lable at ScienceDirect

The Journal of Foot & Ankle Surgery 50 (2011) 648–653

Contents lists avai

The Journal of Foot & Ankle Surgery

journal homepage: www.j fas .org

Trabecular Quality and Cellular Characteristics of Normal, Diabetic,and Charcot Bone

Javier La Fontaine, DPM, MS, FACFAS 1, Naohiro Shibuya, DPM, MS, FACFAS 2, H. Wayne Sampson, PhD 3,Pilar Valderrama, DDS, MS 4

1Associate Professor, Department of Surgery, Podiatry Division, Texas A&M Health Science Center, College of Medicine, and Chief, Podiatry Section, Central Texas Veterans HealthcareSystem, Surgical Service, Temple, TX2Associate Professor, Department of Surgery, Podiatry Division, Texas A&M Health Science Center, College of Medicine, and Central Texas Veterans Healthcare System, Surgical Service,Temple, TX3 Professor, Department of Systems Biology and Translational Medicine, Texas A&M Health Science Center, College of Medicine, Temple, TX4 Private Practice, Cedar Park, TX

a r t i c l e i n f o

Level of Clinical Evidence: 3Keywords:bone mineral densitydiabetes mellitushistologyneuroarthropathytrabeculae

Financial Disclosure: This study was supportedAmerican College of Foot and Ankle Surgery (Chicago

Conflict of Interest: None reported.Address correspondence to: Javier La Fontaine

Professor, Department of Surgery, Podiatry Division, TCollege of Medicine, and Central Texas Veterans HeaTemple, TX 76504.

E-mail address: javierlafontaine@gmail.com (J. LaAudio file online only at http://www.jfas.org

1067-2516/$ - see front matter � 2011 by the Americdoi:10.1053/j.jfas.2011.05.005

a b s t r a c t

Charcot neuroarthropathy (CN) is a disabling and devastating condition that affects many neuropathic diabeticpatients. It can lead to foot deformity, ulceration, and lower extremity amputation. The pathogenesis of CN isnot clear, but 1 possible predisposing factor is increased bone turnover and increased osteoclastic activity.Although the affect of diabetes on bone is not entirely clear, studies have shown increased bone fragility indiabetics with neuropathy. The purpose of the present study was to compare the bone quality histologicfindings and trabecular histomorphometry, including the cellular characteristics, between normal subjects(N ¼ 7), diabetics without CN (N ¼ 8), and diabetics with CN (N ¼ 8). Histologically, the bone in diabetics withCN displayed an inflammatory, myxoid infiltrate. We also observed a statistically significant decrease in thenumber of trabeculae in bone in diabetics with CN compared with normal controls (p < .02). However, thedifference between the trabeculae in diabetics with CN and diabetics without CN was not statisticallysignificant (p > .05). Histologically, the CN bone appeared to be infiltrated with inflammatory myxoid tissueand had a disorganized trabecular pattern compared with diabetic bone without CN and normal bone. Thetrabeculae in patients with CN appeared to have poor quality characteristics compared with that of the othergroups. The findings from the present study might indicate that diabetes mellitus bone is fragile, and thedecrease in the cellular component might impair the reparative process in those with CN foot.

� 2011 by the American College of Foot and Ankle Surgeons. All rights reserved.

Charcot neuroarthropathy (CN), also known as Charcot arthrop-athy and Charcot’s joint, was first described by Jean-Marie Charcot (1)in 1868. CN is characterized clinically by a gradual destruction of thebone and joint tissues in patients with neurosensory loss and is mostcommonly observed in the neuropathic diabetic patient (2,3). Anumber of theories have been proposed to explain the etiology of CN.These theories suggest that osteopenia results from uncontrolledosteoclastogenesis and blood flow after the development of auto-nomic neuropathy (4). Two major etiologic theories are attributed to

by a research grant from the, IL).

, DPM, MS, FACFAS, Associateexas A&M Health Science Center,lthcare System, Surgical Service,

Fontaine).

an College of Foot and Ankle Surgeon

the development of Charcot arthropathy. In 1868, Charcot describedwhat is now referred to as the French neurovascular theory of boneand joint degeneration (1). After studyingmore than 5000 chronicallyill patients, Charcot concluded that the profound joint destruction anddeformities that he observed were secondary to changes in thetrophic centers of the spinal cord, specifically, diseased anterior horncells that caused in autonomic neurogenic and circulatory disruptionthat resulted in hyperemic demineralization, leading to osteopenia. In1870, Volkmann described what is now referred to as the Germanneurotraumatic theory of bone and joint degeneration (5), which isattributed to unperceived trauma to an insensate joint, with a resul-tant stress fracture that leads to progressive and permanent defor-mation and persistent biomechanical imbalance. In general, theassociation of bone degeneration with neuropathy is suggestive ofneural control of skeletal homeostasis (6,7). Also, evidence isemerging that the nervous system and certain neuropeptides mightinfluence bone metabolism (7). Also, CN is exemplified by prolongedinflammation and osteolysis that might be similar to that observed

s. All rights reserved.

J. La Fontaine et al. / The Journal of Foot & Ankle Surgery 50 (2011) 648–653 649

in rheumatoid arthritis, periodontal disease, and hip implant failure(8–10).

Often surgeons encounter soft bone when operating on patientswith Charcot arthropathy, especially in the earlier stages of the disease.In later stages, Charcot bone seems to be harder and more brittle.These different characteristics of bone influence the surgeon’s decisionregarding the use of bone fixation techniques. To our knowledge, nobench or human studies have beenpublished that show the qualitativechanges in Charcot bone. Although a low bone mineral density (BMD)is consistently observed in patients with type 1 diabetes, in type 2diabetes, the BMD is similar to or greater than that in nondiabeticsubjects (11). However, overall, the bone of a diabetic patient,regardless of the BMD, appears to be more fragile than the bone ofnondiabetic patients (12,13). Recent cohort studies have confirmedthat the bone in type 1 and type 2 diabetic patients is associated witha greater risk of fracture than it is in nondiabetic patients (14–16).

Bone morphology and physiology is influenced by cellular activity,organized within different types of structural framework. Trabecularbone enveloped by cortical bone provides resistance to compressiveand tensile forces that enable the skeleton to support the body.Trabecular bone forms a 3-dimensional branching lattice alignedalong areas of stress in which compression seems to predominate.Cortical bone, which surrounds cuboid (small, irregular) bones andthe diaphysis of long bones, is suited to resist bending and torsionalforces, as well as compressive forces.

Microscopically, bone presents in 2 basic forms: woven andlamellar bone. Woven bone is considered primitive, immature bone. Ithas disorganized collagen fibers with a high cellular content and aninconsistent mineral content and is found in disease states such asneoplasm, osteogenesis imperfecta, and Paget’s disease. Lamellar boneis a mature structure that results from remodeling of pre-existingwoven bone. Woven and lamellar bones are structurally organizedinto trabecular (cancellous) and cortical (compact) components, andthe metabolic turnover of trabecular bone has been reported to be 8times greater than that of cortical bone (17). When bone is studiedunder polarized light, a clear lamellar pattern is visible in cortical andcancellous bone.However, if thebone turnover is highordisturbed, thelamellar pattern disappears, and woven bone is formed.

The major types of bone cells include osteoblasts, osteocytes, andosteoclasts. An osteocyte is a differentiated osteoblast, the predomi-nant cell type in bone. The function of the osteocyte is to maintainbone homeostasis by sensing and responding to mechanical, elec-trical, and chemical stimuli. The osteoclast reabsorbs bone and, alongwith the osteoblast, maintains the balance between bone productionand resorption by a process known as coupling. Clinically and surgi-cally, it can be difficult to determine the optimal time to immobilizeand off-load, or to operate on, bone in patients with CN. For manyyears, the treatment of the acute stage of CN has focused on immo-bilization. Lately, however, it has been proposed that early surgicalintervention is indicated in an effort to prevent the development ofulceration and future amputations (18–20).

Because of the poor bone quality, different treatment modalities toimprove healing in patients with CN have been suggested, includingthe use of external fixation, combinations of internal and externalfixation, early use of bone growth stimulation, different types of bonegraft materials, and combinations of these modalities (18–23). More-over, our reviewof the publisheddata showed that a “stable nonunion”has been considered an acceptable outcome (21,24). Clearly, a betterunderstanding of the cellular and metabolic characteristics of bone inpatients with CN could provide information that is likely to be helpfulin the treatment of patients with this problem. Thus, we undertook aninvestigation toquantify thevolumeof trabecular boneand the cellularcharacteristics among samples harvested from normal patients, dia-betic patients, and diabetic patients with CN.

Patients and Methods

Bone tissue was obtained from patients who were undergoing an osseous proce-dure as a part of corrective foot surgery. To investigate the bone quality and cellularfindings, bone specimens from the followingmatched groups of patients undergoing anosseous foot procedure were studied: (1) patients without diabetes (group 1/control;N ¼ 7), (2) diabetic patients with or without peripheral neuropathy (group 2/DM;N ¼ 8), and (3) diabetic patients with CN (group 3/CN; N ¼ 8). Five patients in theCharcot group were in the active (acute, inflammatory) phase. All patients and controlsubjects gave informed consent before their participation in the investigation, and theinstitutional review board of the Central Texas Veterans Healthcare System approvedthe study.

The diagnosis of diabetes mellitus was in accordance with World Health Organi-zation criteria. Neuropathy was diagnosed and confirmed on the basis of the absence ofappreciation of >25 V of electromotive force measured with a biothesiometer(Biomedical Instruments, Newbury, OH), and 5.07-gauge monofilament. CN was diag-nosed using clinical and radiologic parameters, as described by Eichenholtz (25). AcuteCNwas determined radiographically and clinically. Radiographically, it was determinedby the appearance of decreased bone density, periarticular fragmentation, fracture, andjoint collapse, as viewed on standard radiographs. Clinically, the diagnosis was deter-mined by the presence of cutaneous warmth (relative to the adjacent skin), erythema,swelling, pain, and bounding dorsalis pedis and/or posterior tibial pulses, or the use ofskin thermometry with a>4�F increase comparedwith the contralateral, quiescent foot(26).

Bone samples were acquired from consecutive subjects 18 to 25 years of age, maleand female, with type 1 or type 2 diabetes and from the control subjects, with orwithout neuropathy secondary to diabetes, and able to provide informed consent.Patients with an active neuropathic ulcer or a history of neuropathic ulceration wereexcluded from the present study to avoid procurement of specimens potentiallyaffected by osteomyelitis. Also, patients currently or previously taking medicationsdesigned to affect the bonemetabolism, as well as thosewith known osteoporosis, end-stage renal disease, kidney transplantation, systemic inflammatory disease, active orprevious radiotherapy or chemotherapy, chronic corticosteroid use, or osteomyelitiswere excluded.

The patients underwent the surgical procedure as planned. Elective reconstructivesurgical interventions served as the indication for operation for the patients in thecontrol group. Elective or prophylactic surgery was the indication for operation in theDM and CN groups. The bone specimens were procured using sharp excision with anosteotome and mallet or a power saw. The bone specimens from the DM and controlgroups were collected from different locations in the foot, depending on the indicationfor the surgical procedure. All CN bone samples were collected from the cancelloustarsal bones, depending on the location of the Charcot process. Additional bone sampleswere obtained from representative locations in the patients in the DM and CN groupsusing a 3-mm Jamshidi needle, as a part of the intervention, with the intent of differ-entiating the Charcot process from possible osteomyelitis. The area of bone biopsy wasapproached percutaneously with the Jamshidi needle, after preparation of the skin and,if needed, local anesthetic infiltration. Once procured, the study specimens werepreserved in formaldehyde and immediately stored in −80�F freezer.

Histologic Examination

All bone specimens were analyzed at the end of the study. The samples weretransported to the Department of Systems Biology and Translational Medicine of theTexas A&M Health Science Center (Temple, TX), for histologic and histomorphometricanalysis. The bone samples were fixed in Carson’s modified Millonig’s phosphate-buffered formalin for 48 hours and then decalcified in 4% ethylene-diamine-tetra-acetic acid. Thereafter, the specimens were embedded in paraffin and cut into 4-mmsections and mounted on coated slides. Furthermore, every fifth section, 3 per slide,was isolated and stained with hematoxylin and eosin.

Imaging

The sections were analyzed using an osteometric system attached to an OlympusBX60 light microscope with a 100-W mercury light source and a superwide band filtercube for ultraviolet radiation (Olympus, Tokyo, Japan). All histomicrographic imageswere taken at 1.25 to 40� magnification, as indicated in the images depicted in thisreport.

Histomorphometry

The sections for histomorphometry were analyzed using an OsteoMeasure HighResolution Digital Video System (OsteoMetrics, Decatur, GA) attached to an OlympusBX60 light microscope with a 100-W mercury light source and a superwide band filtercube for ultraviolet radiation. The bone volume/total volume ratio was measured as theamount of actual bone tissue in a predetermined area of cancellous bone. The trabec-ular thickness was measured as the mean average thickness of each piece of trabecularbone in a predetermined area of cancellous bone. The number of trabeculae (TbN) in

Table 1Demographic description of diagnostic groups (N ¼ 23)

Variable Normal(n ¼ 7)

DiabetesMellitus (n ¼ 8)

CharcotNeuroarthropathy(n ¼ 8)

2-tailedp Value*

Median age (yr) 49 (37, 68) 52 (32, 72) 53 (46, 66) .0609Male gender 4 (57.14%) 7 (87.5%) 8 (100%) .050 to .388Type 2 diabetes NA 8 (100%) 8 (100%) >.05

Abbreviation: NA, not applicable.* 1-way ANOVA.

Table 3Mann-Whitney U test for difference in trabeculae number between Charcot and normalgroups

Group Patients (N) Rank Sum Mean Rank U Statistic

Normal* 7 76.0 10.86 8.0Charcot 8 44.0 5.50 48.0Median difference .4598 d d d

96% confidence interval .0633 (0.876, 2.845) d d d

Mann-Whitney statistic 8.0 d d d

2-tailed p value .0205 d d d

* The normal group consisted of patients without diabetes, without peripher-alneuropathy, and without Charcot bone.

J. La Fontaine et al. / The Journal of Foot & Ankle Surgery 50 (2011) 648–653650

a predetermined area of cancellous bone was determined using the following mathe-matical formula: [10 � (bone volume/total volume)]/(trabecular thickness).

Statistical Analysis

Nonparametric testing was used because the data were not normally distributed.The Kruskal-Wallis test was used to compare the bone trabecular volume between the 3diagnostic groups, and the Mann-Whitney U test was used to compare the trabecularvolume between the normal and CN groups and the normal and DM groups. Spear-man’s rank correlation test was used to evaluate the associations between the amountof bone trabeculae and age. The a-level was adjusted to 0.025 to account for themultiple testing, using Sidak’s correction method.

Results

The descriptive demographic results for the 3 diagnostic groupsare listed in Table 1. No statistically significant differences were notedin age and gender among the 3 groups.

Histomorphometric Analysis

Comparison of the TbN within all groups is given in Table 2. Nostatistically significant differences were found in the TbN among the3 diagnostic groups or between the control group and the DMgroup or the DM and CN groups. However, a statistically significantdifference was noted in the TbN between the control and CN groups(p < .020; Table 3).

Table 2Age, bone sampled, and number of trabeculae by diagnostic group (N ¼ 23)

Group Age (yr) Bone Sampled Trabecular Number*

Normal* (n ¼ 7)49 Talus 1.505039 Cuneiform 2.845868 Calcaneus 1.787649 Metatarsal 2.343557 Metatarsal 1.779837 Metatarsal 1.591649 Cuneiform 1.9565

Diabetes mellitus (n ¼ 8)32 Phalanx 2.499166 Metatarsal 1.356770 Metatarsal 2.344352 Metatarsal 2.524772 Phalanx 1.561951 Metatarsal 1.599252 Phalanx 1.548657 Metatarsal 0.8765

Diabetes mellitus withCharcot neuroarthopathy (n ¼ 8)

62 Cuneiform 0.912766 Cuneiform 1.465059 Talus 1.215546 Navicular 1.263959 Navicular 1.446346 Cuboid 2.092652 Navicular 1.528353 Talus 1.6517

* The normal group consisted of patients without diabetes, without peripheralneuropathy, and without Charcot bone. p ¼ .3088.

Qualitative Histologic Analysis

A pathologist and bone biologist, whowere unaware of the patientgroup, evaluated each specimen. The bone quality was noted to varywith the location from which the bone sample was harvested. Lightmicroscopy evaluation of the bone sample in decalcified sectionsshowed differences between the trabeculae and cellularity among all3 diagnostic groups. In the normal group, the cortical bone displayedmature lamellar organization with organized osteons and Haversiansystems, osteoblasts were noted to line the trabeculae, and osteocyteswere present in the lamellar bone. Stained areas of new bone

Fig. 1. (A) Hematoxylin and eosin stain (4� magnification) of normal bone demonstratingsurface remodeling (blue arrow) and osteoclasts (Oc) with well organized lamellar bone(Lb) and Haversian system (Hs). (B) Hematoxylin and eosin stain (1.25� magnification) ofnormal bone marrow spaces occupied in most by fatty marrow (Fm); lamellar bone (Lb)can be observed as dark pink stain representing well organized, mature bone.

Fig. 2. (A) Hematoxylin and eosin stain (10� magnification) of DM bone demonstratingorganized lamellar bone with osteocytes present but few empty lacunae (Lc). Note, bonemarrow spaces filled with fatty marrow (Fm), decreased number of osteoblasts (Ob) liningtrabeculae, and calcified vessel (Cv). (B) Hematoxylin and eosin stain (1.25� magnifica-tion) of normal bone demonstrating bone trabeculae (Tb) that appears thinner and fewerin number. Scarce areas of bone remodeling observed. Note, calcified vessels within fattymarrow (Fm).

Fig. 3. (A) Hematoxylin and eosin stain (1.25� magnification) of CN bone demonstratingthin osteoid-like trabeculae (Tb) surrounding inflammatory infiltrate (InF). (B) Hema-toxylin and eosin stain (4� magnification) showing myxoid tissue (Mt) within inflam-matory infiltrate. Also, immature, structurally disorganized bone seen (black arrow). (C)Hematoxylin and eosin stain (10� magnification) showing marked increase of osteoclasticactivity evidenced by presence of Howship’s lacunae (HL) and osteoclast (OC), followed byline of osteoblasts (Ob). Note, inflammatory infiltrate dominated by lymphocytes, eosin-ophils (Es), spindle fibroblasts (Fb), and significant vascular congestion (Vc).

J. La Fontaine et al. / The Journal of Foot & Ankle Surgery 50 (2011) 648–653 651

apposition on the periosteal side of the cortical sections revealedsurface remodeling, and the trabecular sinusoids were occupied inmost cases by fatty marrow (Fig. 1). In the DM group, the trabeculaewere formed mainly by lamellar bone in which osteocytes werepresent but to a lesser degree than that seen in the normal group.Moreover, therewere few lacunae that did not contain osteocytes, andthe marrow spaces were filled with fatty marrow. The trabeculae inthe DM group were thinner and fewer than those observed in thenormal group, evidence of remodeling was scarce, and fewer osteo-blasts were lining the trabecular. Overall, the bone in the DM groupdisplayed less cellularity, and the blood vessels in the marrow spacesshowed marked thickening of the tunic media (Fig. 2).

In the CN group, the bone presented characteristics of reactivebone, including the presence of woven bone that was immature andstructurally disorganized. The osteoid-like trabeculae displayeda surrounding inflammatory infiltrate dominated by lymphocytes andeosinophils. Moreover, the marrow spaces were infiltrated withhypervascular, myxoid tissue with spindle fibroblasts, and substantialvascular congestion was present. Furthermore, there was an increasein the number of Howship’s lacunae, indicative of increased osteo-clastic activity. Furthermore, the number of osteocytes was less thanthat observed in the normal and DM groups (Fig. 3), and some emptylacunae were visible.

Discussion

We do not yet have a clear understanding of how diabetes affectsthe bone metabolism. Several theories, including hyperglycemia,hypoinsulinemia, growth factor deficiency, and neuropathy, have

J. La Fontaine et al. / The Journal of Foot & Ankle Surgery 50 (2011) 648–653652

been proposed as causes of abnormal bone physiology in patientswith DM and CN (27–31). Several experimental and clinical obser-vations have suggested that bone abnormalities associated with dia-betes might differ, at least in part, from those associated with senile orpostmenopausal osteoporosis (13). Lips et al (27) demonstrated, ina small sample, that a linear decrease occurs in bone wall thicknessassociated with aging, especially after 50 years of age. In our study,age did not significantly correlate with the trabecular bone volume.Bone fragility has been shown to increase in patients with diabetesmellitus. Petrova et al (32) suggested that osteopenia can be found atpresentation in the contralateral, unaffected foot in patients with type1, but not type 2, diabetes. Sinacore et al (33) suggested that theprolonged inflammatory response observed in CN contributes toosteolysis and the loss of BMD in the calcaneus. They compared 22patients with diabetes, neuropathy, and midfoot deformities to 29healthy controls. The control group’s BMD was 13% greater than thatof the subjects with diabetes and neuropathy. Among the diabetics,the BMD was 16% lower in the involved foot than in the uninvolvedfoot without deformity (34). Our results have confirmed that anabnormal bone structure is present, namely fewer trabeculae andfewer cells, that might predispose patients with diabetes to developCN and that the traditional neurovascular and neurotraumatic theo-ries might not be the only explanations for development of bone andjoint degeneration in this condition.

Although CN has been recognized for many years, the preciseetiology and exact pathophysiology of this entity remains unknown.Clinically, acute CN presents as an exaggerated inflammatoryresponse secondary to a wide range of traumatic events. Radio-graphically, the bone is fragmented in the acute stage with jointdislocation, although it remodels in the later stages. According toFrykberg and Sanders (35), the most common sites of pedal involve-ment with CN, in descending order from the most prevalent, are thetarsometatarsal joints (30%), metatarsophalangeal joint (30%), inter-tarsal joints (24%), ankle (11%), and interphalangeal joint (4%). Clini-cally, these bones are composed largely of trabecular (cancellous)bone, such as the metaphysis of the metatarsal heads. It is clear thattrabecular bone is important for the framework of bone support, andit has the greatest rate of bone turnover compared with cortical bone(17). The histologic features of the bone specimens in the presentinvestigation demonstrated thin trabeculae and fatty marrow in theDM specimens. Moreover, the diabetic trabeculae showed the samepattern as the control (normal) specimens, although they werethinner, with fewer trabeculae and less remodeling. These findingssuggest that the bone in diabetic patients is more fragile and, as such,might be predisposed to excessive deformation in response to load.The bone in the CN group appeared disorganized and immature,consistent with woven bone, which is known to be less rigid and lesssupportive than normal bone. The CN group also demonstratedsubstantial inflammatory infiltrate with hypervascular myxoid tissueand spindled fibroblasts. Our findings also showed a statisticallysignificant decrease in the volume of trabeculae between the CNgroup and normal group. Although, no statistically significant differ-ence was found between the DM and normal bone, the histologicexamination showed that the diabetic bone had less trabeculae thandid the normal bone. The characteristics observed in the CN groups,moreover, might explain the clinically observed lack of resistance tocompressive forces and an inability to maintain foot structuralalignment, which leads to the pedal collapse associated with theCharcot foot. It is also interesting to consider that the presence ofmyxoid tissue within inflammatory infiltrates might explain whyfracture nonunion is frequently encountered in the Charcot foot.

The effect of neuropathy on bone is also not precisely known,although the existing data suggest a connection between neuropathyand cellular activity. Young et al (36) showed an association of small

fiber neuropathy with increased osteoclast activity, resulting inreduced BMD and increased neuropathic fracture risk. A possiblemechanism by which neuropathy affects bone turnover is theneuropeptide calcitonin-gene related peptide, which is down-regulated in patients with neuropathy (37). In our study, an increasein osteoclastic activity was evidenced by the presence of moreHowship’s lacunae, osteoclasts, and woven bone in the CN group. Thisfinding was also consistent with the findings from Young et al (36), inwhich Charcot bone had a greater rate of bone turnover than normalbone. Also, in the present study, fewer osteocytes were observed inthe DM and CN groups, and this might suggest that chemical signalingto maintain bone integrity in CN could be impaired and thatcalcitonin-gene related peptide availability might be compromised.These findings, in sum, are consistent with the increased osteoclasticactivity found in association with peripheral neuropathy and CN.

As with all observational investigations, it is difficult to makea claim of causality. Thus, we could not state with certainty that thealterations in bone structure were directly responsible for the Charcotfoot in those patients with that diagnosis. A number of other meth-odologic shortcomings could also have influenced our results,including the selection of both acute and chronic Charcot foot in thesample, small sample size, and combination of bone specimens har-vested from short and long bones.

In conclusion, the results of the present study showed that osseousabnormality is present in patients with diabetes and CN, and that thecurrent surgical and adjunct treatment modalities used to manage CNmight not address the true pathologic process. These findings mightalso help in the understanding of the use of diphosphonates for thetreatment of the Charcot foot (38,39). Larger studies are needed toinvestigate the effect of neuropathy and diabetes on inflammatorycytokines and growth factors, as well as the direct effect of fixation,bone grafting, and bone stimulators in diabetic and Charcot bone.

Acknowledgements

The authors acknowledge the capable assistance of Simon Mon-telongo and Lindsay Leissner in assisting with the histologicpreparations.

References

1. Charcot JM. Sur quelques arthropathies qui paraissent d�ependre d’une l�esion ducerveau ou de la mo€elle �epini�ere. Archives de Physiologie Normale et Pathologique1:161–78, 379–400, 1868.

2. Sinha S, Munichoodappa CS, Kozak GP. Neuro-arthropathy (Charcot joints) indiabetes mellitus (clinical study of 101 cases). Medicine (Baltimore) 51:191–210,1972.

3. Armstrong DG, Todd WF, Lavery LA, Harkless LB, Bushman TR. The natural historyof acute Charcot’s arthropathy in a diabetic foot specialty clinic. Diabet Med14:357–363, 1997.

4. Edmonds ME, Clarke MB, Newton S, Barrett J, Watkins PJ. Increased uptake of boneradiopharmaceutical in diabetic neuropathy. Q J Med 57:843–855, 1985.

5. Delano PJ. Pathogenesis of Charcot’s joint. AJR Am J Roentgenol 56:189–200, 1946.6. Zaidi M. Neural surveillance of skeletal homeostasis. Cell Metab 1:219–221, 2005.7. Irie K, Hara-Irie F, Ozawa H, Yajima T. Calcitonin gene-related peptide (CGRP)-

containing nerve fibers in bone tissue and their involvement in bone remodeling.Microsc Res Tech 58:85–90, 2002.

8. Tuominen JT, Impivaara O, Puukka P, Ronnemaa T. Bone mineral density in patientswith type 1 and type 2 diabetes. Diabetes Care 22:1196–1200, 1999.

9. Geusens PP, Landewe RB, Garnero P, Chen D, Dunstan CR, Lems WF, Stinissen P, vander Heijde DM, van der Linden S, Boers M. The ratio of circulating osteoprotegerinto RANKL in early rheumatoid arthritis predicts later joint destruction. ArthritisRheum 54:1772–1777, 2006.

10. Granchi D, Pellacani A, Spina M, Cenni E, Savarino LM, Baldini N, Giunti A. Serumlevels of osteoprotegerin and receptor activator of nuclear factor-kappaB ligandas markers of periprosthetic osteolysis. J Bone Joint Surg Am 88:1501–1509,2006.

11. Vernal R, Chaparro A, Graumann R, Puente J, Valenzuela MA, Gamonal J. Levels ofcytokine receptor activator of nuclear factor kappa B ligand in gingival crevicularfluid in untreated chronic periodontitis patients. J Periodontol 75:1586–1591,2004.

J. La Fontaine et al. / The Journal of Foot & Ankle Surgery 50 (2011) 648–653 653

12. Melton LJ III, Leibson CL, Achenbach SJ, Therneau TM, Khosla S. Fracture risk intype 2 diabetes: update of a population-based study. J Bone Miner Res 23:1334–1342, 2008.

13. Melton LJ III, Riggs BL, Leibson CL, Achenbach SJ, Camp JJ, Bouxsein ML,Atkinson EJ, Robb RA, Khosla S. A bone structural basis for fracture risk in diabetes.J Clin Endocrinol Metab 93:4804–4809, 2008.

14. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM,Recker RR. Bone histomorphometry: standardization of nomenclature, symbols,and units. J Bone Miner Res 2:595–610, 1987.

15. Pinzur MS. Neutral ring fixation for high-risk nonplantigrade Charcot midfootdeformity. Foot Ankle Int 28:961–966, 2007.

16. Fabrin J, Larsen K, Holstein PE. Arthrodesis with external fixation in the unstable ormisaligned Charcot ankle in patients with diabetes mellitus. Int J Low ExtremWounds 6:102–107, 2007.

17. Farber DC, Juliano PJ, Cavanagh PR, Ulbrecht J, Caputo G. Single stage correctionwith external fixation of the ulcerated foot in individuals with Charcot neuro-arthropathy. Foot Ankle Int 23:130–134, 2002.

18. Schwartz AV, Sellmeyer DE, Ensrud KE, Cauley JA, Tabor HK, Schreiner PJ, Jamal SA,Black DM, Cummings SR. Older women with diabetes have an increased risk offracture: a prospective study. J Clin Endocrinol Metab 86:32–38, 2001.

19. Melton LJ III, Leibson CL, Achenbach SJ, Therneau TM, Khosla S. Fracture risk intype 2 diabetes: update of a population-based study. J Bone Miner Res 23:1334–1342, 2008.

20. Melton LJ III, Riggs BL, Leibson CL, Achenbach SJ, Camp JJ, Bouxsein ML,Atkinson EJ, Robb RA, Khosla S. A bone structural basis for fracture risk in diabetes.J Clin Endocrinol Metab 93:4804–4809, 2008.

21. Papa J, Myerson M, Girard P. Salvage, with arthrodesis, in intractable diabeticneuropathic arthropathy of the foot and ankle. J Bone Joint Surg Am 75:1056–1066, 1993.

22. Hanft JR, Goggin JP, Landsman A, Surprenant M. The role of combined magneticfield bone growth stimulation as an adjunct in the treatment of neuroarthropathy/Charcot joint: an expanded pilot study. J Foot Ankle Surg 37:510–515, 1998.

23. Grady JF, O’Connor KJ, Axe TM, Zager EJ, Dennis LM, Brenner LA. Use of electro-stimulation in the treatment of diabetic neuroarthropathy. J Am Podiatr MedAssoc 90:287–294, 2000.

24. Pelton K, Hofer JK, Thordarson DB. Tibiotalocalcaneal arthrodesis using a dynam-ically locked retrograde intramedullary nail. Foot Ankle Int 27:759–763, 2006.

25. Eichenholtz SN. Charcot Joints, Charles C. Thomas, Springfield, IL, 1966.

26. Armstrong DG, Lavery LA. Monitoring healing of acute Charcot’s arthropathy withinfrared dermal thermometry. J Rehabil Res Dev 34:317–321, 1997.

27. Lips P, Courpron P, Meunier PJ. Mean wall thickness of trabecular bone packets inthe human iliac crest: changes with age. Calcif Tissue Res 26:13–17, 1978.

28. Levy JR, Murray E, Manolagas S, Olefsky JM. Demonstration of insulin receptorsand modulation of alkaline phosphatase activity by insulin in rat osteoblastic cells.Endocrinology 119:1786–1792, 1986.

29. Balint E, Szabo P, Marshall CF, Sprague SM. Glucose-induced inhibition of in vitrobone mineralization. Bone 28:21–28, 2001.

30. Machwate M, Zerath E, Holy X, Pastoureau P, Marie PJ. Insulin-like growth factor-Iincreases trabecular bone formation and osteoblastic cell proliferation in unloadedrats. Endocrinology 134:1031–1038, 1994.

31. La Fontaine J, Harkless LB, Sylvia VL, Carnes D, Heim-Hall J, Jude E. Levels ofendothelial nitric oxide synthase and calcitonin gene-related peptide in theCharcot foot: a pilot study. J Foot Ankle Surg 47:424–429, 2008.

32. Petrova NL, Foster AV, Edmonds ME. Calcaneal bone mineral density in patientswith Charcot neuropathic osteoarthropathy: differences between type 1 and type2 diabetes. Diabet Med 22:756–761, 2005.

33. Sinacore DR, Hastings MK, Bohnert KL, Fielder FA, Villareal DT, Blair VP III,Johnson JE. Inflammatory osteolysis in diabetic neuropathic (Charcot) arthropa-thies of the foot. Phys Ther 88:1399–1407, 2008.

34. Sinacore DR, Bohnert KL, Hastings MK, Johnson JE. Mid foot kinetics characterizestructural polymorphism in diabetic foot disease. Clin Biomech (Bristol, Avon)23:653–661, 2008.

35. Frykberg RG, Sanders LJ. The Charcot foot, In The High Risk Foot in Diabetes Mellitus,pp 325–35, edited by R Frykberg, Churchill Livingstone, New York, 1991.

36. Young MJ, Marshall A, Adams JE, Selby PL, Boulton AJ. Osteopenia, neurologicaldysfunction, and the development of Charcot neuroarthropathy. Diabetes Care18:34–38, 1995.

37. Schini-Kerth VB, Fisslthaler B, Busse R. CGRP enhances induction of NO synthase invascular smooth muscle cells via a cAMP-dependent mechanism. Am J Physiol267(6 Pt 2):H2483–H2490, 1994.

38. Naqvi A, Cuchacovich R, Saketkoo L, Espinoza LR. Acute Charcot arthropathysuccessfully treated with pamidronate: long-term follow-up. Am J Med Sci335:145–148, 2008.

39. Anderson JJ, Woelffer KE, Holtzman JJ, Jacobs AM. Bisphosphonates for thetreatment of Charcot neuroarthropathy. J Foot Ankle Surg 43:285–289,2004.