Best Shoes for Accessory Navicular Syndrome: A Podiatrist’s Guide
An accessory navicular hurts two ways: the shoe rubs it, the tendon pulls on it. The usual fix for one makes the other worse — what to buy instead.
On this page
- The contradiction this diagnosis puts you in
- The brief
- What the bone is, and why only one kind of it hurts
- What the pressure data shows
- The shoes on this page
- The shoe we would not choose for this diagnosis — and it is a good shoe
- What the research actually supports — and where it stops
- Questions patients ask
- What happens next on this page
Spec analysis — not yet tested · Published September 2026
Nothing on this page has been measured by us. Every shoe figure below is the manufacturer’s own published number or an independent laboratory’s, attributed in the row it sits in. No shoe here carries a score or the Podiatrist Tested seal, because none has been on our bench. How we test →
Most foot problems are made worse by a shoe. This one is often made by a shoe. An accessory navicular is an extra bone on the inside of your midfoot, sitting where the arch meets the ankle — and the 2024 meta-analysis that pooled 39 studies on it lists, among the things that turn a silent bone into a painful one, “friction against the footwear.” That single phrase is why this page exists, and it is why the advice here is different from every other guide on this site.
The contradiction this diagnosis puts you in
Two different things make an accessory navicular hurt, and the shoe features that fix one of them make the other worse.
The first is pressure from outside. The bone sticks out on the inside of your foot. Anything firm pressed against it — a stiff medial overlay, a seam, a supportive counter wrapped around the arch side of the shoe, a ski boot, a skate, a cleat — loads the tender spot directly. The fix is a soft, smooth, roomy inside edge.
The second is pull from inside. In the type that actually hurts, the posterior tibial tendon — the main muscle holding your arch up — inserts straight onto the extra bone rather than onto the navicular behind it. Every step the arch flattens, the tendon pulls, and the pull crosses a 1–3 mm strip of cartilage joining the two bones. The fix for that is arch support, which on most shoes means firmer material along the inside edge.
So the standard advice for a collapsing arch — buy a stability shoe — can be exactly the wrong advice here, because most stability shoes deliver their support as firmer foam wrapped up the medial side. Our own review of the HOKA Arahi 8 makes the distinction plainly: its firmer foam is “wrapped around the heel and medial side rather than wedged under the arch”, which is the better-tolerated design for a sore arch. For a sore bone on that same medial side, the same design is the problem.
The resolution is the whole point of this page: support the arch from underneath, not from the side. That means a shoe with a deep, removable footbed you can replace with an orthosis, a soft and seam-free inside edge, and enough width that nothing presses on the prominence. Support comes from the insole. The shoe’s job is to stay out of the way and hold the insole.
The brief
- Nothing firm over the bump. Run a finger along the inside of the shoe where the arch meets the ankle. If you can feel a seam, a plastic overlay or a hard counter there, that is the spot that will hurt after an hour.
- A footbed that lifts out whole. The only intervention with a randomised trial behind it in this diagnosis is a custom orthosis. It has to fit in the shoe, and a glued-in sockliner means it will not.
- Width, then width again. Sizing up in length gives you a longer shoe, not a wider one. The prominence sits at the widest part of the midfoot, so a wide or extra-wide last does more here than half a size up.
- Support from below, not beside. A shoe that holds the arch with a deep, well-shaped footbed is doing the same job as a medial post without touching the sore bone.
- Lacing you can skip. Being able to miss out one eyelet over the prominence is a free fix, and it is the reason a five- or six-eyelet lacing pattern beats a slip-on for a bad flare.
What the bone is, and why only one kind of it hurts
An accessory navicular is a normal developmental variant, not a disease and not an injury. It is one of the commonest extra bones in the human foot. A 2024 meta-analysis by Stolarz and colleagues pooled 39 studies, 11,015 patients and 36,837 feet and put the prevalence at 17.5% of people (95% CI 11.5–25.7) and 12.6% of feet. Where one exists it is usually on both sides — bilateral in about half of patients.
Two findings from that paper are worth printing because they contradict what gets repeated online. The first: it is not more common in women. Pooled across the studies that reported gender, it was present in 21.1% of males and 22.0% of females, with no significant difference. The second: prevalence depends heavily on ancestry — 38.4% in East Asian populations against 8.0% in North American ones — so “how common is it” has no single answer.
Which type you have decides almost everything about how it behaves. The classification in use is Coughlin’s, refining Geist’s original 1914 description:
| Type | What it is | How it usually behaves |
|---|---|---|
| Type I (os tibiale externum) | A small round or oval fragment sitting inside the fibres of the posterior tibial tendon, before the tendon reaches the navicular | Mostly silent |
| Type II (subtypes IIA and IIB) | A larger triangular or heart-shaped bone, typically 8–12 mm, joined to the navicular by a 1–3 mm strip of cartilage. The posterior tibial tendon inserts directly onto it | This is the painful one |
| Type III (cornuate navicular) | A prominent but fully fused navicular tuberosity — the accessory bone has bridged into the main bone | Stable; mostly silent, but still prominent |
Classification and measurements: Stolarz et al. 2024 (source 1). The review reports that types IIA and IIB account for more than 70% of symptomatic cases, while types I and III are primarily asymptomatic — though it flags that only 435 of its 11,015 patients were analysed in full for subtype, so treat that split as a strong signal rather than a precise figure.
That cartilage strip is the mechanism. It is not a joint built to take tension, and the posterior tibial tendon pulls across it on every step. That is why the pain is worse with activity, worse on hard ground, worse in a shoe that lets the arch drop — and why a type III, fused solid, can be just as prominent under the skin and still never hurt from the inside. If yours is a type III, your problem is entirely the first one on this page: pressure from outside. The shoe is the whole treatment.
What the pressure data shows
A 2023 study of 90 military recruits — 30 with normal feet, 30 with flat feet and 30 with a painful accessory navicular — measured where the load actually goes. Flat feet, as expected, shifted pressure to the inside: more under the medial forefoot, medial heel and medial arch, less under the outside of the foot. The feet with a painful accessory navicular were not simply flat feet. They pushed further in the same direction, and the difference was stark: medial arch pressure higher again at P < 0.001, lateral arch pressure lower again at P < 0.001. The authors describe the contact as faster and more impulsive over the same area.
Read practically, that says the problem is concentrated load on one strip of the foot. It is the argument for spreading load laterally with a footbed rather than for cushioning the whole shoe and hoping. It is also a small, single-site study of young male recruits, so treat it as a mechanism, not a measurement of you.
The shoes on this page
These are chosen on three properties and nothing else: how many widths the brand actually makes, whether the footbed lifts out so an orthosis can go in, and what sits against the inside of your midfoot. None has been on our bench. Every figure below is the manufacturer’s published number unless the row says otherwise.
| Shoe | Widths the brand publishes | Drop | Toe box | Price |
|---|---|---|---|---|
| Brooks Ghost 18 | Narrow, Normal, Wide, X-Wide | 10 mm | 70.6 mm | $150 |
| New Balance 1080v15 | M: D / 2E / 4E · W: B / D / 2E | 6 mm | 73.2 mm | $169.99 |
| Orthofeet Coral | W: A / B / D / 2E | 15.7 mm | — | $119 |
| HOKA Bondi 9 | Narrow, Medium, Wide, X-Wide | 5 mm | 72.5 mm | $170 |
Widths, drops and prices are the manufacturers’ own published figures. Toe-box width is not — it is an independent laboratory’s caliper measurement taken 28.3 mm back from the tip of the shoe, which is roughly where the big toe ends. It is here because it is the only measured width figure that exists for any of these shoes; nobody publishes a midfoot width, which is the number this diagnosis would actually want. That gap is real and we are not going to paper over it.
Four widths · neutral · the default
Brooks Ghost 18 Spec analysis
The reason to start here is that it does nothing clever on the inside edge. It is a neutral trainer with no medial post, an engineered mesh upper, a footbed that lifts out cleanly, and — unusually in 2026 — four widths from Narrow to X-Wide. Brooks publishes 36 mm at the heel and 26 mm at the forefoot, 289 g, around $150. If you are going to put an orthosis under the arch, this is the shoe that will take it and then leave the bone alone.
Worth knowing before you buy it for a tight Achilles as well: Brooks prints a 10 mm drop, and an independent laboratory that cut a pair in half measured 12.2 mm. Both figures are attributed in our spec review.
Skip it if: the inside of your midfoot is already too tight in a Normal width and the brand’s Wide is not enough — New Balance goes wider.
Men’s listing above. Women’s listing on Amazon · Read our spec review
Widest measured forefoot on file · softest upper
New Balance Fresh Foam X 1080v15 Spec analysis
Where the Ghost gives you a clean inside edge, this gives you room. At 73.2 mm it is the widest big-toe measurement of the neutral shoes we hold, it goes to 4E in men’s and 2E in women’s, and the knit upper is the least structured thing on this list — which is the point when the complaint is a shoe rubbing a bony lump. New Balance publishes 40/34 mm, a 6 mm drop and 261 g at $169.99.
One correction worth carrying: the v15 is no longer a Fresh Foam X shoe in anything but name — New Balance moved it to a new midsole called Infinion, so reviews of the v13 and v14 describe a different shoe.
Skip it if: you need the arch held firmly and you are not going to use an orthosis. This is a soft, unstructured shoe and it will not do that job by itself.
Built for a prominence · extra depth
Orthofeet Coral Spec analysis
This is the only shoe on the page designed from the outset around the two things this diagnosis needs. Orthofeet publishes a quarter-inch removable insole and extra depth, which means an orthosis goes in without stealing the volume your foot needs, and the interior is built to avoid seams — the single feature most likely to sit on the bump. Four women’s widths, A through 2E, at $119, which is the cheapest thing here.
The number nobody mentions: its 15.7 mm drop is the highest we hold by a distance. A high drop takes tension off the back of the leg, which some people find helps and some find changes their gait more than they want. Try before you commit to it as a daily shoe.
Skip it if: you are running. This is a walking and standing shoe.
For a type III · no medial structure at all
HOKA Bondi 9 Spec analysis
If your bone is fused — a type III, or a type II that stopped hurting from the inside years ago — then you do not need arch control at all. You need a shoe that does not touch the prominence, and the Bondi is about as close to structurally absent along the medial side as a cushioned shoe gets, in four widths from Narrow to X-Wide. HOKA publishes 43/38 mm, a 5 mm drop and 297 g at $170.
Know what you are buying: an independent laboratory that cut a pair in half measured the drop at 9.1 mm, not 5, with the forefoot stack nearly 6 mm under the published figure. It is a taller, more sloped shoe than the spec sheet says.
Skip it if: your arch is actively collapsing. A tall, soft platform gives a pronating foot more to roll through, not less.
Men’s listing above. Women’s listing on Amazon · Read our spec review
The shoe we would not choose for this diagnosis — and it is a good shoe
If you search for shoes for a collapsing arch you will be sent to a stability shoe, and the HOKA Arahi 8 is one of the better ones — it even carries the APMA Seal of Acceptance. Our own review describes how it works: the firmer foam is “wrapped around the heel and medial side rather than wedged under the arch”, and notes that in clinic this is usually the better-tolerated design for a tender arch.
Read that sentence again with a painful bone in mind. Firmer material wrapped around the medial side is, for this one diagnosis, a description of the problem rather than the solution. The same is true of any shoe sold on a medial post, a firm medial cage, or a supportive leather upper moulded along the inside edge. None of them is a bad shoe. They are the wrong tool for a foot whose pain is a lump on the inside.
If you need both — real arch control and nothing against the bone — the answer is a neutral shoe with the depth to take a proper orthosis, which is the arrangement the trial in the next section actually tested. It is also why our flat-feet guide and our overpronation guide give different answers from this page, and they are right to.
What the research actually supports — and where it stops
Everything above this heading is clinical reasoning applied to published specifications. This is the part with trials behind it, and the part that is still judgement.
There is exactly one randomised trial of a footwear intervention in this diagnosis. Pu and colleagues, publishing in BMC Musculoskeletal Disorders in 2025, randomised 54 professional athletes with diagnosed accessory navicular syndrome to customised orthoses or conventional insoles and followed them for 12 months, measuring foot shape, plantar pressure and pain at three, six and twelve months.
- The prominence itself moved. Navicular prominence distance fell significantly in the custom-orthosis group against controls, as did heel eversion angle — the two measurements that describe how far the midfoot is collapsing inward.
- Load came off the inside of the foot. Pressure and force–time integral dropped at the first metatarsal head, the medial arch and the medial heel, while loading on the lateral arch went up. That is the pressure pattern from the recruit study, run in reverse.
- The arch held better. Arch angle and arch height were higher in the orthosis group at every assessment point.
- Pain fell. VAS pain scores dropped significantly against the control group, and the authors describe the relief as both faster and larger than with conventional insoles.
Now the limits, in the authors’ own terms. They name two: their orthoses were not fully customised to individual foot shape, and the design did not target the different accessory navicular subtypes — which matters, because a type II and a type III are different problems. They also state plainly that the sample size was not calculated for this study; it was taken from earlier similar trials. And they call for follow-up at three to five years before anyone claims durable benefit. Add the one they do not name: these were professional athletes. Nothing here establishes what an orthosis does for a twelve-year-old with a sore foot after netball, which is the commonest version of this patient.
What happens if you just treat it without surgery
This is the question patients ask and the one almost nothing online answers with a number. A 2019 review at Cincinnati Children’s followed 169 children with 226 symptomatic accessory naviculae treated non-operatively over a decade. Average age at diagnosis was 11.8 years. Type II accounted for 72.7% of them. Patients averaged 2.1 separate non-operative attempts.
The outcomes: 28% got complete pain relief. 41% got partial relief and never needed an operation. 30% went to surgery. Among those who became pain-free, it took an average of eight months. So roughly seven in ten avoid the knife, most people improve, and improvement is slow and often incomplete. That is a more honest expectation than either “just change your shoes” or “you will need surgery”. It is a retrospective chart review from a single hospital — the authors grade it Level III — so it describes what happened, not what a treatment causes.
If it does come to surgery, a 2022 systematic review compared the two operations used — simple excision and the Kidner procedure — and found both effective with low complication rates and high satisfaction. The finding that bears on this page: patients who also had a flat, valgus hindfoot were the ones who came back with recurrent pain and needed revision. The underlying foot shape does not go away when the bone does, which is the same reason a shoe alone rarely finishes the job.
Questions patients ask
Will the bump ever go away?
No. It is a bone you were born with, and about one person in six has one. The goal is not to remove the prominence but to stop it being loaded — from outside by the shoe, and from inside by the tendon. Pain going away while the lump stays exactly where it is is the normal, successful outcome.
Why does it hurt so much more in ski boots, skates or football boots?
Because all three are rigid shells clamped tight across exactly the part of the foot the bone sticks out of, and none of them lets you skip an eyelet. The 2024 meta-analysis lists friction against footwear among the things that make a silent accessory navicular symptomatic, and stiff sports footwear is the extreme case. Punching or heat-moulding the shell over the prominence is a standard fix that a good boot fitter can do.
Custom orthotics, or will an off-the-shelf insole do?
For this diagnosis specifically, the only randomised comparison that exists put custom orthoses ahead of conventional insoles on pain and on every structural measure. That is a real result and it is unusual — across the wider footwear literature, reviews repeatedly fail to find custom beating a well-chosen prefabricated device. Our insoles hub sets out where that general finding holds. Here the evidence points the other way, in athletes, over twelve months, in one trial. Start with a well-fitted prefabricated device if cost matters, and know that the trial evidence for going custom is better in this condition than in most.
My daughter has this and she is twelve. Does she need to stop sport?
That is a question for the clinician who has seen the foot, and it usually depends on whether the pain settles between sessions or accumulates across them. What the data can tell you is the shape of the road: in the largest non-operative series, most children improved without surgery, it took around eight months on average to become pain-free, and it commonly took more than one attempt at treatment before something worked. If your child also has a flat foot that rolls in, treating the flat foot is treating the bone — see our flat-feet guide and our guide to children’s shoes.
Is it more common in women?
The bone is not. The 2024 meta-analysis found it in 21.1% of males and 22.0% of females, with no significant difference — which contradicts a claim repeated widely, including in older papers. What does skew female is who ends up being treated: 78% of the children in the Cincinnati series were girls. Those are two different questions, and conflating them is how the myth survives.
What happens next on this page
The measurement this diagnosis needs does not exist anywhere. Every shoe here publishes a forefoot width or a lab-measured toe-box figure; nobody publishes the internal width or the medial-seam position at the midfoot, which is the dimension that decides whether a shoe rubs an accessory navicular. When shoes reach our bench, that is the figure we will take — internal width at the navicular, and where the first seam or overlay sits — and this page gains a column no competitor can fill. Until then every number here belongs to someone else, and it says so in the row it sits in.
- Stolarz K, Osiowski A, Preinl M, Osiowski M, Jasiewicz B, Taterra D. The prevalence and anatomy of accessory navicular bone: a meta-analysis. Surgical and Radiologic Anatomy 2024;46(10):1731–1743. doi:10.1007/s00276-024-03459-x (PMID 39136746). 39 studies, 11,015 patients, 36,837 feet.
- Pu X, Xing W, Cheng Z, Wang Y, Wang Y, Zhang Y, Jiang L, Liu B. Effects of customized orthoses on foot morphology and pressure in patients with accessory navicular syndrome. BMC Musculoskeletal Disorders 2025;26(1):491. doi:10.1186/s12891-025-08689-7 (PMID 40382595). Randomised, 54 professional athletes, 12 months. Chinese Clinical Trial Registry ChiCTR2500100238, registered retrospectively. Open access.
- Wynn M, Brady C, Cola K, Rice-Denning J. Effectiveness of Nonoperative Treatment of the Symptomatic Accessory Navicular in Pediatric Patients. The Iowa Orthopaedic Journal 2019;39(1):45–49. PMID 31413673, PMC6604528. 169 patients, 226 feet; Level of Evidence III.
- Wariach S, Karim K, Sarraj M, Gaber K, Singh A, Kishta W. Assessing the Outcomes Associated with Accessory Navicular Bone Surgery — a Systematic Review. Current Reviews in Musculoskeletal Medicine 2022;15(5):377–384. doi:10.1007/s12178-022-09772-5 (PMID 35776339).
- Pu X, Xing L, Jiang L, Liu B, Wang Y, Zhang Y, Wang Y, Kang Q. Plantar Pressure Characteristics and Prevention of Painful Accessory Navicular in Military Recruits. Journal of Musculoskeletal and Neuronal Interactions 2023;23(4). PMID 38037362, PMC10696376. 90 recruits, 30 per group.
- Brooks Running, New Balance, Orthofeet and HOKA product pages — stated stack, drop, weight, widths, insole removability and list price, read September 2026. Toe-box widths are RunRepeat laboratory caliper measurements, taken 28.3 mm from the tip of the shoe; each is also cited on that shoe’s own review page here.
- Best shoes for flat feet — the foot shape that loads the posterior tibial tendon, and the one that decides whether this bone stays quiet.
- Best shoes for overpronation — why that page recommends medial support and this one does not.
- Best shoes for Sever’s disease — the other growing-foot diagnosis that hurts at a growth centre and resolves slowly.
- Insoles — what an insole is actually proven to do, condition by condition, with the trials that found nothing.
- Best insoles for flat feet — the devices that do the arch-support job this page hands to the insole rather than the shoe.
- Shoe spec database — every published figure on this site in one table, with who published each one.
Recently reviewed
Individual shoe pages state their evidence tier, published measurements and who should skip them.
HOKA Clifton 11
Spec AnalysisHOKA’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.
Skechers Hands Free Slip-ins
Spec AnalysisUntested. 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.
New Balance Made in USA 990v6
Spec AnalysisNew 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…
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.