Charcot arthropathy is a progressive, non-infectious, destructive inflammatory process. Charcot arthropathy of the knee (CK) is rare and diagnosis is often delayed, resulting in detrimental outcomes. This scoping review aims to investigate the literature on CK, present the pathognomonic features of CK to aid early diagnosis, and suggest gaps in the literature for future research.
A systematic search of PubMed, EMBASE, Web of Science for literature relevant to CK was performed. Primary studies such as case reports, case series, retrospective studies were included. Review articles and animal studies were excluded.
Of the 513 results, 58 were included in qualitative synthesis. Average time from symptom onset to CK diagnosis was 50.5 months. Eighteen and twenty-one studies included patients who had diabetes mellitus and syphilis, respectively. Twenty-one studies reported pain as a presenting complaint, but the degree of pain didn’t correspond with the level of destruction. Oedema and joint effusion were noticed in 34 studies. Twenty-nine studies reported lower limb hypoesthesia and 17 studies reported decreased tendon reflex. Twenty-eight studies reported initial conservative treatment, often in a knee brace with minimal weight bearing, 9 of which needed subsequent surgical management. Twelve studies utilised arthrodesis, with fracture at the intramedullary nail entry site being the most common complication. Twenty-four studies utilised TKA.
The literature on CK remains sparse, with most publications being case reports. Given that CK dramatically reduces quality of life, increases morbidity of patients, there is need for more literature on evidence-based options for early diagnoses and management.
The online version contains supplementary material available at
Charcot neuroarthropathy (CN) is the chronic, progressive, non-infectious destruction of bone and joints, in patients with peripheral neuropathy, as first described by William Musgrave in 1703 [
Early diagnosis is important, and any patient with peripheral neuropathy, presenting with a red, warm, and erythematous knee joint, should be reviewed for CK. The cause of CK is wide-ranging, with patients often having many co-morbidities. In the past, syphilitic causes were common; however, as antibiotic therapy improved, diabetic causes are becoming more prevalent. Neurological causes, such as spinal canal stenosis and disk herniation, have also been reported.
Traditionally, conservative management with braces and limited weight bearing were common treatment options. In recent years, advancements in technology have allowed specific types of prosthesis to yield acceptable results. The paucity of literature and the dearth of large-scale clinical trials comparing the long-term outcomes following different treatment types could explain the lack of a robust management guideline for CK.
To our knowledge, there has not been any systematic or scoping review of the CK literature. This scoping review aims to systematically examine the literature regarding the presentation, diagnosis, management, and long-term outcomes, and point out gaps in the literature that could be filled with rigorously conducted CK studies that use a standard reporting template that we have proposed.
The manuscript does not contain clinical studies or patient data.
We conduced our review using the method described by Arksey and O’Malley [
This scoping review aimed to answer the primary question: What is currently known about Charcot neuroarthropathy of the knee (CK)? Secondary questions were the following: What is the aetiology of CK? How does CK usually present? What are the pathognomonic features of CK? How is CK managed, and what complications could arise?
In order to balance practicality with comprehensiveness and breadth, only English language studies, and literature formally published in sources such as journals were included. Before formulating our search criteria, an initial review of the literature was performed, to gauge the heterogeneity of this field, avoiding the chance of important studies being missed.
On March 2nd 2021, a systematic search was performed on Embase, Medline, and Web of Science, which were considered comprehensive. The search strategy can be found in Online Resource
After importing the studies into Mendeley reference manager, the in-built deduplication function was used. JZ and VL independently completed the title, abstract, and full-text screening, based on the inclusion and exclusion criteria. This was determined a priori based on the research questions. Agreement between authors was assessed for all studies and generated 78% agreement. A third reviewer (AT) was contacted for unresolvable disagreements.
Studies were included if they were conducted on human subjects (all ages, both sexes) and relevant to CK. Primary studies, such as case reports and retrospective observational studies were included, however those that only mention CK as a passing statement, with no further elaboration on specific outcomes measures were excluded. Review articles were also excluded, since they provide little first-hand quantitative or qualitative data for analysis, and different review articles could report on the same patients, thus duplicating data. For seven older studies, full text could not be found after an extensive search, and were excluded. The inclusion and exclusion criteria can be found in Online Resource
Data from each study were split into 4 different categories. Demographics category includes number of patients in each study, ethnicity, mean age, gender, patient BMI, laterality of knee affected. Presentation category included how the patient presented, time from symptom onset to CK diagnosis, any trigger for symptom onset, physical and neurological exam of the knee, joint aspiration, and histopathological outcomes. Imaging category includes the pathognomonic features of CK seen on radiographs. Treatment category includes a description of different treatment modalities, including complications encountered during follow-up, time to partial weight bear (PWB) and full weight bear (FWB), quantitative changes in knee range of motion and knee scores, and patient-reported outcome measures (PROMs).
Qualitative data were presented under the category to which they belong. Extracted quantitative data was analysed with IBM SPSS Statistics version 27. Statistical analyses focused on descriptive statistics such as mean, median and range.
The aforementioned review process was performed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines for scoping reviews (PRISMA-ScR) [
This scoping review was not prospectively registered in the International Prospective Register of Systematic Reviews PROSPERO.
A total of 513 studies were identified in the original search, After de-duplication, 356 studies were identified for title and abstract screening. 239 studies were excluded, leaving 117 studies for full-text screening. Based on the inclusion and exclusion criteria, 58 articles were included for final analysis. Publication dates ranged from 1954 to 2020, the mean and median year of publication were 2003 and 2011, respectively. Online Resource
The majority (n = 40) were single patient case reports, and 9 case series were included. The remaining 9 studies were retrospective studies investigating a group of CK patients treated at a single institute, the largest of which included 27 patients [
A total of 212 patients with 259 CKs were included, with the average age being 52.3 years, and 45.4% being male patients. 76.9% of all patients reported unilateral CK (33.8% right knee, 43.1% left knee). 23.1% of all patients reported bilateral involvement.
Patient BMI was only reported by 7 studies involving 71 patients, averaging 23.51 kg/m2 [ Aetiology of patients with CK Aetiology Number of studies Number of patients
All case reports mentioned how the patient presented initially, apart from two case series published in 1954 [ Twenty-one studies involving 52 patients described a degree of pain and tenderness as a presenting complaint [ Oedema is commonly reported in CK patients and is often indicative of disease stage. As proposed by Eichenholtz, CK is divided into 3 phases, development, coalescence, reconstruction. 68 patients in 34 studies presented during the development or coalescence phase, and hence presented with swelling, joint effusion, erythema, and knee ballottement [ For patients presenting with these features, osteomyelitis, septic arthritis, pseudogout are important differentials, and 13 studies involving 16 patients performed joint aspiration studies to exclude them [ Valgus deformation was reported in 19 studies involving 114 patients [ Thickened, inflamed, indurated synovium was present in 11 studies involving 34 patients [ CK presents with characteristic neurological findings, with 29 studies involving 62 patients reporting lower limb hypoesthesia with diminished pain and temperature sensation [
Most studies did not perform histopathological analysis, yet those that did report findings pathognomonic for CK, such as hyperplastic inflamed synovium with bone and cartilage detritus [
Table Number of studies presenting each imaging finding X Ray MRI CT [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [
Initial radiographic assessment was almost always conducted by X-ray, after which further scans in the form of CT or MRI are often performed, to get a more detailed visualisation of the joint, surrounding soft tissue, vasculature, and guide preoperative surgical planning.
The tibial-femoral (TF) angle is the angle between the anatomical axis of femur and tibia [
One case series involving two patients used single-photon emission computed tomography (SPECT) to confirm the diagnosis of CK, with increased uptake of tracer associated with the increased vascularity in the femoral condyles [
Despite the low incidence of CK among lower limb pathologies, the high morbidity and rapid progression present difficulties in management and recovery. Current literature includes different treatment options; however, there is no universal treatment algorithm, perhaps due to lack of randomised control trails. The paucity of CK literature, and the difficulty in measuring treatment effectiveness makes it unlikely that any advocated treatment regimen will be standardised by the orthopaedic community in the near future. Furthermore, CK patients often have other co-morbidities such as syringomyelia, diabetes mellitus, which dictates a unique treatment pathway for each patient.
Length of follow-up was reported in 29 studies involving 135 patients [ Twenty-eight studies involving 68 patients chose conservative treatment (6,8,13,18,19,20,21,22,25,26,27,29,30,33,35,37,39,41,42,43,46,48,50,55,56,58,59,62). All used a combination of knee brace and consistent immobilisation. Conservative treatment was used for patients with a wide range of aetiologies, including 50% of studies with diabetic CK patients [ Pharmacological treatment is offered when there is a clear underlying cause. Underlying syphilis is often treated by antibiotics, with penicillin being most commonly used [ Among studies utilising conservative treatment, four involving 12 patients, and three involving one patient each reported time to PWB and FWB, respectively, with the average time being 19.1 weeks and 45.3 weeks, respectively. Four studies involving one patient each reported knee scores, with one case report using AKSS knee and function scores. Conservative management can be satisfactory to the patient [ Twelve studies involving 27 patients [ Among studies utilising arthrodesis, four involving 15 patients, and four involving 7 patients reported time to PWB and FWB, with the average time being 16.7 weeks and 21.7 weeks respectively. One case study used AKSS knee and function scores. Knee score improved from 25 to 85. The site of entry for intramedullary nails can be a fragile point prone to fracture [
Twenty-four studies involving 124 patients opted for TKA [
Among studies using TKA, four involving one patient each, and five involving 11 patients reported time to PWB and FWB, with the average time being 7.5 weeks and 10.4 weeks respectively. The Japan orthopaedic association score [
Two patients in two case reports were treated initially with open reduction internal fixation, allowing fracture fixation away from the articular surface. However, both were complicated by delayed union and fixation failure, after which TKA was performed. There was one example of a failed knee osteotomy, although the disease course was complicated by a proximal tibia Schatzker type IV fracture from a previous traumatic incident [
Due to the operation site, the peroneal nerve is at risk, and two case reports documented foot drop post-operatively due to nerve disturbance [
Only 8 studies involving 45 patients present PROMs, with 3 published in the twentieth century [
CK is a rare condition that has been reported in the literature for many decades. With diverse aetiologies, the literature mostly consists of case reports, with a few small-scale retrospective cohort studies. To our knowledge, there has not been any randomised control trials, due to the ethical implications of not providing personalised treatment for such a complex and debilitating condition.
Diagnoses, particularly in the early stages, revolve around differentiating CN with other aetiologies that also produce erythematous swelling of the joint, such as osteomyelitis or an acute attack of gout. This is clearly important for management strategies, especially in those who go on to develop chronic CN with underlying osteomyelitis. Kucera et al. states that diagnosis should be ‘
Radiography is the preferred modality, although changes are typically delayed and have low sensitivity [
Eichenholtz published a radiological classification in 1966, comprising the following pathognomonic features of joint subluxation and bone debris [ Eichenholtz Classification—Temporal staging of CK based on the pathophysiological progression of the disease Stage Radiographic findings Clinical findings Normal Swelling, erythema, warmth Osteopenia, joint subluxation, dislocation Swelling, erythema, warmth, ligamentous laxity Absorption of debris, sclerosis, fusion of larger fragments Decreased warmth, decreased swelling, decreased erythema Deformity consolidation, fibrous ankyloses, rounding and smoothing of bone fragments Fixed deformity, absence of warmth, swelling, erythema
Conservative treatment used to be the main treatment for CN patients. First proposed by Henderson in 1905, those with joint deformity fared better with bed rest [
Treatments aimed at decreasing the pathological inflammatory processes such as intra-articular corticosteroid injections have been used for Charcot’s arthropathy of the knees, shoulders, and hips. However, pain-free periods after the injection removes the protective barrier against self-injury, with excess activity and microtrauma leading to joint destruction [
Early surgical treatment usually involves arthrodesis; however, this severely reduces quality of life, especially for younger patients. Furthermore, varying cultures around the globe means that some groups such as the Japanese spend more time seated, making knee arthrodesis more debilitating [
Until relatively recently, TKA was contraindicated for knee involvement. The risk of incorrectly aligned prosthesis caused complication concerns, with the most common being periprosthetic fractures [
The scoping review is a novel clinical research method for a rapid and easily-presentable way to map out the literature in a particular field, especially ones that are relatively underexplored and heterogenous, and uncover gaps in the literature where future systematic reviews could be conducted.
We used Arksey and O’Malley’s methodological framework for scoping reviews, which was relatively simple to follow. We covered three major multidisciplinary databases (Medline, Embase, Web of Science), which encompassed the vast majority of literature on Charcot knee. Indeed, we could not find any additional studies in the bibliography of included studies that were not found by our search. However, like systematic reviews, the search was performed on a certain day, mapping out the literature at a certain point in time, and very soon became out of date.
Some may regard scoping reviews as a ‘less rigorous’ systematic review, yet they have their unique differences. Systematic reviews aim to critically analyse a specific sub-section of the literature, whereas scoping reviews provide a broad overview, and categorise the literature by common subject matters, helping researchers efficiently identify where further data analyses can be carried out.
Scoping reviews are not meant to assess literature quality. The balance between breadth and depth is challenging, especially when the overall aim was to map out the current literature. It was not feasible to conduct a comprehensive assessment of literature quality given the large volume of studies identified. Furthermore, the studies included in this review spanned multiple decades, hence the reporting style for many outcome measures were heterogeneous, making it difficult for any rigorous systematic review to be performed.
As scoping reviews increase in popularity, it is important to have a recognised framework to ensure the high quality of reviews. The PRISMA-ScR checklist was thus utilised to ensure all aspects were covered, improving the transparency and quality of this scoping review.
CK is rare, with current literature mostly limited to case reports. Given the significant morbidity involved, it’s paramount that prompt diagnosis is made, preferably with the more sensitive MRI scan during the ‘prodromal’ phase, to initiate early conservative treatment and maintain structural integrity. Leukocyte/marrow scintigraphy should be utilised to rule out any concomitant osteomyelitis.
Care must be taken with treatment, given the lack of consensus. Any management strategy should focus on joint and limb alignment, bone defect reconstruction, and ligament balancing. TKA should not be shunned, with recent literature showing its potential to increase quality of life. Given the wide range of causes of CN, it’s still unclear whether the underlying pathophysiology has an effect on long-term outcome.
A summary of the key findings is given in Table Summary of key findings Topic Key findings Common aetiologies of Charcot Knee (in descending order) Syphilis, Diabetes mellitus, idiopathic, spinal cord injury and syringomyelia Biggest challenge to early diagnosis Long delay from symptom onset to diagnosis due to non-specific nature of symptoms Traumatic injury usually acts as trigger for discovery after imaging Common physical findings Minimal pain, oedema of knee joint, valgus/varus deformity of knee, hypoesthesia, reduced/absent tendon reflex Important differential diagnosis to rule out Osteoarthritis and osteomyelitis When is conservative treatment indicated Diabetic causes when deformity was discovered early A clear underlying cause such as syphilis is present, whereby medication is indicated When is arthrodesis indicated Cases of failed TKA operations, such as periprosthetic infection cases When quality of life is significantly improved by pain reduction and increased mobility, and not diminished by the restriction in range of motion When is TKA indicated End-stage neuropathic arthropathy, especially the coalescence and reconstruction stages Those presenting with joint dislocation prior to surgery Proposed reporting template for future CK studies Part Item # Checklist item A-Demographics 1 Ethnicity 2 Patient Age 3 Patient Gender 4 Patient BMI 5 Laterality of knee affected 6 Smoking status B-Presentation 1 Aetiology 2 Time from symptom onset to CK diagnosis 3 Trigger for onset of symptoms (if any) 4 Presenting complaint 5 Physical exam of CK 6 Neurological exam of affected lower limb 7 Initial pre-treatment knee range of motion + knee scores 8 Joint aspiration results (if appropriate) 9 Any other Charcot joints present C-Imaging 1 Primary diagnostic radiographical methodology (e.g. X-ray, CT, MRI) 2 Radiographical findings 3 Diagnostic scoring system using the Eichenholtz classification D-Treatment 1 Treatment modality and reason describing the choice Detailed description of technique used, implants/devices used, and if a retrospective study, if it was performed by the same/different surgeon If technique is adapted/changed for a particular case, describe why and how 2 Histopathological analysis of any intraoperative specimen taken 3 Post-operative management of the patient, including when PWB and FWB is allowed* 4 Time delays that may influence any time-dependent outcome needs to be described 5 Post-operative knee range of motion + knee scores 6 Any complications encountered during follow-up period, and if so, if revision surgery was needed 7 Total follow-up time* 8 PROMs including EQ-5D-5L, SF-36 *All time-dependent outcome measures must include a clear definition.
Currently, there is a lack of large-scale studies with adequate follow-up comparing treatment types or diagnostic imaging modalities between CK patients with similar baseline characteristics. There may be little appetite in surgeons for randomised control trials due to high complication rates and morbidity for patients, and low patient recruitment. PROMs provide a beneficial overview of lower limb functional recovery, and unfortunately were infrequently reported. Newer studies provide a more holistic picture of the patient’s quality of life post-treatment. Nevertheless, most only report anatomical and medical outcomes, rather than lifestyle factors. No study reported EQ-5D-5L or SF-36 indices. Normally, a scoping review could aid researchers to locate areas of the literature that could be further investigated with systematic reviews and meta-analyses. However we feel that this is premature, given the lack of high-quality large-scale studies in the literature conducted with similar baseline measures, and any systematic review would probably suffer from high heterogeneity.
No services of an external publishing agency were used.
The authors confirm that the data supporting the findings of this study are available within the article [and/or] its supplementary materials.
None.
Below is the link to the electronic supplementary material. Supplementary file1 (PDF 172 KB) Supplementary file2 (PDF 168 KB) Supplementary file3 (PDF 293 KB) Supplementary file4 (PDF 184 KB)
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.