Amniotic Membrane Transplantation for Conjunctival Surface Reconstruction

SCHEFFER C. G. TSENG, MD, PHD, PINNITA PRABHASAWAT, MD, AND SHWU-HUEY LEE, MD

• PURPOSE: To determine whether preserved hu­man amniotic membrane can be used to recon­struct the conjunctival defect created during surgical removal of a large lesion or during sym-blepharon lysis. • METHODS: Amniotic membrane transplantation was performed in six consecutive patients (seven eyes) during removal of large conjunctival lesions and in nine patients (nine eyes) during removal of conjunctival scars or symblepharon. • RESULTS: During a mean follow-up period ± SD of 10.9 ± 9 . 1 months (range, 2.2 to 34.0 months), 10 patients (11 eyes) showed successful surface reconstruction without recurrence, five patients (five eyes) showed improved visual acuity, and one patient (one eye) showed epithelialization within 3 weeks and resolution of motility restric­tion. Two patients (two eyes) showed partial suc­cess, with surrounding conjunctival inflammation. Three cases (three eyes) failed and exhibited recurrent scarring: one patient had received mito-mycin treatment and beta radiation, whereas the transplanted amniotic membrane of the second patient was partially, and of the third patient was

Accepted for publication May 5, 1997. From the Ocular Surface and Tear Center, Department of Ophthal­

mology, Bascom Palmer Eye Institute (Drs Tseng, Prabhasawat, and Lee), and Department of Cell Biology and Anatomy, University of Miami School of Medicine, Miami, Florida (Dr Tseng). Supported in part by Public Health Service Research Grant EY 10900 from Department of Health and Human Services, National Eye Institute, National Institutes of Health, Bethesda, Maryland, and in part by an unrestricted grant from Research to Prevent Blindness, Inc, New York, New York.

Reprint requests to SchefFer C. G. Tseng, MD, PhD, Bascom Palmer Eye Institute, William L. McKnight Vision Research Center, 1638 NW 10th Ave, Miami, FL 33136; fax: (305) 326-6306; e-mail: stseng® mednet.med.miami.edu

completely, dissolved or replaced by the inflamed pseudopterygial tissue. Two patients (two eyes) had epithelial cyst formation. • CONCLUSION: Amniotic membrane transplan­tation can be considered an alternative substrate for conjunctival surface reconstruction during re­moval for large tumors, disfiguring scars, or sym­blepharon, especially for those whose surrounding conjunctival tissue remains relatively normal.

ALTHOUGH SURGERIES TO REMOVE A LARGE POR-tion of conjunctiva are generally benign, scarring resulting from granulated tissue for­

mation can take place, resulting in a disfiguring appearance or motility restriction. This unwanted outcome can happen in removing dysplasia, tumors, scars, or symblepharon. To avoid this potential com-

See also pp. 723-728, 729-735, 736-750, 751-757, 758-764, and 825-835.

plication, autografts from the conjunctiva1 or the oral mucosa2,3 have been used to cover such conjunctival defects. Although a high success rate has been reported for the latter two procedures, scarring involv­ing the donor site has also been noted.4,5

Previously, amniotic membrane transplantation has been shown to be effective in the reconstruction of the corneal surface in rabbits6 and, combined with limbal allografts, in patients with advanced ocular cicatricial pemphigoid and Stevens-Johnson syndrome.7 Recently, preserved human amniotic mem­brane has also been shown to be effective in promot­ing healing of persistent corneal epithelial defects

VOL.124, No. 6 © AMERICAN JOURNAL OF OPHTHALMOLOGY 1997,124:765-774 765

TABLE. Demographic and Clinical Data

Patient No.. Age (yrs). Sex Underlying Cause

Group A: Removal of large conjunctival lesions 1.50, F 2, 32, F

3, 79, F

4, 47. M

5 ,41 , M

6, 69, M

Corneal and limbal melanoma Limbal and conjunctival

acquired melanosis Conjunctival chalasis

Limbal and conjunctival CIN with pterygium and granuloma pyogenlcum

Limbal and conjunctival CIN with pterygium

Limbal and conjunctival recurrent CIN with pseudopterygium

Eyo/Graft Size (mm)

LE/12x5 RE/8x7

RE/12X8 LE/12x8 RE/20X8

RE/8 x 5

RE/8 x 5

Visual Acuity

Before After

20/300 20/100 20/20

20/20 20/40 20/30

20/20

20/25

20/20

20/20 20/40 20/20

20/20

20/25

Follow-up (mos)

23 6

4.5

— 22.5

7.9

2.2

Outcome

Success with cyst Success

Success Success Success

Partial success with surrounding conjunctival inflammation

Partial success with cyst and surrounding conjunctival inflammation

Associated Procedures

Lamellar keratectomy

—

— —

Cryotherapy

Group B: Removal of conjunctival scar with or without symblepharon lysis LE/15x6 20/50 20/25 7.2 Success 7. 45, M Symblepharon and

pseudopterygium after thermal burn

8. 46, F Scar after multiple muscle surgeries for strabismus

9. 5, F Scar after multiple surgeries including mucous membrane grafts for forceps injury

10, 41, M Scar and symblepharon with motility restriction after primary nasal and temporal pterygium surgeries

11,42, M Scar after removal of primary RE/10x7 20/20 20/20 pterygium with primary closure

RE/15x5 20/20 20/20

LE/17X5 NLP NLP

RE/12x7 20/20 20/20

12, 43, M Scar and motility restriction after recurrent pterygium surgeries including MMC and 3 irradiation

13, 60, M Symblepharon after inactive OCP

14, 79, M Symblepharon after inactive OCP

15, 41, F Symblepharon after TEN

LE/10X7 20/20 20/20

RE/8x6 20/40 20/40

LE/15X10 HM 20/400

LE/9x8 20/40 20/25

12.5 Success

6 Success

5.5 Success with resolution of vertical motility restriction

3.2 Success

3.2 Failure

14 Success

12 Failure

34 Failure

CIN = conjunctival intraepithelial neoplasia, HM = hand motions, MMC cicatricial pemphigoid, TEN = toxic epidermal necrolysis.

mitomycin C, NLP = no light perception, OCP = ocular

with laceration8 and as a substrate alternative to conjunctival autograft during the "bare sclera" re­moval of pterygia.9 Because of these encouraging results, we examined whether this membrane can also be used for reconstructing conjunctival defects created during surgical removal of tumors, scars, or symblepharon.

PATIENTS AND METHODS

THE STUDY WAS APPROVED BY THE MEDICAL SCIENCE Subcommittee for the Protection of Human Subjects in Research of the University of Miami School of Medicine. At the Bascom Palmer Eye Institute from 1993 to 1996, amniotic membrane surgery was per-

766 AMERICAN JOURNAL OF OPHTHALMOLOGY DECEMBER 1997

formed in six consecutive patients (seven eyes) during the removal of a large conjunctival lesion in each (group A, Table). These lesions included acquired melanosis (one patient, one eye), conjunctival intra-epithelial neoplasia (three patients, three eyes), lim-bal and corneal melanoma (one patient, one eye), and conjunctival chalasis (one patient, two eyes). The procedure was also performed in another nine pa­tients (nine eyes) during the removal of conjunctival scars, with or without symblepharon lysis (group B, Table). The problems requiring the procedure in these patients were caused by previous eye surgeries (five patients, five eyes), thermal burn (one patient, one eye), toxic epidermal necrolysis (one patient, one eye), and inactive ocular cicatricial pemphigoid (two patients, two eyes). Demographic data for these patients are shown in the Table.

Human amniotic membrane was prepared and preserved by our recently described method.8'9 In brief, human placenta was obtained shortly after elective cesarean delivery. Human immunodeficiency virus, human hepatitis types B and C, and syphilis had been excluded by serologic tests. Under a lamellar flow hood, the placenta was cleaned of blood clots with sterile saline solution containing 50 |xg/ml of penicillin, 50 |xg/ml of streptomycin, 100 fig/ml of neomycin, and 2.5 u,g/ml of amphotericin B. The amnion was separated from the rest of the chorion by blunt dissection through the potential spaces situated between these two tissues and flattened onto a nitrocellulose paper with a pore size of 0.45 u,m, with the epithelium/basement membrane surface up. The paper with the adherent amniotic membrane was then cut into 3- x 4-cm disks and stored before transplantation at —80 C in a sterile vial containing Dulbecco modified Eagle medium and glycerol in a ratio of 1:1 (v/v).

After the thorough removal of the conjunctival or limbal lesion up to the episcleral surface in group A, and with the additional removal of subconjunctival scarring in group B, the amniotic membrane was then removed from the storage medium, peeled from the nitrocellulose filter paper, transferred to the recipient eye, and fitted to cover the defect by trimming off the excess portion. This fashioned membrane was then secured to the corneal edge of the defect by interrupt­ed 10-0 nylon sutures (when the limbal circumference

to be covered was less than 2 clock hours in size) or by a purse-string running suture (when it was more than 2 clock hours in size) and was secured to the surrounding conjunctival edge with interrupted 9-0 Vicryl sutures with episcleral bites. This was followed by topical application of Maxitrol ointment (neomy­cin sulfate, polymyxin B sulfate, and dexamethasone; Alcon Laboratories, Inc, Fort Worth, Texas). After surgery, all patients received prednisolone acetate 1% eyedrops every 2 hours while awake and Maxitrol nightly, both of which were tapered off in 1 month. Sutures were removed at 3 weeks.

RESULTS

AS SUMMED UP IN THE TABLE, AMNIOTIC MEMBRANE transplantation in group A (seven eyes) was effective in reconstructing defects created during surgical exci­sion of such conjunctival lesions as melanoma, acquir­ed melanosis, conjunctival intraepithelial neoplasia, and chalasis. For the mean follow-up period ± SD of 11.0 ± 9.3 months (range, 2.2 to 23 months), visual acuity was improved in two eyes and remained unchanged in five eyes after the surgery. There was no recurrence of the original lesion in any patient. In Patient 1, additional lamellar keratectomy was per­formed to remove the invasive corneal and limbal melanoma, which had previously been removed sever­al times. In general, epithelialization was rapid and invariably completed in 3 weeks despite the variable sizes of defects created and grafts used. As shown in Figure 1, which compares preoperative and postopera­tive appearances, Patients 1 to 4 (five eyes) had a good success with the reconstructed surface appearing noninflamed, wet, and smooth (Figure 1, a through f). Furthermore, in these five eyes, the membrane-covered area tended to be less injected than the surrounding host tissue was (for example, see Figure Id), and scars were noted only in the interface where the membrane was sutured to the host conjunctival edge (also Figure Id). We noted focal cyst formation on the transplanted membrane area after healing in Patients 1 and 4. We rated the results of the remaining cases, in Patients 5 and 6, as partial successes because the conjunctiva surrounding the lesion remained inflamed and did not return to normal after amniotic

VOL.124, No. 6 AMNIOTIC MEMBRANE TRANSPLANTATION 767

768 AMERICAN JOURNAL OF OPHTHALMOLOGY DECEMBER 1997

membrane transplantation and because the trans­planted amniotic membrane was partially (Patient 5) and completely (Patient 6) dissolved or replaced by the inflamed pseudopterygial tissue (Figure 1, g through j).

In group B, amniotic membrane transplantation facilitated successful reconstruction in six of nine cases during removal of conjunctival scars or lysis of symblepharon, or both (Table). For the follow-up period of 10.8 ± 9.6 months (range, 3.2 to 34 months), the visual acuity of three eyes (Patients 7, 14, and 15) improved, and that of the remaining six eyes was not changed. As shown in Figures 2 and 3, the reconstructed conjunctival surfaces were success­ful and cosmetically superior in Patients 7 through 11. In Patient 7, the scarred tissue and symblepharon were caused by a remote thermal injury, resulting in tight adhesion of the lower eyelid to the lower cornea and obliteration of the involved bulbar and tarsal conjunctiva (Figure 2, a and b). In Patient 8, multiple muscle surgeries had been performed to correct a childhood esotropia, including reattachment of the lost medial rectus muscle (Figure 2, c and d). In Patient 9, the scarred tissue was part of the mucous membrane graft used to reconstruct the sealed forni-ces with both upper and lower eyelids after a serious forceps injury at birth, which had also led to a total retinal detachment and loss of vision (Figure 2, e and f). In Patients 10 through 12, a disfiguring scar with or without motility restriction developed, although no true recurrence was noted after pterygial surgeries. In Patient 10, the amniotic membrane was placed inferi-orly and successfully released vertical motility restric­

tion caused by scarred tissue and symblepharon in the nasal (Figure 2, g and h) and temporal (not shown) inferior fornices. In Patient 11, the disfiguring scarred tissue developed after excision of primary pterygium with primary closure (Figure 2, i and j). Nevertheless, in Patient 12, the nasal scarred tissue caused disturbing diplopia, but unfortunately the transplanted amniotic membrane was never epithelialized and eventually failed (Figure 3, a and b). A notable difference be­tween Patient 12 and Patients 10 and 11 was that beta irradiation and mitomycin C treatment had been performed during two previous pterygial surgeries, respectively, in the latter.

For symblepharon caused by progressive scarring of the involved conjunctiva in ocular cicatricial pemphi-goid and toxic epidermal necrolysis, amniotic mem­brane transplantation was not as effective. Amniotic membrane transplantation was successful in recreat­ing the inferior fornix and hence in facilitating die formation of the tear meniscus in a focal manner in Patient 13 (Figure 3, c and d, arrows), but it failed to achieve effective reconstruction in Patients 14 and 15 (Figure 3, e through h). It should be noted that symblepharon was a part of total (Patient 14) and partial (Patient 15) limbal stem cell deficiency and that amniotic membrane transplantation was per­formed without limbal stem cell transplantation.

DISCUSSION

WHEN HEALING IS ALLOWED TO TAKE PLACE ON A LARGE conjunctival defect created surgically, it is likely that

FIGURE 1. External photographs of group A (Patients 1 through 6). (a) Patient 1. Preoperative appearance, with amelanotic malignant melanoma involving the 9 to 11 o'clock-position limbus and extending to the peripheral cornea. (b) One-month postoperative appearance after lamellar keratectomy to remove the tumor and reconstruction by amniotic membrane transplantation, (c) Patient 2. Preoperative appearance, showing acquired melanosis involving the 9 to 11 o'clock-position limbus and extending to 7- to 8-mm adjacent bulbar conjunctiva, (d) Six-month postoperative appearance of the site covered by the amniotic membrane (outlined by arrows, which also show the scar tissue formed at the border), which was quiet and less injected than the adjacent tissue, (e) Patient 4. Preoperative appearance, exhibiting biopsy-proven perilimbal conjunctival intraepithelial neoplasia involving the entire limbal circumference except the 7 to 11 o'clock positions, and an associated pterygium and a biopsy-induced granuloma in the nasal bulbar conjunctiva, (f) Five-month postoperative appearance, (g) Patient 5. Preoperative and (h) 4-month postoperative appearance, with limbal conjunctival intraepithelial neoplasia and adjacent pseudopterygium. (i) Patient 6. Preoperative and (j) 2-month postoperative appearance, with perilimbal conjunctival intraepithelial neoplasia and adjacent pseudopterygium. Both Patients 5 and 6 healed with membrane partially (h) or completely (j) dissolved and the surrounding fibrovascular tissue remaining.

VOL.124, No. 6 AMNIOTIC MEMBRANE TRANSPLANTATION 769

AMERICAN JOURNAL OF OPHTHALMOLOGY DECEMBER 1997

scar tissue will develop. When scarring develops in the area covered by the eyelid, for example, in conjunctival autografts for pterygia, medical attention is infrequently needed unless ptosis or motility restric­tion is noted.10 However, when healing occurs on the exposed nasal or temporal conjunctiva, the scarred tissue frequently exhibits annoying redness, irritation, or horizontal motility restriction. Surgical manage­ment may become challenging and frequently leads to autologous grafts from the conjunctiva1 or the oral mucosa.2,3

In this report, we have demonstrated that the preserved human amniotic membrane can be used as an alternative graft. In a total of 15 patients (16 eyes), nine patients (10 eyes) showed successful reconstruc­tion, with the return of satisfactory mucosal appear­ance. It should be noted that all nine of these patients had relatively normal conjunctiva adjacent to the lesion. These were Patients 1 through 4 in group A and 7 through 11 in group B (Table). In these cases, we noted that epithelialization was rapid and invaria­bly was completed in 3 weeks, even when a graft size exceeding 8 X 10 mm was used. Rapid epithelializa­tion has also been noted in our recent reports of using amniotic membrane transplantation for patients with persistent sterile corneal ulceration8 and during pte-rygial excision.9 The resulting conjunctival surface remained stable, without inflammation or erosion, and tended to be less injected than the adjacent host tissue was, for a mean total of 10.9 ± 9 . 1 months (range, 2.2 to 34 months) thereafter. Furthermore, impression cytology confirmed the recovery of normal conjunctival epithelial phenotype with a higher gob­let cell density than that of normal controls.11 We attribute this success to a unique property of the amniotic membrane, which contains a thick base­ment membrane and an avascular stromal matrix.12'15

It has been recognized that the basement membrane facilitates migration of epithelial cells,16 reinforces adhesion of basal epithelial cells,17,18 promotes epi­thelial differentiation,19'22 and prevents epithelial apo-ptosis.23,24

Amniotic membranes have been used for other indications.25 Except in rare situations, live rather than preserved amniotic membranes were used in these instances and therefore might have caused undesirable allograft rejection. The first report in English describing the ophthalmic use of live human fetal membrane, that is, amniotic membrane with chorionic membrane, for conjunctival symblepharon was made by de Rotth in 1940.26 Although claiming a complete "take" in all six patients, the author noted one case of success and progressive shrinkage and dissolution in the other five cases. Because only the substrate without live cells was transplanted in our patients, no adverse effect such as allograft rejection was noted, a finding consistent with those of our recent reports.8,9 In our five patients with symblepha­ron (Patients 7, 10, and 13 through 15), we noted that amniotic transplantation was effective in three, two of whom had relatively normal conjunctiva adjacent to the symblepharon (Patient 7, with ther­mal burn, and Patient 10, after pterygial removal). For pterygia, we have noted9 that surface reconstruction can be achieved effectively by amniotic membrane transplantation, with a recurrence rate significantly lower than that of pure bare sclera excision with or without primary closure. In that report,9 we also noted that, if it is not thoroughly removed, the surrounding pterygial fibrovascular tissue might impede the out­come of amniotic membrane transplantation.

In the remaining three failed cases in the present report, there were diffuse cicatricial changes sur­rounding the symblepharon. Furthermore, prior treat-

FIGURE 2. External photographs of group B (Patients 7 through 11). (a) Patient 7. Preoperative and (b) 6-month postoperative appearance, with symblepharon and pseudopterygium involving adhesion of the lower eyelid margin (asterisk) to the cornea, (c) Patient 8. Preoperative and (d) 1-year postoperative appearance, with conjunctival scar involving the nasal and the rest of 360 degrees of perilimbal conjunctiva (not shown here) after multiple muscle surgeries, (e) Patient 9. Preoperative and (f) 6-month postoperative appearance, with scarring in the nasal and the remaining 360 degrees of perilimbal bulbar conjunctiva after a mucous membrane graft, (g) Patient 10. Preoperative and (h) 7-week postoperative appearance, with inferonasal and inferotemporal (not shown) scar and symblepharon after a pterygium surgery, (i) Patient 11. Preoperative and (j) 6-month postoperative appearance, with conjunctival scar and inflammation after pterygium removal and primary closure.

VOL.124, No. 6 AMNIOTIC MEMBRANE TRANSPLANTATION 771

FIGURE 3. External photographs of group B (Patients 12 through 15). (a) Patient 12. Three-week and (b) 5-week postoperative appearance, with conjunctival scar and motility restriction after multiple procedures. Note the necrosis and dissolution of the amniotic membrane (b). (c) Patient 13. Preoperative and (d) 5-month postoperative appearance, with symblepharon resulting from ocular cicatricial pemphigoid. The area that was reconstructed is marked by arrows. (e) Patient 14. Preoperative and (f) 5-month postoperative appearance, with symblepharon caused by inactive ocular cicatricial pemphigoid. (g) Patient 15. Preoperative and (h) 10-month postoperative appearance, with symblepharon caused by toxic epidermal necrolysis. Note the recurrence of the symblepharon, which was part of limbal stem cell deficiency.

772 AMERICAN JOURNAL OF OPHTHALMOLOGY DECEMBER 1997

ment with mitomycin C or beta irradiation might have also altered the surrounding tissue to the extent that amniotic membrane transplantation became contraindicated, as shown by Patient 12. In Patients 14 and 15, whose symblepharon was a part of limbal stem cell deficiency, amniotic membrane transplanta­tion was not sufficient for effective ocular surface reconstruction. This notion is further supported by our recent impression cytology finding of conjunctival but not corneal epithelial phenotype on the corneal surface.11 Thus, we concur with Tsubota and associ­ates,7 who showed that amniotic membrane trans­plantation together with limbal allografts can recon­struct the corneal surface in patients with advanced ocular cicatricial pemphigoid and Stevens-Johnson syndrome.11 Furthermore, severe dry eye could be another explanation for the graft failure. We attribute the cyst formation in Patients 1 and 6 to the imperfect suturing of the membrane to the host conjunctiva in our early experience, leading to epithelialization un­der the membrane.

We suggest that amniotic membrane transplanta­tion be considered for conjunctival surface reconstruc­tion, especially when a lesion on the interpalpebral zone or a large lesion is removed. Success can be expected from those patients whose surrounding conjunctival tissue is healthy and whose corneal surface is without limbal stem cell deficiency. Further studies are needed to determine the respective role of the ocular surface epithelium and fibroblasts in inter­acting with the amniotic membrane matrix. We recognize that our patient series is relatively small, but further investigations and trials may unravel other potential uses for human amniotic membrane trans­plantation.

ACKNOWLEDGMENT The authors wish to express their deep gratitude to D. Gene Burkett for obtaining donor placentas.

REFERENCES

1. Vastine DW, Stewart WB, Schwab IR. Reconstruction of the periocular mucous membrane by autologous conjunctival transplantation. Ophthalmology 1982;89:1072-1081.

2. Hosni FA. Repair of trachomatous cicatricial entropion using

mucous membrane graft. Arch Ophthalmol 1974;91:49-51. 3. Shore JW, Foster S, Westfall CT, et al. Results of buccal

mucosal grafting for patients with medically controlled ocular cicatricial pemphigoid. Ophthalmology 1992;99:383-395.

4. Vrabec MP, Weisenthal RW, Elsing SH. Subconjunctival fibrosis after conjunctival autograft. Cornea 1993;12: 181-183.

5. Neuhaus RW, Baylis HI, Shorr N. Complications at mucous membrane donor sites. Am J Ophthalmol 1982;93:643-646.

6. Kim JC, Tseng SCG. Transplantation of preserved human amniotic membrane for surface reconstruction in severely damaged rabbit corneas. Cornea 1995;14:473-484.

7. Tsubota K, Satake Y, Ohyama M, et al. Surgical reconstruc­tion of the ocular surface in advanced ocular cicatricial pemphigoid and Stevens-Johnson syndrome. Am J Ophthal­mol 1996;122:38-52.

8. Lee S, Tseng SCG. Amniotic membrane transplantation for persistent epithelial defects with ulceration. Am J Ophthal­mol 1997;123:303-312.

9. Prabhasawat P, Barton K, Burkett G, et al. Comparison of conjunctival autografts, amniotic membrane grafts and pri­mary closure for pterygium excision. Ophthalmology. Forth­coming.

10. Adamis AP, Starck T, Kenyon KR. The management of pterygium. Ophthalmol Clin North Am 1990;3:611-623.

11. Prabhaswat P, Tseng SCG. Impression cytology study of epithelial phenotype of ocular surface reconstructed by preserved human amniotic membrane. Arch Ophthalmol. Forthcoming.

12. Liotta LA, Lee CW, Morakis DJ. New method for preparing large surfaces of intact human basement membrane for tumor invasion studies. Cancer Lett 1980;11:141-152.

13. Russo RG, Thorgeirsson U, Liotta LA. In vitro quantitative assay of invasion using human amnion. In: Liotta LA, Hart IR, editors. Invasion and metastasis. Dordrecht: NijhofF, 1982:173-187.

14. Modesti A, Kalebic T, Scarpa S, et al. Type V collagen in human amnion is a 12-nm fibrillar component of the pericellular interstitium. Eur ] Cell Biol 1984;35:246-255.

15. Campbell S, Allen TD, Moser BB, et al. The translaminal fibrils of the human amnion basement membrane. J Cell Sci 1989;94:307-318.

16. Terranova VP, Lyall RM. Chemotaxis of human gingival epithelial cells to laminin: a mechanism for epithelial cell apical migration. ] Periodontol 1986;57:311-317.

17. Khodadoust AA, Silverstein AM, Kenyon KR, et al. Adhe­sion of regenerating corneal epithelium: the role of basement membrane. Am J Ophthalmol 1968;65:339-348.

18. Sonnenberg A, Calafat J, Janssen H, et al. Integrin a6/b4 complex is located in hemidesmosomes, suggesting a major role in epidermal cell-basement membrane adhesion. J Cell Biol 1991;113:907-917.

19. Guo M, Grinnell F. Basement membrane and human epidermal differentiation in vitro. ] Invest Dermatol 1989;93: 372-378.

20. Streuli CH, Bailey N, Bissell MJ. Control of mammary epithelial differentiation: basement membrane induces tissue-specific gene expression in the absence of cell-cell interaction and morphological polarity. J Cell Biol 1991;115:1383-1395.

21. Kurpakus MA, Stock EL, Jones JCR. The role of the basement membrane in differential expression of keratin proteins in epithelial cells. Dev Biol 1992;150:243-255.

22. Barcellos-Hoff MH, Aggeler J, Ram TG, et al. Functional differentiation and alveolar morphogenesis of primary mam-

VOL. 124, NO. 6 AMNIOTIC MEMBRANE TRANSPLANTATION 773

mary cultures on reconstituted basement membrane. Devel­opment 1989;105:223-235.

23. Boudreau N, Sympson C], Werb Z, et al. Suppression of ICE and apoptosis in mammary epithelial cells by extracellular matrix. Science 1995;267:891-893.

24. Boudreau N, Werb Z, Bissell MJ. Suppression of apoptosis by basement membrane requires three-dimensional tissue organ­

ization and withdrawal from the cell cycle. Proc Nad Acad Sci U S A 1996;93:3500-3513.

25. Trelford JD, Trelford-Sauder M. The amnion in surgery, past and present. Am J Obstet Gynecol 1979;134:833-845.

26. de Rotth A. Plastic repair of conjunctival defects with fetal membrane. Arch Ophthalmol 1940;23:522-525.

Authors Interactive™ Comments and questions regarding this article may be addressed to authors via THE JOURNAL'S Web Site. Selected comments, questions, and author responses are posted on Authors Interactive™ at http://www.ajo.com/

774 AMERICAN JOURNAL OF OPHTHALMOLOGY DECEMBER 1997