The Nerve Between Your Toes: Inject It or Remove It?

The two options are not two versions of the same thing

One is a test you can undo. The other is a change you cannot. A corticosteroid injection places steroid into the web space around the nerve; if it does nothing, nothing about the foot has been permanently altered. Surgery for an interdigital neuroma comes in two broad shapes, and they are not interchangeable.

One removes the nerve. The other releases the ligament crossing above it and leaves the nerve in place. That distinction matters more than most consent conversations make it sound: removing a sensory nerve removes what the nerve did, and those two toes stop reporting sensation, permanently. It is not a complication. It is the mechanism — the nerve carrying the pain is the nerve carrying everything else.

Before either option, Morton's neuroma is a diagnosis — and the ball of the foot has more than one thing that hurts in that spot. Neuroma vs metatarsalgia is a real fork, and a needle or a blade aimed at the wrong structure tests nothing.

Neurectomy — removing the nerve itself. Decompression — releasing the tissue over it and leaving the nerve intact.

Where the evidence on this page comes from

Not from a foot. Every trial cited below was run somewhere else in the body — an elbow, a knee, a shoulder, a back — and this page names each one's home condition where it uses it, rather than quietly letting a tennis-elbow result stand in for a toe. That is a limitation worth stating plainly rather than papering over.

What transfers is not the number but the structure of the question. The sections below lay out three parts of it, each anchored to a trial in the condition where it was run: what an injection's early relief does and does not predict, what waiting costs, and how a tie gets broken. The same fork sits under trigger finger injection or surgery and under de quervain's treatment.

None of that tells you what will happen to your foot. It tells you which questions to put to the person who has examined it, and what a good answer sounds like.

What a steroid injection actually tells you

Less than the relief suggests. The cleanest measurement of this comes from sciatica, where epidural corticosteroid injections were pooled across trials: they produced small, short-term relief of leg pain and disability — and no meaningful long-term benefit, and no reduction in the number of people who went on to have surgery 1Ref 1Pinto RZ, Maher CG, Ferreira ML, et al. (2012).Epidural Corticosteroid Injections in the Management of Sciatica: A Systematic Review and Meta-analysis.Epidural corticosteroid injections give small, short-term relief of leg pain and disability in sciatica but no meaningful long-term benefit and no reduction in subsequent surgery — cited here, and labelled as sciatica, for the principle that an injection's early relief does not predict the longer course..

That last clause is the one to sit with. The injection worked, in the narrow sense: pain went down for a while. And the people who received it were no less likely to end up in an operating theatre than those who did not 1Ref 1Pinto RZ, Maher CG, Ferreira ML, et al. (2012).Epidural Corticosteroid Injections in the Management of Sciatica: A Systematic Review and Meta-analysis.Epidural corticosteroid injections give small, short-term relief of leg pain and disability in sciatica but no meaningful long-term benefit and no reduction in subsequent surgery — cited here, and labelled as sciatica, for the principle that an injection's early relief does not predict the longer course.. Relief and trajectory came apart.

This is the trap inside the phrase the injection wore off. It sounds like proof the injection was working and simply ran out — that another would extend the good stretch. The sciatica data supports a colder reading: short-term effect and long-term course behaved like separate things, and buying the first did not purchase the second.

An injection that helps for six weeks has told you about six weeks. It has not told you about the year.

An injection is not a neutral act

The default assumption — that a steroid injection either helps or does nothing, so it costs only the needle — has been tested, and it did not hold. In tennis elbow, corticosteroid injection produced worse outcomes at one year and higher recurrence than a placebo injection. Physiotherapy added no significant benefit at one year either 2Ref 2Coombes BK, Bisset L, Brooks P, Khan A, Vicenzino B (2013).Effect of Corticosteroid Injection, Physiotherapy, or Both on Clinical Outcomes in Patients With Unilateral Lateral Epicondylalgia: A Randomized Controlled Trial.For lateral epicondylalgia (tennis elbow), corticosteroid injection produced worse one-year outcomes and higher recurrence than placebo injection, and physiotherapy gave no significant added benefit at one year — cited here, and labelled as tennis elbow, to retire the assumption that a steroid injection carries no downside..

Worse than placebo. Not worse than surgery, not worse than waiting — worse than being injected with something inert 2Ref 2Coombes BK, Bisset L, Brooks P, Khan A, Vicenzino B (2013).Effect of Corticosteroid Injection, Physiotherapy, or Both on Clinical Outcomes in Patients With Unilateral Lateral Epicondylalgia: A Randomized Controlled Trial.For lateral epicondylalgia (tennis elbow), corticosteroid injection produced worse one-year outcomes and higher recurrence than placebo injection, and physiotherapy gave no significant added benefit at one year — cited here, and labelled as tennis elbow, to retire the assumption that a steroid injection carries no downside.. Whatever made the elbow feel better early did something to the twelve-month picture that the sham did not.

That result concerns a tendon at the elbow, and nothing here claims it generalises to a nerve in the foot. Its use is narrower: it retires it is only an injection, there is no downside to trying as a safe default. That is an assumption, and in at least one condition where somebody checked, it was wrong. Whether a steroid injection suits a particular foot belongs to the clinician holding the needle — and it is a question, not a formality.

Does waiting cost you the operation later?

Where it has been tested, the delay did not spoil the later operation — and this is the finding that makes a conservative-first sequence rational rather than merely cautious. In young active adults with acute ACL tears, structured rehabilitation with the option of delayed reconstruction was not inferior to rehabilitation plus early reconstruction. About half the rehabilitation-first group avoided surgery altogether, without worse outcomes at two years 3Ref 3Frobell RB, Roos EM, Roos HP, Ranstam J, Lohmander LS (KANON) (2010).A Randomized Trial of Treatment for Acute Anterior Cruciate Ligament Tears.In young active adults with acute ACL tears, rehabilitation with optional delayed reconstruction was not inferior to rehabilitation plus early reconstruction, and about half the rehabilitation-first group avoided surgery without worse two-year outcomes — cited here, and labelled as ACL, for the principle that waiting did not spoil the later operation..

Sciatica showed the other half. Early surgery produced faster relief of leg pain than a prolonged conservative course — genuinely faster, which matters to anyone counting the months — though by twelve months the two strategies had converged 4Ref 4Peul WC, van Houwelingen HC, van den Hout WB, et al. (2007).Surgery versus Prolonged Conservative Treatment for Sciatica.For sciatica from lumbar disc herniation, early surgery gave faster relief of leg pain than prolonged conservative care but one-year outcomes were similar between strategies — cited here, and labelled as sciatica, for the principle that an operation often buys speed rather than a different endpoint..

Put them together: the operation bought time, and the wait did not cost the operation. That is the whole argument for trying the reversible thing first. Someone who rehabilitates, waits, and then has surgery anyway has not wasted a year — they arrive knowing the alternative was genuinely tried, which is the one thing nobody could have told them in advance.

When the options tie, the tiebreaker is harm and cost

Frozen shoulder produced the cleanest example of a three-way tie, and of what to do with one. The trial ran three arms: physiotherapy started early, manipulation under anaesthesia, and arthroscopic capsular release. Twelve months on, the three had not separated on patient-reported outcomes 5Ref 5Rangan A, Brealey SD, Keding A, et al. (UK FROST) (2020).Management of adults with primary frozen shoulder in secondary care (UK FROST): a multicentre, pragmatic, three-arm, superiority randomised clinical trial.In primary frozen shoulder, physiotherapy, manipulation under anaesthesia, and arthroscopic capsular release gave broadly similar twelve-month patient-reported outcomes; arthroscopic release carried more complications and manipulation was most cost-effective — cited here, and labelled as frozen shoulder, for the principle that tied options are broken on harm and cost..

So the outcome did not choose. Something else had to, and the trial said what: the arthroscopic arm accumulated more complications than either alternative, and manipulation came out most cost-effective 5Ref 5Rangan A, Brealey SD, Keding A, et al. (UK FROST) (2020).Management of adults with primary frozen shoulder in secondary care (UK FROST): a multicentre, pragmatic, three-arm, superiority randomised clinical trial.In primary frozen shoulder, physiotherapy, manipulation under anaesthesia, and arthroscopic capsular release gave broadly similar twelve-month patient-reported outcomes; arthroscopic release carried more complications and manipulation was most cost-effective — cited here, and labelled as frozen shoulder, for the principle that tied options are broken on harm and cost..

That is the rule worth carrying into a foot. When two options land in the same place, the one that gets there with fewer ways to go wrong wins — which tends to favour the reversible one. A frozen shoulder treatment decision made on enthusiasm rather than harm profile is how people acquire the complications of an intervention that was never going to outperform the alternative.

When surgery is clearly the right call

None of the above argues against operating, and reading it that way would be a mistake with real costs. The orthopaedic guideline shape for a compressed nerve is explicit about both halves: for carpal tunnel syndrome, the AAOS guideline covers splinting and corticosteroid injection as nonsurgical management — and surgical release for appropriate patients 6Ref 6American Academy of Orthopaedic Surgeons (AAOS) (2016).Management of Carpal Tunnel Syndrome — Evidence-Based Clinical Practice Guideline.The AAOS guideline for carpal tunnel syndrome covers splinting and corticosteroid injection as nonsurgical management and surgical release for appropriate patients — cited here, and labelled as carpal tunnel, as the orthopaedic-society shape of a conservative-first sequence that still ends in surgery for the right person..

Appropriate patients is doing a great deal of work there, and no guideline hands a reader a checklist for their own foot. What it means across these conservative-first arguments is someone for whom the reversible option was genuinely delivered — not mentioned, not half-attempted in the wrong shoes — and genuinely failed. The situations where nobody sensible argues for another injection:

• The conservative option was real, and it failed. A proper trial of a wide toe box and offloading, run for months rather than weeks, diagnosis confirmed first. A failed test only informs if the test was run.
• The pain is setting the terms of the life. When shoes, walking distance, standing at work, and exercise have all reorganised around one square centimetre of foot, wait, it might settle has lost the argument.
• Numbness arriving on its own. A nerve losing function is a different problem from a nerve that is irritated, and a reason to be assessed promptly rather than to wait out another course of anything.

None of this is about avoiding the knife. It is about order — arriving at the operation with the reversible options genuinely spent rather than skipped. The most useful thing to carry out of an appointment is a written answer to one question: what would tell us this has failed, and by when? Someone who books surgery with that answer behind them is deciding. Someone who books because the injection wore off is reacting to a calendar.

What to ask before either needle or knife

Each question below has a real answer, and a clinician who gives it precisely is signalling they have thought past the next appointment. None requires a study of the foot to answer. Writing them out before the visit costs nothing and changes what the visit is.

• Are we sure this is the nerve? Ball-of-foot pain has several sources and they are not treated alike. What confirmed it, and how confident is that?
• If the injection helps, what will we conclude? The sharpest question in the room. If the answer is that it is working, so we will repeat it, the sciatica finding — relief with no reduction in later surgery — is a reason to press 1Ref 1Pinto RZ, Maher CG, Ferreira ML, et al. (2012).Epidural Corticosteroid Injections in the Management of Sciatica: A Systematic Review and Meta-analysis.Epidural corticosteroid injections give small, short-term relief of leg pain and disability in sciatica but no meaningful long-term benefit and no reduction in subsequent surgery — cited here, and labelled as sciatica, for the principle that an injection's early relief does not predict the longer course..
• If the injection does not help, what will we conclude? A test with the same next step regardless of its result is not a test.
• Which operation, and what does it remove? Taking out the nerve and releasing the tissue above it are different trades. Permanent numbness is a price, not a side effect, and deserves pricing out loud before consent.