Postoperative complications associated with the use of sutures in corneal surgery include inflammation, stromal vascularization, astigmatism and prolonged visual rehabilitation. For more than 40 years, ophthalmic surgeons have been investigating a variety of substances, both biologically derived and synthetically produced, in a search for an adhesive material that could replace sutures.1·2
The ideal surgical adhesive should rapidly and strongly adhere tissue planes in an aqueous environment, as well as be biodegradable, microbicidal and non-toxic. Several adhesive materials have been experimentally and, in some cases, clinically evaluated for their efficacy in ocular surgery, particularly in the closure of surgical or traumatic corneal wounds.319 In order to further assess the interaction of potential surgical adhesives and the cornea, we compared the clinical and histopathologic effects of two adhesives when instilled in the corneal stroma of rabbits.
MATERIALS AND METHODS
Young adult (2 to 3 kg) New Zealand white rabbits of both sexes were used in this study. The animals were housed at the Biological Resources Laboratory of the University of Illinois College of Medicine at Chicago. They were maintained and cared for in accordance with the 1983 ARVO Resolution on the Care and Use of Animals in Research.
Two adhesives, one synthetic and the other biologically derived, were evaluated in this study. The synthetic adhesive was a commercial preparation of butyl-2-cyanoacrylate (Histoacryl, B. Braun Melsungen AG, Melsungen, West Germany). It was packaged sterile in 0.5 gn vials and stored at 4°C until time of use. The biologically derived adhesive was a combination of an experimental preparation of a polyphenolic protein with an enzyme polymerizer (COX). The polyphenolic protein is a repeating decapeptide derived from the common blue mussel Mytilus edulis) and is known as a mussel adhesive protein (MAP). Both components were provided in sterile form by BioPolymers Ine (Farmington, Conn). MAP was used at a concentration of 5 mg/mL, while the COX component was used at a concentration of 324 units^L. We previously have used these concentrations in efficacy evaluations of the MAP-COX adhesive (unpublished data). Both MAP and COX were stored at 4°C prior to usage; storage periods were no longer than 2 weeks for MAP and 2 months for COX. The MAP-COX adhesive was freshly mixed at a 10:1 by volume ratio of MAPrCOX immediately prior to each intracorneal instillation. This ratio had been determined previously to produce acceptable corneal tissue adhesion (P. Picciano, PhD, personal communication).
Figure 1: Rabbit comea 14 days following intrastromal placement of cyanoacrylate (closed arrows). Note superior neovascularization (open arrows) of a severity grade of approximately 2. (Original magnification x 16)
Figure 2: Rabbit cornea 7 days following intrastromal placement of MAP-COX. No neovascularization is evident. Note that the incision site is clearly seen (arrows), while the intrastromal adhesive (asterisks) is vaguely visible. (Original magnification x16)
For control studies, commercially available balanced saline solution (BSS, Alcon, Ft Worth, Tex) was used.
The animals were randomly assigned to either one of the two experimental groups or to the control group. Each animal had a baseline biomicroscopic examination in order to rule out pre-existing anterior segment abnormalities. Prior to each procedure, animals were anesthetized with subcutaneous xylazine hydrochloride (4 mg/kg) and ketamine hydrochloride (35 mg/kg); additionally, proparacaine hydrochloride was applied topically.
For this study, we elected to place the adhesive material in the stroma by creating a lamellar pocket, as opposed to direct intrastromal injection. This was done to ensure that a uniform volume of material was routinely distributed over a uniform area of the stroma. Following the administration of general and topical anesthesia, the right eye of each animal was proptosed. A 3 mm circumferential corneal incision was made parallel and 2 mm central to the corneascleral limbus in the superotemporal quadrant; the depth of the incision was approximately 50% of the stromal thickness. A No. 66 Beaver blade (Beaver, Waltham, Mass) was then used to create an intralamellar stromal pocket that extended 5 mm centrally. Five ^h of either adhesive or BSS was instilled into the stromal pocket using a micropipette (Eppendorf, Netheler and Hinz, Hamburg, West Germany). Sixteen animals received MAP-COX, 14 received CA, and five received BSS. Care was taken to keep the animal positioned so that the incision site was upright, in order to minimize gravity-induced leakage. Mild pressure was also applied to the incision site for approximately 3 minutes, in order both to allow for initial adhesive curing and to help retard leakage. At the end of each procedure, the operated eye received a 20 mg subconjunctival injection of gentamicin sulfate. During the postoperative period, erythromycin ointment was applied to each operated eye daily for the first 5 days.
The animals were examined, using a masked observer, on postoperative days 1, 3, 7, 14, 28, and 60. Examinations consisted of slit-lamp biomicroscopic evaluation and photography. A grading scale was established to evaluate the severity of clinical phenomena: 0 = absent, 1 = trace, 2 = mild, 3 = moderate, and 4 = severe. The presence and severity of the following parameters were assessed: conjunctival vascular injection, cornea stromal infiltration, corneal edema, corneal vascularization, anterior chamber cellular reaction, iris vascular congestion, and lens opacification.
Figure 3: Photomicrograph of rabbit cornea 7 days following intrastromal instillation of MAP-COX. The lamellar dissection site remains visible (arrows). There is no evidence of neovascularization or significant inflammatory response. (Hematoxylin & eosin, original magnification ? 150)
Figure 4: Photomicrograph of rabbit cornea immediately following intrastromal instillation of cyanoacrylate. Note the fibrillar appearance of the adhesive visible in the intralamellar pocket. (Hematoxylin & eosin, original magnification x 60).
At each postoperative examination time period, two animals from each study group were randomly selected for histopathological evaluation. After clinical examination, these animals were killed with an overdose of intravenous sodium pentobarbitol. Following sacrifice the corneas were removed and immediately placed in Trump's fixative at 4°C for 24 hours. They were then dehydrated in a series of graded alcohols and embedded in paraffin. Tissue blocks were sectioned at 5 µp? thickness and stained with either hematoxylin-eosin or periodic acid-Schiff (PAS).
Figure 5: Photomicrograph of rabbit cornea 14 days following intrastromal instillation of cyanoacrylate. A marked inflammatory cell response is visible in the stroma Immediately posterior to the lamellar pocket. The pocket has artifactually separated during processing and some of the adhesive has extruded out. Stromal vascularization (arrows) is also present. (Hematoxylin & eosin. original magnification x 100).
In the experimental groups, acute conjunctival vascular injection and local corneal edema were noted. In each group these phenomena were mild in degree, with average clinical grades less than 1.0, and they were totally resolved by day 7. Corneal neovascularization was noted first on day 3 in the CA group and on day 7 in the MAP-COX group. The degree of vascularization in the CA group progressed by day 14 to a mean severity score of 2.0, but then stabilized (Figure 1). In the MAP-COX group of eyes (Figure 2), stromal vascularization was markedly less, with a mean score of 0.8; it did not progress in severity during the course of the study.
Neither of the experimental groups manifested any evidence of stromal infiltration, anterior chamber reaction, iris vascular injection, or lens opacification. Throughout the follow-up period, the bluish-tinged CA was easily visible; the MAP-COX were noted to be slightly brownish in color and was visible only up to day 7.
In the control group, acute conjunctival vascular injection and local corneal edema (mean grade 1.0) was noted; these totally resolved by day 3. One of the control animals developed late (first observed on day 14), mild (clinical grade 1.0) corneal neovascularization. No other adverse effects were noted in this group.
MAP-COX group. Histopathologic sections from post-injection days 1 and 3 animals demonstrated an "acute" inflammatory reaction, manifested by an infiltration of polymorphonuclear leukocytes (PMNs) surrounding the intralamellar pocket (Figure 3). This PMN reaction diminished by day 14, although occasional PMNs were noted at each time point thereafter. Few stromal blood vessels were noted in some of the animals after day 7. There was no histopathologic evidence of the presence of the MAP-COX adhesive.
CA group. In the histopathologic sections of the CA group animals, the adhesive was easily noted (Figures 4 & 5). Specimens from acute time points (days 1 & 3) were characterized by intrastromal infiltration of PMNs; this degree of PMN response was subjectively much greater than that for the MAP-COX animals. Concurrently, the CA group had a loss of keratocytes surrounding the intrastromal pocket. From day 14 on, a mixture of PMNs and lymphocytes were noted in the stroma around the adhesive (Figure 5). Additionally, occasional multinucleated giant cells were noted in the intralamellar pocket, as well as in the surrounding stroma. Stromal neovascularization was first noted at day 3; later examinations revealed a steadily increasing degree of observable blood vessels.
Control Group. In the acute time points (days 1 & 3) a mild PMN response was noted surrounding the stromal incision site and intralamellar pocket. No inflammation was observed after day 3.
Since the 1940s, investigators have been evaluating a variety of adhesives for use in ophthalmic surgery. The first types of adhesives studied were those related to plasma proteins, specifically thrombin, fibrinogen, and fibrin.3-7,20,21 These proteins, either alone or in combination, were used experimentally to close conjunctival, corneal, and cataract wounds. Despite initial success, it became apparent that intraocular and tear film proteases weakened these fibrin-based closures, providing for unacceptable wound weakness.7 Concurrently, as better suturing materials and techniques were developed, the use of plasma protein-derived adhesives fell out of favor.
Interest in the surgical use of adhesives was rekindled following the development of synthetic plastic materials. These compounds, derivatives of CA, have been used in specific ophthalmic surgical indications. Webster and associates8 in 1968 described the use of a butyl-CA compound to seal corneal perforations.
Several other studies914 have confirmed the efficacy of butyl- and isobutyl-CA derivatives in the therapy of descemetoceles, bonding of keratoprostheses, and closure of small corneal perforations and other ocular surface wound leaks. Despite these encouraging results, the use of CAs in other anterior segment surgical procedures has to date been limited, primarily because of the impermeability and the possibilities of inflammatory reaction or tissue toxicity that may be associated with the use of these compounds.15,22
The issue of potential tissue toxicity has been associated with the use of CAs in ophthalmology. The original attempt to use these compounds clinically involved a methyl derivative of CA, known as Eastman 910. 23 Despite encouraging efficacy results, some studies demonstrated toxicity when this material was placed subconjunctival^24 and use of this material was abandoned. Although higher analogues of CA were believed to cause less tissue toxicity than the original methyl monomer, several investigators have expressed concerns regarding the ability of any CA analogue to be well tolerated.1,14,22
In 1970, Gasset and co-workers25 investigated the effects of intracorneal injection of various analogues of CA, including isobutyl, hexyl, octyl, and decyl. In their study, they observed a very low incidence of adverse clinical reactions - including neovascularization - in the experimental corneas. In contrast, in the present study, we consistently noted an early and progressive neovascularization in the corneas instilled with butyl-2-CA. Similarly, on histopathologic evaluation, we noted a greater inflammatory reaction surrounding the intralamellar CA than did Gasset and colleagues.25 These discrepancies in the results of the two studies may be related to a combination of the following factors: the use of a different CA analogue, the use of different amounts of CA, the spreading of the adhesive over a greater area of the cornea, and the selection of postoperative medications.
In the present study, we additionally examined the effects upon the cornea of a biologically derived adhesive material. The instillation of the MAP-COX adhesive was found to produce a much lower incidence and degree of adverse reactions, both clinical and histopathologic, than did the instillation of CA. In a previous efficacy study,18 we used the MAP-COX adhesive in an experimental model of epikeratoplasty and we similarly noted that the adhesive produced a minimal degree of inflammatory or other adverse reactions in the rabbit corneal stroma.
The use of adhesive materials will continue to intrigue ophthalmologists, especially those involved in corneal surgery. For over than 20 years, CA derivatives have been used successfully to stabilize descemetoceles and to close small corneal perforations. In spite of the mild-to-moderate neovascular and inflammatory reactions noted in this study, the rapid curing time of these compounds will probably ensure their continued popularity for use in these limited indications.
Biologically derived adhesive materials have the potential for much wider applicability in ophthalmic surgery. Although their curing time may be slower than that of CA, these compounds potentially have greater degrees of permeability and tissue tolerance. Previous in vitro and in vivo studies18,26 have demonstrated that mussel adhesive is sufficiently permeable to essential corneal nutrients. The present study has illustrated that this adhesive material, when placed in the corneal stroma, produces less tissue reaction than does CA. Because the MAP-COX adhesive has adequate nutrient permeability, produces minimal tissue reaction, and has demonstrated a degree of adhesion efficacy18,27,28 we believe that this compound may eventually prove to be an effective adhesive material for ocular surgery.
1. Refojo MF. Surgical adhesives in ophthalmology. Journal of Macromolecular Science Chemistry. 1970; A4:667-674.
2. Refojo MF, Dohlman CH, Koliopoulos J. Adhesives in ophthalmology: a review. Surv Ophthalmol. 1971; 15:217-236.
3. Parry TG, Laszlo GC. Thrombin technique in ophthalmic surgery. Br J Ophthalmol. 1946; 30:176-178.
4. Katzin HM. Aqueous fibrin fixation of corneal transplants in the rabbit. Arch Ophthalmol. 1946; 35:415-420.
5. Tassman IS. Experimental studies with physiological glue (autogenous plasma plus thrombin) for use in the eyes. Am J Ophthalmol. 1950; 33:870-878.
6. Town AE, Naidoff D. Fibrin closure in eye surgery. Am J Ophthalmol. 1950; 33:879-882.
7. Tassman IS. The use of fibrin coagulum fixation in ocular surgery: in keratoplasty. Transactions of the American Academy of Ophthalmology and Otolaryngology. 1949; 54:134-139.
8. Webster RG Jr, Slansky HH, Refojo MF, Burochoff SA, Dohlman CH. The use of adhesive for the closure of corneal perforation. Arch Ophthalmol. 1968; 80:705-709.
9. Hirst LW, Stark WJ, Jensen AD. Tissue adhesives. New perspectives in corneal perforation. Ophthalmic Surg. 1979; 10:58-64.
10. Hyndiuk RA, Hull DS, Kenyoun JL. Free tissue patch and cyanoacrylate in corneal perforations. Ophthalmic Surg. 1974; 5:50-55.
11. Boruchoff SA, Refojo M1 Slansky HH, Webster RG, Freeman MI, Dohlman CH. Clinical applications of adhesives in corneal surgery. Transactions of the American Academy of Ophthalmohgy and Otolaryngology. 1969; 73:499-505.
12. Weiss JL, Williams P, Lindstrom RL, Doughman DJ. The use of tissue adhesive in corneal perforations. Ophthalmology. 1983; 90:610-615.
13. Hirst LW, Smiddy WE, Stark WJ. Corneal perforations: changing methods of treatment, 1960-1980. Ophthalmology. 1982; 89:630-635.
14. Hirst LW, DeJuan E Jr. Sodium hyaluronate and corneal perforation. Ophthalmology. 1982; 89:1250-1253.
15. Cardarelli J, Basu PK. Lamellar corneal transplantation in rabbits using isobutyl cyanoacrylate. Can J Ophthalmol. 1969; 4:179-182.
16. Freyler H, Kiemen U. Gluing with fibrin in corneal surgery. Graefes Arch Clin Exp Ophthalmol. 1978; 207: 27-39.
17. Kiemen UM, Freyler H, Prskavec FH, Dinges HP. Healing of lamellar corneal grafts in rabbit eyes after gluing with highly concentrated fibrinogen. Graefes Arch Clin Exp Ophthalmol. 1979; 210:261-267.
18. Robin JB, Picciano P, Salazar J, Benedict C. A preliminary evaluation of the use of mussel adhesive protein in epikeratoplasty. Arch Ophthalmol. 1988; 106:973-977.
19. Rostron CK Brittain GPH, Morton DB, Rees JE. Experimental epikeratophakia with biological adhesive. Arch Ophthalmol. 1988; 106:1103-1106.
20. Brown AL, Nantz FA. Corneal healing. The adhesive power of aqueous fibrin in the rabbit. Preliminary report. Am J Ophthalmol. 1944; 27:1220-1224.
21. Town AE, The use of fibrin coagulum fixation in ocular surgery: in intraocular surgery. Transactions of the American Academy of Ophthalmology and Otolaryngology. 1949; 54:131-133.
22. Lehman RAW, Hayes GJ, Leonard F. Toxicity of alkyl-2cyanoacrylates, I: peripheral nerve. Arch Surg. 1966; 93:441-446.
23. Ellis RA, Levine AM. Experimental sutureless ocular surgery. Am J Ophthalmol. 1963; 55:733-736.
24. Straatsma BR, Allen RA, Hale PN, Gomey R. Experimental studies employing adhesive compounds in ophthalmic surgery. Transaction of the American Academy of Ophthalmology and Otolaryngology. 1963; 61:320-325.
25. Gassett AR, Hood Cl, Ellison ED, Kaufman HE. Ocular tolerance to cyanoacrylate monomer tissue adhesive analogues. Invest Ophthalmol Vis Sci. 1970; 9:3-11.
26. Green K, Berdecia R, Cheeks L. Mussel adhesive protein: permeability characteristics when used as a basement membrane. Curr Eye Res. 1987; 6:835-838.
27. Picciano PT, Benedict CV. Mussel adhesive protein: a new ophthalmic tissue adhesive and cell attachment factor. Invest Ophthalmol Vis Sci. 1986; 27(suppl):31.
28. Robin JB, Picciano PT, Benedict CV. A preliminary evaluation of mussel adhesive protein for the acute treatment of corneal perforations. In: Cavanagh HD, ed. The Cornea: Translations of the World Congress on the Cornea III. New York, NY: Raven Press; 1988:55-58.