THE FOOT: Neuroarthropathy
February 18th, 1997
R. M. Marks, M. S. Myerson
Orthopaedic management of the patient with neuroarthropathy can be problematic, from both diagnostic and treatment standpoints. Accurate diagnosis depends on incorporation of a high index of suspicion for patients at risk. Once the diagnosis is made, management (surgical or non-surgical) can proceed in a logical fashion based on the patient's clinical and radiographic presentation. The treatment goal in this population is to create a stable plantigrade foot and ankle, which is either shoeable or braceable.
Orthopaedic management of a patient with neuroarthropathy is often problematic, from both diagnostic and treatment standpoints. This process is labor-intensive, is often associated with complications, and frequently produces less than satisfactory results. The initiation of appropriate treatment depends on the prompt identification not only of the diagnosis of neuroarthropathy, but also of the patient ‘at risk’. Subsequent treatment is based on the patient's clinical presentation, i.e. the presence (or absence) of ulceration, edema, hyperemia, erythema, and/or co-morbid conditions, and on whether the event was acute or chronic. The use of classification schemes that take into account the clinical presentation as well as radiographic appearance can then help formulate a treatment algorithm for the patient with neuroarthropathy.
Background and Etiology
Neuropathic changes were first described in a syphilitic patient by Jean Martin Charcot1 in 1868. The initial recognition of the association between neuroarthropathy and diabetes was made by Jordan2 in 1935. Today, diabetes, syphilis and syringomyelia are the most common disease processes associated with neuropathy, although many other causes can also be implicated.3-10
Neuropathy affects between 6.0 and 41.6% of the diabetic population,11-13 and neuroarthropathy occurs in 0.1-2.5% of these patients.14,15 Factors associated with the development of peripheral neuropathy are varied and not well-defined, although duration of diabetes and advanced age are positive correlates.16, 17
There is no predilection for the development of neuropathy between males and females or between juvenile-onset (Type 1) and adult-onset (Type 11) diabetes. It has been shown that patients with poor glucose control develop peripheral neuropathy at a significantly higher rate18 and that rigorous control of serum glucose may decrease clinical neuropathy by as much as 64%.19,20 A correlation between glucose control and subsequent neuroarthropathic changes, however, has yet to be proven. In an animal model, fracture healing was favorably altered by proper glucose control.21
There is no specific profile for the patient at risk for developing neuroarthropathy. Although such a patient typically has had diabetes for more than 10 years,22 neuroarthropathic injuries can also signify the initial presentation of diabetes.23,24 There is a 30% incidence of bilateral involvement,25 perhaps due to protective weightbearing, with increased stresses being placed on the contralateral extremity.
Those patients with associated renal disease appear to be at higher risk for the development of neuroarthropathy, perhaps due to associated osteopenia and the use of immunosuppressive drugs.9,26 This population also has a higher incidence of complications associated with their treatment, as they are more prone to delayed union, malunion, wound complications and infections. 27
The etiology of neuropathic fracture-dislocations is not well understood, but the one constant is that peripheral neuropathy is universally associated with these injuries. Both neurotraumatic and neurovascular theories have been proposed for the pathogenesis of these injuries. Johnson's28 neurotraumatic theory states that decreased protective sensation permits cumulative mechanical trauma, or 'microtrauma', resulting in fracture or dislocation. This theory, however, does not account for the fact that some patients still have pain with neuropathy. Brower and Allman's29 neurovascular theory proposes that this population exhibits increased blood flow with osteoelastic bone resorption secondary to a neurally initiated vascular reflex, or 'autosympathectomy'. The initial atrophic, destructive changes are followed by a hypertrophic, reparative phase with new bone formation and coalescence. In actuality, it would appear that destructive neuroarthropathic changes can be attributed to both etiologies. Neither theory, however, adequately explains the mechanism of acute ligamentous instability of the midfoot or hindfoot in the absence of fracture; perhaps both neurotrauma and hypervascularity play a role.
Acute neuroarthropathy is classically characterized as a painless deformity associated with erythema and swelling. As this process becomes quiescent, the erythema and swelling subside, and the previously fragmented bone coalesces and becomes more stable. Unfortunately, the diagnosis is either delayed or missed in almost 25% of patients30 since a known history of predisposing trauma may or may not exist.23 Occasionally, the patient recalls a minor twisting injury or contusion that predates the changes of neuroarthropathy.31 Contrary to the common belief that neuroarthropathic injury is a painless condition, pain in the extremity may be present.32
All too often patients with acute neuroarthropathy are initially incorrectly diagnosed with cellulitis or osteomyelitis and are managed with intravenous antibiotics, biopsy and multiple imaging studies. Differentiating these conditions can be aided by laboratory findings (glucose, white cell count, sedimentation rate, uric acid level), radiographic assessment and elevating the limb for 4-6 h, which will decrease the swelling in the patient with neuroarthropathy but will not affect that associated with infection.33-35
One must identify the at-risk diabetic patient who has sustained an acute sprain or fracture. Although the initial clinical and radiographic presentations may be the same as those of the general population, these patients must be recognized as being at risk for the development of neuroarthropathy (Fig. 1). Subsequent treatment of injuries must be modified by prolonged immobilization, typically twice that of the general population, to prevent neuroarthropathic changes. Although these treatments are modified as described below, they should not be compromised due to diabetes since the later development of neuroarthropathy only further complicates treatment.
In a classic treatise, Eichenholtz36 described the three stages of neuroarthropathy based on both clinical and radiographic criteria. In Stage I, an acute inflammatory process is present, characterized by edema, hyperemia and erythema. Radiographically, bone fragmentation is typically seen, although very little change may also be noted since joint subluxation and dislocation are also possible and, at times, quite subtle. With Stage II begins the reparative process, with diminution of inflammatory changes and coalescence of bone at the site of the fracture or dislocation. Stage III entails radiographically evident quiescence of the inflammation and bony consolidation. Stage III, by definition, implies stability - which, however, may not be present despite bony coalescence and the absence of inflammation and swelling. If instability is present, the disease is, by definition, in the Stage II category.
Brodsky32 developed an anatomic classification of neuroarthropathic joints that recognizes the potential for later healing and thereby impacts on treatment and prognosis. Type I occurs in 60-70% of the patients, involves the midfoot, and is rarely unstable. If untreated, however, it results in plantar bony prominences that can lead to ulceration. Type II, which involves the hindfoot and represents approximately 20% of neuroarthropathic joints, is characterized by subluxation and instability of the peritalar joints and requires long periods of immobilization. Type IIIA involves the ankle and results in the most severe pattern of instability, requiring the longest period to bony healing. Type IIIB, which involves the os calcis, presents with a pathologic fracture of the tuberosity of the calcaneus. Progressive pes planus may develop secondary to tuberosity elevation and tendo Achillis incompetence.
The treatment algorithm is determined by the clinical stage of the injury, the radiographic characteristics of the fracture, the location and magnitude of the deformity, the presence of infection or ulceration, and the presence of dislocation or subluxation, which can potentially compromise soft tissues. Co-morbid conditions, such as limb ischemia as well as ulceration with the possibility of osteomyelitis, must be taken into account. The patient's general medical status, as well as pre-existing levels of activity, are additional factors. The goal of treatment is to create a stable, mechanically sound foot that is plantigrade and that can accommodate a shoe or, if necessary, a brace. Ideally, the initial treatment should aim at preventing bony prominences that can lead to ulceration and subsequent abscess or osteomyelitis.
Treatment of acute (Stage 1) neuroarthropathy relies on initial elevation and immobilization of the limb, combined with weightbearing restrictions. The duration of restricted weightbearing varies from 1 to 3 months, depending on the site and magnitude of the deformity. For midfoot and ankle arthropathy, weightbearing typically begins at about 6 weeks and 3 months, respectively. Immobilization is best achieved by the application of a total contact cast, which serves to control edema as well as uniformly distributing pressures about the foot.37-39 This cast generally is changed in 7-10 days due to the often dramatic reduction in swelling. Casts are subsequently changed every 2-3 weeks until the acute inflammation has subsided. More recently, removable foot devices have gained popularity and, in our experience, have a role in treatment.
The transformation to a subacute (Stage II) situation is indicated by small fluctuations in swelling and the radiographic presence of early bony healing. At this point, the patient may be placed into a total contact or ‘clam shell’ ankle foot orthosis. This, typically, is a polypropylene construct lined with plastizote.40 Buckles or straps are used to hold the two halves together.
Treatment of chronic (Stage III) neuroarthropathy is undertaken once bony healing is evident, there is no further collapse, and inflammation has subsided. Custom-molded inner soles are fabricated to accommodate any bony prominence and/or the particular shape of the foot or ankle. These are typically placed inside athletic or custom-made shoes. At least two layers are used to fabricate the molded inner soles. Usually the top layer is composed of polyethylene foam material and the bottom layer is made of a more durable substance. Rigid orthoses should be avoided and commonly rocker-bottom modifications are required on the shoes. This protocol permits adequate treatment for up to 70% of neuroarthropathic joints.41
Surgical intervention is required if:
1. The foot or ankle is grossly unstable and cannot be maintained in a plantigrade position with either shoe wear modification or bracing
2. An acute fracture-dislocation is irreducible and likely to cause skin necrosis with reduction of edema and resumption of weightbearing
3. Recurrent ulcerations occur secondary to bony prominences.
In the setting of acute (Stage I) neuroarthropathy, surgery should be performed only for a severe dislocation which is unstable or where it is anticipated that the resultant tension on the soft tissues would lead to skin necrosis and ulceration, with the subsequent risk of developing osteomyelitis. In this acute phase, however, soft-tissue swelling and bony fragmentation associated with hyperemia and osteopenia generally preclude surgical intervention, with the exception of a severely unstable fracture-dislocation (Fig. 2).
Subacute (Stage 11) neuroarthropathy is characterized by pseudarthrosis, motion through the midfoot (associated with rocker-bottom deformity), and, frequently, associated tendo Achillis contracture with hindfoot equinus. It is in this clinical stage that many reconstructive procedures are performed (Fig. 3). A tendo Achillis lengthening is performed percutaneously in conjunction with the midfoot or hindfoot arthrodesis.41
Once a chronic (Stage 111) situation has been reached, the foot is stiff, stable and amenable to the use of a molded orthosis. At times, however, a shoe or brace is not sufficient for an advanced deformity and reconstructive surgery is required. Exostectomy may be required if a patient presents with a recurrent non- infected ulceration, typically on the plantar surface of the midfoot.42,43 Wide but judicious resection of the offending bony prominence is recommended. Occasionally the exostectomy may lead to further collapse due to resection of capsuloligamentous supporting structures. This circumstance is unusual in the chronic phase but, if the midfoot is unstable, exostectomy should be combined with a reconstructive procedure.44
In this at-risk population, the indications for the treatment of acute fractures, particularly of the ankle, remain the same as those for the general population, i.e. individuals without neuropathy. If open reduction is required, the fixation should be satisfactory due to the tendency toward poor bone quality. Although the same principles apply to midfoot injuries, the latter should be treated with primary arthrodesis, not open reduction and internal fixation.41 With this population, extended non-weightbearing and immobilization is necessary In general, a good rule of thumb is that the immobilization period for the neuropathic population should be double that in the non-neuropathic population.
Presurgical intravenous antibiotic prophylaxis (cephalosporin) should be administered. If there is a possibility of associated infection, such as in the presence of deep ulceration,45 then Gentamycin should be administered for coverage of Gram-negative organisms. Definitive surgical treatment may have to be staged so that the infection can be eradicated.
One should avoid surgical intervention in the presence of acute neuroarthropathy with bony fragmentation.41,44 Hyperemia poses a problem with excessive bleeding and, due to osteopenia, achieving rigid fixation may not be feasible. Since the skin quality is frequently poor, full-thickness skin incisions should be made directly to the periosteum. Although doing so carries a potential risk of damaging superficial sensory nerves, this is, in fact, rarely a problem because of the dense neuropathy that is commonly present. Once the periosteum is reached, the resection of bone and fibrous tissue should be aggressive and extensive enough to achieve realignment.
Neuroarthropathic fracture-dislocation requires primary arthrodesis due to poor bone quality and capsuloligamentous laxity. In the acute stage, surgery is performed for severe dislocations that are unstable and manually reducible. Given the poor quality of bone that is invariably present, rigid internal fixation may be extended into uninvolved joints to achieve adequate bony purchase with screws. Supplemental fixation (intramedullary rod or external fixator7,46) is at times necessary, particularly for hindfoot or ankle reconstructive procedures. External fixation, however, should be used judiciously due to the increased rate of infection associated with its use in diabetics.46,47 It is frequently necessary to bone graft large structural defects either with femoral head allograft or autologous iliac crest graft.44 Pseudarthrosis rates vary from 25 to 75%, but fibrous ankylosis can produce an acceptable result.48
Post-surgically, drains are left in place for 24-48 h. Antibiotics are continued for 48 h. The sine qua non of neuroarthropathic care is restriction of activity and prolonged immobilization. The length of activity restriction is determined by the location of the injury.49 Typically, all patients are kept non-weight-bearing for 3-4 months, after which time protected weightbearing commences. Midfoot procedures or injuries require a total of 6 months of immobilization, and hindfoot or ankle procedures or injuries require an additional 6-12 months of immobilization (Fig. 4). Since pseudarthrosis may be problematic, long-term bracing and/or shoe wear modifications are frequently necessary, particularly after hindfoot and/or ankle procedures or injuries.
Orthopaedic care of the neuroarthropathic patient depends on a prompt, accurate diagnosis. A treatment algorithm based on the clinical and radiographic presentations, as well as the existence of any co-morbid conditions, maximizes the possibility for a successful treatment outcome. The ultimate goal for the neuroarthropathic patient, be it by non-surgical or surgical intervention, is the creation of a stable, plantigrade foot and ankle that are shoeable or braceable.
Charcot J M. Sur quelques arthropathies qui paraiissent dependre d’une lesion du cerveau ou de la moelle epiniere. Arch Physiol 1868; 1:161-178 r, 570-900.
Jordan W R. Neuritic manifestations in diabetes mellitus. Arch Intern Med 1936; 57:307-366.
Jacobs R L, Karmody A. Charcot foot. In: Jahss MH, ed. Disorders of the Foot, Philadelphia: Wb Saunders Co, 1982; 1248-1397.
Medhat M A, Krantz H. Neuropathic ankle joint in Charcot-Marie-Tooth disease after triple arthrodesis of the foot. Orthop Rev 1988; 17(9):873-880.
MacEwan G D, Floyd G C. Congenital insensitivity to pain and its orthopaedic implications. Clin Orthop 1970; 68:100-107.
Horibe S, Tada K, Nagano J. Neuroarthropathy of the foot in leprosy J Bone Joint Surg 1988; 70B (3):481-485.
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.
Wyrsch B, McFerran M, Johnson K, Harper M, McAndrew M, Schwartz H. A randomized, prospective study comparing the complications encountered in the management of pilon fractures. Presented at the Annual Meeting of the Orthopaedic Trauma Association, New Orleans (LA), September 23, 1993.
Meneghello A, Bertoli M. Neuropathic arthropathy (Charcot's joint) in dialysis patients. ROFO Fortschr Geb Rontgenstr Nuklearmed 1984; 141:180-184.
Thornhill HL, Richter RW, Shelton ML, Johnson CA. Neuropathic arthropathy (Charcot forefeet) in alcoholics. Orthop Clin North Am 1973; 4:7-20.
Tesfaye S, Malik R, Ward JD. Vascular factors in diabetic neuropathy. Diabetologia 1994; 37:847-854.
Boulton AJ, Knight G, Drury J, Ward JD. The prevalence of symptomatic, diabetic neuropathy in an insulin-treated population. Diabetes Care 1985; 8:125-128.
Kumar S, Ashe HA, Pamell LN, et al. The prevalence of foot ulceration and its correlates in type 2 diabetic patients: a population-based study. Diabet Med 1994; 11:480-484.
Griffiths HJ. Diabetic osteopathy. Orthopedics 1985; 8(3):401-406.
Lippmann HI, Perotto A, Farrar R. The neuropathic foot of the diabetic. Bull N-Y Acad Med 1976; 52(10):1159-1178.
Knuiman MW, Welborn TA, McCann VJ, Stanton KG, Constable IJ. Prevalence of diabetic complications in relation to risk factors. Diabetes 1986; 35:1332-1339.
Kuroda N, Taniguchi H, Baba S, Yamamoto M. Relationship between age and autonomic neuropathy in diabetes mellitus. Diabetes Res Clin Pract 1990; 9:49-53.
Sotelo JR, Horie H, Ito S, Denech C, Sango K, Takenaka T. An in vitro model to study diabetic neuropathy. Neurosci Lett 1991; 129:91-94.
The Diabetes Control and Complications Trial Research Group. The effect of intensive diabetes therapy on the development and progression of neuropathy. Ann Intem Med 1995; 122:561-568.
Feldman EL, Stevens MJ. Clinical testing in diabetic peripheral neuropathy. Can J Neurol Sci 1994; 21:S3-7.
Macey LR, Kana SM, Jingushi S, Terek RM, Borretos J, Bolander ME. Defects of early fracture-healing in experimental diabetes. J Bone Joint Surg 1989; 71 A: 722-733.
Cofield RH, Morrison MJ, Beabout JW. Diabetic neuroarthropathy in the foot: patient characteristics and patterns of radiographic change. Foot Ankle 1983; 4(l):15-22.
Kristiansen B. Ankle and foot fractures in diabetics provoking neuropathic joint changes. Acta Orthop Scand 1980; 51(6):975-979.
Pedersen LM, Madsen OR, Bliddal H. Charcot arthropathy as an unusual initial manifestation of diabetes mellitus [letter]. Br J Rheumatol 1993; 32:854-855.
Brodsky JW. The diabetic foot. In: Mann RA, Coughlin MJ, eds. Surgery of the Foot and Ankle, 6th edn. St. Louis: Mosby-Year Book Inc, 1993; 877-958.
Thompson RC, Jr., Havel P, Goetz E Presumed neurotrophic skeletal disease in diabetic kidney transplant recipients. JAMA 1983; 249:1317-1319.
Clohisy DR, Thompson RC, Jr. Fractures associated with neuropathic arthropathy in adults who have juvenile-onset diabetes. J Bone Joint Surg 1988; 7OA(8): 1192-1200.
Johnson JTH. Neuropathic fractures and joint injuries. Pathogenesis and rationale of prevention and treatment. J Bone Joint Surg 1967; 49A(l): 1-30.
Brower AC, Allman RM. Pathogenesis of the neurotrophic joint: neurotraumatic vs. neurovascular. Radiology 1981;139(2):349-354.
Lose G, Lindholm R Transient painful osteoporosis of the hip in pregnancy. Int J Gynaecol Obstet 1986; 24:13-16.
Slowman-Kovacs SD, Braunstein EM, Brandt KD. Rapidly progressive Charcot arthropathy following minor joint trauma in patients with diabetic neuropathy. Arthritis Rheum 1990; 33:412-417.
Brodsky J-W. Management of Charcot joints of the foot and ankle in diabetes. Semin Arthroplasty 1992; 3:58-62.
Gold RH, Bassett LW. Radiologic evaluation of the arthritis foot. Foot Ankle 1982; 2:332-341.
Levine SE, Neagle CE, Esterhai JL, Wright DG, Dalinka MK. Magnetic resonance imaging for the diagnosis of osteomyelitis in the diabetic patient with a foot ulcer. Foot Ankle Int 1994; 15:151-156.
Yuh NW, Corson JD, Baraniewski HM, et al. Osteomyelitis of the foot in diabetic patients: evaluation with plain film, 99mTc-MDP bone scintigraphy, and MR imaging. AJR Am J Roentgenol 1989; 152:795-800.
Eichenholtz SN. Charcot Joints, Springfield (IL): Charles C Thomas, 1966.
Pring DJ, Casiebanca N. Simple plantar ulcers treated by below-knee plaster and moulded double-rocker plaster shoe - a comparative study Lepr Rev 1982; 53(4):261-264.
Laing PW, Cogley DI, Kienerman L. Neuropathic foot ulceration treated by total contact casts [see comments]. J Bone Joint Surg 1992; 74B: 1 33-136.
Shaw J E, Boulton A J M. The Charcot foot. The Foot 1995; 5:65-70.
Morgan JM, Biehl WC, 111, Wagner FW, Jr. Management of neuropathic arthropathy with the Charcot Restraint Orthotic Walker. Clin Orthop 1993; 296(Nov):58-63.
Myerson MS, Henderson MR, Saxby T, Short KW. Management of midfoot diabetic neuroarthropathy. Foot Ankle Int 1994; 15(5):233-241.
Brodsky J-W, Rouse AM. Exostectomy for symptomatic bony prominences in diabetic charcot feet. Clin Orthop 1993; 296:21-26.
Leventen EO. Charcot foot-a technique for treatment of chronic plantar ulcer by saucerization and primary closure. Foot Ankle 1986; 6:295-299.
Myerson MS, Alvarez RG, Brodsky JW, Johnson JE. Symposium: Neuroarthropathy of the foot. Contemp Orthop 1993; 26(l):43-64.
Wagner FW A classification and treatment program for diabetic, neuropathic, and dysvascular foot problems. Instr Course Lect 1979; 28:143-165.
Stuart MJ, Morrey BE Arthrodesis of the diabetic neuropathic ankle joint. Clin Orthop 1990; 253(Apr):209-21 1.
Rubin P, Casarett GW Radiopathologic basis of the radiosensitivity and radioresistance of tumors. In: Clinical Radiation Pathology, Philadelphia: WB Saunders Co, 1968; 894-933.
Papa J, Myerson M, Girard P Salvage, with arthrodesis, in intractable diabetic neuropathic arthropathy of the foot and ankle. J Bone Joint Surg 1993; 75A(7):1056-1066.
Brodsky JW, Wagner FW, Kwong PK, et al. Patterns of breakdown in the Charcot tarsus of diabetes and relation to treatment [abstr]. Orthop Trans 1987; 2:484.
Fig. I-A 35-year-old diabetic woman sustained a twisting injury, resulting in a lateral malleolar fracture (A, B). It was treated with a non-weightbearing cast for 4 months, at which point union was felt to have occurred (C, D). One year later, radiographs showed profound destruction of the joint that was, however, stable and braceable (E, F).
Fig. 2-A 37-year-old diabetic man complained of painless swelling and erythema in his left midfoot (A, B). He was initially treated with rest and intravenous antibiotics. Continued swelling and erythema prompted an orthopaedic surgical consultation for evaluation of a possible deep infection. Anteroposterior (C) and lateral (D) radiographs revealed extrusion of the medial cuneiform with marked naviculocuneiform sag. The medial skin was compromised. This acutely unstable midfoot fracture-dislocation was treated with reduction and primary arthrodesis (E, F).
Fig. 3-A 53-year-old diabetic woman experienced progressive midfoot collapse with rocker-bottom deformity not amenable to orthotic modifications (A, B). After recurrent ulceration refractory to nonoperative care, an extended midfoot fusion with plantar exostectomy, utilizing a plantar-applied plate, was performed (C, D).
Fig. 4-A 48-year-old woman with diabetic neuropathy sustained an impaction fracture of the tibial plafond with anterior subluxation of the talus (A, B). This remained unstable and a tibiotalocalcaneal fusion was performed with bone graft obtained from the fibula (C). One month after surgery, the patient experienced an oblique fracture through the distal tibia (D [arrow], E). She was non-compliant with nonweightbearing secondary to dense neuropathy. Prolonged immobilization in an articulated ankle-foot arthrosis resulted in a stable fusion (F). Figs C, D, E and F continued on p. 192.