Charcot Neuroarthropathy of the Foot and Ankle

Abstract

Charcot neuroarthropathy (CN) is a rare, progressive, deforming disease of bone and joints, especially affecting the foot and ankle and leading to considerable morbidity. It can also affect other joints such as the wrist, knee, spine and shoulder. This disease, described originally in reference to syphilis, is now one of the most common associates of diabetes mellitus. As the number of diabetics increase, the incidence of CN is bound to rise. Faster initial diagnosis and prompt institution of treatment may help to reduce its sequelae. There should be a low threshold for ordering investigations to assist coming to this diagnosis. No single investigation is the gold standard. Recent studies on pathogenesis and development of newer investigation modalities have helped to clarify the mystery of its pathogenesis and of its diagnosis in the acute phase. Various complementary investigations together allow the correct diagnosis to be made. Osteomyelitis continues to be confused with acute CN. Hybrid positron emission tomography has shown some promise in differentiating these conditions. A multispecialty approach involving diabetologists, orthopaedists and podiatrists should be used to tackle this difficult problem. The aim of this article is to describe current knowledge about CN with particular reference to the status of diagnostic indicators and management options.

Introduction

Charcot neuroarthropathy (CN) is a progressive, deforming disease of the bone and joints, especially affecting the foot and ankle. However, it can affect other joints such as the wrist, knee, spine and shoulder1, 2, 3, 4. Although it is rare, it leads to considerable morbidity. CN has been associated with many conditions including syphilis, chronic alcoholism, leprosy, syringomyelia, toxic exposure, poliomyelitis, spinal cord injury, rheumatoid arthritis, multiple sclerosis, congenital neuropathy, cellulitis, osteomyelitis, synovitis, foot surgery and renal transplantation surgery5. This disease, described originally in reference to syphilis, is now one of the most common associates of diabetes mellitus. According to various authors, the prevalence of this complication of diabetes is 0.1% to 8%4 ^, 6, 7, 8. However, the true prevalence is unknown because of a lack of population‐based studies as well as a consensus on diagnostic criteria. Acute CN has been a diagnostic dilemma for quite some time now. It can be incorrectly diagnosed as gout, deep vein thrombosis, osteomyelitis or cellulitis9. The latter two conditions are more common in diabetic patients. Stuck and colleagues found obesity is a significant risk factor for CN in diabetics10. Stressing the need for early diagnosis, Wukich et al. found in a retrospective study that a delay in diagnosis of up to eight weeks may lead to more rapid progression and an increased complication rate11. Recent studies on pathogenesis and development of newer investigation modalities have helped to clarify the mystery of its pathogenesis and of its diagnosis in the acute phase. However, some confusion persists. The aim of this article is to describe current knowledge about CN with particular reference to the status of diagnostic indicators and management options.

Pathogenesis

The pathogenesis of CN is not completely understood. The neuro‐traumatic theory proposed by Volkman and Virchow in 1886 maintains that repetitive minor trauma to insensate joints leads to the fractures and disintegration of joints seen in CN. The neuro‐vascular theory maintains that autonomic dysfunction increases the blood supply to the limbs, leading to resorption of bones. Stevens and colleagues reported selective loss of small fibers for cold sensation in CN with preservation of those for perception of warmth and light touch12. The role of pro‐inflammatory cytokines in the pathogenesis of CN has been widely researched. Receptor activator of nuclear factor‐κb ligand (RANKL) is expressed in response to activation by cytokines such as tumor necrosis factor‐α and interleukin‐1β. RANKL leads to maturation of osteoclasts by triggering production of nuclear factor‐κβ. In CN, uncontrolled production of these cytokines, RANKL and osteoclasts leads to local osteolysis13. Witzke and colleagues found fewer receptors for advanced glycosylation end‐products (RAGEs) in diabetic patients with CN14. RAGE has been implicated in increasing activation of RANKL, which in turn leads to bone resorption through osteoclastogenesis as discussed earlier. In a recent study, Chantelau and colleagues found a high threshold for cutaneous pressure pain perception in Charcot feet. They further suggested that these patients fail to perceive inflammatory pain caused by trauma15.

Classification

The most widely recognized classification system for CN is the Eichenholtz system16. He described three stages of progression of CN, namely fragmentation\dissolution, coalescence and reconstruction. In stage 1 there is localized warmth, redness and swelling. X‐ray films show bony fragmentation, debris, fractures and subluxation/dislocation. In stage 2 there is reduction of warmth and swelling clinically and absorption of bony debris, bony/joint fusion and sclerosis on X‐ray films. In stage 3, there is absence of warmth, swelling and redness. X‐ray films in this stage show decreased sclerosis, osteophyte formation, joint arthrosis and deformity. Shibata et al. modified this classification, adding a stage 0 in which there is localized warmth, redness and swelling but no visible abnormal radiological findings17. Anatomical classifications based on pattern of destruction have been described by Sanders and Frykberg18 and Brodsky19. Sanders and Frykberg described five patterns of joint destruction. In Pattern I, metatarsophalangeal and interphalangeal joints are involved; in Pattern II Lisfranc's joint (tarsometatarsal); in Pattern III Chopart's (talonavicular, calcaneocuboid) and naviculocuneiform joints; in Pattern IV ankle and subtalar joints and in Pattern V the calcaneum18. According to this classification, Patterns II and III are associated with the greatest complication rates. Brodsky described four types of disease pattern based on frequency of joints affected, the commonest involvement being of the tarsometatarsal joint19. Type 1 includes Lisfranc's joint (tarsometatarsal); Type 2 Chopart's joint and/or the subtalar joint; Type 3A the ankle joint (Type 3B the calcaneum)19. Rogers and Bevilacqua proposed a classification system based on location of foot involvement and associated complications18 (Fig. 1). They also postulated a high risk of amputation with hindfoot/ankle involvement as well as in the presence of osteomyelitis in CN.

Figure 1.

Rogers and Bevilacqua classification. As we move to the right or down, the risk of amputation increases. Reproduced from reference 18 with permission from Elsevier (Philadelphia, PA, USA).

Clinical Presentation

Peripheral neuropathy is a prerequisite for CN21 ^, 22. CN may present as an acute or chronic condition depending on the natural course of the disease23 ^, 24, which is usually self‐limiting25. Patients usually present with redness, swelling and pain. Although any part of the foot and ankle may be involved, the mid‐foot is most frequently affected13. Although the affected limb is insensate, pain and swelling are the usual complaints by patients seeking medical consultation22 ^, 25. The vascularity of the limb is usually preserved and the peripheral pulses may be bounding. However, swelling or edema may make their palpation difficult13. Radiographs of the foot often reveal bony erosion, fracture and subluxation or dislocation of multiple joints. Over time, the redness and warmth decrease and deformities develop. Radiologically, the bones show osteosclerotic features and the fractures unite. A typical deformity associated with this condition is the rocker‐bottom foot, which is caused by collapse of the longitudinal arch of the foot. This, along with instability of the ankle joint, is reportedly the major cause of morbidity in CN8. It leads to a high pressure area at the base of the foot which in turn predisposes to ulceration and subsequent infection26 (Fig. 2). Recently, Jones et al. described three cases of acute CN with unusual presentations; they had primary bony resorption without subluxation and dislocation27. The most problematic differential diagnosis of acute CN is osteomyelitis. Osteomyelitis and CN are not mutually exclusive and can occur simultaneously. To add to the confusion is the lack of inflammatory response seen in diabetic patients because of immunopathy. Any associated ulcer should be thoroughly probed. Osteomyelitis usually spreads through a contiguous break in the skin. Presence of an ulcer tilts the diagnosis in favor of osteomyelitis.

Figure 2.

Rocker‐bottom deformity in Charcot neuroarthropathy with ulceration at the apex of the deformity.

Diagnosis

The diagnosis is frequently delayed9 ^, 28. The reason for this delay is the rarity of this condition and frequent misdiagnosis in the acute stage. Furthermore, there are no clear‐cut set criteria for its diagnosis. Misdiagnosis can be avoided only by having a high index of suspicion, taking a thorough history, performing a careful clinical examination and requesting relevant investigations29. Some authors maintain that there is always a history of trauma that is frequently not elicited20 ^, 29.

Physical Examination

Neurological Testing

A variety of neuropathies are associated with diabetes. These include sensorimotor neuropathy, distal symmetric polyneuropathy, diabetic mononeuropathy, focal neuropathy, mononeuritis multiplex, diabetic amyotrophy and diabetic autonomic neuropathy30. Unger and Cole stated that “Peripheral neuropathy is the most common type of neuropathy seen in clinical practice, and causes either pain or loss of feeling in the toes, feet, legs, hands, and arms”30. It is necessary to examine for loss of pressure and vibration sense using Semmes‐Weinstein monofilament and a 128 hz tuning fork, respectively. Autonomic neuropathy is diagnosed clinically by measuring heart rate variability (HRV) with deep breathing or orthostatic blood pressure20. Decrease or absence of HRV is the most consistent and earliest indicator of autonomic neuropathy in the diabetic patient30.

Temperature of the Foot

In the acute stage of CN, the foot is warm. A temperature difference between the feet of 2 °C is considered significant. This can be measured using an infrared thermometer31 ^, 32. Furthermore a temperature difference of <2 °C between the limbs is used as a determinant of the end of the acute phase33 ^, 34. One study found no relapse when a foot skin temperature difference of <2 °C was used as a clinical marker for immobilization withdrawal35. In a recent study, Najafi et al. used a thermal imager to study variability of plantar thermal response caused by walking in Charcot and non‐Charcot feet36. They found that after 50 steps there was a significant increase in the plantar temperature in CN patients. However in the later stages of CN, this temperature differential may not be found31.

Medical Imaging

X‐Ray Films

Plain weight‐bearing X‐ray films of the foot and ankle are a readily available and relatively cheap mode of investigation. However, they are not of much help during the acute phase (“stage 0” as described by Shibata and colleagues)17. Rajbhandari et al. summarized the findings on X‐ray films as “The five D's: joint distension, dislocation, debris, disorganisation and increased density”22. According to the Eichenholtz classification, which is based on radiographic changes, stage 1 shows fragmentation of bone, bony proliferation and joint destruction/dislocation; Stage 2 resorption of bone debris, fusion of bony fragments and sclerosis of ends of bone; and Stage 3 sclerosis decreases and joint arthrosis and osteophytes become evident16.

Computerized Tomography and Magnetic Resonance Imaging

Computerized tomography is more valuable for surgical reconstruction and preoperative planning than for diagnosing CN9. CT is useful in the early stages of spinal neuroarthropathy37; however, the role of CT in diagnosis of the Charcot foot is still not validated.

Magnetic resonance imaging is becoming a popular investigation for a variety of foot problems. In chronic CN, MRI shows low signal intensity in subchondral bone on both T₁ and T₂ weighted images; this correlates with sclerosis on radiographs. In contrast, in osteomyelitis there is low signal intensity on T₁‐weighted and high signal intensity on T₂‐weighted images38. However, differentiating acute Charcot foot from osteomyelitis remains a challenge13. The very similar clinical presentation of these two entities causes confusion in the diagnosis and overlap of the MRI findings adds to this confusion. However, there are some clues that can help to clinch the diagnosis. On MRI, a Charcot foot will show localized juxta‐articular edema, whereas in osteomyelitis the edema is more on one side of the joint and is not confined to the juxta‐articular area38. Clinically, osteomyelitis affects a single bone in the forefoot or hindfoot, whereas CN affects many bones, commonly in the midfoot9. In their study of MRI findings and clinical symptoms in the early stage of CN, Schlossbauer et al found a strong correlation between MRI findings and clinical assessment39. They also proposed using MRI to monitor pressure relieving treatment.

In their retrospective study of MRI findings of foot and ankle in neuropathic joint disease and degenerative arthritis, Halstead et al. proposed that the origin of neuropathic arthropathy is ligamentous and osteogenic rather than articular40. Zampa et al. measured contrast medium (gadolinium) uptake on dynamic MRIs in 40 patients with acute CN and found this to be a reliable means of monitoring the treatment outcome in the acute stage41. However, differentiating early Sudeck disease from the acute stage of CN is still a grey area42. In diabetic neuropathic feet, Thorning et al. found areas of increased bone marrow edema in the midfoot and hindfoot, remote from the area shown to be affected by T₂ weighted MRI43. They reported the prevalence of this phenomenon to be 30%. However, the clinical significance of these lesions is not known.

Positron Emission Tomography

Positron emission tomography (PET) has been recently advocated for differentiating between early stage CN and osteomyelitis44, 45, 46. PET has a distinct advantage over MRI in patients with metal implants. Addition of fluorodeoxyglucose (FDG) further adds to the diagnostic capability of PET. FDG enters the cells and is phosphorylated by hexokinase into fluorodeoxyglucose‐6‐phosphate, which cannot be metabolized further. Hence it starts accumulating inside activated white blood cells at the site of infection and inflammation47. There is some controversy regarding FDG uptake in the presence of hyperglycemia in diabetes. Keidar et al. found no relationship between abnormal FDG uptake and glycemia48. Zhuang et al. found PET–FDG to be 100% sensitive and 87.5% specific for excluding chronic osteomyelitis49. Basu et al. found FDG–PET to have high negative predictive value in ruling out infection in diabetic patients with foot ulcers45. Höpfner et al. recommended PET over radiography and MRI for preoperative evaluation of CN patients44. Addition of CT/MRI to the scintigraphic images of PET (hybrid PET) further enhances the specificity of PET45. However, the high cost and limited availability may limit its use.

Other Investigations

Bone Scintigraphy

Bone scintigraphy is not specific enough to diagnose CN22. However, it can detect changes in patients with CN much earlier than radiographic techniques. The commonly used technetium 99 m methylene diphosphonate (Tc‐99 m MDP) bone scan is positive in all three phases. There is increased bone turnover in both osteomyelitis and Charcot neuroarthropathy. Combining the Tc‐99 m MDP bone scan with indium In‐111–labelled leukocyte scintigraphy increase the sensitivity and specificity50 ^, 51. Bem and colleagues proposed that quantitative bone scan parameters could be used as an adjunct for treatment monitoring purpose52. They found significant correlations between quantitative bone scan parameters and bone turnover markers.

Bone Turnover Markers

In Charcot neuroarthropathy there is increased bone turnover. Some studies have found the carboxy‐terminal telopeptide region of type 1 collagen53 and bone‐specific alkaline phosphatase useful markers of bone turnover in patients with CN54. These markers are often used as indicators of disease activity and to measure the response to therapeutic agents in various studies. However, because they are not specific for CN, they are not very useful for its diagnosis.

In a prospective study measuring calcaneal bone mineral density (BMD) in CN patients, Petrova and Edmonds found a significant decrease in calcaneal BMD during the initial three months of CN55. However, Witzke and colleagues found no statistically significant difference in calcaneal BMD between normal, diabetic and Charcot feet14. They found calcaneal bone stiffness to be a better indicator of bone quality in CN. La Fontaine and colleagues performed an observational study of cellular characteristics and bone trabeculae in diabetic feet with and without CN and noted inflammatory myxoid tissue and disorganised trabeculae in the CN group56.

In summary, a wide range of investigations ranging from X‐ray films to PET scan is available for the diagnosis of CN. X‐ray films, although relatively inexpensive, are not useful in the acute stage of CN. PET scans have a distinct advantage over MRI in patients with implants but are expensive and not readily available.

Management

The treatment of CN depends on the stage during which the disease is diagnosed. However other important factors such as the presence of infection, severity of the deformities and location of involvement strongly influence the treatment regime. Furthermore, any other patient comorbidities affect the overall prognosis.

Offloading

Historically, early off‐loading of the affected foot has been the essential treatment measure for acute CN. The aim of treatment is to immobilize the affected foot until inflammation has subsided and the fractures have healed. A Total Contact Cast (TCC) for offloading the affected foot during the acute phase is currently the most favored therapy23, 24, 25. The TCC assists even distribution of forces across the plantar surface of the foot, preventing further destruction of involved bones and joints and eventually limiting the degree of deformity. Generally, non‐weight‐bearing with a TCC is continued until swelling and hyperemia have resolved and foot skin temperature differences have decreased. A TCC is usually continued for 8–12 weeks. To avoid cast complications, frequent cast changing is recommended. Cast changing allows for inspection of the skin and adjustment for changes in foot size with resolution of edema29. Pinzur et al. recommended weight‐bearing with a TCC for the initial 6–8 weeks with biweekly cast changing for acute CN57. Based on a retrospective study, De Souza also recommended a weight‐bearing TCC for the acute stage of CN58. However, Clohisy and Thompson highlighted the risk of fracture due to increased stress on the contralateral weight‐bearing limb59. The location of the arthropathy also determines how long the TCC is indicated. Sinacore stated that, when a TCC is used, forefoot arthropathy heals faster than that of ankle, midfoot and hindfoot60.

Armstrong et al. conducted a randomized control trial comparing efficacy of a TCC, removable cast walker and half‐shoe in patients with noninfected, nonischemic diabetic plantar foot ulcers. They found TCC to be the most effective modality61. Mueller et al. also found TCC superior to traditional dressings for treatment of plantar ulcers on the insensate diabetic foot62. However, use of a TCC requires specially trained personnel and entails high costs due to frequent changing, thus limiting its utility18 ^, 63. Alternatively, removable cast walkers such as Aircast walkers have been used. These walkers allow for surveillance of skin and dressing changes. However, they are not suitable for non‐compliant patients nor those with gross deformity. In a retrospective study, Verity et al. found prefabricated removal cast walker to be more economical than TCCs in the long run64. They further stated that these removable walkers are light‐weight and conform to the shape of the limb better. In a small case series using a vacuum stabilization below‐knee boot in non‐compliant patients with Eichenholtz stage 1 CN, Cook and Cook found no severe deformity despite patients full weight‐bearing65.

After the acute stage has subsided, the patient is advised to wear an ankle‐foot orthosis or a custom‐made shoe59. Traditionally Charcot Restraint Orthotic Walkers have been used at this stage. Again, their application requires expertise. With any form of offloading further support in the form of crutches and wheelchair are also advocated66.

A temperature difference of <2 °C between the affected and non‐affected foot is a prerequisite for cast removal and shifting to custom footwear. Because the bones are fragile and joints usually stiff after the initial immobilization, gradual restoration to full weight‐bearing is advised, starting with partial weight‐bearing. Vigorous mobilization can cause relapse of the disease process and is therefore to be avoided. In one series, the recurrence rate of CN in the ipsilateral foot was 7.1 percent; however no correlation with the initial location of the primary lesion was found67. Custom‐made footwear is used after cast removal. The goal is to achieve a plantigrade biomechanically stable foot, which helps the patient ambulate with minimal difficulty. The optimal type of the footwear depends on the deformity of the foot. Patients with severe deformity and instability are better managed with a solid ankle foot orthosis. Long term continued follow‐up and self‐awareness on the part of patients with CN is advised to prevent future foot problems such as relapse, ulceration and involvement of the contralateral foot.

Pharmacological Therapy

Although bisphosphonates have shown some promise in the acute stage of CN, the evidence for them is inconclusive. They act by inhibiting osteoclast‐mediated bone resorption. Their antiresorptive effect on bone is mediated by inhibition of hydroxyapatite breakdown and osteoclast apoptosis68. Pamidronate (oral and intravenous), alendronate and zolendronic acid have been assessed in some randomized control studies (RCT)69, 70, 71, 72. One study found intranasal calcitonin to be effective in arresting excessive bone turnover in patients with acute CN73. Further larger clinical trials are required to establish the benefits of the pharmacological agents recommended for CN.

Surgical Management

Surgery for Charcot foot is replete with many technical challenges and uncertain outcomes. Giovinco et al. stressed the need for thorough preoperative planning before performing these complex procedures74. Based on a pilot study, they reported that preoperative 3‐D templating was useful for refinement of surgical technique. Lack of randomized control studies (due to the rarity of the disease) and the diversity of joints affected by this disease preclude recommendation of any one specific surgical technique75. Surgery is usually reserved for patients in whom conservative treatment has failed or for treating complications like deformity, joint instability, infection and recurrent ulceration. The sequels of CN increase the morbidity and mortality of affected patients. The aim of surgery is to stabilize and align the foot/ankle and make it amenable to wearing appropriate braces or footwear. The procedures used for treatment of Charcot feet are realignment and arthrodesis, exostectomies and amputation. For managing associated ulcers and infection, debridement is performed with or without the use of antibiotics. Some centers advocate early surgical intervention to avoid the potential high cost of, and complications associated with, prolonged immobilization76 ^, 77. Simon et al. performed arthrodesis of the tarsal‐metatarsal region in 14 patients with Eichenholtz stage‐1 Charcot feet and reported 100% success rate without any postoperative complications78. Both external and implantable electric bone stimulators have been recommended to enhance bone healing after arthrodesis. They stimulate production of growth factors like bone morphogenetic protein, transforming growth factors and insulin like growth factor 279. Lau et al. advised caution in their use for Charcot feet79. A variety of surgical fixation techniques have been described but the jury is still out on the optimal technique to use.

Exostectomy

Strictly speaking, exostectomy means resection of an exostosis. True exostoses do not occur in CN; however, bony prominences caused by joint subluxation or dislocation do. These cause high plantar pressure areas in insensate limbs even during normal walking. Furthermore, they hinder the healing of ulcers or cause recurrent ulceration. According to the Brodsky classification, Type‐1(midfoot) involvement commonly leads to formation of these problematic bony prominences19. In patients with deformity, exostectomy is a viable option provided the midfoot deformity is stable; otherwise there is a risk of worsening the deformity80, 81, 82. Garapati and Weinfeld recommended an operative approach via separate incisions on the non‐weight‐bearing surface of the foot82. Further, they advocated going through the ulcer for centrally located prominences and primary closure of any small and superficial ulcers during exostectomy. Many authors have recommended Achilles tendon lengthening in cases of recurrent ulceration and equinus deformity83 ^, 84. Theoretically, Achilles lengthening decreases stress across midfoot joints by increasing the calcaneal inclination angle. Mueller et al. performed a randomized controlled trial to compare the recurrence rate of ulceration in patients with neuropathic plantar ulcers treated with Achilles tendon lengthening plus total contact casting and total contact casting alone85. They reported a 15% percent ulcer recurrence in the Achilles tendon lengthening with total contact casting group as compared to 59% in the group with total contact casting alone at seven months follow‐up. Further, the group with total contact casting alone developed reulceration earlier than did the other group. Exostectomy is followed by appropriate bracing or a TCC to facilitate soft tissue and wound healing.

Arthrodesis

Arthrodesis is performed in patients with Eichenholtz stage 2 and 3 CN. Various techniques and implants have been described for this procedure in CN. They depend on the location of deformity and the surgeon's preference25. Because the midfoot is the most commonly affected area, it is frequently arthrodesed. However, arthrodesis should not be considered a substitute for bracing or appropriate footwear80. In Charcot feet with ulcers, the procedure can be staged with initial debridement and an external fixator followed by internal fixator and bone grafting. The choice of implant for arthrodesis depends on the surgeon's preference as none has been proven superior to the others. Screws, plates, intramedullary devices and external fixators have all been used for arthrodesis. Adequate autologous bone graft is used after preparing the joint surfaces. In a retrospective study, Grant et al. highlighted the stability of a construct using large stainless steel intramedullary screws as beams for lateral and medial columns along with arthrodesis of the subtalar joint for reconstruction of Charcot feet86. Sammarco used the term “superconstructs” for a combination of plantar plating, locking plate technology and axial screw fixation during reconstruction of midfoot deformities in Charcot feet87. These superconstructs entail extending the fusion beyond the affected zone by inclusion of non‐affected joints, reduction of deformity without causing excessive soft tissue tension by shortening the extremity, using the strongest devices that can be tolerated by the soft tissue envelope and applying the device in the optimal position to increase mechanical function. The complications of arthrodesis in patients with CN include non‐union, infection and delayed wound healing. Lately, external fixators have been advocated by many authors88, 89, 90. Their main advantages are that they permit monitoring of soft tissue healing and avoidance of more invasive surgery in patients with infection25. However, their main drawback is pin tract infection. Lamm et al. recommended a staged minimally invasive technique wherein initially an external fixator is applied for gradual distraction followed by arthrodesis using percutaneous internal fixation91. They also recommended percutaneous Achilles tendon lengthening along with external fixator application. Autografts required for arthrodeses can be harvested from iliac crest, tibia or fibula. Hollawell recommended the use of the allogenic cellular bone matrix “Osteocel Plus” (NuVasive, San Diego, CA, USA) as an autograft alternative in cases of high risk ankle and foot arthrodesis92. The author found 100% radiological fusion at the end of 6 months. However, only three patients with CN were included in the study.

Amputation

Amputation of the limb is a last resort in cases of failed previous surgery. In a meta‐analysis, Sohn et al. found that the overall risk of amputation in patients with CN is sevenfold less than in patients with diabetic foot ulcers93. They also reported a many fold increase in risk in the presence of ulceration. Moreover, if the infection or ulcer extends into the midfoot or hindfoot, major amputation is usually required94. Altindas and colleagues reported a two‐staged Boyd operation technique for late stage Charcot feet. Most of the patients in their study had hindfoot involvement94. The first stage includes talectomy, joint surface preparation (for calcaneotibial arthrodesis), and fixation of the calcaneum to the tibia with Kirschner wires after fashioning the joint surface and pushing the calcaneum forward. Second stage closure of the flaps is performed 2–3 weeks later. They reported no complications with a mean follow‐up of 2.1 years.

In summary, offloading is the treatment of choice for acute stage CN of the foot and ankle. The role of surgery is mainly in correction of deformities and management of complications. Pharmacological therapy is not yet of proven efficacy; further studies are required to determine its role.

Conclusion

While rare, CN is a deforming disease associated with high morbidity. As the number of diabetics increase, the incidence of CN is bound to rise. More rapid initial diagnosis and prompt institution of treatment may help reduce its sequelae. There should be a low threshold for ordering investigations to assist diagnosis of CN. It is better to over‐diagnose CN than to miss this diagnosis. No single investigation is the gold standard. Various investigations complement each other to arrive at the correct diagnosis. Osteomyelitis continues to be a confounding factor in the diagnosis of acute CN. Development of new techniques, such as hybrid PET, has shown some promise in this regard. A multispecialty approach involving diabetologists, orthopaedists and podiatrists is recommended to tackle this difficult problem.