Reviews/Commentaries/ADA Statements CONSENSUS REPORT

The Charcot Foot in

1 8 LEE C. ROGERS, DPM ALEXANDRA JIRKOVSKA, MD associated with this condition is midfoot 2 4 ROBERT G. FRYKBERG, DPM, MPH EDWARD JUDE, MD “ ” 3 9 collapse, described as a rocker-bottom DAVID G. ARMSTRONG, DPM, PHD STEPHAN MORBACH, MD 4 10 foot (Fig. 1), although the condition ap- ANDREW J.M. BOULTON, MD WILLIAM B. MORRISON, MD 5 11 pears in other and with other pre- MICHAEL EDMONDS, MD MICHAEL PINZUR, MD 6 12 sentations. Pain or discomfort may be a GEORGES HA VAN, MD DARIO PITOCCO, MD 6 13 AGNES HARTEMANN, MD LEE SANDERS, DPM feature of this disorder at the active 7 14 FRANCES GAME, MD DANE K. WUKICH, MD (acute) stage, but the level of pain may 7 15 fi WILLIAM JEFFCOATE, MD LUIGI UCCIOLI, MD be signi cantly diminished when com- pared with individuals with normal sen- sation and equivalent degrees of injury. The diabetic Charcot foot syndrome is a serious and potentially limb-threatening lower-extremity The set of signs and symptoms that complication of diabetes. First described in 1883, this enigmatic condition continues to chal- occur together with CN qualifies this lenge even the most experienced practitioners. Now considered an inflammatory syndrome, the condition as a syndrome, the “Charcot diabetic Charcot foot is characterized by varying degrees of bone and disorganization foot syndrome.” secondary to underlying neuropathy, trauma, and perturbations of bone metabolism. An in- ternational task force of experts was convened by the American Diabetes Association and the Definition and classification American Podiatric Medical Association in January 2011 to summarize available evidence on the pathophysiology, natural history, presentations, and treatment recommendations for this entity. recommendations c Nomenclature should be standardized Diabetes Care 34:2123–2129, 2011 to CN or the Charcot foot. c Existing classifications do not provide prognostic value or direct treatment. he Charcot foot in diabetes poses DEFINITION—Charcot neuropathic Active or inactive should be used to many clinical challenges in its di- osteoarthropathy (CN), commonly re- T describe an inflamed or stable CN, re- agnosis and management. Despite ferred to as the Charcot foot, is a condi- spectively. Acute and chronic can also the time that has passed since the first tion affecting the bones, joints, and soft be used in this regard, but there is publication on pedal osteoarthropathy in tissues of the foot and ankle, character- no accepted measure that defines the 1883, we have much to learn about the ized by inflammation in the earliest phase. transition point. pathophysiology, and little evidence ex- The Charcot foot has been documented to ists on treatments of this disorder. The occur as a consequence of various periph- PATHOGENESIS—There is no sin- international task force was convened in eral neuropathies; however, diabetic neu- gular cause for the development of the January 2011 at the Salpêtrière Hospital ropathy has become the most common Charcot foot, but there are factors that in Paris, France, to review the literature etiology. The interaction of several com- predispose to its development, as well as and report on the definition, pathogene- ponent factors (diabetes, sensory-motor a number of likely precipitating events. sis, diagnosis, and treatment of the dia- neuropathy, autonomic neuropathy, The current belief is that once the disease betic Charcot foot. Recommendations in trauma, and metabolic abnormalities of is triggered in a susceptible individual, this report are solely the opinions of the bone) results in an acute localized in- it is mediated through a process of un- authors and do not represent the official flammatory condition that may lead to controlled inflammation in the foot. This positions of the American Diabetes Asso- varying degrees and patterns of bone inflammation leads to and is ciation or the American Podiatric Medical destruction, subluxation, dislocation, indirectly responsible for the progressive Association. and deformity. The hallmark deformity fracture and dislocation that characterizes ccccccccccccccccccccccccccccccccccccccccccccccccc its presentation (1). The evidence to sup-

1 2 port this hypothesis is largely circumstan- From the Prevention Center at Valley Presbyterian Hospital, Los Angeles, California; the Carl T. fl Hayden Veterans Affairs Medical Center, Phoenix, Arizona; the 3Southern Arizona Limb Salvage Alli- tial. A neurally mediated vascular re ex ance (SALSA), University of Arizona College of Medicine, Tucson, Arizona; the 4Department of Medicine, leading to increased peripheral blood University of Manchester, Manchester, U.K.; the 5Diabetic Foot Clinic, King’s College Hospital, London, flow and active has 6 U.K.; the Diabetes and Metabolic Diseases Department, Pitié-Salpêtrière University Hospital, Paris, France; been proposed as an etiological factor in the 7Department of Diabetes and Endocrinology, Nottingham University Hospitals NHS Trust, Nottingham, U.K.; the 8Institute for Clinical and Experimental Medicine, Diabetes Centre, Prague, Czech Republic; the the development of bone and joint de- 9Department of Diabetes and Angiology, Marienkrankenhaus, Soest, Germany; the 10Department of Radiology, struction in neuropathic patients. How- Thomas Jefferson University Hospital, Philadelphia, Pennsylvania; the 11Department of Orthopedic Surgery, ever, the relationship between increased Loyola University Health System, Maywood, Illinois; the 12Diabetes Center and Foot Care Unit A, Gemelli fl 13 blood ow to bone and active bone resorp- Hospital, Catholic University, Rome, Italy; the Podiatry Service, Veterans Affairs Medical Center, Lebanon, tion has not been conclusively defined. Pennsylvania; the 14Department of Orthopedic Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania; and the 15Department of Internal Medicine, University of Rome Tor Vergata, Rome, Italy. Corresponding author: Lee C. Rogers, [email protected]. Uncontrolled inflammation DOI: 10.2337/dc11-0844 When a bone is fractured, the release of This article is copublished in Diabetes Care and the Journal of the American Podiatric Medical Association, under proinflammatory cytokines including tumor joint copyright. a b © 2011 by the American Diabetes Association and the American Podiatric Medical Association. Readers may necrosis factor- and interleukin-1 leads use this article as long as the work is properly cited, the use is educational and not for profit, and the work is to increased expression of the polypep- not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details. tide receptor activator of nuclear factor-kB care.diabetesjournals.org DIABETES CARE, VOLUME 34, SEPTEMBER 2011 2123 Consensus Report

Predisposition The researchers also reported an associa- Neuropathy is a universal feature of the tion between BMD and the relative preva- affected limb. Although it has been sug- lence of fracture and of dislocation in the gested that people with a Charcot foot affected foot (11). may have particular patterns of sensory Diabetes may be associated with os- loss reflecting involvement of different teopenia, but the available evidence sug- fibers (5,6), this is not generally accepted. gests that reduction of BMD is a feature of Nevertheless, three groups have shown more so than type 2 dia- that people who have had an acute Char- betes (12). Any reduction of BMD in type cot foot exhibit retention of vasodilatory 1 diabetes may relate to loss of islet pep- reflexes in contrast to diabetic individuals tides such as and amylin (IAPP), with distal symmetrical neuropathy with- both of which act as growth factors for out CN (7–9). bone. Despite this, the fracture risk in Despite these observations, it should may be no less than in be noted that the syndrome might also type 1 diabetes (12), and this would ex- occur in patients with a spectrum of plain the fact that the presentation does unrelated diseases complicated by nerve not appear to differ between the two types damage. These include distal neuropathies of the disease. Any associated deficiency caused by toxins (ethanol, drug related) of vitamin D—with or without renal failure and infection (), as well as diseases and secondary hyperparathyroidism— of the spinal cord and nerve roots (tabes would increase the possibility of reduced Figure 1—The typical appearance of a later- dorsalis, trauma, ) and a BMD in diabetes. The use of thiazolidine- stage Charcot foot with a rocker-bottom de- number of other conditions (Parkinson’s diones could theoretically increase the like- formity. disease, HIV, sarcoidosis, rheumatoid dis- lihood of an acute Charcot foot through an ease, and psoriasis). Although the neuro- effect on bone density, but this has not yet is typically more proximal in been reported. The use of corticosteroids ligand (RANKL) from any of a number of those with disease of the spinal cord, the as immunosuppressants in people with di- local cell types. RANKL triggers the syn- presentation may be otherwise indistin- abetes who have had a renal and/or pan- thesis of the nuclear transcription factor guishable. creatic transplant (13) may explain the nuclear factor-kb (NF-kb), and this in Loss of protective sensation will in- apparent high incidence of the Charcot turn stimulates the maturation of osteo- crease the likelihood of trauma to the foot, foot in this group. clasts from osteoclast precursor cells. At while motor neuropathy could result in Although the Charcot syndrome may the same time, NF-kb stimulates the pro- altered structure of the foot (with exag- occur in a variety of conditions, diabetes duction of the glycopeptide osteoprotegerin geration of the plantar arch and clawing) is ostensibly the most common world- (OPG) from osteoblasts. This “decoy re- and changed gait with resultant abnormal wide. Diabetes may predispose to its ceptor” acts as an effective antagonist of loading. occurrence through a number of mech- RANKL (2). The fracture will also be asso- Finally, it is possible that peptides anisms. Apart from the presence of neu- ciated with pain, and this leads to splinting normally secreted from nerve terminals ropathy and possible osteopenia, these of the bone, and the rise in proinflamma- are also important in the underlying include the effects of advanced glycation tory cytokines is usually relatively short- pathophysiology. Of these, calcitonin end products, reactive oxygen species, lived. In the person who develops an acute gene–related peptide (CGRP) is a likely and oxidized lipids, which may all enhance Charcot foot, however, the loss of pain sen- candidate because it is known to antago- the expression of RANKL in diabetes sation allows for uninterrupted ambulation, nize the synthesis of RANKL. Hence, any (10). The effect of local inflammation with repetitive trauma. It has been sugges- reduction of CGRP through nerve damage on this pathway would similarly com- ted that this results in continual production will result in an increase in RANKL ex- pound the expression of RANKL. Fur- of proinflammatory cytokines, RANKL, pression. It is of particular interest that thermore, a single study has reported NF-kb, and osteoclasts, which in turn CGRP has been reported to be necessary an apparent association between two leads to continuing local osteolysis (1). for the maintenance of the normal integ- OPG-related polymorphisms in people This has subsequently been shown by an rity of joint capsules, and it follows that with a history of an acute Charcot foot increase in proinflammatory phenotypes of any reduction in CGRP release by nerve in diabetes (14). monocytes in those with active Charcot terminals could facilitate joint disloca- Many patients recall that the onset of foot when compared with diabetic control tion (10). the condition was precipitated by trauma subjects (3). Because it is not possible to identify that is often minor in nature (15). Other Osteoclasts generated in vitro in the those most likely to develop the Charcot cases may be triggered by different causes presence of macrophage colony-stimulating syndrome, it is impossible to determine of local inflammation, including previous factor and RANKL from patients with with any degree of confidence whether ulceration, infection, or recent foot sur- active CN have been shown to be more preexisting osteopenia is a significant gery. In this respect the occurrence of an aggressive and exhibit an increase in their predisposing factor. One group has, how- acute Charcot foot as a complication of resorptive activity compared with control ever, reported an apparent reduction in is increasingly recognized subjects. However, these changes are only bone mineral density (BMD) of the fem- in people with diabetes. Very occasion- partially inhibited by OPG, indicating that oral neck in the contralateral (unaf- ally, the onset of an acute Charcot foot other cytokines may also be important (4). fected) limb at the time of presentation. may follow successful revascularization.

2124 DIABETES CARE, VOLUME 34, SEPTEMBER 2011 care.diabetesjournals.org Rogers and Associates

DIAGNOSIS—The initial manifesta- However, radiographic changes of CN tions of the Charcot foot are frequently are typically delayed and have low sensi- mild in nature, but can become much tivity (28). more pronounced with unperceived re- Magnetic resonance imaging (MRI) petitive trauma. Diagnostic clinical find- allows detection of subtle changes in the ings include components of neurological, early stages of active CN when X-rays vascular, musculoskeletal, and radio- could still be normal. MRI primarily graphic abnormalities. There have been images protons in fat and water and can no reported cases of CN developing in the depict anatomy and pathology in both absence of neuropathy. Accordingly, pe- soft tissue and bone in great detail. Be- ripheral sensory neuropathy associated cause of its unique capability of differen- with reduced sensation of pain is the tiating tissues with high detail, MRI has a Figure 2—Lateral X-ray of a Charcot foot essential predisposing condition that per- high sensitivity and specificity for osteo- deformity showing a dislocation of the tarso- fi mits the development of the arthropathy myelitis and has become the test of choice metatarsal joint with break in the talo- rst (16–19). Because of the very presence of for evaluation of the complicated foot in metatarsal line (dashed lines) and a reduced calcaneal inclination angle (solid lines). insensitivity, a personal history concern- diabetic patients (29). Although not re- ing antecedent trauma is often unreliable quired for diagnosis when X-rays are di- (18,20,21). Typical clinical findings agnostic for Charcot bone and joint noncontrast MRI is very effective at ex- include a markedly swollen, warm, and changes, MRI is very useful in making the cluding osseous disease. If the patient often erythematous foot with only mild diagnosis at its earliest onset before such has an ulceration with a high likelihood to modest pain or discomfort (16,18– changes become evident on plain films. of deep infection, MRI is the best diagnos- 20,22). Acute local inflammation is often Nuclear medicine includes a number tic modality. Nonetheless, one test may the earliest sign of underlying bone and of exams based on the use of radioisotopic not be adequate for full evaluation. In joint injury (23). This initial clinical picture tracers. Three-phase bone scans, based on 99m this setting where MRI diagnosis is in- resembles cellulitis, deep vein thrombosis, technetium-99m ( Tc), are highly sen- determinate, a subsequent labeled white or acute and can be misdiagnosed as sitive for active bone pathology. How- blood cell scan can provide more speci- such. There is most often a temperature ever, diminished circulation can result ficity and should be correlated with clin- differential between the two feet of several in false-negative exams and, perhaps ical findings. The decision of nuclear degrees (20,24). The affected population more importantly, uptake is not specific imaging versus MRI is largely based on typically has well preserved or even exag- for osteoarthropathy. Labeled white blood 111 99m personal preference, availability, and local gerated arterial blood flow in the foot. cell scanning (using In or Tc) pro- experience. In general, if metal is present Pedal pulses are characteristically bound- vides improved specificity for infection in in the foot, nuclear medicine exams are ing unless obscured by concurrent the setting of neuropathic bone changes preferred, whereas diffuse or regional is- . Patients with chronic deformities, (30), but it can be difficult to differentiate chemia makes MRI the preferred exam. however, can develop subsequent limb- soft tissue from bone. Therefore, this exam threatening ischemia. Musculoskeletal can be combined with a three-phase bone deformity can be very slight or grossly scan or sulfur colloid marrow exam when Diagnostic recommendations for evident most often due to the chronicity superimposed osteomyelitis is suspected active CN of the problem and the anatomical site of (31). More recently, positron emission to- c The diagnosis of active Charcot foot is involvement (16,17,19,25). The classic mography scanning has been recognized primarily based on history and clinical rocker-bottom foot, with or without as having potential for diagnosis of infec- findings but should be confirmed by plantar ulceration, represents a severe tion and differentiating the Charcot foot imaging. chronic deformity typical for this condi- from osteomyelitis (32,33). However, c Inflammation plays a key role in the tion (16,26,27). Radiographic and other this remains investigational at this time. pathophysiology of the Charcot foot imaging modalities can detect subtle Evaluation of bone density may be and is the earliest clinical finding. changes consistent with active CN. useful in those with diabetes to assess c The occurrence of acute foot/ankle frac- onset of CN as well as fracture risk. BMD tures or dislocations in neuropathic Imaging of the Charcot foot can be assessed using dual-energy X-ray individuals is considered active CN Radiographs are the primary initial imag- absorptiometry or calcaneal ultrasound. because of the inflammatory process of ing method for evaluation of the foot in BMD has been related to the pathological bone healing, even in the absence of diabetic patients. Easily available and in- pattern of CN, whereby deformity. expensive, they provide information on is more prevalent in those with normal c X-rays should be the initial imaging bone structure, alignment, and minerali- mineralization versus fracture in those performed, and one should look for zation. X-rays may be normal or show with diminished BMD (11). subtle fractures or subluxations if no subtle fractures and dislocations or later Experts agree that radiographs are obvious pathology is visible. show more overt fractures and subluxa- important as the first exam in virtually c MRI or nuclear imaging can confirm tions. In later stages, the calcaneal inclina- all settings (33,34). However, a negative clinical suspicions in the presence of tion angle is reduced and the talo-first result obviously should not offer any con- normal-appearing radiographs. metatarsal angle is broken (Fig. 2). Me- fidence regarding lack of disease. In a dial calcification of the arteries is present patient with low clinical suspicion of MEDICAL TREATMENT—The med- in most Charcot feet and is a frequent osteomyelitis and no sign of CN on radio- ical treatment of CN is aimed at offload- secondary finding on radiographs (25). graphs, either three-phase bone scan or ing the foot, treating bone disease, and care.diabetesjournals.org DIABETES CARE, VOLUME 34, SEPTEMBER 2011 2125 Consensus Report preventing further foot fractures (34). Be- Antiresorptive therapy weight-bearing devices such as pre- cause of the various etiologies of increased Treatment by antiresorptive drugs has scription shoes, boots, or braces. local bone resorption and/or secondary os- been proposed because bone turnover in c Lifetime surveillance is advised to mon- teoporosis in patients with CN and limited patients with active CN is excessive. itor for signs of recurrent or new CN randomized placebo-controlled trials in However, there is little evidence to sup- episodes as well as other this area, treatment guidelines are largely port their use, but both oral and intrave- complications. based on professional opinion rather than nous bisphosphonates (38) have been the highest level of clinical evidence. studied in the treatment of CN in small SURGICAL TREATMENT—Surgi- randomized, double-blind, controlled cal treatment of Charcot arthropathy of Offloading trials (39,40) or in retrospective con- the foot and ankle is based primarily on Offloading at the acute active stage of the trolled studies (41). Some patients can- expert opinion. Surgery has generally Charcot foot is the most important man- not tolerate oral bisphosphonates but been advised for resecting infected bone agement strategy and could arrest the may benefit from intravenous therapy us- (osteomyelitis), removing bony promi- progression to deformity. Ideally, the ing pamidronate or zoledronic acid (42). nences that could not be accommodated foot should be immobilized in an irremov- Intranasal calcitonin is another antiresorp- with therapeutic footwear or custom able total contact cast (TCC), which is tive agent that has been studied in CN. orthoses, or correcting deformities that initially replaced at 3 days, then checked This treatment was associated with a sig- could not be successfully accommodated each week. Edema reduction is often nificantly greater reduction in cross- with therapeutic footwear, custom ankle- remarkable in the first few weeks of treat- linked carboxy-terminal telopeptide of foot orthoses, or a Charcot Restraint Or- ment. The cast should be changed fre- type I collagen and bone-specificalka- thotic Walker (50). This clinical approach quently to avoid “pistoning” as the edema line phosphatase than standard treatment is based on expert opinion and small, un- subsides. If possible, the patient should in the control group that received only controlled retrospective case series. There use crutches or wheelchair and should calcium supplementation and offloading. has been an increasing trend in the litera- be encouraged to avoid weight bearing Calcitonin has a safer profile in renal fail- ture to advise earlier surgical correction of on the affected side. The casting is con- ure when compared with bisphosphonate deformity and arthrodesis, based on the as- tinued until the swelling has resolved and therapy (43–46). However, a single dose sumption that surgical stabilization would the temperature of the affected foot is of intravenous bisphosphonate generally lead to an improved patient-perceived within 2°C of the contralateral foot (35). does not require renal adjustment. There quality of life (51). An alternative device for offloading the is no conclusive evidence for using bi- Several investigators have suggested acute active stage of CN is a prefabricated sphosphonates in active Charcot foot, that Achilles tendon lengthening com- removable walking cast or “instant TCC” and our understanding is evolving as bined with total contact casting has the technique, which transforms a removable more trials are currently underway. potential to decrease the deforming forces cast walker to one that is less easily re- at the midfoot and decrease the morbidity moved (36,37). It is important to take Bone growth stimulation associated with CN (52–57). Exostec- into consideration that TCC may actually There is limited evidence for the use of tomy offers the potential to reduce pres- have unfavorable consequences on the external bone stimulation in CN. Ultra- sure caused by bony prominences. This non-Charcot limb and induce unnatural sonic bone stimulation was reported for treatment is often combined with accom- stress patterns causing ulcerations and the treatment of CN of the ankle and for modative bracing and appears to obtain even fractures. Furthermore, patients the healing of fresh fractures. Direct cur- more favorable results in patients without with CN have increased instability and rent electrical bone growth stimulators associated ulcers (58–60). risk for falling and fracture as a result have been used specifically in CN patients Surgery has generally been avoided of multiple comorbidities including loss undergoing arthrodesis and clinically during the active inflammatory stage be- of proprioception and postural hypoten- tested to promote healing of fractures cause of the perceived risk of wound in- sion. Nonetheless, it should be noted that in the acute phase of CN in small case fection or mechanical failure of fixation. total immobility has disadvantages in itself series. Although these findings are prom- Two recent case series would suggest po- with a loss of muscle tone, reduction in ising, there have been no subsequent tentially favorable outcomes with early bone density, and loss of body fitness. studies to validate this method, and its correction of deformity combined with Duration and aggressiveness of off- use has been supported only as an ad- arthrodesis (61,62). Most case series have loading (nonweight bearing vs. weight junct therapy during the postsurgical focused on reconstruction of the deformity bearing, nonremovable vs. removable period (45–49). by reduction and arthrodesis using stan- device) are guided by clinical assessment dard methods of internal fixation. Because of healing of CN based on edema, ery- Recommendations for medical of the poor bone quality, expert opinion thema, and skin temperature changes therapy has advised an extended period of non- (34,35). Evidence of healing on X-rays c Offloading the foot and immobilization weight bearing after surgery to account or MRI strengthens the clinical decision are the most important treatment rec- for the poor bone healing and inherent to transition the patient into footwear. ommendations in active CN and can weakness of the underlying osseous struc- To prevent recurrence or ulceration on prevent further destruction. tures. Early surgical series showed im- subsequent deformities, various devices c There is little evidence to guide the use provement in restoring a plantigrade foot are recommended after an acute or active of available pharmacological therapies and preventing recurrence of ulceration, al- episode has resolved, including prescrip- to promote the healing of CN. though nonunion, failure, and loss of initial tive shoes, boots, or other weight-bearing c Protective weight bearing is required correction were common (63–67). The con- braces. Frequent monitoring is required. after an active episode, involving cept of an internal fixation “superconstruct”

2126 DIABETES CARE, VOLUME 34, SEPTEMBER 2011 care.diabetesjournals.org Rogers and Associates that extends internal fixation beyond the zone of fusion has evolved to address these issues (68). The combination of poor bone quality and a tenuous soft tissue envelope in a relatively immune-impaired popula- tion has led many surgeons to use a mod- ification of the external fixation method of Ilizarov to correct deformity with a limited risk for surgical-associated mor- bidity (69–74).

Charcot arthropathy of the ankle Given the common failures of nonsurgical management of CN of the ankle, the task force members agree that surgical man- agement could be considered a primary treatment. Surgical correction of defor- mity at the level of the ankle is likely more necessary due to the poor tolerance of deformity in the coronal plane (i.e., ankle varus or valgus) and the resultant prom- inence of the malleoli and their vulnera- bility to pressure-induced ulceration. Several small uncontrolled series have recommended augmented internal fixa- tion followed by prolonged periods of immobilization and nonweight bearing in neuropathic patients who sustain acute ankle fractures (75–77). Acute ankle frac- tures in patients with complicated diabe- tes are associated with significantly higher Figure 3—An algorithm depicting the basic approach to the Charcot foot. *Osteomyelitis can be rates of noninfectious complications and difficult to distinguish from the Charcot foot. The reader is referred to the “Imaging of the Charcot need for surgical revision when compared foot” section of the article for techniques to improve specificity of various imaging modalities. with diabetic patients without other organ system comorbidities (78). Numerous the ankle and hindfoot to the midfoot techniques have been reported without — – and forefoot. Acknowledgments The American Diabetes comparative effectiveness (79 82). All of Association/American Podiatric Medical As- c Arthrodesis can be useful in patients the surgical studies are retrospective in sociation Task Force meeting was supported nature without a control group and are with instability, pain, or recurrent ul- by unrestricted educational grants from Small basedonalimitednumberofpatients. cerations that fail nonoperative treat- Bone Innovations, sanofi-aventis, and Integra While a strong, stable construct is required, ment, despite a high rate of incomplete LifeSciences. No other potential conflicts of inconclusive data exist to recommend one bony union. interest relevant to this article were reported. form of fixation over another (i.e., inter- c For severe CN of the ankle, surgical The authors thank Cécile Swiatek, chief nal, external, or combined) in the surgical management could be considered a librarian, from the J.-M. Charcot library at the reconstruction of the foot and ankle in pa- primary treatment. Hospital Pitié-Salpêtrière for her assistance tients who are not infected. with the site preparation for the task force and references in this article. CONCLUSIONS—The Charcot foot Recommendations for surgical syndrome is a complex complication of treatment diabetes and neuropathy. Its destructive References c Surgical treatment is beneficial in CN effects on the foot and ankle begin with a 1. Jeffcoate WJ, Game F, Cavanagh PR. The cases refractory to offloading and im- cycle of uncontrolled inflammation. The role of proinflammatory cytokines in the mobilization or in the case of recalcitrant classic rocker-bottom foot deformity is a cause of neuropathic osteoarthropathy ulcers. late stage of the syndrome and can be (acute Charcot foot) in diabetes. Lancet c The initial management of acute neuro- avoided by early recognition and man- 2005;366:2058–2061 pathic fractures and dislocations should agement. Offloading is the most impor- 2. Boyce BF, Xing L. Functions of RANKL/ not differ from other fractures. tant initial treatment recommendation. RANK/OPG in bone modeling and re- modeling. Arch Biochem Biophys 2008; c Exostectomy is useful to relieve bony Surgery can be helpful in early stages 473:139–146 pressure that cannot be accommodated involving acute fractures of the foot or fl 3. Uccioli L, Sinistro A, Almerighi C, et al. with orthotic and prosthetic means. ankle or in later stages when of oading is Proinflammatory modulation of the sur- c Lengthening of the Achilles tendon or ineffective. An algorithm summarizing face and cytokine phenotype of mono- gastrocnemius tendon reduces forefoot the approach to the Charcot foot can be cytes in patients with acute Charcot foot. pressure and improves the alignment of seen in Fig. 3. Diabetes Care 2010;33:350–355 care.diabetesjournals.org DIABETES CARE, VOLUME 34, SEPTEMBER 2011 2127 Consensus Report

4. Mabilleau G, Petrova NL, Edmonds ME, 18. Cofield RH, Morrison MJ, Beabout JW. 34. Frykberg RG, Eneroth M. Principles of Sabokbar A. Increased osteoclastic activity Diabetic neuroarthropathy in the foot: pa- conservative management. In The Diabetic in acute Charcot’s osteoarthropathy: the tient characteristics and patterns of radio- Charcot Foot: Principles and Management. role of receptor activator of nuclear factor- graphic change. Foot Ankle 1983;4:15–22 Frykberg RG, Ed. Brooklandville, MD, Data kappaB ligand. Diabetologia 2008;51:1035– 19. Johnson JT. Neuropathic fractures and Trace Publishing Company, 2010, p. 93–116 1040 joint injuries. Pathogenesis and rationale 35. Armstrong DG, Lavery LA. Monitoring 5. Stevens MJ, Edmonds ME, Foster AV, of prevention and treatment. J Bone Joint healing of acute Charcot’s arthropathy Watkins PJ. Selective neuropathy and Surg Am 1967;49:1–30 with infrared dermal thermometry. J Re- preserved vascular responses in the di- 20. Armstrong DG, Todd WF, Lavery LA, habil Res Dev 1997;34:317–321 abetic Charcot foot. Diabetologia 1992; Harkless LB, Bushman TR. The natural 36. Armstrong DG, Lavery LA, Wu S, Boulton 35:148–154 history of acute Charcot’s arthropathy in a AJ. Evaluation of removable and irremov- 6. Young MJ, Marshall A, Adams JE, Selby PL, diabetic foot specialty clinic. Diabet Med able cast walkers in the healing of diabetic Boulton AJ. Osteopenia, neurological dys- 1997;14:357–363 foot wounds: a randomized controlled function, and the development of Charcot 21. Henderson VE. Joint affections in tabes trial. Diabetes Care 2005;28:551–554 neuroarthropathy. Diabetes Care 1995;18: dorsalis. J Pathol 1905;10:211–264 37. Armstrong DG, Short B, Espensen EH, 34–38 22. Caputo GM, Ulbrecht J, Cavanagh PR, Abu-Rumman PL, Nixon BP, Boulton AJ. 7. Veves A, Akbari CM, Primavera J, et al. Juliano P. The Charcot foot in diabetes: six Technique for fabrication of an “instant Endothelial dysfunction and the expres- key points. Am Fam Physician 1998;57: total-contact cast” for treatment of neu- sion of endothelial nitric oxide synthetase 2705–2710 ropathic diabetic foot ulcers. J Am Podiatr in , vascular disease, 23. Jeffcoate W. The causes of the Charcot Med Assoc 2002;92:405–408 and foot ulceration. Diabetes 1998;47: syndrome. Clin Podiatr Med Surg 2008; 38. Selby PL, Young MJ, Boulton AJ. Bi- 457–463 25:29–42, vi sphosphonates: a new treatment for di- 8. Shapiro SA, Stansberry KB, Hill MA, et al. 24. McGill M, Molyneaux L, Bolton T, abetic Charcot neuroarthropathy? Diabet Normal blood flow response and vaso- Ioannou K, Uren R, Yue DK. Response of Med 1994;11:28–31 motion in the diabetic Charcot foot. J Di- Charcot’s arthropathy to contact casting: 39. Jude EB, Selby PL, Burgess J, et al. Bi- abetes Complications 1998;12:147–153 assessment by quantitative techniques. sphosphonates in the treatment of Charcot 9. Baker N, Green A, Krishnan S, Rayman G. Diabetologia 2000;43:481–484 neuroarthropathy: a double-blind rando- Microvascular and C-fiber function in di- 25. Sinha S, Munichoodappa CS, Kozak GP. mised controlled trial. Diabetologia 2001; abetic Charcot neuroarthropathy and di- Neuro-arthropathy (Charcot joints) in di- 44:2032–2037 abetic . Diabetes Care abetes mellitus (clinical study of 101 cases). 40. Pitocco D, Ruotolo V, Caputo S, et al. Six- 2007;30:3077–3079 Medicine (Baltimore) 1972;51:191–210 month treatment with alendronate in 10. Jeffcoate WJ, Game FL. New theories 26. Sanders LJ, Frykberg RG. The Charcot acute Charcot neuroarthropathy: a ran- on the causes of the Charcot foot in dia- foot (Pied de Charcot). In Levin and domized controlled trial. Diabetes Care betes. In The Diabetic Charcot Foot: Princi- O’Neal’s The Diabetic Foot. 7th ed. Bowker 2005;28:1214–1215 ples and Management. Frykberg RG, Ed. JH, Pfeifer MA, Eds. Philadelphia, Mosby 41. Anderson JJ, Woelffer KE, Holtzman JJ, Brooklandville, MD, Data Trace Publishing Elsevier, 2007, p. 257–283 Jacobs AM. Bisphosphonates for the Company, 2010, p. 29–44 27. Morrison WB, Shortt CP, Ting AYI. Im- treatment of Charcot neuroarthropathy. 11. Herbst SA, Jones KB, Saltzman CL. Pattern aging of the Charcot foot. In The Diabetic J Foot Ankle Surg 2004;43:285–289 of diabetic neuropathic arthropathy as- Charcot Foot: Principles and Management. 42. Hofbauer LC, Hamann C, Ebeling PR. sociated with the peripheral bone mineral Frykberg RG, Ed. Brooklandville, MD, Approach to the patient with secondary density. J Bone Joint Surg Br 2004;86: Data Trace Publishing Company, 2010, osteoporosis. Eur J Endocrinol 2010;162: 378–383 p. 65–84 1009–1020 12. Hofbauer LC, Brueck CC, Singh SK, 28. Morrison WB, Ledermann HP. Work-up 43. Bem R, Jirkovská A, Fejfarová V, Skibová J, Dobnig H. Osteoporosis in patients with of the diabetic foot. Radiol Clin North Am Jude EB. Intranasal calcitonin in the treat- diabetes mellitus. J Bone Miner Res 2007; 2002;40:1171–1192 ment of acute Charcot neuroosteoarthropa- 22:1317–1328 29. Morrison WB, Ledermann HP, Schweitzer thy: a randomized controlled trial. Diabetes 13. Matricali GA, Bammens B, Kuypers D, ME. MR imaging of the diabetic foot. Magn Care 2006;29:1392–1394 Flour M, Mathieu C. High rate of Charcot Reson Imaging Clin N Am 2001;9:603– 44. Molines L, Darmon P, Raccah D. Charcot’s foot attacks early after simultaneous 613, xi foot: newest findings on its pathophysi- pancreas-kidney transplantation. Trans- 30. Larcos G, Brown ML, Sutton RT. Di- ology, diagnosis and treatment. Diabetes plantation 2007;83:245–246 agnosis of osteomyelitis of the foot in di- Metab 2010;36:251–255 14. Pitocco D, Zelano G, Gioffrè G, et al. As- abetic patients: value of 111In-leukocyte 45. Ulbrecht JS, Wukich DK. The Charcot sociation between osteoprotegerin G1181C scintigraphy. AJR Am J Roentgenol 1991; foot: medical and surgical therapy. Curr and T245G polymorphisms and dia- 157:527–531 Diab Rep 2008;8:444–451 betic Charcot neuroarthropathy: a case- 31. Palestro CJ, Mehta HH, Patel M, et al. 46. Wukich DK, Sung W. Charcot arthropa- control study. Diabetes Care 2009;32: Marrow versus infection in the Charcot thy of the foot and ankle: modern con- 1694–1697 joint: indium-111 leukocyte and technetium- cepts and management review. J Diabetes 15. Foltz KD, Fallat LM, Schwartz S. Useful- 99m sulfur colloid scintigraphy. J Nucl Med Complications 2009;23:409–426 ness of a brief assessment battery for early 1998;39:346–350 47. Strauss E, Gonya G. Adjunct low intensity detection of Charcot foot deformity in 32. Keidar Z, Militianu D, Melamed E, Bar- ultrasound in Charcot neuroarthropathy. patients with diabetes. J Foot Ankle Surg Shalom R, Israel O. The diabetic foot: Clin Orthop Relat Res 1998;(349):132– 2004;43:87–92 initial experience with 18F-FDG PET/CT. 138 16. Eichenholtz SN. Charcot Joints. Spring- J Nucl Med 2005;46:444–449 48. Hockenbury RT, Gruttadauria M, McKinney field, IL, Charles C. Thomas, 1966 33. Hopfner S, Krolak C, Kessler S, Tiling R. I. Use of implantable bone growth stim- 17. Charcot J-M, Fere C. Affections osseuses Preoperative imaging of Charcot neuro- ulation in Charcot ankle arthrodesis. Foot et articulaires du pied chez les tabétiques arthropathy. Does the additional appli- Ankle Int 2007;28:971–976 (Pied tabétique). Archives de Neurologie cation of 18F-FDG-PET make sense? 49. Petrisor B, Lau JT. Electrical bone stimu- 1883;6:305–319 [in French] Nuklearmedizin 2005;45:15-20 lation: an overview and its use in high risk

2128 DIABETES CARE, VOLUME 34, SEPTEMBER 2011 care.diabetesjournals.org Rogers and Associates

and Charcot foot and ankle reconstruc- 59. Rosenblum BI, Giurini JM, Miller LB, 71. Fabrin J, Larsen K, Holstein PE. Arthrod- tions. Foot Ankle Clin 2005;10:609–620, Chrzan JS, Habershaw GM. Neuropathic esis with external fixation in the unstable vii–viii ulcerations plantar to the lateral column or misaligned Charcot ankle in patients 50. Pinzur MS. Surgical management: history in patients with Charcot foot deformity: with diabetes mellitus. Int J Low Extrem and general principles. In The Diabetic a flexible approach to limb salvage. J Foot Wounds 2007;6:102–107 Charcot Foot: Principles and Management. Ankle Surg 1997;36:360–363 72. Wukich DK, Belczyk RJ, Burns PR, Frykberg RG, Ed. Brooklandville, MD, 60. Laurinaviciene R, Kirketerp-Moeller K, Frykberg RG. Complications encountered Data Trace Publishing Company, 2010, Holstein PE. Exostectomy for chronic with circular ring fixation in persons with p. 165–188 midfoot plantar in Charcot deformity. diabetes mellitus. Foot Ankle Int 2008;29: 51. Dhawan V, Spratt KF, Pinzur MS, J Wound Care 2008;17:53–55, 57–58 994–1000 Baumhauer J, Rudicel S, Saltzman CL. 61. Mittlmeier T, Klaue K, Haar P, Beck M. 73. Bevilacqua NJ, Rogers LC. Surgical man- Reliability of AOFAS diabetic foot ques- Should one consider primary surgical re- agement of Charcot midfoot deformities. tionnaire in Charcot arthropathy: stabil- construction in Charcot arthropathy of Clin Podiatr Med Surg 2008;25:81–94, vii ity, internal consistency, and measurable the feet? Clin Orthop Relat Res 2010;468: 74. Rogers LC, Bevilacqua NJ, Frykberg RG, difference. Foot Ankle Int 2005;26:717– 1002–1011 Armstrong DG. Predictors of postoperative 731 62. Simon SR, Tejwani SG, Wilson DL, complications of Ilizarov external ring fix- 52. Lavery LA, Armstrong DG, Boulton AJ; Santner TJ, Denniston NL. Arthrodesis as ators in the foot and ankle. J Foot Ankle Diabetex Research Group. Ankle equinus an early alternative to nonoperative man- Surg 2007;46:372–375 deformity and its relationship to high agement of Charcot arthropathy of the 75. Connolly JF, Csencsitz TA. Limb threat- plantar pressure in a large population with diabetic foot. J Bone Joint Surg Am 2000; ening neuropathic complications from an- diabetes mellitus. J Am Podiatr Med Assoc 82-A:939–950 kle fractures in patients with diabetes. Clin 2002;92:479–482 63. Papa J, Myerson M, Girard P. Salvage, with Orthop Relat Res 1998;(348):212–219 53. Armstrong DG, Lavery LA. Elevated peak arthrodesis, in intractable diabetic neuro- 76. Jani MM, Ricci WM, Borrelli J Jr, Barrett plantar pressures in patients who have pathic arthropathy of the foot and ankle. SE, Johnson JE. A protocol for treatment Charcot arthropathy. J Bone Joint Surg J Bone Joint Surg Am 1993;75:1056– of unstable ankle fractures using trans- Am 1998;80:365–369 1066 articular fixation in patients with diabetes 54. Hastings MK, Mueller MJ, Sinacore DR, 64. Pakarinen TK, Laine HJ, Honkonen SE, mellitus and loss of protective sensibility. Salsich GB, Engsberg JR, Johnson JE. Peltonen J, Oksala H, Lahtela J. Charcot Foot Ankle Int 2003;24:838–844 Effects of a tendo-Achilles lengthening arthropathy of the diabetic foot. Current 77. Perry MD, Taranow WS, Manoli A 2nd, procedure on muscle function and gait concepts and review of 36 cases. Scand J Carr JB. Salvage of failed neuropathic an- characteristics in a patient with diabetes Surg 2002;91:195–201 kle fractures: use of large-fragment fibular mellitus. J Orthop Sports Phys Ther 2000; 65. Stone NC, Daniels TR. Midfoot and hind- plating and multiple syndesmotic screws. 30:85–90 foot arthrodeses in diabetic Charcot J Surg Orthop Adv 2005;14:85–91 55. Maluf KS, Mueller MJ, Strube MJ, arthropathy. Can J Surg 2000;43:449– 78. Wukich DK, Joseph A, Ryan M, Ramirez Engsberg JR, Johnson JE. Tendon Achilles 455 C, Irrgang JJ. Outcomes of ankle fractures lengthening for the treatment of neuro- 66. Pinzur M. Surgical versus accommodative in patients with uncomplicated versus pathic ulcers causes a temporary reduction treatment for Charcot arthropathy of the complicated diabetes. Foot Ankle Int 2011; in forefoot pressure associated with changes midfoot. Foot Ankle Int 2004;25:545– 32:120–130 in plantar flexor power rather than ankle 549 79. Ayoub MA. Ankle fractures in diabetic motion during gait. J Biomech 2004;37: 67. Garapati R, Weinfeld SB. Complex re- neuropathic arthropathy: can tibiotalar 897–906 construction of the diabetic foot and an- arthrodesis salvage the limb? J Bone Joint 56. Holstein P, Lohmann M, Bitsch M, kle. Am J Surg 2004;187(5A):81S–86S Surg Br 2008;90:906–914 Jørgensen B. Achilles tendon lengthening, 68. Assal M, Stern R. Realignment and ex- 80. Dalla Paola L, Volpe A, Varotto D, et al. the panacea for plantar forefoot ulcera- tended fusion with use of a medial column Use of a retrograde nail for ankle arthrodesis tion? Diabetes Metab Res Rev 2004;20 screw for midfoot deformities secondary in Charcot neuroarthropathy: a limb sal- (Suppl. 1):S37–S40 to diabetic neuropathy. J Bone Joint Surg vage procedure. Foot Ankle Int 2007;28: 57. Mueller MJ, Sinacore DR, Hastings MK, Am 2009;91:812–820 967–970 Lott DJ, Strube MJ, Johnson JE. Impact 69. Pinzur MS. Neutral ring fixation for high- 81. Caravaggi C, Cimmino M, Caruso S, Dalla of Achilles tendon lengthening on func- risk nonplantigrade Charcot midfoot de- Noce S. Intramedullary compressive nail tional limitations and perceived disability formity. Foot Ankle Int 2007;28:961–966 fixation for the treatment of severe Char- in people with a neuropathic plantar ul- 70. Farber DC, Juliano PJ, Cavanagh PR, cot deformity of the ankle and rear foot. cer. Diabetes Care 2004;27:1559–1564 Ulbrecht J, Caputo G. Single stage cor- J Foot Ankle Surg 2006;45:20–24 58. Catanzariti AR, Mendicino R, Haverstock B. rection with external fixation of the ul- 82. Pinzur MS, Noonan T. Ankle arthrodesis Ostectomy for diabetic neuroarthropathy cerated foot in individuals with Charcot with a retrograde femoral nail for Charcot involving the midfoot. J Foot Ankle Surg neuroarthropathy. Foot Ankle Int 2002; ankle arthropathy. Foot Ankle Int 2005; 2000;39:291–300 23:130–134 26:545–549

care.diabetesjournals.org DIABETES CARE, VOLUME 34, SEPTEMBER 2011 2129