Revisional First Metatarsophalangeal Joint Arthrodesis

Revisional First Metatarsophalangeal Joint Arthrodesis

Alan Ng, Brian Derner


The first metatarsophalangeal joint (MTPJ) can be a source of extreme pain and disability for a wide age range. Originally, it was described by Davies-Colley in 1887 as “hallux flexus” due to a plantarflexed position of the hallux (great toe) on the first metatarsal.1 Radiographically, osteoarthritis of the first MTPJ is very common around the world, with a prevalence between 5% and 42% among patients 18 years or older. Traditionally, females have a higher prevalence of hallux rigidus.29 Radiographic changes include osteophyte formation, loose bodies, periarticular lipping of the joints, subchondral sclerosis, cyst formation, decreased joint space narrowing, enlarging sesamoids, sesamoid arthritis, widening of the metatarsal head, and (in end stages) bony ankylosis of the joint.10 There are not only radiographic changes to a patient’s foot but also biomechanical implications. Patients with hallux rigidus and limitus have decreased joint range of motion, leading to decreased push off during the gait cycle, and increased load bearing on the lesser metatarsals. In turn, this could lead to lesser metatarsalgia and potential for arthritis or more proximal joint arthritis.

Multiple classification systems have also been created to guide treatment for patients with hallux limitus and rigidus.11,12 Most classifications indicate worsening osteophytic lipping and decreased joint space along with pain during range of motion. Specifically, the Coughlin and Shurnas classification delineates stage 3 and stage 4 (end-stage) deformities based on whether the patient has pain at mid-range of motion or not, despite the same radiographic parameters in both stages.11

A myriad of different surgeries or interventions have been published and performed to decrease pain and increase function of the first MTPJ. As the first MTPJ decreases in range of motion and increases with pain or crepitus present, the surgical options change. Regardless, the etiology of pain must be addressed for optimal surgical outcomes. In patients with metatarsus primus elevatus and early-stage first MTPJ arthritis, joint preserving procedures like plantarflexory metatarsal osteotomies, dorsiflexory phalangeal osteotomies, and cheilectomies have been described and used with positive results.1316 Joint destructive procedures, such as implant arthroplasties (either hemi or total implant), the Valenti osteotomy, and the Keller arthroplasty have also been used with long-term follow-up. For older and low-demand patient populations, joint destructive procedures can provide satisfactory outcomes.17,18 However, these procedures are not without complications. For example, implant arthroplasties can have aseptic loosening or implant subsidence, leading to further pain and dysfunction.19,20 Joint preservative osteotomies can also lead to shortening, nonunion, or transfer metatarsalgia. Regardless of the initial procedure of choice, patients can have worsening arthritis of the first MTPJ, necessitating another surgical intervention.

The gold standard for correction of hallux rigidus is the first MTPJ arthrodesis. It was originally described by McKeever in 1952 for hallux valgus, hallux rigidus, and metatarsus primus varus.21 This procedure has been shown to correct first MTPJ pathology in a predictable manner with high patient satisfaction and union rates. Roukis22 performed a systematic review in over 2500 patients with a nonunion rate of 5.4% in their patient population; 32.7% of this population had a symptomatic nonunion. Many different types of fixation have been described in the literature for performing a first MTPJ arthrodesis, including Kirschner wires, various screw orientation constructs, dorsal locking plates (with or without interfragmentary compression screws), flexible nails, and staples.22

A primary concern for patients is the lack of first MTPJ motion after arthrodesis; however, Donegan and Blume in their case series of over 200 first MTPJ arthrodesis procedures had a >87% patient satisfaction rate with their surgery.23 Doeselaar and colleagues showed a statistically significant increase in the Foot Function Index for their first MTPJ arthrodesis cohort, regardless of an etiology of hallux valgus or hallux rigidus.24 Da Cunha et al., in their young cohort between 18 and 55 years of age, found that their patient population had an 88.6% return to maximal physical activity with a 96% satisfaction rate after an average of 5.1 years.25 These studies illustrate that even in young patient cohorts, there is a very high patient satisfaction rate with long-term outcomes.

In the case of revision surgery for first MTPJ pathology, arthrodesis is the procedure of choice owing to its biomechanical stability and lower complication profile, with reliable subjective and objective outcomes. Myerson et al. performed revisional first MTPJ arthrodesis in 22 patients with an average of 62.7 months of follow-up. They used either iliac crest autograft or femoral head allograft for their definitive fixation constructs with a 79.1% arthrodesis rate at an average of 13.3 weeks. The mean American Orthopedic Foot and Ankle Society (AOFAS) score improved from 39 to 79. They also noted an increase of 13 mm in length compared with initial preoperative radiographs.26 Garras et al. published on 21 hemiarthroplasties that were converted to first MTPJ arthrodesis with calcaneus, distal tibia, or iliac crest autograft. All 21 patients had radiographic and clinical union in their cohort. They noted statistically significantly improved time to arthrodesis and Foot and Ankle Ability Measure (FAAM) scores in their patients with local bone grafting versus iliac crest autograft.27 Interestingly enough, Usuelli and colleagues reported on their 11 patients with prior first MTPJ arthroplasty and revision with calcaneal dowel autograft and first MTPJ arthrodesis. However, they only had a 75% arthrodesis rate (3 nonunions, 1 symptomatic). They did note statistically significant improvements in the Visual Analog Score, AOFAS score, and Foot Ankle Disability Index in their patient population.28 Overall, union rates of autograft bone block revision arthrodesis for first MTPJ arthrodesis range from 75% to 100%.2629

Further expanding on Myerson et al.’s hypothesis of allograft bone block arthrodesis, Luk and colleagues performed this revision technique on 15 patients with an average follow-up of 46 weeks. Their study showed an 87% union rate with a 93% patient satisfaction rate. An average of an 18-mm bone graft was placed along with bone morphogenic protein (BMP) in 7 of the 15 patients. In 14 of 15 patients, they fixated the graft with a dorsal plate and screws. Mean FAAM and Activities of Daily Living scores improved from 63.9 to 83.5; most notably, patients had difficulty squatting and rising up on their toes.30 Bei et al performed their bone block arthrodesis with 90-90 locking plates in 11 patients. They noted a 90.9% union rate, with radiographic union at a mean of 10.7 weeks. Of note, they used allograft iliac crest, demineralized bone matrix (DBM), and bone marrow aspirate (BMA) from the calcaneus to supplement their osseous arthrodesis. Mean AOFAS score statistically significantly improved from 44.5 to 87.75; however, they did have an 18.2% complication rate.31

The major benefit of allograft is the lack of donor site morbidity, which can lead to additional complications in revision arthrodesis procedures. Although there are few studies on allograft bone grafting, the results for union rate are similar compared with the autograft cohort. However, in these types of revision cases, the use of autograft allows for the additional osteogenic properties to aid in osseous arthrodesis of the first MTPJ versus only osteoconductive and osteoinductive properties in allografts. The results with allograft bone block arthrodesis are promising and should have further comparative cohort studies regarding its outcomes compared with autograft.

Another option for revision and segmental bone loss are titanium truss cages, either computed tomography (CT) 3D reconstructed or stock shapes. These are reconstructive new options in foot and ankle surgery and have minimal to no studies regarding their efficacy and patient outcomes. There are only a handful of cases in the literature without long-term follow-up.32 However, this is a salvage option for the reconstructive surgeon with first ray (metatarsal and great toe) bony defects.

If a patient has significant shortening of the first ray from prior procedures and revisions, callus distraction of the first ray has been described. In a systematic review, complications of this procedure were up to 33% and led to double the healing time of a single-staged autograft bone block arthrodesis; however, this was for lesser ray brachymetatarsia.33 There is also a case series with regards to reconstruction of 7 juvenile patients with Apert syndrome. Patients had positive functional outcomes but a 35.7% complication rate including further revisions due to early union, pin site infections, or pin loosening.34 Callus distraction should be used in extreme situations, with minimal to no other options and multiple attempts at single-stage procedures.


The need for a revisional first MTPJ arthrodesis arises from multiple etiologies and prior procedures. In most literature, failed prior procedures such as hemi or total joint arthroplasties (metallic or silastic) leave large bone gaps after removal and must have concomitant bone graft in conjunction with a revision arthrodesis.26 Other reasons include prior bunion surgery, failed cheilectomy, prior distal metatarsal osteotomy for hallux limitus or rigidus with continued pain and dysfunction, prior first MTPJ arthrodesis with malunion or symptomatic nonunion, hallux varus, or prior interpositional arthroplasty. In addition, Keller arthroplasties with a cock-up hallux deformity and/or worsening lesser metatarsalgia are also candidates for revisional first MTPJ arthrodesis.35 Less common indications may include benign neoplasm with bony deficit or infected hardware or infected nonunion with subsequent bone loss.36 Geriatric patients or patients with rheumatoid arthritis with large bunion deformities also have indications for first MTPJ arthrodesis procedures. In addition, prior avascular necrosis of the first metatarsal head from a bunion could lead to a potential bone block arthrodesis as a salvage procedure.37 If biomechanical deformities of the first ray are not corrected, central or lateral column pathology can occur. If one does not appropriately manage the etiology of this pathology, this could lead to worsening pain and dysfunction.


Patients with active infection should not have definitive revisional first MTPJ arthrodesis. The patient can be staged with cultures and biopsy, placement of polymethylmethacrylate (PMMA) cement (antibiotic or not), and eventual definitive arthrodesis once an infection is eradicated.26 Consultation with infectious disease specialists and appropriate antibiotic administration should be completed prior to definitive treatment. Some relative contraindications to first MTPJ arthrodesis are hallux interphalangeal joint arthrosis, prior hallux interphalangeal joint (IPJ) arthrodesis, or inability for patient to have no motion of the first MTPJ. The latter could be due to their occupational demands or hobbies, such as gardening or plumbing.38 A possible hallux IPJ takedown and subsequent first MTPJ arthrodesis could be a viable option. On the contrary, biomechanical sequelae from arthrodesis of the distal first ray (IPJ and MTPJ) lead to poorer patient outcomes than a single joint arthrodesis of the IPJ or MTPJ.

Other relative contraindications include past smoking history or a current smoker, severe vitamin D deficiency, or osteoporosis. Any medical condition or medication that may prevent osseous union or wound healing compromise could be perceived as a relative contraindication. Therefore, proper workup is needed in these patients with medically complex condition with appropriate consultants.

Preoperative Considerations

Prior to revisional first MTPJ surgery, a thorough history and physical examination of the patient needs to be completed. Specifically, preoperative radiographs and procedure type with knowledge of prior implants are important for definitive revisional surgical plan.

If a patient had a past history of a wound complication, infection, or inflammatory episode, these factors must also be weighed for appropriate surgical plan and outcome. Inflammatory markers like C-reactive protein and erythrocyte sedimentation rate must be taken prior to surgery in patients with history of infection or clinical signs of infection around the first MTPJ. In these patients, intraoperative STAT Gram stains should be performed along with bone biopsies to determine if a patient is ready for definitive fixation or a staged surgery with antibiotic cement placement.

Patients with medical comorbidities should be medically optimized by the appropriate specialists. Specifically, patients with osteopenia or osteoporosis should be seen by an endocrinologist and have proper supplementation of their vitamin D. Typically, if a patient is undergoing an arthrodesis, laboratory tests for vitamin D3, ionized calcium, and parathyroid hormone levels should be obtained. Patients with a vitamin D level of less than 20 nmol/L should be supplemented with 50,000 international units (IU) of ergocalciferol (vitamin D2) once weekly in addition to 5000 IU of cholecalciferol (vitamin D3) daily. Patients with a vitamin D3 level of greater than 20 nmol/L should be supplemented with 5000 IU vitamin D3 daily.

If patients have absent pedal pulses, skin atrophy, loss of digital hair, and delayed capillary refill time, they should be referred to a vascular surgeon for further management. However, these types of patients are not typically suited for arthrodesis procedures of the distal extremities and would be better suited with conservative management or a Keller-type procedure, if appropriate. Patients with an extended smoking history should be counseled on possible risks and complications with regards to the procedure and arthrodesis in general. Patients who actively smoked in the perioperative period having forefoot surgery historically have a 20% greater chance of a surgical complication than patients with a prior history of smoking.39 Patients currently smoking should attempt smoking cessation for at least 8 weeks prior to surgery. It has been shown that each week of cessation prior to surgery can lead to a relative risk decrease of complications by 19%.40 If patients have a past allergy history to specific metals or may possibly have a metal allergy, consultation to an allergy specialist should be considered. Once completed, a more definitive plan with specific hardware with considerations of the patient’s metallic allergies should be used.

If a patient had a prior first MTPJ arthrodesis that was malpositioned or a symptomatic nonunion, a CT scan should be performed. This helps finalize a surgical plan and estimates the osseous void to anticipate the use of an allograft or autograft for revision. In addition, the CT scan can show cystic formation within the proximal phalanx and first metatarsal head, which could further shape the surgical plan.

Once these parameters are met and the patient is ready to undergo surgery, these patients may undergo monitored anesthesia care in conjunction with local anesthesia, if allografts are used for revision. However, if the surgical plan involves iliac crest autograft, surgical coordination with the orthopedic team must be made and general anesthesia is used by the anesthesiology team. In addition, if calcaneal autograft is used, a popliteal and adductor canal block could be performed prior to procedure. In the day and age of the pandemic COVID-19, patients also can undergo spinal anesthesia to avoid aerosolization of respiratory droplets from the patient and putting the surgical team at less risk of transmission.

A sterile ankle or midcalf tourniquet can be applied once preparation and draping is completed if no autograft is needed. A nonsterile thigh or calf tourniquet could also be utilized in order to gain access to the tibia or calcaneus for BMA or autograft harvesting. Patients are placed supine on the operating table with a bump underneath the ipsilateral hip to obtain orthogonal positioning of the foot, with the toes perpendicular to the operative bed. The bump could be increased to obtain an increased internal rotation of the leg, if calcaneal grafting is desired.

In addition, a sagittal saw or cup and cone reamers should be available depending on the shape of the osseous void or joint preparation for the surgeon. The sagittal saw could also be used for autograft harvesting. Additional DBM or BMP adjuvants can be on standby to increase osteoconduction and osteoinduction to the revision construct. These products should be used in patients with poor ability to form bone or older age to increase the chance of union. Cancellous bone chips can also be used to fill some caps and provide additional scaffold for osseous turnover and potential arthrodesis. BMA can also be used to soak an allograft or autograft bone block. A small drill bit (ie, 2.0 mm in size) should be used for subchondral drilling; however, one may choose to use a Kirshner wire for joint preparation instead. In the case of prior history of infection, the surgeon may choose to use external fixation techniques for the resected osteomyelitis of the first MTPJ arthrodesis. With any fixation construct, the surgeon must be able to pivot to other fixation techniques in the case of poor bone purchase, history of infection, allergies, or other reasons.

Surgical Description

A standard dorsal incision, just lateral to the extensor hallucis longus tendon, is advised to allow for lengthening of the incision as necessary for longer fixation constructs. Once the periosteal and capsular dissection is completed, the first MTPJ is visualized. If there is a prior implant or hardware existing, it would be removed at this time. If an implant such as a stemmed silastic implant was utilized, the canals of the implant should be burred to remove any synovitis or membrane that prevents osseous bleeding to occur that will aid in osseous union. In the author’s experience, cup and cone reaming is performed, along with subchondral drilling, in these revisions to allow for appropriate placement of an anatomically contoured allograft. The graft is convex distally and concave proximally and cylindrical in shape. These allografts come in numbers of different lengths and widths to custom fit the bony deficit made by the prior surgery, if one exists. In addition, DBM products and recombinant human bone morphogenic protein-2 (RHBMP-2) adjuvants are used in combination with the allograft. In patients younger than 50 years, calcaneal autografting is used to supplement the allograft to aid in osseous arthrodesis.

Once all biological elements are placed, appropriate positioning of the first MTPJ is determined and provisionally fixated with a Kirschner wire. Typically, the great toe is abducted 5° to 10°, with 0° of frontal plane rotation and minimal dorsiflexion of the great toe. A simulated weight-bearing surface should be placed under the foot and the distal hallux IPJ should be just purchasing the surface. The foot should be biomechanically loaded and the pressure of the first and fifth metatarsals should be similar to re-establish the tripod of the foot for appropriate long-term stability. The length of the first ray should be established as to not cause lesser metatarsalgia. A trial graft is placed in the arthrodesis site and inspected on intraoperative C-arm fluoroscopy. The length should allow for lengthening of the first ray but should not cause increased tension on the soft tissue structures that may cause neurovascular compromise.

Stay updated, free articles. Join our Telegram channel

Oct 22, 2022 | Posted by in ORTHOPEDIC | Comments Off on Revisional First Metatarsophalangeal Joint Arthrodesis

Full access? Get Clinical Tree

Get Clinical Tree app for offline access