A challenging problem for any podiatric surgeon is surgically managing cases in which a silicone elastomer implant in the first metatarsophalangeal joint (MPJ) has failed. When patients initially present with this problem, they will complain of pain, deformity or both at either the first or lesser metatarsophalangeal joints.
The cause of the pain or deformity can be multifactorial. The possible causes may include: chronic synovitis and swelling around the implant; chronic skin fistulas; implant breakage or fragmentation; severe periarticular bony subsidence and erosion; or biomechanical disruption of first ray stability and resultant lesser metatarsophalangeal joint overload.
The use of silicone implants in the first MPJ was described initially in 1972 when the implant was a single-stemmed hemiprosthesis designed to replace the resected base of the proximal phalanx.1 Results of this hemi-implant proved unsatisfactory. Consequently, a flexible, double-stemmed, hinged prosthesis was developed.2 Advocates of these implants cited such benefits as restoration of hallux length, good joint mobility, restoration of nearly normal joint biomechanics, adequate correction of hallux valgus and a shorter recovery period.1-8
Other clinical studies were less supportive as researchers reported high complication rates and poor functional outcomes, particularly for the hemi-implants.9-12 In one study, 36 percent of patients who received a single-stem implant were dissatisfied with the results of the operation.13 Researchers have also reported distal amputation after implant insertion.14 More recent data suggest these implants have poor rates of survival in weightbearing joints and in younger patients and that over time, these devices have high rates of mechanical failure with resultant metatarsalgia.11-13,16-19
Despite the failure associated with use of first metatarsophalangeal joint silicone prostheses, there is a paucity of literature on how to salvage these failures. Salvage can be particularly difficult in younger, active patients. Options include implant removal and synovectomy; implant removal with reinsertion of an alternate implant; or bone block distraction arthrodesis.
For older, more sedentary people with low physical demands, implant removal with synovectomy is an option. However, both the surgeon and the patient must remember that this option does not provide a functional forefoot even for patients who are only minimally active. When considering this implant removal procedure, one must counsel patients on the strong possibility of lesser metatarsal overload symptoms and postoperative need for orthoses.
In a study of 11 failed silicone first MPJ implants, 10 patients received a mean 4.9 years of follow-up and seven patients had excellent clinical results.9 The authors observed a general clinical trend toward toe extension and used pressure studies to show increased load on the lateral forefoot.9 Other authors also observed postoperative lesser metatarsalgia in active patients who had received resectional arthroplasty.20,21
Another option for salvage is implant replacement. Koenig published a three-year follow-up study of 10 patients who received revisional arthroplasty after silicone implant removal and reinsertion of a dual-component great toe device.10 In that study, the author did not mention the ages of the study cohort and defined success as one or more years with no pain from the first MPJ.10 How these revised implants would perform over a longer period is difficult to determine. Gaining patient acceptance of this option may also prove difficult because patients with implant failure are usually extremely reluctant to have another prosthesis inserted. In addition, for any subsequent implant replacement, bony subsidence about the silicone implant would make it challenging to maintain positional stability of the new prosthesis.
The final and most biomechanically sound option for salvage is bone block distraction fusion of the first metatarsophalangeal joint. This procedure accomplishes subjective improvement of the patient’s level of pain and walking tolerance as well as objective restoration of first ray stability (thus alleviating symptoms of lesser metatarsophalangeal joint overload). The procedure also restores great toe alignment and length, and usually enables one to correct lesser digital malalignment after restoring the hallux position. To date, only one study (14 patients) has reported functional outcomes of distraction arthrodesis after a failed implant.22 The authors of that study stipulated this technically demanding procedure provides long-term stability to the hallux, restores weightbearing and allows patients to maintain a propulsive gait.22 
Distraction arthrodesis is preferable for medically fit, active individuals with a failed silicone implant. Although the age of the patient is a factor in considering surgical options, this procedure can be a reasonable choice for medically fit, active patients as old as 70.
Surgeons should obtain preoperative anteroposterior, lateral and medial oblique radiographs of both feet in weightbearing stance so they can compare the unaffected foot with the other foot in regard to digital malalignment, shortening and bony erosion about the implant. These preoperative radiographs also enable surgeons to assess implant fragmentation and the metatarsal parabola.
One should counsel patients regarding the need for autogenous bone graft and the risk of further morbidity from the bone graft donor site. In most cases, two donor sites are available for use, depending on the size of the graft you need. If a piece measuring 1.0 to 1.5 cm in length is required, then one can harvest a bicortical block of bone from the superior aspect of the calcaneus. If one needs a larger piece, obtaining a tricortical bone block from the anterior iliac crest is the best option. Preoperative radiographs will indicate the size of structural graft needed.
In my experience with these cases, I have found it a good idea always to anticipate the need for more bone. The bony interfaces with the implant are usually sclerotic and adequate resection to healthy bleeding bone can result in a large defect. In my experience, nearly all cases of failed first metatarsophalangeal joint implants have required iliac crest bone grafting. Mixing demineralized bone matrix and other osteogenic products with autogenous cancellous bone also can be useful to “fill” defects created by the implant. Rigid fixation is required until one achieves radiographic union. I usually employ internal fixation but one can use external fixators to supplement the construct.
In many of these cases, one will observe varus malposition of the hallux and second and third digits. These digits can usually be anatomically aligned after the hallux is anatomically positioned.
For the salvage procedure, one should ensure the patient is in a supine position and proceed to prepare and drape the foot and ipsilateral hip. If the heel is the site for bone harvesting, place a bump under the hip. If possible, make the incision over the first metatarsophalangeal joint over any existing scars. The dorsal linear incision usually measures 7 to 8 cm in length and one should make this incision just medial to the extensor tendon of the big toe. Carry the incision to bone and enter the joint. Remove the implant and debride the synovitis. Resect the bone interfaces with an oscillating saw and a high speed burr until one sees bleeding. Depending on the type of implant the patient has, erosion of the medullary canal of the first metatarsal head or proximal base may occur. This erosion can be extensive and require extensive burring until one achieves a healthy “bleeding” bone interface.
After removing the sclerotic margins, pack the defects with cancellous bone and products that promote osteogenesis. Proceed to fashion the structural graft to fit the defect, restoring a plantigrade hallux with good length.
When the graft is in place and is of appropriate length, proceed to fit the graft by using an obliquely placed 3.5-mm AO positional screw and a five- to seven-hole, one-third tubular dorsal neutralization plate. One can usually select a screw measuring 40 to 60 mm in length. Proceed to direct the screw from the medial aspect of the first metatarsal head through the graft to the lateral aspect of the base of the proximal phalanx. Position the plate slightly lateral to the midline of the metatarsal and hallux for coverage with the extensor tendon. One should engage at least four cortices proximal and distal to the graft piece. However, this goal is not always achievable given the extent of sclerosis in the proximal phalanx and the fact that bone length after resection can be extremely short. Proceed to close the extensor tendon over the plate and close the remaining layers.
Postoperatively, one should have the patient wear a modified Jones compression splint for 10 to 14 days and then remove the sutures. Apply a short-leg non-weightbearing cast for four more weeks. Take non-weightbearing radiographs of the foot at six weeks postoperatively.
Depending on the radiographic evidence of healing, one may advance patients to a removable fiberglass walking boot with partial weightbearing for two weeks and follow up with two weeks of full weightbearing in the walking boot. Evidence of radiographic consolidation determines the advancement to regular shoes.
Repeat follow-up visits and weightbearing radiographs of the foot at three and six months postoperatively. Given the higher risk of delayed union or nonunion, utilizing external bone stimulation is routinely recommended, particularly if the length of the structural graft piece exceeds 2 cm.
Although surgical correction of the failed silicone implant in the first metatarsophalangeal joint can be technically difficult, three surgical treatment options are available: implant removal, implant replacement with a new device and bone block arthrodesis. Distraction arthrodesis is the only option that affords first metatarsophalangeal joint stability and a biomechanically sound first ray, and is the best option for an active person. With careful preoperative and postoperative management, one can achieve optimal outcomes with this procedure.
Dr. Hamilton is affiliated with the Department of Orthopedics and Podiatric Surgery at the Kaiser Permanente Medical Centers in Oakland and Richmond, Calif. He is the Director of Education at the San Francisco Bay Area Foot and Ankle Residency Program, and is a Fellow of the American College of Foot and Ankle Surgeons. The author thanks the Kaiser Foundation Hospital’s Department of Medical Editing for its assistance with this article.
Editor’s Note: For related articles, check out the archives at www.podiatrytoday.com.
1. Swanson AB. Implant arthroplasty for the great toe. Clin Orthop 1972;85:75-81.
2. Swanson AB, Lumsden RM, Swanson GD. Silicone implant arthroplasty of the great toe. A review of single stem and flexible hinge implants. Clin Orthop 1979 Jul-Aug;(142):30-43.
3. Albin RK, Weil LS. Flexible implant arthroplasty of the great toe: An evaluation. J Am Podiatry Assoc 1974 Dec;64(12):967-975.
4. Cracchiolo A 3rd, Swanson A, Swanson GD. The arthritic great toe metatarsophalangeal joint: a review of flexible silicone implant arthroplasty from two medical centers. Clin Orthop 1981 Jun;(157):64-9.
5. Gould N. Hallux rigidus: cheilotomy or implant? Foot Ankle 1981 May;1(6):315-20.
6. Kampner SL. Total joint prosthetic arthroplasty of the great toe--a 12-year experience. Foot Ankle 1984 Mar-Apr;4(5):249-61.
7. LaPorta GA, Pilla P Jr, Richter KP. Keller implant procedure: a report of 536 procedures using a Silastic intramedullary stemmed implant. J Am Podiatry Assoc 1976 Mar;66(3):126-47.
8. Wenger RJ, Whalley RC. Total replacement of the first metatarsophalangeal joint. J Bone Joint Surg Br 1978 Feb;60(1):88-92.
9. Kitaoka HB, Holiday AD Jr, Chao EY, Cahalan TD. Salvage of the failed first metatarsophalangeal joint implant arthroplasty by implant removal and synovectomy: clinical and biomechanical evaluation. Foot Ankle 1992 Jun;13(5);243-50.
10. Koenig RD. Revision arthroplasty utilizing the Biomet Total Toe System for failed silicone elastomer implants. J Foot Ankle Surg 1994 May-Jun;33(3);222-7.
11. Granberry WM, Noble PC, Bishop JO, Tullos HS. Use of a hinged silicone prosthesis for replacement arthroplasty of the first metatarsophalangeal joint. J Bone Joint Surg Am 1991 Dec;73(10):1453-9.
12. Ris HB, Mettler M, Engeloch F. [Long-term results using a Swanson silastic endoprosthesis of the metatarsophalangeal joint of the great toe. Discrepancy between clinical and radiologic findings] [Article in German]. Z Orthop Ihre Grenzgeb 1988 Sep-Oct;126(5):526-9.
13. Shankar NS. Silastic single-stem implants in the treatment of hallux rigidus. Foot Ankle Int 1995 Aug;16(8):487-91.
14. Neumann R, Reisch P. [Silastic arthroplasty of the great toe metatarsal joint] [Article in German]. Orthopade 1996 Aug;25(4);332-7.
15. Kitaoka HB, Cracchiolo A 3rd. Stress fracture of the lateral metatarsals following double-stem silicone implant arthroplasty of the hallux metatarsophalangeal joint. Clin Orthop 1989 Feb;(239):211-6.
16. Lemon RA, Engber WD, McBeath AA. A complication of Silastic hemiarthroplasty in bunion surgery. Foot Ankle 1984 Mar-Apr;4(5):262-6.
17. Shereff MJ, Jahss MH. Complications of silastic implant arthroplasty in the hallux. Foot Ankle 1980 Sep;1(2):95-101.
18. Verhaar J, Bulstra S, Walenkamp G. Silicone arthroplasty for hallux rigidus. Implant wear and osteolysis. Acta Orthop Scand 1989 Feb;60(1):30-3.
19. Papagelopoulos PJ, Kitaoka HB, Ilstrup DM. Survivorship analysis of implant arthroplasty for the first metatarsophalangeal joint. Clin Orthop 1994 May;(302):164-72.
20. Bonney G, MacNab L. Hallux valgus and hallux rigidus: a critical survey of operative results. J Bone Joint Surg Br 1952 Aug;34-B(3):366-85.
21. Cleveland M, Winant EM. An end-result study of the Keller operation. J Bone Joint Surg Am 1950 Jan;32A(1);163-175.
22. Hecht PJ, Gibbons MJ, Wapner KL, Cooke C, Hoisington SA. Arthrodesis of the first metatarsophalangeal joint to salvage failed silicone implant arthroplasty. Foot Ankle Int 1997 Jul;18(7):383-90.