Can Orthoses Take the Heat Out of a Painful Midfoot? What MRI Reveals | Find Your Stride | Edinburgh Podiatrist
Can foot orthoses reduce midfoot pain?
For clinicians treating painful feet, one of the perennial questions is deceptively simple:
If foot orthoses reduce someone's pain, what are theyactually doing? Are they changing joint motion? Redistributing plantar pressure? Altering internal tissue loading? Or simply providing a different sensory environment that makes the foot feel better?
A 2025 study by Halstead et al., published in Arthritis Care & Research, titled offers an intriguing piece of the puzzle. Rather than looking only at pain and function, the researchers used MRI to investigate whether foot orthoses could also alter bone marrow lesions (BMLs) in people with midfoot pain. The result is interesting, but perhaps more importantly for podiatrists and sports clinicians, it raises some bigger questions about how we understand mechanical loading, bone stress and orthotic therapy.
What did the researchers actually do?
The study recruited people with mechanical midfoot pain lasting more than three months who also had MRI-confirmed BMLs. Importantly, the researchers excluded people with established clinical/radiographic midfoot osteoarthritis, inflammatory arthritis, neurological symptoms, recent foot surgery and several systemic conditions. The intention was to create a group with midfoot pain and MRI evidence of bone abnormality without established radiographic OA. Forty-five people were randomised, with 42 ultimately included:
27 received firm, arch-contouring foot orthoses.
15 received cushioning control insoles.
Participants were followed for 12 weeks.
Pain was assessed at baseline, 6 weeks and 12 weeks.
MRI was performed at baseline and 12 weeks to measure BML volume.
The intervention was not an individually prescribed custom orthosis. Participants received a pre-formed Vectorthotic device, incorporating a polypropylene shell and medial rearfoot wedges of either 4° or 6°, with a variable-depth full-length top cover. The control was a cushioning insole. That distinction matters when we consider how or whether the findings should influence clinical practice.
The headline finding: pain improved
The orthoses group experienced a greater reduction in foot pain. At six weeks, pain fell by approximately 15 mm on a 100-mm visual analogue scale, compared with about 7 mm in the control group. By 12 weeks, the orthoses group had an overall reduction of approximately 18.5 mm from baseline, compared with about 7.8 mm in the control group.
There was also a striking difference in patient perception. At 12 weeks:
86% of the orthoses group reported improvement.
40% of the control group reported improvement.
For a patient sitting in a consultation room wondering whether foot orthoses might make their painful midfoot more tolerable, that is clinically relevant. But there is a wrinkle. The study was small, and several of the secondary functional measures did not demonstrate a clear sustained difference between groups. Both groups improved in functional impairment during the first six weeks, with little further change by 12 weeks. So the most convincing clinical signal is pain and patient-reported improvement, rather than a dramatic restoration of function.
What did the MRI tell us?
This is arguably the most interesting part of the paper. Across the orthoses group, mean BML volume decreased by approximately 1,544 mm³ over 12 weeks. The control group decreased by approximately 316 mm³. Expressed as an overall percentage, the authors report a 25.8% reduction in BML volume in the orthoses group versus 4.4% in the control group.
That sounds impressive, and it certainly is biologically interesting. But this is where clinicians need to resist the temptation to jump from:
"BML volume decreased"
to:
"The orthosis definitively reduced harmful bone stress."
The latter has not been demonstrated conclusively.
BMLs are not simply "bone bruises"
One of the most important aspects of interpreting this study is understanding what a BML actually represents. The MRI appearance described as a bone marrow lesion is not a diagnosis in itself. BMLs can reflect abnormal bone physiology and have been associated with mechanical loading, bone stress and osteoarthritis. They can also occur in asymptomatic individuals and in a variety of pathological contexts. This is particularly relevant to runners and athletes. The paper highlights previous evidence linking foot BMLs with repetitive loading in running, sport, military activity and dance. The authors also cite work showing changes in BMLs following alterations in loading and footwear.
That makes the concept attractive: mechanical loading → bone response → BML → pain.
But biology is rarely that linear. A BML may be part of the pain-generating process, but its presence does not automatically tell us that it is the source of pain. Nor does a reduction in BML necessarily mean that a foot has become mechanically "normal." Indeed, the authors themselves are careful to describe BMLs as a surrogate measure of bone stress rather than a direct measurement of bone loading.
The fascinating contradiction: the orthoses were designed to increase midfoot contact force
This is perhaps the most thought-provoking aspect of the study for podiatrists. The orthoses were selected partly because previous work suggested that they could increase midfoot contact force and maximum midfoot force. Yet BML volume subsequently decreased. At first glance, that seems counterintuitive. If we assume: more force = more bone stress = more BML
then why did the group wearing the firmer orthoses demonstrate less BML? The answer may be that external force, internal tissue stress and pathological loading are not interchangeable concepts.
Foot orthoses can change where forces are transmitted, how they are distributed, and how joints and tissues share load without simply reducing total force through the foot.
The study therefore provides an important reminder: Orthotic therapy is not necessarily about "taking load off the foot." It may be about redistributing load more favourably.
That is a much more sophisticated and clinically useful way of thinking about foot orthoses.
What does this mean for runners?
The findings are potentially relevant to runners, but there is a substantial gap between the study population and the typical sports population. The participants had a mean age of 53 years, with 71% female, a mean BMI of approximately 30, and multiple participants reporting pain in other joints. This is not a study of recreational marathon runners, elite athletes or competitive trail runners. So we should not conclude that runners with midfoot pain and suspected bone stress injuries should automatically be given this type of foot orthoses.
In fact, that could be an inappropriate extrapolation.
A runner with an acute bone stress injury requires a much more comprehensive assessment of training load, recovery, nutrition, biomechanics and risk factors. Foot orthoses may potentially form part of management in selected cases, but this study does not establish orthoses as a treatment for running-related bone stress injuries. What the paper does provide is a fascinating hypothesis:
Changing foot mechanics may influence the biological response of bone.
That deserves further investigation in athletic populations.
The study's biggest strength: looking beyond pain
Orthotic research often gets trapped in the question: "Does it hurt less?" That's important but incomplete. Halstead et al. attempted to look underneath the symptom by combining clinical outcomes with MRI. The imaging was also analysed with researchers blinded to group allocation and scan sequence, reducing the possibility that knowledge of the intervention influenced BML measurements. The authors also investigated measurement repeatability, reporting a 6.2% repeatability error for BML volume measurements. This is a genuine strength. The study therefore moves the conversation from: "Patients say their feet feel better."
towards: "Is there an accompanying biological change?" That is exactly the kind of question we need more of in podiatric biomechanics.
But the limitations are substantial
1. It's a very small study
This is the biggest problem. Only 42 participants entered the study, and by 12 weeks only 22 orthoses participants had completed the follow-up compared with all 15 controls. Five participants in the orthoses group failed to complete the three-month follow-up. The authors explicitly state that no sample-size calculation was performed and describe the investigation as exploratory. That means the findings should be considered hypothesis-generating rather than definitive.
2. The groups weren't perfectly balanced
The 2:1 randomisation deliberately favoured the orthoses group. After randomisation, the groups differed in some characteristics, including age and the number of painful joints.
With a small sample, even apparently modest baseline imbalances can matter.
This makes it more difficult to confidently attribute differences at follow-up exclusively to the orthoses.
3. Participants and assessors weren't blinded
This is another major issue for a pain study. The intervention was a firm orthosis while the control was a cushioning insole. Participants and the clinical assessor were therefore aware of the treatment allocation. The researchers attempted to minimise expectation bias by presenting both devices with equal merit, but they acknowledge that participant beliefs were not fully assessed. This matters because pain is influenced by expectation, context and treatment beliefs. The MRI measurements were blinded, which helps considerably for the structural outcome but the primary clinical outcome was still vulnerable to performance and reporting bias.
4. The MRI result wasn't statistically definitive
This is probably the most important point to emphasise when communicating the study.
The reduction in BML volume was:
Orthoses: −1,544 mm³
95% CI: −3,660 to +572 mm³
P = 0.144
The control group showed:
−316 mm³
95% CI: −1,528 to +897 mm³
P = 0.587
Likewise, the percentage change in BML per bone in the orthoses group had a P value of 0.078. So although the numerical difference is intriguing, the study does not provide strong statistical evidence that the orthoses caused a significant reduction in BML volume. That distinction should not get lost beneath an attractive headline.
There is another fascinating finding: BMLs didn't simply disappear
The study found new BML formation in both groups at 12 weeks. In the orthoses group, some bones demonstrated resolution while other sites developed new BMLs. That is a useful reminder that bone adaptation is dynamic. The foot isn't simply: BML → orthosis → BML disappears. Instead, loading appears to produce a constantly changing pattern of adaptation.
The intermediate cuneiform, for example, showed both increases and decreases in BML volume. For clinicians, this reinforces the danger of viewing MRI findings in isolation. A single scan provides a snapshot; the mechanical environment of the foot is constantly changing.
What should clinicians take from this?
The paper doesn't justify prescribing one particular pair of foot orthoses to every patient with midfoot pain. It does, however, support several useful clinical concepts.
1. Think beyond "support"
Orthoses aren't necessarily passive props that simply hold the foot in a particular position.
They can alter the distribution of mechanical forces through the foot.
2. Pain reduction doesn't prove a mechanical correction
A patient feeling better is clinically important, but it doesn't tell us precisely why they improved.
3. MRI abnormalities need clinical context
BMLs may be relevant to pain and mechanical loading, but they are not synonymous with pain or tissue damage.
4. Load redistribution may be more important than load reduction
This study is particularly interesting because the intervention was associated with greater midfoot contact forces in previous work, yet BMLs decreased. That challenges simplistic ideas about orthoses merely "reducing load."
5. Don't extrapolate too quickly to athletes
The participants were predominantly middle-aged adults with chronic mechanical midfoot pain not an athletic cohort. For runners, the study should be viewed as evidence worth watching, not as a prescription for treating bone stress injuries.
The bigger question: could orthoses influence bone adaptation?
This is where the paper becomes genuinely exciting. We already know that bone responds to mechanical loading. The authors cite research involving running, military training and other high-loading activities showing associations between repetitive stress and BMLs. They also point to experimental work demonstrating that foot orthoses can alter bone strain and modelling work suggesting that the stiffness and arch height of foot orthoses can influence navicular and cuneiform loading.
Put together, these findings suggest an intriguing research pathway: Orthosis design → altered foot mechanics → altered internal bone loading → altered bone adaptation → potential change in symptoms. We are nowhere near being able to draw that complete causal chain from this study alone. But it is a compelling hypothesis, and importantly, it moves orthotic research away from crude questions about whether a particular "arch type" requires "support" and towards a more nuanced understanding of how individual tissues respond to altered mechanical environments.
The bottom line for podiatrists
This is an interesting, innovative and genuinely thought-provoking study but it is not definitive evidence that orthoses reduce bone stress. The strongest clinical message is that firm, arch-contouring orthoses were associated with greater reductions in midfoot pain and greater patient-reported improvement than cushioning insoles over 12 weeks.
The MRI findings add something new: BML volume also decreased more in the orthoses group. But the small sample, unequal groups, lack of blinding, participant drop-out and exploratory statistical design mean that the structural findings should be interpreted cautiously. The authors themselves acknowledge that the study's size limits generalisability and direct comparison between groups.
For podiatrists, perhaps the most valuable takeaway is not:
"Orthoses reduce BMLs." It is: "Changing foot mechanics may change the biological environment of the midfoot and we need much better research to understand exactly how." For runners, that translates into an equally important message: Foot orthoses are a load-management tool, not a magic correction. If midfoot pain is limiting training, the answer is unlikely to be found in an orthotic prescription alone. The diagnosis, training load, footwear, recovery, strength, running mechanics and, where appropriate imaging all need to be considered together.
Find Your Stride!
Citation
Halstead J, Keenan AM, Conaghan PG, McGonagle D, Redmond AC. Effect of Foot Orthoses on Midfoot Pain and the Volume of Bone Marrow Lesions in the Midfoot: A Randomized Mechanism of Action Study. Arthritis Care & Research. 2025. doi:10.1002/acr.25648
Comments