High density Porous Polyethylene (Medpor) as a Successful Anophthalmic Socket Implant

High density Porous Polyethylene (Medpor) as a Successful Anophthalmic Socket Implant

by James W. Karesh, MD, and Steven C. Dresner, MD, November 1993

  • Presented at the American Academy of Ophthalmology Annual Meeting, Chicago, November 1993.
  • Published in Ophthalmology, Vol. 101, No. 10, October 1994 (Ophthalmology 1994;101:1688-1696)
  • The authors have no proprietary interest in the development or marketing of Medpor or in Porex Technologies Corporation.

Background: High-density porous polyethylene (Medpor) has been used successfully as an implant in orbital fracture repair and in the management of both cosmetic and post-traumatic facial deformities.  The material is well tolerated, resists infection, is nonantigenic, and promotes tissue ingrowth.  Additionally, sutures can be passed through it.  These characteristics led to its use as an implant in anophthalmic socket surgery.

Methods: Twenty-one patients with anophthalmia underwent implantation of spherically shaped high-density porous polyethylene implants.  The implant was used in six primary enucleations with direct extraocular muscle attachment, ten secondary implant insertions, one repeat secondary implant insertion, and five eviscerations.

Results: Postoperative follow-up averaged 19 months.  During this period, there were no extrusions, infections, significant inflammatory responses, or implant exposures.  One implant was removed 4 months after insertion due to inadequate initial placement secondary to a severe post-traumatic orbital deformity.  Successful re-implantation was performed without complication.  Postoperatively, all sockets showed good to excellent motility.  Results of histologic examination of the removed implant demonstrated minimal inflammatory response and extensive fibrovascular ingrowth involving 60% of the implant substance.

Conclusions: This initial report indicates that high-density porous polyethylene [Medpor] can be used successfully as an implant in anophthalmic socket surgery.  Its advantages over other similar implants include a significantly lower material cost and the ability to suture the extraocular muscles directly to it without the need for a covering material such as fascia or sclera.  Experimental studies are in progress to determine if this implant can be integrated with an ocular prosthesis to promote improved motility and cosmesis.

The search for the ideal implant for the anophthalmic socket continues to evolve.  The purpose for such an implant is to mimic as closely as possible the presence of a normal eye in the anophthalmic socket.  This implant must be able to occupy sufficient orbital volume to compensate for the absent globe.  It also should have minimal rates of extrusion, exposure, and infection without significant inflammation.  Finally, the implant must promote socket motility and, more particularly, prosthesis motility to duplicate as much as possible the normal movement of the globe.

Many materials and implant types have been used to achieve cosmesis and prosthesis motility in the anoph thalmic socket.  Totally buried quasi-integrated implants with mounds such as the Universal and Allen implants and partially buried integrated implants have been used most commonly to realize these ends (1-3).  However, the motility associated with the former in many cases is not much better than that associated with a simple methylmethacrylate sphere and the latter implant almost invariably results in extrusion.  Polytetrafluoroethylene (Goretex) covered spherical implants with the muscles attached to the polytetrafluoroethylene improve socket motility and promote tissue ingrowth but do not allow for true integration with the prosthesis (4) (authors’ personal experience).

Current interest in integrated implants centers around those made with coralline hydroxyapatite (Integrated Orbital Implants, Inc, San Diego, CA).  This is a naturally occurring porous material that promotes tissue ingrowth, is well tolerated, and can be coupled to a prosthesis through the use of a peg and sleeve mechanism placed into a hole drilled into the hydroxyapatite after adequate vascular ingrowth has been demonstrated by bone scan or magnetic resonance imaging (3,5-7). There are several possible drawbacks to this implant.  A scleral or facial covering over the hydroxyapatite usually is needed to attach the extraocular muscles to the implant (7,8).  This may represent a risk to the patient for transmission of slow viruses, hepatitis, or human immunodeficiency virus.  If uncovered or inadequately vascularized, the rough surface of the implant may cause erosion of the overlying conjunctiva and Tenon’s capsule and implant exposure (3).  Fibrovascular ingrowth into the hydroxyapatite may take a variable amount of time or may be incomplete, preventing the insertion of the peg and sleeve mechanism due to inadequate epithelialization of the hole drilled into the implant.  A bone scan or magnetic resonance imaging is necessary to confirm adequate fibrovascular ingrowth (3,5-7). Finally, the implant itself represents a considerable expense to the patient.  This may present a problem for medical insurance coverage in an age of cost-effective medical care.

The current study was undertaken to determine if an implant composed of high-density porous polyethylene (Medpor, Porex Technologies Corporation, Fairburn, GA) could be successfully used in the an ophthalmic socket as an alternative to hydroxyapatite to overcome some of the drawbacks associated with this latter material.  The current report represents a preliminary clinical study which examines the results of using this material as an anophthalmic socket implant and the histologic changes associated with its implantation.

Patients and Methods

During the 3 1/2-year period, ending January 1993, 21 patients underwent implantation of a spherical high-density porous polyethylene anophthalmic socket implant.  These implants are manufactured from medical-grade high-density polyethylene (molecular weight, 21,000).  Seventeen patients underwent surgical intervention between June 1991 and January 1993 (18 months).  Four patients underwent surgery during 1989 and 1990 before this material was commonly available as a spherical orbital implant.  This only occurred in mid-1991.  After this time, spherical porous polyethylene implants were used in all patients requiring enucleation, evisceration, or removal and replacement of an an ophthalmic socket implant.  In this group, there were 5 eviscerations for non-seeing, phthisical eyes, 11 secondary anophthalmic socket implants for extruding or inadequately positioned implants, and 6 enucleations (3 for tumor removal, 2 for non-seeing, painful eyes, and 1 for globe rupture after trauma).  In one patient (case 18), an inadequately placed high-density porous polyethylene implant was exchanged with a smaller and more posteriorly placed implant of the same material to facilitate the fitting of an ocular prosthesis.

Porous polyethylene implants are provided by the Porex Technologies Corporation in sterile, protective envelopes.  They are only removed from these containers before use.  These implants cannot be resterilized.  Before handling the implant, sterile gloves must be thoroughly rinsed with sterile physiologic saline to remove all traces of powder and other foreign material.  After removal from the sterile envelope, the implant should be placed in a basin containing sterile physiologic saline.  This reduces the incidence of contaminating the implant with powder or foreign material.  The implant should not be placed on any surface that might contaminate it with lint or other particulate material.  Before insertion, the implant was soaked for 20 minutes in saline to which 1 g vancomycin HCI was added.  All patients received 1 g cefazolin sodium intravenously just before the initiation of the surgical procedure.  Patients received cephalexin (250 mg orally every 6 hours for 48 hours) after surgery.  All patients were fitted with a permanent prosthesis 4 to 6 weeks after surgery .

Evisceration is performed by making a 360° incision through the conjunctiva and Tenon’s capsule as close to the limbus as possible.  The corneal button is excised, and an evisceration spoon is used to remove the entire contents of the globe.  A sterile gauze pad is used to abrade and remove any adherent uveal tissue.  Direct unipolar cautery of any persistently bleeding vessels is performed as necessary.  Transscleral relaxing incisions are made at the 2-,4-, 8-, and 10-o’clock positions to allow placement of a 16- or 18-mm high-density porous polyethylene [Medpor] sphere within the evisceration cavity.  The vertical and horizontal scleral flaps created by the relaxing incisions then are sutured over the implant in two layers using 5-0 polyglactin 910 suture.  The conjunctiva is closed with similar material, and a conformer is placed in the fornices until the permanent prosthesis is fashioned.

Enucleations are performed in standard fashion.  After the conjunctiva and Tenon’s capsule is opened for 360° as close to the limbus as possible, blunt dissection with scissors is performed in each of the four quadrants between the rectus muscles.  The rectus muscles are isolated individually and hooked with a muscle hook and a double-armed 5-0 polyglactin 910 suture with small spatulated needles is passed through the muscle insertion and locked at either edge of the muscle and its tendon.  The muscles then are cut individually from their insertions, taking care not to cut either the sutures or the sclera.  The inferior and superior oblique muscles are hooked and incised without placing any sutures in their tendons.  A gently curved enucleation scissors then is used to bluntly dissect through the posterior layer of Tenon’s capsule and incise the optic nerve as far posteriorly in the orbit as possible.  Bleeding is controlled with gauze soaked in 1% phenylephrine HCI and digital pressure for 5 minutes.  Unipolar cautery is used as necessary to control any persistent bleeding from the posterior orbit.

An appropriately sized high-density porous polyethylene spherical implant, usually 18 or 20 mm as determined using a methylmethacrylate sphere for sizing, is placed into the enucleation cavity after all bleeding is controlled.  Using the previously placed 5-0 polyglactin sutures in the muscle insertions, the muscles are sutured to the implant sphere to approximate their normal insertions along the spiral of Tillaux (Fig 1).  Only very superficial bites are required to attach the muscles.  Deep suture bites will bend or break the needle.  However, some pressure is necessary to push the needle through the implant pores.  High-density porous polyethylene is adherent only minimally to the surrounding orbital tissue and can be placed easily into the orbit.  Wetting the implant is helpful in performing this placement.  Small malleable retractors also can be used.  After the implant is correctly positioned in the orbit, the anterior portion of Tenon’s capsule and the conjunctiva are closed separately over it with interrupted 5-0 polyglactin 910 sutures.  A conformer is placed in the fornices and the eyelids are sutured closed over this for 7 to 10 days.  A double eye patch is taped in position for 24 hours.  The conformer is left in place until a permanent prosthesis is fashioned.

When secondary implant placement is performed, a horizontal incision is made across the remaining conjunctiva and other tissue present over the previously placed implant.  The implant is removed along with any capsular material covering it.  No attempt is made to isolate the rectus muscles. An appropriately sized high-density porous polyethylene implant is placed, as was done after enucleation, and Tenon’s capsule and the conjunctival are closed over the implant with interrupted 5-0 polyglactin 910 sutures.  The high-density porous polyethylene markedly adheres to the surrounding tissues; therefore, suturing of the Tenon’s capsule in the area of any presumed muscle insertions to the implant is unnecessary.  A conformer is placed as already described.


Postoperative follow-up for this group of patients averaged 19 months (range, 7-43 months).  During this period, there were no infections, significant episodes of inflammation, or implant exposures or extrusions (Table 1).  No patient experienced significant postoperative edema or pain.  All except one patient were easily fit with a custom-formed prosthesis 4 to 6 weeks after surgery.  One patient (case 18) with a severely deformed orbit after previous orbital trauma required removal of a high-density porous polyethylene implant 4 months after its placement.  At the time this implant was inserted, placement deep into the orbit was difficult due to the significant post-traumatic bony changes that were present.  When the permanent prosthesis was fashioned, the ocularist found that the superior nasal aspect of the socket was inadequate, despite several attempts at redesigning the prosthesis.  For this reason, the original high-density porous polyethylene implant was removed and a smaller implant was placed.  After this surgery, the patient was fitted successfully with a custom conformer without difficulty.

Table 1.  Clinical Data for Patients Undergoing Medpor Implantation


Age (yrs)


Follow-up (mos)


 Type of Surgery

Implant Size

764F21OSSecondary implant18
922F20OSSecondary implant20
1073F16OSSecondary implant18
1163M16ODSecondary implant18
1253M16OSSecondary implant18
1426F15OSSecondary implant18
1743F12ODSecondary implant18
1863F4 (implant removed)OSSecondary implant18
1811OSSecondary implant16
1910M11ODSecondary implant18
2131F7ODSecondary implant16
  OD = right eye; OS = left eye

Socket motility was subjectively determined by observing the movement of the socket and fornices while asking the patient to move the contralateral normal eye into the various cardinal positions of gaze.  Clinically, all patients with muscles attached to the high-density porous polyethylene implant at the time of enucleation demonstrated excellent socket motility (Fig 2).  This was also true for the postevisceration sockets.  Good to very good socket motility was present in all except one (case 18) of the patients who had secondary implant.  That patient had only fair socket motility which was believed to be due to the fact that she had previously undergone multiple interventions to her orbit and socket with the formation of significant scarring.

At the time of surgical removal, the one high-density porous polyethylene implant to be examined histopathologically failed clinically to demonstrate the presence of any capsule formation.  Rather, the surrounding fibrous tissue was deeply adherent into the implant material.  Extensive sharp dissection was required for removal.  Results of examination of this implant after removal demonstrated numerous tufts of tissue embedded into all areas of the implant.  Upon hemisectioning, it appeared that this tissue was present through at least the outer 60% of the implant.

The removed implant was submitted for histopathologic examination.  On both routine staining and Masson-trichrome staining, it was noted that extensive fibrovascular ingrowth was present throughout the outer 6-mm thick rim of the implant, leaving all except the most central 6-mm spherical area nonvascularized (Fig 3).  The central portion of the implant was missing, presumably due to histologic processing and the absence of any fibrovascular skeleton to hold it to the surrounding well-vascularized material.  Only minimal fibrous capsule or pseudocapsule formation was observed.  Significant vascular tissue and vascular channels with erythrocytes were present (Fig 4).  No giant cells were noted, and only few inflammatory cells and macrophages were present.  A network of fibrovascular tissue could be seen within the high-density porous polyethylene matrix in all areas examined.

Hemisection of Medpor implant, 1X magnification Outer section of Medpor implant, 10X magnification. Outer section of Medpor implant, 200X magnification.

Top, Figure 3.

Top left, hemisection of the implant removed from case 18 demonstrates tissue ingrowth throughout the external 6 mm of the high-density porous polyethylene substance [Medpor] (hematoxylin-eosin; original magnification, X1).

Top right, higher power view of the outer section
of the high-density porous polyethylene implant. Notice the minimal pseudocapsule formation (hematoxylin-eosin; original magnification, X10).

Bottom, Figure 4. High-power view of the high-density porous polyethylene implant shows vascularized channels and dense fibrovascular ingrowth (hematoxylin-eosin; original magnification, X200).


The fabrication of the “ideal” implant for facial and orbital reconstruction and for socket implantation has been elusive.  Such an implant would resist infection, incite little or no inflammation, result in minimal rates of extrusions, and be nonantigenic and chemically and biologically inert.  In addition, this material would have minimal weight and be malleable and moldable while maintaining structural rigidity.  Certain other characteristics specific for anophthalmic socket reconstruction would include the ability to promote implant motility by attaching the extraocular muscle directly to the implant and the potential for the prosthesis and implant to be linked to achieve the widest range of motion, approximating normal ocular motility.  Naturally, factors such as ease of manufacture and the costs associated with the material itself as well as any specialized techniques or processes associated with the surgical implantation of this material are also important.

Recent research into the “ideal” implant has centered around the use of porous materials that facilitate tissue ingrowth thereby improving long-term implant retention and reducing the rate of extrusion.  The most important requirement for tissue ingrowth is a pore size of at least 40 µm (9).  Certainly, larger pore sizes are more desirable with a pore size of at least 100 µm for bony ingrowth and at least 150 µm or more for ensuring the most consistent ingrowth (10) (Porex Technologies Corporation, personal communications). In addition, complete vascularization and tissue integration into such an implant would allow it to be coupled directly to a prosthesis through the use of an external peg placed in a epithelialized hole within the implant.  The peg then could fit into an excavation in the back of the prosthesis, transferring movement from the implant to the prosthesis.  This would avoid the extrusion problems associated with partially exposed integrated implants.

Currently, corallin hydroxyapatite implants with a linear pore structure and pore diameter of 500 µm are being used successfully as buried integrated implants (8,11).  They are well tolerated, but can incite significant inflammatory response on histopathologic evaluation (Goldberg RA, unpublished data).  While readily vascularized and integrated with fibrovascular tissue, there are increasingly more reports of implant exposure and extrusion (12-14).  Nonetheless, they provide excellent prosthesis motility more closely approximating the normal eye than any other currently available implant.  However, there are drawbacks to this material.  Due to a rough surface and rigid structure, they should be covered with sclera or fascia to reduce the incidence of postoperative exposure and for attaching the extraocular muscles (7,8).  Unfortunately, this contributes to the expense of this implant which is already twice as costly as other porous implants.  In addition, donor sclera is associated with a theoretical risk for the transmission of human immunodeficiency syndrome, slow viruses, and hepatitis.  Another expense in addition to the surgical procedure associated with this material is the need to drill it for insertion of the coupling peg.  Finally, as vascularization of hydroxyapatite proceeds at a variable rate, a bone scan or magnetic resonance imaging is required to ensure complete vascularization before drilling the implant (3,5-7).  Without first ensuring adequate vascularization, the hole drilled into the hydroxyapatite may not epithelialize, resulting in exposure and chronic infection (3,7).

Polyethylenes are straight chain aliphatic hydrocarbon chains created by the polymerization of ethylene molecules under high pressure and temperature.  Because the polymerization process determines the length of the polyethylene chains, the molecular weight of this material can vary from as low as 1000 to as much as 38,000 (15).  Lower weight polyethylenes are used in lubricants, whereas higher weight ones form solid materials that can be molded into a variety of shapes and are found in toys, containers, and building materials.  The high-density porous polyethylene used in Medpor is a medical grade implantable material that received Food and Drug Administration approval for use in reconstructive surgery in 1985 (Porex Technologies Corporation, personal communication).  However, it has been used in facial reconstructive procedures since 1947 (15).  The molecular weight of this material is approximately 21,000 (15).  This implantable material is formed through sintering powdered polyethylene.  In this procedure, powdered material is heated to a temperature just below its melting point and then molded into the desired form or compressed into sheets (Porex Technologies Corporation, personal communication). By controlling this process, a finished product can be produced which is both porous and malleable and at the same time structurally rigid.  Through the manufacturing process, the pore size can be controlled within a fairly precise range.  In the high-density porous polyethylene sheets used for facial and orbital reconstruction, the pore size ranges from 100 to 250 µm (average, 150 µm), whereas in the spherical implants used for anophthalmic socket surgery the range is 100 to 500 µm (average, 200 µm) (Fig 5)Close-up views of Medpor(Porex Technologies Corporation, personal communication).  In these implants, over 85% of the pores have a diameter greater than 150 µm (Porex Technologies Corporation, personal communication).  The pore structure in all high-density porous polyethylene implants is omnidirectional due to the nature of the sintering process (Porex Technologies Corporation, personal communication).

Experimental studies of fibrovascular ingrowth within high-density porous polyethylene have indicated that complete tissue ingrowth into 3-mm sheets of this material occurs over a 5- to 12-week period (16).  One study examining implants removed from dogs 2 years after implantation demonstrated complete integration of this material into surrounding bone and soft tissue (Porex Technologies Corporation, personal communication).  More recently, a study of 14-mm high-density porous polyethylene implants placed in rabbit orbits for periods ranging from 4 to 48 weeks showed that at 24 to 48 weeks the implant was vascularized throughout except in its most central third segment (Goldberg RA, unpublished data).  A similar picture was demonstrated by the one implant we examined histopathologically.  However, this contrasts with studies of hydroxyapatite implants which appear to achieve complete vascularization at a much more rapid rate (8,17,18).  Nonetheless, this same study (16) (as well as two other unpublished studies) has demonstrated that high-density porous polyethylene is associated with significantly less inflammation and capsule formation when compared with hydroxyapatite (Goldberg RA, unpublished data; Porex Technologies Corporation, unpublished data).

The patients followed in this clinical study of high-density porous polyethylene as well as those in other longer-term studies demonstrate that this material is well tolerated and is associated with a very minimal risk of implant exposure or extrusion (15).  This is due to the excellent fibrovascular ingrowth that high-density porous polyethylene induces.  However, only long-term studies and more extensive use of this material will determine if it can stand the test of time as an anophthalmic socket implant.  Another benefit associated with this material is that the extraocular muscles can be directly sutured to it without the necessity for an external covering of sclera or another donor material.  As the patients in this study demonstrated, there is good to excellent socket motility associated with this material after its implantation.  In addition, the cost associated with a high-density porous polyethylene implant is less than one half of that associated with hydroxyapatite.  In cases of evisceration, both hydroxyapatite and high-density porous polyethylene offer little advantage over methylmethacrylate spheres when there is no integration of the implant with the prosthesis.  Their main benefit in such cases may be a reduced incidence of extrusion due to tissue ingrowth.

Currently, the most significant drawback to high-density porous polyethylene is that, as yet, there has been no attempt to determine if it can be used as a true integrated implant.  Therefore, prosthesis motility with this implant is similar to other implants to which the extraocular muscles have been attached but which have not been directly coupled with the overlying prosthesis.  However, unlike hydroxyapatite, it may not be necessary to create a hole and peg relation to achieve this effect.  Although drilling high-density porous polyethylene may be successful, as with hydroxyapatite, this would be a second surgical procedure requiring either a bone scan or a magnetic resonance imaging scan.  Another more effective method for achieving integration would be to mold the external surface of the high-density porous polyethylene implant.  Unlike hydroxyapatite, which is a very rigid material, high-density porous polyethylene can be molded easily into other shapes, thereby improving its usefulness in eviscer ation. Clinical and experimental studies are now under way to evaluate a modified quasi-integrated implant design for achieving the motility results seen when hydroxyapatite is integrated with a prosthesis.

While hydroxyapatite remains the implant of choice for producing the most natural prosthesis motility, high-density porous polyethylene shows promise as another useful entry into the field of anophthalmic socket implants.  It is certainly more effective than the traditional methylmethacrylate sphere both in terms of motility and a reduced rate of extrusion.  In those cases where the characteristics of hydroxyapatite are not necessary for forming a true integrated implant, high-density porous polyethylene is a less costly and equally efficacious alternative.


1. Anderson RL, Thiese SM, Nerad JA, et al. The universal orbital implant: indications and methods. Adv Ophthal Plast Reconstr Surg 1990;8:88-99.

2. Danz W. Mobility implants: a review. Adv Ophthal Plast Reconstr Surg 1990;8:46-52.

3. Perry AC. Advances in enucleation. Ophthalmol Clin N Am 1991;4:173-82.

4. Karesh JW. Polytetrafluoroethylene as a graft material in ophthalmic plastic and reconstructive surgery. Ophthal Plast Reconstr Surg 1987;3:179-85.

5. Baumgarten D, Wojno T, Taylor A Jr. Evaluation of biomatrix hydroxyapatite ocular implants with technetium-99m-MDP. J Nucl Med 1993;34:467-8.

6. Hamilton HE, Christianson MD, Williams JP, Thomas RA. Evaluation of vascularization of coralline hydroxyapatite ocular implants by magnetic resonance imaging. Clin Imaging 1992; 16:243-6.

7. Perry AC. Integrated orbital implants. Adv Ophthal Plast Reconstr Surg 1990;8:75-81.

8. Dutton JJ. Coralline hydroxyapatite as an ocular implant. Ophthalmology 1991 ;98:370- 7.

9. Shanbhag A, Friedman HI, Augustine J, von Recum AF. Evaluation of porous polyethylene for external ear reconstruction. Ann Plast Surg 1990;24:32-9.

10. Klawitter JJ, Bagwell JG, Weinstein AM, Sauer BW. An evaluation of bone growth into porous high density polyethylene. J Biomed Mater Res 1976;10:311-23.

11. Shields CL, Shields JA, De Potter P. Hydroxyapatite orbital implant after enucleation. Experience with initial 100 consecutive cases. Arch OphthalmoI 1992;110:333-8.

12. Buettner H, Bartley GB. Tissue breakdown and exposure associated with orbital hydroxyapatite implants. Am J OphthalmoI 1992;113:669-73.

13. Goldberg RA, Holds JB, Ebrahimpour J. Exposed hydroxyapatite orbital implants. Report of six cases. Ophthalmology 1992;99:831-6.

14. Nunery WR, Heinz GW, Bonnin JM, et al. Exposure rate of hydroxyapatite spheres in the anophthalmic socket: histopathologic correlation and comparison with silicone sphere implants. Ophthal Plast Reconstr Surg 1993;9:96-104.

15. Rubin LR. Polyethylene as a bone and cartilage substitute: a 32-year retrospective. In: Rubin LR, ed. Biomaterials in Reconstructive Surgery. St. Louis: CV Mosby, 1983; chap 30.

16. Sauer BW, Weinstein MA, Klawitter JJ, et al. The role of porous polymeric materials in prosthesis attachment. J Biomed Mater Res 1974;8: 145-53.

17. Shields CL, Shields JA, Eagle RC Jr, De Potter P. Histopathologic evidence of fibrovascular ingrowth four weeks after placement of the hydroxyapatite orbital implant. Am J Ophthalmol 1991;111:363-6.

18. Kazim M, Katowitz JA, Fallon M, Piest KL. Evaluation of a collagen/hydroxylapatite implant for orbital reconstructive surgery. Ophthal Plast Reconstr Surg 1992;8:94-108.

Please Call Us Today to Schedule a Consultation