Interdigital neuroma

Latest evidence check: March 2010.

Principal authors: Gillian Jackson and Jim Barrie

Also known as Morton's metatarsalgia or Morton's neuroma, although Morton described neither (Morton thought this was a problem in the 4th MTP joint and Betts described the "neuroma" 70 years later).


The "neuroma" consists of degenerative and fibrotic changes in the common digital nerve near its bifurcation. However, there may be similar changes in adjacent unaffected nerves and it is not known why one becomes symptomatic. A number of causative factors have been suggested including:

Clinical features

The symptoms may be quite non-specific:

Symptoms are commonest in the 3rd interdigital space, then the 2nd. Symptoms in the 4th space are rare and should make one doubt the diagnosis. Symptoms in the first space are virtually unknown.

The condition may remain undiagnosed for many years.

The diagnosis is often strongly suspected within the first minute of the consultation. However, it may be arrived as part of the assessment of a more generalised metatarsalgia. In any case, a full assessment of the foot should be carried out.

Ask about:

Examination should begin with assessment of any suggested nerve entrapment in the spine, proximal limb or tarsal tunnel.

The whole foot should be examined, looking for any other factors likely to produce metatarsalgia.

On local examination look for:

A digital nerve stretch test is described by Cloke (2006) with high sensitivity in the presence of other positive signs of neuroma but no reference to its use to differentiate from other diagnoses.

A local anaesthetic injection into the affected space may be useful - if it relieves the symptoms this is supportive of the diagnosis. The differential/ concurrent diagnosis of MTPJ synovitis can be confirmed with MTPJ injection. (Miller 2001) However, further critical study of the diagnostic validity of injection would be helpful.


Both ultrasound and MRI have been described for imaging a neuroma, but the evidence for their value is not strong. However, if the clinical situation is atypical or there appear to be multiple diffential diagnoses imaging may be useful. If there is a suggestion of other forefoot pathology standing AP and lateral forefoot films should be obtained.

Ultrasound has a high sensitivity(95%) for web-space abnormality; however it may not accurately assess the size of a neuroma or separate it from other intermetatarsal pathologies (Read 1999). Sensitivity of up to 98% and specificity of 95% in diagnosis of neuroma are quoted by Gomez (2005). However they presume that because the 39 cases with negative USS resolved with conservative measures that a diagnosis of neuroma had not been missed by the scan.

Saragas (2006) had USS confirmation of all 43 clinically suspected neuromas, with subsequent histological confirmation in 97.6%. In the same paper another 45 clinically suspected neuromas all responded to a diagnostic steroid and local anaesthetic injection  and when excised were all histologically confirmed. Whilst USS is non invasive, injection is as reliable, can be done at initial consultation and can also be a definitive treatment.

Zanetti (1999) used clinician questionnaires to review the effect of MRI findings on diagnosis and treatment in suspected neuromas. They found an alteration in primary clinical diagnosis and/ or proposed treatment in 31/54 (57%) of feet following MRI, however several of these patients had other suspected differential diagnoses which may have lead to the initial request for the scan. There is no reference to how many patients were treated for neuroma by these clinicians without a scan.

Bencardino (2000) found that a third of neuromas found on MRI may exhibit no clinical symptoms. Biasca (1999) in only a small number of patients (19) relied on MRI to size the neuroma and predict surgical outcome.

Sharp (2003) reviewed 29 cases of clinically diagnosed neuroma who underwent USS and MRI. They found no requirement for imaging where the clinical diagnosis is clear.

Non-surgical management

All patients should be advised on the use of shoes with adequate room in the toe-box and high heels should be avoided. There is no proven role for orthoses (Kilmartin 1994).

If simple measures do not control the pain then a local anaesthetic and steroid injection into the intermetatarsal space can be offered. The patient is warned that it may be quite painful for several days and they may need to rest more than usual. Also warn about the small risks of infection and cutaneous atrophy. The published results of this treatment are variable. Greenfield (1984) found that 90% of patients had little or no pain two years later, even if they got temporary or no benefit from the initial injection. Bennett (1995) found that about 50% of patients were relieved of pain by a single injection, the authors imply, but do not substantiate, that this result was maintained at follow-up 2.5-5 years later. Rasmussen 1996, however, found that although 80% were relieved of pain by a single injection, 47% eventually had a neurectomy and most of the rest were symptomatic at review 2-6 years later. 64% of 171 patients in our series had a resolution of symptoms with a steroid and local anaesthetic injection, the other 36% proceeding to surgery.

Saygi 2005 found 82% of patients with injection  had complete or partial relief of pain compared with 63% treated with footwear modifications alone at one year.

Hughes 2007 reports 94% partial or total symptomatic relief in 101 patients who prospectively underwent USS guided alcohol injection. Only 3 patients went on to have surgery. There is a reduced risk of fat pad and cutaneous atrophy but 16.8% had increased pain for up to 3 weeks. A trial of alcohol vs steroid injection may be useful.
If symptoms persist despite non-surgical treatment and the diagnosis is regarded as firm enough the patient may be offered a surgery.



The standard operation is a digital neurectomy done most commonly through either a dorsal or plantar incision (see below for discussion). The nerve is divided 2-3cm proximal to the bifurcation and excised. The deep transverse metatarsal ligament may be wholly or partially released. Post-operatively the patient can mobilise fully weight bearing.

The largest clinical series is that of Pace (2010). 78 patients had neurectomies through a dorsal approach. 69/78 patients were women and the neuroma was in the 3rd space in 43, the 2nd in 20 and the 4th in 18. Mean follow-up was 4.6y. Pain scores improved markedly but pre-op scores were by recollection. 58/78 patients were satisfied with few or no reservations but 20 had major reservations or were dissatisfied. There were 8 re-operations for recurrent neuroma. In Coughlin’s well-documented but retrospective series of neurectomy via a dorsal approach (2001), overall satisfaction was good or excellent in 85% of 66 patients. 65% of feet were pain free at final follow-up, and this is consistent with several other studies. In our own retrospective series of 64 neurectomies by 2 surgeons 90% had a good or excellent result.

The common causes for recurrent symptoms after excision of an interdigital neuroma are inadequate resection of the nerve and neuroma formation in a place of movement, friction or pressure. A retrospective review by Myerson re-explored 60 interspaces. He suggests that a plantar approach and/or intermuscular transposition may reduce these recurrences. One should ensure an appropriate size of dorsal incision to allow adequate visualisation.

Dorsal versus plantar approach

The theoretical advantages of a dorsal approach are early ambulation as the incision is not on a weight bearing surface. The plantar approach, however, provides best access to the neuroma and preserves the deep intermetatarsal ligament, with a reduction in complications such as inadequate resection. The disadvantage of the plantar approach can be a painful scar on the weight bearing area.

Nashi (1997) compared 26 dorsal with 26 plantar neurectomies. They found increased satisfaction, faster weight bearing and return to work and a less painful scar in the dorsal group. In this study patients had a number of confounding variables which have not been accounted for in the analysis.

Two recent papers from Akermark in Sweden published  in 2008 address this issue. Their retrospective review of 69 longitudinal plantar incisions and 56 dorsal incisions found similar clinical outcomes and patient satisfaction (good or excellent in 88% and 84% respectively). However in the plantar group there was a significant reduction in long term sensory loss, post operative sick leave and complications. There was no difference in scar tenderness. Akermark’s prospective study of 59 plantar approaches found patient satisfaction to be excellent or good in 86%. 90% of patients had none or minimal scar tenderness. The group state that they have completed but not yet submitted a randomised, prospective and comparable study of plantar versus dorsal incisions. The results of this may help to resolve this issue.

Jerosch (2006) describe their plantar approach and review 415 cases. 328 of 356 cases were satisfied with the results; only 10 patients had persistant scar problems.

Su (2006) examined 674 consecutive pathological specimens after neurectomy - 638 via a dorsal and 36 via a plantar approach. The specimens contained digital artery in 38.9%. There was no significant difference between the approaches and rate of arterial resection. No adverse clinical effects were noted.

Neurectomy versus Neurolysis

Gauthier 1979 first described a decompression of the interdigital space with division of the deep transverse metatarsal ligament and neurolysis of the common digital nerve without resection of the neuroma. Although his outcome measures are not clear, 83% of his 206 patient series, had a ‘rapid and stable improvement’. Diebold (1996) reported satisdactory relief of pain in 35/40 (87.5%) patients. Okafor reprted complete relief of pain 17/35 patients and minimal pain with activity in 12. The results of neurolysis and neurectomy in case series seem comparable and a randomised controlled trial comparing these techniques would be helpful.

Colgrove (2000) reported excellent results with neurectomy and also with an alternative transposition technique, where the distal end of the transacted nerve is implanted in the intrinsic muscle without neurectomy. The theoretical advantage is a reduction in the post operative formation of a symptomatic transection neuroma. No further work has been done on this. A paper by Vito (2003) reports good results with a decompression and relocation procedure. Zelent (2007) describes a minimally invasive technique to release the intermetatarsal ligament in 14 patients.