Hallux rigidus · Coughlin-Shurnas Grade I–II · joint-preserving resection of the dorsal 30% of the first metatarsal head
- Cheilectomy is the joint-preserving operation for Coughlin-Shurnas Grade I–II hallux rigidus with at least 50% of the metatarsal head plantar cartilage preserved. Grade IV, or Grade III with less than 50% of the head cartilage remaining at surgery, is treated with arthrodesis.
- The 30/70/50 rule governs the whole operation: resect the dorsal 30% of the metatarsal head (the osteophyte plus the adjacent articular surface) with an osteotome or saw at 30–45°, confirm at least 70° of intra-operative dorsiflexion before closure, and preserve at least 50% of the plantar cartilage.
- The benchmark evidence is Coughlin and Shurnas (2003 JBJS): 92% of cheilectomies (86 of 93 feet) were successful at a mean 9.6 years, and 97% of patients had a good or excellent result.
- Two structures are critical in the dorsomedial approach — the medial branch of the dorsal cutaneous nerve (a branch of the superficial peroneal nerve) and the extensor hallucis longus tendon with its medial expansion. Identify and protect both before any capsulotomy.
- Cheilectomy does not burn bridges — overall revision is 8.8% (Roukis 2010, 706 cases), most commonly conversion to arthrodesis, and salvage arthrodesis outcomes are comparable to primary arthrodesis.
When & Why
Indication. Symptomatic hallux rigidus — dorsal first MTP joint pain on push-off and at end-range dorsiflexion, a firm dorsal bony prominence, reduced dorsiflexion with preserved plantar flexion, and a dorsal metatarsal osteophyte on weight-bearing radiographs — graded Coughlin-Shurnas Grade I–II (and selected Grade III with at least 50% of the metatarsal head cartilage preserved), that has failed conservative management (a minimum 3–6 month trial of stiff-soled or rocker-bottom footwear, activity modification, anti-inflammatories and orthotics). Absolute indications - Coughlin-Shurnas Grade I — dorsal osteophyte impingement with pain at end-range dorsiflexion, full or near-full passive motion, normal or near-normal plantar articular cartilage.
- Coughlin-Shurnas Grade II — moderate restriction of dorsiflexion (less than 50% of normal), with at least 50% of the plantar cartilage preserved on pre-operative assessment.
- Documented failure of conservative management over at least 3–6 months. Relative indications - Isolated dorsal pain with active dorsiflexion impingement in a younger, active patient.
- A dorsal ganglion or synovitis at the first MTP joint refractory to aspiration or injection.
- Combined hallux valgus and early hallux rigidus where the hallux valgus correction is the primary procedure and the cheilectomy is performed adjunctively. Contraindications — these point to arthrodesis instead (Coughlin-Shurnas 2003): - Grade III with less than 50% of the metatarsal head articular cartilage remaining at surgery.
- Grade IV — a stiff, painful joint with pain through mid-range motion and severe cartilage loss.
- Active infection at or around the first MTP joint, or severe vascular insufficiency limiting wound healing.
- Inflammatory arthropathy with active synovitis (treat the systemic disease first).
- Unrealistic patient expectations about residual stiffness. The one decision that matters — the cartilage threshold. Cheilectomy is a joint-preserving operation, so its success rests on the quality of the plantar metatarsal head cartilage. Confirm the grade and the cartilage status before committing, and counsel the patient that if less than 50% of the head cartilage is preserved, arthrodesis is the better operation.
For Grade I–II and selected Grade III with at least 50% of the head cartilage preserved. Resects the dorsal osteophyte and 30% of the dorsal metatarsal head to relieve impingement. Does not burn bridges.
Adds a dorsal closing-wedge osteotomy of the proximal phalanx in the high-demand Grade II patient, shifting preserved plantar flexion into functional dorsiflexion through the MTP-to-IP axis.
The procedure of choice for Grade IV, or Grade III with less than 50% of the head cartilage, and the definitive salvage for a failed cheilectomy. Outcomes after failed cheilectomy match primary arthrodesis.
Consent specifically for medial-side numbness or a painful neuroma, wound problems or superficial infection, some permanent loss of terminal dorsiflexion, a small risk of incomplete pain relief or recurrent impingement, and the roughly 1-in-12 chance of later conversion to arthrodesis (Roukis 2010). Setup. Supine with a sandbag under the ipsilateral buttock to rotate the foot and improve medial access, foot at the end of the table. An ankle tourniquet at 250 mmHg is sufficient (a thigh tourniquet is an alternative). An ankle block (posterior tibial, superficial peroneal, deep peroneal, sural, saphenous) with long-acting local anaesthetic gives excellent post-operative analgesia for day-case surgery. Standard small-joint set with a 6–10 mm osteotome, rongeur, small burr, and an oscillating saw if preferred.
The Operation
The goal is to expose the first MTP joint through a dorsomedial approach, remove the dorsal osteophyte plus the dorsal 30% of the metatarsal head while protecting the medial dorsal cutaneous nerve and the EHL tendon, confirm at least 70° of intra-operative dorsiflexion, and mobilise the patient immediately in a stiff-soled shoe. The exposure is laid out in full below (and in depth on the dorsomedial approach to the first MTP joint page).
Operative sequence
- Supine, sandbag under the ipsilateral buttock, foot at the end of the table; ankle tourniquet at 250 mmHg.
- Palpate and mark the first MTP joint line, the dorsal osteophyte on the metatarsal head, and the course of the EHL tendon centrally.
- Plan a 5–7 cm dorsomedial incision in the internervous plane between the medial branch of the dorsal cutaneous nerve (medial) and the EHL (central).
- A longitudinal 5–7 cm incision centred over the dorsomedial aspect of the first MTP joint, from about 3 cm proximal to 3 cm distal to the joint line.
- Incise skin only; deepen carefully with scissors through subcutaneous fat — the medial branch of the dorsal cutaneous nerve lies just superficial to the capsule in this plane.
- Identify the medial branch of the dorsal cutaneous nerve (superficial peroneal) early and retract it gently with a vessel loop — it crosses the incision zone and is the structure most often injured.
- Confirm the EHL tendon and its medial expansion; retract it laterally with a narrow Langenbeck or self-retaining retractor before any capsulotomy.
- Avoid electrocautery near the nerve.
- Longitudinal capsulotomy just medial to the EHL, from 2 cm proximal to 2 cm distal to the joint line, elevating medial and lateral flaps for exposure (a Z-plasty or T-shaped capsulotomy is used only if the capsule is contracted).
- Elevate the periosteum off the dorsal metatarsal head to give about 270° of exposure — dorsal, medial and lateral — to fully visualise the osteophyte. Do not elevate the plantar periosteum (it risks avascular necrosis of the metatarsal head).
- Plantarflex the MTP joint to inspect the metatarsal head and assess the plantar cartilage directly.
- If less than 50% of the plantar cartilage is preserved and arthrodesis was discussed and consented, convert at this sitting; otherwise stage it.
- Document loose bodies, synovitis and the extent of the osteophyte on every aspect of the joint.
- Place a 6–10 mm osteotome at the dorsal margin of the normal articular cartilage, angled at 30–45° to the metatarsal shaft to resect the dorsal 30% of the metatarsal head (the osteophyte plus the adjacent dorsal articular surface).
- Drive with a single controlled mallet strike (bimanual control of the osteotome); an oscillating saw angled at 30–45° with copious irrigation is the alternative.
- The cut should emerge at or just proximal to the dorsal cortical apex of the osteophyte — too little leaves impingement, too much violates the plantar cartilage.
- Inspect the dorsal base of the proximal phalanx — a phalangeal osteophyte is present in 70–80% of patients and must be removed with a rongeur or small osteotome.
- Smooth every cut bone edge and any sharp cortical margin with a rongeur and small burr to prevent post-operative bony irritation and heterotopic recurrence.
- With the tourniquet released (or manual compression eased), test passive dorsiflexion of the hallux at the MTP joint.
- The target is at least 70° of dorsiflexion from neutral. If it is not achieved, re-inspect for a retained lateral cortex, a sharp dorsal ridge or a missed phalangeal osteophyte, and resect further under direct vision.
- Do not close until 70° is confirmed — failure to reach this is the strongest predictor of persistent symptoms and early failure.
- Decompress the joint with a small laminar spreader to inspect the plantar recess.
- Remove all osteocartilaginous loose bodies with a small rongeur or grasper.
- Lavage the joint with at least 500 mL of saline before closure.
- Close the capsule with absorbable sutures (2-0 Vicryl or PDS), apposing without tension with the toe in neutral — avoid over-tight closure, which recreates the dorsiflexion restriction.
- Subcutaneous layer 3-0 Vicryl (confirm the nerve is not caught in a suture); skin with 3-0 or 4-0 nylon or absorbable interrupted sutures.
- Non-adherent dressing with soft padding; no plaster. Apply a stiff-soled post-operative shoe and mobilise, full weight-bearing, immediately.
The medial branch of the dorsal cutaneous nerve (superficial peroneal) runs just superficial to the capsule along the medial border of the EHL and crosses the incision zone — identify it in the subcutaneous plane and retract it gently; electrocautery near it causes a painful neuroma. The EHL tendon and its medial expansion lie centrally over the joint, immediately adjacent to the capsulotomy line — confirm and retract it laterally before any capsule incision. Both must be named and protected before bone work begins.
Before closure, confirm at least 70° of passive dorsiflexion from neutral. Failure to reach 70° is the leading cause of early failure and persistent impingement. If the target is not met, return to the joint and resect under direct vision — typically a retained lateral cortex or a missed phalangeal osteophyte — and do not close until it is achieved. Document the intra-operative range in the operative note.
The dorsomedial approach allows medial capsule repair and keeps the EHL retracted laterally on its natural expansion, away from the line of bone resection. A central dorsal approach brings the cut directly over the EHL and risks the tendon and its expansion. Reserve the dorsal approach for specific indications.
Angle the osteotome or saw at 30–45° to the metatarsal shaft and visualise the transition from osteophyte to healthy articular cartilage before committing to the resection plane. Over-resection into the plantar weight-bearing surface destroys the bearing cartilage and mandates conversion to arthrodesis; subperiosteal elevation must stay dorsal only.
Aftercare & Complications
Rehabilitation | Phase | Timing | Weight-bearing & footwear | Therapy | |-------|--------|---------------------------|---------| | 1 | 0–2 weeks | Immediate full weight-bearing in a stiff-soled post-operative shoe; elevate above the heart for 48 hours | Gentle active dorsiflexion from day 1–2; ankle-block analgesia covers the first 12–24 hours | | 2 | 2–6 weeks | Stiff-soled shoe to about week 3, then a wide toe-box shoe; avoid high heels for at least 3 months | Progressive active and passive ROM, scar mobilisation once healed | | 3 | 6 weeks – 3 months | Regular wide footwear | Strengthening; return to low-impact sport at 6–8 weeks | | 4 | 3–6 months | Normal footwear | Return to running and high-impact sport at 3–4 months; expect some permanent loss of terminal dorsiflexion | Most patients are in comfortable regular footwear by 6 weeks. Pain relief is typically early (2–6 weeks) and range of motion continues to improve for 3–6 months. Follow up at 2 weeks (wound check and suture removal), 6 weeks (range, footwear, return to activity), 3 months (pain, function, AOFAS score) and 12 months (final outcome, with a weight-bearing radiograph if symptomatic). Warn the patient to return urgently for increasing pain, erythema or wound discharge after day 5 (deep infection), loss of dorsiflexion after initial improvement (heterotopic bone or inadequate resection), or new numbness or dysaesthesia (nerve injury). Complications
- Incidence
- 10–20% when the intra-operative ROM check is omitted
- Prevention
- Resect the dorsal 30% of the metatarsal head; confirm at least 70° dorsiflexion before closure
- Management
- Revision cheilectomy if dorsiflexion is under 70° and symptoms persist beyond 6 months; arthrodesis if the plantar cartilage is now compromised
- Incidence
- 2–5%; highest with a mis-angled oscillating saw
- Prevention
- Angle the osteotome or saw at 30–45° to the shaft; visualise the osteophyte-to-cartilage transition before cutting
- Management
- If recognised intra-operatively and less than 50% of the plantar cartilage remains, convert to arthrodesis at the same sitting if consented
- Incidence
- 5–10%; the commonest nerve complication
- Prevention
- Identify the nerve early in the superficial dissection; retract gently; avoid electrocautery near it
- Management
- Neuropraxia: observe 3–6 months. Painful neuroma: desensitisation and a corticosteroid injection; excision and burial of the proximal stump if persistent beyond 12 months
- Incidence
- 2–5%; higher with diabetes, peripheral vascular disease or steroids
- Prevention
- Avoid skin tension at closure; use interrupted sutures; stiff-soled shoe to protect the wound; perioperative antibiotics
- Management
- Superficial: local wound care and oral antibiotics. Deep infection: surgical debridement and IV antibiotics; arthrodesis if the joint is destroyed
- Incidence
- 8.8% overall revision after isolated cheilectomy (Roukis 2010); concentrated in Grade IV (55.6%)
- Prevention
- Select correctly — avoid isolated cheilectomy in Grade IV or Grade III with under 50% head cartilage; counsel every patient pre-operatively
- Management
- First MTP arthrodesis is the definitive salvage; outcomes are comparable to primary arthrodesis — cheilectomy does not burn bridges
Viva & Exam Focus
CHORDCHORD — the five key steps of cheilectomy
Clinical Decision Scenarios
Practise clinical reasoning and management decisions out loud
“A 48-year-old recreational runner has 18 months of dorsal first MTP pain on push-off, a firm dorsal bony prominence, dorsiflexion of 25° (normal 70–90°), preserved plantar flexion and pain only at the extremes of dorsiflexion. Weight-bearing films show a dorsal metatarsal osteophyte with minimal joint-space narrowing. How do you grade this patient and what are the surgical options?”
“Walk me through your technique for cheilectomy. How much bone do you remove, how do you know when you have removed enough, and what are the two structures most at risk during the approach?”
“A 55-year-old woman had a cheilectomy 2 years ago with initial improvement, but her pain has returned and she now has only 20° of dorsiflexion. Radiographs show a recurrent dorsal osteophyte and moderate joint-space narrowing. What has happened and what is your management?”
Indication
- Coughlin-Shurnas Grade I–II (and selected Grade III) hallux rigidus with at least 50% of the head cartilage preserved
- Failed 3–6 months of conservative care: stiff-soled or rocker-bottom footwear, NSAIDs, activity modification, orthotics
- Contraindicated in Grade IV and in Grade III with under 50% head cartilage — arthrodesis instead (Coughlin-Shurnas 2003)
Exposure
- Dorsomedial 5–7 cm incision centred on the first MTP joint
- Internervous plane between the medial branch of the dorsal cutaneous nerve and the EHL
- Longitudinal capsulotomy medial to EHL; elevate dorsal periosteum 270° — dorsal, medial, lateral, never plantar
The 30 / 70 / 50 rule
- Resect 30%: the dorsal 30% of the metatarsal head with an osteotome or saw at 30–45°
- Achieve 70°: confirm at least 70° of passive dorsiflexion before closure — the non-negotiable check
- Preserve 50%: at least 50% of the plantar cartilage must be intact — below it, do arthrodesis
- Include the phalangeal osteophyte (present in 70–80%) and smooth every edge
Two danger structures
- Medial branch of the dorsal cutaneous nerve (superficial peroneal) — identify and retract gently
- EHL tendon and its medial expansion — confirm and retract laterally before any capsulotomy
- Plantar metatarsal head (70%) — do not elevate plantar periosteum; over-resection mandates arthrodesis
Aftercare
- Immediate full weight-bearing in a stiff-soled shoe from day 0
- Stiff-soled shoe to about week 3, then wide toe-box footwear; no heels for 3 months
- Return to running and high-impact sport at 3–4 months
Landmark evidence
- Coughlin-Shurnas 2003: 92% of cheilectomies (86/93) successful at a mean 9.6 years
- Roukis 2010: 8.8% overall revision across 706 cheilectomies, Grade IV highest (55.6%)
- Cheilectomy does not burn bridges — salvage arthrodesis matches primary arthrodesis
Background & Evidence
Pathoanatomy. Hallux rigidus is degenerative arthritis of the first metatarsophalangeal joint characterised by dorsal osteophyte formation and progressive loss of dorsiflexion. The dorsal articular surface of the metatarsal head (about 30% of the head) bears the brunt of the wear; the plantar 70% is usually well preserved in early and moderate disease and is the functional weight-bearing and toe-off surface that cheilectomy is designed to spare. As the dorsal osteophyte grows it impinges against the dorsal base of the proximal phalanx at end-range dorsiflexion, producing the characteristic pain on push-off; a dorsal phalangeal osteophyte is present in 70–80% of cases and adds to the impingement. Why the Coughlin-Shurnas grade drives every decision. The Coughlin-Shurnas system grades hallux rigidus on dorsiflexion range, radiographic appearance (osteophytes, joint space, sclerosis) and the clinical pain pattern. It is the most prognostically validated of the many systems described (Beeson 2008 critically reviewed more than 18 systems and found most lacked formal validation, while Coughlin-Shurnas remains the most clinically applied), and its Grade IV category — pain through mid-range motion with severe cartilage loss — is the line beyond which a joint-preserving operation fails. The 50% plantar cartilage threshold. The single most important selection criterion is the quality of the metatarsal head articular cartilage. Coughlin and Shurnas (2003) showed cheilectomy was reliable for Grade I, Grade II and selected Grade III disease, but that Grade IV, or Grade III with less than 50% of the metatarsal head cartilage remaining at surgery, should undergo arthrodesis. Assess the plantar cartilage intra-operatively by plantarflexing the joint, and convert to arthrodesis at the same sitting if the threshold is not met and the patient has been consented. Classification — Coughlin-Shurnas (2003)
- Dorsiflexion / loss
- 40–60° (10–20% loss)
- Radiographic features
- No osteophyte; normal joint space; stiffness only
- Management
- Non-operative — splint, stiff-soled footwear, NSAIDs
- Dorsiflexion / loss
- 30–40° (20–50% loss)
- Radiographic features
- Dorsal osteophyte; minimal joint-space narrowing
- Management
- Cheilectomy — the first-line operation
- Dorsiflexion / loss
- 10–30° (50–75% loss)
- Radiographic features
- Moderate osteophytes; mild–moderate narrowing and sclerosis
- Management
- Cheilectomy appropriate; add a Moberg osteotomy for the high-demand patient
- Dorsiflexion / loss
- 10–30°, plantarflexion also reduced
- Radiographic features
- Severe osteophytes; marked narrowing; subchondral cysts or sclerosis
- Management
- Cheilectomy only if at least 50% head cartilage remains; otherwise arthrodesis
- Dorsiflexion / loss
- As Grade 3 but pain throughout mid-range
- Radiographic features
- Severe cartilage loss
- Management
- Arthrodesis — cheilectomy is contraindicated
Key evidence. Coughlin and Shurnas (2003) prospectively evaluated one surgeon's practice over 19 years: 110 of 114 patients returned, 80 patients (93 feet) underwent cheilectomy at a mean 9.6-year follow-up, 97% of patients (107/110) had a good or excellent result and 92% of cheilectomies (86/93) were successful for pain relief and function — the benchmark reference examiners expect. Mann and Clanton (1988) established the early series (25 patients; pain relieved in 22/25; a mean 20° motion gain) showing that decompressing dorsal impingement reliably relieves pain. Roukis (2010) quantified the revision burden in a systematic review of 706 isolated cheilectomies: 8.8% were revised, most commonly to arthrodesis, with the revision rate concentrated in Grade IV (55.6%) — confirming that cheilectomy is a sound first-line treatment that does not burn bridges, while advanced disease belongs to arthrodesis. Stevens et al (2020) showed minimally invasive cheilectomy carried higher reoperation (12.8% versus 2.6%) and complication rates than the open dorsal technique, so the open approach remains the evidence-based default. The Moberg (1979) proximal phalangeal dorsal closing-wedge osteotomy remains a useful adjunct in the high-demand Grade II patient.
References
Hallux rigidus: grading and long-term results of operative treatment
- 110 of 114 patients reviewed; 80 patients (93 feet) had cheilectomy at a mean 9.6-year follow-up
- 97% of patients (107/110) a good or excellent subjective result; 92% of cheilectomies (86/93) successful for pain and function
- Cheilectomy reliable for Grade 1, Grade 2 and selected Grade 3; arthrodesis for Grade 4 or Grade 3 with under 50% head cartilage
- No association between hallux rigidus and first-ray hypermobility, functional hallux limitus or metatarsus primus elevatus
Hallux rigidus: treatment by cheilectomy
- 25 patients, mean follow-up 56 months; pain relieved in 22 of 25 (three failures)
- First MTP motion improved by a mean of 20°, reaching an acceptable arc in 23 patients
- No patient required additional surgery during follow-up
- Cheilectomy judged superior to arthrodesis, resection arthroplasty or flexible implant arthroplasty for this cohort
The need for surgical revision after isolated cheilectomy for hallux rigidus: a systematic review
- 23 studies, 706 isolated cheilectomies; 62 (8.8%) required surgical revision, most commonly arthrodesis
- Grade-specific revision: Grade I 20%, Grade II 14.8%, Grade III 9.1%, Grade IV 55.6%
- The very high Grade IV revision rate confirms advanced disease should not have isolated cheilectomy
- Overall low revision rate supports cheilectomy as a first-line surgical treatment
Comparison of complication and reoperation rates for minimally invasive versus open cheilectomy of the first metatarsophalangeal joint
- 171 cheilectomies (38 open, 133 minimally invasive), mean 3-year follow-up
- Reoperation: open 2.6% (1/38) versus MIS 12.8% (17/133), relative risk 4.86
- Complications: open 2.6% versus MIS 11.3%, relative risk 4.29
- MIS offers faster recovery and cosmesis but carries a higher relative risk of further surgery
Surgical management of hallux rigidus
- Comprehensive review: the surgical choice is dictated by the grade of involvement
- Early-to-mid-stage disease is best managed by cheilectomy, or cheilectomy plus a proximal phalanx (Moberg) osteotomy
- Arthrodesis and arthroplasty are reserved for late-stage disease
- Non-surgical management should be exhausted before operative referral in many patients
Further reading 1. Maceira E, Monteagudo M. Functional hallux rigidus and the Achilles-calcaneus-plantar system. Foot Ankle Clin. 2014;19(4):669–699. PMID 25456716. Review of functional hallux rigidus (normal non-weight-bearing dorsiflexion blocked under load); supports gastrocnemius release in selected patients. 2. Moberg E. A simple operation for hallux rigidus. Clin Orthop Relat Res. 1979;(142):55–56. PMID 498648. Original description of the dorsal closing-wedge osteotomy of the proximal phalanx as an adjunct to increase functional dorsiflexion. 3. Beeson P, Phillips C, Corr S, Ribbans W. Classification systems for hallux rigidus: a review of the literature. Foot Ankle Int. 2008;29(4):407–414. PMID 18442456. Critically reviews more than 18 classification systems; highlights the lack of formal validation while Coughlin-Shurnas remains the most clinically applied. 4. Smith RW, Katchis SD, Ayson LC. Outcomes in hallux rigidus patients treated nonoperatively: a long-term follow-up study. Foot Ankle Int. 2000;21(11):906–913. PMID 11103761. At a mean 14.4 years, 75% of feet would again choose non-operative care; pain often stable despite radiographic progression — supports a conservative trial before surgery. 5. Blyth MJ, Mackay DC, Kinninmonth AW. Dorsal wedge osteotomy in the treatment of hallux rigidus. J Foot Ankle Surg. 1998;37(1):8–10. PMID 9470110. Retrospective series of 18 dorsal closing-wedge (Moberg-type) osteotomies; good/excellent results in 14/18 at a mean 4 years, supporting the proximal phalangeal osteotomy as an effective adjunct.