Disorders of the Foot

Etiology

An accessory navicular is considered to be a normal anatomic variant that may become symptomatic for a variety of reasons. Several reports describe autosomal dominant inheritance with incomplete penetrance. ^,

Pathology

In an examination of accessory naviculars removed from symptomatic patients, Grogan and associates found areas of microfracture of the cartilaginous synchondrosis, hemorrhage, acute and chronic inflammation within and around the synchondrosis, and cellular proliferation in tissues surrounding the fractures. All the accessory naviculars exhibited chronic inflammation indicative of chronic injury with a prolonged inflammatory response. All the fractures were partial separations, with no cases of complete separation from the primary navicular. These changes were considered to be the result of chronic stress, occurring from overuse.

Bareither and colleagues studied 38 cadaveric feet in a cadaver study of feet with a prominent navicular area ; half had an accessory navicular bone and half had a hypertrophic posterior tibial tendon. Some of the accessory navicular bones were true sesamoids, lying in the tendon before it split and separated from the navicular bone by a 3-mm distance. Most accessory naviculars were connected to the navicular by fibrous tissue and were within the main insertion of the tendon. Another cadaver study found that the posterior tibial tendon inserted directly into the accessory navicular without extending to the sole of the foot; the second part of the posterior tibial tendon extended from the accessory navicular to the normal plantar insertions. No connection was present between these two portions of the tendon, and when traction was placed on the proximal tendon, the distal portion showed no movement, suggesting that the presence of this anomaly would lead to a pronated foot.

A more recent analysis using MRI to assess the insertions of the posterior tibial tendon in cadaveric feet confirmed that in all feet with a type I accessory navicular, the tendon inserted directly into the accessory bone with a slip less than 1.5 mm in thickness extending to the medial aspect of the navicular.

The accessory navicular was thought to interfere with normal leverage of the tibialis posterior and result in a weak longitudinal arch and flatfoot ^, ; subsequent studies have shown no relationship between the two.

Clinical Features

Controversy exists regarding how often the entity is painful and how often its presence goes unnoticed. Many children have asymptomatic accessory navicular bones that may be noticed incidentally on clinical examination or on radiographs. In addition, the true native navicular extends well medially and toward the plantar surface of the foot, and a prominence in this area may often be due to pressure over this normally large bone.

A child with a symptomatic accessory navicular will have pain over an enlarged area at the medial aspect of the midfoot ( Figs. 19.10 and 19.11 ). The enlarged site is seldom larger than a centimeter in diameter and is generally somewhat smaller. This area, which is just at the insertion of the tibialis posterior tendon, is frequently callused or red. Tight-fitting shoes aggravate the pain, especially those worn for sports. The pain is alleviated by wearing less constraining footwear.

On examination, there will be some tenderness over the enlarged area. Skin irritation over the prominence is not uncommon. It may be possible to feel motion between a prominent accessory navicular and the primary navicular. Resisted inversion is sometimes painful, and there may be tenderness over the tibialis posterior tendon.

Radiographic Findings

An accessory navicular is best seen on an oblique radiograph directed medially to laterally (the external oblique view) (see Figs. 19.10 and 19.11 ). It may also be seen on a standard anteroposterior (AP) projection. The navicular is the last tarsal bone to ossify. Ossification occurs in girls between 1 and 3 years old and in boys between ages 3 and 5 years. ^, An accessory navicular ossifies at an even later age. Feet with accessory navicular bones have been found to have wider and more prominent native navicular bones with greater medial prominences than feet without accessory naviculars. Radiographic diagnosis of the condition is usually made in later childhood or adolescence.

CT can be useful to better delineate the extent of an accessory navicular but is rarely necessary for diagnosis or preoperative planning. Technetium bone scans may help identify symptomatic accessory naviculars, although some asymptomatic accessory naviculars will also have increased uptake of tracer ( Fig. 19.12 ). ^{, ,} In a study of patients with focal pain over the navicular, MRI showed edema of the marrow.

A bipartite navicular (which is an entity distinct from an accessory navicular; Fig. 19.13 ) appears on radiographs as a dorsally displaced, comma-shaped, separate segment of the navicular. Despite the dorsal displacement, the segment still articulates with the talus.

Treatment

Treatment varies from observation and symptomatic management to excision. A child with an accessory navicular initially should be treated with soft pads between the foot and the sole of the shoe and should avoid wearing tight, stiff shoes. Elevated arch pads are not beneficial for these patients because the pads may aggravate the pressure over the navicular. If these treatment measures fail and there is a planovalgus deformity of the foot, a valgus-correcting shoe insert may be effective. Such inserts relieve pressure over the navicular by inverting the patient’s heel during gait rather than by pushing up on the arch of the foot.

In more recalcitrant cases, the surgeon may consider injecting the joint between the accessory navicular and the primary navicular with steroids and an analgesic agent. Immobilization in a short-leg cast has also been recommended.

A number of surgical procedures have been used for this condition ( Video 19.2 ). One report cites good results in type II accessory naviculars with percutaneous drilling to achieve union between the accessory and primary navicular bones. Simple excision of the navicular by shelling it out of the substance of the posterior tibial tendon accompanied by anatomic repair of the tendon is the procedure of choice for patients in whom conservative therapy fails. We have not found it necessary to detach the tendon from its broad insertions. The incision extends along the medial surface of the foot, directly over the accessory navicular from the midtalus to the base of the first metatarsal. The subcutaneous tissue and the deep fascia are divided, and the wound margins are retracted to expose the posterior tibial tendon and the medial portion of the navicular. The posterior tibial tendon inserts into the tuberosity of the navicular and into the plantar surfaces of the three cuneiform bones, as well as into the bases of the second through fourth metatarsals and laterally into the cuboid. The accessory navicular is dissected free from the central portion of the posterior tibial tendon while the remainder of the tendon insertions are left intact. The accessory navicular is excised and the medial and plantar portions of the navicular are resected until it is flush with or slightly depressed relative to the adjacent talus and cuneiform. If the primary navicular is still prominent after excision of the accessory navicular, the surgeon should consider complete removal of the medial prominence of the true navicular. The wound is then closed and soft dressings applied. A short-leg walking cast may be used postoperatively for four to six weeks for ease of ambulation and to allow healing of the posterior tibial tendon to the native navicular.

Other authors prefer the Kidner procedure, ^{, , ,} which entails rerouting the central slip of the tibialis posterior laterally onto the plantar surface of the navicular where it is sutured under tension to the local ligaments or to the navicular itself using suture anchors. A short-leg walking cast is then generally applied for 3 to 4 weeks postoperation.

Results and Complications

In many cases, patients obtain full relief of symptoms after simple excision of an accessory navicular (>90%) and after the Kidner procedure (96%). ^{, , , ,} Comparing the results of simple excision with the Kidner procedure, no differences in results or patient satisfaction have been identified and we therefore recommended the simpler approach. ^,

Occasionally, symptoms persist after surgical excision of the accessory navicular. If the primary navicular is prominent medially, pain and tenderness may continue over the area even though the accessory navicular has been removed. In other cases, the scar itself and the area beneath the scar remain tender despite adequate removal of bone. The cause of this tenderness is unclear, however. Over time the symptoms usually diminish, but they can be annoyingly persistent.

For References, see expertconsult.com .

Etiology

The cause of Köhler disease is not known. It has been suggested that because of its late ossification relative to the other tarsals, the navicular is vulnerable to mechanical compression injury. In a study of normal navicular ossification, Karp found that ossification occurs much earlier in girls than in boys. The navicular had ossified in half of the studied girls by 2 years old and in all girls by 3 years old. In half of the boys, the navicular did not ossify until 3 years old, and more than one- third were 3 years old before the nucleus ossified. In patients with slow ossification, the nucleus often appeared similar to Köhler disease. Karp thought that the delay in ossification in boys predisposed them to development of the disorder.

Another possible cause of the disease process is avascular necrosis (AVN) secondary to periodic compression of the bone. The abundant blood supply of the navicular would allow rapid, spontaneous healing, unlike AVN occurring in areas of marginal blood supply (e.g., the hip). Blood is supplied to the navicular bone by a dense perichondrial network of vessels on the nonarticular surfaces. As a child matures, one or more penetrating arteries appear, and ossification begins around these vessels. When more than one vessel penetrates, there will be more than one ossification center. At maturity, the navicular is supplied by five or six arteries, which anastomose with the bone. In rare cases a different pattern is found in which a single dorsal or plantar artery supplies most of the nutrition to the bone. The vascular etiology of Köhler disease is supported by biopsy studies showing areas of necrosis, resorption of dead bone, and formation of new bone. One author reported bilateral involvement in identical twins, suggesting a genetic etiology.

Clinical Features

The disorder occurs more often in boys than in girls, usually between 2 and 7 years old. The child will complain of pain in the midfoot and limping ( Fig. 19.14 ). Symptoms last from a few days to more than a year. There is no apparent relationship between the duration of symptoms and radiographic changes. In one study, a fifth of the patients had no symptoms, and the diagnosis of Köhler disease was made incidentally from radiographs. A small number of patients will have a distinct history of trauma.

Physical signs include tenderness, swelling, and sometimes redness over the dorsum of the foot (occasionally this clinical picture has been mistaken for an infection). The foot is generally held in pronation. Occasionally, however, it will be in supination as the child walks on the lateral side of the foot to relieve stress on the painful medial arch.

The natural history of the disorder is one of spontaneous resolution of the clinical symptoms and radiographic abnormalities over a period ranging from 18 months to 3 years. Persistence of symptoms into adulthood is extremely rare.

Radiographic Findings

The radiographic findings in Köhler disease are distinct ( Fig. 19.15 ). Often there is dense sclerosis of the navicular, with narrowing and apparent flattening of the bone (especially on a lateral projection). On an AP view, both sclerosis and lucency of the navicular are seen. These changes gradually disappear over a period of several years, with the radiographic appearance of the navicular ultimately returning to normal ( Fig. 19.16 ).

Treatment

Karp in 1937 found that the mainstay of treatment was restriction of weight bearing. Supportive measures, such as shoe inserts and casts, did not seem to affect the course of the disorder. Reports indicate that symptoms resolve faster in patients treated with walking casts than in patients who did not wear a cast. ^, Short-term cast treatment for 3 to 4 weeks is currently recommended for patients with persistent symptoms that limit their activities.

Outcomes

Köhler disease is a self-limiting disorder that in virtually all cases resolves over time. Radiographic changes return to normal, and persistence into adulthood does not occur ( Fig. 19.17 ). In a 31-year follow-up study, Borges and co-workers found that only two patients had persistent symptoms after being treated by cast immobilization. One patient had a talocalcaneal coalition and the other had a large accessory navicular. The authors concluded that patients could expect full resolution and normal function with symptomatic treatment.

Etiology

The cause of the disorder is not known. It is commonly thought to be due to AVN of the metatarsal head, and the histologic findings resemble those of AVN of other bones. Repetitive stress on the metatarsal head, caused by microfracture secondary to abnormal stress on the metatarsal head, trauma, and abnormal circulation have been proposed as etiologies. ^{, ,} Stanley and colleagues, however, found no evidence to support trauma as the cause and, through pedobarographic studies, found that pressure was not increased at the affected metatarsal head. Interestingly, 85% of the affected metatarsals were the longest in the foot, and the authors believed that this was an etiologic factor.

Clinical Features

The disorder appears most often in adolescents, usually after 13 years old. It occurs more frequently in girls than in boys. Pain under the second metatarsal head is the most common complaint, with resultant limping and a decrease in physical activities. An antalgic gait with poor push-off is generally present as well. Physical findings are normally limited to tenderness over the affected metatarsal head, with occasional swelling noted.

Radiographic Findings

Radiographs of the second metatarsal head show areas of lucency and collapse with flattening and loss of the normal shape of the condyles ( Fig. 19.18 ). This area shows increased uptake on technetium bone scans.

Treatment

An initial trial of conservative treatment is strongly recommended. A period in a hard-soled shoe or a trial in a short-leg walking cast will often relieve the symptoms. Subsequent use of a metatarsal pad in the shoe may reduce pressure on the metatarsal head.

Excision of the metatarsal head, which is recommended in very refractory cases, has been reported to relieve symptoms. Interpositional arthroplasty using the extensor digitorum brevis can be performed as an adjunct to metatarsal head excision. Tachdjian advocated curettage of the metatarsal head, with a cancellous bone graft placed in the cavity in the head. Dorsiflexion osteotomy of the metatarsal head, and metatarsophalangeal (MTP) joint débridement also have been shown to improve symptoms. ^{, ,} Shortening of the second metatarsal provided excellent relief; however, persistent stiffness was a problem. Long-term follow-up in patients managed with an extra-articular dorsal wedge closing osteotomy of the metatarsal has demonstrated high satisfaction rates, minimal pain, and excellent quality-of-life indices.

For References, see expertconsult.com .

Metatarsus Adductus

Occasionally, children are born with inward deviation of the forefoot relative to the hindfoot. The deformity may be very mild and resolve spontaneously, it may be slightly fixed and persist to walking age, or it may be rigid and associated with valgus of the hindfoot ( Fig. 19.19 ). Milder deformities require only parental reassurance. Moderate deformities respond well to manipulation and casting. Those rare cases with severe, rigid deformation may require surgical correction.

The terminology of these conditions is confusing because different authors use various terms, such as metatarsus adductus , hook-foot , bean-shaped foot , serpentine foot , and congenital metatarsus adductus , with little agreement among authors (see Fig. 19.19 ).

For simplicity, we will call most of these abnormalities metatarsus adductus and qualify them as actively correctable, passively correctable, or rigid. In this text the term skewfoot will be reserved for a foot with fixed adductus of the forefoot, increased valgus of the hindfoot, and lateral subluxation of the navicular on the talus.

Clinical Features

The deformity is usually noted at birth, but it may initially be recognized at any age. The clinical hallmark of the condition is medial deviation of the forefoot relative to the hindfoot. When the foot is viewed from the dorsal surface, the entire foot often appears to be turned inward. When the foot is viewed from the plantar surface, the sole of the foot has the shape of a bean. The base of the fifth metatarsal is usually prominent and the arch frequently appears higher than normal. In addition, the space between the first and second toes is wider than normal, and the first toe seems to be reaching medially. Kite listed six characteristic features of the disorder ( Box 19.1 ).

Box 19.1

Characteristic Features of Metatarsus Adductus

Data from Kite JH. Congenital metatarsus varus. J Bone Joint Surg Am . 1967;49:388–397.

• Spontaneous active medial deviation of the foot by the child

• High arch

• Concave medial border of the foot

• Separation of the first and second toes

• Fixed adductus of the forefoot when the hindfoot is held in neutral

• Bean-shaped sole of the foot

The full extent of the deformity can be best appreciated when the examiner grasps the heel and compares the alignment of the heel with that of the forefoot. Although it is not seen on a casual examination, careful evaluation of the hindfoot reveals slight valgus of the heel. There is always full range of ankle and subtalar motion.

It is important to establish the degree of flexibility of the deformity. In mild cases the foot will correct actively when the lateral border of the foot is stimulated. In less flexible cases the foot will not correct actively but can easily be corrected passively. The examiner should maintain the hindfoot in neutral during this maneuver, grasp the heel with one hand, and with the web space of the hand placed against the head of the first metatarsal, push the foot laterally. A gentle push will align the metatarsals in most children. A rigid deformity has a medial soft tissue crease at the tarsometatarsal level and a medial soft tissue contracture that prevents passive correction of the foot. As the forefoot is abducted, a tight abductor hallucis can be palpated medially ( Fig. 19.20 ).

Many of these children will also have internal tibial torsion, which will contribute to an intoed appearance. This is the most common parental complaint, and this component of the deformity should be noted separately.

Treatment

If the metatarsus adductus is flexible and actively corrects as the foot is stimulated, the condition does not need to be treated. These mild deformities will resolve gradually. Parents should be reassured and shown how to gently stretch the foot and how to stimulate it to actively correct.

There is controversy as to whether or not metatarsus adductus of intermediate severity, in which the deformity does not actively correct but is easily corrected passively, requires treatment. Many physicians believe that these deformities will self-correct over time without intervention. Rushforth found asymptomatic moderate deformity in 10% of children with untreated metatarsus adductus, and residual deformity and stiffness in 4% at 7-year follow-up. Some physicians advocate passive parental stretching of the foot by the child’s parent ( Fig. 19.21 ). Parents are taught to hold the heel in a neutral position with one hand and abduct the forefoot with the web space of the other hand. Some physicians also recommend that the child wear straight last shoes. For a child younger than 6 months, a short series of short-leg plaster casts will easily correct the foot position, and this treatment approach may be warranted for patients with obvious deformity. Some authors prefer long-leg bent-knee casts but reports indicate that only 10% of patients require such treatment.

The technique of cast correction is similar to the stretching procedure taught to the parent. The child’s heel is grasped and held in a neutral position while the forefoot is abducted. The thumb of the hand holding the heel should reach the cuboid so that the fulcrum for abducting the forefoot is at the level of the cuboid metatarsal joint. Eversion of the foot should be avoided. Before discontinuing cast treatment, the convexity of the lateral border of the foot should be straightened or reversed, the prominence of the base of the fifth metatarsal should no longer be noticeable, and active adduction should no longer occur when the child moves the foot.

Metatarsus adductus that cannot be passively corrected should be more vigorously treated. These feet usually have significant rigidity and a medial crease with an overactive abductor hallucis. Most such feet can be corrected with serial stretching and cast application, which should be started as soon as possible. If the foot is resistant to cast correction, surgical release of the abductor hallucis and capsulotomy of the first tarsometatarsal joint may be indicated, followed by further cast correction. ^, Good results have been reported in children (3.5 years old) at an average follow-up of 3.5 years.

The best way to manage older children with fixed adductus is also controversial. The Heyman-Herndon release of the tarsometatarsal joints has had mixed results; some authors have reported excellent outcomes, whereas others have had numerous failures ( Fig. 19.22 ). ^, We occasionally recommend this procedure for treating severe adductus in children too young to undergo osteotomies. Children with significant deformity who are older than 3 years may benefit from metatarsal osteotomies to realign the foot ( Video 19.3 ). Care must be taken, though, to avoid forcing the hindfoot into valgus and creating a skewfoot deformity. Napiontek and others reported good results with the use of an opening wedge osteotomy of the medial cuneiform in children younger than 4 years old. Their cases were mostly clubfoot residua, but four cases were primary adductus abnormalities. A percutaneous method of treatment that incorporates osteotomies of the central metatarsals and medial soft tissue releases has been described with favorable results.

Two other surgical procedures that have been recommended for treating metatarsus adductus are anteromedial release with capsulotomy of the medial metatarsocuneiform and naviculocuneiform joints and transfer of the posterior tibial tendon with capsulotomy of the naviculocuneiform joint. We have no experience with either of these approaches.

For References, see expertconsult.com .

Etiology and Clinical Features

This postural deformity of infancy is characterized by an oftentimes dramatic hyperdorsiflexion of the foot that appears to be “plastered” up against the anterior surface of the tibia ( Figs. 19.23 and 19.24 ). Plantar flexion of the foot is frequently limited as a result of contracture of the anterior ankle and foot structures. This deformity results from abnormal in utero positioning of the foot, described as a “packaging” defect, in which all anatomic structures form normally but are deformed at the end of the pregnancy because of in utero crowding. Unlike a “manufacturing” or “parts” problem, such as congenital clubfoot, in which embryologic formation is defective, postural deformities such as calcaneovalgus resolve spontaneously in the vast majority of cases.

Not only is the foot hyperdorsiflexed, but the heel is also frequently in marked valgus, with the forefoot appearing abducted. The calcaneus is palpable in the heel pad and is noted to be in the dorsiflexed (“calcaneus”) position. This differentiates a calcaneovalgus foot from the more serious pathologic congenital vertical talus, where the heel is in equinus, giving the foot a “rocker bottom” appearance. The forefoot may be equally dorsiflexed in both conditions, but it is the heel equinus and rocker bottom deformity that distinctly differentiate congenital vertical talus from a calcaneovalgus foot. It is important to look for associated problems accompanying the packaging problem of the foot—in particular, hip dysplasia. Although no initial instability may be appreciated during the newborn hip examination, the association of hip dysplasia with calcaneovalgus seems somewhat stronger when a contralateral metatarsus adductus is present, giving the feet a “windblown” appearance. Although the association of hip dysplasia with metatarsus adductus is controversial, with some investigators reporting it as a close association ^{, ,} and others refuting such an association, the simultaneous occurrence of calcaneovalgus in one foot and metatarsus adductus in the other seems to heighten the likelihood of at least silent dysplasia.

Less subtle is the association of posteromedial bowing of the tibia with a calcaneovalgus foot. The deformation of the tibia is easily understood as a packaging defect associated with the hyperdorsiflexed foot (see Figs. 19.23 and 19.24 ). However, there is more to this “packaging” combination in that there is usually growth inhibition of the lower part of the leg as a result of the posteromedial bow, generally mandating some form of limb equalization procedure later in childhood and adolescence. Thus, as opposed to a pure packaging defect, which should resolve without sequelae, calcaneovalgus with posteromedial bow of the tibia represents a more serious defect in morphogenesis.

Perhaps most commonly associated with calcaneovalgus is external rotation of the lower extremities. This becomes most obvious as the child begins to take weight on the lower extremities, with the feet sometimes pointing outward nearly 90 degrees. The source of the foot position may be persistent eversion and external rotation from the calcaneovalgus, or it may be an external rotation contracture of infancy at the hip. The latter is destined to resolve spontaneously once the child begins walking, again with the caveat that silent hip dysplasia should be ruled out with an imaging study (plain radiography or ultrasound), especially if there is any suggestion of asymmetry in the hip examination or hip rotational profiles.

The incidence of talipes calcaneovalgus has been reported to be as high as 30% to 50% in newborns. This estimate is likely too high as this same investigator found an increased incidence of flatfoot on long-term follow-up of patients with calcaneovalgus in infancy. The study was skewed toward following patients with severe flatfoot. A more likely incidence is the 1 in 1000 live births reported by Wynne-Davies. The incidence, like that of congenital dislocation of the hip, is higher in first-born children (because of intrauterine crowding) and girls, and is associated with hip dysplasia, as mentioned earlier. Because a calcaneovalgus foot position is either relatively common or extremely common, depending on which investigator is believed, the role of the orthopaedist in assessing what may be a normal variant of foot position is to eliminate true pathologic foot conditions (congenital vertical talus), associated tibial anomalies (posteromedial bow of the tibia), and most importantly, associated hip dysplasia.

Treatment

The prognosis of calcaneovalgus foot is excellent. Only in the most severe cases, with marked restriction of plantar flexion and supination/inversion, is anything more than gentle stretching exercises by the parents necessary. Generally, the foot position normalizes within 3 to 6 months. In more severe cases there may be a role for corrective casting or splinting (or both) in association with stretching exercises. The clinical experience of several authors has identified a correlation between talipes calcaneovalgus and a symptomatic form of hypermobile pes planovalgus in an older child. ^{, , ,} There is thus little downside to the application of two or three sets of corrective casts in the newborn period when there is significant limitation of plantar flexion and inversion. Casting may then be followed by stretching exercises and an ankle-foot orthosis type of splint for a few additional months to ensure satisfactory foot position when the infant begins to pull to stand.

For References, see expertconsult.com .

Definition

Although the exact incidence of flatfoot in children is unknown, it is undoubtedly one of the most common “deformities” evaluated by pediatric orthopaedists. Whether or not flatfoot represents a true deformity is questionable. For example, Staheli and colleagues regarded flatfeet as “usual in infants, common in children, and within the normal range in adults” in assessing and documenting the spontaneous development of the longitudinal arch. Others have concurred with these observations. ^, It may be surprising, therefore, that flatfoot is evaluated and treated, often prophylactically, with significant fervor by certain nonorthopaedic branches of medicine, where the controversy surrounding the “deformity” has not been resolved. On the other hand, because concern about abnormal-looking feet will never appreciably decrease among parents and pediatricians, especially in feet with a more clinically severe appearance ( Fig. 19.25 ), children with flatfoot will continue to be referred to the pediatric orthopaedist for the treatment of pain, perceived disability, and abnormal shoe wear.

The heel shows excessive eversion during weight bearing, and the forefoot is usually abducted, producing a midfoot sag with lowering of the longitudinal arch (see Fig. 19.25 ), so that the talar head and navicular tuberosity appear to be in contact with the floor and to participate excessively in weight bearing. The medial column of the foot appears longer than the lateral column. The entire foot is often described as pronated, although this description is misleading because the forefoot is actually supinated in relation to the hindfoot, a fact that is most appreciated when the hindfoot is corrected operatively or stabilized manually during physical examination. ^, The relationship of the forefoot to the hindfoot may also be appreciated when one contemplates a cavovarus foot, the anatomic reverse of a flatfoot, in which the forefoot is pronated in relation to the hindfoot and an excessively high longitudinal arch is produced.

Clinicians might be tempted to use radiography as the defining diagnostic examination, with flatfoot being considered a foot with measurements greater than two standard deviations from the mean. Because flatfeet are relatively common and generally benign, radiographs to document this diagnosis are rarely obtained, thereby perpetuating the lack of a specific definition of flatfoot. A standing lateral radiograph allows measurement of the lateral talus–first metatarsal angle, or Meary angle ( Fig. 19.26 ). This angle is normally 0 degrees (a straight line). In a flexible flatfoot an apex-plantarward angle will be present. The normal range of this angle also varies with age, with spontaneous improvement in plantar sag seen until age 8 years. The location of the sag—talonavicular or naviculocuneiform joint—can be determined and may suggest the cause of an abnormal measurement (i.e., a tight heel cord producing a plantar flexed talus and talonavicular sag; Fig. 19.27 ). The degree of plantar flexion of the talus—the angle formed by the longitudinal axis of the talus and the horizontal—can also be measured (normal, 26.5 ± 5.3 degrees ) as can the calcaneal pitch angle, which is formed by the axis of the calcaneus and the horizontal.

Perhaps the most compelling reason to obtain radiographs in cases of flatfoot is to rule out causes of the “deformity” other than idiopathy. The differential diagnosis includes such bony abnormalities as tarsal coalition, congenital vertical talus (convex pes valgus), persistent talipes calcaneovalgus, an accessory navicular, and various arthritic and inflammatory conditions. Most of these conditions are diagnosed primarily from the history and physical examination findings, and in such situations radiographs should be used to confirm a suspected diagnosis.

Clinical Features

Regardless of the exact definition of flatfoot, it is estimated that a depressed longitudinal arch occurs in approximately 23% of the adult population. Of this population, approximately two thirds have a flexible, hypermobile flatfoot with normal or increased mobility of the subtalar complex and ankle joint. Approximately one fourth of flatfeet exhibit a contracture of the triceps surae associated with an otherwise typical hypermobile flatfoot, and this form of flatfoot is a known cause of disability in army recruits. ^, The remainder of flatfeet are characterized by more rigidity of the subtalar joint, typically seen with tarsal coalitions. Differentiating between flexible and more rigid deformities is therefore a wise first step in the assessment of all flatfoot deformities.

The height of the longitudinal arch is determined by bony structure and degree of ligamentous laxity. Electromyographic studies have documented that neither the intrinsic nor the extrinsic muscles of the foot have electrical activity during standing at rest. Therefore, the static structure of the longitudinal arch is independent of any muscular activity. Obviously, with walking or more vigorous activity, both sets of muscles are active, thus suggesting dynamic stabilization of the arch. However, flatfoot is a well-known sequela of a lacerated or insufficient tibialis posterior tendon, so there must be some contribution to static integrity of the arch from this muscle because its absence results in the deformity.

In the typical flexible flatfoot, the longitudinal arch reconstitutes when the foot is in a non–weight-bearing position. The arch should also reconstitute during active plantar flexion, such as when a patient is asked to stand on tiptoes ( Fig. 19.28 ). Inversion of the heels and arch reconstitution during toe standing are requisite examination findings for a diagnosis of flexible flatfoot. Because the arch reconstitutes during such active muscle function, it is tempting to prescribe muscle-strengthening exercises of the plantar flexors and invertors in a patient with a planovalgus deformity. Although Tachdjian in the second edition of this textbook denigrated such exercises, no scientific study has evaluated their effectiveness.

If flexibility of the hindfoot and arch reconstitution are not demonstrated on the tiptoe test, then other conditions must be considered, especially if there is a complaint of pain. The general neurologic assessment—observation of gait, coordination, and reflexes—will uncover neurologic or myopathic conditions associated with flatfeet in which the foot position may be due to weakness (poliomyelitis, peripheral neuropathy), weakness with Achilles tendon contracture (Duchenne muscular dystrophy), or spasticity with equinus (cerebral palsy). Abnormal hindfoot motion, especially if painful or rigid, suggests tarsal coalition or inflammatory arthritis. A rigid rocker bottom deformity with equinus suggests a congenital vertical talus. Specific areas of pain, such as the navicular, may point to an accessory navicular or osteochondritis. Radiographs or consultation with a neurologist or rheumatologist may be appropriate if an idiopathic, hypermobile flatfoot is ruled out by the history and physical examination.

Particular attention to the Achilles tendon is important because a contracture tends to make hypermobile flatfeet symptomatic. After tiptoeing to confirm subtalar flexibility, a child should be asked to walk on the heels. An Achilles tendon contracture will make this activity difficult. Passive dorsiflexion of the foot, with the heel locked in varus (inverted), will further demonstrate this contracture.

As with any lower extremity deformity, joint range of motion and torsional profile must be evaluated to assess for more proximal causes that may have encouraged development of the flatfoot.

Natural History

The arch is usually obscured in an infant’s foot because of subcutaneous fat, and spontaneous resolution of “fat foot” can be anticipated as the young child matures and such fat atrophies. Both footprint ^, and radiographic studies of the child’s foot demonstrate that the longitudinal arch develops during the first decade of life. As discussed, the lateral talus–first metatarsal angle demonstrates a decrease in the amount of plantar sag of the midfoot until 8 years old. Such improvement in the sag of the medial ray of the foot would suggest that ligamentous laxity in a toddler spontaneously resolves as the ligaments become more taut. This observation also leads to the overwhelming conclusion that prophylactic treatment of a typical flatfoot is unnecessary, with profound implications for the corrective shoe and insert-orthosis industry. Development of the arch is independent of the use of such external orthoses or the wearing of corrective shoes. Studies from countries where shoes are not worn at all tend to substantiate the opinion that symptomatic deformities do not develop with aging in flatfooted children. The classic study of 3600 army recruits by Harris and Beath documented that the presence or absence of a longitudinal arch did not correlate with disability and that a flatfoot was compatible with normal function unless an Achilles tendon contracture was present. ^, There has even been a suggestion that shoes may be detrimental to development of the longitudinal arch. ^, Indeed, controlled prospective randomized studies on the effect of shoe modifications and inserts on development of the arch have failed to demonstrate any effect. ^, The use of these devices for correction of such a “deformity” must be considered ineffective and probably unnecessary. It is therefore the orthopaedist’s primary role to educate parents seeking treatment of their child’s flatfeet that the condition is both benign and unaffected by prophylactic treatment with such devices.

The development of bunions has also been related to the presence of a flatfoot. However, in Kilmartin and Wallace’s longitudinal study of children with flatfeet, bunion development was independent of the longitudinal arch.

Treatment

Conservative Treatment

In the typical case of a hypermobile (postural) flatfoot, no treatment is indicated in an asymptomatic pediatric patient. Education and reassurance are the mainstays.

Orthoses and medial arch supports have traditionally been prescribed, even though there is no scientific evidence that such modifications are efficacious. ^, With the evidence that flatfooted army recruits and children who do not wear shoes have essentially normal function as adults, and with the lack of objective studies demonstrating a lasting change in the radiographic anatomy of the foot with the use of corrective devices, there is no medical indication for the treatment of asymptomatic flatfeet. In light of the not insubstantial cost of some of the custom-molded inserts and orthoses, there is little justification for prescribing such devices, and the tradition of prescribing special shoes or inserts for orthopaedic management of the child’s foot should be abandoned.

If an Achilles tendon contracture is present, stretching exercises—both manually by the parents and actively by the child, if old enough to cooperate—are an appropriate form of management. These children may have symptomatic calluses under the head of the plantar flexed talus associated with the Achilles tendon contracture. Emphasis should be placed on stretching the heel cord with the heel inverted and the knee straight, and in the case of an older child, exercises involving the use of an elastic Thera-Band and dorsiflexion stretching with the heel maintained on the ground (with the patient leaning forward while the hands are supported on the wall) are recommended ( Fig. 19.29 ).

In symptomatic patients, arch supports and orthoses may be of benefit. Typical symptoms include medial arch pain, callosities, and fatigue. Lateral ankle pain may occur as well due to impingement between the everted calcaneus and the distal fibula. In addition to gastrocsoleus stretching exercises, we have found that the footwear sold in sporting goods stores, especially that designed for running, is often more readily accepted for social reasons by children and adolescents than the more traditional devices placed inside shoes. Running shoes designed for a “pronated” or “hyperpronated” foot have significant heel and arch support built into the shoe itself, thus making prescription of additional orthoses superfluous. Because running shoes usually support the relaxed portion of the arch or hindfoot, the suggestion to use such footwear may be all that is necessary to resolve the problem.

In more recalcitrant cases, formal orthotic management with custom devices such as a University of California Biomechanics Laboratory (UCBL) insert can be attempted. Such an orthosis can acutely change the talonaviculocuneiform axis and improve calcaneal pitch ( Fig. 19.30 ); it has been reported to alleviate symptoms and improve shoe wear in symptomatic patients. ^, Acceptance of this more rigid device—the orthosis is made from a plaster cast of the patient’s foot and molded from rigid plastic to invert the valgus heel and support the arch—may be problematic in that the rigid orthosis can be somewhat uncomfortable, similar to the proverbial “rock in the shoe”; because there is no evidence that it has any lasting effect on flatfeet, its use should not be pursued if an initial prescription fails. We have used soft inserts (Plastazote; Fig. 19.31 ) in symptomatic patients who have rejected the UCBL type of device, with better acceptance and probably the same efficacy, and have prescribed a rigid formal orthosis much less over the past several years.

It should again be emphasized that no permanent change in foot anatomy or arch structure has been documented with the use of any orthosis or shoe modification.

Surgical Treatment

Indications

Surgical management of a true hypermobile flatfoot is reserved for a patient who has intractable symptoms unresponsive to shoe or orthotic modifications and who is unable to modify the activities that produce pain. Thus patients with talonavicular calluses, medial arch “strain,” or calcaneofibular abutment whose daily activities are limited by foot pain are the only true candidates for surgical management. Although surgery can alter the shape of the arch by reconstructing it with either soft tissue imbrication or bony procedures, with generally good short-term results, long-term evidence of continued foot health after such procedures is generally lacking.

Indeed, unsatisfactory results from surgery have been reported in 49% to 77% of longer-term studies, ^{, ,} and all longer-term follow-up studies of limited tarsal fusion procedures have described osteoarthritic changes at adjacent unfused joints. ^a

a References , , , , , , , , .

Because such changes are known to occur de novo in untreated tarsal coalitions, it should come as no surprise that creation of an iatrogenic coalition during reconstruction of the longitudinal arch by tarsal bone fusion, however limited, produces the same degenerative changes at the adjacent joints. Thus surgical correction of flatfoot must emphasize joint-sparing procedures, usually combining extraarticular osteotomy with soft tissue imbrication.

Arthroereisis

Arthroereisis of the subtalar joint, using a metal, silicone, or Silastic implant, has been reported as an alternative to more complex joint reconstruction. The rationale of the procedure is to limit the amount of valgus motion in the subtalar joint by using an interposition peg. Long-term results of this procedure are lacking, and because of potential complications of intraarticular placement of foreign material, especially in the normal cartilaginous surfaces of a child’s hindfoot, this procedure is not warranted given that the natural history of a flexible flatfoot is generally benign. ^{, , ,} Nevertheless, the use of silicone, Silastic, and metal spacers interposed in the subtalar joint is common in the podiatric literature. ^{, , ,} The potential for synovitis necessitating implant removal is real ( Fig. 19.32 ).

Heel Cord Lengthening

An Achilles tendon contracture should always be considered and treated during any surgery for flatfoot. If the patient has severe enough symptoms to warrant surgery, a heel cord lengthening or gastrocnemius recession should be part of a comprehensive procedure to reconstruct the longitudinal arch. The selection between these two options for lengthening of the triceps surae is usually based upon intra-operative Silfverskiöld testing, often following provisional correction of the hindfoot valgus deformity.

Subtalar Fusion

Subtalar fusion as a primary procedure for hypermobile flatfoot should probably be condemned. While there is no question that excessive heel valgus and restoration of the longitudinal arch can be achieved through this procedure, the sacrifice of subtalar motion for this purpose is too great a cost. The mechanics of the hindfoot are completely altered by subtalar fusion, and the mobility of the remaining midfoot joints—talonavicular, calcaneocuboid, and the entire midfoot complex—is irretrievably altered by subtalar fusion. The more extensive triple arthrodesis eliminates all hindfoot mobility, and although deformity is effectively corrected by such a procedure, it is again indicated only as a salvage procedure in a foot in which other surgical procedures have failed. Furthermore, it should only be considered after selective joint injection with local anesthetic has demonstrated that it is the mobility of the hindfoot-midfoot complex that is producing the pain and disability ( Fig. 19.33 ).

Lateral Column Lengthening

Lateral column lengthening by insertion of a bone graft into an osteotomy of the calcaneal neck is currently the most attractive procedure to correct a flatfoot deformity and not sacrifice joint motion. ^{, ,} The lateral column is lengthened by inserting a trapezoid-shaped tricortical iliac crest allograft between the anterior and middle facets of the calcaneus ( Fig. 19.34 ). A transverse osteotomy of the neck of the calcaneus, approximately 1.5 to 2 cm proximal to the calcaneocuboid joint, is gently spread apart to receive a graft of the same length. Prior to spreading the osteotomy site, it is critical to stabilize the calcaneocuboid articulation with a percutaneous, centrally placed Steinmann pin to prevent subluxation. In addition, we typically lengthen the peroneus brevis to decrease tension across the osteotomy site as lengthening occurs. The technique of spreading the osteotomy is crucial to success of the procedure. Forceful spreading, with a lamina spreader for example, can crush the cortical edges of the two sides of the osteotomy and make it difficult to prop the osteotomy open. One technique is to place threaded Steinmann pins transversely into each osteotomy segment so that the osteotomy can be opened using the pins as handles. This technique has the advantage of not obstructing the osteotomy site during insertion of the graft. Once the graft has been properly impacted into place, internal fixation with screws or a staple can be selected though we prefer to simply advance the Steinmann pin which can be cut outside the skin for later removal in clinic (see Fig. 19.33 ). Soft tissue tensioning will usually hold the graft in place and would obviate the need for pin stabilization if it were not important to prevent joint subluxation. Calcaneocuboid subluxation should be avoided with longitudinal pin insertion though no disability has been identified in patients who have healed with this malunion.

Postoperatively, short-leg cast immobilization is maintained for 8 to 10 weeks to ensure healing of the osteotomy. Results of calcaneal lengthening have been considered satisfactory when there is relief of medial arch pain, resolution of calluses, correction of heel valgus, improvement in the appearance of the arch, radiographic restoration of the Meary angle and the lateral talocalcaneal angle, and maintenance of subtalar motion. This outcome is achieved in more than 90% of patients observed in short-term reviews, ^, and 75% of patients show no evidence of degenerative changes in longer-term follow-up.

Imbrication of Talonaviculocuneiform Complex

Imbrication of the talonaviculocuneiform complex medially is performed in combination with calcaneal lengthening. It is not recommended as a single procedure because of progressive stretching of the medial repair with weight bearing, especially when the lateral column has not been lengthened. The technique has evolved from limited fusions of the talonavicular and naviculocuneiform joints with tendon imbrication, to “tightening” of the naviculocuneiform joint with plantar imbrication, to opening wedge osteotomy of the cuneiform to re-create the arch. Combined use of a lateral column lengthening and medial opening wedge osteotomy has been shown in a cadaveric model to result in greater deformity correction and significantly diminished pressure under the lateral forefoot.

One technique involves initial detachment and later imbrication of the tibialis posterior tendon and raising of an osteoperiosteal flap of the cuneiform-navicular capsules by sharply dissecting a tongue of the medial capsules from proximal talonavicular to distal and leaving the flap attached at the cuneiform ( Fig. 19.35 ). ^, The talonaviculocuneiform alignment is corrected (usually after lateral column lengthening); the osteoperiosteal flap is advanced proximally and plantarward, and is reattached to the talar neck with heavy suture. We usually protect the medial reconstruction with a smooth K-wire, which is removed at the time of cast removal. As mentioned, the tibialis posterior should be shortened and advanced to restore appropriate tension after “shortening” of the medial column by soft tissue imbrication.

Our preferred technique involves cutting the tibialis posterior tendon in a z-type fashion without detachment from its insertion distally. A segment of the oftentimes redundant medial capsule of the talonavicular joint is then removed and the remaining capsule is imbricated plantarward. Lastly, the posterior tibial tendon is repaired in a shortened position. Postoperative care of the soft tissue generally requires an additional 4 to 6 weeks of casting to allow complete healing of the repair. Although excellent results have been reported with this medial reconstruction alone, ^, we continue to use it only in conjunction with a calcaneal lengthening osteotomy.

Medial imbrication can also be combined with a sliding calcaneal osteotomy. ^{, ,} Although one may argue that displacement of the posterior half of the calcaneus medially, to reestablish the weight-bearing line in the center of the ankle-subtalar coronal plane, merely creates a compensatory varusization for talocalcaneal valgus, the effect of such a shift seems to be helpful in supporting the plantar flexed talus and decreasing overall eversion and midfoot abduction ( Fig. 19.36 ). Thus, when combined with medial column shortening–imbrication, good results are achieved ( Fig. 19.37 ). Significant improvement in both pain measures and radiographic parameters has been reported in adolescents undergoing this combined correction. The need for Achilles tendon lengthening, peroneal lengthening, and corrective metatarsal or medial cuneiform osteotomy if forefoot supination is excessive after a calcaneal osteotomy with medial imbrication must be assessed during the combined procedures. In particular, careful attention must be paid to a forefoot supination deformity which is often unmasked once hindfoot correction is achieved. If the first ray is elevated on simulated weight bearing intra-operatively, a medial cuneiform osteotomy should be performed to plantarflex the first ray.

Complications of the combined procedure include nonunion of the calcaneal graft or displacement of the graft requiring revision, displacement of the calcaneocuboid joint, recurrence of the deformity, or pain that develops with time or prolonged weight bearing. Nonunion may be particularly difficult to treat because it is frequently accompanied by calcaneocuboid joint degeneration with subsequent pain. A painful arthrosis of this nature often mandates treatment via a triple arthrodesis. It goes without saying that the need for a salvage fusion procedure is an extremely untoward outcome following a “joint-sparing” operation. It cannot be overemphasized, therefore, that attention to sizing and stability of the graft is essential to avoiding nonunion.

Summary

In summary, a hypermobile flatfoot is a normal variant of foot structure and does not require prophylactic treatment. Should an Achilles tendon contracture accompany the deformity, vigorous stretching of the heel cord should be instituted to lessen the possibility of symptoms later in life. Nonoperative management of painful flatfeet in adolescents is generally successful and entails shoe modifications (running shoes suffice for this purpose), orthoses, and strengthening exercises. Surgical correction, truly a last resort for this normal variant, should emphasize joint-sparing procedures, including lateral column lengthening or a calcaneal medial sliding osteotomy, often combined with heel cord lengthening, medial soft tissue imbrication, and possible medial cuneiform osteotomy to provide symptomatic relief by realigning the talonaviculocuneiform complex and improving the hindfoot valgus deformity.

For References, see expertconsult.com .

Incidence and Natural History

By all reports, skewfoot is a rare deformity; Peterson found only 50 cases and Kite just 12 in 2818 cases of forefoot adduction. ^{, ,} The first accurate description in the American literature was published in 1933.

The natural history of skewfoot is unknown. In some patients the deformity undoubtedly resolves, just as metatarsus adductus and flexible pes planovalgus can resolve. On the other hand, with more rigid feet, calluses on each side of the foot, over the talar head, and over the cuboid–fifth metatarsal articulation, children may become symptomatic. ^{, , , ,}

Radiographic Findings

Radiographically, there is an increase in the talocalcaneal angle in both the AP and lateral projections, along with lateral and dorsal subluxation of the navicular on the talus. On the AP radiograph, a line drawn along the first metatarsal through the navicular, then to the head of the talus and through the body of the talus, makes a Z shape. Similarly, on the lateral radiograph, there is dorsiflexion of the talonavicular joint (probably secondary to increased talar plantar flexion) and plantar flexion of the tarsometatarsal joints, which creates a sagittal plane Z deformity.

Treatment

Because of the lack of criteria for definition of the deformity and its relative severity, early treatment of skewfoot is often based on the perception of need to treat the metatarsus adductus. Thus a more severe adductus that is not passively correctable and is rigid can be treated by stretching and serial casting. The cast technique must carefully mold the hindfoot into varus to avoid exacerbating the existing valgus while correcting the forefoot. ^, Because hindfoot valgus cannot be corrected by casting, this treatment basically converts skewfoot into flatfoot. Such an outcome is not unreasonable, because flatfeet are generally less symptomatic and are responsive to nonoperative treatment later in life, if necessary. Reverse-last shoes and the Denis Browne bar are probably contraindicated in true skewfoot because they can exacerbate hindfoot valgus.

In children beyond 10 years old, symptomatic skewfoot usually has an Achilles contracture, not unlike symptomatic flatfoot. Stretching and orthoses can be tried, but because skewfoot deformities are more inflexible than those present in pes planovalgus, these approaches are generally unsuccessful. As a result, operative treatment, ideally the last resort, is often pursued for lack of a useful alternative.

Earlier reports of extensive soft tissue releases (tarsometatarsal capsulotomies) combined with subtalar fusion or calcaneal osteotomy included only small numbers of patients, and because of recurrence and degenerative arthritis arising from the joint-damaging procedures, ^, have fallen into disfavor. Mosca proposed an Evans calcaneal lengthening procedure combined with a Fowler-type opening wedge osteotomy of the medial cuneiform and Achilles tendon lengthening to completely avoid any joint incursion and correct all the different deformities. ^{, , ,} With this combination, 9 of 10 feet were well corrected while joint motion was preserved. This approach would appear to be the procedure of choice for a skewfoot requiring surgical correction.

For References, see expertconsult.com .

Etiology

Clubfoot has long been associated with neuromuscular diseases and syndromes, and therefore an underlying neuromuscular or syndromic/dysmorphic etiology for all “idiopathic” clubfeet has always been suspected. In the second edition of this book, Tachdjian listed arthrogryposis, diastrophic dysplasia, Streeter dysplasia (constriction band syndrome), Freeman-Sheldon syndrome, Möbius syndrome, and other conditions resulting from chromosomal deletions as just a few of the more recognizable systemic conditions with associated clubfeet. In contrast, idiopathic clubfoot is commonly due to a single musculoskeletal deformity in an otherwise normal infant. Because the final outcome in this latter situation is often diminished function, ^, the conclusion that idiopathic clubfoot represents a primary but local dysplasia of all tissues of the affected extremity from the knee down is supported by the historical inability of treatment to completely reverse this congenital dysplasia and produce a normal foot and extremity. That is not to say that current treatment modalities, including surgery, cannot produce a functional foot and extremity that serve the patient well. It simply recognizes that treatment of true idiopathic clubfoot can never produce a fully normal extremity.

Over the past 5 years, several genetic factors have been implicated in population and family studies on clubfeet. As maternal smoking during pregnancy increases the risk for clubfoot, genes that may have an impact on modulating tobacco smoke have been studied. These include the N-acetylation genes, NAT1 and NAT2, as well as other xenobiotic metabolism genes such as CYP1A1. ^, Following investigation, these genes are thought not to play a major role in the development of clubfeet. Other genes that are involved in limb and muscle morphogenesis ( HOXA , HOXD , and IGFBP3 ), involved in the development of the lower extremity ( CAND2 and WNT7a ), that encode contractile proteins of skeletal myofibers, and that are hind limb–specific genes (TBX4) have also been studied. ^{, , ,} Variation in these genes may increase the susceptibility toward the development of clubfeet, but none are considered to be a direct cause. Continued study in genetics may ultimately provide specific answers with regard to the etiology of clubfeet.

Many other theories on the etiology of congenital clubfoot have been proposed, including an arrest in embryonic development. In normal fetal development of the lower limb, the foot in a 6- to 8-week-old fetus has many characteristics of a congenital clubfoot, including equinus, supination, forefoot adduction, and medial deviation of the talar neck. Bohm proposed that an arrest in fetal development at this stage was responsible for the clinical deformities noted at birth. Normally, the supinated, adducted, and equinus position seen in an 8-week embryo gradually corrects with continued development, and the fetal foot becomes normal at 12 to 14 weeks. If this theory is accepted, a true congenital clubfoot has already existed for approximately 7 months in utero by the time the full-term infant is born. However, the characteristic dysmorphic talar head and the medial dislocation of the navicular have never been observed at any stage of normal fetal development. ^{, , ,} Thus, an arrest in normal fetal development fails to account for this primary dysplasia.

The innate stiffness of clubfeet was clarified by Zimny and colleagues, who identified myofibroblastic retractile tissue in the medial ligaments. This finding confirmed earlier studies by Ippolito and Ponseti, who identified an increase in collagen fibers and fibroblastic cells in the ligaments and tendons of a clubfoot. Thus a second hypothesis about the etiology of clubfoot proposes a retractive fibrotic response, not unlike Dupuytren contracture, as a primary factor. This hypothesis is supported by studies demonstrating abnormal ligamentous and fascial restraints in “soft” tissues that inherently resist correction of deformity. ^, These histopathologic findings help explain the maintenance of a clubfoot deformity and resistance to correction, if not the actual cause. Transforming growth factor-β and platelet-derived growth factor are expressed at higher levels in these contracted tissues. Growth factor blockade with neutralizing antibodies is reported to have the potential to lessen the severity of the contractures and ultimately positively influence the outcome of clubfoot treatment. A decreased density of nerve fibers in the synovium of clubfeet has been reported. This lack of sensory input may also be responsible for the fibrosis and contractures associated with clubfoot. However, the association of clubfeet with syndromes of inherent ligamentous laxity (Down, Larsen) confounds the hypothesis that fibrotic retractile tissue is a primary etiology. In addition, a recent report using light and transmission electron microscopy failed to reveal any myofibroblast-like cells in the capsule, fascia, ligaments, or tendon sheaths of nine clubfoot specimens.

Studies of stillborn and fetal clubfeet have led some authors to propose that a primary germ plasm defect in the cartilaginous talar anlage produces the dysmorphic neck and navicular subluxation. ^{, , ,} Such a proposal is consistent with the observation that the dysmorphic talar head and navicular position are not seen in normal embryonic development and thus must be present from initial limb bud differentiation in an affected extremity. The association of clubfoot with various neurologic entities is well known, with some of the most severe clubfeet being associated with paralytic disorders, such as arthrogryposis and spina bifida. Not surprisingly, a theory postulating localized neuromyogenic imbalance, especially involving the peroneals, has been proposed. ^, Congenital fiber type disproportion, with an imbalance between type I and type II muscle fibers and atrophy of type I fibers, has been found in both peroneal and triceps surae histopathologic specimens. ^{, ,} A study on the histologic and histochemical analysis of 431 muscle specimens in idiopathic clubfeet reported that 86% showed no evidence of a pathologic diagnosis with normal fiber-type ratios and no type I fiber grouping indicative of neuromuscular pathology. Only four specimens (0.9%) showed type I fiber predominance, and 12.8% revealed muscle fiber atrophy. This study did not support the theory that a neuromuscular abnormality is responsible in the etiology of clubfoot. It is safe to conclude that the etiology of idiopathic clubfoot is multifactorial and modulated significantly by developmental aberrations early in limb bud development. Clubfoot does cluster in families but does not fit typical mendelian inheritance patterns. ^, Studies conducted on twins, different incidences in various ethnic groups, and transmission between generations all suggest a genetic etiologic component. Investigation of the genetic sequences of early embryonic limb development will eventually yield a more unified etiologic picture, as well as possible new therapeutic avenues for interrupting or correcting these aberrations.

Pathologic Anatomy

Descriptions of the pathologic anatomy in clubfoot can be found in some of the earliest orthopaedic writings and continue to be essentially correct today, even as we have more sophisticated methods of imaging to quantitate that deformity. Scarpa in 1803 reported medial and plantar displacement of the navicular, cuboid, and calcaneus around the talus. Displacement of the navicular and calcaneus produces an inverted or varus hindfoot, and the entire complex rests in equinus. Contracture of the soft tissues (ligaments, joint capsules, and tendons) maintains this pathologic malalignment of joints, described as equinovarus. Multiple subsequent authors have added to the body of knowledge by describing deformities that can be separated into intra osseous deformities, or deformities within the bone itself, and inter osseous deformities, or deformities resulting from the relationship of one bone to another. Scarpa, Adams in 1866, and Elmslie in 1920 did not implicate the talus as the main pathologic structure but emphasized the midtarsal subluxation—the navicular and cuboid displaced medially, with plantar and medial rotation of the calcaneus. ^, This, as well as other deformities in the clubfoot, has been nicely demonstrated in the past several years by MRI studies. ^, Others have emphasized the talonavicular subluxation and dislocation of the head of the talus out of its “socket” (acetabulum pedis). ^{, , ,} Actual deformity of the talar body and neck has been described in the more recent literature based on intraoperative observations and imaging studies. ^c

c References , , , , , , , .

Finally, Ponseti, in defining clubfoot, has emphasized the cavus component, especially how it relates to nonoperative correction.

The deformity in the talus itself includes medial and plantar deviation of the anterior end, with a short talar neck projecting medially from a dysmorphic, small body that is poorly placed within the ankle joint. The talar neck-body declination angle is invariably decreased, with the neck axis approximating 90 degrees to the axis of the body in some specimens as compared with the normal 150 to 160 degrees ( Fig. 19.39 ). ^, The articular surface of the talar head may be found so close to the body that a true neck is not present ( Fig. 19.40 ). On the inferior aspect of the talus, the anterior and medial facets of the subtalar joint are absent, fused, or significantly misshapen, so the overall impression of talar development is consistent with the proposed primary cartilaginous anlage defect. ^{, , ,} The fact that the radiographic appearance of the ossification center of the talus is delayed further supports the hypothesis of a primary germ plasm defect. Intraosseous deformity in the calcaneus, navicular, and cuboid, though similar to the dysplasia of the talus, is usually much less severe. The contour of the calcaneus, for example, is generally normal, although the calcaneus is often small. The sustentaculum tali is usually underdeveloped, consistent with dysplasia of the talar facets above, and the anterior articular surface of the calcaneus is medially deviated and deformed because of the interosseous deformity of the calcaneocuboid joint (see Fig. 19.40 ). ^{, ,} Both the navicular and the cuboid tend to have more normal shapes and are misshapen only by their interosseous relationships with the talus and calcaneus. The medial tuberosity of the navicular may be hypertrophied as a result of the excessively thick ligamentous structure tethering the navicular to the medial malleolus and calcaneus.

Controversy exists concerning the presence or absence of excessive medial or internal tibial torsion. Evidence for ^{, ,} and against ^{, ,} this deformity has been reported, and it is our experience that true medial tibial torsion can exist in the presence of clubfoot but is generally unusual. More important is the intraarticular (interosseous) deformity known as medial, or internal, spin. This deformity, which involves both the talus and the calcaneus within the mortise, is also a source of controversy. In the transverse plane, the talus has been described as medially rotated and supinated within the mortise, laterally rotated in its body, and neutrally rotated. ^{, , , ,} The controversy has been due to the difficulty of observing the interosseous deformity at surgery. Exposure of the tibiotalar joint and other structures of the clubfoot necessarily eliminates some of the interosseous deformity because the joint capsules and ligaments have been cut to expose these deformities. It is therefore not surprising that surgical observations by different investigators have been somewhat contradictory, which underscores the need for preoperative imaging to evaluate the interosseous deformities noninvasively and thus not disturb the deformity while in the process of observing it.

The study of Herzenberg and colleagues was a landmark in this regard. By digitizing multiple microtome slices of normal and clubfoot fetal specimens, the investigators were able to create three-dimensional reconstructions of the deformities by computer, similar to what would ultimately become three-dimensional CT and MRI reconstruction technology. The significantly dysmorphic talus (see Fig. 19.40 ) was found to have a neck-body axis of 60 degrees in their specimen. More important, the talar neck was found to be internally rotated 45 degrees relative to the tibia-fibula axis (ankle mortise), whereas the calcaneus was internally rotated 22 degrees. Both these rotations were approximately 20 degrees more than normal. Herzenberg and colleagues further commented that the body of the talus appeared to be externally rotated within the mortise but noted that the overall axis gave the impression of internal rotation because of the marked intrinsic deformity of the talar neck and medial displacement of the articular surface.

In the coronal plane of the ankle, deformity of the talus around its longitudinal axis has been found to actually be a pronation or “intorsion” deformity, ^, reminiscent of the deformity seen in embryonic specimens. The calcaneus, often described as inverted or supinated in surgical observations, has also been found to be intorted or pronated, especially its posterior segment, though not nearly to the same degree as the talus. Although it may be difficult to visualize a pronation deformity as being responsible for what is anatomically an inversion or varus of the heel, the presence of this deformity in the tibiotalar joint, perhaps exacerbated by the equinus positioning of both hindfoot bones, cannot be denied because it has been observed both arthrographically and visually at surgery ( Fig. 19.41 ).

The navicular is consistently displaced medially and plantarward on the talar head and has a false articular relationship to the medial malleolus (see Figs. 19.39 and 19.40 ). The articular cartilage of the talar head may be uncovered laterally as a result of medial displacement of the navicular.

As mentioned, the cuboid is similarly displaced medially on the anterior end of the calcaneus (see Fig. 19.40C ). ^{, ,} Because the calcaneus is also medially rotated in relation to the ankle mortise in the transverse plane, this contributes to a significant midfoot “varus” or adductus.

Contractures of the periarticular soft tissues must be stretched—or occasionally surgically released—to successfully restore clubfoot anatomy to a more normal appearance. Thickening and contracture of tendon sheaths and ligaments (as well as inelastic muscle tissue) have been reported by multiple investigators. From studies of muscle, evidence of neurogenic disease has been described, ^, as well as fibrotic (collagen-producing) protein synthesis. ^, Denervation and neuromyogenic changes in the tibialis posterior, peroneals, triceps surae, and long toe flexors appear to be a result of the condition itself as opposed to being the result of nonoperative or operative treatment. Shortened musculotendinous units are a consistent finding at any stage of clubfoot treatment and are obstacles to correction of the bony deformity described earlier. In addition, fibrosis of tissues such as the plantar fascia, the calcaneonavicular (“spring”) ligament, the tibionavicular ligament, and the so-called master knot of Henry (which engages the flexor hallucis longus and flexor digitorum longus at their decussation) is suspected in the pathogenesis, ^, and all must be addressed by nonoperative manipulative stretching programs or on occasion by surgical release. Mobilizing the navicular depends on successfully stretching the tibialis posterior and the master knot; mobilizing the talus and calcaneus out of equinus often requires lengthening the Achilles tendon; and the ability to externally rotate the calcaneus to restore normal talocalcaneal divergence requires peripheral subtalar capsular stretching. The increased fibrosis and contractile myofibroblasts in these “soft” tissues, the interosseous restraints maintaining deformity, must be successfully stretched or occasionally surgically released if there is to be any remodeling after anatomic correction of the bony dysmorphic structures.

A flexor digitorum accessorius longus muscle may be identified in 7% of children whose deformities require surgical correction. Children who have first-degree relatives with clubfoot are seven times more likely to have the anomalous flexor muscle than children without first-degree relatives with clubfoot.

Finally, there have been separate reports of congenital deficiencies of the dorsalis pedis artery and the posterior tibial artery ^, associated with idiopathic clubfeet. These deficiencies, though rare, are more prevalent in clubfeet with greater deformity. As a result of these abnormalities, ischemic necrosis of a portion of the foot after surgery has been described. ^,

Diagnostic Features and Differential Diagnosis

It is rarely difficult to identify a true clubfoot in a newborn ( Fig. 19.42 ). The classic appearance of the heel in marked equinus, with the foot inverted on the end of the tibia, giving the foot an upside-down appearance in more severe cases, is difficult to mistake for anything else. Lack of correctability separates a true clubfoot from the milder postural clubfoot. The milder manifestations represent an in utero postural deformity, identified by the fact that it is fully (or nearly fully) correctable passively and by the conspicuous absence of the significant contractures and deep skin creases of a true clubfoot. A postural clubfoot exhibits none of the atrophy and rigidity of true talipes equinovarus. Postural deformity can frequently be passively corrected at initial evaluation by several minutes of gentle stretching.

In addition to distinguishing the severity, it is also essential to search for associated anomalies and neuromuscular conditions that define a nonidiopathic deformity. The prognosis for a nonidiopathic, syndromic clubfoot is generally worse than that for an idiopathic clubfoot, although there are certain exceptions, such as Down syndrome or Larsen syndrome. In these syndromes, because of the significant ligamentous laxity underlying the syndrome itself, correction may be achieved with nonoperative treatment. If surgical release of a clubfoot is needed, it must be done judiciously rather than aggressively or completely because the foot will have a propensity to be overcorrected, and overcorrection will result in an equally severe and perhaps unreconstructable calcaneovalgus deformity if the laxity of the underlying syndrome is not taken into account. On the other hand, patients with arthrogryposis, diastrophic dysplasia, Möbius or Freeman-Sheldon syndrome, spina bifida and spinal dysraphism, and fetal alcohol syndrome have clubfeet that are notorious for defying correction and subject to severe recurrence. Although there have been recent reports of correction of patients with arthrogryposis and spina bifida by Ponseti nonoperative technique, the need for surgical correction should be expected. ^{, , ,} Such techniques as primary bone resection (e.g., lateral column shortening, talectomy) and complete division of tendons rather than lengthening are often used in the management of these syndromic types of clubfeet. The importance of presurgical diagnosis of these conditions cannot be overemphasized because it will eventually affect the technique of management.

Some evaluation of the severity of the deformity is recommended, both for prognostic value and for monitoring the progress of treatment. Methods of treatment cannot be compared for efficacy if the initial severity of the deformity is not known or described. Determination of the initial severity index is an important assessment of each component of talipes equinocavovarus because it alerts the surgeon and family to the need for heel cord tenotomy or possible surgical release. ^, Although Goldner and Fitch, Carroll and colleagues, Pirani and associates, and others ^{, ,} have proposed evaluation schemes, we favor the method of Dimeglio ^{, , ,} because of its more objective and reproducible method of scoring ( Fig. 19.43 ). In fact, rating the severity of the clubfeet before nonoperative treatment using this method is predictive of the outcomes at 2 years old ( Fig. 19.44 ).

The overall rate of hip dysplasia in those with idiopathic clubfeet is less than 1%. Thus, screening hip radiographs are probably not warranted.

Nonoperative Treatment

Almost all orthopaedists agree that the initial treatment of idiopathic clubfoot should be nonoperative. Most also agree that the earlier the treatment is begun, the more likely that it will be successful because of the relatively viscoelastic character of the newborn foot. The underlying philosophy of advocates of nonoperative treatment is that this should be a definitive method, thereby eliminating or significantly reducing the incidence and amount of surgery that might eventually be required. Doing so avoids the scarring and stiffness that develops after surgery. Much of this philosophy is predicated on observations in earlier surgical experience that neonatal clubfoot surgery almost always produces a more scarred and stiff foot. ^, With this in mind, nonoperative treatment proposes to gradually correct the clubfoot deformity without producing the scar tissue that inevitably diminishes the result. The detrimental role of retracting fibrosis and the observations of myofibroblasts in retractile tissue ^{, , ,} certainly add histologic confirmation to the clinical observations concerning the results of neonatal surgery.

Perhaps the most determined early protagonist of nonoperative treatment was J.H. Kite, who in the period from 1924 to 1960 nonoperatively treated more than 800 patients at the Atlanta Scottish Rite Hospital. In his 1964 monograph The Clubfoot , Kite described in great detail the method of manipulation and cast correction that had served him well over many years. Kite corrected each component of the clubfoot deformity separately and in order, beginning with forefoot adduction and proceeding to correction of heel varus (inversion) and finally to correction of equinus. He was adamant that one could not proceed to correct the next deformity until the previous one had been corrected. Ideally, Kite believed, the forefoot should be slightly overcorrected into a mild flatfoot position before the foot was brought up out of equinus. Today, the Kite method is rarely used because of the inability of others to match his results and the excessive amount of time (26–49 weeks) required for infants to remain in casts. However, enthusiasm for nonoperative treatment remains at an all-time high on account of the success that is currently being achieved with the Ponseti method, and the less frequently used French physiotherapy technique. ^d

d References , , , , , , , , , , , , , , , , , , , , , , .

Ponseti Method

In the early 1940s, Ignacio Ponseti developed his nonoperative approach to the treatment of clubfeet. Similar to Kite, he was stimulated to investigate a less aggressive correction than that used by surgeons with their attendant high rates of complications, stiffness, and overcorrection. Careful anatomic dissection of stillborn babies with clubfeet was critical for Ponseti to define the pertinent pathoanatomy and to rationalize a mechanism for correction. In addition, detailed investigations into the biology of collagen were fundamental in supporting a gradual correction. His method of weekly manipulation and cast application to hold correction allows relaxation of the collagen and atraumatic remodeling of joint surfaces without the fibrosis and scarring associated with surgical release.

In general, the ability to obtain full correction via this approach is enhanced when treatment is instituted within the first month of life. In these instances, Ponseti reports that the need for posterior medial and lateral release is avoided in more than 95% of cases. Although the success rate in older infants (7–10 months old) is lower than that in younger ones, there is some merit in attempting to obtain as much correction as possible before surgical release.

For this method to be successful, correction of deformity should proceed in an orderly fashion. The acronym CAVE ( c avus, a dductus, v arus, e quinus) is helpful because it not only describes the clinical position of the clubfoot, but it also outlines the general order of deformity correction via the Ponseti method.

Technique

The protocol consists of stretching and manipulating the foot and applying holding casts until the next session 5 to 7 days later. To stretch the ligaments and gradually correct the deformity, the foot is manipulated for 1 to 3 minutes. The correction is maintained for 5 to 7 days with a plaster cast extending from the toes to the upper third of the thigh and the knee at 90 degrees of flexion. Five or six cast changes are sufficient to correct most clubfeet. Casting is usually timed to coincide with routine feedings; after manipulation, the infant is fed a bottle, which tends to relax the infant and allow easier cast application.

The first goal is correction of the cavus deformity by forefoot supination relative to the hindfoot. This manipulation seems counterintuitive because it tends to exaggerate the appearance of overall foot inversion. Elevation of the first metatarsal and supination of the forefoot is in contradistinction to other methods of manipulation that propose correction of the cavus by pronation of the first metatarsal. At the first session the forefoot is simultaneously supinated and abducted. The cavus is almost always corrected with the first cast.

At successive manipulation and casting sessions, metatarsus adductus and hindfoot varus are simultaneously corrected by abducting the foot while counterpressure is applied laterally over the talar head. With this technique, the calcaneus, navicular, and cuboid are gradually displaced laterally. This key maneuver corrects the majority of the clubfoot deformity and must be performed at each session with attention to three points ( Fig. 19.45 ). First, forefoot abduction should be performed with the foot in slight supination. As such, correction of the cavus deformity is preserved, but collinearity of the metatarsals is also maintained, thereby producing an efficient lever arm for abduction. Second, the heel should not be constrained by premature dorsiflexion. It is important that abduction be accomplished with the foot in equinus to allow the calcaneus to abduct freely under the talus and evert to a neutral position without touching the heel. It is also important to avoid forceful dorsiflexion before correction of hindfoot varus because a rocker bottom deformity could develop. Third, care is taken to locate the fulcrum for counterpressure on the lateral head of the talus. Correction of hindfoot varus and calcaneal inversion would be hindered if counterpressure were applied to the lateral column of the foot or at the calcaneocuboid joint as opposed to the talar head. In general, three or four weekly manipulation and casting sessions are required to loosen the medial ligamentous structures of the tarsus and partially mold the joints. After each cast, foot supination is gradually decreased to correct the inversion of the tarsal bones while the foot is further abducted under the talus.

Equinus is the last deformity that is corrected, and correction should be attempted when the hindfoot is in neutral to slight valgus and the foot is abducted 70 degrees relative to the leg. This degree of abduction seems excessive but is needed to prevent recurrence of deformity. Equinus may be corrected by progressively dorsiflexing the foot after the varus and adduction of the foot have been corrected. The foot is dorsiflexed by applying pressure under the entire sole of the foot and not much under the metatarsal heads to avoid a rocker bottom deformity. Equinus may be completely corrected through further progressive stretching and casting. However, to facilitate more rapid correction, subcutaneous heel cord tenotomy is performed in the vast majority (at least 85%) of patients. In this procedure the entire Achilles tendon is transected. Heel cord tenotomy has been performed in children up to 1 year old without the occurrence of overlengthening or weakness. Tenotomy may be performed with a thin cataract knife in the clinic under sterile technique (after lidocaine/prilocaine cream has anesthetized the skin locally for 30 minutes; Fig. 19.46 ). Though the procedure is done this way in the majority of patients, some physicians elect to perform tenotomy in the operating room in children older than 3 months. It is easier to apply a better cast without the resistance encountered in older and therefore stronger infants.

Certain techniques should be adhered to when performing a percutaneous heel cord tenotomy. After standard sterile preparation, the foot is held by an assistant with mild to moderate dorsiflexion pressure. Excessive pressure may tend to tighten the skin and hinder the ability to palpate the tendon well. The blade enters the skin along the medial border of the Achilles tendon. Because the calcaneus is usually elevated in the fat pad, it is important to cut the tendon 0.5 to 1 cm proximal to its insertion, where it tends to fan out onto the tuberosity of the calcaneus. After insertion, the blade is pushed medial to the tendon and rotated underneath it. Counterpressure with the opposite index finger will push the tendon onto the blade and prevent inadvertent and unnecessary skin laceration. Excessive motion of the blade laterally places the lesser saphenous vein and sural nerve at risk. Successful tenotomy is heralded by a palpable pop and immediate ability for further dorsiflexion of approximately 15 to 20 degrees. No stitches are needed and sterile cotton cast padding is applied, followed by the application of a long-leg cast in maximal dorsiflexion with abduction to 70 degrees. The foot is immobilized for 3 to 4 weeks; most infants require immobilization for 3 weeks, the slightly longer immobilization being reasonable in children older than 6 months.

An alternative to percutaneous heel cord tenotomy has been suggested by Alvarez and colleagues. Botulinum A toxin is injected into the triceps surae muscle complex to weaken its function. Very short-term success with this approach, as opposed to tenotomy, was reported in 50 of 51 infants with clubfeet.

Well-molded long-leg plaster casts are applied over a thin layer of cotton padding at all steps during the treatment ( Fig. 19.47 ). Benzoin is not applied to the skin and fiberglass is not used because of poor molding characteristics. Casts extending above the knee are necessary to maintain the foot in abduction and external rotation, and to improve results. The casts are molded with care taken to avoid pressure spots directly over the heel or malleoli. The plaster on top of the toes may be trimmed off, but a platform of plaster is left under the toes to favor stretching of the toe flexors. As previously mentioned, it is advantageous to cast the infants when they are feeding. This may distract the child and facilitate application of the cast ( Video 19.4 ).

After removal of the last cast, a foot abduction orthosis (often called a Denis Browne bar and shoes) is prescribed to prevent recurrence of the deformity, to favor remodeling of the joints with the bones in proper alignment, and to increase leg and foot muscle strength. The orthosis consists of two straight-last open-toe shoes, or alternatively more flexible leather shoes with rubber inserts, connected by a bar that allows the shoes to be placed at shoulder width ( Fig. 19.48 ). The bar should hold the shoes at 70 degrees of external rotation and 5 to 10 degrees of dorsiflexion. In unilateral cases, the normal foot should be in 40 degrees of outward rotation. Maintaining the feet at shoulder width facilitates foot abduction. The orthosis is worn full time for at least 3 to 4 months, and afterward it is worn at nap and nighttime for 2 to 4 years. A flexible abduction bar may increase wear compliance.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here