Porter & Schon: Baxter's The Foot and Ankle in Sport, 2nd ed.

Section 2 - Sport Syndromes

Chapter 2 - Impingement syndromes of the foot and ankle

William G. Hamilton,
Mihir M. Patel,
Roman A. Sibel






General technique tips for osteophyte removal



Specific anatomic areas





“Impingement” is derived from the Latin verb impingere, meaning “to force against.” The periosteum is composed of two layers: the superficial (fibrous layer) and the deep (cambium) layer. The cambium layer has osteogenic potential. This can be seen following conditions that strip the periosteum from the underlying bone (e.g., tumors and fractures). In impingement, one bone's repetitively striking the other can stimulate the cambium layer to form osteophytes. Once the osteophytic prominence forms, impingement occurs more easily, so that the impingement spur, once formed, often increases in size and eventually may break off, forming a loose body. Conservative treatment should be aimed at breaking this repetitive cycle so that the impingement spurs do not enlarge and produce irritation. In ballet we tell the dancer not to “hit bottom” in the plié (knee-bend) when he or she lands from a jump and the ankle is forced into maximal dorsiflexion. These restrictions often are difficult to follow or are too restrictive. Therefore if the symptoms warrant, and conservative treatment is not working, surgery usually is indicated.


General Technique Tips for Osteophyte Removal



Get adequate exposure and visualization. If one is using the arthroscope and struggling, one should open it up and do the job right.



Make the skin incisions carefully to avoid incisional neuromas. Nothing is more discouraging than a good clean-out that is spoiled by a hypersensitive scar.



Be sure to get all the osteophytes out; there can be hidden spurs or more than one. If there is any doubt, it is best to take an x-ray in the operating room at the end of the case to make sure nothing has been missed.


Specific Anatomic Areas

The interphalangeal joint of the hallux

Dorsal impingement with spur formation similar to that seen in the first metatarsophalangeal (MP) joint can occur in this joint. It can be a sign of degenerative joint disease (DJD) but usually is secondary to stiffness and lack of motion in the adjacent MP joint. When hallux rigidus forms in the first MP joint, the interphalangeal (IP) joint will be forced into excessive dorsiflexion in an attempt to compensate for the lack of motion in the proximal joint. At times this can be dramatic. I once saw a female dancer who was born with congenital ankylosis of both first MP joints. They were totally rigid. She had Grecian (Morton's) feet with short first rays and had developed 90 degrees of dorsiflexion in her IP joints so that she had a full demi-pointe relevé. Rigidity in this joint can be treated similarly to the condition found in the first MP joint. One should remember, however, that this joint is forgiving, and surgery is rarely necessary.

The lesser metatarsophalangeal joints

Dorsal impingement in these joints usually is associated with Freiberg's disease.[1] This condition is no more common in athletes and dancers than it is in the general population. One should remember, however, that it can be symptomatic for as long as 6 months before it appears on x-ray and should be considered in unexplained metatarsalgia in young patients. A bone scan or magnetic resonance imaging (MRI) usually will confirm the diagnosis before plain radiographic changes are evident. Freiberg's infarction comes in the following four variations.[1]

Type I

The head of the metatarsal (MT) dies and then heals by “creeping substitution” (Phemister[2]). In this form it may heal completely, with little or no collapse, leaving the articular surface intact and almost as good as it was before the event occurred. Surgery often is not necessary.

Type II

The head collapses during revascularization and the articular surface settles and remains intact, but peripheral osteophytes form along the dorsal margin of the joint, limiting dorsiflexion. This type is amenable to a dorsal clean-out (cheilectomy), which should leave the joint intact and restore dorsiflexion. (The surgeon should remember to remove more bone than he or she thinks is necessary when performing this operation.)

Type III

The head collapses and the articular surface loosens and falls into the joint, leaving the joint totally destroyed. Obviously simply removing the osteophytes will not suffice in this case—an arthroplasty is required. All the necrotic bone must be excised from the MT head and all the dorsal osteophytes must be removed. Usually the plantar portion of the head is left when this has been done. The surgeon should be generous in the excision to permit full dorsiflexion later, but the entire MT head should not be removed. Either a dorsiflexion osteotomy of the MT head or a capsular arthroplasty, similar to the one described for use in the first MP joint, can be useful in this situation.

Type IV

Multiple heads are involved in the process ( Fig. 2-1 ). This type is rare and actually may be a form of epiphyseal dysplasia. Each MT head must be evaluated and treated individually.


Figure 2-1  A rare case of multiple Freiberg's disease.



A Freiberg-like syndrome can occur in the fifth MT head following a nondisplaced or minimally displaced fracture of the distal MT shaft, similar to the “boxer's fracture” of the fifth metacarpal ( Fig. 2-2 ).


Figure 2-2  “Boxer's fracture” of the distal fifth metatarsal (arrow).



Lesser MP joint instability

Metatarsalgia is not common in the young, healthy athletic population. When it is encountered, one should suspect either early Freiberg's disease or MP instability.[1] This subtle problem often goes unrecognized because the x-rays are normal. The patient presents with isolated metatarsalgia. There is plantar tenderness under the MT head and dorsal tenderness where the phalanx subluxes on top of the head when the patient relevés or goes up on the ball of the foot. The subluxated phalanx pushes the head of the metatarsal downward, producing the metatarsalgia, the so-called dropped metatarsal. It is recognized easily on physical examination if one remembers to observe for it. The Lachman test of the MP joint will be positive.[1] The base of the proximal phalanx is grasped in the fingers, and a dorsal-plantar force is applied. The instability is recognized easily when the phalanx subluxes on the top of the MT head ( Fig. 2-3 ).


Figure 2-3  Lachman test of the metatarsophalangeal (MP) joint.



Conservative treatment consists of padding to unload the painful MT head and taping or wearing a toe retainer to try to control the instability ( Fig. 2-4 ). It often is a frustrating situation for the dancer or athlete because he or she does not want to undergo surgery, but once the ligaments and plantar plate are stretched, they can be tightened again only by surgery. The surgical options for this problem in a dancer are tricky. The usual operations for this condition (stabilizing procedures such as the Girdlestone-Taylor operation[3]) are inappropriate for athletes because they stabilize the joint but also limit dorsiflexion—an unacceptable solution for dancers, gymnasts, and so forth. We have had success in a limited number of dancers and athletes with a resection arthroplasty and partial syndactyly, especially in the fourth MP joint, which seems especially prone to this problem. As previously noted, one should not remove too much of the proximal phalanx (one-fourth to one-third at most), and one should remove the plantar condyles of the MT head, use a toe wire, and remove it early (2 weeks).


Figure 2-4  Taping to control metatarsophalangeal (MP) instability.



Idiopathic synovitis

Idiopathic synovitis[4] is characterized by MP swelling, the so-called sausage toe. Its cause is controversial. (It usually is not caused by systemic inflammatory diseases, but of course these must be ruled out.) It usually is associated with laxity of the joint and MP instability. Whether the looseness irritates the joint and leads to chronic synovitis or the synovitis loosens the joint is not known. Conservative therapy involves stabilizing the joint by using tape or toe retainers (see Fig. 2-4 ); having the patient reduce activities and take anti-inflammatory medication; and, if necessary, giving the patient one or two (at most) intra-articular injections of steroids. If this fails, exploration and appropriate surgery are indicated. Surgical options include (1) extensor tendon lengthening with resection of the plantar condyles of the MT head, (2) Girdlestone-Taylor[3] procedure, (3) DuVries-type arthroplasty,[3] and (4) resection arthroplasty with partial syndactyly to the adjacent digit.

Medial midfoot impingements

True impingements in the midfoot are rare. Occasionally, accessory ossicles can be seen between the bases of the metatarsals or the cuneiforms; symptoms may warrant their removal. These bones, more often than not, will be asymptomatic.

An isolated osteophyte on the dorsum of the midfoot occasionally can cause entrapment of the deep peroneal nerve or irritation of the extensor hallucis longus (EHL) tendon as they pass over it. Initially, extra padding on the undersurface of the tongue of the shoe may resolve pain. If this fails, removal of the osteophyte may be necessary. The painful accessory navicular is not due to an impingement and is discussed in Chapters 8, 14, and 27 [] [] [] .

Lateral midfoot impingements

In the lateral midfoot, three related conditions are a combination of impingement and subluxation: (1) derangement of the cuboid—base of the fourth and fifth metatarsal joints, (2) cuboid subluxation, and (3) sinus tarsi syndrome.

Subluxation of the cuboid[5] and derangement of the cuboid—base of the fourth and fifth metatarsal joints—often are seen together. The subluxing cuboid is a common but poorly recognized condition. It presents as lateral midfoot pain and an inability to “work through” the foot. Pressing on the plantar surface of the cuboid in a dorsal direction is painful. The normal dorsal-plantar joint play is reduced or absent when compared with the uninjured side. (Because of this immobility, the condition sometimes has been referred to as a “locked cuboid.”) Often there is a shallow depression on the dorsal surface, a palpable fullness on the plantar aspect of the cuboid, and subtle forefoot valgus. Documentation by x-ray, computed tomography (CT) scan or MRI is difficult because of the normal variations found in the relationship between the cuboid and its surrounding structures. The diagnosis must be made on the basis of the history and physical findings. Treatment involves recognition of the pathology and manual reduction by a therapist or physician familiar with the condition and follow-up to be certain that it remains in place. Therapists and orthopaedists involved in the care of athletes and dancers should be aware of the subluxed cuboid and be able to recognize it when it occurs. When the cuboid subluxes plantarward, the bases of the fourth and/or fifth metatarsals often are elevated, causing the head of the fourth metatarsal to be depressed ( Fig. 2-5 , Table 2-1 ).


Figure 2-5  A “dropped” fourth metatarsal head resulting from elevation of its base.  From Marshall PM, Hamilton WG: Am J Sports Med 20:170, 1992.


Table 2-1   -- Cuboid subluxation



Lateral midfoot pain

Tender plantar mass

Weakness in push-off

Decreased joint play

Inability to “work through” the foot

Shallow depression over the cuboid

Function limited by pain

Subtle forefoot abduction



There usually are two types of cuboid subluxations: acute and chronic/recurrent. Treatment consists of recognition and manual reduction by a therapist familiar with the condition. The cuboid then must be held in place by taping and padding for several weeks to prevent recurrence. If the subluxation has gone unrecognized and the joint has been subluxed for any length of time, reduction can be difficult. The forefoot valgus must be corrected and the lateral column lengthened manually before the reduction can be performed. In the chronic condition, it may not be possible to keep the cuboid reduced if it goes in and out at random. In these cases, athletes often can be taught to reduce the subluxation themselves ( Fig. 2-6 ).


Figure 2-6  Reduction of a subluxed cuboid by the patient.



Sinus tarsi syndrome[4] is a controversial condition that produces pain deep in the sinus tarsi that increases with activity and is exacerbated by impact (jumping and running) and pronation. It is often, but not always, a sequel to a sprained ankle. On physical examination, there is discrete tenderness, or a “trigger point” deep in the sinus tarsi, and forceful abduction-pronation of the heel and midfoot may be painful. The condition usually can be confirmed with an injection of 0.5ml of lidocaine into the trigger point. If the pain is relieved by the local anesthetic, a second injection of 0.15ml of corticosteroids often can be highly effective. The condition is thought to have several etiologies: (1) soft tissue entrapment or partial tear of the interosseous talocalcaneal ligament, (2) osteophyte impingement ( Fig. 2-7 ), (3) neural entrapment (motor nerve to the extensor digitorum brevis [EDB]), (4) degenerative arthrosis, and (5) arthrofibrosis. It can be difficult to differentiate this syndrome from subtalar dysfunction, and osteophytes can be found in the sinus tarsi that are not causing symptoms. The two areas are anatomically close together. One of the best ways to differentiate one from the other is to pay close attention to subtalar motion. Mann and Coughlin[3] have shown how important subtalar motion is to normal foot mechanics. Subtle loss of this motion, such as arthrofibrosis of the subtalar joint from bleeding into the joint in conjunction with an ankle sprain, can cause residual symptoms after the sprain has healed.


Figure 2-7  Sinus tarsi syndrome; note osteophytes (arrow).



Conservative treatment consists of anti-inflammatory medication, physical therapy, a medial heel wedge or arch support to open up the sinus tarsi, and, if necessary, the previously mentioned cortisone injection. If symptoms persist and the diagnosis has been confirmed with lidocaine injection, surgical exploration and clean-out is indicated. This is one area in which an injection—if placed in the right spot—often is dramatically effective and will avoid surgery. Finally, the sinus tarsi syndrome often is found in conjunction with lateral ankle ligament laxity, and, in these cases, sinus tarsi exploration and debridement should be considered if ankle ligament reconstruction is planned. More about the subtalar joint can be found in Chapter 15 .

The ankle

When considering ankle impingement, one should remember the basic anatomy of the ankle. The talus sits sidesaddle on the os calcis so that the axis of the talus is roughly in line with the first web space of the foot and the axis of the os calcis is in line with the fourth web space ( Fig. 2-8 ). In dorsiflexion, bony impingement occurs anteromedially between the neck of the talus and the anterior lip of the tibia. In plantarflexion, bony impingement occurs posterolaterally between the os calcis and the posterior lip of the tibia. Therefore anteromedial and posterolateral problems usually are associated with bony impingement, whereas anterolateral and posteromedial problems usually are soft tissue in origin (there is no bony impingement in these areas). The anterior ankle is an extremely common location for impingement, but impingements can be found in all quadrants around the ankle: anterior, lateral, posterior, and medial.


Figure 2-8  Axis of talus versus axis of os calcis.



Anterior (medial, central, lateral)

Anteromedial ankle pain often comes from impingement of the anterior portion of the medial malleolus against an impingement spur on the medial shoulder of the talus. This spur is hard to see on x-ray because it cannot be visualized on standard lateral radiographs. However, nonweight-bearing, oblique x-rays of the foot may detect these spurs. The spur often can be palpated on physical examination and should be looked for in any anterior ankle clean-out. It is easy to miss, the “hidden spur.” To visualize and resect this spur arthroscopically, the surgeon must hold the ankle in dorsiflexion.

Anterocentral is the location of the classic anterior ankle impingement. It comes in the following four types:



Spurs are primarily on the lip of the tibia. This type is ideal for arthroscopic debridement. Under direct vision, the lip of the tibia can be removed fairly easily with a thin osteotome or burr. Care should be taken not to damage the dome of the talus, either by holding the ankle in maximal dorsiflexion or by using an osteotome with blunt edges, as described by Scranton and McDermott.[6]



Spurs are primarily on the neck of the talus. This type is more difficult to treat with the arthroscope, because the osteophytes often are within the capsular insertion on the neck of the talus, and it is necessary to strip off the capsule distally to visualize the pathology. It is easy to miss some of the osteophytes; therefore intraoperative imaging may be needed to ensure adequate removal.



Spurs are present on both the lip of the tibia and the neck of the talus, sometimes with loose bodies that have broken off the osteophytes. This type is common and is the most difficult to deal with. In the early 1980s, I thought professional dancers would return to dancing sooner if their anterior debridement could be done with the arthroscope. I found that, in all but the most uncomplicated cases, it took 3 months for them to return whether the operation was performed with the arthroscope or with a small anterior arthrotomy. Use of the arthroscope often was taking more than an hour and required an x-ray to be certain that I had not missed anything, whereas the arthrotomy took 30 to 40 minutes and rarely required intraoperative imaging. I thought that I was doing a better job, with less surgical time, with the open technique. I have therefore gone back to a small anterior arthrotomy on these complex cases, with the use of a headlamp and Army-Navy retractors to allow adequate visualization.



Multiple anterior osteophytes can be present secondary to frank osteoarthritis of the ankle. Anterior clean-out in these cases is of questionable effectiveness and probably should not be performed for this condition.

Anterolateral ankle pain usually is not due to bony impingement because the tibia and talus do not come together in this location. Difficulties in this area usually are due to one of two conditions: Basset's ligament and Ferkel's phenomenon.

Basset's ligament[7] is an abnormal distal slip of the anterior tibiofibular ligament extending so far distally on the lateral malleolus that the lateral shoulder of the talus impinges against it when the ankle is plantarflexed ( Fig. 2-9 ). It is difficult to diagnose but, when present, can be resected with the arthroscope.


Figure 2-9  Basset's ligament.  Arthroscopic view seen from the medial portal.


Ferkel's phenomenon[8] is an accumulation of scar tissue and synovitis in the anterolateral gutter of the ankle, usually following trauma. It causes symptoms similar to Basset's ligament and also is amenable to arthroscopic debridement.

Anterior syndesmosis pathology usually is not the result of impingement but can cause anterolateral ankle pain that is exacerbated by dorsiflexion of the ankle because the wide portion of the talus spreads the malleoli and places tension on the anterior tibiofibular ligament. There are three types of pathology:



A sprain of the syndesmosis, the “high ankle sprain” sometimes can take an extraordinarily long time to heal.



The Tillaux fracture is an avulsion fracture of the insertion of the distal tibiofibular ligament into the tibia. On rare occasions, the avulsion can occur on the fibular side.



Synovial hernias into rents in the tibiofibular ligament have been described.[9]

The lateral ankle can be a complex site of pain and discomfort, and an accurate diagnosis in this area can be difficult. Symptoms in this area often have their onset following ankle sprains. The original trauma often can be mild—for example, a first-degree sprain with no resultant lateral instability. Cuboid subluxation and the sinus tarsi syndrome were discussed earlier. Other conditions to consider are as follows:



The “meniscoid” of the ankle[10] is thought to be soft tissue trapped between the lateral shoulder of the talus and the lateral malleolus. McCarroll et al.[11] described this lesion in four soccer players who had a history of frequent ankle sprains and, after failing conservative treatment, underwent arthroscopic debridement of the lesion. After appropriate rehabilitation, all four returned to competition with cessation of symptoms, with one player having only rare pain.



An avulsion fracture of the anterior process of the os calcis[12] is not an impingement. It is an avulsion fracture of the origin of the EDB and EHB. It usually can be seen on an x-ray ( Fig. 2-10 ) and diagnosed on physical examination by the specific tenderness at the anatomic site or pain with pronation-supination of the forefoot. Persistent symptoms may warrant excision of the fragment (see Fig. 2-10 ).



Fracture of the lateral process of the talus[13] also can cause impingement beneath the lateral malleolus. This fracture has been labeled the “snowboarder's fracture” for its increased incidence in this patient population. A high index of suspicion is required to identify this injury, often misdiagnosed as an ankle sprain, because routine plain radiographs often are read as normal. CT is the study of choice for diagnosis. Surgical treatment options range from excision to open reduction internal fixation (ORIF), depending on the size of the fragment.[14] Talus pathology is covered in more detail in Chapter 14 .



An accessory ossicle, the os subfibulare, which had been asymptomatic, can be loosened by injury and become symptomatic.



Avulsion fractures of the tip of the fibula can become trapped in or under the lateral ankle joint and become symptomatic. The bony fragment often is in the insertion of the calcaneofibular ligament. If it is small, it should be excised and the stump of the ligament sutured into the tip of the lateral malleolus. If it is large, it often can be reattached with a screw or K-wire. Infrequently the same situation can be found at the anterior edge of the lateral malleolus at the insertion of the anterior talofibular ligament (ATFL) ( Fig. 2-11 ).



Figure 2-10  Fracture of the anterior process of the os calcis.




Figure 2-11  Fracture of the tip of the fibula trapped under the lateral malleolus.



Impingement under the tip of the fibula following os calcis fractures is a common complication of this injury. Often it is difficult to differentiate the impingement pain from subtalar joint pain. A small injection of a local anesthetic beneath the tip of the lateral malleolus, but not into the subtalar joint, can indicate how much of the pain is coming from the impingement, versus an arthritic subtalar joint. If the pain relief with the local anesthetic is dramatic, it might be worthwhile to excise this portion of the os calcis before recommending subtalar arthrodesis ( Fig. 2-12 ). For a more detailed discussion of calcaneal fractures, refer to Chapter 5 .


Figure 2-12  Impingement under the tip of the fibula, following a fracture of the os calcis and lateral malleolus.



Peroneal dysfunction, although not an impingement, also can produce pain in this area and should be considered in the differential diagnosis. This includes peroneal subluxation, longitudinal splits in the tendons,[15] and even fracture of the os perineum with retraction of the peroneus longus[16] ( Fig. 2-13 ).


Figure 2-13  Retraction of the os peroneum (arrow), following rupture of the peroneus longus tendon.



Posterior ankle pain is common in athletes such as dancers, gymnasts, soccer players, and skaters who must work or kick in the equines position. A review of posterior ankle anatomy will help to explain the two common pain syndromes found in this area (Tables 2-2 and 2-3 [0020] [0030]). The posterior aspect of the talus has two tubercles: the medial and the lateral ( Fig. 2-14 ). The lateral tubercle is the origin of the posterior talofibular ligament. The tubercle can be small or large. When it is large it is referred to as the posterior process of the talus or Stieda's process. In 7% to 11% of people, this posterior process is separate from the talus and connected by a fibrous synostosis; then it is called the os trigonum (OT).[17] The OT is the second most common accessory bone in the foot, the accessory navicular being the most common.[18] Bony impingement can occur posterolaterally when the trigonal process or OT is compressed between the posterior lip of the tibia and superior portion of the os calcis in extreme plantarflexion ( Fig. 2-15 ).

Table 2-2   -- FHL tendonitis versus posterior impingement of the ankle

FHL tendonitis

Posterior impingement



Tenderness over FHL tendon

Tenderness behind fibula

Pain or triggering with motion of the hallux

Pain with plantar flexion of the ankle

± Thomasen's sign[14]

Plantar flexion sign

Mistaken for PT tendonitis

Mistaken for peroneal tendonitis

FHL, flexor hallucis longus; PT, posterior tibial.




Table 2-3   -- Medial versus lateral posterior ankle pain in athletes and dancers



FHL tendinitis

Posterior impingement (OT syndrome)

Soleus syndrome

Fx. trigonal process (Shepherd's fracture)

PT tendonitis

Peroneal tendonitis

Posteromedial fibrous tarsal coalition

Pseudomeniscus syndrome

FHL, flexor hallucis longus; Fx., fracture; OT, os trigonum; PT, posterior tibial.





Figure 2-14  Anatomy of the posterior talus.




Figure 2-15  Posterior impingement on the os trigonum.



The flexor hallucis longus (FHL) tendon passes through the fibro-osseous tunnel between the two posterior tubercles as it runs from its origin on the fibula (laterally) to its insertion in the distal phalanx of the hallux (medially) ( Fig. 2-16 ). Chronic tendinitis and dysfunction within this tunnel can produce posterior medial pain, “dancer's tendinitis.” [0190] [0200] [0210] [0220] Thus there usually are two sources of posterior ankle pain: lateral (trigonal impingement) and medial (FHL tendinitis), or a combination of the two.


Figure 2-16  Medial anatomy of the flexor hallucis longus.



Posterolateral ankle pain

The posterior impingement syndrome of the ankle, or talar compression syndrome, [0210] [0230] [0240] is the natural consequence of full weight bearing in maximal plantarflexion of the ankle in the demi-pointe or full pointe position, especially if an OT is present. It presents as posterolateral pain in the back of the ankle when the posterior lip of the tibia closes against the superior border of the os calcis. It can be confirmed on physical examination by tenderness behind the peroneal tendons in the back of the lateral malleolus (it often is mistaken for peroneal tendinitis) and pain with forced passive plantarflexion of the ankle, the “plantarflexion sign.”

This syndrome is often, but not always, associated with an OT or trigonal process in the back of the ankle. As previously noted, the posterior aspect of the talus normally has two tubercles: the medial and the lateral. Between the two lie the fibro-osseous tunnel and the FHL tendon ( Fig. 2-17 ). Most people who have an OT are not aware of its presence, and the posterior impingement syndrome is rare in most athletes. In dancers it may or may not be symptomatic, and the degree of symptoms is not always related to its size. Large OTs can be minimally symptomatic and small ones sometimes can be disabling. Usually the symptoms are mild and, on the whole, the OT often is more asymptomatic than symptomatic. Many world-famous ballerinas have asymptomatic OTs, and they work with them without any trouble. It is important to stress this fact to the dancer when discussing the problem, because the condition often is overdiagnosed by paramedical practitioners, who may recommend surgery unnecessarily, perhaps because of the dramatic appearance of the bone on x-ray. It is seen best on a lateral view of the ankle en pointe or in full planter flexion ( Fig. 2-15 ). The diagnosis can be confirmed, if necessary, by injecting 0.5ml of a local anesthetic into the posterior soft tissues behind the peroneal tendons. If the pain that was present is relieved by this small injection, the diagnosis is almost certain.


Figure 2-17  Superior view of relationship of flexor hallucis longus to os trigonum.



Treatment of the posterior impingement syndrome should be graded. The first step, similar to the treatment for tendinitis, is modification of activities (“Don't do what hurts!”), nonsteroidal anti-inflammatory drugs (NSAIDs—if the dancer is older than age 16), and physical therapy. In cases in which this approach has failed, or the symptoms recur, an injection of 0.25 to 0.5ml of a mixture of a long-acting and a short-acting corticosteroid often can give dramatic and permanent relief of symptoms. This can be done accurately with the use of sonography. Before injecting the steroid preparation, the clinician should confirm the diagnosis with a local anesthetic. If the local anesthetic does not relieve the symptoms, there is no point in injecting the steroids. It should be stressed that the OT usually is not a surgical problem; most dancers with an OT do not need to have it removed surgically.

Occasionally, the OT does cause enough disability to warrant surgical excision, but, as with most elective surgery, it is indicated only after the failure of conservative treatment in a serious dancer at least 16 years of age or older. If the problem is an isolated OT with no medial symptoms, it can be approached posterolaterally between the FHL and the peroneal tendons (with the sural nerve protected). Not infrequently, there is a combined problem of FHL tendinitis and posterior impingement. The posteromedial approach is used in these patients so that the neurovascular bundle can be isolated and protected. A tenolysis of the FHL and removal of the adjacent OT then can be performed safely.

Other causes of posterolateral ankle pain include the following:



A previously asymptomatic OT may become persistently symptomatic following an ankle sprain, resulting from disruption of its ligamentous connections and a subtle shift in position.



Posterior impingement can follow an ankle sprain that stretches out the lateral ligaments that hold the talus under the tibia in the relevé.[22] As the talus slips forward, the posterior lip of the tibia comes to rest on the os calcis. The treatment for this type of posterior impingement is to tighten the lateral ankle ligaments (preferably by the Broström-Gould procedure).[25] If the drawer sign can be corrected, the posterior impingement usually disappears.



A posterior pseudomeniscus or plica in posterior ankle,[22] with or without an OT, can cause the posterior impingement syndrome in the absence of an OT or ligament laxity. We have seen bucket-handle tears in this structure causing locking and other mechanical symptoms that are seen more often in the knee than the ankle.

Excision of the OT using the lateral approach

Under anesthesia, the patient is placed in the lateral decubitus or prone position with a pneumatic tourniquet on the leg or thigh over cast padding. Because dancers have increased external rotation of the hip, it is extremely difficult to perform this operation with the patient supine. A curvilinear incision is made at the level of the posterior ankle mortise in line with the posterior border of the peroneal tendon sheath. The sural nerve is identified or carefully avoided in the subcutaneous tissues. The dissection is carried down to the interval between the peroneal tendons laterally and the muscle belly of the FHL medially. A posterior capsular incision then is made with the ankle in neutral or slight dorsiflexion. The OT or trigonal process (a Stieda process) can be found on the superior surface of the posterior talus, just on the lateral side of the FHL tendon, between the ankle and subtalar joints. It has attachments on all its sides: (1) superior—the posterior capsule of the talocrural joint; (2) inferior—the posterior talocalcaneal ligament, at times thick and fibrous; (3) medial—the FHL tunnel with its sheath; and (4) lateral—the origin of the posterior talofibular ligament.

The bone can be removed by circumferential dissection. One should be careful not to stray too far medially—the posterior tibial nerve rests on the FHL tunnel. The proximal entrance of the FHL tunnel can be opened if there are muscle fibers attaching distally on the FHL tendon that crowd into the tunnel when the hallux is brought into dorsiflexion (see Tomasen's sign[10]). One should not dissect medial to the FHL tendon without adequate visualization; the posterior tibial nerve lies directly medial to the FHL tendon. The surgeon should check thoroughly for loose bodies; I have found them even in the FHL tunnel. The foot should be brought into maximal plantarflexion to look for any residual impingement. At times it is necessary to remove more of the posterior lateral tubercle. Often there is a facet on the cephalad portion of the os calcis that articulated with the OT, and this can be large enough to impinge against the posterior lip of the tibia after the OT has been removed. Careful hemostasis will prevent a postoperative hematoma, which can delay recovery and make early motion difficult for the patient. A layered closure then is performed with plain catgut suture with the ankle in neutral dorsiflexion. We usually close the wound with a running absorbable suture and Steri-Strips. The patient is placed in a bulky dressing, and weight bearing with crutches is begun as tolerated. Early motion is essential to prevent fibrosis and resultant limited range of motion. The dancer is encouraged to swim and progress to barre exercises as discomfort subsides. Average return to full dancing is 2 to 3 months.

Posteromedial ankle pain

Tendinitis of the FHL tendon behind the medial malleolus of the ankle is so common in dancers that it is known as dancer's tendinitis. [0190] [0200] [0210] [0220] [0260] The FHL is the “Achilles tendon of the foot” for the dancer. It passes through a fibro-osseous tunnel behind the talus like a rope through a pulley. As it passes through this pulley, it is easily strained. When this occurs, rather than moving smoothly in the pulley, it begins to bind. This binding causes irritation and swelling, which, in turn, causes further binding, irritation and swelling—setting up the familiar cycle: because it is swollen and irritated, it binds; and because it binds, it is swollen and irritated. If a nodule or partial tear is present, triggering of the big toe may occur. This is known as hallux saltans ( Fig. 2-18 ). At the extreme, the tendon may become completely frozen in the sheath, causing pseudo hallux rigidus. This tendinitis typically responds to the usual conservative measures. Rest is an important component of the therapy so that the chronic cycle previously described can be broken. NSAIDs can help, but they should be used only as part of an overall treatment program and not as medicine to kill the pain so that the patient can continue dancing and ignore the symptoms. As with other tendon problems, steroid injections should be avoided in the office setting because of the danger of injecting the steroid into the tendon. However, diagnostic and therapeutic injections can be performed into the FHL tendon sheath more accurately and safely under sonographic control. On some occasions, in professional or high-level amateur dancers and athletes, FHL tendinitis may be recurrent and disabling. In refractory cases, operative tenolysis may be indicated, but only after failure of conservative therapy. The situation is similar to de Quervain's stenosing tenosynovitis in the wrist.


Figure 2-18  A nodule in the flexor hallucis longus (FHL) tendon causing triggering of the great toe; “hallux saltans.”



FHL tendinitis usually occurs behind the medial malleolus, but occasionally it can be found at the knot of Henry under the base of the first metatarsal where the flexor digitorum longus (FDL) crosses over the FHL, and under the head of the first metatarsal where it passes between the sesamoids. A fibrous subtalar coalition may be present in the posteromedial ankle, mimicking FHL tendinitis or tarsal tunnel syndrome. This condition should be suspected when there is less than normal subtalar motion on physical examination.

The differential diagnosis of posterior ankle pain includes the following:



Posterior process (Shepherd's)[27] fracture; nondisplaced or stress.



FHL tendinitis (dancer's tendinitis).



Peroneal tendinitis.



Posteromedial localized talocalcaneal coalition.



Osteoid osteoma.

Operative treatment is indicated when conservative therapy has failed. The posterior clean-out can be performed from either the medial or lateral side of the Achilles tendon. The lateral approach should be used if the patient has an isolated posterior impingement without a history of FHL tendinitis or medial difficulties. A medial incision in indicated if the patient has a combined problem of FHL tendinitis and posterior impingement, or if he or she primarily has FHL tendinitis with an incidental OT that the surgeon wishes to remove along with an FHL tenolysis. The medial incision is safer and more utilitarian because one can work safely on the lateral side from the medial approach, but it is dangerous to work medially from the lateral approach because the neurovascular bundle cannot be isolated and protected from the lateral side.

Tenolysis of the FHL and excision of the OT from the medial approach

This procedure can be performed with the patient supine because dancers usually have increased external rotation of the hip and knee that allows easy visualization of the posterior ankle from the medial side. A bloodless field is desirable, so we use a tourniquet on the thigh over cast padding. For this reason, the procedure cannot be performed with the patient under local anesthesia or ankle block. A curvilinear incision is made over the neurovascular bundle behind the medial malleolus beginning just above the superior border of the os calcis and continuing to a line just posterior to the tip of the medial malleolus ( Fig. 2-19, A ). This incision should be made carefully. The deep fascia and lacinate ligament in this area often are thin. If the incision is made too enthusiastically, the surgeon may find himself or herself in the midst of the neurovascular bundle before he or she had planned to be there. The deep fascia then is divided carefully to avoid damage to the artery and nerve beneath it.


Figure 2-19  (A) Posteromedial incision. (B) Neurovascular bundle beneath a thin layer of fascia. (C) Neurovascular bundle taken down from the posterior medial malleolus. (D) Posterior tibial nerve protected with a blunt retractor. Underneath lies the flexor hallucis longus (FHL) sheath. (E) FHL sheath opened.



At this point one must decide whether to go in front of the bundle or behind it. The posterior approach can take the surgeon into the variable neural branches to the os calcis. It is safer to go anteriorly to the bundle. All branches of the tibial nerve at this level go posteriorly, thus the safe plane is between the posterior aspect of the medial malleolus and the neurovascular bundle. The bundle can be taken down off the malleolus by blunt dissection ( Fig. 2-19, B ). Often there are several small vessels here that need to be ligated, but once the bundle is mobilized it can be held with a blunt retractor such as a loop or Army-Navy retractor (never with a rake) ( Fig. 2-19, C ). The posterior tibial nerve is much larger than one expects; it usually is about the size of a pencil ( Fig. 2-19, D ). The surgeon should examine the nerve and the artery and note where they each divide into medial and lateral plantar branches as they leave the tarsal canal. It is not unusual for either the artery or the nerve, or both, to divide above this area, leading to reduplication within the tunnel. There also may be reduplication of the tendons—the flexor hallucis accessorius. With the neurovascular bundle retracted posteriorly, the FHL is identified easily by moving the hallux ( Fig. 2-19, E ). The thin fascia overlying the muscle fibers of the FHL is opened proximally, and a tenolysis is performed by opening the sheath from proximal to distal. Usually it is stenotic and tough, and the FHL often can be seen entering it at an acute angle. Care should be taken distally because the FHL tunnel and the nerve are close together here. As the tenolysis approaches the area of the sustentaculum tali, the sheath thins so that there no longer seems to be anything to divide. The tendon should be retracted with a blunt retractor and inspected for nodules or longitudinal tears. If present, these should be carefully debrided or repaired. At this point the FHL can be retracted posteriorly with the neurovascular bundle. The OT or trigonal process will be found just on the lateral side of the FHL tunnel. If the posterior aspect of the talus cannot be visualized, a capsulotomy should be performed. If there is difficulty in visualizing the OT, it helps to identify the superior border of the os calcis and the subtalar joint (by moving the os calcis into adduction and abduction). The subtalar joint then is dissected from medial to lateral, and this will take the surgeon underneath the OT. Once identified, it can be removed by circumferential dissection. Care should be taken to stay on the bone when performing this part of the procedure. This can be somewhat difficult, especially if the OT is large. Once it is removed, the posterior ankle joint should be inspected for remnants, bone fragments or loose bodies, soft tissue entrapment, or a large articular facet on the upper surface of the os calcis that articulated with the OT. If this articulation is large, it may need to be removed with a thin osteotome. The FHL sheath is not closed. The wound then is irrigated, checked for any residual impingement by putting the foot in maximal plantarflexion, and closed in layers with plain catgut and with the ankle in neutral flexion. We begin weight bearing as tolerated with crutches as soon as possible and proceed with swimming and physical therapy when the wound is healed. If the tenolysis is performed without excision of the OT, the recovery period is about 6 weeks. If the OT is removed along with the tenolysis, the recovery time is 8 to 12 weeks. It is important to get patients moving early to prevent stiffness. In dancers with a rather large OT, it is necessary to warn them that, once it is removed, the ankle does not just drop down into maximal plantarflexion. They must realize that the bone has been there since they were born, and removing it does not lead to immediate motion. The increased plantarflexion is obtained slowly and can be accompanied by many strange symptoms, both anteriorly and posteriorly, as the soft tissues adjust to the new range of motion.

The medial ankle

Posterior tibial tendinitis, so common in athletes, is rare in dancers—an example of altered kinesiology producing altered patterns of injury. Working primarily in the equinus position produces less stress on the posterior tibial tendon but more on the FHL tendon as it passes through its pulley behind the medial malleolus. In addition, dancers are selected for, and usually have, cavus feet, which are less prone to posterior tibial (PT) tendinitis. Indeed, more often than not, a dancer diagnosed with PT tendinitis, on careful examination, will be found to have FHL tendinitis instead (dancer's tendinitis).

Medial sprains of the ankle are rare because the medial structures are strong and rigid in comparison with the lateral ones. Persistent symptoms on the medial side may be due to an unrecognized fracture of the sustentaculum tali, which can be picked up on a bone scan, or to a localized fibrous tarsal coalition. Sprains of the medial ankle do occur, usually from landing off balance with sudden pronation, but again, this is more likely to produce a sprain of a portion of the deltoid ligament than a strain of the PT tendon, although this can occur. The sprain usually affects the portion of the ligament under tension when the force was applied: the anterior deltoid if the foot was in equinus, the middle deltoid if the foot was plantigrade, and the posterior portion if the foot was in dorsiflexion (rare). An accessory bone, the os subtibiale, may be present in the deep layer of the deltoid; this bone can be involved in the sprain, becoming symptomatic when it had not been before. The treatment of these medial sprains and strains in the acute phase consists of the usual RICE regimen (i.e., rest, ice, compression, and elevation), an aircast stirrup brace, crutches if necessary, and physical therapy. An x-ray should be taken to rule out bone or physeal injury. Recovery usually is uneventful. Occasionally a trigger point can form in the deltoid, usually around a chip fracture or accessory ossicle. These may require a corticosteroid injection if they do not respond to conservative therapy. Only rarely will surgical excision be necessary. Nodules may form on the EDL or PT tendons following medial strains, but these usually are asymptomatic.

In dancers, the most common cause of pain around the medial malleolus comes from “rolling in” (pronating) to obtain proper turnout ( Table 2-4 ). This produces a chronic strain on the deltoid ligament, particularly the deep portion, and is one of many overuse syndromes seen in dancers.

Table 2-4   -- Differential diagnosis of medial ankle pain in athletes and dancers

Most common

PT tendonitis (athletes) FHL tendonitis (dancers)


Deltoid ligament strain


FDL tendonitis Soleus syndrome

FDL, flexor digitorum longus; FHL, flexor hallucis longus; PT, posterior tibial.




Contusion of the medial prominence of the tarsal navicular can occur. This usually happens when one foot is brought forward past the other and, as it passes the navicular, strikes the medial malleolus of the other ankle. These contusions usually heal with symptomatic treatment. On rare occasions, a fracture of the medial tubercle or disruption of an accessory navicular can occur. In this setting, the injury should be treated in a short-leg walking cast or cam walker boot for 4 to 6 weeks to prevent the injury from becoming chronic.

Strains of the spring ligament and plantar fascia can be mistaken for medial ankle pain, but a careful physical examination should make the diagnosis apparent.

A rare cause of medial ankle pain is an unrecognized fracture of the colliculus located on the medial portion of the posterior tibia. This occult injury can be difficult to diagnose. It usually can be documented by a bone scan and CT scan.

Another cause of medial pain just above the medial malleolus is the soleus syndrome.[28] This presents as chronic pain resembling a shin splint but is too far distal on the posteromedial tibial metaphysis to be a true shin splint. It is caused by an abnormal slip in the origin of the soleus muscle. The condition, similar to the exertional compartment syndrome, is much more common in athletes than dancers. It usually responds to conservative therapy, but, on rare occasions, release of the tight band may be necessary.

In summary, treatment of dancers can be as challenging as it is rewarding. Dancers often have unusual difficulties related to the altered kinesiology required by their individual dance form—ballet, modern dance, jazz, tap, ethnic, Broadway, and so forth. A thorough understanding of these movements will help guide the physician to the cause of the disability, particularly in the setting of overuse injuries. This knowledge, coupled with a careful physical examination, is essential for the accurate diagnosis and treatment of the dancer, who is both athlete and artist.



  1. Thompson FM, Hamilton WG: Problems of the second metatarso-phalangeal joint.  Orthopedics1987; 10:83.
  2. Phemister DB: Necrotic bone and the subsequent changes which it undergoes.  JAMA1915; 64:211.
  3. Mann RA, Coughlin MJ: Surgery of the foot and ankle,  ed 6. St Louis, Mosby Year Book, 1993.
  4. Taillard W, et al: The sinus tarsi syndrome.  Int Orthop1981; 5:117.
  5. Marshall PM, Hamilton WG: Cuboid subluxation in ballet dancers.  Am J Sports Med1992; 20:169.
  6. Scranton PE, McDermott JE: Anterior tibiotalar spurs: a comparison of open versus closed debridement.  Foot Ankle1992; 13:125.
  7. Bassett FH, et al: Talar impingement by the anteriorinferior tibiofibular ligament.  J Bone Joint Surg1990; 72-A:55.
  8. Ferkel RD, et al: Arthroscopic treatment of anteriorlateral impingement of the ankle.  Am J Sports Med1991; 19:440.
  9. McLaughlin HL: Trauma,  Philadelphia, WB Saunders, 1960.
  10. Tomasen E: Diseases and injuries of ballet dancers,  Denmark, Universitetsforlaget I.Arhus, 1982.
  11. McCarroll JR, et al: Meniscoid lesions of the ankle in soccer players.  Am J Sports Med1987; 15:255.
  12. Harburn T, Ross H: Avulsion fracture of the anterior calcaneal process.  Phys Sports Med1987; 15:
  13. Hawkins LG: Fractures of the lateral process of the talus.  J Bone Joint Surg1970; 52A:991.
  14. Valderrabano V, et al: Snowboarder's talus fracture: treatment outcome of 20 cases after 3.5 years.  Am J Sports Med2005; 33:871.
  15. Sammarco JG, DiRaimondo CV: Chronic peroneus brevis tendon lesions.  Foot Ankle1989; 9:163.
  16. Thompson FM, Patterson AH: Rupture of the peroneus longus tendon, report of three cases.  J Bone Joint Surg1989; 71-A:293.
  17. Grant JCB: A method of anatomy,  Baltimore, Williams & Wilkins, 1985.
  18. Sarrafian SK: Anatomy of the foot and ankle,  Philadelphia, JB Lippincott, 1983.
  19. Hamilton WG: “Dancer's tendonitis” of the FHL tendon,  Durango, CO, American Orthopedic Society for Sports Medicine, 1976.
  20. Hamilton WG: Tendonitis about the ankle joint in classical ballet dancers; “Dancer's tendonitis”.  J Sports Med1977; 5:84.
  21. Hamilton WG: Stenosing tenosynovitis of the flexor hallucis longus tendon and posterior impingement upon the os trigonum in ballet dancers.  Foot Ankle1982; 3:74.
  22. Hamilton WG: Foot and ankle injuries in dancers.   In: Yokum L, ed. Sports clinics of North America,  Philadelphia: Williams & Wilkins; 1988.
  23. Howse AJG: Posterior block of the ankle joint in dancers.  Foot Ankle1982; 3:81.
  24. Quirk R: The talar compression syndrome in dancers.  Foot Ankle1982; 3:65.
  25. Hamilton WG, Thompson FM, Snow SW: The modified Broström procedure for lateral ankle instability.  Foot Ankle1993; 14:1.
  26. Hamilton WG: Ballet.   In: Reider B, ed. The school-age athlete,  Philadelphia: WB Saunders; 1991.
  27. Shepherd FJ: A hitherto undescribed fracture of the astragalus.  J Anat Physiol1882; 17:79.
  28. Michael RH, Holder LE: The soleus syndrome.  Am J Sports Med1985; 13:87.