ORIGINAL ARTICLES

The Natural History of Acute Charcot’sArthropathy in a Diabetic FootSpecialty Clinic†

D.G. Armstrong*1,2,3, W.F. Todd3, L.A. Lavery1,2, L.B. Harkless1, T.R. Bushman3

1Department of Orthopaedics, University of Texas Health ScienceCenter, San Antonio, Texas, USA2The Diabetic Foot Research Group, San Antonio, Texas, USAcThe Diabetic Foot Centers of America, Washington, DC, USA

The aim of this longitudinal study was to report on the clinical characteristics andtreatment course of acute Charcot’s arthropathy at a tertiary care diabetic foot clinic.Fifty-five diabetic subjects, with a mean age of 58.6 ± 8.5 years, were studied. All patientswere treated with serial total contact casting until quiescence. Following casting andbefore transfer to prescription footwear, patients were eased into unprotected weightbearingvia a removable cast walker. This cohort was followed for their entire treatment courseand for a mean 92.6 ± 33.7 weeks following return to shoes. Pain was the most frequentpresenting complaint in these otherwise insensate patients (76 %). The mean duration ofcasting was 18.5 ± 10.6 weeks. Patients returned to footwear in a mean 28.3 ± 14.5weeks. Nine per cent of the population had bilateral arthropathy. These subjects werecasted significantly longer than the unilateral group (p , 0.02). Surgery was performedon 25 % of patients, with approximately two-thirds of these procedures involving plantarexostectomies and one-third fusions of affected joints. Patients receiving surgery remainedcasted significantly longer than non-surgical patients (p , 0.05). Additionally, men werecasted longer than women (p , 0.008).

Acute Charcot’s arthropathy requires prompt, uncompromising reduction in weight-bearing stress. Our data show that the ambulatory total contact cast is very effective forthis. Regardless of the specific treatment method instituted, it is imperative that appropriateand aggressive treatment be undertaken immediately following diagnosis to help preventprogression to a profoundly debilitating, limb-threatening deformity. 1997 by JohnWiley & Sons, Ltd.

No of Figures: 3. No of Tables: 2. No of Refs: 47

KEY WORDS Charcot’s arthropathy

Received 3 October 1996; revised 12 November 1996; accepted 17 November 1996

of Charcot’s Arthropathy,11–20 we have been unable toIntroductionidentify studies specifically reporting on long-term surgi-cal and non-surgical outcomes of the acute form of thisNeuropathic osteoarthropathy, commonly referred to as

Charcot’s joint or Charcot’s arthropathy was probably disease. We are also unaware of any reports of theduration of treatment of acute Charcot’s joint fromfirst described by Musgrave in 1703.1 He described it

as complication secondary to venereal desease. Since inciting event to prescription footware. Furthermore, wehave been unable to identify studies reporting on thethen, numerous theories have been propagated regarding

its pathogenesis.2–5 The current theory suggests that, efficacy of total contact casting in the treatment ofCharcot’s arthropathy in a large cohort of patients. Thefollowing the development of autonomic neuropathy,

there is increased blood flow to the extremity, resulting purpose of this longitudinal study was to report on theclinical characteristics and treatment course of acutein osteopenia. In addition, motor neuropathies result in

muscle imbalance, placing abnormal stress on the Charcot’s arthropathy at a tertiary care diabetic foot clinic.affected extremity, and sensory neuropathy renders the

Patients and Methodspatient unaware of the often profound osseous destructiontaking place.6–8,9,10 We abstracted data retrospectively from patient visits to

While there are a number of studies in the medical a multidisciplinary tertiary care diabetic foot clinicliterature discussing characteristics and surgical treatment between 1 February 1991 and 1 July 1994. The initial

diagnosis of acute Charcot’s arthropathy was based onprofound swelling, locally increased skin temperature,* Correspondence to: David G. Armstrong, Assistant Professor, Depart-

ment of Orthopaedics, University of Texas Health Science Center, San erythema, joint effusion, and bone resorption and frag-Antonio 7703 Floyd Curl Drive, San Antonio, Texas, 78284-7776, USA. mentation in an insensate foot. All subjects studied had† This study was presented as an abstract at the 54th annual scientific

clinical loss of protective sensation to the 10 grammesymposium of the American College of Foot and Ankle Surgeons, NewOrleans, Louisiana, USA, 1996. Semmes-Weinstein monofilament wire using the method

357CCC 0742–3071/97/050357–07$17.50 1997 by John Wiley & Sons, Ltd. DIABETIC MEDICINE, 1997; 14: 357–363

ORIGINAL ARTICLESand criteria described by Birke.21 All subjects were concomitant osteomyelitis were also excluded (n = 8).

The preliminary (clinical) diagnosis of osteomyelitis wasdiagnosed with diabetes mellitus based on the criteriaset forth by the World Health Organization.22 All subjects made utilizing a sterile blunt surgical probe.28,29 This was

subsequently confirmed in all cases by both microbiologichad palpable pedal pulses on initial evaluation.We classify Charcot’s arthropathy into two treatment- and histologic analysis via bone and synovial biopsy.

Location of Charcot’s arthropathy was described usingoriented phases based on radiographic,23 dermal thermo-metric,24,25 and clinical signs.26 The initial clinical the Sanders pattern classification. Sanders described five

different patterns of Charcot’s arthropathy based on site.diagnosis of acute Charcot’s arthropathy (as describedabove) has been well documented in the literature.4 Pattern 1 indicates arthropathy in the forefoot. Pattern 2

refers to involvement at Lisfranc’s joint. Pattern 3 affectsFollowing resolution of acute neuropathic osteoarthropa-thy, patients are converted to the post-acute (quiescent) the bones of the lesser tarsus. Pattern 4 affects the ankle

joint. Pattern 5 affects the posterior calcaneus.phase, during which time progressive weight-bearing isintroduced (Figure 1). All patients were initially treated with total contact

casting. The total contact cast consists of an inner layerOnly those patients diagnosed with acute Charcot’sarthropathy were selected for study. Diagnosis was based of plaster with thin felt first applied to the malleoli and

tibial crest and foam to the digits for protection. Theon radiographic, dermal thermometric, and, if necessary,histological analysis. Histological analysis was based on outer splints and remaining layers were made of fibreglass,

with a rubber cast plug secured to the plantar aspect ofbone and synovial biopsy consisting of multiple shardsof bone and soft tissue embedded in the deep layers the cast to increase durability. All casts were applied

using the material and technique described by Komin-of synovium.27 Patients clinically and histologicallydiagnosed or treated prior to presentation to our facility sky.32 Casts were checked weekly and evaluated for

proper fit.were excluded from analysis (n = 18). Patients with

Figure 1. Generalized treatment algorithm

358 D.G. ARMSTRONG ET AL.

ORIGINAL ARTICLESCasts of patients with concomitant ulceration were Results

changed weekly for ulcer evaluation and debridement.The study included 55 subjects, 27 male and 28 female,Cast change intervals for patients without ulcerationwith a mean age of 58.6 ± 8.5 years. Fifty-fourwere dependent on cast comfort and integrity (3 weekssubjects (98 %) were diagnosed with Type 2 (non-insulinmaximum). Casting was discontinued based on clinical,dependent) diabetes mellitus with a mean duration ofradiographic, and dermal thermometric signs of quiesc-15.9 ± 5.7 years. One subject was diagnosed with Typeence. Skin temperatures were objectively monitored1 diabetes, present for 12 years prior to presentation.using a portable infrared thermometric probe (ExergenThere was no significant difference in age or durationModel DT 1001, Exergen Products, Newton, MA, USA).of diabetes among male and female subjects. TheUse of dermal thermometry has been well described inprevalence of acute-onset Charcot’s arthropathy amongthe literature.33,34 If patients presented with bilateralall diabetic patients presenting to the Diabetic Footacute Charcot’s arthropathy, the patients remained inCenter during the study period was 12.9 % (55/426).bilateral total contact casts until both feet/ankles nor-

The average initial thermometric elevation on themalized clinically and radiographically.affected side was 5.1 ± 1.4 °C. There was no significantFollowing casting, patients progressed from removabledifference between skin temperature gradient and locationcast walkers to accommodative footwear with ankle footof neuroarthropathy, nor was there a significant differenceorthoses, as required. Removable cast walkers (Easystepbetween skin temperature gradient between males orWalker, Centec Orthopaedics, Camarillo, California,females. Bilateral Charcot’s arthropathy was present inUSA) were used to ease the transition from total contact9 % (5/55) of the patients presenting in acute Charcot’scasting to full, unprotected weight-bearing in prescriptionarthropathy. Twenty-two patients (40 % of total) initiallyfootwear. The transition to removable cast walker waspresented with a concomitant grade 1a ulceration usingmade when skin temperature gradients were within 1 °Cthe University of Texas Wound Classification System.36,37

for 2 consecutive weeks at the affected site comparedDistribution of these ulcerations correlated with locationwith the corresponding site, contralaterally. The transitionof maximum deformity as described by Sanders’ patternfrom removable cast walker to prescription footwear wasclassification in every case (Table 1).based on 1 month of skin temperature equilibrium on

Location of the deformity, based on the Sandersthe affected site compared to the contralateral extremityclassification, revealed 3 % forefoot, 48 % at the tarso-(± 1 °C). This period of protected weight-bearing providedmetatarsal (Lisfranc’s) joint, 34 % at Chopart’s joint, 13 %

the pedorthic shoe specialist time in which to make andat the ankle joint, and 2 % involving the posterior

fit prescription footwear.calcaneus. Nine patients presented with multiple foci of

Reconstructive surgery was performed only if a deform-Charcot’s arthropathy. Of those 9 patients, 6 presented

ity existed that placed the foot at risk for ulceration andwith a combination of type II and type III. There was

could not be safely accommodated in prescription no significant difference in location distribution offootwear. Surgery was undertaken only after radiographic Charcot’s arthropathy between males and females.signs of coalescence (trabecular bridging) or if clinically When questioned, 73 % could not recall a singleunstable pseudarthroses existed. Following surgery, precipitating event prior to the onset of their symptoms;patients were total contact casted until skin temperatures 22 % recalled a specific traumatic episode within 1and postoperative oedema had returned to normal. As month of onset of symptoms; 4 % related recent footwith those treated non-surgically, following casting, surgery. Although all subjects were clinically insensatepatients progressed to a removable cast walker followed on examination, 76 % of subjects complained of painby permanent prescription footwear. Following transition in the foot upon presentation. There was no significantinto footwear, patients were followed at 2-month intervals difference in those subjects who recalled a singlefor clinical signs of recurrence. Patients were followed traumatic event and duration of casting and returnfor a minimum of 1 year following return to prescription to footwear.footwear (mean 92.6 ± 33.7 weeks). All patients were treated with serial total contact casts

A Mann-Whitney U-test was used to compare differ- and remained casted for an average of 18.5 ± 10.6ences in age, duration of diabetes, length of treatment weeks (range 4 to 56 weeks). All patients returned toamong various dichotomous groups including male and permanent footwear. Progression to permanent footwearfemale, surgical and non-surgical, bilateral and unilateral took 28.3 ± 14.5 weeks. All concomitant ulcerationsarthropathy. A Kruskal-Wallis non-parametric test with healed during the total contact casting regimen prior toone-way analysis of variance was used to measure the quiescence of the neuroarthropathic process. Time ofdifferences between location of neuropathic osteoarthro- casting and time to footwear is further divided by locationpathy, duration of therapy, and skin temperature gradient. in Figure 2. There was no significant correlation betweenFisher’s exact test was used to compare risk of reulceration location of acute Charcot’s arthropathy and duration castfollowing the acute phase of Charcot’s arthropathy.35 therapy or time to footwear. Men were contact castedValues are expressed as mean ± standard deviation. For for a significantly longer period (21.8 ± 12.4 weeks)

than women (15.2 ± 7.6 weeks) (p , 0.008). While menall analyses, we used an alpha level of 0.05.

359ACUTE CHARCOT’S ARTHROPATHY AT DIABETIC FOOT CLINIC

ORIGINAL ARTICLESTable 1. Site of ulceration on presentation with acute Charcot’s arthropathy

Site of ulceration n Corresponding Sanders % of totalClassification ulcerations

Hallux interphalangeal joint 1 1 4.5Cuboid metatarsal articulation 14 2 63.6Talonavicular joint 6 3 27.3Medial malleolus 1 4 4.5

underwent fusions remained casted for an average of29.0 ± 15.5 weeks and returned to permanent footwearin an average of 48.1 ± 16.5 weeks (Figure 3). Patientswho underwent fusions remained casted for a longerperiod of time (p , 0.05) and returned to footwear later(p , 0.02) than the rest of the study population. Fourof five patients undergoing fusions proceeded to bonyunion. One patient had an asymptomatic pseudoarthrosisof the ankle, subsequently controlled in an ankle-footorthosis and custom moulded footwear.

Following the total contact casting regimen, 15 % ofpatients experienced transient increases in temperatureon the formerly symptomatic side. Of these patients,88 % (7/8) exhibited these elevations in the first monthfollowing return to permanent footwear. One patientexperienced transient skin temperature elevation at 35months. None of these patients showed radiographic orclinical signs of recurrent destructive changes. All ofthese patients were placed again on a regimen of totalcontact casting for an average of 2.9 ± 1.2 weeks untiltemperatures normalized.

Figure 2. Treatment duration by location of osseous destruction.Following resolution of acute Charcot’s arthropathy,The Sanders pattern classification refers to five areas of

four patients (7.3 % of total) returned during the follow-osseous destruction. These include the forefoot (pattern 1), thetarsometatarsal articulations (pattern 2), the bones and joints up period with a new-onset neuropathic ulceration.of the lesser tarsus (pattern 3), the ankle joint (pattern 4), and These wounds appeared at a mean 10.0 ± 2.5 monthsthe posterior calcaneus (pattern 5) following return to footwear. All recurrent wounds were

grade 1a in nature. All four patients who ulceratedtook longer to return to footwear (30.2 ± 14.5 weeks) during the follow-up period had midfoot ulcerations onthan women (26.4 ± 14.8 weeks), this association did initial presentation with acute Charcot’s arthropathy, butnot reach significance. reulcerated in the forefoot (Table 2).

Patients presenting with bilateral arthropathy remainedcasted for an average of 28.0 ± 14.7 weeks and returned Discussionto footwear in an average of 48.0 ± 18.2 weeks. Therewas a significant difference in time to return to permanent This study reports the distribution of acute Charcot’s

arthropathy, prevalence of bilateral involvement andfootwear between bilateral and unilateral Charcot’sarthropathy (p , 0.02). All patients with bilateral ulceration, duration to healing, and recurrence of acute

arthropathy and ulceration in people with diabetesarthropathy presented with both feet involved. No patientdeveloped contralateral Charcot’s arthropathy during the mellitus. Our results indicate that the majority of patients

with Charcot’s arthropathy present with pain and a highperiod reviewed.Surgery was performed on 25 % of patients. Of those prevalence (40 %) of neuropathic ulceration at the site

of joint damage. Additionally, the treatment of the diseaseprocedures, 9 were exostectomies (i.e. midfoot planing)and 5 included fusion of affected joints with 2 of the 5 is lengthy, spanning several months of immobilization

in contact casts and cast walkers. After return tofusions receiving tendo-achilles lengthening. There wereno significant differences in surgery or type of surgery prescription footwear, relapse was infrequent with regular

follow-up in a specialty foot clinic. There were noperformed on male versus female patients. Patients withexostectomy procedures remained casted for an average instances of fracture or refracture of the treated or of the

contralateral limb. No amputations resulted as a resultof 15.4 ± 4.2 weeks and returned to permanent footwearin an average of 27.6 ± 14.0 weeks. Patients who of Charcot’s arthropathy in this series.

360 D.G. ARMSTRONG ET AL.

ORIGINAL ARTICLES

Figure 3. Treatment duration

Table 2. Reulceration following resolution of acute Charcot’s arthropathy

Charcot location Site of reulceration Time of reulceration Method of Return to footwear(months following off-loading (weeks)return to footwear)

Chopart’s joint Distal 2nd digit 7 Healing sandal 3Chopart’s joint Distal 3rd digit 13 Healing sandal 4Chopart’s joint Plantar 2nd MT 10 Total contact cast 5Lisfranc’s joint Plantar 3rd MT 10 Total contact cast 5

MT, metatarsal.

While it is generally thought that people with acute The average time of total contact casting (18.5 weeks)is by no means a suggested duration of cast therapy.Charcot’s arthropathy require a long duration of treatment

until quiescence, we have been unable to find any The range of cast duration in this study (between 1month to over 1 year) should indicate to the clinicianreports in the medical literature that discuss the time

period until patients could return to footwear in a cohort that individual patients with acute Charcot’s arthropathyseldom adhere to a standard treatment timetable. Clinicalof patients treated with the same modality. The mean

time of immobilization (casting followed by removeable judgment, coupled with radiographic and thermometricdata, should dictate the appropriate casting regimen andcast walker) prior to return to permanent footwear was

approximately 6 months in our study. Myerson and progression to prescription footwear.The patients who underwent prophylactic surgerycoworkers reported that, following open reduction and

internal fixation of acute Charcot fracture-dislocations in (25 % of the total study population) fell into twocategories. The first category contained patients with a8 patients, mean time of casting was 5 months. There

was no report on duration to permanent prescription relatively stable foot and a readily definable and excisabledeformity. These patients underwent exostoses to removefootwear. Similarly, Brodsky16 reported that patients

receiving an exostectomy for plantar prominences sec- the offending bony apex. This group was casted for farless time than the second category of surgery, whichondary to Charcot’s arthropathy were casted for a

mean 2.4 months. These exostectomies, however, were consisted of patients with clinically unstable deformitiesfollowing coalescence in the post-acute period. Inperformed after the patients reached quiescence. The

duration to quiescence was not reported. The results of general, surgery in the Charcot’s arthropathy patientshould be considered prophylactic in nature. If under-our study indicate that time to unprotected ambulation

in acute Charcot’s arthropathy was approximately two taken, the goal of any procedure should be to enable thepatient to successfully return to accommodative footwear.to five times longer than that of midfoot fractures in

sensate patients without diabetes.38,39 Only 22 % of subjects could recall a specific traumatic

361ACUTE CHARCOT’S ARTHROPATHY AT DIABETIC FOOT CLINIC

ORIGINAL ARTICLESepisode prior to the onset of their condition. This is in responses and thereby effectively eliminating refractures.

This approach is part of a diabetic foot classificationcontrast to an earlier report from Pinzur et al. whoreported that 53 % of their patient population of Charcot’s and treatment system that was implemented in this

specialty foot clinic.26arthropathy gave a history of trauma prior to presentation.It should be considered that, in many cases of neuropathic The four patients who had recurrent neuropathic

ulcerations during the follow-up period all had neuro-arthropathy, a single traumatic event does not take placeat all. Rather, it has been suggested that, much in the pathic ulcerations present in the midfoot on initial

presentation. It has been reported that, following healingway a neuropathic plantar ulcer forms, it is repetitiveminor trauma to an insensate region that may precipitate a of a neuropathic ulcer, 19 % to 58 % of patients develop

another ulcer within a year.45–47 Interestingly, in ourlarge percentage of acute neuroarthropathic events.7,40,41

Although neuropathy may have obscured sensation and study, all recurrent ulcerations occurred in the forefoot.We attribute this to changes in foot structure andthus recognition of any specific trauma, nearly three-

quarters of the subjects related ill-defined foot pain. mechanics which, with progression of neuropathicosteoarthropathy (and sensorimotor neuropathy inThe overall prevalence of acute Charcot’s arthropathy

among all patients presenting for care at the clinic general), may cause clawing of the digits and subsequentplantarflexion of the metatarsal heads, increasing theirreviewed during the study period (13 %) was higher than

in other published reports,42,43 in which prevalences prominence thus predisposing them to neuropathic ulcer-ation.ranged from 0.08 % to 7.5 %. Cofield et al. indicated

that in diabetic patients with peripheral neuropathy, the The most important facet of treatment of diabeticneuropathic osteoarthropathy is prevention. Diabeticprevalence climbs to 29 %.43 None of these reports make

a clear distinction between acute and post-acute Charcot’s patients with sensory neuropathy should be examinedregularly for increases in pedal skin temperature.25,34arthropathy. For this reason, the comparison made may

not be entirely accurate. It should also be noted that the Because temperature assessment by palpation is relativelyinsensitive and qualitative, the use of infrared dermalstudy site was a referral-based specialty diabetic foot

clinic. However, we believe that our relatively high thermometry has been employed in our clinic to objec-tively and quantitatively assess subtle temperature differ-prevalence is at least partially attributable to heightened

awareness of the signs and symptoms of the condition ences. Those at greatest risk must be educated as to theappropriate signs and symptoms of acute Charcot’sin its acute phase.

The prevalence of bilateral Charcot’s arthropathy was arthropathy, with any traumatic episode being reportedas soon as possible.9 % in this study population. This is substantially lower

than other published studies, which report prevalence With the exception of osteomyelitis, diabetic neuro-pathic osteoarthropathy is perhaps the most debilitatingof bilateral arthropathy as high as one-third to two-thirds

of cases.13,15 Perhaps more interesting than prevalence orthopaedic condition to strike the diabetic lowerextremity. Its insidious onset and rapid progression makesof bilateral acute arthropathy is whether contralateral

events occur during treatment. In our study no patients early diagnosis and rapid institution of treatment criticalto arriving at an acceptable long-term outcome. Wedeveloped contralateral arthropathy. We attribute this to

the use of total contact casting. Patients who are placed have outlined the history and pathogenesis of diabeticneuropathic osteoarthropathy and suggested an evalu-in these casts are able to ambulate freely during the

majority of treatment. This, coupled with reduced stride ation and treatment protocol for the acute form of thisdevastating disease. The analysis of our treatment methodslength and decreased cadence, expose the contralateral

extremity to less repetitive trauma that might occur if suggest they are competitively effective. The central tenetto be followed in treatment of acute Charcot’s arthropathythe patient walked with crutches.44 These factors com-

bined with frequent monitoring and appropriate prescrip- is prompt, uncompromising reduction in weight-bearingstress, frequent monitoring, and gradual progression totion footwear may have reduced the risk of precipitation

of a bilateral episode. unprotected weight-bearing in prescription footwear. Theambulatory total contact cast has proven very effectiveFollowing return to permanent footwear, 15 % of

the population exhibited unilateral elevations in skin in this endeavour. Regardless of the specific treatmentmethod instituted, it is imperative that appropriatetemperatures without other objective signs of Charcot’s

arthropathy, mostly in the first month after returning to and aggressive treatment be undertaken immediatelyfollowing diagnosis to help prevent progression to apermanent footwear. We believe that in the majority of

these cases this transient increase in temperature may profoundly debilitating, limb-threatening deformity.have been a subclinical vascular response to theadditional stress placed on the affected limb. Whetherthis response was a precursor to a recurrent attack of ReferencesCharcot’s arthropathy is not clear. If it were, the rapidreinstitution of total contact casting, frequent temperature 1. Kelly M. William Musgrave’s De arthritide symptomaticamonitoring, and slow reintroduction to footwear was (1703): his description of neuropathic arthritis. Bull Hist

Med 1963; 37: 372–376.probably instrumental in treating early inflammatory

362 D.G. ARMSTRONG ET AL.

ORIGINAL ARTICLES2. Kelly M. John Kearsley Mitchell and the neurogenic theory 27. Horwitz T. Bone and cartilage debris in the synovial

membrane. Its significance in the early diagnosis ofof arthritis. J Hist Med 1965; 20: 151–157.3. Eloesser L. On the nature of neuropathic affections of the neuroarthropathy. J Bone Joint Surg 1948; 30A: 579–588.

28. Grayson ML, Balaugh K, Levin E, Karchmer AW. Probingjoint. Ann Surg 1917; 66: 201–206.4. Sanders LJ, Frykberg RG. Charcot’s Joint. In: Levin ME, to bone in infected pedal ulcers. A clinical sign of

underlying osteomyelitis in diabetic patients. J Am MedO’Neal LW, Bowker JH, eds. The Diabetic Foot, 2nd edn.Saint Louis: Mosby-Year Book, 1993. Assoc 1995; 273: 721–723.

29. Caputo GM. Infection: investigation and management. In:5. Charcot JM. Sur quelques arthropathies qui paraissentdepender d’une lesion du cerveau ou de la moele Boulton AJM, Connor H, Cavanagh PR, eds. The Foot in

Diabetes, 2nd edn. Chichester: Wiley, 1994.epiniere. Arch Des Physiol Norm et Path 1868; 1: 161–171.6. Edmonds ME, Clarke MB, Newton JB, Barrett J, Watkins 30. Sanders LJ, Mrdjencovich D. Anatomical patterns of bone

and joint destruction in neuropathic diabetics (Abstract).PJ. Increased uptake of radiopharmaceutical in diabeticneuropathy. Q J Med. 1985; 57: 843–855. Diabetes 1991; 40 (suppl 1): 529A.

31. Harris JR, Brand PW.Patterns of disintigration of the tarsus7. Brower AC, Allman RM. The neuropathic joint: a neurovas-cular bone disorder. Radiologic Clinics of North America in the anesthetic foot. J Bone Joint Surg 1966; 48B: 4–16.

32. Kominsky SJ. The ambulatory total contact cast. In: RG1981; 19: 571–579.8. Todd WF, Laughner T, Samojla BG. The diabetic foot. Frykberg, ed. The High Risk Foot in Diabetes Mellitus,

1st edn. New York: Churchill Livingstone, 1991: 449–455.In: Robbins JM, ed. Primary Podiatric Medicine, 1st edn.Philadelphia: WB Saunders, 1994: 213–245. 33. Armstrong DG, Lavery LA. Monitoring neuropathic ulcer

healing with infrared dermal thermometry. J Foot Ankle9. Finsterbush A, Friedman B. The effect of sensory dener-vation on rabbits’ knee joints. J Bone Joint Surg 1975; Surg 1996; 35: 335–338.

34. Armstrong DG, Lavery LA, Liswood PL, Todd WF, Tredwell57A: 949–957.10. Banks AS, McGlamry ED. Charcot foot. J Am Podiatr J. Infrared dermal thermometry of the high risk diabetic

foot. Physical Therapy 1997; 77: 169–177.Med Assoc 1989; 79: 213–217.11. Norman A, Robbins H, Milgram JE. The acute neuropathic 35. Kirkwood BR. Essentials of Medical Statistics. Oxford:

Blackwell, 1988.arthropathy. Radiology 1968; 90: 1159–1164.12. Papa J, Myerson M, Girard P. Salvage with arthrodesis in 36. Lavery LA, Armstrong DG, Harkless LB. Classification of

diabetic foot wounds. J Foot Ankle Surg 1996; 35:intractable diabetic neuropathic arthropathy of the footand ankle. J Bone Joint Surg 1993; 75A: 1056–1066. 528–531.

37. Lavery LA, Armstrong DG, Walker SC. Healing rates of13. Clohisy DR, Thompson RC. Fractures associated withneuropathic arthropathy in adults who have juvenile- diabetic foot ulcers associated with midfoot fracture due

to Charcot’s arthropathy. Diabetic Medicine 1997; 14:onset diabetes. Journal of Bone and Joint Surgery 1988;70A: 1192–1200. 46–49.

38. Arntz CI, Veith RG, Hansen ST. Fractures and fracture-14. Sinha S, Munichoodapa CS, Kozak GP. Neuroarthropathy(Charcot joints) in diabetes mellitus. Medicine 1972; 51: dislocations of the tarsometatarsal joint (Abstract). J Bone

Joint Surg 1988; 70A: 171–181.191–210.15. Myerson MS, Henderson MR, Saxby T, Wilson-Short K. 39. Cass JR. Overview of injuries to the anterior tarsus and

navicular. In: Jahss M, ed. Disorders of the Foot andManagement of midfoot diabetic neuroarthropathy. FootAnkle 1994; 15: 233–241. Ankle. Philadelphia: WB Saunders, 1991: 1321–1332.

40. Delano PJ. The pathogenesis of Charcot’s joint. Am J16. Brodsky JW, Rouse AM. Exostectomy for symptomaticbony prominences in diabetic Charcot feet. Clinical Roentgenol 1946; 56: 189–200.

41. Young MJ, Marshall A, Adams JE, Selby PL, BoultonOrthopaedics and Related Research 1993; 296: 21–26.17. Shibata T, Tada K, Hashizume C. The results of arthrodesis AJM. Osteopenia, neurological dysfunction, and the

development of Charcot neuroarthropathy. Diabetes Careof the ankle for leprotic neuroarthropathy. J Bone JointSurg 1990; 72A: 749–756. 1995; 18: 34–38.

42. Cavanaugh PR, Young MJ, Adams JE, Vickers KL, Boulton18. Cohn BT, Brahms MA. Diabetic arthropathy of the firstmetatarsal cuneiform joint. Introduction of a new surgical AJM. Radiographic abnormalities in the feet of patients

with diabetic neuropathy. Diabetes Care 1994; 17:fusion technique. Orthop Rev 1987; 16: 465–470.19. Stuart MJ, Morrey BF. Arthrodesis of the diabetic neuro- 201–209.

43. Cofield RH, Morrison MJ, Beabout JW. Diabetic neu-pathic ankle joint. Clin Orthop 1990; 253: 209–211.20. Samilson RL, Sankaran B, Bersani FA, Smith AD. Ortho- roarthropathy in the foot: patient characteristics and

patterns of radiographic change. Foot Ankle 1983; 4:pedic management of neuropathic joints. Arch Surg 1959;78: 119–121. 15–22.

44. Armstrong DG, Liswood PL, Todd WF. The contralateral21. Birke JA, Sims DS. Plantar sensory threshold in theulcerated foot. Lepr Rev 1986; 57: 261–267. limb during total contact casting: a dynamic pressure and

thermometric analysis. J Amer Podiatr Med Assn 1995;22. World Health Organization. Second Report on DiabetesMellitus. Geneva: WHO, 1980. 85: 733–737.

45. Uccioli L, Faglia E, Monticone G, Favales F, Durola L,23. Eichenholz SN. Charcot Joints. Springfield: Charles C.Thomas, 1966. Aldeghi A, et al. Manufactured shoes in the prevention of

diabetic foot ulcers. Diabetes Care 1995; 18: 1376–1378.24. Armstrong DG, Lavery LA, Liswood PL, Todd WF, TredwellJ. Infrared dermal thermometry of the high risk diabetic 46. Helm PA, Walker SC, Pulliam GF. Recurrence of neuro-

pathic ulcerations following healing in a total contactfoot (Abstract). Diabetologia. 1996; 39 (suppl 1): A266.25. Armstrong DG, Lavery LA. Monitoring healing of acute cast. Arch Phys Med Rehabil 1991; 72: 967–970.

47. Bergtholdt HT. Temperature assessment of the insensitiveCharcot’s arthropathy with infrared dermal thermometry.J Rehabil Res Dev 1997: in press. foot. Phys Ther 1979; 59: 18–22.

26. Armstrong DG, Lavery LA, Harkless LB. Treatment-basedclassification system for assessment and care of diabeticfeet. J Am Podiatr Med Assn 1996; 86: 311–316.

363ACUTE CHARCOT’S ARTHROPATHY AT DIABETIC FOOT CLINIC