First MTPJ Arthrodesis

Primary indication. Symptomatic end-stage first metatarsophalangeal joint (MTPJ) disease — most commonly Coughlin and Shurnas Grade IV hallux rigidus (bone-on-bone, rigid, globally painful joint) — that has failed non-operative care (stiff-soled or rocker-bottom footwear, NSAIDs, orthoses, and where appropriate a cheilectomy or injection). Absolute indications

• Grade IV hallux rigidus: bone-on-bone with a rigid, globally painful first MTPJ
• Failed cheilectomy for advanced hallux rigidus with persistent or worsening symptoms
• Rheumatoid arthritis forefoot with first MTPJ destruction (often combined with lesser MTPJ procedures)
• Salvage of a failed Keller's (resection) arthroplasty — requires structural bone graft to restore length Relative indications
• Grade III hallux rigidus in heavy-demand patients where cheilectomy alone is unlikely to maintain function
• Hallux valgus with coexisting first MTPJ osteoarthritis where correction and arthrodesis provide a combined solution
• Neuromuscular hallux valgus with MTPJ degeneration (e.g. cerebral palsy, poliomyelitis) where soft-tissue correction alone is insufficient
• First ray hypermobility with symptomatic MTPJ osteoarthritis
• Septic arthritis with irreversible joint destruction after infection clearance The threshold rule (Coughlin and Shurnas, 2003). Arthrodesis is indicated for Grade IV, or for Grade III with under 50 percent of the metatarsal head cartilage remaining at surgery. Cheilectomy remains the procedure of choice for Grades I and II and for selected Grade III joints where more than half the head cartilage is preserved. The full five-grade classification is detailed under Background and Evidence. Contraindications
• Active infection at the operative site (relative — treat the infection first; delayed arthrodesis is acceptable)
• Severe peripheral vascular disease precluding healing (absolute)
• Avascular necrosis of the first metatarsal head without revascularisation potential (relative)
• Established interphalangeal joint (IPJ) arthritis — not an absolute contraindication, but the patient must be counselled that fusion accelerates IPJ loading
• Unrealistic expectations about post-fusion footwear limitations Consent specifically for: permanent stiffness of the joint (the mechanism of pain relief), a non-union rate of roughly 2 to 10 percent (higher in smokers), hardware prominence requiring later removal, transfer metatarsalgia or IPJ arthritis if the fusion is malpositioned, wound problems in thin rheumatoid skin, and the higher complexity of any Keller's salvage. Setup. Supine, foot at the end of the table so the surgeon can sit or stand at the foot. Thigh or calf tourniquet for a bloodless field. Image intensifier (mini C-arm) available for intraoperative position and fixation checks. Have local reamings available, and iliac crest graft if a Keller's salvage is planned. Anaesthesia: general, spinal or ankle block. Position the foot so it can be laid flat on the table or a sterile back surface for the ground-clearance test.

The goal is to excise the arthritic joint surfaces to bleeding cancellous bone, hold the hallux in the functional fusion position, and achieve union. The exposure is straightforward but the position is the operation — every step serves the final alignment. The standard route is a dorsal midline approach (preferred when a dorsal plate is used); a medial approach is an alternative for crossing-screw fixation. Surgical anatomy that drives the technique

• Osseous. The first metatarsal head is large and rounded with medial and lateral condyles and a plantar sulcus cradling the sesamoids; its articular surface extends further plantarward than dorsally, so plantar cartilage must also be removed. The proximal phalanx base is concave — the rationale for cup-and-cone (dome) reaming, which reproduces this convex–concave geometry and maximises cancellous contact.
• Soft tissue. The medial capsule is thick (contains the medial collateral ligament); the dorsal capsule is thinner and elevated as a full-thickness flap. The plantar plate resists dorsiflexion. Extensor hallucis longus (EHL) crosses dorsally and is retracted; flexor hallucis longus (FHL) runs plantarly in its fibro-osseous tunnel and is not directly at risk.
• At risk. The medial plantar digital nerve runs along the medial plantar surface of the phalanx and metatarsal head — vulnerable during medial periosteal dissection and phalanx reaming. The first dorsal metatarsal artery is safe with standard approaches.

View larger

View larger

Rehabilitation | Phase | Timing | Immobilisation / weight-bearing | Therapy focus | |-------|--------|---------------------------------|---------------| | 1 | 0 to 2 weeks | Stiff-soled shoe or cast; non- or heel weight-bearing | Wound check and suture removal at 2 weeks | | 2 | 2 to 6 weeks | Heel weight-bearing in a stiff-soled shoe | Protect fusion until early callus on radiograph | | 3 | 6 to 12 weeks | Progressive weight-bearing as tolerated | Radiological union typically at 8 to 12 weeks | | 4 | Long term | Rocker-bottom or stiff toe-box footwear | Activity modification; reduce IPJ loading | Most patients return to desk work by about 6 weeks. Around 85 to 90 percent of well-positioned fusions unite; radiological union is typically seen at 8 to 12 weeks.

Non-union work-up. Serial radiographs at 6 and 12 weeks; CT if there is doubt at 12 to 16 weeks. Exclude infection (CRP, ESR, white cell count) and identify contributors (smoking, nutrition, bone density). Revision is repeat arthrodesis with bone graft plus upgraded fixation — a dorsal plate if one was not used primarily. Keller's salvage specifics. Remove all soft-tissue interposition and periosteum from the resected proximal phalanx stump, measure the length deficit and match it with a structural bone block (iliac crest autograft remains the gold standard), and span both graft–host interfaces with a dorsal locking plate — crossed screws alone are mechanically inadequate across a graft–host interface. Reported union in specialist salvage series is lower than for primary fusion (roughly 85 to 90 percent).

Pathoanatomy. Hallux rigidus is degenerative arthritis of the first MTPJ. It begins dorsally with osteophyte formation and dorsal impingement at terminal dorsiflexion, then progresses circumferentially to global cartilage loss, subchondral sclerosis and cyst formation, sesamoid involvement, and ultimately a rigid bone-on-bone joint. The Coughlin and Shurnas five-grade classification (below) captures this progression and directs treatment.

The landmark classification study. The Coughlin and Shurnas 2003 long-term series is the basis for the modern five-grade clinical-radiographic classification and for the threshold at which fusion replaces cheilectomy. A common exam trap is to misquote the headline numbers: in that paper the 92 percent good-or-excellent figure refers to the cheilectomy procedures (86 of 93), and 97 percent of patients overall had a good or excellent subjective result. The authors' actual treatment rule is that Grade IV — or Grade III with under 50 percent of the metatarsal head cartilage remaining at surgery — should be treated with arthrodesis, whereas cheilectomy was used with predictable success for Grades I, II and selected Grade III. Arthrodesis versus joint-sacrificing alternatives. The best single piece of comparative evidence is the Gibson and Thomson RCT, in which arthrodesis gave significantly greater pain relief than total toe replacement and all fusions united, while 6 of 39 phalangeal implant components required removal for loosening. Raikin and colleagues separately showed arthrodesis outperformed metallic hemiarthroplasty (24 percent hemiarthroplasty failure). The McNeil, Baumhauer and Glazebrook evidence-based review then placed this in context: arthrodesis is the only procedure for hallux rigidus that reaches fair evidence (Grade B), with cheilectomy, osteotomy, implant, resection and interpositional arthroplasty all at Grade C. Fixation and joint preparation. Goucher and Coughlin reported the benchmark dome-reamer technique (50 patients, 54 feet) using a low-profile dorsal titanium plate with crossed lag screws — 96 percent satisfaction, a 92 percent union rate (4 of 54 nonunions, 8 percent). Flavin and Stephens described an analogous low-profile contoured plate plus compression screw over a ball-and-socket preparation, with radiographic union at 6 weeks and dorsiflexion set at 20 to 25 degrees relative to the first metatarsal. The honest position on fixation is that no high-quality study has shown one construct is decisively superior: in Hyer and colleagues' comparative series, time-to-fusion did not differ between crossed screws and dorsal plate (the plate group actually had numerically more nonunions, 3 versus 1, in a small cohort) while the plate cost significantly more. The mechanical argument for a dorsal plate plus interfragmentary (lag) screw is that it is a tension-band-style construct resisting the dorsal bending moment generated at push-off; it is reasonable to prefer it for revision, poor bone stock or graft interfaces, but a candidate should not overstate the union-rate evidence. Functional and positional rationale. Brodsky and colleagues' prospective gait analysis objectively confirmed that successful fusion increases maximal ankle push-off power and single-limb support time — the biomechanical justification for fusing the joint in slight dorsiflexion rather than flat. The positional convention dates to Mann and Oates (95 percent fusion in 41 toes) and is best expressed against two reference frames: relative to the first metatarsal shaft aim for roughly 15 to 25 degrees dorsiflexion and 5 to 10 degrees valgus; relative to the floor (weight-bearing) the hallux pulp should sit only a few millimetres above the ground because of the metatarsal's natural inclination. Insufficient dorsiflexion overloads the IPJ at push-off; excessive dorsiflexion causes dorsal footwear pressure and lesser-ray transfer metatarsalgia. Relevant biomechanics. The first ray bears approximately 40 percent of forefoot load during gait, rising to 60 percent or more at push-off. Following arthrodesis the IPJ becomes the primary remaining mobile joint of the hallux, so excessive MTPJ dorsiflexion concentrates ground-reaction force at the IPJ during push-off — the mechanical basis for the IPJ arthritis that follows an over-dorsiflexed fusion. The sesamoids give flexor hallucis brevis a mechanical advantage and protect the plantar metatarsal head; after fusion they are fixed in position, so accurate position prevents dorsal sesamoid impingement.

Verified citations. According to PubMed, the following references have been individually verified (title, authors, journal, year, pages and PMID). DOI links are provided where PubMed returned a DOI. 1. Coughlin MJ, Shurnas PS. Hallux rigidus: grading and long-term results of operative treatment. J Bone Joint Surg Am. 2003;85(11):2072–2088. PMID 14630834. Landmark five-grade classification; 97 percent of patients good/excellent overall and 92 percent of cheilectomies successful; arthrodesis indicated for Grade IV or Grade III with under 50 percent metatarsal head cartilage remaining. 2. McNeil DS, Baumhauer JF, Glazebrook MA. Evidence-based analysis of the efficacy for operative treatment of hallux rigidus. Foot Ankle Int. 2013;34(1):15–32. PMID 23386758. DOI. Systematic review of 135 studies; arthrodesis the only procedure reaching fair evidence (Grade B); all alternatives Grade C. 3. Gibson JNA, Thomson CE. Arthrodesis or total replacement arthroplasty for hallux rigidus: a randomized controlled trial. Foot Ankle Int. 2005;26(9):680–690. PMID 16174497. DOI. RCT (22 arthrodesis vs 27 arthroplasty); greater pain relief and all-united fusions versus 6 of 39 implant components removed; cost ratio 2:1 favouring fusion. 4. Goucher NR, Coughlin MJ. Hallux metatarsophalangeal joint arthrodesis using dome-shaped reamers and dorsal plate fixation: a prospective study. Foot Ankle Int. 2006;27(11):869–876. PMID 17144945. DOI. 50 patients/54 feet; dome reaming with low-profile dorsal titanium plate and crossed lag screws; 92 percent union (8 percent nonunion), 96 percent satisfaction. 5. Flavin R, Stephens MM. Arthrodesis of the first metatarsophalangeal joint using a dorsal titanium contoured plate. Foot Ankle Int. 2004;25(11):783–787. PMID 15574236. DOI. Prospective series; low-profile contoured plate plus compression screw over ball-and-socket preparation; union at 6 weeks, dorsiflexion 20 to 25 degrees relative to first metatarsal. 6. Mann RA, Oates JC. Arthrodesis of the first metatarsophalangeal joint. Foot Ankle. 1980;1(3):159–166. PMID 7319432. DOI. 41 toes in 28 patients; 95 percent fusion; foundational outcome series for first MTPJ arthrodesis including rheumatoid forefoot. 7. Raikin SM, Ahmad J, Pour AE, Abidi N. Comparison of arthrodesis and metallic hemiarthroplasty of the hallux metatarsophalangeal joint. J Bone Joint Surg Am. 2007;89(9):1979–1985. PMID 17768195. DOI. Severe first MTPJ OA; 24 percent hemiarthroplasty failure with plantar stem cut-out; all 27 arthrodeses united; AOFAS-HMI, pain and satisfaction superior with fusion. 8. Brodsky JW, Baum BS, Pollo FE, Mehta H. Prospective gait analysis in patients with first metatarsophalangeal joint arthrodesis for hallux rigidus. Foot Ankle Int. 2007;28(2):162–165. PMID 17296132. DOI. Instrumented gait analysis in 23 patients; significant increase in maximal ankle push-off power and single-limb support after fusion. 9. Hyer CF, Glover JP, Berlet GC, Lee TH. Cost comparison of crossed screws versus dorsal plate construct for first metatarsophalangeal joint arthrodesis. J Foot Ankle Surg. 2008;47(1):13–18. PMID 18156059. DOI. 45 fusions; overall fusion 91.1 percent; no significant difference in time-to-fusion between crossed screws and dorsal plate; plate implants cost significantly more.