Acute Charcot's Arthropathy of the Foot and Ankle

Transcription

1 Acute Charcot's Arthropathy of the Foot and Ankle harcot's joint (neuropathic osteoarthropathy) is a progressive condition affecting the musculoskeletal system and is characterized byjoint dislocation, pathologic fractures, and often debilitating deformities (Figs. 1 and 2).l The condition most commonly occurs in patients with diabetes mellitus who have severe peripheral neuropathies. The prevalence of Charcot's joint is variable, ranging from 0.16% of all patients with diabetes2 to as high as 13% of patients receiving care at a high-risk diabetic foot clinic.' The frequency of diagnosis of this condition appears to be increasing as a result of increased awareness of its signs and symptoms. The Etiology of Neuropathic Osteoarthropathy (Charcot's Joint) Neuropathic osteoarthropathy was first reported by Musgrave in 1703.We described it as an arthralgia secondary to venereal disease. In 1868, the noted French neurologist Jean-Martin Charcot became the first investigator to concisely describe the neuropathic component of the di~ease.~ Charcot linked the degenerative condition to syphilis, which was then a common malady.4 Syphilis was the disease most commonly associated with this type of arthropathy until 1936, when Jordan linked it to diabetes mellit~s.~ Since these first descriptions, numerous theories on its etiology have been pr~moted.~."~ Charcot believed that neuropathic osteoarthropathy was secondary to deficiencies in trophic centers in the spine.4 It was for this concept, along with his brilliant description of the malady, that neuropathic osteoarthropathy was subsequently renamed "Charcot's arthropathy." This spinal-centric view of Charcot's arthropathy, however, was not shared by all of Charcot's contemporaries. Volkman, Virchow, and other members of the "German school" vehemently opposed Charcot's theory, which they believed was based solely on observation and assumption. They believed that the etiology of Charcot's arthropathy was neurotraumatic in nature. That is, an insensate foot subjected to trauma would fracture and heal with exuberant bone formation. In testing this theory, Eloesser, in 1917, sectioned the posterior nerve roots to the forelimbs in 38 cats.' Following a period of activity, Eloesser noted neuropathic Key Words: Arthropathy, neurogenic; Joint diseases; Lowm extremity, ankle and foot. [Armstrong DG, Lavery LA. Acute Charcot's arthropathy of the foot and ankle. Phys Thm. 1998;78: David G Amtrung Lawrence A Lavety Physical Therapy. Volume 78. Number 1. January 1998

2 bony changes in 71% of the animals. Six decades later, Finsterbush and FriedmanQepeated Eloesser's experiments using a rabbit model. After sectioning the posterior nerve roots, the rabbits' hind limbs were casted. A difference was noted in the response to immobilization between normal and denervated groups. Finsterbush and Friedman concluded that trauma was important but not the primary factor leading to the deterioration of insensate joints." Finsterbush and Friedman's work%pened the way for further conjecture about the nature of Charcot's arthropathy. Subsequent investigatorsi0 hypothesized that trauma alone could not explain the sometimes striking osteopenia seen in patients with Charcot's arthropathy. This hypothesis led to the notion that increased blood flow was at least partially responsible for the arthropathy, causing a resorption of bone and a subsequent weakening of the supporting structure. Thus, fractures could be caused by even trivial stress. In an attempt to lend more credence to this concept, E:dmonds and coworkers,ll using scintigraphy, demonstrated that blood flow within bone was greater when neuropathy was present. This neurovascular theory has gained a great deal of favor among many clinicians treating patients with this condition. The actual etiology of Charcot's arthropathy may lie somewhere between these neurovascular and neurotraunlatic theories. The current theory suggests that, following the development of autonomic neuropathy, there is an increased blood flow to the extremity, resulting in osteopenia. Subsequently, motor neuropathies result in muscle imbalance, which places abnormal stress on the affected extremity, while sensory neuropathy renders the With the possible exception of Charcot's arthropathy is perhaps the most debilitating of diabetes mellifus. patient unaware of the often profound osseous destruction taking place with each step during amb~lation.~-~" Diagnosis of Acute C harcot's Arthropathy The initial diagnosis of acute Charcot's arthropathy is often based on ire found unilateral swelling, locally increased skin temperature, erythema, joint effusion, and bone resorp tion in an insensate foot. These characteristics, in the presence of intact skin and loss of protective sensation, are often pathognomonic of acute Charcot's arthropathy. Armstrong et all have also noted that there is some degree of pain in an otherwise insensate extremity in over 75% of patients with acute Charcot's arthropathy. Diagnosis is complicated by the fact that 40% of patients with acute Charcot's arthropathy have a concomitant ulceration,' thus raising the issue of whether there is contiguous oste~myelitis.~~ When faced with a warm, edematous, erythematous, insensate foot with a concomitant wound, clinicians may find it difficult to differentiate between acute Charcot's arthropathy and osteomyelitis solely on the basis of plain radiographs. Additional laboratory studies may prove to be useful in arriving at a correct diagnosis. The white blood cell count will often be elevated, with a left shift, on differential analysis in patients with acute osteomyeli- DG Armstrong, DPM, is Assistant Professor, Department of Orthopaedics, University of Texas Health Science Center, 7703 Floyd Curl Dr, San Antonio, TX (USA) (armstrong@usa.net). Address all correspondence to Dr Armstrong. L4 Lavery, DPM, is Associate Professor, Department of Orthopaedics, University of Texas Health Science Center. Physical Therapy. Volume 78. Number 1. January 1998 Armstrong and Lavery. 75

3 tis. Generalized white blood cell count elevation with a left shift is strongly indicative of the presence of an infectious disease such as osteomyelitis. The term "left shift" implies the premature release of immature polymorphonuclear leukocytes (band forms) into the circulation in response to overwhelming physiologic demand. This premature release of immature polymorphonuclear leukocytes into the circulation is also typical in infection. Although the erythrocyte sedimentation rate may also be elevated in the case of acute infection, the sedimentation rate increases in response to any inflammatory process and is therefore nonspecific. The white blood cell count with white cell differential assay remains the most clinically useful hematologic means of detecting infection, and a glaringly elevated white blood cell count or substantial left shift should alert the clinician to the possibility of sepsis. A white blood cell count or white cell differential assay that is within normal limits should not deter the clinician from instituting appropriate treatment. l5-i7 Technetium bone scans are expensive and nonspecific in assisting in the differentiation between osteomyelitis and acute Charcot's arthropathy.18 We do not believe that technetium scanning is very useful for the diagnosis of acute pedal sequelae to peripheral neuropathy. Indium scanning, although expensive, has been shown to be far more specific.iy Indium-1 11 scintigraphy may be used in two instances. First, indium-] 11 scin tigraphy may be used to initially assist in differentiation between osteomyelitis and Charcot's joint in the presence of a pedal ulcer. Second, several weeks following debridement of osteomyelitic bone, indium-1 11 scanning may provide some benefit in determining the adequacy of bony resection. If an indium-1 11 scan is returned positive, a bone biopsy is indicated to confirm the diagnosis of osteomyelitis and rule out Charcot's joint.' If a scan is returned negative, the presumptive diagnosis is Charcot's joint until proven otherwise. Additional studies currently used for assistance in differentiating Charcot's joint from osteomyelitis include bone scans utilizing white blood cells labeled with technetium hexamethyl propylenamine oxime and magnetic resonance imaging. Although imaging studies may be useful at many centers, including our own, we prefer to use a sterile blunt probe. Probing to bone, combined with radiographic and clinical evaluation, may be the most practical and costeffective means of diagnosing osteomyelitis prior to surgical debridement and definitive bone biopsy. If a wound is probed directly to bone, osteomyelitis is frequently assumed. This diagnosis may then be confirmed with a bone A bone biopsy is currently the "gold standard" by which all other diagnostic modalities are measured."' Bone biopsies have a very low complication profile and are less expensive than many advanced Figure 1. Clinical presentation of acute Charcot's arthropathy. imaging technique^.^^ A positive histologic diagnosis of Charcot's joint is less clinically important than a negative diagnosis of osteomyelitis because a prolonged course of parenteral antibiotics or surgical ablation may be obviated by a negative diagnosis. Nonetheless, a biopsy consisting of multiple shards of bone and soft tissue embedded in the deep layers of synovium is pathogne monic for neuropathic osteoarthropathy.2' The Classification of Charcot's A~hropathy The most common classification system used in the treatment of patients with Charcot's arthropathy was described by Eichenholz in 1966.Z3 This classification system is primarily radiographic in nature and is divided into developmental, coalescent, and reconstructive stages. The developmental stage is characterized by profound osseous destruction, with frequent dislocation. The coalescent stage is marked by evidence of repair of large fracture fragments. The reconstructive stage denotes bony ankylosis and often large amounts of hypertrophic proliferation. Although this system is very descriptive, it is not very clinically useful. Sanders and MrdjencovichZ4 introduced a classification system based on the location of arthropathy. Loosely based on Harris and Brand's classic this system is highly descriptive, and, because it denotes the location of the arthropathy, it is clinically useful. The reason that Sanders and Mrdjencovich's system is more clinically useful is that location is pivotal when considering potential complications and fracture healing. For instance, midfoot fractures, which frequently lead to a "rockerbottom" foot type where the majority of the patient's weight is on the midfoot, are often the most debilitating and result in permanent deformity. Using Sanders and Mrdjencovich's system for location of arthropathy, we further classify Charcot's arthropathy, based on radiographic,23 dermal therm0metric,2"~ and clinical signs,z8 as consisting of two treatment-oriented phases: (1) an 76. Armstrong and Lavery Physical Therapy. Volume 78. Number 1. January 1998

4 acute phase and (2) a postacute (quiescent) phase. The initial clinical diagnosis of acute Charcot's arthropathy (as described earlier) is well documented in the literat~re.~ Following resolution of acute neuropathic osteoarthropathy, patients a.re converted to the postacute phase, during which uncasted weight bearing is introduced (Fig. 3).l Management of Acute Charcot's Arthropathy lmmobilization and reduction of stress are essential in the treatment of patients with acute Charcot's arthropath~.l.~' Many investigators advocate no weight bearing, through the use of crutches or other assistive devices, during the acute phase of Charcot's arthropathy. Although this is an ~i~~~~ accepted form of treatment, a threepoint gait may increase pressure to the contralateral limb, thus predisposing it to repetitive stress, ulceration, and neuropathic fract~re.~" Armstrong et all reported that, through the use of appropriately applied total contact casts (Fig. 4), most patients may ambulate during the entire period of treatment. All patients at our center were initially treated with total contact casting.30 The total contact cast consists of an inner layer of plaster with thin felt applied to the malleoli and tibia1 crest and foam applied to the digits for protection. The outer splints and remaining layers are made of fiberglass, with an optional rubber cast plug secured to the plantar aspect of the cast to increase durability. Casts are routinely checked weekly and evaluated for proper fit. Casts of patients with concomitant ulceration (Fig. 5) are changed weekly for ulcer evaluation and debridement. Cast change intervals for patients w~thout ulcers are dependent on cast comfort and integrity (3 weeks maximum). Casting is discontinued based on clinical, radiographic, and dermal thermometric signs of quiescence. Skin temperatures are monitored using a portable infrared thermometric probe. Use of dermal thermometry in the diabetic foot has been well de~cribed.z~,~~ Patients with bilateral acute Charcot's arthropathy present a unique dilemma. Because of increased inflammation (and subsequent increase in temperature) on both sides, dermal thermometry is less effective in providing clinically useful information. These patients' lower limbs, therefore, remain in bilateral total contact casts until both feet and ankles normalize clinically and radiographically. Clinical signs of quiescence include reduction of edema and erythema and 2. ~ ~ d i ~ ~, nphic presentation of acute Charcot's arthropathy. return of skin lines to the foot. Radiographic signs of quiescence are evidenced by trabecular bridging on serial radiographs. Although the prevalence of bilateral arthropathy has been reported to be as high as two thirds of cases, our recent report of a large series of patients showed a prevalence of only 9%.1,",33 Perhaps more interesting than the prevalence of bilateral acute Charcot's arthropathy is whether contralateral Charcot's joint events occur during treatment. We have observed no contralateral events during treatment of patients with unilateral acute Charcot's arthropathy. Patients whose lower limbs are placed in these casts are able to ambulate freely during the majority of treatment. We believe that the resultant reduced stride length and decreased cadence expose the contralateral extremity to less repetitive trauma than might occur if a patient walked with crutches. These factors, combined with frequent monitoring and appropriate prescription of footwear (eg, depth-inlay shoes versus custom-molded shoes), may reduce the risk of precipitating a bilateral episode of acute Charcot's arthropathy. Until recently, there were no reports concisely detailing treatment of patients with acute Charcot's arthropathy through the postacute period. The mean time of immobilization (casting followed by removable cast walker) prior to return to permanent footwear was approximately 6 months in our study of 55 patients.l Patients receiving arthrodeses and patients who had bilateral Charcot's arthropathy were casted for longer periods of time (approximately 6 months versus 4 months) and took longer to return to permanent footwear (approximately 11 months versus 6 months).' Myerson and Physical Therapy. Volume 78. Number 1. January 1998 Armstrong and Lavery. 77

5 Charcot's Figure 3. Charcot's arthropathy treatment algorithm. Reprinted with permission from Armstrong et 01.1 coworkersmreported that, following open reduction and internal fixation of acute Charcot's fractures or dislocations in eight patients, the mean time of casting was 5 months. This report, however, did not discuss when patients returned to permanent prescription footwear and fully ambulaton functional status. Figure 4. The ambulatory total contact cast. Following casting until quiescence, the patients move into the postacute phase of treatment. Patients may, as required, then progress from casting to removable cast walkers to accommodative footwear with ankle-foot orthoses. Removable cast walkers or braces may be used to ease the transition from total contact casting to full, unprotected weight bearing in prescription footwear. Certain modcls of removable cast walkers (eg, EasyStep Walker*) have been shown to be as effective as total contact casts for reducing vertical plantar pressures."4 Several models of removable cast walkers, however, have poor off-loading characteristics. Therefore, care must be taken in using a removable cast walker that will not off-load effectively. The transition to a removable cast walker should be made when skin temperature gradients are within 1"C for 2 consecutive weeks at the affected site compared with the corresponding site on the contralatera1 extremity.' The transition from a removable cast walker to prescription footwear is based on 1 month of * Kendall Health Care, 1.5 Hampshire St, Mansfield, MA Armstrong and lavery skin temperature equilibrium (21 C) at the affected site compared with the corresponding site on the cantralatera1 ext~wnity.this period of protected weight bearing provides the pedorthic shoe specialist time in which to fabricate and appropriately fir prescription footwear. Reconstructive surgery should be performed if a deformity places the foot at risk for ulceration and if the deformity cannot be safely accommodated in prescription footwear. If the arthropathy is identified in its early stages, surgery is often unnecessary. Only 25% of the patients in the recent study by Armstrong et all ultimately required any form of surgical intervention, with about two thirds of those patients requiring an exostosectomy to remove a bony prominence and about one third of the patients needing an arthrodesis. The goal of any surgery for patients with Charcot's arthropathy is to create a stable, plantigrade foot that may be appropriately shod and that can support an ambulatory adult. Surgery is generally undertaken only after radiographic, Physical Therapy. Volume 7 8. Number 1. January 1998

6 to be of some assistance as an adjunctive modality.77we refer the reader to the references in this report to thoroughly investigate this complex neuropathic sequela of diabetes mellitus. Summary With the possible exception of osteomyelitis, Charcot's arthropathy is perhaps the most debilitating orthopedic sequela of diabetes mellitus. For this reason, early diagnosis and aggressive, noncompromising immobilization, pressure reduction, and consistent follow-through are paramount to effecting an acceptable result. References 1 i\rmstrong DG, Todd WF, Lavery LA, Harkless LB. The natural history of acute Charcot's arthropathy in a diabetic foot specialty clinic. D~abetMed. 1997;14: Sinha S, Munichoodapa CS, Kozak GP. Neurnarthropathy (Charcot joints) in diabetes mellitus. Medicine. 1972;51: Kelly M. William Musgrave's De Al-thrilide Symptomatica (1703): his description of neuropathic arthritis. Bull Hist hfed. 1963;37: Charcot JM. Sur quelaques arthropathies qui paraissent depender d'une lesion du cerveau ou de la moele epiniere. Arch Dm Phy~iolNorm et Path. 1868;1: Jordan WR. Ncuritic manifes~aiio~ls in diabetes mellitus. drch Intern Med. 1936;57: Kelly M. John Kearsley Mitchell and the neurogenic theory of arthritis. J Hist Med Allied Sci. 1965;20: Eloesser L. O n the nature of neuropathic affections of the joint. Ann SUW 1917;66: Figure 5. Acute Charcot's arthropathy of the midfoot, with concomitant neuropathic ulceration. 8 Sanders LJ, Frykberg RG. Charcot'sjoint. In: Levin ME, O'Neal LW, Bowker JH, eds. The Diabetic Fool. 2nd ed. St Louis, Mo: Mosby-Year Book; 1993: Finsterbush A, Friedman B. The effect of sensory denemation on rahhits' knee joints. JBone Joint Surg Am. 1975:57: dermal thermometric, and clinical signs of Charcot's joint quiescence. After surgery, the patient's lower limb is inlmobilized until skin temperatures and postoperative edema normalize. Following immobilization, the patient is progressed to a removable cast walker, followed by prescription of permanent footwear. Limitations of This Review The focus of this update is to provide current information regarding the etiology, diagnosis, and treatment of neuropathic osteoarthropathy. Most of the literature on this topic has focused on surgical treatment for neuropathic osteoarthropathy, often neglecting the nonsurgical aspects of the treatment protocol. We believe (and we have reported1) that the majority of patients with Charcot's arthropathy do not require surgical intervention. For this reason, we have concentrated on outlining this important aspect of care. There have been brief disc~lssionsof pharmacologic augmentation to current treatments, using bisphosphonates to retard bone resorption,-l5." but this is in need of further investigation. Additionally, electrical bone stimulation may prove Physical Therapy. Volume 78. Number 1. January Brower AC, Allman RM. The neuropathic joint: a neurovascular bone disorder. Radio1 Cltn North Am. 1981;19: Edmonds ME, Clarke MB, Newton JB, ct al. Increased uprake of radiopharmaceutical in diabetic neuropathy. QIM. 1985;57: Todd WF, Laughner T, Samojla BG. The diabetic foot. In: Robbins JM, ed. Pnmary Podiatnc Medicine. Philadelphia, Pa: WB Saunders Co; 1994: Banks AS, McGlamry ED. Charcot foot. J Arn Podiatr &fed A J J O ~. 1989;79: Lavery LA, Armstrong DG, Walker SC. Healing rates of diabetic foot ulcers associated with midfoot fracture due to Charcot's arthropathy. Diabet Med. 1997;14: Armstrong DG, Lavery LA, Saraya M, Ashry H. Leukocytosis is a poor indicator of acute osteomyelitis of the foot in diabetes mellitus. JFoot Ankle Surg. 1996;34: Armstrong DG, Perales TA, Murff R, et al. Value of white blood cell count with diffeer-entialin the acute diabetic foot infection. J Am Podiatr Med Assoc. 1996;86: Lave~yLA, Armstrong DG, Quebedeaux TL, 'Walker SC. Puncture wounds: the frequency of normal laboratory values in the face of severe foot infections of the foot in diabetic and nondiabetic adults. Am J Med. 1996;101: Armstrong and Lavery. 79

7 18 Keenen AM, Tindel NL, Alavi A. Diagnosis of pedal osteomyelitis in diabetic patients using current scintigraphic techniques. Arch Intern Med. 1989; 149: Schauwecker DS. The scintigraphic diagnosis of osteomyelitis. American Jou.rrra1 of Roentgenology. 1992;158:9. 20 Grayson ML, Balaugh K, Levin E, Karchmer AW. Probing to bone in infected pedal ulcers: a clinical sign of underlying osteomyelitis in diabetic patients.,jam. 1995;273: Caputo GM. Intection: investigation and management. In: Boulton AJM, Connor H, Cavanagh PR, eds. The Foot in Diabetes. 2nd ed. Chichester, England: John Wiley & Sons Ltd; 1994: Honvitz T. Bone and cartilage debris in the synovial membrane: its significance in the early diagnosis of neuroarthropathy. JBoneJoint Surg Am. 1948;30: Eichenholz SN. Charrot Joints. Springfield, Ill: Charles C Thomas, Publisher; Sanders 14, Mrdjencovich D. Anatomical patterns of bone andjoint destruction in neuropathic diabetics. Diabetes. 1991;40(suppl 1):529A. 25 Harris JR, Brand PW. Patterns of disintegration of the tarsus in the anesthetic foot. JBone Joint Surg Br. 1966;48: Armstrong DG, Lavery LA, Liswood PL, et al. Infrared dermal thermomet~y for the high-risk diabetic foot. Phys Thm. 1997;77: Armstrong DG, Lavery L4. Monitoring healing of acute Charcot's arthropathy with infrared dermal thennomet~y. J Rehabil Res Dm. 1997;34: Armstrong DG, Lavery LA, Harkless LB. Treatment-based classification system for assessr~~ent and care of diabetic feet. JAm Podiatr Med Assoc. 1996;86: Arrnsc1-ong DG, Liswood PL, Todd WF. The contralateral limb during total contact casting: a dynamic pressure and thermometric analysis. JAm Podiatr Med iissoc. 1995;85: Kominsky SJ. The ambulatoly total rontact cast. In: Frykberg RG, ed. The High-Risk Foot in Diabetes Mellitus. New York, NY Churchill Livingstone Inc; 1991: Armstrong DG, Lavely IA. Monitoring neuropathic ulcer healing with infrared dermal thermometly. JFoot AnkleSurg. 1996;35: Clohisy DR, Thompson RC. Fractures associated with neuropathic arthropathy in adults who have juvenile-onset diabetes. JBoneJoint Surg Am. 1988;70: Myerson MS, Henderson MR, Saxby T, Wilson-Short K Management of midfoot diabetic neuroarthropathy. Foot Ankle. 1994;15: Lavery LA, Vela SA, Lavery DC, Quebedaux TL. Reducing dynamic foot pressures in high-risk diabetics with foot ulcerations: a comparison of treatments. Diabetes Care. 1996;19: Selby PL, Young MJ, Boulton AJ. Bisphosphonates: a new treatment for diabetic Charcot neuroarthropathy? Diabet Med. 1994;11: van der Pluijm G, Binderup L, Bramm E, et al. Disodium-I-hydro? 3(1-pyrrolidiny1)-propylidene-1,l-bisphophonate (EB1053) is a potent inhibitor of bone resorption in vitro and in vivo. J Bone Miner Res. 1992;7: Cohen M, Roman A, Lovins JE. Totally implanted direct current stimulator as treatment for a nonunion in the foot. JFoot Ankle Surg. 1993;32: Armstrong and Lavery Physical Therapy. Volume 78. Number 1. January 1998