The term ‘flat foot’ applies to a range of conditions, some of which are normal, and some of which portend major disease. Generally, the term means loss of the medial arch; this is usually accompanied by increased weight bearing on the medial border of the foot, that is eversion of the subtalar joint. This contrasts to the situation seen in club foot, in which the subtalar joint is inverted. Eversion is not necessarily associated with flat foot: 97 per cent of infants have no arch, not because of abnormal weight bearing, but because the arch in infants is filled with fat. This is, of course, of no importance.

The most common type of everted foot is the ‘loose-ligamented’ or ‘supple’ flat foot seen in young children. Ligamentous structure is most lax in the young, presumably becoming tighter as the collagenous structures become more crosslinked with age.

The loose-ligamented flat foot is always bilateral, since laxity cannot be a phenomenon only on one side of the body. When the child is non-weight bearing, or when he stands on his toes, and the muscles contract, the arch looks normal. Treatment by strengthening the foot musculature to support the arch was shown to be ineffective: electromyographic studies revealed no electrical activity in the foot musculature when the patient was standing quietly. No matter how much the muscles were strengthened, the foot remained flat. Laxity can often be demonstrated in other joints: the patient can often bring the thumb to the forearm, and there is usually hyperextension of the elbow.

It is important to look for inversion and eversion of the subtalar joint, and to make certain that the heel cord is not tight. To test for heel cord tightness, not only should the knee be straight (to stretch the gastrocnemius muscle), but the foot should be held in a varus position. Since the subtalar joint goes into dorsiflexion when it everts, a tight heel cord causes eversion of the foot, creating a flat foot which almost looks as if it is broken in the middle (Fig. 1) 2603. A tight heel cord could be the result of spasticity or congenital low insertion of the soleus. Limitation of subtalar motion usually implies either a congenital vertical talus (congenital convex pes valgus) or a coalition syndrome. The subtalar joint also moves into a valgus position when there is inflammation, because there is more space in the joint. Patients with juvenile rheumatoid arthritis or trauma to the articular surfaces therefore tend to have a fixed valgus deformity.

The loose-ligamented flat foot is not painful, even in adults. However, flat feet arising from other causes may become painful as the patient matures. The tight heel cord foot has already been described. The talus is usually prominent on the medial sole; if the foot is held in equinus and inverted, the navicular swings in front of the talus, and the alignment of the foot is neutral.

There are two varieties of flat feet, in which there is no subtalar motion. In the congenital vertical talus, the talar head is prominent on the medial border of the foot. This deformity has also been described as the congenital rocker bottom foot: as the forefoot is everted and dorsiflexed, and the hindfoot is in equinus (Figs. 2 - 4) 2604,2605,2606. The navicular bone is dislocated on the neck of the talus; the talar-calcaneal angle is more than 50° on the lateral radiograph, and may be completely vertical. However, not all vertically oriented tali are associated with navicular dislocation. Any really severe flat foot may have a vertically oriented talus. Unfortunately, because the navicular bone does not ossify until the third year of life, this is not a helpful radiographic finding in the newborn. The only important finding in congenital vertical talus as a cause of flat foot is that the calcaneus is in an equinus position. At birth, the congenital vertical talus is most commonly confused with the calcaneal-valgus foot, a benign condition in which the forefoot is also everted (Figs. 5 and 6) 2607,2608. The difference here is that the calcaneus is in a dorsiflexed position. Another diagnostic feature is lack of alignment of the first metatarsal with the talus on plantar flexion of the foot. MRI clearly show the navicular bone and whether or not it is dislocated.

Lack of inversion is also characteristic of the coalition syndrome, or peroneal spastic flat foot. Clinically, this does not usually present as much of a flat foot. These children usually have pain, mostly on the lateral side of the foot, over the subtalar joint. Attempts at inversion lead to pain over the peroneal tendons. However, injection of steroid or severance of the tendons does not lead to improved motion or decreased pain. Pain usually starts around adolescence, probably because the coalitions start to ossify around that time. There are bony, cartilaginous, and fibrous coalitions, the most common being the calcaneonavicular and the talocalcaneal coalitions. Coalitions occasionally involve other pairs of bones, but these are quite rare. The aetiology is related to failure of segmentation in utero, and not to incorporation of accessory bones, as once was thought. This tendency is inherited as a dominant gene, with variable penetrance: 55 per cent of first-degree relatives may be affected, but many coalitions never become symptomatic. Development of symptoms may follow some prolonged episode of strain.

There are clues to the diagnosis on the regular lateral radiograph. The talus may appear foreshortened, and beaking of the top of the navicular or the neck of the talus is common. The calcaneonavicular coalition is best seen on oblique radiographs. It may appear as a bony bridge, or the space between the bones may be narrowed, and the bony surfaces appear serrated (Fig. 7) 2609, suggesting a cartilaginous coalition. The talocalcaneal coalitions are seen best on axial views of the calcaneus shot with the beam angled down on the foot, making an angle of between 30 and 45° with the floor (Harris-Beath views). The attempt here is to follow the angle of the subtalar joint. The more common coalition involves the medial facet, but the posterior facet may also be involved. The anterior facet is more rarely affected. A lot of artefacts are produced with these views, which may not be in the plane of the joint. If the joint is involved, not only is the facet partially or completely obliterated, but it appears angled, instead of horizontal. Lateral tomograms may help, especially with anterior facet involvement, but the current gold standard for diagnosis of talocalcaneal coalitions is the CT scan, which is particularly useful in helping to determine surgical treatment.

The flat foot with an unattached accessory navicular bone, or ostibiale externum, is also painful (Fig. 10) 2612. Accessory navicular bones are present in 10 per cent of individuals; in 90 per cent of these the bone attaches to the main body of the navicular which is prominent or cornuted, but is not painful. When the accessory bone is not attached, pain occurs not only over the prominence, but on resisted inversion of the foot. This indicates that this is a traction injury, and that the motion produced by the pull of the posterior tibial tendon is giving the symptoms. Again, pain usually starts in adolescence.

The loose-ligamented flat foot requires no treatment. The main problem is the wearing down of shoes on the medial border. Various supports are available: most are uncomfortable and are not used by the patient. There is no evidence to suggest that such supports correct flat foot due to ligamentous laxity—patients who improve would probably have shown spontaneous correction. Supports may be of use in patients with severe deformity, to hold the foot in a better position until spontaneous ligament tightening occurs. Surgical treatments for persistent pain or deformity with abnormal shoe wear have been described. A Grice-Green subtalar extra-articular arthrodesis may be used in young children; another approach is an opening wedge osteotomy of the calcaneus on its lateral border. Evans has described enlarging the lateral column of the foot by placing opening a wedge on the lateral border of the foot. Many types of partial fusion have been described, but these are not often used.

Congenital vertical talus almost always needs surgical treatment. Early casting will improve the appearance of the foot, by plantar flexing and inverting the forefoot. It also makes surgery somewhat easier, but casting probably never corrects the foot with a true navicular dislocation. There is a lot of controversy over the correct surgical procedure. A medial approach, similar to a club foot release, allows a capsulotomy to be performed to reduce the navicular bone on the head of the talus. It may be necessary to lengthen the peroneal tendons or the toe extensors, although lengthening of the toe muscles is rarely needed. The tendon Achilles also has to be lengthened to bring the heel up. The difficulty comes in elevating the talus so that it stays reduced on the calcaneus. Some surgeons move the anterior tibial tendon to the neck of the talus; others use the posterior tib as a sling underneath. Some have proposed removing the navicular bone, wedging it underneath the talus. Release of the underlying ligaments binding the talus to the calcaneus, including the interosseous ligaments and the ligaments of the sinus tarsi, allows the talus to spring up. This can be pinned on top of the calcaneus; the navicular can be pinned to the talus. Earlier surgery is better; in older patients a Grice extra-articular subtalar arthrodesis may be added to the treatment regimen. It is important to remember that congenital vertical tali are frequently associated with underlying neurological problems, and muscle imbalance has to be considered.

The pain associated with coalition syndrome should initially be treated conservatively. Casting may have no effect, or may relieve the pain for several months, or even permanently. If conservative treatment is unsuccessful, the coalition may need to be resected (Fig. 11) 2613: calcaneonavicular coalitions are most amenable to such treatment. Contraindications to resection vary. Some authorities feel that neither calcaneonavicular nor talocalcaneal coalitions should be resected if secondary arthritis changes are present. These changes are not, however, necessarily representative of degenerative changes, and may be more related to capsular stripping brought on by the restricted motion in other joints. They should probably not be considered a contraindication to resection, but surgery may not be successful, particularly in older patients. Talo-calcaneal coalitions may be resected, if less than 50 per cent of the facet is affected. Involvement of a second joint, which is not infrequent, is a contraindication to resection.

The approach to the calcaneonavicular coalition is lateral, over the sinus tarsi. The coalition is actually a very dense wall of bone, extending deeply. To prevent its reforming, we frequently place the origin of the extensor digitorum brevis muscle into the hole. The medial talocalcaneal coalition is approached between the flexor tendons and is removed with a small Hall-type air drill. Fat is placed into the hole. If no motion can be produced at surgery, none will be achieved. Resection of calcaneonavicular coalitions tends to improve motion more than surgical treatment of talcalcaneal coalitions. If the coalition cannot be resected, or if resection fails to improve the pain a triple arthrodesis may be undertaken.

A triple arthrodesis is performed through a lateral approach, running from the peroneal tendons, across the sinus tarsi, to the top of the talonavicular joint. The fat in the sinus is lifted out as one layer, so it may be replaced to fill a large dead space. The joint surfaces are denuded of cartilage, usually with a large flat osteotome or saw, and if there is any deformity, appropriate wedges are removed. The talocalcaneal and calcaneocuboid joints are fixed with staples: the talonavicular joint may need a threaded wire or a staple. Patients remain in a cast for 3 months, non-weight-bearing for 6 weeks, and weight-bearing for another 6 weeks. Patients undergoing simple resection of coalitions are not usually put into casts, but are started on early motion.

Patients with accessory navicular bones are treated using the Kidner procedure. The posterior tibial tendon is identified, and followed down to the navicular bone. The superior surface is peeled back, and the extra bone is removed. The tendon is then lifted off the lateral surface of the navicular, without severing the extension of the tendon under the foot. An ‘0’ suture is brought through the part of the tendon that was attached to the navicular bone, and the tendon is pulled forwards to fit into a groove under the bone. This area is curetted to give a bleeding surface, and the suture ends are brought through a hole in the navicular and tied on the dorsum. This procedure not only relieves the pain, but may also improve the arch, although the latter is less certain. A cast is applied for 6 weeks with the foot inverted.

Asher C. Flat foot and valgus heel. In: Postural Variations in Childhood. London: Butterworth & Co., 1975; 76 - 101.

Asher M, Mosier MK. Coalition of the talocalcaneal middle facet: treatment by surgical excision and fat graft interposition. Orthop Trans 1983; 7: 149.

Barry RJ, Scranton PEJ. Flat feet in children. Clin Orthop 1984; 181: 68.

Basmajian JR, Stecko G. The role of muscles in arch support of the foot. An electromyographic study. J Bone Joint Surg 1963; 45A: 1184.

Bleck EE, Berzins UJ. Conservative management of pes valgus with plantar flexed, talus flexible. Clin Orthop 1977; 22: 85.

Bordelon RL. Correction of hypermobile flatfoot in children by moulded insert. Foot and Ankle, 1980; 1: 132.

Chang FM. The flexible flatfoot. American Association of Orthopedic Surgeons Instruction Course Lectures, 1988; 37: 109.

Coleman SS, Stelling FH, Jarrett J. Pathomechanics and treatment of congenital vertical talus. Clin Orthop 1970; 70: 62.

Cowell HR. Diagnosis and management of peroneal spastic flatfoot. American Association of Orthopedic Surgeons Instruction Course Lectures, 1975; 24: 94.

Cowell HR, Elener V. Rigid painful flatfoot secondary to tarsal coalition. Clin Orthop 1983; 177: 54.

Drennan JC, Sharrad WJW. The pathological anatomy of convex pes valgus. J Bone Joint Surg 1971; 53B: 455.

Glancy GL. Surgical management of the flatfoot. American Association of Orthopedic Surgeons Instruction Course Lectures 1988; 37: 111.

Harris RI. Retrospect: peroneal spastic flat foot (rigid valgus foot). J Bone Joint Surg 1965; 47A: 1657.

Mann RA, et al. Subtalar fusion for isolated subtalar disorders. Preliminary report. Clin Orthop 1988; 226: 260.

Miller GR. The operative treatment of hypermobile flatfeet in the young child. Clin Orthop 1977; 122: 95.

Mitchell GP. Spasmodic flat foot. Clin Orthop 1970; 70: 73.

Penneau K, Lutter LD, Winter RB. Pes planus: radiographic changes with foot orthoses and shoes. Foot and Ankle 1982; 2: 299.

Powell HD. Pes planovalgus in children. Clin Orthop 1983; 177: 133.

Rose GK, Welton EA, Marshall T. The diagnosis of flat foot in the child. J Bone Joint Surg 1985; 67B: 71.

Seitz DG, Carpenter EB. Triple arthrodesis in children. A ten-year review. South Med J 1974; 67: 1420.

Swiontkowski MF, Scranton PE, Hansen S. Tarsal coalitions: long-term results of surgical treatment. J Pediatr Orthop 1983; 3: 287.