SALVAGE OF DIABETIC NEUROPATHIC ARTHROPATHY WITH ARTHRODESIS
March 7th, 1994
Mark S. Myerson, MD
General Concepts of Treatment
Neuroarthropathy of the foot and ankle in the diabetic patient is a common orthopaedic problem and the outcome depends on proper management. The incidence of neuroarthropathy and diabetes ranges from 1 to 2.5%.1,2 The most common areas of neuroarthropathy are the midfoot and hindfoot.3,4 Due to the increasing longevity of patients with diabetes, it is anticipated that many orthopaedic surgeons will have an opportunity to manage this problem.
The Charcot joint in the diabetic patient (neuroarthropathy) is a common event which may lead to ultimate amputation of the limb. Although other elements of neuropathy and ischemia are often present, they may be, to some extent, addressed without resorting to amputation through revascularization. Structural breakdown of the foot, which leads to wound problems, and osteomyelitis are far more difficult to treat.
The treatment of the Charcot foot and ankle is influenced by the stage of arthropathy. Eichenholtz5 described three stages of development of the Charcot joint that are fairly typical of the course of events from the initial phase of the arthropathy through healing. In Stage I, characterized by acute inflammation associated with hyperemia and erythema, the bone dissolves and fragments, and dislocations are common. In the second stage, characterized by bony coalescence and decreasing swelling, radiographic evidence of periosteal new bone formation is present, even when the initial injury was a joint dislocation rather than fracture. During Stage III, bony consolidation and healing occurs. The initial diagnosis of acute neuroarthropathy is relatively straightforward, based on painless swelling associated with warmth of the affected area. Although radiographs are helpful, the typical changes of neuroarthropathy, such as fragmentation and periosteal new bone formation, are not often present during this acute stage.
Treatment concepts of neuroarthropathy have changed significantly over the past decade.6,7 The goal of any treatment program for the Charcot foot is to achieve a plantigrade weight-bearing surface that is free of infection. In addition to these parameters, the foot and ankle must be stable and braceable. Whether the phase is acute or chronic and involves the midfoot, hindfoot or ankle, most forms of Charcot breakdown are treated conservatively. The accepted treatment for neuropathic arthropathy of the foot and ankle has been prolonged immobilization in a brace or plaster cast until consolidation and healing are radiographically evident and clinical stability of the foot and ankle have been restored.6-13 These methods of treatment are effective in most patients, particularly when promptly instituted. However, severe deformity may develop despite appropriate immobilization and protected weight-bearing, and immobilization of the limb does not guarantee that additional deformity will not occur.
Although some reports indicate that deformity may progress in the immobilized limb even if it is not subjected to weight-bearing, my experience has been otherwise.14 Deformity may indeed worsen in a cast, but I have observed, without exception, that patients with these conditions do bear weight on the extremity, due to a combination of neuropathy, noncompliance, and inability to use crutches or a walker because of problems with proprioception and balance. One wonders whether a long leg cast with the knee flexed would achieve a more reliable end result under these circumstances.
The mainstay of treatment for neuropathic arthropathy, once diagnosed, continues to be prolonged external immobilization in a plaster cast or brace. A brace is the traditional method of managing chronic deformity of the hindfoot and ankle. This brace may be a vertical double upright type, a molded polypropylene ankle foot orthosis, a posterior AFO type, or a clam shell type. The patellar tendon bearing (PTB) brace has recently been recommended by Saltzman et al,15 who assessed the effect of shoe wear, custom-made inserts, PTB braces, and extra-padded PTB braces on load transmission to neuroarthropathic feet. They found that a properly fitted PTB brace can reduce load transmission to the Charcot foot, but in a reliable manner only to the hindfoot. Adding extra padding to the brace decreased the mean peak force further by 32%. They recommended that a PTB brace not be used to reduce vertical load transmission to the midfoot or forefoot. For long-term use, the PTB brace should be used for treatment of hindfoot disorders only, and adjusted regularly to ensure adequate fit. However, deformity of the hindfoot and ankle may not be amenable to bracing, since these structures are far too unstable to be maintained in adequate alignment by either a brace or a cast. The stresses of weight bearing in these patients may preclude the possibility of bracing. To avoid the complications of unstable neuroarthropathy, operative treatment is occasionally warranted, particularly when consolidation and healing of the neuropathic fracture and dislocation has not taken place. This is well exemplified in this patient who was treated with a double upright brace for neuropathic deformity of the ankle. Recurrent ulceration over the fibula developed secondary to ankle instability while in the brace. The problem was not appreciated until weight-bearing radiographs of the foot inside the brace were obtained.
With either a varus or valgus deformity of the hindfoot or ankle, it is particularly difficult to keep the foot in a plantigrade position centered under the leg. Although a Syme or below-the-knee amputation eliminates the immediate problem, many amputees experience limitations in day-to-day functional activities, necessitating a significant modification in their lifestyle. Amputees have an increased energy requirement, frequently have limited cardiac reserve, are often overweight, and have a marked risk for future contralateral amputation. For these reasons, we believe that limb salvage should be attempted wherever possible.
Although operative treatment for neuropathic arthropathy of the foot and ankle is an option, under most circumstances, this should not be the initial treatment of choice. We have reported previously on salvage of intractable diabetic neuropathic arthropathy of the hindfoot and ankle treated with open reduction and arthrodesis. At an average of approximately 4 years after arthrodesis, salvage proved to be successful in 93% of these patients. Certainly, operative treatment must be considered part of one's armamentarium for managing the deformed diabetic foot. This treatment program was, however, complex, since we encountered 20 complications in 19 of the 29 patients. Enthusiasm for operative treatment therefore has to be tempered by the potential for complications during a difficult salvage procedure.
Other authors have reported on the results of operative treatment of neuropathic arthropathy.11,13,16-18 Shibata and colleagues18 reported on results of extended hindfoot fusions in patients with leprotic neuropathic deformity. They used an intramedullary nail for fixation and an arthrodesis occurred in 19 of the 26 patients. Stuart and Morrey13 have reported on the results of ankle and hindfoot arthrodesis in 13 patients with insulin-dependent diabetes. Their results were satisfactory in only five of 13 patients, and complications developed in seven of the nine patients who showed radiographic signs of new neuroarthropathy. One of the problems that may have led to the high rate of failure in their patients was that external fixation was used in nine of the 13 patients. As discussed below, I do not believe that external fixation is the ideal form of stabilization for these deformities. Healing from the acute inflammatory phase through coalescence and ultimate consolidation takes a long time. Healing of the tarsometatarsal joint and midfoot takes approximately 6 to 12 months, and that of the hindfoot and ankle takes 12 to 24 months. The mainstay of treatment for most feet is adequate stabilization during this period of time to allow the tissues to heal and ultimately consolidate. Although deformity leading to recurrent ulceration may require surgery for claw toes, metatarsal head resections, and exostectomy, intervention with open reduction and arthrodesis is less commonly required. It is important to recognize that open reduction on its own is not sufficient to stabilize these joints.
Ideally, surgery should not be performed in the presence of an open wound. Therefore, if feasible, wounds are first healed with either a total contact cast19,20 or a split thickness skin graft. If surgery is performed in the presence of an open wound, the infection rate is increased. Therefore, in the presence of an open wound with or without osteomyelitis reconstruction and salvage has to be staged. This problem exists for example over the malleoli where ulceration may be present associated with deeper infection. In these instances, I prefer to debride the wound and bone, and initiate treatment for osteomyelitis.
The Midfoot
The treatment goal for patients with neuroarthropathy of the midfoot is to obtain a durable plantigrade foot for ambulation. Amputation, however, remains a necessary management option in patients with severe infection with or without ischemia. However, by controlling the rate and severity of ulceration, a carefully designed treatment program can decrease the frequency with which amputations are required. It is important to identify the stage of neuroarthropathy to initiate treatment. Once immobilization and rest brings active neuroarthropathy under control, a total contact cast becomes the mainstay of treatment. Nonoperative treatment measures are usually successful for the midfoot compared to deformity of the hindfoot and ankle where these conservative treatment modalities are more likely to fail.
Acute Midfoot Neuroarthropathy
For acute neuroarthropathy, the mainstay of treatment is rest of the limb, restricted ambulation and activities and total contact cast immobilization. The indication for surgery in this acute setting is very specific, and should be performed only for a severe dislocation which is unstable and manually reducible. Any bone fragmentation or periosteal new bone formation is a contra- indication to surgery. Open reduction and arthrodesis is indicated in these patients to prevent skin necrosis and acute ulceration after reduction of edema. Since the disease process is acute, the bone density is usually adequate for fixation. I emphasize that surgery should not be attempted if preoperative radiographs show evidence of bone absorption and fragmentation, since profound osteopenia may be present precluding stable reduction and fixation.
Surgery should be performed once swelling has resolved, since operating on a swollen foot increases problems with wound closure, and likelihood of infection. This has to be monitored very closely, since the displaced tarsal and metatarsal bones will often cause skin necrosis (Fig. 1). I use the AVI foot pump (Kendall, Mansfield, MA) to reduce edema of the foot.21 We have used the foot pump successfully in the presence of neuropathic swelling of the foot. This has to be monitored closely to prevent pressure and ischemic necrosis of prominent parts of the foot. Between 4 and 6 hours of intermittent compression are often sufficient to decrease all edema in the foot and prepare the foot for surgery. Prophylactic cephalosporin antibiotics are used routinely. I do not use a tourniquet, and surgery is performed under regional ankle block anesthetic. Bone graft is occasionally required for arthrodesis of acute neuroarthropathy, and depends on the magnitude of the deformity. In the event of significant bone loss, iliac crest bone should be used, but we are able to harvest smaller quantities of graft from the calcaneus. This is usually unnecessary in the setting of acute fracture dislocation, since minimal bone resorption and destruction has occurred. For reconstruction of chronic neuroarthropathy, bone graft is invariably required.
The incisions are planned according to the pattern of dislocation, the most common of which is disruption of the medial column involving the first metatarsal and medial cuneiform. Here, a single dorsal incision is made medial to the extensor hallucis longus tendon. In patients with middle column disruption involving the second and third metatarsocuneiform joints, a long incision is made centered over the space between the second and third metatarsals and a second smaller incision medially over the first metatarsal and medial cuneiform. Thick skin flaps are raised without much attention paid to superficial nerves. The bones are handled cautiously since they are fragile and with too much periosteal dissection a bone may fall out of the wound. Fibrous scar and granulation tissue has to be removed from the joint spaces, and thin, flexible chisels or fine, sharp osteotomes are used to denude the articular surfaces. Manual reduction is now performed and temporary internal fixation obtained using 0.062-inch Kirschner wires or the guide pins for cannulated screws. It is important to obtain intraoperative anteroposterior and lateral radiographs to assess the reduction. Although fluoroscopic imaging is useful to guide the insertion of the pins, it is not sufficient to determine alignment. The reduction of the dislocation is usually accomplished by a maneuver that involves grasping the hallux and pulling it into varus while simultaneously pushing with the thumb against the base of the first metatarsal and medial cuneiform. If this is not sufficient to reduce the medial cuneiform, this can be done using a small periosteal elevator. The base of the second metatarsal is the key to an anatomic reduction. Although achieving a precise reduction and alignment of the midfoot is not as important as for the patient without neuropathy, it is still preferable to restore a medial longitudinal arch and prevent recurrent pressure on the medial foot. I therefore prefer to use compression screws for fixation to reduce the medial cuneiform directly into the second metatarsal and vice versa. The first screw is introduced medially from the medial cuneiform into the middle and lateral cuneiforms or the second metatarsal depending on the pattern of dislocation. If the second metatarsal is laterally displaced, I direct the screw obliquely and distally from the medial cuneiform into the second metatarsal base.
Subacute Midfoot Neuroarthropathy
It is important to distinguish between subacute and chronic neuroarthropathy of the midfoot. Although many feet end up deformed, if the chronic stage is reached the foot is stiff and deformed. These feet are generally easy to protect in an extra depth shoe with a molded orthosis and it is unlikely that patients with true chronic neuroarthropathy will require an arthrodesis. Some feet permanently remain in a subacute phase, the midfoot is deformed and unstable, and a "spongy" fibrous arthrosis is present at the apex of the deformity. A rocker bottom deformity is present, with an apex medial or lateral, depending on which bones are prominent on the plantar surface of the foot. In these feet, sagittal plane motion occurs through this pseudoarthrosis and the hindfoot remains fixed in equinus. The rocker bottom deformity is always associated with a fixed hindfoot equinus. The posterior soft tissues, including the Achilles and the flexor tendons, are contracted. An Achilles tendon lengthening is integral to the success of this procedure. The remaining long flexor tendons rarely require lengthening. The magnitude of this equinus contracture is best evaluated on a weight-bearing lateral radiograph of the foot. It is then appreciated that the forefoot must be redirected (plantarflexed) in line with the hindfoot, which must be corrected through posterior soft tissue releases.
Prior to the midfoot approach, soft-tissue lengthening is performed posteriorly with percutaneous triple hemisection technique for the Achilles tendon. A stab incision is made centrally in the posterior aspect of the Achilles tendon and then directed subcutaneously. The incisions on the tendon are spaced approximately 3/4" apart. The Achilles tendon should be lengthened before working on the midfoot, as no effective dorsiflexion lever on the foot is present once the midfoot is open. If the hindfoot is still in equinus after Achilles tendon lengthening, the other long flexor tendons may require lengthening. In severe rocker bottom deformity, this would have to be approached through a lateral incision over the peroneal tendons and a posteromedial incision posterior to the medial malleolus.
I use three incisions to reconstruct these feet: dorsomedial, central, and lateral. Sammarco has recommended a transverse incision across the midfoot with the advantage of improved exposure (GJ Sammarco, personal communication). Transection of superficial nerves is not important; however, I have some concerns about this incision since it disrupts the superficial veins and potentially could cause problems with wound healing, which have not as yet been reported by Sammarco. Thick skin flaps are raised regardless of the method of these incisions. The dislocated joints are approached by resection of the fibrous scar. Although an osteotome may be used here, I find that a small microsagittal saw blade is preferable to perform these planar cuts. The extruded bone fragments on the plantar surfaces must be removed and are usually accessible through the dorsal incisions. A laminar spreader is placed into the wound between the tarsal and metatarsal bones and placed on distraction, and the bone fragments on the plantar surface are removed with a rongeur.
Once the bone fragments, debris, and fibrous tissue have been removed, the forefoot is reduced to the midfoot by adduction and plantar flexion. Large gaps are usually present between the tarsal and metatarsal bones, and approximation of the bone ends should not be attempted at this time. Temporary internal fixation with Kirschner wires is used and anteroposterior and lateral radiographs are obtained. Once the intraoperative radiographs are obtained, permanent fixation and final correction is planned. Bone graft is usually necessary, and although I prefer to use iliac crest autograft, allograft bone has been recommended in the diabetic patient by Alvarez.14 The bone graft is morsellized and only cancellous bone fragments (2 x 3 mm) are used. Permanent fixation of the midfoot is not easy due to osteopenia and the irregular size and shapes of the remaining bones and joints. Although the insertion of crossed pins from the first and fifth metatarsals is relatively easy, this fixation construct is not very stable, and I use compression lag screws. One screw which is always helpful in these chronic midfoot reconstructions is a 6.5-mm lag screw introduced from the medial cuneiform transversely across the foot into the cuboid. The other screws are introduced from the medial and lateral aspects of the foot obliquely. Cannulated screws are much easier to insert since the guide pins can be introduced and radiographs obtained before inserting the screws. Occasionally threaded Steinmann pins are needed due the orientation and quality of the bones. Threaded pins are preferable since they cause less motion at the skin interface and are therefore less likely to be associated with pin-tract infections. These are left in place for approximately 3 months and should be used cautiously since they are left protruding from the skin.
Chronic Midfoot Neuroarthropathy
The need for arthrodesis of the midfoot in the setting of chronic neuroarthropathy is less common. Most feet can be successfully treated with accommodative orthoses in a wide, extra-depth shoe with a rocker bottom sole. If this type of shoe is unsatisfactory, a custom molded shoe with a steel shank or a molded ankle foot orthoses can be used. Unlike the unstable foot seen in subacute neuroarthropathy, these feet are stiff and generally stable, so that the deformity can usually be protected with the appropriate shoe. If recurrent ulceration occurs, tarsal ostectomy is an alternative form of treatment. Unlike the hindfoot and ankle, ostectomy of the midfoot is a good option and should be considered when recurrent ulceration occurs. Ostectomy only works, however, where the arthropathy is truly chronic and the deformity is rigid and stable and if no pseudoarthrosis nor false motion occurs across the site of the original fracture dislocation process. Therefore, this is generally easier to treat since the foot can be fitted with an appropriate accommodative shoe and orthosis. If recurrent ulceration occurs despite this treatment program, then ostectomy is indicated and the offending bone is shaved down or removed. This may involve the cuneiform, cuboid, navicular, or a combination of these bones.
Rarely, patients with stable chronic neuroarthropathy have pain. Although we do not associate pain with neuroarthropathy, this may occur when severe forefoot abduction and midfoot pronation is present. The forefoot is usually abducted, the midfoot pronated, and the hindfoot in equinus. All of these should be corrected as described above for subacute deformity. In most of these feet, the medial soft-tissue structures of the hindfoot are stretched out, including the posterior tibial tendon, spring ligament, talonavicular capsule and deep deltoid ligament. I have seen perfect correction of the midfoot with a reconstructive arthrodesis only to find that the hindfoot later collapses into severe planovalgus. This is not necessarily a reactivation of the neuroarthropathy, but more likely represents partial treatment of the initial deformity. Following correction of the midfoot, attention should be directed to the medial hindfoot. If from the preoperative planning, these structures were noted to be attenuated, then soft tissues are corrected as one would do with a posterior tibial tendon reconstruction. In these patients, I perform an osteotomy of the calcaneus, and translate it medially. This is performed in conjunction with a flexor digitorum longus tendon transfer to the navicular or cuneiform. An alternative would be to perform is a subtalar or transverse tarsal arthrodesis simultaneously. The combination of hindfoot and midfoot arthrodesis is, however, required infrequently.
Hindfoot and Ankle Deformity
Regardless of the magnitude of the varus or valgus deformity, if the ankle is stable, then bracing usually succeeds. If the foot can be maintained under the axis of the leg during the acute or subacute phases of neuroarthropathy, despite bone dissolution or destruction, the foot will usually be stable. This, of course, requires prolonged immobilization until the consolidation phase has been reached. This contrasts with the foot where gross dislocation of the tarsus or ankle is present, which may occur during either the acute or chronic phases of neuroarthropathy. Operative treatment should as a rule be performed only once a chronic phase is reached. During the acute hyperemic stage, fragmentation of bone occurs and adequate rigid internal fixation is not usually possible. However, some feet do not reach a chronic phase of bony consolidation. Dislocation or subluxation persists, and in these feet reconstruction is performed during the subacute stage of arthropathy.
The indications for arthrodesis in the hindfoot and ankle are limited to the severely unstable joint that is not amenable to bracing. In this group of patients are also those for whom bracing has been attempted but repeated ulceration occurs (Fig. 2). Surgery is also occasionally indicated during the acute phase of hindfoot neuroarthropathy associated with an acute fracture or dislocation followed by disorderly fragmentation and loss of alignment. An acute fracture dislocation of the ankle in a patient with diabetes which is amenable to open reduction and internal fixation should be treated in the same manner as patients without neuropathy. Diabetes is not a contraindication to internal fixation of acute fractures of the ankle. In fact, I would exercise even more caution to ensure that the fracture heals uneventfully. Unless these fractures are immobilized postoperatively for a long period of time, malunion and fragmentation of the talus may occur. I prefer to sacrifice full range of motion in these patients by immobilizing them until complete healing with no signs of any warmth nor swelling are evident. The patient is not allowed to bear weight for 3 months, and for a further 3 to 4 months cast immobilization is continued until warmth and swelling has dissipated. Despite this caution, in patients with neuropathy, deformity may occur very rapidly, and an early arthrodesis should be considered before inevitable ulceration and infection occurs, as in the patient shown in Figure 3 who was successfully treated with open reduction and internal fixation for a displaced bimalleolar fracture of the ankle. Weight-bearing on the ankle was begun at 2 months when healing was thought to be present. The patient presented for treatment with significant deformity and a superficial ulcer over the medial malleolus (Fig. 3). The same principles apply to an acute dislocation of the hindfoot and tarsal joints where anticipated deformity is ultimately going to be difficult to brace. An example of this situation is presented in a patient with acute neuropathic dislocation of the naviculum with considerable deformity (Fig. 4). The alternatives for treatment in this patient would be to follow the course of neuroarthropathy with limb elevation, immobilization, and bracing in the hope of avoiding eventual ulceration. With this deformity, however, it is highly likely that ulceration will occur due to the prominence of the naviculum on the plantar medial aspect of the foot. Salvage could be performed later with excision of the naviculum or a later arthrodesis. However, it is likely that with time, the deformity will increase, which may subsequently be difficult to salvage. Early surgery may therefore be indicated.
As for the midfoot, it is preferable not to operate on the hindfoot and ankle in the absence of an open wound. This is not always easy, since the nature of the instability causing and also perpetuating the wound may preclude the possibility of adequate skin coverage. In these patients, strict bed rest and a split-thickness skin graft may be used to obtain coverage followed by reconstructive surgery as soon as is feasible before breakdown occurs again. If the hindfoot is grossly unstable, or if osteomyelitis is present, an external fixator is used to achieve temporary stability and once the wound has settled down, coverage is obtained followed by arthrodesis. Osteomyelitis is treated aggressively with debridement and appropriate antibiotics, and the definitive surgery is delayed until completely resolved. External fixation, however, should not be used routinely to secure fixation, since the risk of infection is markedly increased. The only time therefore that I use external fixation is when severe focal sepsis is present that cannot be managed by other means. I prefer to use a rigid internal fixation for the hindfoot and ankle with large cannulated 6.5- or 7.0-mm cancellous screws (Fig. 5).
Prophylactic intravenous antibiotics and a pneumatic tourniquet should be used for the hindfoot and ankle. A standard approach for a triple arthrodesis is used with a lateral incision dorsal to the peroneal tendons and a second dorsomedial incision medial to the anterior tibial tendon. A transfibular lateral approach is used for all tibiotalar fusions, and extended distally toward the sinus tarsi if a tibiotalocalcaneal or pantalar fusion is performed. A small medial incision is used to explore the medial ankle and resect the medial malleolus and debridement of the talonavicular joint when it is included in the fusion. Full-thickness skin flaps should be preserved wherever possible. Congruent surfaces are fashioned in the ankle and hindfoot to allow good bony contact and maximum inherent stability.
If the talus is fragmented and avascular, then a talectomy and tibiocalcaneal fusion is performed (Fig. 5). If talectomy is to be performed, then supplemental bone graft should be used. This has been reported successfully using a femoral head allograft.14 Due to the plantar inclination of the posterior facet the calcaneus cannot be apposed directly to the tibial plafond. If this is attempted, the hindfoot and calcaneus dorsiflex into a calcaneus position. Instead, a triangular-shaped bone graft needs to be inserted into the space created by resection of the joint surfaces. This can either be fashioned using a strut of the fibula, femoral head allograft or with trapezoidal pieces of tricortical bone from the iliac crest. This is augmented with copious cancellous bone chips. One way to obtain copious cancellous bone is to use a morsellized fibula graft by grinding down the fibula using a small acetabular reamer. One of the more difficult problems arises when determining the final position in which the foot is to be fused. This is particularly a problem following talectomy. The height of the foot is shortened, and the soft tissue bulk on the medial and lateral ankle obscures and distorts the alignment. For this reason, the limb should be draped proximal to the knee joint to allow full visualization of the foot and knee prior to definitive fixation. The foot is positioned plantigrade with the ankle in neutral dorsiflexion, 5o of hindfoot valgus, and slight external rotation. While slight malunion is tolerated by the patient with neuropathy, varus deformity, however mild, will lead to ulceration along the lateral border of the foot. Temporary fixation is obtained using cannulated pins, and the position of the foot is checked with anteroposterior and lateral radiographs before definitive fixation with long, partially threaded cancellous screws. The wound is closed in layers and a large bulky dressing is applied, with a plaster splint incorporated into the bandage.
Postoperative Management
Prophylactic intravenous antibiotics are utilized for 48 hours, the limb is elevated, and strict bed rest is enforced. Oral antibiotics are used until the wound is clean or the sutures are removed. Patients commence protected ambulation on the second postoperative day. No weight is allowed on the affected extremity for 3 months, and the foot is immobilized in a short leg cast. If patients are not able to comply with these restrictions midfoot procedures, a rubber heel can be attached posteriorly on the cast to minimize pressure under the midfoot. Weight-bearing is allowed when the warmth and swelling in the foot decreases, usually by 3 months. A reliable method of determining reduction of warmth is to use a skin thermistor and chart the changes in temperature every 2 to 3 weeks with the cast changes. The cast should be changed every 2 weeks during the first 6 weeks postsurgery and then at 3-week intervals. At approximately 3 months, weight bearing is commenced in a short leg cast for a further 3 months for the midfoot, and for 6 to 12 additional months for the hindfoot and ankle. Once weight bearing has commenced, the foot should be inspected more regularly. After midfoot procedures, patients should ambulate in a shoe with a rocker bottom sole, and a protective molded orthoses that should be provided and ready for the patient once casting is discontinued. We usually take the mold for the orthosis 4 weeks prior to the anticipated time for discontinuing the cast. A brace is not usually necessary for the midfoot, but is preferable for the hindfoot and ankle, where it is continued indefinitely.
References
1. Griffiths HJ, Diabetic osteopathy, Orthopedics (1985) 8(3):401-406.
2. Kristiansen B, Ankle and foot fractures in diabetics provoking neuropathic joint changes, Acta Orthop Scand (1980) 51:975-979.
3. Brodsky JW, Wagner FW, Kwong PK, et al, Patterns of breakdown in the Charcot tarsus of diabetes and relation to treatment, Orthop Trans (1987) 2:484.
4. Sinha S, Munichoodappa CS, Kozak GP, Neuroarthropathy (Charcot joints) in diabetes mellitus, Medicine (1972) 51:191-210.
5. Eichenholtz SN, Charcot Joints (Charles C Thomas: Springfield (IL), 1966).
6. Papa J, Myerson MS, Girard P, Salvage with arthrodesis in intractable diabetic neuropathic arthropathy of the foot and ankle, J Bone Joint Surg (1993) 75A(7):1056-1066.
7. Myerson MS, Henderson M, Short K, The treatment of neuropathy of the midfoot, Foot Ankle (1994), in press.
8. Clohisy DR, Thompson RC,Jr., Fractures associated with neuropathic arthropathy in adults who have juvenile-onset diabetes, J Bone Joint Surg (1988) 70A(8):1192-1200.
9. Harris JR, Brand PW, Patterns of disintegration of the tarsus in the anaesthetic foot, J Bone Joint Surg (1966) 48B:4-16.
10. Johnson JTH, Neuropathic fractures and joint injuries. Pathogenesis and rationale of prevention and treatment, J Bone Joint Surg (1967) 49A:1-30.
11. Lesko P, Maurer RC, Talonavicular dislocations and midfoot arthropathy in neuropathic diabetic feet. Natural course and principles of treatment, Clin Orthop (1989) 240(Mar):226-231.
12. Newman JH, Spontaneous dislocation in diabetic neuropathy. A report of six cases, J Bone Joint Surg (1979) 61B(4):484-488.
13. Stuart MJ, Morrey BF, Arthrodesis of the diabetic neuropathic ankle joint, Clin Orthop (1990) 253(Apr):209-211.
14. Myerson MS, Alvarez RG, Brodsky JW, et al, Symposium: Neuroarthropathy of the foot, Contemp Orthop (1993) 26(1):43-64.
15. Saltzman CL, Johnson KA, Goldstein RH, et al, Pate;;a tendon bearing brace as treatment for neurotrophic arthropathy. A dynamic force monitoring study, Foot Ankle (1992) 13:14-21.
16. Cleveland M, Surgical fusion of unstable joints due to neuropathic disturbance, Am J Surg (1939) 43:580-584.
17. Cohn BT, Brahms MA, Diabetic arthropathy of the first metatarsal cuneiform joint. Introduction of a new surgical fusion technique, Orthop Rev (1987) 16(7):465-470.
18. Shibata T, Tada K, Hashizume C, The results of arthrodesis of the ankle for leprotic neuroarthropathy, J Bone Joint Surg (1990) 72A(5):749-756.
19. Myerson M, Wilson K, Management of neuropathic ulceration with the total contact cast. In: Sammarco GJ, ed, The Foot in Diabetes (Lea & Febiger: Philadelphia, 1991), 145-152.
20. Myerson M, Papa J, Eaton K, et al, The total-contact cast for management of neuropathic plantar ulceration of the foot, J Bone Joint Surg (1992) 74A(2):261-269.
21. Myerson MS, Henderson MR, Clinical applications of a pneumatic intermittent impulse compression device after trauma and major surgery to the foot and ankle, Foot Ankle (1993) 14(4):198-203.
Figure Legends
Fig. 1. Acute neuropathic dislocation of the tarsometatarsal joint associated with swelling of the foot.
Fig. 2. Chronic dislocation of the transverse tarsal joint is demonstrated. This dislocation was grossly unstable, and not braceable (A). This was treated with a modified pantalar arthrodesis by resecting a portion of the talus (B, C).
Fig. 3. Acute neuropathic dislocation of the ankle after open reduction and internal fixation for an ankle fracture associated with an ulcer (A). Dislocation was present on the anteroposterior and lateral radiographs (B, C). This was treated with a tibiotalocalcaneal arthrodesis which was quite stable 5 years later (D).
Fig. 4. This patient (also presented in Fig. 1) sustained an acute dislocation of the tarsometatarsal joints. On the anteroposterior (A) and lateral (B) radiographs, medial and dorsal dislocation of the medial cuneiform was present which could cause skin necrosis. This was treated with open reduction and primary arthrodesis of the tarsometatarsal and midtarsal joints (C, D).
Fig. 5. A brace was used unsuccessfully in this patient to control the deformity of the hindfoot. Anteroposterior and lateral radiographs demonstrate the fragmentation of the talus, and varus dislocation of the ankle (A, B). This was treated by resecting the body of the talus and a tibiocalcaneal arthrodesis (C).
|
 |
