Buying guide

Best Shoes for Baxter’s Nerve Entrapment

Heel pain from a pinched nerve, not the fascia. The one randomised footwear study found a HARD insole changed how the foot works, not a soft one.

By Tom Biernacki, DPM, FACFASDouble board-certified foot & ankle surgeonUpdated 14 Sep 2026

Spec analysis — not yet tested · Published September 2026

Nothing on this page has been measured by us. Every shoe figure below is either the brand’s own published specification or a laboratory measurement published by RunRepeat, and the source is printed beside each one. No shoe here carries a score or the Podiatrist Tested seal. How we test →

Baxter’s nerve entrapment is heel pain caused by a compressed nerve rather than an inflamed ligament. It is the commonest reason a heel keeps hurting after months of doing everything right for plantar fasciitis — and it is the condition on this site where the published evidence most sharply contradicts what people buy. The only randomised footwear study anyone has run on it found that a hard insole changed how the foot worked. Not a soft one.

Why the heel still hurts after the plantar fasciitis treatment worked

The nerve in question leaves the inside of the ankle, turns sharply under the heel, and passes between two muscles and along the inner edge of the heel bone on its way to the outside of the foot. It is a mixed sensory and motor nerve taking a long, awkward path through a crowded space, which is why it is vulnerable to compression. Entrapment tends to produce pain deep in the heel that people describe as burning or electric rather than sharp, often worse as the day goes on rather than first thing in the morning, and frequently tender slightly further toward the inner edge of the heel than plantar fasciitis is.

It matters because the treatment diverges. Plantar fasciopathy responds to load management, stretching and time. A compressed nerve does not care how well the fascia is doing, and someone who has diligently finished a fasciitis programme and still hurts is one of the classic presentations of this diagnosis. One paper, explaining why it studied the condition at all, reports that Baxter’s entrapment has been described as responsible for up to 20% of foot disorders — a figure worth treating as a claim from the literature rather than a settled fact, and reason enough not to assume every stubborn heel is a fascia problem.

What you noticeMore consistent withWhere to read next
Sharp pain on the first steps out of bed, easing after a few minutesPlantar fasciopathyShoes for plantar fasciitis
Burning or electric pain deep in the heel, worse as the day goes on, sometimes at restA nerve — this pageThis page, and an examination
Pain that is worse the longer you stand, at night, and in both heelsHeel fat pad syndromeShoes for heel fat pad atrophy
Tingling or numbness spreading from the inside of the ankle into the soleTarsal tunnel syndrome, further up the same nerveShoes for tarsal tunnel syndrome
A bone spur on an x-ray, and a clinician telling you that is the causeA common finding that often explains nothingShoes for heel spurs

An orientation table, not a diagnosis — and these conditions coexist more often than they separate. A heel that has not improved in three months is a reason to be re-examined rather than re-shod. See our editorial policy.

What the research actually supports — and where it stops

This is an unusual evidence section, because the most important thing in it is a set of negative findings about how the condition is diagnosed. Anyone reading about Baxter’s neuropathy online will quickly meet the idea that it shows up on MRI as fatty change in a small muscle on the outside of the sole, the abductor digiti minimi. Two recent papers from the same research group went looking for that association properly.

What they found
  • The systematic review could not establish the association at all. Searching to March 2024, it found four studies and 1,052 participants, only one of which it rated ‘good’. Reported prevalence of fatty infiltration on MRI was 4% to 11% in the general population, and similar in people with and without generalised foot pain — roughly 8% against 6%. No study reported the prevalence in people who actually had plantar heel pain or Baxter’s neuropathy. The authors conclude the association remains unknown for lack of robust evidence.
  • Then a cross-sectional study measured it directly, and the result went the wrong way. Fifty people with plantar heel pain were compared with 25 controls matched for age, sex and body mass index, all imaged by MRI. Some degree of fatty infiltration was present in 78% of the heel-pain group and 96% of the group without heel pain. The paper’s title states the finding plainly: plantar heel pain is not associated with fatty infiltration of that muscle.
  • And the anatomy itself is contested. A 2026 paper in Microsurgery is titled “Baxter’s Nerve Exists but Is Usually NOT the First Branch of the Lateral Plantar Nerve” — which is the definition most textbooks and most websites use. We have read the title and not the paper, so we will not summarise its argument; the existence of the argument is the point, before anyone quotes the anatomy at you as settled.

Chen JSC, Abbott M, Landorf KB. Association of Baxter’s Neuropathy and Fatty Infiltration of the Abductor Digiti Minimi Muscle on Magnetic Resonance Imaging: A Systematic Review. J Foot Ankle Res 2025;18(3):e70075 · Chen JSC, Osborne JWA, Menz HB, Abbott M, Munteanu SE, Entwisle T, Connell DA, Landorf KB. Plantar Heel Pain Is Not Associated With Fatty Infiltration of the Abductor Digiti Minimi Muscle on Magnetic Resonance Imaging. J Foot Ankle Res 2026;19(2):e70155. Both open access; full citations below.

None of that says the condition is not real. It says the imaging shortcut people reach for is not doing the work it is credited with, which is why the diagnosis still rests on the story, the examination and — where it is used — a diagnostic nerve block. For a reader the practical consequence is short: an MRI report mentioning fatty change in that muscle is not a diagnosis, and an MRI without it does not rule this out.

The one footwear study, and it says hard

There is exactly one randomised piece of footwear evidence for this diagnosis, and it is small. Eighteen people diagnosed with entrapment of the first branch of the lateral plantar nerve wore insoles of differing hardness while the activity of the abductor hallucis — a muscle on the inner side of the sole, close to the nerve’s path — was recorded with surface electromyography, standing and walking, in a cross-over design.

ConditionWhat happened to abductor hallucis activitySignificance
Standing still, with and without an orthosisNo detectable differencep > 0.05
Walking, insoles of differing hardnessA hard insole reduced peak muscle activityp < 0.01

⚠️ Eighteen people, and the outcome is muscle activity, not pain. It shows a hard insole changes how the foot is working. It does not show that anyone hurt less. Nobody should describe it as proof that firm shoes treat this condition, and that includes us. Full citation below.

Taken for what it is, it still points somewhere useful, because it points against the default. The instinct with heel pain is to buy the softest thing on the wall, and the softest thing on the wall is very soft indeed: the softest midsole we hold a laboratory figure for measures 29.2 on the Asker C scale against a laboratory average of 35.6. If the mechanism here is a nerve compressed in a space that changes shape as the foot loads, then how much the foot is allowed to move under load is the variable — and a surface that lets it move more is not obviously the safer choice.

The hardness figures, and how few there are

Every shoe box tells you how tall the midsole is. None tells you how hard it is. Hardness is measured with a durometer and reported on the Asker C scale, where lower is softer. No brand publishes a durometer figure for any shoe. Eight of the 43 shoes on this site have one, all from the same laboratory, and the spread across them is wide enough to matter.

ShoeMidsole hardnessWhere it sits
Brooks Addiction Walker 248.7 ACFirmest on the site — the end the one study points toward
Brooks Adrenaline GTS 2444.1 ACFirm, and 9.6 points firmer than the generation after it
HOKA Bondi 943.0 ACThe shoe people buy for cushioning, and one of the firmest measured
ASICS Gel-Kayano 3241.3 ACFirmer than the laboratory average
Brooks Ghost 1736.2 ACAlmost exactly the laboratory average of 35.6
Brooks Adrenaline GTS 2534.5 ACSofter than average, in a stability chassis
HOKA Clifton 1034.2 ACSecond-softest, and the tallest of the eight
Brooks Glycerin 2329.2 ACSoftest on the site — the end the one study points away from

RunRepeat laboratory measurements of a retail pair, read September 2026; laboratory average 35.6 Asker C, lower is softer. The other 35 shoes on this site have no hardness figure from anybody. Every figure and its publisher is in the shoe spec database, and the full working — including the finding that stack height and hardness are uncorrelated across these eight — is on our heel fat pad page.

Read the top and bottom rows together and you have the whole awkwardness of this page. The HOKA Bondi 9, at 43.0, is the shoe most often handed to someone who says the words “heel pain” — and it is one of the firmest foams here, which on this diagnosis is arguably a point in its favour and is certainly not what it was sold on. The Glycerin 23, at 29.2, is the shoe that actually delivers what people think they are buying, and it is the one the single study argues against. Neither is a recommendation. Both are a reason to stop choosing by adjective.

The shoes whose published figures fit the brief

Two shoes, and a deliberate refusal to name a third. None has been tested by us, none carries a score, and the brief they are chosen against is one small EMG study — so the drawbacks below are not a formality.

Firmest measured midsole on the site

Brooks Addiction Walker 2 Spec analysis

If the direction the one study points is firm, this is the firm end: 48.7 on the Asker C scale against a laboratory average of 35.6, on the shortest heel stack of the eight at 35.7 mm. It is a motion-control walking shoe built to resist the foot collapsing inward, which is the movement that closes the space the nerve runs through, and it fits to 4E in men’s and 2E in women’s — useful, because a nerve problem is one of the few where a shoe being slightly too tight is a mechanism rather than a nuisance.

The honest caveat: firm is uncomfortable to many people, and comfort is the one footwear variable with a real literature behind it. It is also the heaviest shoe we hold a weight for at 411 g, and its heel counter graded only 3 out of 5, the most flexible of the four we have a grade for. Nothing here has been shown to reduce pain in this condition — the study measured muscle activity in eighteen people.

Laboratory figureValue
Midsole hardness48.7 AC (lab average 35.6) RunRepeat lab
Heel-counter stiffness3 / 5 RunRepeat lab
Heel stack · forefoot stack35.7 mm · 23.5 mm RunRepeat lab
Heel-to-toe drop12.2 mm derived by us from the two lab stacks
Weight411 g Brooks
WidthsM: B/D/2E/4E · W: 2A/B/D/2E Brooks
List price$140 Brooks

Read the full specification analysis →

Firm, but with a stiffer heel and more room

Brooks Adrenaline GTS 24 Spec analysis

At 44.1 Asker C it is the second-firmest foam we hold a figure for, and unlike the Addiction Walker it comes with a 4-out-of-5 heel counter — a firmer cup around the heel, which holds the rearfoot still rather than letting it splay as it loads. Narrow through X-Wide, and considerably lighter at 283 g. For anyone who wants the firm end of the range in a shoe they can walk distances in, this is the more liveable of the two.

The honest caveat: it is a superseded generation. The GTS 25 that replaced it measured 34.5 — 9.6 points softer, which moves it to the other end of this page’s argument. If you buy the current model expecting what the figure above describes, you will not get it. That is a real and recent example of why an adjective on a box is not a specification.

Laboratory figureValue
Midsole hardness44.1 AC (lab average 35.6) RunRepeat lab
Heel-counter stiffness4 / 5 RunRepeat lab
Heel stack · forefoot stack39 mm · 27 mm Brooks
Heel-to-toe drop12 mm stated · 13.5 mm measured Brooks · RunRepeat lab
Toe box, at the big toe73.7 mm RunRepeat lab
Weight283 g Brooks
WidthsNarrow, Medium, Wide, X-Wide Brooks
List price$140 Brooks

Read the full specification analysis →

And the third card we are not writing. The obvious one would be an insole rather than a shoe, because the study tested insoles and not shoes. We are not going to name one, because the study did not report the hardness of the insoles it compared in any unit we could match to a product — it says “hard” and “soft” — and no insole brand publishes a durometer either. Recommending a specific firm insole off the back of it would be inventing a match between a study and a product that nobody has established. The honest version is the instruction: if you are going to try this, try a firmer one than you would have chosen, and notice what happens over two or three weeks.

The one we would not choose for this diagnosis

Whatever is marketed hardest at heel pain, bought on the word “cushioned”, and worn without anything else changing. That is not a swipe at a brand — it is the actual failure pattern. The Brooks Glycerin 23 at 29.2 Asker C is a genuinely excellent shoe and it is the softest thing we hold a figure for, which on this one diagnosis is the end of the scale the only study argues against. And the HOKA Bondi 9, bought by more people for heel pain than anything else on this list, is not even soft: at 43.0 it is one of the firmest measured. Buying it for cushioning is buying it for a property it does not have, whichever diagnosis you have.

The second thing we would avoid is a shoe that is snug across the inside of the heel. A nerve being compressed does not need help. If a shoe presses anywhere along the line between the inside of your ankle bone and the centre of your heel, that is the wrong shoe here regardless of what anything else about it measures.

Watch: Heel Spurs Won’t Heal? Here’s What You’re MissingThe heel that does not settle despite doing everything right is the story this diagnosis usually arrives with — and the spur on the x-ray is usually not the reason.Michigan Foot Doctors · general education, not a diagnosis · editorial policy

Questions people ask

How is this actually diagnosed?

On the history and the examination, and — where a clinician uses one — a diagnostic injection of local anaesthetic around the nerve, on the logic that if numbing it stops the pain, the nerve is involved. What the evidence above says is that the MRI sign widely quoted for it does not carry the weight people give it: a systematic review could not establish the association, and a study of 50 people with heel pain and 25 without found the sign in more of the controls. So a scan is worth having to exclude other things, and is not the thing that confirms this one.

Can I have this and plantar fasciitis at the same time?

Yes, and it is common enough that the two are studied together — the heel-pain literature usually groups them under the heading “plantar heel pain” precisely because separating them clinically is hard. Practically it means a partial response to fasciitis treatment is not evidence against a nerve being involved: it can mean one problem improved and the other did not. It also means the shoe advice pulls in two directions, because the fascia generally likes cushioning and the nerve, on the one study there is, may not.

Should I be stretching?

Stretching is the best-supported thing in the plantar heel pain literature generally — a 2021 guideline analysis put it ahead of orthoses and everything else for first-step pain in the short term. Whether it does anything for a compressed nerve specifically has not been tested. What is worth knowing is that aggressive stretching of a structure a nerve runs through is not automatically neutral, which is a conversation for whoever is examining you rather than a rule a shoe page should hand out. Our plantar fasciitis page carries the guideline evidence in full.

Is a heel cup or a gel pad a good idea?

Nobody has tested one for this. The single relevant study compared insoles by hardness and found the harder one changed muscle activity in walking — which cuts against a soft gel pad rather than for it, though nothing in it measured pain. There is also a volume point that applies to anything you add: a pad takes up room in a shoe, and this is a compression problem. If adding something makes the heel feel tighter, that is a reason to take it out rather than persist.

Why does it hurt more in the evening than the morning?

Because a nerve compressed by load accumulates the load. Plantar fasciopathy classically hurts most on the first steps after rest, when the tissue has been shortened and is suddenly stretched; a compressed nerve more often builds through the day and can still be complaining at rest in the evening. That difference in timing is one of the more useful things you can bring to an appointment, and it costs nothing to notice for a week before you go.

Does surgery work?

Release of the nerve is done, and the literature on it is mostly retrospective case series rather than controlled trials — which is the same problem as everywhere else on this page, one level up in stakes. This is a question for a foot and ankle surgeon who has examined you and knows what has already been tried, and there is nothing a footwear site can usefully add to it. What we can say is that the conservative side has one small randomised study to its name, so anyone presenting either route as settled is ahead of the evidence.

What happens next on this page

Two numbers would move this page from reasoning to measurement. Midsole hardness for every shoe we hold rather than eight, on a durometer of our own to a stated method — it is the specification this diagnosis turns on and the whole industry declines to print it. And the same figure for insoles, which is arguably more useful here, because the one study that exists tested insoles and reported their hardness only as the words “hard” and “soft”. A durometer reading for the twenty most-sold insoles would let anybody match a product to that study for the first time. Both are on the bench list; when they exist this page names products on a number instead of on a direction.

Sources
  1. Pinto-Franco V, Losa Iglesias ME, Casado-Hernández I, Navarro-Flores EM, López-López D, Martínez-Jiménez EM, Pérez-Boal E, Muñoz-Sánchez JL, Becerro-de-Bengoa-Vallejo R. Influence of the Hardness of Insoles on the Abductor Hallucis Muscle in Baxter’s Entrapment: A Cross-Over Randomized Study. Journal of the American Podiatric Medical Association 2026;116(2). doi:10.7547/23-241 (PMID 42042531). Source of the cross-over design, the 18 participants, the absence of a difference standing (p>0.05), the reduction in peak abductor hallucis activity with a hard insole in walking (p<0.01), and the framing that Baxter’s entrapment has been reported as responsible for up to 20% of foot disorders.
  2. Chen JSC, Abbott M, Landorf KB. Association of Baxter’s Neuropathy and Fatty Infiltration of the Abductor Digiti Minimi Muscle on Magnetic Resonance Imaging: A Systematic Review. Journal of Foot and Ankle Research 2025;18(3):e70075. doi:10.1002/jfa2.70075 (PMID 40836398). Open access. Source of the four studies and 1,052 participants, the single ‘good’ quality rating, the 4–11% general-population prevalence, the ~8% versus ~6% comparison, the absence of any study in a plantar-heel-pain or Baxter’s population, and the conclusion that the association remains unknown.
  3. Chen JSC, Osborne JWA, Menz HB, Abbott M, Munteanu SE, Entwisle T, Connell DA, Landorf KB. Plantar Heel Pain Is Not Associated With Fatty Infiltration of the Abductor Digiti Minimi Muscle on Magnetic Resonance Imaging: A Cross-Sectional Observational Study. Journal of Foot and Ankle Research 2026;19(2):e70155. doi:10.1002/jfa2.70155 (PMID 42033738). Open access. Source of the 50 cases and 25 matched controls and the 78% versus 96% fatty-infiltration comparison.
  4. Williams NH, Senatore JR, Dellon AL, Williams EH. Baxter’s Nerve Exists but Is Usually NOT the First Branch of the Lateral Plantar Nerve. Microsurgery 2026;46(3):e70215. doi:10.1002/micr.70215 (PMID 41834215). Cited for the existence of its argument only — we have read the title and bibliographic record, not the paper.
  5. RunRepeat laboratory measurements of midsole hardness on the Asker C scale, stack heights, heel-counter stiffness, toe-box width and weight for the eight shoes named above; read September 2026. Each figure is also cited on that shoe’s own page here.
  6. Brand-published specifications from Brooks and HOKA — stated stack heights and drop, widths, weight and list price; read September 2026. The Addiction Walker 2’s 12.2 mm drop is our own subtraction of the two laboratory stack figures, not a figure Brooks publishes.
  7. Podiatrist Tested shoe spec database — the 43-shoe dataset behind every count on this page, with the publisher of each individual figure.

No shoe on this page has been tested by us, and nothing here is a diagnosis or a treatment recommendation. Persistent heel pain deserves an examination; see our editorial policy.

Related reference pages and guides
Measured, not guessed

Recently reviewed

Individual shoe pages state their evidence tier, published measurements and who should skip them.

HOKA

HOKA Clifton 11

Spec Analysis

HOKA’s product page carries a podiatric seal claim the APMA database does not support for this version, and a women’s stack height that contradicts HOKA’s own drop. Not tested by us.

$155Not scoredRegular, Wide, X-Wide
Skechers

Skechers Hands Free Slip-ins

Spec Analysis

Untested. A heel panel engineered to stay open cannot also be a rigid heel counter, and Skechers publishes no stiffness figure for it — nor a drop, a weight or a width.

$110Not scoredNot published at point of sale
New Balance

New Balance Made in USA 990v6

Spec Analysis

New Balance publishes no weight, drop or stack height for the 990v6, but it does publish B through 6E widths — and that range is why it still matters. Not tested by…

$199.99Not scoredM: B/D/2E/4E/6E

All reviews

Measured, worn, scored — never sponsored

No brand has ever paid for a review, a score, a ranking or the seal. Where we earn a commission the link says so, and the score does not change.