Anterior Tarsal Tunnel Syndrome

What Works

The honest headline first: nobody has run a trial on this condition. No randomized trials, no systematic reviews, no meta-analyses, and no clinical guideline from NICE, APTA, BOA, EULAR, ACR or JOSPT names it. That is a gap in the research, not a verdict on treatment. Everything below is reasoned from where the nerve runs and what presses on it, and each item is labelled with exactly what backs it so you can weigh it yourself.

1. Take the pressure off the top of the foot Mechanistic

Re-lace around the sore point, re-pad or loosen a stiff boot tongue, change the shoe. Shoe modification is named in the review literature (DiDomenico 2006, PMID 16958392; Bojovic 2025, PMID 40042611) and the compression mechanism is consistent across every source retrieved.

What backs it: mechanism plus expert consensus. No outcome has ever been measured. This is still the highest-value action available, because it is free, reversible, and aimed squarely at the thing causing the problem.

2. Stay out of sustained pointed-toe positions Mechanistic

Follows the stretch mechanism proposed by Borges and colleagues, who presented evidence that this syndrome may result from abnormal stretch of the nerve rather than pure compression (1981, PMID 7463126).

What backs it: mechanism only. No trial has tested position modification.

3. Keep training, at full load Mechanistic

This is a compression problem, not a torn tissue. There is no healing structure to protect, so blanket rest costs you fitness and buys nothing. Change the footwear, not the program.

What backs it: mechanism, plus the general principle that unnecessary deloading has its own cost.

4. Corticosteroid injection Consensus only

Named as an option in both available reviews (PMID 16958392, PMID 40042611).

What backs it: nothing measured. No trial, no case series with outcomes, no dose, and no injection-accuracy data for this site. An artery shares the tunnel, so if it is used, image guidance is sensible reasoning rather than a published recommendation.

5. Surgical decompression, for identified structural causes that failed conservative care Emerging

Thirteen consecutive patients, minimum two-year follow-up: AOFAS hindfoot score went from 55 to 83 at twelve months and 88 at twenty-four months (Yassin 2015, PMID 26209470). The causes found were localized swelling of the joint lining in five, a bone spur in four, a small cyst in two, and unknown in two.

What backs it: one uncontrolled series of thirteen. The improvement is large and it lasted, and the design cannot separate the surgery from removing a lesion that was going to be removed anyway. There has never been a comparison group.

Cinematic anatomical rendering of the foot and ankle in low key light

Exercise Prescription

These do not fix the nerve. They keep the foot and ankle moving well while the irritation settles, and they keep you training.

Exercise	How	Sets × Reps	Frequency	Pain guide
Ankle circles and pumps	Sitting, move the ankle slowly through its comfortable range both directions. Stay out of the end position that tingles	2 × 15 each way	2× daily	No tingling, no burning
Toe spread and lift	Foot flat. Lift all the toes, spread them, lower slowly	2 × 10	Daily	Effort, not sharp or electric
Calf stretch, knee straight and bent	Standard wall stretch. Do not force the pointed-toe position, which stretches the nerve	3 × 30 sec each	Daily	Felt in the calf. Stop if felt on top of the foot
Barefoot walking	On a safe surface, with nothing pressing on the top of the foot	10 minutes	Daily where practical	Comfortable throughout

Dosing note: these numbers are conventional safe starting points chosen for that reason, not values taken from a study of this condition. No study of this condition has ever published one.

What Doesn't Work

Rest and generic strengthening. The problem is a local pressure or a local end position. Cutting your training volume addresses neither.
Compressive taping or strapping over the top of the foot. That is adding compression to a compression problem. It is the one intervention that actively works against the mechanism.
Treating a nerve test result in a foot that is not complaining. The largest group ever studied described this condition as often causing no symptoms at all, found while investigating something else.
Treating a normal nerve test as an all-clear. Up to 28% of people have an extra nerve branch supplying the test muscle that bypasses the pinch point entirely.
MRI as the first scan. In the only comparative series, MRI missed it in every patient who had one, while ultrasound found it in all four.
Borrowing numbers from the tarsal tunnel on the inside of the ankle. Different nerve, different side of the joint, different condition. Its 75% success figure does not belong here.

Return to Training

Strength lifting the foot is normal and equal on both sides
No numbness or burning between the first two toes at rest for 7 straight days
The problem footwear can be worn for a full session, with symptoms settling within 10 minutes of taking it off
No lingering numbness the morning after a full session
Muscle bulk on the top of the foot equal to the other side, or at minimum not getting worse
Full training load tolerated for 2 straight weeks with no flare

These are clinical-reasoning criteria. No return-to-activity milestone has ever been published for this condition, so nobody can hand you a validated checklist here, and you should be suspicious of anyone who does.

! Red Flags — see a doctor, not a search bar

You are struggling to lift your foot, or it slaps the floor when you walk. This is not this condition. The nerve branches that lift your foot split off much higher up the leg. Something further up needs looking at.
The numbness spreads well past the small patch between your first two toes, or the same thing starts in the other foot.
You can feel a lump on the top of your foot.
The pulse on the top of your foot is weak or absent. An artery runs through the same tunnel as the nerve, and that is a circulation question first.
The muscles on the top of your foot are visibly shrinking, or things are getting steadily worse rather than steadily better.
You badly injured or twisted that ankle at some point and the pain never fully settled.
You have diabetes and any new numbness in the foot. Do not wait on this one.

Refer to: your GP as the default route for all of these. Emergency care is not needed for this condition unless the foot is acutely cold and pulseless, or the injury is fresh.

Cinematic anatomical rendering of the anterior ankle and dorsal foot

Take the shoe off and look at the lace line. If a lace or a stiff tongue crosses the exact spot that hurts, re-lace so it doesn't.

Skip the eyelets sitting over the sore point and lace above and below it instead. Then walk for ten minutes and notice what changes. The single most common cause of this is something pressing on the top of your foot, and it is usually free to fix.

Takes less than 2 minutes. No equipment needed.

Conviction

Low endpoint-stratified, because the parts of this are not equally supported

The anatomy and the clinical picture it forces	Moderate
Nerve testing fails in both directions	Moderate
The pinch is often further forward than the name suggests	Mod-Low
Decompression helps selected structural cases	Mod-Low
Accuracy of any diagnostic test	No evidence
Conservative and rehabilitation management	No evidence

What would change my mind: the diagnostic claims

A prospective study of 300 or more consecutive first-contact patients with dorsal foot pain, testing the dorsal tap, first-web sensation and the small extensor muscle against blinded ultrasound and nerve studies, reporting sensitivity, specificity and likelihood ratios with confidence intervals. Nothing remotely like this exists. It would move the entire assessment section from reasoning to evidence, and it is the single highest-value study anyone could run here.

What would change my mind: the treatment claims

A randomized trial of footwear and lacing modification versus usual care in 80 or more people with ultrasound-confirmed compression, over twelve weeks, with a validated foot and ankle score and first-web sensory testing as the endpoints, and the intervention specified down to the lacing pattern. A positive result would give this condition its first quantified conservative treatment. A null result would be just as useful, because it would move these patients into an imaging pathway sooner.

Foot and ankle nerve problems get mistaken for each other constantly, and the difference decides whether you change your shoes or get a scan. The Verdict sends one evidence-checked protocol a week, free, with the honest confidence level attached.

Join The Verdict — free

The Full Picture — Anatomy, Diagnosis & Evidence

What's Actually Going On

The deep peroneal nerve splits off near the outside of the knee, runs down the front of the shin, and passes under a tight band called the inferior extensor retinaculum at the front of the ankle, sitting over the talus and navicular bones. That fibrous tunnel also carries an artery, a vein and four tendons.

The single most important fact about this condition is where the branches leave. The branches that power the muscles lifting your foot and your big toe all leave the nerve up in the leg, well above the tunnel. Nothing happening under that band can weaken them. What is left downstream is one branch to the small extensor muscles on the top of the foot, and one purely sensory branch to the skin between the first and second toes.

That is why the picture is so distinctive and why it is missed so often: the foot lifts perfectly, and a coin-sized patch of skin has gone numb. Anyone expecting a weak foot will look straight past it.

Two different mechanisms are described and they point at different fixes. Compression is the classic one: a lace, a boot tongue, a bone spur on the top of the midfoot, a small cyst, or swollen joint lining. Stretch is the other: holding the foot in a pointed position tensions the nerve across the front of the ankle (Borges 1981, PMID 7463126). Because those are different loads, "rest the foot" misses both.

Cinematic anatomical rendering of the deep nerve pathway across the front of the ankle

How to Identify It

Four things, in this order.

Test the strength lifting the foot Sn/Sp: none published — this should be normal. Normal strength is the finding, not the absence of one. Weakness sends the whole search up the leg or to the back.
Map the numbness with one finger Sn/Sp: none published — confined to the web between the first and second toes is the condition. Wider than that, or the same on both feet, is a different problem.
Tap at two places, not one Sn/Sp: none published — over the front of the ankle, and then further forward over the middle of the foot. A tingle running into those toes localises it. The forward site is where all four surgically proven cases in the only imaging series actually were (Bianchi 2019, PMID 30215106).
Look at the actual shoes — bring in the pair worn when it is at its worst. This is the highest-yield question in the whole assessment, and it is also the treatment.

Every accuracy figure above says "none published" because that is true. Four literature searches across 82 papers returned no diagnostic accuracy study of any test for this condition. If you see a sensitivity quoted for it somewhere, it has been carried over from the tarsal tunnel on the inside of the ankle, which is a different nerve.

On nerve conduction studies, and this is the part worth knowing: the test fails in both directions at once. In as many as 28% of people, the muscle being tested is supplied partly or wholly by an extra nerve branch that never enters the tunnel, so a normal result rules nothing out (Sinanović 2013, PMID 24399869). And the largest group ever collected, 85 patients over eleven years, described this condition as frequently causing no symptoms at all, turning up as a misleading result during testing for something else entirely (Logullo 2014, PMID 24337947). A test that both misses cases and over-calls them is a contributing piece of information, not a referee.

Cinematic anatomical rendering of clinical examination of the dorsal foot

The Debate

There is no clinical guideline for this condition anywhere, so there is no guideline-versus-trial argument to referee. The genuine disagreements sit between smaller sources, and they matter more than a guideline dispute would.

Is this a disease, or a finding?

Surgical series (Yassin 2015, PMID 26209470; Krause 1977, PMID 75254)

A real, underdiagnosed, symptomatic condition that responds well to decompression.

Largest cohort, N=85 (Logullo 2014, PMID 24337947)

"Often asymptomatic or oligosymptomatic", showing up as a misleading result during testing for suspected nerve disease elsewhere.

Both are true, and they are describing different people. A surgical clinic enrols patients who came in complaining. A neurophysiology lab finds the same electrical lesion in people who came in about something else. Only one of those groups contains a patient who needs treating, and this is the whole reason to treat the person rather than the number.

Where is the pinch, actually?

The name, and the classical description

Under the band at the front of the ankle. That is what "anterior tarsal tunnel" means.

Bianchi 2019 (PMID 30215106), 4 surgically proven cases

All four were compressed further forward, in the middle of the foot, by a small extensor tendon. Ultrasound found it after examination, X-ray and MRI had not.

The name fixes a location the pathology does not respect. A review from 2006 already named two separate release sites for exactly this reason. A clear examination at the ankle does not rule the condition out. It just moves it forward.

Does conservative treatment work?

Reviews (Bojovic 2025, PMID 40042611; DiDomenico 2006, PMID 16958392)

Conservative treatment, orthotics and rehabilitation are "often effective".

The case literature (e.g. Reed 1995, PMID 7497372)

Conservative treatment repeatedly recorded as having failed before surgery.

Neither side is evidence. Nobody writes up the patient who got better by loosening their boots, so the published case record is a register of conservative failures by construction. Treat conservatively first because it is cheap, safe and reversible, not because it has been proven.

Honest Limitations

Every study population is a surgical or nerve-lab one

What the research shows: decompression lifted foot-and-ankle scores from 55 to 88 over two years; prolonged nerve latencies and muscle changes characterise the condition.

The real-world gap: those patients were already booked for surgery, or already in a neurophysiology lab, or already being tested for back pain. Nobody has ever studied the population that walks into a first-contact clinic with pain on the top of the foot, the overwhelming majority of whom have something else. That is a gap in who was studied, not just how much. More research on operated feet will never close it.

What to do with that: treat this as an uncommon answer that has to earn its place. It earns it through the numb patch plus normal strength, never through dorsal foot pain alone.

The published record over-represents the dramatic causes

What the research shows: the documented causes are bone spurs, cysts, swollen joint lining, an enlarged muscle in a ballet dancer, an extra nerve, a blocked artery, an old fracture, and in one case leprosy.

The real-world gap: every one of those got published because it was unusual and needed an operation. The person whose symptoms went away when they changed their boots generates no paper. So the literature reads as though this is usually structural and usually surgical, and no dataset supports that impression.

What to do with that: give the boring fix a genuine trial before escalating, because the boring fix is the one the literature is structurally blind to.

The reviews recommend rehabilitation and no study says what it was

What the research shows: "conservative treatments, such as physical therapy, orthotics, and corticosteroid injections, are often effective" (2025 review), echoing the same sequence from 2006.

The real-world gap: neither statement is attached to an outcome. There is no exercise, no dose, no frequency, no duration and no review interval published for this condition anywhere. This is a hole in what researchers have chosen to study, not a verdict on the treatment.

What to do with that: say so out loud. It changes what anyone should promise about timelines, and it is the difference between a clinician reasoning transparently and one quoting a protocol that does not exist.

The Nuance

Surgery versus conservative care, honestly: the comparison cannot be made for this condition, and pretending otherwise would be the dishonest option. One uncontrolled series of thirteen operated patients is the entire outcome literature. It shows a large improvement that held to two years, and it cannot separate the surgery from natural history, from removing a lesion that was coming out anyway, or from the attention of being in a study. Zero patients have ever been followed under conservative care with a published outcome.

What can be said confidently is narrower and still useful: when a structural cause is found and removed, the patients in that one series did well and stayed well. What cannot be said is how many would have done just as well by loosening their boots, because nobody has looked.

The nuance that changes practice most: a diagnostic test that fails in both directions gets handled as though it failed in only one. Clinicians are careful with a test known to miss things, and sceptical of a test known to over-call. When a test does both at once, it tends to get used exactly as before, because the two failure modes are written up in separate papers and never meet on the same page. That pattern is not specific to this nerve, and it is worth carrying to every borderline result you read.

And the part that is genuinely reassuring: the differentiator here costs nothing and takes seconds. Test the strength, map the numbness with one finger, look at the shoe. The person best placed to catch this is whoever actually examines the foot and asks what the patient wears on it, which is a question the scan cannot ask.

Cinematic anatomical rendering distinguishing nerve territories of the lower limb

Sources

Logullo F, et al. (2014). Anterior tarsal tunnel syndrome: a misunderstood and a misleading entrapment neuropathy. Neurological Sciences. PMID 24337947. N=85, largest cohort in the literature. Establishes the condition is often asymptomatic and misleads testing done for suspected nerve disease elsewhere.
Yassin M, et al. (2015). Treatment of anterior tarsal tunnel syndrome through an endoscopic or open technique. The Foot (Edinburgh). PMID 26209470. N=13, minimum 24-month follow-up. The only quantified treatment outcome that exists for this condition.
Sinanović O, et al. (2013). The accessory deep peroneal nerve and anterior tarsal tunnel syndrome. Acta Myologica. PMID 24399869. The extra nerve branch supplying the test muscle in as many as 28% of people, bypassing the tunnel entirely.
Bianchi S, et al. (2019). Partial anterior tunnel syndrome: ultrasound findings in four surgically proven cases. Skeletal Radiology. PMID 30215106. Compression further forward than the name implies; ultrasound diagnostic where examination, X-ray and MRI were not.
Aktan Ikiz ZA, et al. (2007). Dimensions of the anterior tarsal tunnel and features of the deep peroneal nerve. Surgical and Radiologic Anatomy. PMID 17607504. 18 cadavers, 36 limbs. Nerve branching varies more than the textbook picture suggests.
Krause KH, et al. (1977). The anterior tarsal tunnel syndrome. Journal of Neurology. PMID 75254. The classical description, including the partial forms where only the motor or only the sensory branch is affected.
Borges LF, et al. (1981). The anterior tarsal tunnel syndrome. Report of two cases. Journal of Neurosurgery. PMID 7463126. Proposes abnormal stretch of the nerve alongside compression.
Bojovic M, et al. (2025). Overview of nerve entrapment syndromes in the foot and ankle. International Orthopaedics. PMID 40042611. The most recent review covering this nerve. Conservative first, surgery for stubborn cases, no effect sizes and no dosing.
DiDomenico LA & Masternick EB (2006). Anterior tarsal tunnel syndrome. Clinics in Podiatric Medicine and Surgery. PMID 16958392. Names the conservative sequence and, importantly, the two separate release levels.
Golovchinsky V (1998). Double crush syndrome in lower extremities. Electromyography and Clinical Neurophysiology. PMID 9553749. N=169. Peripheral entrapment in 5.3%, with a concurrent problem higher up in 4.4%.
Huang KC, et al. (1999). Anterior tarsal tunnel syndrome: case report. Chang Gung Medical Journal. PMID 10584426. Bone spurs on the top of the midfoot as the cause; symptom-free with a normal examination one year after decompression.
Ferkel E, et al. (2015). Entrapment Neuropathies of the Foot and Ankle. Clinics in Sports Medicine. PMID 26409596. Places the condition among foot and ankle nerve entrapments; the first web space as the typical radiation.

Every source above resolves to a machine-collected literature record and was rendered from that record rather than retyped. All 21 citations in the underlying research file passed the citation gate at 21/21 matched, zero unverified. Evidence ceiling for this condition: no source rises above small uncontrolled series or expert review, and every one is abstract-level.

If the skin between your first two toes feels dead but your foot lifts fine, look at your shoes.