Can a Diabetic Foot With Bone Infection Be Saved? Bone-Preserving Surgery and the Honest Limits of Limb Salvage
The foot we no longer have to lose — and the feet we still do.
An X-ray showing infection in the bone has, for generations, meant the foot comes off. In selected patients that is no longer true. But the decision that matters most is not which operation you have — it is whether your foot is one that can be saved at all, and that question has only two parts.
Before you read on. If a foot wound comes with fever, spreading redness, foul discharge, or blackening tissue, go to an emergency department now. That situation is measured in hours, not appointments.
The same person, the same lighting, thirty seconds apart. Left: front camera, elbow bent (~30 cm). Right: rear camera, 1.5 m.
Nothing about the face changed — only the distance did.
Key takeaways
-
Infected bone is not an automatic amputation. The infected marrow can be drilled out through small openings and the cavity packed with an antibiotic-loaded material, keeping the bone that carries your weight.
-
The two-question test decides everything. Can the infection be cleared? Can blood flow be restored? Both yes, salvage is reasonable. Either no, a timely amputation is the operation that protects your life.
-
Delay kills. In a 2021 cohort of patients advised to have a below-knee amputation, those who declined and continued with dressings and antibiotics were more likely to die during follow-up. Salvage is not a reason to wait.
-
Conservative surgery is not fringe. In a published series of 185 patients, removing only the infected bone and sparing the rest cleared the infection in close to half of cases.
-
The soft-tissue hole can often be rebuilt in place with a scaffold and the body's own healing signals, avoiding a flap. The evidence here is case series, not trials.
-
A saved foot is only saved if it is protected. Around 4 in 10 healed ulcers return within a year without offloading footwear, steady glucose, and a nightly look at both feet.
What is a diabetic foot ulcer, and why does it threaten the limb?
Three things fail together. Nerve damage removes the alarm, so you do not feel the injury. Poor circulation removes the healing, so the wound stalls. High blood sugar weakens the immune response, so bacteria move in.
All three have to be assessed. Fixing one while ignoring another is the commonest reason treatment fails.
Think of a leaking pipe behind a wall. The leak itself is trivial. The damage comes from nobody noticing for six weeks. A diabetic foot has no alarm, which is why a scratch that would be nothing on a healthy foot can reach bone in a fortnight.
When infection gets into the bone, it is called osteomyelitis. That is the point at which most people are told the foot must go.
Clinical Vignette: A foot three operations had already given up on
A 55-year-old man arrived with an acutely infected diabetic foot: the forefoot hot and swollen, toes turning black, pus tracking across the sole. Two hospitals in two other cities had already operated twice, removing toes each time. The gangrene had kept advancing, and he had now been advised a midfoot amputation, cutting the foot back and leaving just enough heel to stand on. It was not bad advice. After a spreading infection and two failed attempts, it was the cautious call.
I offered first to find out whether the foot was salvageable, because that question had not yet been asked. Nothing could be judged until the infection was controlled. In the first operation I removed the gangrenous toe and its metatarsal, cleared all dead and purulent tissue, and opened a long incision along the sole to drain a deep abscess tracking from forefoot to heel, under broad-spectrum antibiotics later narrowed to the cultures. Antibiotic-loaded biocomposite was packed into the soft-tissue defects. Within days the spreading necrosis had stopped and the foot's colour returned.
Only then did imaging answer the real question. An MRI showed infection seeded through the marrow of several bones, including some well away from the original wound. Each was explored, drilled, its infected marrow evacuated and replaced with antibiotic-loaded biocomposite; the open ulcers were later coaxed to heal with collagen scaffolds and platelet-derived growth factors. He healed conservatively at home, and was standing on the foot, bandaged and supported, within weeks of the last operation. A year on there has been no recurrence, and he walks on his own foot in custom diabetic footwear.
The earlier surgeons were not wrong. Removing dead toes was correct, and the midfoot amputation was a defensible plan. But one question had been skipped: could the infection be cleared, on a foot with enough blood supply left to heal?The CT angiogram gave a borderline answer, the arteries fading out near the ankle, circulation partial rather than full. Enough to justify the attempt; not enough to make success a foregone conclusion. Where circulation is gone altogether and cannot be restored, the answer runs the other way, and the amputation he was offered would have been the operation that protected him. Judgement at the edge of what was viable decided this, not the instinct to save.
Published with the patient's consent.
When is amputation the right answer?
Before anything else, the honest case for taking the foot off, because it is stronger than salvage enthusiasts admit.
An amputation is definitive. It removes the infection completely, in one operation, with a predictable recovery.
Salvage is the opposite: several operations, weeks to months, strict non-weight-bearing, and no guarantee at the end. A patient who spends four months on failed salvage and then loses the leg anyway has had the worst of both.
And the mortality data is real. In a 2021 cohort of patients for whom a below-knee amputation had been recommended, those who declined and continued with wound care were more likely to die during follow-up. Sepsis from an uncontrolled foot infection kills.
So the burden of proof sits on salvage, not on amputation. What follows has to clear it.
How can infected bone in the foot be saved?
First, how bone infection is found, because this is where assessments go wrong. Diabetic foot infections are icebergs: a small surface ulcer can hide infected tendon sheaths and bone. There is a simple bedside test. If a sterile probe passed into the ulcer touches bone, bone infection is likely, and imaging and deep tissue samples should follow. Note deep tissue samples. A surface swab grows whatever is living on the surface and routinely misleads the antibiotic choice.
The old reasoning was then that infected bone is dead bone, and dead bone must come out. That is true of bone that has lost its blood supply. It is not true of all infected bone.
The alternative treats bone infection the way we treat infection anywhere else. The infected marrow is drilled out through small openings, a sample goes to the laboratory to name the organism, and the cavity is filled with an antibiotic-loaded calcium-sulphate material: an absorbable paste, mixed with the antibiotic the culture calls for, that releases a high concentration of drug exactly where tablets and drips reach poorly. Over weeks it is absorbed and replaced by the patient's own bone. The weight-bearing architecture stays.
The evidence is real but not settled. Conservative surgery removing only infected bone succeeded in close to half of 185 patients in a published series. Adding a local antibiotic carrier is supported by observational cohorts reporting good infection control and low major-amputation rates. One small randomised trial exists but was stopped early and left underpowered, so it points in the right direction without settling anything.
The counter-signal deserves airing too. One 45-case cohort reported osteomyelitis recurring in around four in ten feet at twelve months, and concluded the material should be used cautiously. Read closely, though, that study describes a different operation from the one above: the beads were placed in the soft tissues near the skin rather than packed into drilled bone, at each surgeon's discretion rather than to a protocol, with no imaging or biopsy standard stated for deciding what counted as cure or recurrence, in a population where nearly half had peripheral arterial disease and a third had kidney disease. Its authors name the absent control group and the pandemic-disrupted follow-up themselves. Series that deliver the antibiotic into bone through drilled tunnels report markedly better results. That gap is the honest headline: outcomes here depend less on the material than on whether the infected bone is actually staged, reached and cleared.
Which is why the imaging matters. Plain X-rays underestimate how far marrow infection has travelled. MRI shows which bones are truly involved, including ones distant from the visible wound, and treating every involved bone rather than only the obvious one is what separates a durable result from a recurrence.
Two things about the sequence, so nothing comes as a shock. All dead and infected tissue has to come out first, and the wound gets bigger before it gets better. Families should be warned of this in advance; a wound that looks worse after surgery is usually a wound that has finally been cleaned properly. And arteries come before reconstruction. Nothing is rebuilt onto a foot that cannot feed it.
How is the wound closed without a flap or amputation?
Clearing the infection leaves a hole. A wound with exposed tendon or bone cannot be stitched shut, and a skin graft will not take over bare bone, which has no blood-supplied bed to feed it.
For large or complex defects the definitive answer remains a free flap: a block of living tissue, often from the thigh, moved to the foot with its own artery and vein, which are then joined to vessels in the leg under a microscope. Those vessels are one to three millimetres across and the stitches are finer than a hair. It works. In experienced centres, flap survival in properly selected diabetic patients exceeds 90%. The costs are equally real: a second wound at the donor site, hours of general anaesthesia, and hourly monitoring of the flap for the first 72 hours, because a vessel that blocks can only be rescued if it is caught fast.
That is the operation the alternative below has to justify replacing. For many diabetic patients the deciding factor is not the foot at all: it is that their heart, kidneys or vessels make hours of general anaesthesia the riskiest part of any plan.
There is now a third option for some of these wounds. A porous scaffold is laid over the exposed structures, acting as a trellis for the body's own cells and new vessels to grow into. The patient's own blood is concentrated for its platelets, the natural store of healing signals, and worked in. A negative-pressure dressing removes fluid and draws new vessels forward. Over planned stages the scaffold dissolves and living tissue replaces it.
The honest limit: this is case reports and small series, including the case above, plus one prospective study in fingertip injuries. That is an early signal, not a success rate. It is for wounds where a flap is not feasible and the alternative is removal. For a wound a simple graft handles, it is needless complexity.
What if the blood supply is the problem?
Some feet fail because the arteries are blocked. Nothing heals until inflow is restored, so the first answer is always to reopen the vessels by bypass or angioplasty. A minority cannot be reopened by either.
For that group, injecting cells into the calf muscles is worth trying before removal. Mostly this means the patient's own cells, drawn and concentrated the same day: platelet-rich fibrin, which costs almost nothing, and where the patient wishes it, a bone marrow concentrate that adds little. Both are low-risk and within reach of most families. India also has an approved off-the-shelf donor-cell product, but its approval covers critical limb ischaemia specifically, not diabetic foot ulcers. It is expensive, and it has not proved to be a game-changer.
The evidence is mixed. Some cohorts report improved salvage and pain relief; several larger randomised trials showed no clear benefit. This belongs at the no-option end, after bypass and angioplasty are excluded. It buys a leg a chance. It is not a substitute for blood flow.
Can surgery prevent recurrent foot ulcers?
For a foot that ulcerates in the same spot repeatedly, there is one more option, and it is contested.
On the textbook understanding it should not work. Diabetic nerve damage is a diffuse axonopathy, a dying-back of fibres along the whole limb, not a single pinch point, and you cannot fix a diffuse disease by opening one tunnel. That objection is sound and it is why the operation never became mainstream.
The counter-argument is narrower. A diabetic nerve is also swollen, and a swollen nerve is more easily trapped where it passes through a tight tunnel, above all the tarsal tunnel at the inner ankle. The claim is not that surgery cures neuropathy, but that a releasable compression can sit on top of the diffuse disease. Releasing the tibial nerve from the ankle down into its plantar branches, muscles left intact, can restore enough protective sensation to stop ulcers forming. The approach is associated with A. Lee Dellon, building on the chronic nerve compression work he and Susan Mackinnon established. The candidates are those with a positive Tinel's sign: a tingle into the foot when the nerve is tapped behind the inner ankle, marking a nerve that is squeezed rather than worn out.
Cohort studies following decompression after a first healed ulcer report recurrence rates among the lowest published, including analyses using the patient's own untreated opposite leg as comparison. There is no randomised trial, so neurology reviews label it unproven. Unproven means not yet settled by a trial, not shown to be useless, and the two are constantly confused. The randomised bar is also applied unevenly: sharp debridement is universal practice and has never been tested in a trial either.
My position follows the data rather than the label. In a carefully chosen patient with recurrent ulcers in the same place, adequate circulation and a positive Tinel's sign, it can help. Offering it to everyone with a numb foot is a mistake. Some surgeons have done exactly that. I do not.
What limb salvage does not prove
The other side, stated plainly:
-
Salvage is not always better than amputation. A well-fitted prosthesis on a healed stump can restore more function than a preserved but painful, deformed, repeatedly ulcerating foot.
-
The strong evidence is observational. Cohorts and case series, not randomised trials, for bone preservation, tissue regeneration and nerve decompression alike. They cannot exclude that surgeons chose the feet most likely to do well.
-
Trying and failing is not free. It can cost months, several operations, and occasionally a higher amputation than the one first offered.
-
Mortality after amputation is mostly about the arteries, not the foot. Critical limb ischaemia carries a five-year mortality of 50 to 60%, largely cardiovascular, because the disease blocking the leg also blocks the coronaries. Keeping a foot does not fix that, and no salvage claim should imply otherwise.
How do you stop the ulcer coming back?
Prevention is unglamorous and it outperforms everything above.
-
Offload the pressure. While an ulcer is healing this means proper offloading, usually a total-contact cast, a specialised boot, or enforced rest, not simply comfortable shoes. Afterwards it means covered shoes with fitted insoles, and where there is deformity or a healed ulcer, custom-moulded footwear with individually made silicone inserts. A rebuilt foot returned to the shoe that caused the first ulcer will produce a second.
-
Look at both feet every night. Ten seconds. Soles to a light, or use a mirror, or ask someone. You cannot feel the injury that matters.
-
Never go barefoot, indoors or out. Test bath water and hot ground with an elbow.
-
Keep the skin dry, especially between the toes.
-
Keep blood sugar steady. It is what makes healing possible at all.
-
Have nails and calluses cut by someone trained, never with bathroom scissors at home.
-
Treat redness, swelling, warmth, discharge or odour as urgent. Not something to watch for a few days.
When is a diabetic foot an emergency?
Same-day assessment if the foot is hot, swollen, foul-smelling or discharging pus; if redness spreads up the leg; if any area turns black; or if you have fever or feel unwell. A rapidly changing diabetic foot is a limb-threatening emergency, and here speed protects you.
The one presentation people misjudge in the other direction: a hot, swollen, red foot with no wound at all. That still needs urgent assessment, but it may be Charcot foot rather than infection, and the treatment is immediate offloading rather than antibiotics alone. Either way it is not a wait-and-see.
In my own practice
I no longer treat infected bone as an automatic path to amputation. I clear the marrow and preserve the bone with an antibiotic-loaded material, rebuild soft tissue in place rather than reaching first for a flap, and use nerve decompression selectively for recurrent ulcers. These are procedures I perform, which is precisely why the limits above are stated as firmly as the possibilities.
What I do not do is treat salvage as automatically correct. Some feet should come off, and saying so late is worse than saying so early.
Thinking about a second opinion?
If amputation has been recommended and there is time, the test is the one from the vignette: ask whether the infection can be cleared, and whether the blood supply can be restored. Ask what the honest downside is if salvage is attempted and fails.
An amputation decided in a single visit, without vascular imaging, without deep tissue cultures, and without a reconstructive opinion, deserves a second look.
Five questions are worth putting to any team:
-
Has my circulation been imaged, and can it be improved?
-
Have deep tissue samples been taken, rather than a surface swab?
-
Is bone infection proven, and what would treating that bone alone look like?
-
Has a reconstructive surgeon assessed the options for covering the wound?
-
What would happen if we staged this and reassessed in two weeks?
Ask also who is on the team. Proper salvage needs at minimum a vascular surgeon, a reconstructive surgeon, a physician or endocrinologist for glucose control, and infectious-disease input for antibiotics. If your care involves only one of these, ask why. Glucose matters here too: teams commonly want an HbA1c below roughly 8% before definitive reconstruction, though urgent infection control cannot always wait for good numbers.
If the answers are no, that is your answer, and a timely amputation is the operation that protects your life. If nobody has asked the questions, they are worth asking.
Anyone promising a guaranteed save is as much a warning sign as anyone rushing to remove. Arrange a consultation if you want the questions answered properly, including the possibility that the answer is amputation.
Frequently asked questions
Can a diabetic foot be saved if the bone is infected?
Often, yes. The infected marrow can be cleared through small openings and the cavity filled with an antibiotic-loaded material, preserving the weight-bearing bone. Whether it is possible in your case turns on two things: whether the infection can be cleared, and whether blood flow can be restored.
Does osteomyelitis in the foot always mean amputation?
No. In a published series of 185 patients, conservative surgery removing only the infected bone succeeded in close to half of cases. Amputation becomes necessary when the infection cannot be controlled or the blood supply cannot be restored, not simply because bone is involved.
Is it dangerous to delay an amputation to try saving the foot?
Yes, if the limb is truly beyond saving. A 2021 cohort found that patients who declined a recommended below-knee amputation were more likely to die during follow-up. Attempting salvage is reasonable only when infection can be cleared and circulation restored, and it should be decided quickly, not drifted into.
How long does limb salvage take?
Weeks to months, in stages. Clearing infection, restoring blood flow and rebuilding tissue happen over several planned operations, followed by strict offloading before full weight-bearing. Amputation is faster. That is a real trade-off, not a detail.
How long should a diabetic foot ulcer take to heal?
When circulation, infection and offloading are all corrected, shallow ulcers commonly heal in six to twelve weeks. Reconstructed wounds follow the timeline of grafts or flaps, several weeks to a few months. The rule worth remembering: an ulcer that has not changed after four weeks of good care is a signal to escalate and reassess, not to carry on with the same dressing.
What is the success rate of limb salvage?
In experienced centres free-flap survival exceeds 90%, and structured multidisciplinary programmes report saving a substantial majority of limbs referred for amputation. Those numbers carry heavy selection: the teams publishing them choose the feet most likely to do well. The figure that matters is yours, and a credible team quotes it after assessing your circulation, infection depth, heart and kidney status, not before.
Is major reconstruction safe if I am elderly?
Age alone rules out very little; physiology decides. Heart, kidney and vascular fitness are assessed formally. Where hours of general anaesthesia is the dominant risk, the staged regenerative route exists precisely to offer reconstruction on gentler terms.
Can stem cells save a diabetic foot?
ometimes, at the margins. For feet failing because the blood supply cannot be surgically reopened, injecting cells into the calf is low-risk and worth trying before removal, most often using the patient's own cells at little cost. The evidence is mixed and several randomised trials showed no clear benefit. It is not a cure and never replaces restoring blood flow.
Can surgery stop diabetic foot ulcers coming back?
In selected patients. Releasing a pinched nerve at the ankle has been followed by strikingly low recurrence rates in cohort studies, in people with a positive Tinel's sign and adequate circulation. There is no randomised trial, so it is formally unproven, which means not yet settled rather than shown to be useless. It is not for everyone with a numb foot.
How do I stop an ulcer coming back after it heals?
Offloading footwear with proper insoles, steady blood sugar, and a nightly look at both feet, for life. Around 4 in 10 healed ulcers return within a year without them. This matters more than any operation described here.
My foot doesn't hurt. Does that mean it is fine?
No, and this is the most dangerous misunderstanding in diabetic foot care. Nerve damage removes pain, so the wounds that matter are the ones you cannot feel. Absence of pain is not evidence of safety. Looking is.
References
-
Aragón-Sánchez FJ, Lázaro-Martínez JL, Hernández-Herrero C, et al. Surgical treatment of limb- and life-threatening infections in the feet of patients with diabetes and at least one palpable pedal pulse. Diabetologia. 2008. (Retrospective surgical series, n = 185; conservative bone-sparing surgery successful in close to half.)
-
Armstrong DG, Boulton AJM, Bus SA. Diabetic foot ulcers and their recurrence. New England Journal of Medicine.2017;376(24):2367–2375. (Narrative review; source of the ~40% one-year recurrence figure.)
-
Antibiotic-loaded calcium-sulphate / calcium-sulphate–hydroxyapatite biocomposite for diabetic foot osteomyelitis — observational evidence.
-
Favourable series: comparative retrospective cohort (n = 133), Foot & Ankle Surgery: Techniques, Reports & Cases, 2025; the CLOSE-UP limb-sparing protocol (n = 97, 2017–2024), major amputation in 4.1%, Foot & Ankle Surgery, 2025; Drampalos E, et al., gentamicin-loaded biocomposite delivered into drilled bone tunnels for calcaneal diabetic foot osteomyelitis, no major amputations; Niazi et al. (n = 70), 90% healing.
-
Cautionary series: Alkayali T, Casciato D, Wynes J, Chua J, Doub JB. Are biodegradable calcium sulfate antibiotic beads effective and safe adjuvants for diabetic foot osteomyelitis? Cureus. 2024;16(1):e52444. doi:10.7759/cureus.52444 — retrospective, n = 45, 42% osteomyelitis recurrence at 12 months and 62% wound drainage; beads placed in soft tissue near the skin rather than into bone, at individual surgeon discretion, with no stated imaging or histological definition of cure or recurrence; no control group; 47% peripheral arterial disease, 36% chronic kidney disease; follow-up disrupted by the COVID-19 pandemic (all limitations stated by the authors).
-
One small randomised trial (BIG D-FOOT, Monami et al.) favoured the approach but was stopped early and underpowered. — Evidence is observational and heterogeneous, and outcomes appear highly protocol-dependent; large randomised trials are still lacking.
-
Saha S. Hybrid regenerative therapy for successful reconstruction of an infected, traumatised diabetic foot wound. Plastic and Reconstructive Surgery – Global Open. 2023;11(8):e5213. doi:10.1097/GOX.0000000000005213. PMID 37593695. (Single case report, n = 1; 62-year-old man, 14 × 10 cm dorsal foot wound after trauma, 18-month follow-up. Author's own; open access. A case report establishes feasibility, not a success rate.)
-
Saha S. Soft-tissue regeneration over an exposed skull using a biomaterial scaffold and platelet-rich plasma. Surgical Neurology International. 2021. (Case report, author's own, establishing the scaffold-plus-platelet method later applied to the foot; subsequently cited as precedent by a Children's Hospital of Philadelphia / University of Pennsylvania group.)
-
Dellon AL. Treatment of symptomatic diabetic neuropathy by surgical decompression of multiple peripheral nerves. Plastic and Reconstructive Surgery. 1992;89(4):689–697. (Foundational technique description.)
-
Nickerson DS, Rader AJ. Nerve decompression after diabetic foot ulceration may protect against recurrence: a 3-year controlled, prospective analysis. Journal of the American Podiatric Medical Association. 2014;104(1):66–70. (Prospective controlled cohort; observational, not randomised.)
-
Chaudhry V, Stevens JC, Kincaid J, et al. Practice Advisory: utility of surgical decompression for treatment of diabetic neuropathy. Neurology. 2006;66:1805–1808. (AAN advisory; rated unproven, Level U. The sceptical position, echoed by the 2008 Cochrane review.)
-
Retrospective cohort of patients recommended below-knee amputation for diabetic foot osteomyelitis. Open Forum Infectious Diseases. 2021. (Deferral of a recommended amputation associated with higher mortality; population already judged non-salvageable.)
-
Reviews of critical limb ischaemia reporting ~50–60% five-year mortality, predominantly cardiovascular. Vascular Health and Risk Management, 2018, and related vascular literature. (Narrative reviews.)
-
Meta-analyses of cell therapy for critical limb ischaemia. (Mixed results; several randomised trials showed no clear benefit.)
-
International Working Group on the Diabetic Foot (IWGDF). Guidelines on the prevention and management of diabetic foot disease. iwgdfguidelines.org. (International consensus guideline; source for offloading and probe-to-bone principles.)
-
Lipsky BA, et al. Infectious Diseases Society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections. (Consensus guideline; deep tissue culture over surface swab.)
This is a patient-education article. The references support the general claims; individual cases differ, and the evidence for several newer techniques is early-stage, as stated in the text.
Author and disclosure: This article is written by Dr Srinjoy Saha, Adjunct Professor at the Apollo Hospital Educational and Research Foundation and Senior Consultant Plastic and Reconstructive Surgeon at Apollo Multispeciality Hospital, Kolkata. He performs diabetic foot limb salvage, bone-preserving surgery for osteomyelitis, regenerative soft-tissue reconstruction and selective nerve decompression — a declared interest directly relevant to this article. He has no commercial relationship with the manufacturer of any material referred to here.
Medical Disclaimer: This article is for education only and does not constitute medical advice, diagnosis, or a treatment recommendation. Individual suitability for any procedure can only be assessed in consultation. If concerns about your appearance are causing significant distress, please speak to a doctor or a mental-health professional.
Published July 2026 · Last reviewed 20 July 2026 by the author.