Periocular Autografts in Socket Reconstruction

Hilary A. Beaver, MD, 1 James R. P atrinely, MD, 1,2 John B. Holds, MD /,4

Marcus P. Soper, Be01

Background: Current enucleation and socket reconstruction techniques often re­quire reinforcement of an orbital implant or wound by the use of a tissue graft. Commonly, allograft tissue (cadaveric sclera, fascia, etc.) is used. Disadvantages of allografts include possible inflammatory reaction, unpredictable vascularization rate, variable resorption, antigenicity, and cost. Another alternative to implant reinforcement is autogenous tissue which usually is harvested from a remote site (fascia lata, split dermis, temporalis fascia, pericranium, etc.). An overlooked source of readily available autogenous grafts is the connective tissue in the periorbital region and the enucleated eye itself. These sources include autogenous lamellar sclera, the corneoscleral button, capsular tissue from a migrated implant, and orbital rim periosteum.

Methods: The authors used periocular autografts in primary and secondary socket reconstructions in 24 patients, with excellent success over a 2-year period. Seven autoge­nous corneoscleral buttons and two autogenous scleral grafts were used to cover biointe­grated implant spheres. Ten implant capsules from migrated nonintegrated spheres were used to cover and reinforce secondary implants. In five patients, an autogenous periosteal graft was taken from the supraorbital rim and used to cover an exposed implant.

Results: Complications included one pyogenic granuloma, one conjunctival inclu­sion cyst, and one recurrent exposure after a periosteal graft, which necessitated explana­tion and a dermis fat graft. Postoperative motility was judged to be good to excellent in 21 patients.

Conclusions: These techniques are presented as a new alternative to using human bank tissue or remote incision autografts for reconstruction of the anophthalmic socket. Ophthalmology 1996; 103: 1498-1502

There is often a need for implant or wound reinforcement to reconstruct the anophthalmic socket. Both autologous tissue (e.g., fascia lata, temporalis fascia, dermis, etc.)l-5 and homologous tissue (e.g., cadaveric sclera)2,6-8 have

been used. Autologous tissue is viable, compatible, has more rapid vascularization and less inflammation, but usually requires a remote incision and added surgical time. Homologous tissue does not require a second wound and is easily available but is inert, provides only a colla­gen matrix and barrier and not a living graft, and may elicit a foreign body response with variable resorption and inflammation. Periocular tissue offers the advantages of an autologous graft but neutralizes the problems associ­ated with remote incisions.2,9-11 We report our technique of using autologous periocular tissue to reinforce primary and secondary enucleation implants in 24 patients.

Originally received: May 10, 1995. Revision accepted: May 14, 1996.

1 Department of Ophthalmology, The Cullen Eye Institute, Baylor Col­lege of Medicine, Houston.

2 Division of Plastic Surgery, Baylor College of Medicine, Houston.

3 Department of Ophthalmology, The Anheuser-Busch Eye Institute, St. Louis University School of Medicine, St. Louis.

4 Department of Otorhinolaryngology, The Anheuser-Bu'sch Eye Insti­tute, St. Louis University School of Medicine, St. Louis.

Presented at the American Society of Ophthalmic Plastic and Re­constructive Surgery Annual Meeting, San Francisco, October 1994.

Each author states that he/she has no proprietary interest in tjJe develop­ment or marketing of this or a competing product.

1498

Supported in part by an unrestricted grant to the Department of Ophthal­mology, Baylor College of Medicine, and St. Louis University from Research to Prevent Blindness, Inc, New York, New York.

Reprint requests to James R. Patrinely, MD, Department of Ophthalmol­ogy, Cullen Eye Institute, 6501 Fannin, NC-200, Houston, TX 77030.

Beaver et al . Periocular Autografts in Socket Reconstruction

Materials and Methods

We retrospectively reviewed the medical records of pa­tients who underwent periocular autologous tissue graft­ing after primary enucleation or secondary implant revi­sion from August 1991 to May 1994 at two regional referral medical centers. Indications for periocular auto­grafts in primary enucleation were severe ruptured globe or advanced phthisis bulbi. Indications for revision of an exposed biointegrated enucleation implant were moder­ate-sized anterior defect ( <7 mm), otherwise normal con­junctiva, no evidence of infection, good implant motility, and lack of forniceal shortening. Exposed primary im­plants that did not meet the above guidelines usually were explanted with secondary implant placement. Parameters examined include patient demographics, surgical indica­tion, and procedure. Information obtained on follow-up examinations included the presence or absence of con­junctival inflammation and wound healing; general socket condition; conjunctival forniceal depth; socket motility; implant position; and postoperative complications.

Surgical Techniques

Enucleation techniques with anatomic reattachment of the recti muscles is well described.3

•6

•7 In our technique, after

a standard primary enucleation is performed, a corneal button with a surrounding 1- to 2-mm scleral rim is ex­cised from the enucleated eye. The corneal epithelium and endothelium are debrided completely, using cotton swabs and a #15 Bard-Parker blade (Storz Ophthalmic Instruments, St. Louis, MO). The corneoscleral button and a biointegrated spherical implant (porous polyethyl­ene or hydroxyapatite) are soaked in gentamicin or baci­tracin solution. The corneoscleral button is sutured di­rectly to the planned anterior surface of a porous polyeth­ylene sphere with interrupted 4-0 polyglactin suture on a P-2 needle. This provides an additional tissue barrier anteriorly to prevent exposure. The majority of the im­plant is left unwrapped to promote more rapid vasculari­zation. Hydroxyapatite implants are too brittle to allow direct suturing of the corneoscleral button. Therefore, the graft either may be placed directly on the anterior surface of the hydroxyapatite sphere or secured with sutures passed through preplaced holes in the implant made with a 19-9auge needle or drill (Fig 1).

Temporarily wrapping the orbital implant with a piece of plastic drape facilitates its subsequent placement within Tenon space. This decreases the capture of the orbital tissue on the rough posterior surface of the implant and facilitates deep orbital positioning. The implant is placed within the Tenon capsule and held gently in place with one finger while the drape is removed.6 The rectus mus­cles then are attached directly to the autologous graft with preplaced muscle sutures and the Tenon capsule and conjunctiva are closed anteriorly over the implant in sepa­rate layers.

When inadequate autologous corneal tissue is available

~~d~-~la;om'rn.II~:aor'lCscl~leraral~~raft

A.

~1995 Baylor College of Medicine

Tenon's capsule

Figure 1. Technique of ocular autografting to a porous orbital implant. A, full-thickness comeoscleral button or similar diameter lamellar scleral graft is harvested from the enucleated eye. B, the ocular graft is sewn onto the planned anterior face of the porous implant. C, the extraocular muscles are attached at the implant-graft junction. Anterior Tenon layer and conjunctiva are closed over the smooth graft surface.

for grafting, autologous sclera may be harvested from the enucleated eye using at least a lO-mm diameter section. The sclera is harvested and the inner lamella is removed with sharp blade dissection. The outer lamellar graft then is sutured or placed on the anterior surface of the implant as previously described.

In secondary implant revisions of extruded or migrated primary implants, there is no available autologous cornea or sclera. In these cases, the implant and pseudocapsule are removed in one of two ways. In the first way, the migrated sphere is dissected free from the orbit with its surrounding capsule intact. The implant and pseudocap­sule are removed as one unit after tagging and detaching the extraocular muscles. The pseudocapsule then is re­moved from the implant, soaked in antibiotic solution, and attached to a replacement sphere as an anterior cap graft before implantation in the orbit. The muscles are reattached to the autograft in the anatomic position. Addi­tional inner surface debridement or cryotherapy is per­formed if there is a concern about epithelial ingrowth in cases of extruding implants. The second technique in­volves first removing the migrated implant, leaving the pseudocapsule in the orbit. The posterior two thirds of the implant capsule then is excised and used as an anterior cap graft over the secondary implant. The anterior portion of the pseudocapsule is left intact in the orbit to help reinforce the wound and maintain its relation to the extra­ocular muscles. The choice of technique depends on the preoperative socket motility, thickness of the pseudocap­sule, and ease of dissection.

Frontal periosteum also may be used to reinforce ex­posed biointegrated enucleation implants. An incision is made in the temporal eyelid crease, and the orbicularis muscle is separated bluntly to expose the superior orbital rim. A 15 X l5-mm area of frontal bone periosteum is

1499

Ophthalmology Volume 103, Number 9, September 1996

Figure 2. Frontal periosteum harvest technique through an eyelid crease incision.

incised with a #15 Bard-Parker blade. The periosteal graft is dissected loose with a Freer elevator (Storz Ophthalmic Instruments) (Fig 2). The donor site is closed in layers, and the periosteal graft is sutured as before to the anterior surface of the orbital implant. Frontal periosteum also may be used to cover exposed but viable biointegrated implants. The conjunctiva surrounding the exposed site sharply is dissected and elevated, and the patch graft of periosteum is sutured anterior to the implant but deep to the conjunctiva.8

,12

In this study, any patient with sufficient viable ocular tissue and no contraindications to ocular auto grafting (i.e., tumor, infection, etc.) who has undergone primary enucle­ation had a cornea or lamellar sclera graft placed to mini­mize operating room time and to avoid a second incision for another graft. Periosteal autografts also could be used in primary enucleation if the patient wanted to minimize any risk of sympathetic ophthalmia or had poor quality or amount of ocular tissue (Fig 3), Patients with prior enucleation or evisceration (1 patient) who required a secondary implant had a regional periosteal graft if the pseudocapsule was of poor quality.

The relative risks and benefits of regional periosteal grafts versus ocular autografts and the theoretical risk of sympathetic ophthalmia were explained to each patient through informed consent.

Results

The patients ranged in age from 10 to 52 years (mean, 30.2 years). Of the 24 patients who underwent surgery,

1500

9 had primary implants (38%), 11 had secondary implants (46%), and 4 had implant revisions (17%), all in porous polyethylene spheres. None of the exposed primary im­plants requiring these techniques were hydroxyapatite. Autologous graft tissue included cornea in seven patients (29%), sclera in two (8%), implant capsule in ten (42%), and periosteum in five (21 %).

Follow-up ranged from 1 to 49 months (mean, 18 months). The surgical indications in these 24 patients included a blind, severely phthisical, painful eye in one patient (4%); no light perception vision after attempted primary repair of extensive globe rupture in eight (33%); migration or extrusion of primary implants in ten (42%); exposed but viable biointegrated implant in four (17%); and chronic inflammation with epithelial downgrowth after evisceration in one (4%).

In 23 patients (96%), postoperative conjunctival in­flammation was mild and resolved within the first 2 weeks. One patient had protracted chemosis after a gun­shot wound with extensive globe rupture and primary repair. Wound healing was rapid, and thick, well-vascu­larized conjunctiva was present in 23 patients (96%). Im­plant position was central and the socket was healthy in all 24 patients (100%). The conjunctival fornices were unchanged or improved in 23 patients (96%) but were slightly shortened in one patient who had an enucleation after extensive globe rupture (4%). Socket motility was good to excellent in 21 patients (88%), fair in 1 after a secondary implant (4%), poor in 1 after a gunshot wound with extensive rupture (4%), and not available in 1 (4%).

Figure 3. Three potential harvest sites for periocular autografts are the orbital rim periosteum, the corneoscleral button, and a lamellar scleral graft.

Beaver et al . Periocular Autografts in Socket Reconstruction

In addition, chronic inflammation and pain resolved com­pletely in one patient with epithelial downgrowth due to corneal perforation after undergoing a secondary implant after evisceration (4%). Two patients undergoing second­ary implants for implant migration had improvement of previous superior sulcus deformities (8%).

None of the patients had donor site complication, sym­pathetic ophthalmia, or postoperative infection. Compli­cations were reported in three patients. These included one polyethylene sphere with residual exposure eventu­ally requiring explanation and dermis-fat graft (4%). One patient who had a primary polyethylene sphere covered with a corneoscleral autograft developed anterior expo­sure 4.5 months postoperatively during aggressive topical steroid use to treat a pyogenic granuloma (4%). This ex­posure healed on discontinuing the steroids and the patient has maintained excellent socket contours with 35 months of follow-up. One patient had an uncomplicated small conjunctival inclusion cyst (4%) after primary polyethyl­ene sphere implantation with a corneoscleral autograft after a severe orbital injury due to a gunshot wound to the orbit with severe soft tissue injury. No further revi­sions were required, and the patient was asymptomatic at 4.5 months of follow-up.

Although the implant revision group represents a differ­ent subset of patients than primary enucleations, who are not exposed to the risk of sympathetic ophthalmia, we be­lieved it was important to review both groups of patients. These patients represent a spectrum of needs of autologous periocular tissue that can be harvested from the globe (enu­cleation group), frontal periosteum (either group), or implant pseudocapsule (revision group).

Discussion

A variety of orbital implants have been used to reconstruct the anophthalmic orbit in primary enucleation. These in­clude spheres made of silicone, methylmethacrylate (acrylic), porous polyethylene, and hydroxyapatite.2.12-14 Dermis-fat grafts are useful in patients with mild conjunc­tival deficiency. I Biointegrated motility implants (i.e., hy­droxyapatite and porous polyethylene) have an irregular, rough surface that makes insertion difficult and may incite an inflammatory reaction, increasing the chance of expo­sure.5,12.13 To prevent this, many surgeons reinforce the biointegrated implant anteriorly with autologous or ho­mologous tissue.

Homologous tissue for wrapping implants, such as hu­man donor sclera or donor fascia lata, is easier to use, but adds both to cost and to the theoretical risk of trans­mission of infectious agents. In addition, allogenic tissue is recognized as a foreign material and may result in a greater postoperative host inflammatory response, with subsequent variable tissue resorption and unpredictable vascularization of the graft. The homologous material is, at best, a barrier and matrix for tissue ingrowth.

Autologous grafts, such as dermis, temporalis fascia, fascia lata, and pericranium, induce less inflammatory response and, being living tissue, integrate faster,5 Recent

retrospective studies of porous spherical implant compli­cations showed that autologous grafts were superior to homologous grafts in preventing exposure. 15 A remote incision site, however, is required to harvest the graft. This second site may require the time and skills of an additional surgeon, increase the risk of scarring and infec­tion, and prolong recovery. Although the harvest of fron­tal periosteum requires a second incision, it is performed through a standard eyelid crease approach which is famil­iar to most ophthalmic surgeons and has little potential to leave a visible scar, Also, the periosteal graft may be raised with some of the overlying fat and fascia, providing more thickness than other fascial (temporalis, fascia lata, etc.) grafts.

While evisceration generally provides superior motility as compared with enucleation, there are specific instances in which enucleation is preferred over evisceration. Ex­cepting intraocular tumors, these include a severely rup­tured globe and advanced phthisis bulbi. In these patients, an excellent source of autologous graft tissue, the globe, usually is discarded. A few reports have demonstrated success using autologous periocular tissue in orbital re­construction. YulO described 11 implants wrapped in au­tologous sclera with no episodes of extrusion or migration after follow-up ranging from 6 months to 2 years. Nunery et a12 also reported the use of autologous sclera for implant reinforcement. Tenzel et al9 reported using the fibrous capsule from an extruded primary implant to reconstruct the inferior fornix. We describe the use of autologous frontal periosteum, implant capsule, cornea, and sclera as tissue grafts for orbital implants in selected patients. These techniques represent an intermediary between enu­cleation and evisceration in which some of the ocular tissue is preserved and used in the reconstruction. Due to the theoretical risk of sympathetic ophthalmia, we use the less antigenic cornea and outer sclera. Autologous ocular tissue should not, of course, be used from eyes enucleated for intraocular malignancy or infection. The use of re­gional periosteum alleviates these concerns. None of our 24 patients had significant wound complications, sympa­thetic ophthalmia, or postoperative infections.

In summary, autologous periocular tissue grafts can be used successfully in most patients with primary or secondary orbital implants with decreased postoperative inflammation, excellent socket healing and motility, and low complication rates.

Acknowledgment. The authors thank Edward K. Chan, MD, for translating reference lO.

References

l. Nunery WR, Hetzler KJ. Dermal-fat graft as a primary enucleation technique. Ophthalmology 1985; 92: 1256-61.

2. Nunery WR, Cepela MA, Heinz GW, et al. Extrusion rate of silicone spherical anophthalmic socket implants. Oph­thalmic Plast Reconstr Surg 1993;9:90-5.

3. Oberfield S, Levine MR. Diagnosis and treatment of com­plications of enucleation and orbital implant surgery. In: Bosniak SL, ed. Advances in Ophthalmic Plastic and Re-

1501

Ophthalmology Volume 103, Number 9, September 1996

constructive Surgery, the Anophthalmic Socket. New York: Pergamm Press, 1990; 107 -17.

4. SolI DB. Donor sclera in enucleation surgery. Arch Oph­thalmol 1974; 92:494-95.

5. Petrelli RL. Use of autogenous materials in reconstructing the anophthalmic socket. In: Bosniak SL, ed. Advances in Ophthalmic Plastic and Reconstructive Surgery, the Anoph­thalmic Socket. New York: Pergamm Press, 1990; 153-69.

6. Perry AC. Advances in enucleation. Ophthalmic Plast Re­constr Surg 1991;4:173-82.

7. Walter WL. Update on enucleation and evisceration sur­gery. Ophthalmic Plast Reconstr Surg 1985; 1:243-52.

8. Goldberg MF. A simplified scleral graft technique for cov­ering an exposed orbital implant. Ophthalmic Surg 1988; 19:206-1l.

9. Tenzel DP, Conn H, Doss RP. Fibrous capsule from ex­truded implant used as graft material in cul-de-sac recon­struction. Ophthalmic Plast Reconstr Surg 1989;5:216-9.

1502

10. Yu CT. Clinical use of orbital implant in autogenous sclera. Chung -Hua Yen Ko Tsa Chih, Chinese J oumal of Ophthal­mology 1989;25:146-7.

11. McCarthy RW, Swann ES. Autogenous scleral graft in im­plant extrusion. Ophthalmic Surg 1980; 11 :686-87.

12. Kim Y, Goldberg RA, Shorr N, Steinsapir KD. Manage­ment of exposed hydroxyapatite orbital implants. Ophthal­mology 1994; 101:1709-15.

13. Karesh JW, Dresner SC. High density porous polyethylene (Medpor) as a successful anophthalmic socket implant. Ophthalmology 1994; 101: 1688-96.

14. Perry AC. Integrated orbital implants. In: Bosniak SL, ed. Advances in Ophthalmic Plastic and Reconstructive Sur­gery, the Anophthalmic Socket. New York: Pergamm Press, 1990; 75-8l.

15. Remulla HD, Rubin PAD, Shore JW, et al. Complications of porous spherical orbital implants. Ophthalmology, 1995; 102:586-593.