Posterior ankle impingement syndrome

Posterior ankle pain that wakes up when the foot points, in a ballet dancer, a gymnast or a footballer. Posterior impingement is a clinical diagnosis that is easy to raise, and a diagnosis that imaging will not confirm, because it finds the same thing in people who have no pain anywhere.

What this article covers, and what it does not. Posterior impingement is a mechanical block in forced plantarflexion, between the tibia behind and the calcaneus below, on an os trigonum or a prominent posterior process of the talus. It is not an Achilles tendinopathy, see Achilles tendinopathy ; not a sprain, see lateral ankle sprain ; not instability, see chronic ankle instability ; not osteoarthritis, see talocrural osteoarthritis. And above all: it is not Haglund's disease, from which the differential diagnosis chapter separates it in a single sentence.

In brief: clinical summary

Posterior impingement cannot be seen on an image, it is reproduced under the hand. An MRI showing an os trigonum in a dancer says almost nothing: half of her fellow dancers have one and dance without pain.

What is posterior impingement, and why forced plantarflexion?

The posterior compartment of the ankle is a space that closes. Understanding what lies in it, and what becomes of that space when the ankle goes to the end of its plantarflexion, is enough to understand the whole condition.

A space that closes on its contents

In the neutral position, the back of the talocrural joint is a comfortable space. The posterior rim of the tibial plafond overhangs the dome of the talus; behind the talus, the posterior process runs down towards the greater tuberosity of the calcaneus; between the two tubercles of that process runs the flexor hallucis longus tendon, in its sheath, as though in a pulley.

Take the ankle into maximal plantarflexion, and that space closes. The tibia tilts backwards and comes into contact with whatever projects behind the talus. It is a normal phenomenon: every ankle reaches a block at the end of plantarflexion. The impingement becomes pathological when that block is repeated thousands of times, or when it occurs on a bony structure bulkier than average.

Imaging in the en pointe position shows it directly. In six professional dancers imaged on 3 tesla MRI in the position in which they dance, the posterior articular surface of the tibial plafond becomes incongruent with the talar dome and rests on the posterior talus, while the posterior rim of the plafond compresses Kager's fat pad.13 This is not a mechanical hypothesis: it is what the scanner records.

Two bony variants, one and the same consequence

Two morphological variations dominate the dance literature, the os trigonum and a prominent posterior process of the talus, to which are added less well-known bony causes and soft tissue causes.10 They have to be told apart, because their prognosis differs.

The distinction is not academic. Across 111 ankles arthroscoped for trigonal impingement, the os trigonum was at fault in isolation in 20.3 % of cases, against only 5.4 % for the Stieda process (P = 0.040). Conversely, the Stieda process was far more often accompanied by flexor involvement together with other lesions: 59.5 % against 35.1 % (P = 0.015).3 In other words: when it is a Stieda, it is even less likely that the bone sums up the problem.

Is the os trigonum really the cause?

That is the question that decides everything else. If the os trigonum is the cause, you remove it and the pain goes. The data say something more nuanced, and more useful.

A prevalence that everybody quotes and almost nobody sources

“The os trigonum is present in 7 to 25 % of the population”: the sentence circulates in reviews, dissertations and rehabilitation websites, generally with no primary reference. Two meta-analyses published a year apart now allow it to be replaced with traceable figures, and the gap between them is instructive.

The larger, published in 2025, pools 41 studies and 36,612 feet : the os trigonum is present in 9.0 % of feet (95 % CI 7.4-10.8), and bilaterally in 32,7 % of cases (95 % CI 23.3-43.7).1 The earlier one, published in 2024 on 18 studies and 17,626 ankles, found 10,3 % (95 % CI 7-14.1), with no difference by sex or side.2 The two therefore agree on roughly one foot in ten.

But the overall figure hides the essential point. The prevalence depends massively on the way you look :

An MRI finds five times more os trigona than a dissection. It is obviously not that cadavers have fewer: it is that MRI sees synchondroses and small ossification centres that dissection classes differently and that radiography does not show. The practical consequence is immediate: the more finely you image a painful ankle, the more likely you are to find an os trigonum in it, whether or not it is the culprit.

What arthroscopy finds when it really looks

The most decisive piece of data on this subject comes from a series of 111 ankles operated on by posterior arthroscopy for trigonal impingement, between 2011 and 2016. Arthroscopy has the advantage of being dynamic and magnifying : it sees what static slices miss.

Result: bony impingement was at fault in isolation in only 15.3 % of cases. In the remaining 84.7 %, it was accompanied by something else, and not by a single thing: the commonest number of additional pathologies was three.3

Out of seven ankles operated on for posterior bony impingement, only one had nothing but that bony impingement. Removing the bone and stopping there means treating a sixth of the problem in six cases out of seven.

How do you recognise posterior impingement on examination?

The test is simple, it is the pivot of the diagnosis, and it has never been evaluated. That sentence sounds contradictory; it describes exactly the state of the literature, and it is better to know it before assigning a confidence you do not have.

The history already does half the work

Three elements point the way even before you touch the ankle.

• The site. A pain that is deep and posterior, which the patient localises poorly, often by pinching the back of the ankle between thumb and index finger, on either side of the Achilles tendon, rather than by pointing to the tendon itself.
• The triggering movement. The forced pointed foot : relevé and pointe work in the dancer, instep kicking in the footballer, landing and take-off in the gymnast, entry into the water in the diver. The patient says spontaneously “it is when I point my foot”.
• The mode of onset. Gradual in most cases, but not always: in the professional footballer, 53 % of posterior impingements have an acute onset, against 31 % of anterior impingements (P = 0.03).14 Posterior impingement can therefore reveal itself after a single injury.

This last point deserves remembering, because it generates errors. An inversion sprain followed by persistent posterior pain is not necessarily a sprain that is dragging on: it may have unmasked a posterior impingement, or a tenosynovitis of the flexor. That is exactly the case reported in a 37-year-old man whose initial diagnosis of lateral sprain was corrected, two weeks later, to a tenosynovitis of flexor hallucis longus confirmed on MRI.25

The forced passive plantarflexion test

The patient is seated, knee flexed to relax the triceps surae. The examiner grasps the forefoot and passively takes the ankle into maximal plantarflexion, with a quick, repeated movement rather than a slow sustained push. The test is positive if the movement reproduces the patient's posterior pain, not a simple sense of a block, which every normal ankle experiences at the end of range.

Two refinements increase the clinical yield:

• Adding a rotatory component sometimes separates a posterolateral from a posteromedial impingement, the two sites of synovitis found at arthroscopy in 22.5 % and 25.2 % of cases respectively.3
• Adding forced subtalar pronation to maximal plantarflexion points towards the subtalar joint rather than the talocrural compartment. That manoeuvre corrected the diagnosis in an adolescent dancer initially labelled “os trigonum syndrome” who in fact had arthritis of the posterior talocalcaneal facet.26

The flexor test, which must not be forgotten

Since the flexor is involved in nearly seven cases out of ten, examining it is not an extra, it is part of the diagnosis.

• Dynamic palpation. The tendon is palpated behind the medial malleolus, just behind the posterior tibial neurovascular bundle. Ask for repeated flexion and extension of the hallux, with the ankle in slight plantarflexion: you are looking for crepitus, a trigger, or a painful point that moves with the tendon.
• Triggering of the hallux. Locking or an audible click on active extension of the hallux signals a stenosing tenosynovitis: hallux saltans, or “trigger toe”. In a 16-year-old dancer, that sign preceded the diagnosis by two years, then recurred on the opposite side eight months after the first operation.27
• Tensioning test. Dorsiflexion of the ankle and extension of the hallux, compared with the sound side, looking for restriction or reproduction of pain.

The diagnostic injection: what it is worth, and what is unknown

The principle is logical: if injecting a local anaesthetic into the posterior compartment abolishes the pain on the plantarflexion test, the source really is there. The technique is described by a posterolateral approach, between the posterior process of the talus and the posterior rim of the tibia, and its use is long-standing.2220

What is unknown: its diagnostic accuracy has never been measured. No study provides sensitivity, specificity or predictive value for this procedure in posterior impingement. It is used as confirmation before a surgical decision, and that is a reasonable use; it is not a validated test. The distinction matters when you discuss it with the patient.

What is imaging really worth, and when should you ask for it?

Four converging studies, including one systematic review that could include no study at all, say the same thing: imaging of the posterior compartment nearly always shows something, and that something is not correlated with pain.

The systematic review that found nothing to include

In 2022, an Australian team looked for every study that had compared the imaging of patients with posterior impingement with that of asymptomatic controls. It screened 8,394 titles and abstracts, read 156 full texts, and included… no study at all. Not a single publication had ever made that elementary comparison.4

An empty systematic review is not a methodological failure: it is a result. It establishes that the diagnostic use of MRI in this condition rested, until then, on nothing comparative.

What is found in those who have no pain

The same authors filled the gap. Thirty-eight professional dancers and thirty-eight matched elite athletes, all in full training load, underwent a standardised 3 tesla MRI of one ankle:

Left-hand column: Baillie et al., Skeletal Radiol 2021 (PMID 34013446). Right-hand column: Katakura et al., Orthop J Sports Med 2024 (PMID 39143984). The two populations differ, and so do the protocols: the figures are to be read as orders of magnitude, not as a direct comparison.

The decisive point lies elsewhere than in the percentages. Among the 31 asymptomatic dancers followed for a year, only two developed symptoms requiring medical care.7 In other words: an abnormal MRI in a dancer with no pain predicts almost nothing.

The study that tested the association directly

It remained to check that, in genuinely symptomatic patients, imaging correlated with the clinical picture. Eighty-two elite athletes ( 43 ballet dancers, 24 cricket fast bowlers, 15 footballers) were assessed clinically then imaged on 3 tesla MRI. The result is clear and negative right across the board:

• the MRI signs are associated neither with posterior pain reported on a body chart, nor with the plantarflexion provocation test ;
• they are associated with no functional score reported by the patient (OSTRC overuse questionnaire, FAAM sport subscale);
• they do not differ between athletes carrying a clinical diagnosis of posterior impingement and those who do not.6

The authors' conclusion deserves quoting for its intent: the absence of association questions the very role of imaging in this condition, and clinicians should rely first on clinical assessment for diagnosis as well as for management.

An MRI will not tell you whether your patient has posterior impingement. It will tell you what is in their posterior compartment, which is not the same question, and does not have the same answer.

So when should you ask for imaging?

Not to make the diagnosis, but for three precise uses :

On this last point, a useful nuance: the flexor tendon does not thicken simply by adapting to dance. Comparing healthy non-dancers, healthy dancers and dancers with clinically diagnosed tendinopathy, an ultrasound study found thickening only in the symptomatic dancers, no difference between healthy dancers and non-dancers.30 A thick tendon in a dancer is therefore not “normal for a dancer”: it is a sign. The sample remains small, ten subjects per group, and no micromorphological abnormality was demonstrated.

Why does ballet concentrate this diagnosis?

An article on posterior impingement that did not talk about dance would miss most of the clientele of this diagnosis. The en pointe position is not a slightly exaggerated plantarflexion: it is a constraint of another order, and the figures show it.

What pointe work demands of the ankle

Ballet asks of the ankle a range that almost no other activity demands. In fifteen dancers of university, vocational and professional level, the measured ranges are as follows:

Two practical lessons. The first: measuring without load underestimates what the dancer actually does. Assessing a dancer's ankle lying down means measuring something other than the position in which it hurts. The second: goniometry and inclinometry do not measure the same thing: 77° against 89° on the same ankles. Choosing one tool and sticking to it is better than comparing values from two methods.12

One observation from that same study deserves particular attention: dorsiflexion decreases and plantarflexion increases as the level of dance rises. The professional dancer's ankle is therefore an ankle that has drifted towards plantarflexion: the very movement that closes the posterior compartment.

Where the range comes from: the talocrural joint carries 70 % of the load

You might hope that the 83° of pointe would be shared between the ankle and the joints of the foot. Superimposed weight-bearing radiographs in seven dancers show that it is not: the talocrural joint alone provides 57.6° ± 5.2 in the en pointe position, that is about 70 % of total plantarflexion ; the remaining 30 % come from the joints of the foot.11 In demi-plié, the talocrural joint provides 24.6° ± 9.6 of dorsiflexion.

That is a mechanically weighty piece of data: the load is not diluted. It concentrates exactly where the impingement sits.

Tenosynovitis of the flexor, almost the rule in the dancer

Flexor hallucis longus runs through a narrow fibrous sheath between the two tubercles of the posterior process of the talus. In the dancer, that tendon works at full tension in exactly the position in which its pulley is compressed: en pointe, the hallux is extended, the tendon is taut, and the compartment is closed. Hence its historical nickname of “dancer's tendinitis”.

Its frequency explains why it cannot be treated as a detail. Across 111 ankles operated on for trigonal impingement, the flexor was involved in 69,4 % of cases.3 The systematic review devoted to dancers in fact deals with the two entities together, while specifying that they can exist in isolation and that they are distinct pathological entities, which has direct consequences for the delay before return to dance, as the surgical chapter comes back to.18

A population whose workload is the first factor

The context in which this diagnosis arises is not incidental. Prospective weekly follow-up of three professional companies ( 57 dancers, 44 weeks, 1,627 reports) gives the measure of the background symptom load:

• 82.2 % of weeks include musculoskeletal pain;
• the ability to dance at full potential is reduced one week in two (52,6 %) ;
• 96.5 % of dancers report at least one injury over the season, with a mean of 5.6 health problems per dancer ;
• the ankle is the leading region affected, ahead of the thigh, foot and lower back;
• the subjective causes cited first are excessive workload (35.3 %), fatigue (22.4 %) and overload with insufficient recovery (21.6 %).15

These figures govern physiotherapy management. In a population that is in pain four weeks out of five and that itself names load as the primary cause, the most profitable intervention is not one more exercise: it is a negotiation of load with the dancer and their staff.

How do you tell posterior impingement from the other posterior pains?

The back of the ankle is a crossroads in the literal sense: bones, tendons, bursae, joints and supernumerary muscles all sit side by side. The movement that triggers the pain separates them better than the topography does.

A question of movement, not of area

Asking “where does it hurt?” of a patient with pain at the back of the ankle produces much the same answer whatever the cause. Asking “what were you doing when it hurt?” discriminates immediately. The table below is built on that principle.

In the dancer, the list of causes to rule out is well established: subluxation of the peroneal tendons, posterior impingement on a painful os trigonum, osteochondritis dissecans of the posterior talus, flexor hallucis longus tendinopathy and tibialis posterior tendinopathy.29 For the last of these, see tibialis posterior tendinopathy, whose pain sits at the medial border and is accompanied by collapse of the arch.

Table sources: the dancer's differential list after Luk et al., J Dance Med Sci 2013 (PMID 23759482); accessory soleus muscle after Pogliacomi et al., Ann Ital Chir 2026 (PMID 42304152); the Haglund distinction after Desai et al., Phys Sportsmed 2023 (PMID 35583477); forced subtalar pronation after Futamura et al., Cureus 2025 (PMID 41098269).

Posterior impingement or Haglund's disease: the most deceptive neighbour

This is the confusion that costs most, because both conditions give posterior ankle pain in an athlete and the treatment diverges completely. The distinction fits into one sentence.

Posterior impingement hurts when the foot points; Haglund's disease hurts when the shoe rubs.

Let us set out what that sentence covers.

One honest qualification to finish: the two can coexist, and posterior endoscopy in fact treats both through the same approach: posterior impingement, subtalar osteoarthritis and retrocalcaneal bursitis appear together among the best-supported indications for hindfoot endoscopy.21 That the surgical route is shared does not excuse you from making the diagnosis: the conservative management has nothing in common. Adapting a shoe counter will not relieve posterior impingement, and working on pointe technique will not relieve a Haglund.

Bony or soft tissue impingement: does the treatment change?

Yes, and this is probably the most useful distinction in this article. A bony impingement and a tenosynovitis of the flexor do not respond to the same levers, do not recover in the same time, and do not force the same decisions.

Two mechanisms, two therapeutic logics

Delays from Rietveld et al., J Dance Med Sci 2018 (PMID 29510786): 11 weeks (range 4 to 36) for all surgery in the dancer, 16 weeks (range 8 to 36) for isolated tenosynovitis of the flexor. The rest of the table is an editorial synthesis of the mechanisms described by Mansur et al. (PMID 38559392) and Ishibashi et al. (PMID 36368844).

The counter-intuitive point deserves underlining: it is the soft tissue involvement that takes longest to recover, not the bony lesion. Five weeks more on average, and a range that goes up to 36 weeks. Telling a dancer whose real complaint is tendinous “we take the bone out, you dance in two months” is promising a timeline that will not be kept.

Why the flexor deserves treatment in its own right

Three arguments converge.

• It is involved in 69.4 % of cases at arthroscopy.3 In the dancer, the systematic review of the field deals with the two entities together precisely because they are rarely separable.18
• It is treatable. Unlike a bony block, a tendon responds to progressive loading and to restoration of its glide. The case of a 37-year-old patient treated for 14 weeks with soft tissue therapy, open then closed kinetic chain work, proprioception and conditioning, along with a cortisone injection, ended in a pain-free return to play.25
• It progresses if neglected. A tenosynovitis can become stenosing and produce permanent triggering of the hallux: a stage at which arthroscopic tenolysis becomes the solution, with excellent results but one more operation.27

What rehabilitation, and at what level of evidence?

The colours must be nailed to the mast before the modalities are set out: the first-line management recommended is the one with the weakest evidence. That is not a contradiction to hide, it is the state of the field.

The literature's admission

The systematic review devoted to the treatment of posterior impingement and flexor tenosynovitis in the dancer identified 27 publications reporting surgical results in 376 ankles. On the conservative side, it found only six publications, covering 33 ankles in 28 dancers, and concludes bluntly that this “allows no evidence-based recommendation”.18 The levels of evidence available are 4 and 5.

That does not mean conservative treatment does not work: the prospective series show the opposite. It means that nobody has compared the modalities with each other, and that no superiority of one protocol over another has been demonstrated. What can be stated with figures is the overall success rate.

The four levers of management

1. Adapt the load, and do it precisely

This is the lever best documented indirectly, through epidemiology rather than through trials. Recall that 35.3 % of professional dancers name excessive workload as the primary cause of their problems.15 Adapting the load does not mean stopping:

• Take away the end range, keep the volume. A dancer can carry on working at the barre, in the centre and on physical conditioning while temporarily limiting pointe work and maximal relevés. A footballer can run and play while reducing instep kicking.
• Count, do not estimate. Number of relevés per class, number of kicks per session, minutes on pointe: those are the variables you modulate, and they have to be quantified to be modulated.
• Use a heel lift to limit dorsiflexion range passively and change the loading, a classic measure described in the reviews of the field.23
• Adapt the footwear : worn pointe shoes, a box that is too soft, unsuitable football boots ; these are parameters that can be changed for free.

2. Restore the capacity of the triceps surae

This is the functional deficit that has been measured. Compared with dancers and athletes matched for age, sex and activity, athletes with posterior impingement perform significantly fewer repetitions on the single-leg heel raise endurance test (P = 0,02).9

The caveat to lay down straight away : this study counts ten subjects per group. It is a solid lead, consistent with clinical practice, not robust evidence. It justifies measuring the test in every patient and working on it if it is deficient; it does not demonstrate that improving the test cures the impingement.

In practice: a classic progression of heel raises, first double-leg then single-leg, knee extended then knee flexed to load the soleus, avoiding the painful end range at the start: you work in the available, pain-free range, which is widened as sensitivity settles.

3. Treat flexor hallucis longus

Since it is involved in nearly seven cases out of ten, it deserves dedicated work:

• Tendon glide : active mobilisation of the hallux in different ankle positions, to restore the tendon's excursion in its pulley.
• Progressive strengthening of the flexor : resisted flexion of the hallux, then loaded work on demi-pointe with control of first ray loading.
• Correction of the loads upstream : a hallux that lacks extension forces the tendon to work under permanent tension on demi-pointe.
• Local soft tissue therapy, described in the detailed conservative protocol of the only published case report with a full description of the programme.25

4. Work on technique, and on perceived instability

The second finding of the study on functional deficits is perhaps the most surprising: athletes with posterior impingement report significantly greater perceived instability on the Cumberland questionnaire (P = 0.004), with no difference in joint range or Beighton score.9 A patient with posterior impingement can therefore complain of giving way: see chronic ankle instability for the full framework of that complaint, whose mechanism is different.

On pointe technique, the usual targets: alignment of the foot-ankle-knee unit on demi-pointe and on pointe, avoiding supination of the calcaneal block, control of the descent from pointe rather than dropping out of it, quality of the cushioning demi-plié . These elements belong to cooperation with the dance teacher more than to the rehabilitation gym.24

The modalities against their level of evidence

Two patients out of three recover without surgery, and we cannot say which of our actions contributes to it. Owning that ignorance is better than selling a protocol the literature does not support.

When does surgery become legitimate, and what should you expect from it?

After failure of a conservative treatment carried through for three months. The published results are good, consistent, and rest almost entirely on retrospective series: both pieces of information have to be given together.

Open or endoscopic

The two-portal endoscopic technique in the prone position, described in 2000 in the case of a professional dancer with a bilateral os trigonum and chronic flexor tendinitis, is the most widely used today.20 It allows access to the posterior compartment, the subtalar joint and the flexor through the same approach.

A meta-analysis of 32 studies compared the two approaches:

Source: Zwiers R, Miedema T, Wiegerinck JI, Blankevoort L, van Dijk CN. Open Versus Endoscopic Surgical Treatment of Posterior Ankle Impingement: A Meta-analysis. Am J Sports Med. 2022;50(2):563-575. PMID 34048272.

This table reads both ways, and it has to be read honestly. The functional result and the delay before return do not differ significantly : endoscopy is not magic. On the other hand, when the methodologically poor studies are excluded, the gap in complications becomes clear: 2 % against 24 %. It is on safety, not on performance, that endoscopy stands out.

What can be promised, and to whom

What the series allow you to announce:

• In the operated dancer : a good to excellent result in 89 % of cases, return to dance at 11 weeks on average but with a real range of 4 to 36 weeks, and 16 weeks if the flexor alone is at fault.18
• In the professional footballer : of eight players operated on, all returned to their previous level in 49.8 days on average, with no complication or recurrence at a mean follow-up of 36.5 months.16
• In a non-athletic population : AOFAS from 66.4 to 96.8, VAS from 6.4 to 0.9, return to work in 4.2 weeks, but 11.1 % sural nerve dysaesthesia.17

What do concrete clinical cases teach us?

Five published, indexed cases, chosen because each illustrates an error or a decision that the review literature does not show. None is reconstructed: these are real observations, with their identifiers.

Case 1: When the os trigonum diagnosis is right, and incomplete

Case 2: The sprain that was no longer one

Case 3: Rehabilitation after excision, described step by step

Case 4: When the tenosynovitis becomes stenosing

Case 5: The perfect mimic

How do you apply this concretely in practice?

This chapter condenses the article into actions. It is written to be reread before a consultation.

At the first consultation

1. Get the movement specified, not the area. “What exactly were you doing when it hurt?” Forced pointing of the foot points to impingement; rubbing of the shoe to Haglund; running and jumping to the Achilles.
2. Date it and put it in context. A return after a break, a rise in rehearsals, a move onto pointe, a change of footwear, a recent sprain. Load is the first factor cited by the dancers themselves.
3. Do the forced passive plantarflexion test, knee flexed, with quick repeated movements. Positive only if it reproduces THE patient's pain.
4. Add the variants : a rotatory component to lateralise, forced subtalar pronation to test the subtalar joint.
5. Examine the flexor systematically : dynamic medial retromalleolar palpation, looking for crepitus and triggering of the hallux, comparative tensioning.
6. Measure the single-leg heel raise test on both sides, and record the number of repetitions. It is the only documented functional deficit.
7. Go through the red flags before mobilising forcefully.

What to tell the patient

A twelve-week progression framework

This framework is an editorial construction assembling the principles of the protocols described by Kudaş et al. (PMID 27919560), Cengiz et al. (PMID 35271461), Senécal & Richer (PMC4915470) and Filipa & Barton (PMID 29113569). No rehabilitation protocol has been validated by a controlled trial in this condition; the durations are indicative and follow the three-month interval used by the published series.

When to refer

• Failure of three months of well-conducted and well-followed conservative treatment.
• Permanent triggering of the hallux with locking: established stenosis calls for tenolysis.
• Persistent diagnostic doubt in a typical picture that does not yield: image to look for an accessory soleus, an unrecognised fracture, subtalar involvement.
• A constrained professional context, a dancer or player mid-season, where the decision has to be taken with the medical staff, knowing the respective timelines of the two approaches.
• Any red flag, without delay.

Frequently asked questions

Should an os trigonum be removed because it is there?

No. It is present in about 9 % of the general population1 and in half of professional dancers with no pain.5 Its presence is not an indication. What indicates a procedure is a concordant clinical picture resisting conservative treatment.

Is an MRI needed to make the diagnosis?

No. In 82 elite dancers and athletes, the MRI signs were associated neither with pain, nor with the provocation test, nor with the functional scores, and did not differ between those with and without the clinical diagnosis.6 Imaging serves to rule out a differential or to prepare a surgical decision.

Does posterior impingement affect only dancers?

No, even if ballet accounts for a large share of it. It is described in the footballer, where it represents 62 % of ankle impingements and remains 1.7 times commoner than anterior impingement14, in the gymnast, the diver, the cricket fast bowler, and in a non-athletic population, where a series of 46 patients has been published.17

Can you carry on dancing or playing during treatment?

Usually yes, provided the provoking movement, the end range, is removed precisely while the rest is kept. Of 26 professional footballers treated according to a conservative algorithm, 18 returned to training in 36 days on average without surgery.16

What exactly is the difference from Haglund's disease?

The triggering movement. Posterior impingement hurts when the foot points: forced pointe, relevé, kicking. Haglund's disease hurts when the shoe rubs the posterosuperior angle of the calcaneus, and settles barefoot or in an open shoe.31 Impingement is deep and invisible, Haglund gives a visible swelling. The populations differ too: young athletes on one side, adults around forty on the other.32

Is the forced plantarflexion test reliable?

It is the pivot of the diagnosis, but its sensitivity and specificity have never been published for posterior impingement. The values of 95 % and 88 % often attributed to it come from a different test, concerning synovial impingement that is anterolateral.8

How long before returning after an operation?

In the dancer, 11 weeks on average for all surgery together, with a real range of 4 to 36 weeks; 16 weeks if tenosynovitis of the flexor is isolated.18 In the professional footballer, about 50 days.16 And a detailed case report of rehabilitation after excision describes 20 weeks of follow-up before unrestricted resumption.24

Is endoscopy or the open approach better?

The functional score, satisfaction and the delay before return do not differ significantly between the two. Endoscopy stands out on complications : 2 % against 24 % when methodologically poor studies are excluded.19

Does an injection settle the problem?

It occupies the place of an intermediate step in the published algorithms, between rehabilitation and surgery,16 and it allowed a return to sport at 60 days in one case of associated subtalar arthritis.26 But no controlled trial has measured its own efficacy in this indication.

Can this impingement appear after a sprain?

Yes. In the professional footballer, 53 % of posterior impingements have an acute onset.14 And a published case describes an inversion sprain whose persistent pain turned out to be a tenosynovitis of the flexor.25 See lateral ankle sprain for the initial management.

Bibliography

1. Preinl M, Osiowski A, Stolarz K, Osiowski M, Taterra D. Prevalence and clinical aspects of os trigonum: a meta-analysis. Anat Sci Int. 2025;100(3):287-297. PMID 39586987. DOI 10.1007/s12565-024-00811-4.
2. Ráfare AL, Rodriguez BC, de Souza Ribeiro BF, Fernandes RMP, Babinski MA, Pires LAS. Os trigonum: a discussion of its role in posterior ankle impingement syndrome and a meta-analysis of its prevalence. Surg Radiol Anat. 2024;46(7):1137-1143. PMID 38780789. DOI 10.1007/s00276-024-03389-8.
3. Mansur NSB, Femino JE, Chinnakkannu K, Fayed A, Glass N, Phisitkul P, Amendola A. Posterior Ankle Impingement: It's Not Only About the Os Trigonum. Foot Ankle Orthop. 2024;9(1):24730114241241326. PMID 38559392. DOI 10.1177/24730114241241326.
4. Baillie P, Mayes S, Lam J, Ferrar K, Cook J. Associations between clinical and imaging findings in posterior ankle impingement syndrome: a systematic review. Acta Radiol. 2022;63(5):652-657. PMID 33874783. DOI 10.1177/02841851211008389.
5. Baillie P, Cook J, Ferrar K, Smith P, Lam J, Mayes S. Magnetic resonance imaging findings associated with posterior ankle impingement syndrome are prevalent in elite ballet dancers and athletes. Skeletal Radiol. 2021;50(12):2423-2431. PMID 34013446. DOI 10.1007/s00256-021-03811-x.
6. Baillie P, Ferrar K, Cook J, Smith P, Lam J, Mayes S. Posterior Ankle Impingement Syndrome Clinical Features Are Not Associated With Imaging Findings in Elite Ballet Dancers and Athletes. Clin J Sport Med. 2022;32(6):600-607. PMID 36315819. DOI 10.1097/JSM.0000000000001021.
7. Katakura M, Clark R, Lee JC, Mitchell AWM, Shaw JW, Tsuchida AI, Jones M, Kelly S, Calder JDF. Foot and Ankle MRI Findings in Asymptomatic Professional Ballet Dancers. Orthop J Sports Med. 2024;12(8):23259671241263593. PMID 39143984. DOI 10.1177/23259671241263593.
8. Molloy S, Solan MC, Bendall SP. Synovial impingement in the ankle. A new physical sign. J Bone Joint Surg Br. 2003;85(3):330-333. PMID 12729103. DOI 10.1302/0301-620x.85b3.12873. Concerns anterolateral synovial impingement, not posterior impingement.
9. Baillie P, Cook J, Ferrar K, Mayes S. Single-Leg Heel Raise Capacity is Lower, and Perceived Ankle Instability is Greater, in Dancers and Athletes With Posterior Ankle Impingement Syndrome. Clin J Sport Med. 2024;34(4):376-380. PMID 38507243. DOI 10.1097/JSM.0000000000001217.
10. Russell JA, Kruse DW, Koutedakis Y, McEwan IM, Wyon MA. Pathoanatomy of posterior ankle impingement in ballet dancers. Clin Anat. 2010;23(6):613-621. PMID 20821398. DOI 10.1002/ca.20991.
11. Russell JA, Shave RM, Kruse DW, Koutedakis Y, Wyon MA. Ankle and foot contributions to extreme plantar- and dorsiflexion in female ballet dancers. Foot Ankle Int. 2011;32(2):183-188. PMID 21288419. DOI 10.3113/FAI.2011.0183.
12. Russell JA, Kruse DW, Nevill AM, Koutedakis Y, Wyon MA. Measurement of the extreme ankle range of motion required by female ballet dancers. Foot Ankle Spec. 2010;3(6):324-330. PMID 20581228. DOI 10.1177/1938640010374981.
13. Russell JA, Yoshioka H. Assessment of female ballet dancers' ankles in the en pointe position using high field strength magnetic resonance imaging. Acta Radiol. 2016;57(8):978-984. PMID 26567962. DOI 10.1177/0284185115616295.
14. D'Hooghe P, Waldén M, Hägglund M, Bengtsson H, Ekstrand J. Anterior ankle impingement syndrome is less frequent, but associated with a longer absence and higher re-injury rate compared to posterior syndrome: a prospective cohort study of 6754 male professional soccer players. Knee Surg Sports Traumatol Arthrosc. 2022;30(12):4262-4269. PMID 35689100. DOI 10.1007/s00167-022-07004-4.
15. Junge A, Hauschild A, Stubbe JH, van Rijn RM. Health Problems of Professional Ballet Dancers: an Analysis of 1627 Weekly Self-Reports on Injuries, Illnesses and Mental Health Problems During One Season. Sports Med Open. 2024;10(1):79. PMID 39017734. DOI 10.1186/s40798-024-00753-1.
16. Kudaş S, Dönmez G, Işık Ç, Çelebi M, Çay N, Bozkurt M. Posterior ankle impingement syndrome in football players: Case series of 26 elite athletes. Acta Orthop Traumatol Turc. 2016;50(6):649-654. PMID 27919560. DOI 10.1016/j.aott.2016.03.008.
17. Cengiz B, Moradi R, Karaoglu S. Posterior Ankle Impingement Syndrome in a Nonathletic Population: Causes, Treatment Modalities, and the Results of Endoscopic Treatment. J Am Podiatr Med Assoc. 2023;113(4). PMID 35271461. DOI 10.7547/21-001.
18. Rietveld ABMB, Hagemans FMT, Haitjema S, Vissers T, Nelissen RGHH. Results of Treatment of Posterior Ankle Impingement Syndrome and Flexor Hallucis Longus Tendinopathy in Dancers: A Systematic Review. J Dance Med Sci. 2018;22(1):19-32. PMID 29510786. DOI 10.12678/1089-313X.22.1.19.
19. Zwiers R, Miedema T, Wiegerinck JI, Blankevoort L, van Dijk CN. Open Versus Endoscopic Surgical Treatment of Posterior Ankle Impingement: A Meta-analysis. Am J Sports Med. 2022;50(2):563-575. PMID 34048272. DOI 10.1177/03635465211004977.
20. van Dijk CN, Scholten PE, Krips R. A 2-portal endoscopic approach for diagnosis and treatment of posterior ankle pathology. Arthroscopy. 2000;16(8):871-876. PMID 11078550. DOI 10.1053/jars.2000.19430.
21. Spennacchio P, Cucchi D, Randelli PS, van Dijk NC. Evidence-based indications for hindfoot endoscopy. Knee Surg Sports Traumatol Arthrosc. 2016;24(4):1386-1395. PMID 26744282. DOI 10.1007/s00167-015-3965-1.
22. Yasui Y, Hannon CP, Hurley E, Kennedy JG. Posterior ankle impingement syndrome: A systematic four-stage approach. World J Orthop. 2016;7(10):657-663. PMID 27795947. DOI 10.5312/wjo.v7.i10.657.
23. Ishibashi MA, Doyle MD, Krcal CE Jr. Posterior Ankle Impingement Syndrome. Clin Podiatr Med Surg. 2023;40(1):209-222. PMID 36368844. DOI 10.1016/j.cpm.2022.07.014.
24. Filipa A, Barton K. Physical Therapy Rehabilitation of an Adolescent Preprofessional Dancer Following Os Trigonum Excision: A Case Report. J Orthop Sports Phys Ther. 2018;48(3):194-203. PMID 29113569. DOI 10.2519/jospt.2018.7508.
25. Senécal I, Richer N. Conservative management of posterior ankle impingement: a case report. J Can Chiropr Assoc. 2016;60(2):164-174. PMC PMC4915470.
26. Futamura H, Kawabata M, Nakagawa H, Futamura R, Sugimoto K. Posterior Ankle Pain in a Ballet Dancer: Dynamic Ultrasonography of Simultaneous Onset of Subtalar Joint Arthritis and Os Trigonum Syndrome. Cureus. 2025;17(9):e92241. PMID 41098269. DOI 10.7759/cureus.92241.
27. Gasti AV, James R, Neumayer F. Bilateral hallux Saltans, a Rare Finding: Diagnosis and Arthroscopic Treatment in a Young Ballet Dancer: Report of a Case. J Orthop Case Rep. 2025;15(4):161-166. PMID 40212508. DOI 10.13107/jocr.2025.v15.i04.5484.
28. Pogliacomi F, Battaglia S, Quattrini F, Ceccarelli F, Calderazzi F. Accessory Soleus Muscle and Persistent Posterior Ankle Pain: Report of Three Cases. Ann Ital Chir. 2026;97(6):992-998. PMID 42304152. DOI 10.62713/aic.4588.
29. Luk P, Thordarson D, Charlton T. Evaluation and management of posterior ankle pain in dancers. J Dance Med Sci. 2013;17(2):79-83. PMID 23759482. DOI 10.12678/1089-313x.17.2.79.
30. Mikkelsen P, Andersen A, Shih HS, Rowley KM, Kulig K. Flexor hallucis longus tendon morphology in dancers clinically diagnosed with tendinopathy. J Ultrasound. 2024;27(1):41-49. PMID 37356071. DOI 10.1007/s40477-023-00793-5.
31. Desai SS, Wong TT, Crockatt WK, Tedesco LJ, Trofa DP, Popkin CA. The 'Bauer bump:' ice hockey skates as a common cause of Haglund syndrome. Phys Sportsmed. 2023;51(5):414-419. PMID 35583477. DOI 10.1080/00913847.2022.2079962.
32. Alessio-Mazzola M, Russo A, Capello AG, Lovisolo S, Repetto I, Formica M, Felli L. Endoscopic calcaneoplasty for the treatment of Haglund's deformity provides better clinical functional outcomes, lower complication rate, and shorter recovery time compared to open procedures: a systematic review. Knee Surg Sports Traumatol Arthrosc. 2021;29(8):2462-2484. PMID 33216187. DOI 10.1007/s00167-020-06362-1.

Methodological note. Every reference was checked against the NCBI E-utilities API before being cited: existence of the PMID, author list, journal, year, pagination and DOI, then the abstract was read to verify that the source really establishes what is attributed to it. The levels of evidence in the modalities table are an editorial appraisal applying GRADE principles, and not the reproduction of a published GRADE assessment: none exists on this subject. The strongest formal recommendation in the field is grade Cf, that is, based on low-quality evidence.21 ICD-11 contains no entity corresponding to posterior ankle impingement or to the os trigonum: the structured mark-up of this page therefore carries no code, rather than an approximate one. Article written on 15 August 2026.