Metatarsal Phalangeal Joint Replacement

Number: 0708

Policy

1. total prosthetic replacement arthroplasty with silastic implants; and
2. hemiarthroplasty.

Aetna considers metatarsal phalangeal joint replacement for other indications, and for joints other than the first metatarsal phalangeal joint (e.g., tarsal metatarsal joint) experimental and investigational because its value is unproven.

Aetna considers ceramic prostheses (e.g., the Moje implant) experimental and investigational for replacement of the first metatarsal phalangeal joint and for other indications because their long-term effectiveness has not been established.

Aetna considers modular implants (e.g., the Cartiva Synthetic Cartilage Implant, the METIS prosthesis, the OsteoMed ReFlexion 1st MTP Implant System, and the ToeFit-Plus prosthesis) experimental and investigational for replacement of the first metatarsal phalangeal joint and for other indications because their long-term effectiveness has not been established.

Aetna considers interpositional arthroplasty with biologic spacers (e.g., the InterPhlex interdigital implant) and total prosthetic replacement arthroplasty using total metallic implants experimental and investigational for hallux rigidus, degenerative arthritis, and other indications involving the metatarsal phalangeal joints because their effectiveness has not been established.

See also CPB 0629 - Bunionectomy.

Background

Most clinical presentations of the hallux (big toe) concern the metatarsal phalangeal joint (MPJ).  The underlying causes of disease/disorder of the MPJ include osteoarthritis, rheumatoid arthritis, disease of the hallucal sesamoids and post-traumatic degeneration.  Both types of arthritis often affect the first MPJ located at the base of the big toe.  The MPJ may become stiff (hallux rigidus), or deformed (hallux valgus).  Hallux rigidus is characterized by pain as well as a reduction in the range of motion (ROM), especially dorsiflexion, at the first MPJ.  Hallux valgus is classified as an abnormal deviation of the great toe towards the midline of the foot.  Disease/disorder of the MPJ affects shoe wear, ambulation, and other activities of daily living.  Although the literature addressing treatments of conditions that affect the hallux often focuses on surgical interventions, the use of conservative therapies is emphasized before surgery is considered.  Conservative treatments include exercise, physiotherapy, supportive shoes worn alone or worn with soft/semi-rigid orthoses, non-steroidal anti-inflammatory drugs, and steroid injections.  Many surgical procedures have been described for the treatment of congenital and acquired conditions of the big toe.  They include arthrodesis (fusion of the joint), arthroplasty, cheilectomy (trimming of the joint), Keller procedure (simple excision of the joint), osteotomy, and plantar release.  Metatarsal phalangeal arthrodesis remains the gold standard for arthritis and salvage of the painful first MPJ (Weinfeld and Schon, 1998; Giannnini et al, 2004; Sammarco and Nichols, 2005; Kelikian, 2005). 

Brage and Ball (2002) stated that when approaching patients with a painful first MPJ that has failed conservative therapies and first-line surgical treatments (cheilectomy or minor bunion procedures), the surgeon should stratify these patients based upon diagnosis, age, and activity level.  For the young, active patient, an arthrodesis is the gold standard, and the primary predictors of clinical and radiographical success are proper fusion angle alignment and maintenance or restoration of length.  In the elderly, inactive patient, arthrodesis is a safe and reliable treatment option.  However, the Keller procedure may be preferable because it provides excellent early symptomatic relief and has a less debilitating post-operative rehabilitation program.  The patients between these two extremes fall into a treatment gray zone.  The arthrodesis should again be considered the gold standard because it is reliable and durable with time and activity.  However, the authors noted that biologic or prosthetic inter-positional arthroplasty are exciting investigational treatment options for these patients.  If a prosthetic implant is to be used, the double-stemmed, hinged silastic implant with protective titanium grommets, or a metallic hemi-arthroplasty prosthesis, appear to be the 2 best choices of implant.  With the continuous advances in material engineering and tissue engineering, prosthetic and biologic inter-positional arthroplasties hold the greatest promise for the painful first MPJ in the future.  These treatment modalities allow restoration of alignment and maintenance of motion, length, and strength, which are fundamental in attaining a good clinical result.  The authors stated that when the optimal material is developed (whether it is prosthetic, biologic, or a combination of both), these treatment advantages will be realized without the attendant complications associated with the use of the current implants.  The observations of Brage and Ball (2002) were in agreement with those of Sizensky (2004).

Coughlin and Schurnas (2004) reported their experience with cheilectomy or MPJ arthrodesis in the treatment of hallux rigidus.  Of the original 114 patients with a diagnosis of hallux rigidus, 110 returned for the final evaluation.  Eighty patients (93 feet) had undergone a cheilectomy, and 30 patients (34 feet) had had an arthrodesis.  The mean durations of follow-up were 9.6 and 6.7 years following cheilectomies and arthrodeses, respectively.  These authors reported that 97 % of patients (107/110) had a good or excellent subjective result, and 92 % of cheilectomy (86/93) were successful in terms of pain relief and function.  Cheilectomy was used with predictable success to treat grade-1, grade- 2 and selected grade-3 cases.  Patients with grade-4 hallux rigidus or grade-3 hallux rigidus with less than 50 % of the metatarsal head cartilage remaining at the time of surgery should be treated with arthrodesis. 

A systematic evidence review of treatments for hallux rigidis by Yee and Lau (2008) found that the consistently favorable results reported in several level IV studies constituted fair evidence (grade B recommendation) to support the use of cheilectomy in persons with grade I and II hallux rigidis.  The authors noted that 2 separate studies observed poor results in a small subset of patients with advanced degeneration of the MPJ.  The authors concluded that, based on this evidence, cheilectomy can not be recommended for grade III hallux rigidus.

Taylor et al (2004) stated that arthrodesis has emerged as the primary salvage procedure for severe osteoarthritis of the first MPJ.  These investigators reported that 43 patients underwent arthrodesis of the first MPJ with stabilization provided by either 2 crossed lag-screws or a dorsal plate and screws.  First MPJ arthrodesis was the primary procedure for 46 of the 54 treated feet.  At a mean of 21.7 months (median of 13.5 months), 34 of the 43 patients completed a brief telephone survey about surgical outcomes.  Radiographical measurements of inter-metatarsal, hallux valgus, inclination, and dorsiflexion angles were made pre-operatively and post-operatively.  Mean time to fusion was 7.3 weeks; arthrodesis was successful for 50 of 52 feet (radiographs were missing for 2 of the 54 feet treated).  Internal fixation devices were removed from 5 feet.  Thirty of the 34 patients (88.2 %) rated their result as excellent or good; the other 4 (11.8 %) rated their result as poor.

A systematic evidence review by Yee and Lau (2008) found that the consistently favorable results in many level II and IV studies constitute fair evidence (grade B recommendation) to support the use of arthrodesis for the treatment of stage III hallux rigidus.

Although joint replacement remains the ultimate solution for hip osteoarthritis, and may be a viable option in ankle osteoarthritis, replacement of the MPJ has not been established as a standard of care for osteoarthritis of the hallux.  Despite its initial success in relieving symptoms, the use of total joint replacements of the first MPJ with a flexible hinged silicone prosthesis for replacement arthroplasty was initially abandoned because of the high and increasing rate of failure of the implant, as demonstrated radiographically (Granberry et al, 1991).  These first generation silastic implants failed because of the high shear forces concentrated at the prosthetic hinge.

To address this, new systems were redesigned for insertion with titanium grommets to reduce the stress applied to the silastic in order to increase the survival of the arthroplasty (Yee and Lau, 2008).  Sebold et al (1996) investigated the use of double-stem silicone implants protected by titanium grommets.  These were placed in the hallux metatarso-phalangeal joints of 32 patients (47 feet).  All patients had a painful destroyed joint and most were women.  Three patients (6 feet) were lost to follow-up.  Nineteen patients had a diagnosis of rheumatoid arthritis (25 feet) and 10 had degenerative joint disease (16 feet).  The average age for the group was 57 years and the average follow-up was 51 months (range of 34 to 76 months).  Twenty patients (30 feet) were completely satisfied with their result.  Eight patients (10 feet), all with rheumatoid arthritis, had some minor post-operative complaints, usually involving the lateral toes.  Two patients (3 feet) in this group had no pain, but would have preferred more hallux motion.  One patient with rheumatoid arthritis (1 foot) had a poor result due to implant removal for deep sepsis.  Radiographical analysis of these patients showed no evidence of implant fracture and the implant composite appeared to be well-tolerated by the surrounding bone in which it was placed.  The investigators reported that, when compared with another, similar group of patients in whom grommets were not used, this implant appeared to be much more stable, as there was significantly less evidence of radiolucency seen around those implants protected by the grommets.  The investigators stated that the titanium grommets may protect the silicone implant and may help provide a longer life for the silicone implant.

In a systematic evidence review of treatments for hallux rigidus, Yee and Lau (2008) stated that, despite these improvements, concerns persist regarding the potential effects of silicone debris leading to foreign-body reaction, synovitis, and bone erosion in the hallux.  In addition, the systemic effects of silicone microfragments invading the lymphoreticular system are still unknown.  The authors concluded that conflicting evidence weakly supports total prosthetic replacement arthroplasty with silastic implants with hallux rigidus (grade C recommendation).

Other MPJ prostheses include the titanium hemi-great toe implant (Leavitt et al, 1991), the 2-component first MPJ implant (Gerbert et al, 1995), and the Moje press-fit ceramic implant (Malviya et al, 2004).  Malviya et al reviewed their results with the Moje press-fit ceramic implant.  This study included 7 procedures in 6 patients with a mean age of 60.2 years followed for a mean of 35 months (a range of 24 to 43 months).  There was a significant (p < 0.001) improvement of the visual analog score from 7-8 to 1-2 and of the Foot Function Index from 75.6 to 8.6.  A mean post-operative dorsiflexion of 29.2 degrees and plantar flexion of 12.1 degrees were recoded.  Apart from slight cortical recession in 1 case, probably related to overuse, there was little evidence of osteolysis or loosening of implants and no major complication has been noted in any of the patients.  The authors stated that the press-fit design appears to have overcome the disadvantages of the previous screw-fit prosthesis that had been reported to have complications related to metallosis around the titanium screw.

The Clinical Practice Guideline First Metatarsophalangeal Joint Disorders Panel (Vanore et al, 2003) noted that total joint replacement systems have been designed for the first MPJ generally as 2-component non-constrained articulations in an attempt to allow motion in more than 1 plane.  Materials used for opposing articular surfaces are chosen for their low coefficient of friction and for their minimum wear characteristics.  Numerous implant systems have been developed during the years, and several are still used clinically, although long-term clinical usefulness has yet to be established.  The panel stated that judicious use and strict criteria are recommended to avoid complications and problematic revisions.

The National Institute for Health and Clinical Excellence (NICE, 2005a) released an assessment on MPJ replacement of the hallux.  It concluded that available evidence on the safety and effectiveness of MPJ replacement of the hallux appears adequate to support the use of this procedure, but there is limited evidence of the durability of this procedure.  The NICE assessment stated that clinicians should ensure that patients fully understand the uncertainties about the place of this procedure in relation to alternative treatment options such as arthrodesis (NICE, 2005a).  The assessment stated that patient selection is important, and should take into consideration the likely intensity and duration of use of the joint based on the patient's activities and aspirations.  This report also stated that further research will be useful in establishing the long-term outcomes of different types of prostheses.  These conclusions were based on a review of the available evidence (NICE, 2005b).  The main outcome measures reported were pain relief and patient satisfaction.  Three studies reported that 73 % (8/11), 79 % (46/58) and 100 % (7/7) of joints with implants were pain-free after mean follow-ups of 17 months, 12 years, and 35 months, respectively.  Another study including 86 implants reported a statistically significant improvement in pain scores after the procedure.  Two further studies reported pain relief in 66 % (59/90) of implants and 94 % (30/32) of patients (mean follow-ups of 3 years and 8 years, respectively).  Four studies reported that between 74 % (29/39) and 88 % (7/8) of patients were completely satisfied with the procedure (mean follow-ups of 12 months and 17 months, respectively).  Although most of the specialist advisors to NICE stated that this was an established technique, these advisors noted that there is limited evidence on the durability of the newer implants.  The specialist advisors also stated that potential adverse events included persistent pain, infection, implant loosening, implant fracture, osteolysis, bone over-production, cyst formation, silastic granulomas, and transfer metatarsalgia.  Some of these complications may require removal of the joint.  Radiological follow-up may show fracture of prostheses or immobility of joints in the long-term.  However, the influence of these changes on symptom relief remains unclear.

On the other hand, some recent reviews did not find MPJ replacement to be a standard treatment for diseases/disorders of the hallux (Fuhrmann et al, 2003; Giannnini et al, 2004; Ferrari et al, 2004; Wulker, 2004; Keiserman et al, 2005; Esway and Conti, 2005).  Fuhrmann et al (2003) reported their experience with replacement of the first MPJ.  After a 3-year follow-up, most patients who had an MPJ replacement were extremely satisfied with the outcome.  Plantar pressure distribution revealed a marked improvement.  However, recovery of metatarsophalangeal (MTP) dorsiflexion was limited and joint stability worsened.  Radiologically, 1/3 of the prostheses showed radiolucent lines indicating loosening of the implant.  These researchers stated that MPJ replacement offers distinct advantages in the treatment of end-stage hallux rigidus, but more research is needed on implant design and osseous fixation. 

A review on surgical interventions for hallux rigidus (Giannnini et al, 2004) did not list MPJ replacement as one of the options, which include arthrodesis, cheilectomy, Keller procedure, osteotomy, plantar release, and arthroplasty with the use of a spacer.  Additionally, a Cochrane review on interventions for treating hallux valgus (Ferrari et al, 2004) did not mention the use of MPJ replacement.  Furthermore, Wulker (2004) stated that conservative treatments for hallux rigidus mainly consist of local anti-inflammatory applications and orthopedic appliances to decrease load at the MPJ.  With progression of the arthrosis, joint-preserving procedures such as cheilectomy are used.  In complete destruction of the joint space, arthrodesis is the technique of choice in the mostly active, younger patients.  The author stated that resection arthroplasty is mainly used in the elderly, less active patients, and the results of first MPJ replacement are inferior to arthrodesis.

Keiserman et al (2005) noted that many surgical procedures are available for the treatment of hallux rigidus.  The choice depends on the severity of the disease, activity level of patient, and expectations about the surgery.  These investigators said that cheilectomy is recommended for early disease and may be associated with an osteotomy of the proximal phalanx.  For active patients who have severe hallux rigidus, arthrodesis and biological inter-position arthroplasties have shown good results.  These investigators stated that Keller arthroplasty is reserved for patients with low functional demand; and prosthetic replacements are not recommended at this time.  In addition, Esway and Conti (2005) stated that replacement of the hallux MPJ does not have the same success as hip and knee arthroplasties.  Silastic joint implants have a high patient satisfactory rate; however they have caused many complications, including silicone synovitis and lymph node inflammation.  Metal and polyethylene hemi-arthroplasties and total toe replacements appear to be more promising although results are preliminary.  Problems with these implants seem to be related to soft tissue instability of the joint; patients who have hallux rigidus have more success than patients who have hallux valgus or rheumatoid arthritis.  Severe complications can be treated with removal and synovectomy or arthrodesis, depending on the length and alignment of the foot, as well as the functional demands of the patients.  These researchers further stated that it would be beneficial to have more data on these implants so that improvements can be made in design and patient selection.

Silastic hemiarthroplasty was initially abandoned as a treatment for hallux rigidus because of their failure in terms of poor durability, foreign body reaction and dislodgement of components (Shankar, 1995; Rahman and Fagg, 1993).  The metallic hemiarthroplasty was developed in response to these failures of silastic implants (Townley and Taranow, 1994).  The procedure involves the resection of the proximal portion of the proximal phalanx in addition to the resection remodeling of the metatarsal head (Townley and Taranow, 1994).  The metallic prosthesis is then implanted to replace the articulating surface of the proximal phalanx.  Current metallic hemiarthroplasty has not been proven to offer significant benefits compared with other surgical alternatives.

Available published peer-reviewed evidence for metallic hemiarthroplasty consists primarily of retrospective case series (Townley and Taranow, 1994; Taranow et al, 2005).  In the only prospective study of metallic hemiarthroplasty for hallux rigidus published to date, Roukis and Townley (2003) reported similar short-term outcomes with metallic hemiarthroplasty and periarticular osteotomy.  The investigators compared metallic hemiarthroplasty to periarticular osteotomy in 44 patients (47 feet) with hallux rigidus.  A subjective evaluation, physical examination, and radiographical analysis were performed pre-operatively and at a 1-year follow-up.  Twenty patients (20 feet) underwent a peri-articular osteotomy, with 16 patients (16 feet) returning.  Seven patients (9 feet) underwent metallic hemiarthroplasty, with all patients returning.  The subjective evaluation was based on a modified American Orthopaedic Foot and Ankle Society Hallux Metatarsophalangeal-Interphalangeal 100-point scale.  The physical examination included first MPJ ROM.  Radiographical analysis included the metatarsal protrusion distance, transverse plane angulation of the second digit, lateral talo-first metatarsal angle, sagittal plane relationship of the first and second metatarsals, and hallux equinus angle.  Statistically significant differences between pre-operative and post-operative values were found for the periarticular osteotomy group for the metatarsal protrusion distance (p = 0.000), transverse plane angulation of the second digit (p = 0.000), and lateral talo-first metatarsal angle (p = 0.015).  No other statistically significant differences between the pre-operative and post-operative values for either procedure group were found to exist.  The investigators noted that there were equally significant improvements in subjective scores and a high percentage of patient satisfaction in both groups.  However, both procedures resulted in only in minimal increases in first MPJ ROM.

Raikin et al (2007) reported better long-term outcomes with arthrodesis than with metallic hemiarthroplasty in hallux rigidus.  Patients with hallux rigidus were treated with either a metallic hemiarthroplasty or an arthrodesis between 1999 and 2005.  Post-operative satisfaction and function were graded with use of the American Orthopaedic Foot and Ankle Society Hallux Metatarsophalangeal Interphalangeal (AOFAS-HMI) scoring system, and pain was scored with use of a visual analog scale.  The investigators reported that 21 hemiarthroplasties and 27 arthrodeses were performed in 46 patients.  Five (24 %) of the hemiarthroplasties failed; 1 of them was revised, and 4 were converted to an arthrodesis.  Eight of the feet in which the hemiprosthesis had survived had evidence of plantar cut-out of the prosthetic stem on the final follow-up radiographs.  At the time of final follow-up (at a mean of 79.4 months), the satisfaction ratings in the hemiarthroplasty group were good or excellent for 12 feet, fair for 2, and poor or a failure for 7.  The mean pain score was 2.4 of 10.  All 27 of the arthrodeses achieved fusion, and no revisions were required.  At the time of final follow-up (at a mean of 30 months), the satisfaction ratings in this group were good or excellent for 22 feet, fair for 4, and poor for 1.  The mean pain score was 0.7 of 10.  Two patients required hardware removal, which was performed as an office procedure with the use of local anesthesia.  The AOFAS-HMI and visual analog pain scores and satisfaction were significantly better in the arthrodesis group.  The investigators reported that arthrodesis is more predictable than a metallic hemiarthroplasty for alleviating symptoms and restoring function in patients with severe hallux rigidus.

A systematic evidence review by Yee and Lau (2008) found that, except for a study co-authored by the developer of the hemiarthroplasty (Townley and Taranow, 1994), the use of hemiarthroplasty in the management of hallux rigidus is supported by conflicting or poor quality evidence (grade C recommendation).  The long-term consequences for hemiarthroplasties that have not failed but are malpositioned, subsided or surrounded by radiolucencies remains uncertain.  The authors stated that further studies designed to yield level I or II evidence are warranted to address these concerns.

In a randomized controlled clinical study, results of arthrodesis were also found to be superior than total joint replacement with metallic implants.  Gibson et al (2005) reported on the results of a randomized controlled trial to evaluate clinical outcomes after MPJ arthrodesis and replacement arthroplasty in end-stage hallux rigidus.  Between November, 1998, and January 2001, 63 patients between the ages of 34 and 77 years, with unilateral or bilateral MPJ arthritis were recruited and randomly selected to have either MPJ arthrodesis or arthroplasty.  Twenty-two patients (38 toes) had arthrodesis and 27 patients (39 toes) had arthroplasty.  A single surgeon performed all surgery.  The primary outcome measure determining successful surgery was a decrease in pain as measured on a visual analog scale (VAS).  Functional outcome was assessed at 6 months and 1 and 2 years.  The investigators reported that, at 24 months, pain improved in both groups (p < 0.001), but there were significantly greater improvements after arthrodesis (p = 0.01).  All 38 arthrodeses united at a mean dorsiflexion angle of 26 degrees, with few complications.  In contrast, in the arthroplasty group, 6 of the 39 inserted implants had to be removed because of phalangeal component loosening.  In the remainder the ROM gained was poor, and the patients tended to bear weight on the outer border of their foot.  The investigators concluded that outcomes after arthrodesis were better than those after arthroplasty.  The investigators reported that the results were partially attributable to an unacceptably high incidence of loosening of the phalangeal components, which resulted in removal of the implants.  The investigators noted, however, even when data from the failures were excluded, arthrodesis was clearly preferred by most patients.

A systematic review of the evidence for treatment of hallux rigidus by Yee and Lau (2008) found that given the unfavorable results in multiple studies with different implants, total prosthetic replacement arthroplasty with metallic implants "cannot be recommended at this time for the management of hallux rigidus."  The authors stated that the results of the prospective, randomized trial by Gibson et al (2005) constitute a grade B recommendation of arthrodesis instead of arthroplasty.  The authors stated, however, that more level I or II evidence is warranted to confirm these findings.

Deheer (2006) argued against first MPJ implant arthroplasty.  The author noted that intermediate- and long-term studies raise concerns about implant failure and longevity.  Other causes for concern are silicone-induced synovitis and lymphadenopathy.  Furthermore, the lack of any significant long-term results and the documented metallic breakdown from 2-piece metallic implants make their use in hallux rigidus questionable.  Meanwhile, a comparative study showed the superiority of arthrodesis to implant arthroplasty.  Also, alternatives to joint-destructive procedures are emerging.  These include arthrodiastasis and the osteochondral autograft transfer procedure.  The author concluded that this evidence proves that implant arthroplasty is not the best treatment for patients with hallux rigidus or other first MPJ pathology.

Interpositional arthroplasty combines a standard resection arthroplasty with the insertion of a biologic spacer into the joint to avoid some of the difficulties associated with an isolated resection arthroplasty.  Various tissues, including tendons, have been utilized in the interpositional graft.  Theoretically, this procedure necessitates less bone resection from the proximal phalanx and better maintains joint stability and motion (Yee and Lau, 2008).

Hamilton et al (1997) reported on their experience with 30 patients (37 feet) with severe hallux rigidus who underwent interpositional arthroplasty over a 10-year period.  The authors reported that pain and function were significantly improved.  The American Orthopaedic Foot and Ankle Society (AOFAS) scores improved from an average of 23 pre-operatively to 37 post-operatively.  Average dorsiflexion improved from 10 to 50 degrees.  Transfer metatarsalgia was not seen.  All patients had at least 4/5 plantarflexion strength and averaged 50 degrees of dorsiflexion.  The authors concluded that, in patients with severe hallux rigidus and nearly equal length of first and second metatarsals, capsular interposition arthroplasty offers a surgical option that relieves pain without sacrificing motion or strength.

Kennedy et al (2006) examined 18 patients with severe articular cartilage loss who received 21 interposition arthroplasties.  The patients a mean age was 56 years.  They had a mean follow-up of 38 months.  All patients had substantial loss of articular cartilage when examined intra-operatively.  Patients were evaluated using the AFOAS and Short Form-36 scores.  All 18 patients had pain relief, and 17 of 18 patients said they would have the procedure again.  The mean post-operative increase in ROM of the first MPJ was 37 degrees.  The mean AFOAS and Short Form-36 scores were 78.4 and 96.3, respectively.  The complication rate was 6 %.

Lau and Daniels (2001) conducted a retrospective review of 19 patients (24 feet) with grade 2 osteoarthritis and 11 patients (11 feet) with grade 3 osteoarthritis.  The patients with grade 2 osteoarthritis were managed with a cheilectomy and the patients with grade 3 osteoarthritis with an interpositional arthroplasty.  All patients were individually assessed with a subjective questionnaire, physical examination, AOFAS hallux scale, SF-36 and pedobarographic analysis.  Cheilectomy patients (51.9 years) were younger than interpositional arthroplasty (59 years).  Follow-up between the interpositional arthroplasties (2.0 years) and cheilectomies (2.1 years) were comparable.  Post-operative motion, VAS and SF-36 scores were comparable between groups.  Cheilectomies had a higher mean AOFAS score (77.3) than interpositional arthroplasties (71.6).  Weakness of the great toe was reported in 72.7 % of interpositional arthroplasty patients compared to only 16.7 % of patients with a cheilectomy.  Patient satisfaction was 87.5 % in cheilectomies and 72.7 % in interpositional arthroplasties.  Pedobarographic analysis demonstrated a decreased load under the great toe with increased weight transfer to the lesser metatarsal heads in all patients.  The weight transfer to the lesser metatarsal heads was greatest in patients with interpositional arthroplasty.  The authors concluded that management of moderate hallux rigidus with a cheilectomy and phalangeal osteotomy is a reliable method of relieving pain and improving function.  Management of severe osteoarthritis of the joint with an interpositional arthroplasty should be considered a salvage procedure with less reliable results.

Other studies published of interpositional arthroplasty using a biologic spacer have also been level IV evidence, including Barca (1997) (12 patients followed for an average of 21 months) as well as Coughlin and Shurnas (2003) (7 patients followed for an average of 41 months).  In a systematic evidence review of treatments for hallux rigidus, Yee and Lau (2008) concluded that considering the limited quantity and quality of the data, there is insufficient evidence (grade I recommendation) to recommend interpositional arthroplasty for the treatment of hallux rigidus.

The Moje ceramic toe implant is made of zirconium oxide and was developed in 1994 by Dieter Werner (an orthopedic surgeon) and Hans Jurgen Moje (a ceramic engineer).  The original implant was screw-fit but complications of osteolysis and metallosis led to the replacement of the design with the press-fit one.  The press-fit implant is a 2-component ceramic prosthesis coated with apatite and fosterite crystals (Bioverit).  It relies mainly on interference fit coupled with osseo-integration encouraged by the Bioverit coating.  The coating forms a closed contact with the substrate and possesses a good adhesive strength (Malviya et al, 2004).

In a single-surgeon series study, Barwick and Talkhani (2008) evaluated the clinical outcome of the Moje arthroplasty using objective and subjective assessment tools.  A retrospective outcome study of 24 implants was performed in 22 patients undergoing first MPJ replacement for osteoarthritis from 2004 to 2006.  Each patient underwent clinical assessment using the AOFAS for the hallux and a patient outcome satisfaction questionnaire.  All pre- and post-operative radiographs were reviewed.  Average follow-up was 26 months with a median AOFAS score of 80 out of a maximum 100.  The revision rate at 3 years was 12.5 %.  Only 63 % of patients were "very satisfied" with the overall outcome from the procedure.  AOFAS for the hallux correlated strongly with patient satisfaction.  Radiographical mal-alignment in 4 patients was significantly associated with lower AOFAS (p = 0.01).  The authors concluded that the Moje ceramic prosthesis offers less reliable outcomes than the "gold standard" arthrodesis and caution is advised regarding its use for osteoarthritis of the first MPJ.

McGraw et al (2010) assessed the mid-term clinical and radiographical results of the Moje hallux MPJ replacement.  These investigators described their single-surgeon experience of 63 components in 48 patients at a mean follow-up of 44 months.  Patient satisfaction was assessed by questionnaire and radiographical assessment performed immediately post-operatively and at the latest follow-up.  Mean AOFAS hallux score increased from 56 to 72 (p < 0.01) and mean satisfaction score was 7.6 (scale 1 to 10).  A total of 67 % of subjects reported minimal or no pain.  Five implants have been removed (8 %), 4 because of pain associated with implant loosening and subsidence, and 1 because of deep infection.  Fifty-seven percent of metatarsal and 56 % of phalangeal components had subsided and radiographical evidence of loosening in 58 % of X-rays analyzed at latest follow-up was found.  Prosthetic subsidence was associated with greater margin of uncovered bone under the prosthesis (p = 0.05 for metatarsal, p = 0.03 for proximal phalanx component) and longer follow-up (p < 0.001).  The authors concluded that in spite of the good clinical outcome at the mid-term stage with 91 % implant survival, given the widespread loosening and subsidence encountered in this study, the long-term outcome following this procedure is uncertain.

In a case series study, Brewster and colleagues (2010) reported the functional results of the Moje first MPJ replacements performed between February 2001 and November 2006.  All patients who underwent Moje arthroplasty under the care of a single surgeon were included; outcome scores and complications were recorded annually.  A total of 32 joints in 29 consecutive patients were followed for a mean duration of 34 (range of 6 to 74) months, and the mean patient age at the time of operation was 56 (range of 38 to 79) years.  Hallux rigidus was the primary diagnosis in 28 (87.5 %) of the cases.  The mean AOFAS-HMI score at final follow-up was 74/100 (range 9 to 100), with 13 (40.63 %) joints rated good-to-excellent.  Two (6.25 %) joints were revised to arthrodesis at a mean of 52 (range of 41 to 63) months following the arthroplasty procedure, and the overall prevalence of post-operative complications was 6 (18.75 %).  Based on these results, the authors concluded that first MPJ joint replacement with the Moje device remains promising, but still has room for improvement before the results match those obtained with larger joint (knee, hip) arthroplasty.  Thus, more studies including larger number of patients with longer follow-up are needed to evaluate the long-term results of the Moje ceramic prosthesis for MPJ replacements.  Furthermore, Gutteck and colleagues (2011) stated that the high loosening rate of the Moje prosthesis in the treatment of hallux rigidus caused disappointing medium-term results.  Arthrodesis using an iliac crest bone graft is the standard salvage procedure.

Metatarsophalangeal implants have been proposed as treatment for disorders effecting joints other than the first MTP joint, for other toe joints (e.g., interphalangeal joints), and for the tarsal metatarsal (TMT) joint.  However, there is insufficient evidence regarding the use of MTP implants for these indications.

Nagy et al (2014) noted that ceramic first MPJ replacement has been reported for treatment of hallux rigidus (HR), but there are no published mid- or long-term studies available.  These investigators presented their mid-term results using a 2nd-generation ceramic first MPJ implant.  A retrospective review of clinical data and radiographs was performed for 31 feet (24 women; mean age at surgery was 55 ± 6 years) who had first MPJ replacement with a 2nd-generation ceramic prosthesis (primary, 29 feet; revision, 2 feet).  Mean follow-up was 81 ± 27 months after surgery.  Mean first MP passive ROM was 32 ± 17 degrees (dorsiflexion and plantarflexion).  Mean AOFAS score was 72 ± 19 points and Foot Function Index was 27 ± 26 points (all 31 feet).  Clinical rating for 29 feet that had surgery as a primary procedure was excellent in 5 feet (17 %), good in 8 feet (28 %), fair in 3 feet (10 %), and poor in 13 feet (45 %).  Patients were satisfied with the outcome in 24 feet (77 %).  Follow-up radiographs showed that radiolucency, change in angulation, sinkage, and mal-alignment of the metatarsal or proximal phalanx components were common.  Complications included 1 superficial wound infection, and revision was performed in 5 feet (16 %) because of loosening, sinkage, subluxation, pain, or fractured prosthesis.  Implant survival was 92 % at 5 years, 85 % at 7 years, and 68 % at 9 years.  The authors concluded that these findings of 2nd-generation ceramic first MPJ replacement in this series demonstrated poor clinical and radiological results with a high revision rate.

Stone and colleagues (2017) stated that the optimal operative management of HR is still a matter for debate among surgeons.  Despite arthrodesis widely considered to be the gold standard treatment, many surgeons advocate arthroplasty as a suitable alternative.  There are, however, few long-term or high-quality studies evaluating these modalities.  These researchers presented the 15-year follow-up of a randomized controlled trial (RCT).  These data were the follow-up to the original study published in 2005.  In the original study, 63 patients (77 toes) were recruited to and randomized to have either metatarsophalangeal joint (MTPJ) arthrodesis or arthroplasty.  The primary outcome measure was a decrease in pain on a (VAS at 24 months.  In the present study, data were available for all surviving patients (52 patients, 66 toes).  Data were collected in the form of satisfaction scores, VAS for pain, the VAS foot and ankle and survivorship data.  The results of the original study demonstrated that pain relief was greater following arthrodesis at 2 years.  At 15 years, patients with an arthrodesis experienced less pain and were more satisfied compared to those with an arthroplasty.  No functional differences were seen between these 2 groups.  There were more revisions in the arthroplasty group.  The authors concluded that despite the hope of better function, less pain, and greater satisfaction from MTPJ replacement, this was not found in the authors’’ patient population.  The long-term results of this study showed that arthrodesis out-performed arthroplasty.  If an arthroplasty failed, then salvage was likely to be technically difficult, with significant potential for complications.  Level of Evidence = I.

The METIS Prostheses

In a retrospective study, Kolodziej and colleagues (2013) evaluated functional and radiographic results of the first MPJ replacement with use of unconstrained, modular, 3- component, porous titanium and hydroxyapatite coated, press-fit METIS® prosthesis.  According to author's knowledge, results of this type of prosthesis have never been published before.  A total of 25 prostheses were implanted in 24 patients (were 20 females and 4 males) between February 2009 and May 2011; AOFAS-HMI was used to assess functional results.  Patients were also asked if they would undergo procedure again or recommend it to other people.  Weight-bearing radiographs ware made at final follow-up and analyzed for presence of osteolysis and radiolucency.  In 8 patients total joint replacement was introduced as a salvage treatment after failure of previous surgery like Keller resection arthroplasty, failed arthrodesis, avascular necrosis and post-operative arthritis.  The reasons for prosthetic replacement were HR (n = 11), rheumatoid arthritis (n = 4) and gout (n = 1).  Additional procedures were performed in 3 cases (Akin phalangeal osteotomy in 2 cases and fifth metatarsal osteotomy in 1 case).  The mean age at the operation was 56 years.  The average follow-up period was 18 months (range of 12 to 36 months).  The median post-operative value of AOFAS-HMI scores was 88 points (range of 75 to 95 points).  First metatarsophalangeal joint motion (dorsiflexion plus plantarflexion) was classified according to AOFAS-HMI ranges as: moderately restricted (between 30 to 70 degrees) in 19 patients 80 % (20 prosthesis) and severely restricted (less than 30 degrees) in 5 patients (20 %).  Overall, 15 (64 %) patients were completely satisfied, 5 (20 %) reported moderate satisfaction and 4 (16 %) were totally disappointed and would not undergo this procedure again.  A limited hallux dorsiflexion was the main dissatisfaction reason.  Partial radiolucent line was seen in 1 patient (4 %).  There were 2 serious complications.  In 1 patient, with rheumatoid arthritis, deep infection occurred 12 months after prosthesis implantation.  In the second case phalangeal implant was revised due to misalignment.  The authors concluded that the METIS® MPJ replacement allowed alleviate of pain relating to HR and partial restoration of joint movement, even in patients after failures of primary MPJ surgery.  AOFAS-HMI results were better than previously reported in the literature in assessment of the first MPJ replacement. These preliminary findings need to be validated by well-designed studies.

In a prospective study, Silva and colleagues (2015) evaluated the preliminary results from the METIS-Newdeal metatarsophalangeal prosthesis for treating hallux rigidus grade III/IV. A total of 8 metatarsophalangeal prostheses that were placed in 6 patients between November 2007 and July 2009 were included in this report.  The patients' mean age was 55 years and the mean follow-up after the surgery was 50 weeks.  The results were evaluated using the AOFAS-MTP score and x-ray images as controls.  The AOFAS-MTP score increased significantly from 42p before the surgery to 82p after the surgery (↑ 1.95x), mainly due to improvement in the functional level.  No intercurrences were identified radiologically.  Among the 5 patients who underwent operations, only 1 expressed dissatisfaction with the surgery: this was expressed after early infection appeared at the surgical site, and it was the only post-operative complication found.  The authors concluded that total metatarsophalangeal arthroplasty using METIS-Newdeal presented promising short-term results.  However, they stated that evaluations on a larger number of cases with a longer follow-up are needed in order to draw more consistent conclusions.

Dygrynova and associates (2017) evaluated the results of cheilectomy and TJR in patients with hallux rigidus.  Minimum duration of follow-up was 18 months.  The study included 59 patients who underwent surgery due to hallux rigidus between January 2013 and December 2014; 37 patients underwent cheilectomy and 22 patients had total joint arthroplasty using the METIS.  The outcomes were assessed by comparing pre-operative and post-operative ROM, VAS, AOFAS-HMI and patients' satisfaction with operative treatment.  Pre-operative and post-operative outcomes were compared for the individual types of surgery using the repeated measures ANOVA.  The level of statistical significance was set at p < 0.01.  The mean age was 47.9 ± 7.0 years in patients who underwent cheilectomy and 62.5 ± 5.5 years in patients after TJR METIS.  There was a significant decrease (p < 0.001) in the VAS pain score and a significant improvement in dorsiflexion, ROM, AOFAS-HMI scores in both the treatment groups.  In both the groups more than 75 % of patients reported good or excellent subjective results.  The authors stated that these findings were in agreement with findings of other studies assessing the results of cheilectomy and TJR surgery in patients with hallux rigidus.  Direct comparison of the VAS pain score, AOFAS-HMI and ROM across studies was difficult because of variability in the evaluation systems.  They stated that cheilectomy is mostly recommended for young active patients with mild osteoarthritis.  Moreover, it is also possible to use minimally invasive surgery with early and reliable outcomes.  These investigators performed cheilectomy also in younger patients with moderate osteoarthritis in order to extend the period of clinically acceptable results and thereby to postpone the TJR indication.  They stated that TJR (similarly to arthrodesis of the 1st MTP joint) is a procedure performed in elderly patients with low physical activity and more advanced deformities.  The authors concluded that both the reported methods offer reliable and valuable short-term clinical outcomes with relatively low complication rate.  They stated that cheilectomy is undoubtedly more appropriate for younger patients with mild or moderate arthritic changes.  Although it did not appear to alter the natural progression of the disease process, it provided satisfactory pain relief, motion improvement and overall patient gait comfort for patients in a short-term period.  They stated that TJR appeared to be a better solution for less active older patients to whom it provides a loadable, painless, and moving joint.  This was a small study (n = 22 for the total joint replacement group) with short-term follow-up (18 months).

The ToeFit-Plus Prosthesis

The ToeFit-Plus is a modular implant system for the hemi-arthroplasty or total replacement of the first MPJ.  The non-cemented implant is fixed in the host bone by means of a self-tapping threaded taper.

Duncan et al (2014) investigated the outcomes of first MP replacement for HR using the Smith & Nephew ToeFit-Plus implant.  These investigators assessed the outcomes of 69 first MPJ replacements using the AOFAS score pre-operatively and annually post-operatively, with retrospective radiologic review.  All operations were performed by the same surgeon within 2 centers.  A total of 69 arthroplasties were performed within the study period (57 patients).  The median AOFAS score at 1 year was 100 (interquartile range [IQR] 100 to 100), at 2 years was 100 (IQR 95 to 100), at 3 years was 100 (IQR 87.5 to 100), and at 4 years, it was 100 (IQR 91.25 to 100).  Radiolucencies around the phalangeal component were seen in 23 cases; however, this was symptomatic in only 2 patients, who required revision surgery.  The authors concluded that these early results showed that first MPJ replacement surgery with the ToeFit-Plus™ prosthesis resulted in significant improvements in the AOFAS scores for most patients; however, longer term follow-up is needed to monitor the clinical effect of radiolucency around the phalangeal component.

Erkocak et al (2013) stated that although MTP arthrodesis has been advocated by many authors, implant arthroplasty appears to be successful option in advanced HR.  These investigators evaluated the early results of the ToeFit-Plus prosthesis for the treatment of HR.  Between December 2007 and January 2011, a total of 26 toes of 24 patients with MTP arthritis of the great toe were treated with ToeFit-Plus implant.  The average follow-up time was 29.9 (range of 25 to 62) months.  All patients were evaluated clinically and radiographically.  Post-operative satisfaction and function were scored according to the AOFAS score.  Pain was assessed with the use of a VAS.  Mean pre-operative AOFAS score improved from 42.7 (range of 36 to 59) to 88.5 (range of 59 to 98) at the final follow-up (p < 0.01).  Pre-operative average VAS pain scores improved from 7.4 pre-operatively to 1.9 at the final follow-up (p < 0.01).  The average MTP joint ROM improved from 25.9 degrees pre-operatively to 53.8 degrees at the final follow-up.  No radiologic loosening was found, but radiolucency was observed in 2 patients with this implant.  No revision was required for any of the patients during the follow-up period.  The authors concluded that this total first MTP joint prosthesis yielded good functional outcome and high patient satisfaction level with low early complication rate.  Salvage arthrodesis remains an option if future revisions are indicated.  These preliminary findings need to be validated by well-designed studies with larger sample size and longer follow-up.

In a single-surgeon, case-series study, Titchener et al (2015) evaluated the mid-term results of first metatarsophalangeal joint replacement for hallux rigidus using ToeFit-Plus. These investigators prospectively studied the outcomes of 86 toes in 73 patients using the AOFAS-HMI score and radiological follow-up.  The mean follow-up was 33 months (2 to 72); 10 patients have been lost to follow-up; 8 patients sustained intra-operative fractures or impending fractures requiring circlage wiring; 18 joints have either been revised or listed for revision giving a revision rate of 24 %; this occurred at a mean of 33 months post-surgery.  The authors concluded that 1st MTPJ replacement is an option in hallux rigidus and patients who are not revised experienced significant improvement in AOFAS scores in the medium term.  These investigators had previously published satisfactory results with this prosthesis.  However the revision rate was unacceptably high and they had discontinued its use completely.

Mermerkaya and Adli (2016) evaluated the short- to mid-term outcomes of metatarsal head-resurfacing hemi-arthroplasty and total MTPJ arthroplasty (total joint replacement [TJR]) as surgical treatments for advanced-stage HR.  From 2012 to 2014, all data from patients who underwent surgery for the treatment of grades 2 to 3 HR were retrospectively reviewed, and 45 patients were included in this study.  Of these patients, 26 underwent metatarsal head-resurfacing hemi-arthroplasty (Group I) and 19 underwent TJR (Group II).  All patients were clinically graded prior to surgery and at their final follow-up visits using the AOFAS-HMI scale, VAS, and the 1st MTPJ ROM score.  Metatarsal head resurfacing was performed on 26 patients; 2 patients underwent bilateral procedures, yielding a total of 28 cases in Group I; TJR was performed on 19 patients in Group II.  Of the 26 Group I patients, 12 (46.2 %) were men and 14 (53.8 %) were women, with a mean age of 56.3 ± 4.5 years (range of 47 to 63 years); the mean follow-up duration was 29.9 ± 5.2 months.  Of the 19 Group II patients, 8 (42.1 %) were men and 11 (57.9 %) were women, with a mean age of 57.1 ± 5.8 years (range of 45 to 66 years); the mean follow-up duration was 27.1 ± 7.5 months.  Significant improvements were evident in the AOFAS scores, and the VAS scores decreased, in both groups.  No significant difference was evident between groups I and II.  The authors concluded that after failure of conservative treatment in patients with moderate-to-severe HR, both MTPJ hemi-arthroplasty and TJR were associated with effective recovery of toe function and MTPJ ROM, as well as good short- to mid-term functional outcomes.

This study was limited by its observational and retrospective design and relatively small sample size (n = 45).  Another drawback was that all procedures were performed by 2 surgeons, using a standardized technique, in 2 centers.  Furthermore, these investigators did not evaluate patient satisfaction (e.g., by using the Medical Health Outcomes Short-Form 36-item survey instrument).  The authors stated that a prospective, multi-center randomized trial is needed.  Comparative studies of the long-term outcomes of various surgical techniques, with larger case series of similar patients, are needed.  If conservative treatment failed in patients with moderate-to-severe HR, 1st-MTPJ hemi-arthroplasty and TJR effectively permit recovery of toe function and 1st-MTPJ ROM, in addition to affording good short- to mid-term functional outcomes.

Gupta and Masud (2017) stated that HR is osteoarthritis (OA) affecting the MTPJ of the 1st toe.  Patients often complain of pain and stiffness with pain being aggravated by walking, particularly during toe-off in the gait cycle.  Osteoarthritis of the MTPJ is commonly treated with arthrodesis or resection arthroplasty.  Metallic replacement of this joint is used sometimes but is not widely accepted.  The use of silastic joints has problems with synovitis and implant failure.  These investigators used titanium implants, which can be screwed into the metatarsal and phalanx, allowing good fixation without the use of bone cement.  Release of the tight plantar capsule and tissues is necessary to achieve better ROM and correct implant positioning.  In this study, a total of 55 cases of OA of the 1st MTPJ were treated surgically with Toefit-Plus joint replacement.  The implant consists of both metatarsal and phalangal components and a fixed-bearing polyethylene insert.  All patients had a release of tight soft tissues on the plantar side.  Follow-up occurred at 84 to 144 months after surgery (mean of 134 months), and the results showed increasing numbers of implant failures and revisions (21 %) of Toefit-Plus implants; 47 patients were available for review; 24 (51 %) out of 47 patients reported satisfactory results with Toefit-Plus arthroplasty; 10 of these patients (21 %) had removal of implants and further surgical procedures were needed due to implant failure; 11 (23 %) out of 47 patients still complained of pain despite having joint replacement with the Toefit-Plus  implant.  There was a high rate of complications with the Toefit-Plus implant resulting in revision surgery.  Patients should have the risks associated with arthroplasty clearly explained, including the risk of revision, and the option of arthrodesis should be discussed when planning surgery.  The authors conclude that further trials and re-design of implants may help to improve results; they would not recommend the Toefit-Plus implant due to poor results seen in 1/3 of patients.

Interpositional Arthroplasty of the First Metatarsophalangeal Joint

Aynardi and colleagues (2017) stated that for patients with HR seeking a motion-sparing procedure, interposition arthroplasty is an alternative to fusion.  In a retrospective, case-series study, these researchers reported patient outcomes after interpositional arthroplasty for HR.  All patients undergoing interpositional arthroplasty at the authors’ institution from 2001 to 2014 were identified and a retrospective chart review was performed.  Follow-up was conducted through a telephone survey to obtain survivorship, satisfaction, and functional scores.  Survivorship of the interpositional arthroplasty procedure was defined as no subsequent surgery on the hallux after the index procedure.  Patients were excluded for incomplete records; complications were recorded.  From 2001 to 2014, a total of 183 patients were identified.  Of these, 14 were excluded for incomplete data, leaving 169 patients.  Of these, 133 had an average follow-up of 62.2 months (range of 24.3 months to 151.2 months).  The overall failure rate was 3.8 % (5/133).  Patient-reported outcome was rated as excellent in 65.4 % (87/133) or good in 24.1 % (32/133) of patients and fair or poor in 10.5 % (14/133) of patients.  Of 133 patients, 101 (76 %) were able to return to fashionable or regular foot-wear.  The infection rate was 1.5 % (2/133).  Patient-reported cock-up deformity of the 1st MTPJ occurred in 4.5 % (6/133) of patients.  In addition, 17.3 % (23/133) of patients reported metatarsalgia of the 2nd or 3rd MTPJ at the time of final follow-up, and there was no significant difference between interposition types (p = 0.441).  The authors concluded that interpositional arthroplasty for HR was found to have excellent or good results in most patients at a mean follow-up of 62.2 months.  Level of Evidence = IV.

Modular Implants

Dulgeroglu and Metineren (2017) noted that the treatment of advanced hallux rigidus remains controversial.  Only a few studies have analyzed the short- and mid-term results of MTP joint arthroplasty to treat patients with advanced hallux rigidus.  These researchers presented the short-term follow-up results of patients who underwent MTP joint arthroplasty.  They reviewed the medical records of 15 consecutive patients (3 men and 12 women) who had had grade 3 or 4 hallux rigidus diagnosed according to the Coughlin and Shurnas classification.  The age range at surgery was 44 to 74 (mean of 61.6) years.  The mean follow-up period was 21.7 (range of 18 to 28) months.  The mean change in the overall AOFAS Hallux-First Ray scale score was from 26.9 ± 2.3 pre-operatively to 78.7 ± 8.8 post-operatively (p < 0.005).  The mean change in the overall VAS score was from 8.3 ± 0.8 pre-operatively to 1.7 ± 0.7 post-operatively (p < 0.005).  The mean pre-operative 1st MTP joint ROM was 22.3° ± 7.7° (range of 15° to 45°), which had increased to 77° (range of 65° to 90°) at the final follow-up visit.  No patient required revision surgery or removal.  The authors concluded that these findings indicated that for patients with advanced-stage hallux rigidus refractory to conservative treatment, total joint arthroplasty could lead to good satisfaction and good functional results in the short-term.

Popelka and associates (2017) stated that the 1st MTP joint replacement ranks among the treatment methods of patients with hallux rigidus.  These researchers evaluated the short-term to mid-term outcomes and presented their clinical experience with the Medin PH-flex implant.  From January 2011 to 2016, these investigators performed total replacement of the first MTP joint in 31 patients, in 4 cases bilaterally.  In total, 35 implants were evaluated.  The mean age of the patient at the time of surgery was 57.7 years (39 to 72 years).  The surgery was conducted in 29 women and 2 men.  The patients were evaluated using the AOFAS score, the radiographs were assessed as to the potential occurrence of radiolucent lines, with major stress put on the assessment of the mobility in MTP joint and its position.  The pain was assessed based on the VAS score.  Prior to the joint replacement surgery, the mean AOFAS score in patients was 55.6 (35 to 65).  Post-operatively, the mean AOFAS score was 80.8 (65 to 95).  The pain suffered by patients was evaluated with the use of the pain VAS score.  The pre-operative mean VAS score was 5 (2 to 8), whereas the post-operative score improved to mean VAS 2 (0 to 4).  The ROM was clinically assessed with a goniometer.  The mean ROM of plantar flexion and dorsiflexion was 16.00° (5 to 35°) and 28.60° (10 to 55°), respectively.  The mean ROM was 36° (15 to 60°).  No intra-operative complications were observed.  In all the patients, the surgical wound healed per primam.  In 2 female (5.7 %) of the whole group of patients who underwent surgery a deep infection occurred, namely 10 and 21 months following the implantation.  In both the female patients their condition was managed by joint revision operation and by a simple removal of the implant.  The authors noted that joint replacement related matters were repeatedly discussed in professional literature.  There were many papers published in the literature on this topic.  A whole range of the first MTP joint implants of different shapes have been developed, with extremely different clinical results.  The authors concluded that an appropriately chosen type of the implant, a fitting indication and a correctly applied implantation technique can lead to the desired good outcome.  The 1st MTP joint replacement should be indicated after careful consideration since the management of a potential joint replacement failure can often be very technically challenging and quite mutilating for the patient.  The mid-term outcomes of the Medin 1st MTP implant appeared to be promising.  Moreover, they stated that it will be necessary to wait for long-term outcomes in order to evaluate the final benefits of this type of implant in patients with hallux rigidus.

Wassink and co-workers (2017) evaluated the results following total 1st MTP (FMTP) joint replacement arthroplasty using a modular 3-component press fit prosthesis at 2- year follow-up.  All patient data were collected in a prospective way in 4 study centers.  Both pre-operative and post-operative evaluation consisted of an assessment using the AOFAS-HMI score, VAS for pain, evaluation of the ROM and patient satisfaction scores.  Post-operative X-rays were reviewed for loosening and radiolucency up to 2 years.  A total of 55 feet were available for analysis at 24 months; 2 implants were removed during the study; 6 more feet had additional surgery due to stiffness or mal-alignment.  Post-operative AOFAS-HMI scores improved significantly by 32.4 points at 2-year follow-up (p < 0.001).  The VAS for pain improved significantly from 6.8 (std 1,6) pre-operatively to 1.6 (std 1,9) post-operatively (p < 0.0001).  Mean dorsiflexion improved from 12.6 (std 10,1) degrees pre-operatively to 31.2 (std 16,8) degrees post-operatively; 87 % of patients were moderately to well-satisfied with the end result; 18 prostheses showed radiolucency at 24 months.  The authors concluded that implantation of a Metis modular 3-component press fit prosthesis for the MTP joint in hallux rigidus showed significant improvement in AOFAS-HMI scores and a decrease in pain.  Moreover, they stated that concerns remained with regard to early re-operation rate (14.5 %) and long-term survival of the implant.  They stated that future studies will have to address these aspects.

Stevens and colleagues (2017) stated that hallux rigidus is a common cause of foot pain in the elderly and has a negative impact on quality of life (QOL).  Several operative therapeutic options are available for feet that are refractory to conservative treatment.  Of these, TJR and arthrodesis of the 1st MTP joint are the most commonly performed interventions.  Nevertheless, it is still unclear which intervention results in the best clinical outcome and the fewest complications.  These investigators performed a systematic review on the clinical outcome following TJR and arthrodesis for hallux rigidus.  PubMed/Medline, Embase, and the Cochrane Library were systematically searched for studies assessing outcome with the AOFAS-HMI score, Foot Function Index (FFI), VAS for pain, or SF-36 in patients who underwent an arthrodesis or TJR for the treatment of symptomatic hallux rigidus.  Secondary outcomes were complications and revision rates.  The screening of titles and abstracts, data collection, data extraction, and study quality assessment were performed independently by 2 reviewers.  Study quality was determined with use of risk-of-bias tools.  Results of included studies were presented in a qualitative manner, and the results of high-quality studies were pooled.  A total of 33 studies, describing a total of 741 arthrodeses and 555 TJRs, were included in the qualitative analysis; 6 different prostheses were used for TJR, and various fixation techniques were used for arthrodesis.  The results of 6 arthrodesis studies and 7 TJR studies were pooled in the quantitative analysis.  Pooled results showed superiority of arthrodesis compared with TJR for improving clinical outcome (by 43.8 versus 37.7 points on the AOFAS-HMI score) and reducing pain (a decrease of 6.56 versus 4.65 points on the VAS pain score).  Because of the rare reporting of the FFI and SF-36, no comparison could be made for these outcomes.  Fewer intervention-related complications (23.1 % versus 26.3 %) and revisions (3.9 % versus 11 %) were reported after arthrodesis as compared with TJR, with pain and nonunion and prosthetic loosening being the most commonly reported complications after arthrodesis and TJR, respectively.  The authors concluded that the findings of the present systematic review of the literature indicated that arthrodesis is superior for improving clinical outcome and reducing pain, and is less often accompanied by intervention-related complications and revisions, compared with TJR in patients with symptomatic hallux rigidus.  Moreover, they stated that prospective RCTs are needed to verify this conclusion.  Level of Evidence = IV.

Cartiva Synthetic Cartilage Implant

Sciarretta (2013) stated that despite the various treatment options available, symptomatic articular cartilage defects continue to represent a therapeutic challenge for knee surgeons.  These researchers retrospectively evaluated and presented long-term results, from 5 to 8 years, of PVA-H hydrogel implants ("Cartiva") in the treatment of knee chondral focal defects.  Presented were the clinical and MRI evaluation of 18 patients with III and IV degree knee chondral or osteochondral defects treated by PVA-H hydrogel implants.  Defects were no larger than 20 mm in diameter.  Average age at time of surgery was 54 years.  There were 11 male and 7 female patients.  A total of 20 implants were implanted, 16 with 10-mm and 4 with 15-mm Cartiva implants.  One patient was treated bilaterally.  The majority (80 %) of surgeries were performed arthroscopically.  Patients have been assessed by IKDC, SF36 scores and by MRI imaging.  All patients have shown improvement of knee function and knee scores, in many cases over 50 points of IKDC, except 3.  Of these, the 1st maintained a rather good quality of life for over 5 years with approximately the same level of functionality she enjoys now, after implant removal and knee replacement during 2008.  The 2nd case was a 43-year old female with a post-traumatic chondral defect in a valgus knee.  The patient experienced post-op pain: the implant was removed among another institution at 6 months post-op and was converted to OATS.  The 3rd case was a 49-year old male with a severe arthritic pre-op knee and may not have been ideal candidate for Cartiva, but was too young for a total knee replacement (TKR) at the time of operation and, despite a severe knee worsening during the last year that will need in the short future a knee replacement, has, at over 6 years follow-up, an IKDC score of 33.33 from a pre-op of 37.93.  The authors concluded that 5- to 8-year follow-ups enabled them to conclude that the use of PVA-H synthetic implants in knee chondral defects in middle aged patients can guarantee critical knee function improvement and severe pain reduction.  Even the patients, who have needed a knee replacement, had done well for the first 4 to 5  years; meaning that this type of treatment with the correct indications and future implant and instrumentation improvements, already in course, may guarantee a several year-period of knee health and active life style.  This was a small study (n = 18) that examined the use of Cartiva for knee chondral or osteochondral defects (not the big toe); and 5 to 8 years follow-up provided only medium-term, not long-term, results.

Baumhauer and colleagues (2016) stated that although a variety of great toe implants have been tried in an attempt to maintain toe motion, the majority have failed with loosening, mal-alignment/dislocation, implant fragmentation and bone loss.  In these cases, salvage to arthrodesis is more complicated and results in shortening of the ray or requires structural bone graft to reestablish length.  This prospective study compared the safety and effectiveness of this small (8/10 mm) hydrogel implant to the gold standard of a great toe arthrodesis for advanced-stage hallux rigidus.  In this prospective, randomized non-inferiority study, patients from 12 centers in Canada and the United Kingdom were randomized (2:1) to a synthetic cartilage implant or first metatarsophalangeal (MTP) joint arthrodesis. VAS pain scale, validated outcome measures (Foot and Ankle Ability Measure [FAAM] sport scale), great toe active dorsiflexion motion, secondary procedures, radiographic assessment, and safety parameters were evaluated.  Analysis was performed using intent-to-treat (ITT) and modified ITT (mITT) methodology.  The primary end-point for the study consisted of a single composite end-point using the 3 primary study outcomes (pain, function, and safety).  The individual subject's outcome was considered a success if all of the following criteria were met: (i) improvement (decrease) from baseline in VAS pain of greater than or equal to 30 % at 12 months; (ii) maintenance of function from baseline in FAAM sports sub-score at 12 months; and (iii) absence of major safety events at 2 years.  The proportion of successes in each group was determined and 1-sided 95 % confidence interval (CI) for the difference between treatment groups was calculated.  Non-inferiority of the implant to arthrodesis was considered statistically significant if the 1-sided 95 % lower confidence interval was greater than the equivalence limit (less than 15 %).  A total of 236 patients were initially enrolled; 17 patients withdrew prior to randomization, 17 patients withdrew after randomization, and 22 were non-randomized training patients, leaving 152 implant and 50 arthrodesis patients.  Standard demographics and baseline outcomes were similar for both groups.  VAS pain scores decreased significantly in both the implant and arthrodesis groups from baseline at 12 and 24 months.  Similarly, the FAAM sports and activity of daily living sub-scores improved significantly at 12 and 24 months in both groups.  First MTP active dorsiflexion motion improvement was 6.2 degrees (27.3 %) after implant placement and was maintained at 24 months.  Subsequent secondary surgeries occurred in 17 (11.2 %) implant patients (17 procedures) and 6 (12.0 %) arthrodesis patients (7 procedures); 14 (9.2 %) implants were removed and converted to arthrodesis, and 6 (12.0 %) arthrodesis patients (7 procedures [14 %]) had isolated screws or plate and screw removal.  There were no cases of implant fragmentation, wear, or bone loss.  When analyzing the ITT and mITT population for the primary composite outcome of VAS pain, function (FAAM sports), and safety, there was statistical equivalence between the implant and arthrodesis groups.  The authors concluded that the findings of this study showed equivalent pain relief and functional outcomes.  The synthetic implant was an excellent alternative to arthrodesis in patients who wished to maintain first MTP motion.  The percentage of secondary surgical procedures was similar between groups.  Less than 10 % of the implant group required revision to arthrodesis at 2 years.  This study provided only short-term results (12 to 24 months).

Baumhaurer and associates (2017) conducted a prospective, randomized study to examine outcomes of arthrodesis compared to synthetic cartilage implant in patients with hallux rigidus.  Patients underwent pre-operative clinical examination, radiographic assessment, hallux rigidus grade assignment, and intra-operative assessment of cartilage loss; VAS score for pain was obtained pre-operatively and at 24 months.  Correlation was made between active peak dorsiflexion, VAS pain, cartilage loss, and hallux rigidus grade.  Fisher's exact test was used to assess grade impact on clinical success (p < 0.05).  In 202 patients, 59 (29 %), 110 (55 %), and 33 (16 %) were classified as Coughlin grades 2, 3, and 4, respectively.  There was no correlation between grade and active peak dorsiflexion (-0.069, p = 0.327) or VAS pain (-0.078, p = 0.271).  Rank correlations between grade and cartilage loss were significant, but correlations were small.  When stratified by grade, composite success rates between the 2 treatments were nearly identical.  The authors concluded that irrespective of the grade, positive outcomes were demonstrated for both fusion and synthetic cartilage implant.  Clinical symptoms and signs should be used to guide treatment, rather than a grade consisting of radiographic, symptoms, and ROM factors.  This study provided only short-term follow-up (24 months).

Goldberg and co-workers (2017) examined data from a clinical trial of 1st metatarsophalangeal joint (MTPJ1) implant hemiarthroplasty and arthrodesis to determine the association between patient factors and clinical outcomes.  Patients greater than or equal to 18 years with hallux rigidus grade 2, 3, or 4 were treated with synthetic cartilage implant MTPJ1 hemiarthroplasty or arthrodesis.  Pain VAS, Foot and Ankle Ability Measure (FAAM) sports and activities of daily living (ADL) scores, and SF-36 Physical Function (SF-36 PF) sub-score were obtained pre-operatively, and at 2, 6, 12, 24, 52, and 104 weeks post-operatively.  Final outcome data, great toe active dorsiflexion motion, secondary procedures, radiographs, and safety parameters were evaluated for 129 implant hemiarthroplasties and 47 arthrodeses.  The composite primary end-point criteria for clinical success included VAS pain reduction greater than or equal to 30 %, maintenance/improvement in function, no radiographic complications, and no secondary surgical intervention at 24 months.  Predictor variables included hallux rigidus grade; gender; age; body mass index (BMI); symptom duration; prior MTPJ1 surgery; pre-operative hallux valgus angle, ROM, and pain.  Two-sided Fisher exact test was used (p < 0.05).  Success rates between implant MTPJ1 hemiarthroplasty and arthrodesis were similar (p > 0.05) when stratified by hallux rigidus grade, gender, age, BMI, symptom duration, prior MTPJ1 surgery status, and pre-operative VAS pain, hallux valgus, and ROM.  The authors concluded that synthetic cartilage implant hemiarthroplasty was appropriate for patients with grade 2, 3, or 4 hallux rigidus.  Its results in those with associated mild hallux valgus (less than or equal to 20 degrees) or substantial pre-operative stiffness were equivalent to MTPJ1 fusion, irrespective of gender, age, BMI, hallux rigidus grade, pre-operative pain or symptom duration. 

The above policy is based on the following references: