the patient developed secondary angle closure glaucoma 2.5 months after PKP. Despite maximal medical anti-glaucoma medications for one month, his intraocular pressure was 44mm Hg. Since the bulbar conjunctiva was scarred, a trabeculectomy was not possible. A trans-scleral cyclophotocoagulation was done in 2 quadrants. At last follow-up 4.5 months after PKP, the visual acuity in his right eye was counting fingers at 0.1m and intraocular pressure was 20mm Hg on 0.5% timolol malleate eyedrops. The graft was oedematous. There was no evidence of conjunctivalisation. In the left eye the visual acuity improved from accurate projection of rays before limbal transplantation, to 6/18 after living related conjunctival limbal allograft. This eye did not require keratoplasty. The density of corneal scarring reduced, thus eliminating the need for further surgical intervention.

Figure 1
Figure 1:
a. Right eye of patient 2, showing a dense central corneal scar with no recurrence of conjunctivalisation following living related conjunctival limbal allograft covering the entire limbus. There are no areas of localised congestion or chemosis. b. Same eye showing a stable ocular surface with healthy epithelium 2 months after PKP. The best-corrected visual acuity at this point of time was 0.1 meter and intraocular pressure was under control following limited trans-scleral cyclophotocoagulation.


Table 1 summarises the demographic, preoperative, surgical and postoperative data of the four patients who underwent LT. The duration between LT and PKP ranged from 3 months to 4.5 (mean 3.8) months and the duration of follow up after PKP ranged from 4 to 11 (mean 6) months. In all patients the ocular surface was stable after LT and it remained stable after PKP. The post-operative complications were superficial punctuate keratopathy, corneal allograft rejection, and secondary glaucoma. There were no cases of limbal allograft rejection or evidence of recurrent stem cell deficiency. In case 1, superficial punctate keratopathy in the corneal graft two months post PKP resolved with topical preservative-free lubricants. The lenticules were not engorged, oedematous or associated with overlying epithelial defects. Corneal allograft rejection developed 3.5 months post PKP following discontinuation of immunosuppressants of the patient′s own accord. The rejection could not be reversed by restarting oral and topical immunosuppression and resulted in graft failure. Secondary angle closure glaucoma unresponsive to maximal medical therapy developed in case 2. There was no evidence of raised intraocular pressure pre-operatively. Limited trans-scleral cyclophotocoagulation to control the intraocular pressure subsequently resulted in graft oedema, and reduction in vision.

Side effects of immunosuppression were not experienced by any of the patients. The donors are continuously monitored and none have exhibited any signs of limbal stem cell deficiency.


LT aims to restore ocular surface integrity by improving surface lubrication, to replace conjunctival phenotypic epithelium by corneal phenotypic epithelium and to promote the barrier function of the limbus. Improvement in corneal clarity and visual acuity as a result of this procedure may obviate the need for a PKP. However, in cases where there is significant stromal scarring, a corneal graft is necessary to visually rehabilitate the patient.

The advantages of simultaneous LT and PKP have been addressed earlier.23 There is a greater risk of rejection following limbal transplant than with conventional PKP.89 However, it is also known that a corneal graft in an inflamed and vascularised bed carries a greater risk of rejection.10 Therefore, we prefer visual rehabilitation of these legally blind patients in a phased manner. The first stage involves ocular surface reconstruction by transplantation of limbal stem cells and the second aims at restoration of vision by a PKP. The outcome of simultaneous and sequential LT and PKP has been summarised in a recent review.1 The longterm outcome of PKP either combined with or done after keratolimbal allograft was poor.5

In our series, the visual acuity ranged from perception of light with accurate projection of rays to 6/120 preoperatively. All 4 eyes improved visual acuity in the early postoperative period following PKP. In case 1 the visual acuity reduced from 6/36 to counting fingers at 10 cm following corneal allograft rejection and improved to 6/60 with treatment. In case 2 the vision reduced to counting fingers at 0.1m due to graft oedema.

Retrospectively we felt that the graft-host disparity of 0.25 mm (instead of 0.5 mm) in case # 2 was partly responsible for shallow anterior chamber resulting in peripheral anterior synechiae and secondary glaucoma. A 0.5mm larger graft should be used in all cases to ensure adequate anterior chamber depth. Currently there is no definite guidelines for management of glaucoma in such situations if the medical therapy fails.11 In addition, during LT, if the recipient bed is extremely thin, a simultaneous lamellar keratoplasty may be beneficial. In grade 3 or 4 chemical injuries with a thinned out cornea, a large-diameter lamellar keratoplasty as a single-stage procedure may provide a tectonically stable cornea obviating the need for limbal transplantation.12

In our series, 3 eyes received conjunctival limbal allografts from living related donors. In one eye, cadaveric donor tissue was used. Immune histocompatibility matching was not attempted due to financial constraints. However, rejection may occur even in haplo-identical recipients.13 In one eye a cadaveric kerato-limbal allograft was performed. The yield of limbal stem cells from cadaveric corneo-scleral rims is much lower (43%) than that from living related donors (100%).14 In addition, several reports have shown progressive failure with time despite continuous and aggressive immunosuppression.4515 It is likely that living related conjunctival limbal allografts may survive longer than kerato-limbal allografts, though prospective randomised case-controlled studies might address the question better.

In 3 eyes, limbal tissue was transplanted from livingrelated donors. The stem cell deficiency was 360 degrees in two of these, and complete coverage of the limbus was undertaken by harvesting 6 clock hours from one eye of each parent of the recipient. A major concern is the well being of the donor eyes. It is unknown how these eyes will behave in future in response to an epithelial insult. There are reports1617 of complications at these iatrogenic stem cell deficient sites. Harvesting the anterior stroma along with the Bowman′s membrane and limbal epithelium predisposes to pseudopterygium formation. None of the donor eyes in our series had any complications (pseudopterygium or epithelial defects) as our technique of harvesting the tissue involved minimal invasion of the corneal stroma.

Limbal allograft rejection rates are rather high, from 25% -33.3%.1318 Limbal allograft rejection was not observed in any of our patients. But prolonged follow-up is necessary. One patient (case 1) developed endothelial rejection on discontinuation systemic Cyclosporine A of his own accord. Until the last follow up, the endothelial rejection had not completely reversed. It is clear that immunosuppression is necessary in limbal allografts even in HLA matched recipients13 despite earlier conflicting reports.19 Aggressive immunosuppression is the key to success with limbal transplantation, as in any other solid organ transplantation. All our patients received oral corticosteroids in the immediate preoperative period and continue to be on maintenance doses of Cyclosporine A.

We have used diltiazem as an adjuvant to cyclosporine A. It inhibits cyclosporine metabolism. In vitro studies have demonstrated that it enhances the suppressive effect of cyclosporine A and methyl prednisolone in lower doses. Thus cyclosporine A dosage is reduced by 30-50% with a considerable reduction in medication cost.20 In high non-therapeutic doses however, it exerts a suppressive effect.21 It also provides renal protection from cyclosporine A nephrotoxicity.22 Its use in limbal allografts and high-risk corneal transplants has not been described in the literature. The efficacy of diltiazem as an adjuvant to cyclosporine A in limbal allotransplantation has to be assessed.

The major limitation of this study is the short duration of follow up and poor visual outcome despite satisfactory ocular surface stability.

1. Fernandes M, Sangwan VS, Rao SK, Basti S, Sridhar MS, Bansal AK, Dua HS. Limbal Stem Cell Transplantation Indian J Ophthalmol. 2004;52:5–22
2. Theng JT, Tan DT. Combined penetrating keratoplasty and limbal allograft transplant for severe corneal burns Ophthal Surg Lasers. 1997;28:765–68
3. Reiner T, Sundmacher R, Spelsberg H, Althaus C. Homologous penetrating central limbokeratoplasty (HPCLK) in bilateral limbal stem cell insufficiency Acta Ophthalmol Scand. 1999;77:613–17
4. Solomon A, Ellies P, Anderson DF, Touhami A, Grueterich M, Espana EM, Ti SE, Goto E, Feuer WJ, Tseng SC. Long-term outcome of keratolimbal allograft with and without penetrating keratoplasty for total limbal stem cell deficiency Ophthalmology. 2002;109:1159–66
5. Ilari L, Daya SM. Long-term outcomes of keratolimbal allografts for the treatment of severe ocular surface disorders Ophthalmology. 2002;109:1278–84
6. Croasdale CR, Schwartz GS, Malling JV, Holland EJ. Kerato limbal allograft: Recommendation for tissue procurement and preparation by eye banks and surgical technique Cornea. 1999;18:52–58
7. Tsubota K, Santake Y, Ohyama M. Surgical reconstruction of the ocular surface in advanced ocular cicatricial pemphigoid and Stevens-Johnson syndrome Am J Ophthalmol. 1996;122:38–52
8. Tsubota K, Toda I, Santo H. Reconstruction of the corneal epithelium by limbal allograft transplant for severe ocular surface disorder Ophthalmology. 1995;102:1486–96
9. Pels E, van der Gaag R. HLA-A, B, C and HLA-DR antigens and dendritic cells in fresh and organ-culture preserved corneas Cornea. 1984;3:231–39
10. Gottsch JD, Jakobs FM, Stark WJBrightbill FS. Regrafting Corneal Surgery: Theory, Technique and Tissue. 19993rd edition St Louis Mosby, Inc:508
11. Ayyala RS. Penetrating keratoplasty and glaucoma Surv Ophthalmol. 2000;45:91–105
12. Vajpayee RB, Thomas S, Sharma N, Dada T, Tabin GC. Large-diameter lamellar keratoplasty in severe ocular alkali burns Ophthalmology. 2000;107:1765–68
13. Kwitko S, Marinho D, Barcaro S. Allograft conjunctival transplantation for bilateral ocular surface disorders Ophthalmology. 1995;102:1020–25
14. Vemuganti GK, Kashyap S, Sangwan VS, Singh S. Ex vivo potential of cadaveric and fresh limbal tissues to regenerate cultured epithelium Indian J Ophthalmol. 2004;52:113–20
15. 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;340:1697–703
16. Basti S, Mathur U. Unusual intermediate outcome in 3 cases of limbal autograft transplant Ophthalmology. 1999;106:958–63
17. Morgan S, Murray A. Limbal autotransplantation in the acute and chronic phases of severe chemical injuries Eye. 1996;10:349–54
18. Rao SK, Rajagopal R, Sitalakshmi G, Padmanabhan P. Limbal allografting from related live donors for corneal surface reconstruction Ophthalmology. 1999;106:822–28
19. Tan DTH, Ficker LA, Buckley RJ. Limbal transplantation Ophthalmology. 1996;103:29–36
20. Martin JE, Daoud AJ, Schroeder TJ, et al The clinical and economic potential of cyclosporin drug interactions Pharmacoeconomics. 1999;15:317–37
21. D′Ambrosio A, Segoloni G, Quintieri F. The modulatory effect of diltiazem on human in vitro alloreactivity when used along or in combination with cyclosporine A and or methyl prednisolone Transpl Int. 1997;10:426–3
22. Rodicio JL. Calcium antagonists and renal protection from cyclosporin nephotoxicity and long term trial in renal transplant patients J Cardiovasc Pharmacol. 2000;35:S7–11

Proprietary Interest: None


Limbal stem cell transplant; penetrating keratoplasty; limbal allograft

© 2005 Indian Journal of Ophthalmology | Published by Wolters Kluwer – Medknow