Charcot arthropathy of the diabetic foot in a sub-Saharan tertiary hospital: a cross-sectional study

Diabetes mellitus is the most prevalent non-communicable chronic disease worldwide [1], with Africa having the highest prevalence of undiagnosed diabetes [2]. The prevalence of diabetes mellitus in Uganda is on the increase and it is becoming a major public health problem that is being compounded by limited access to quality health care [3‐6]. Population-based studies in Uganda have reported diabetes prevalence rates ranging from 1.4 to 7.4% [3, 7]. Diabetes mellitus predisposes patients to foot complications that range from ulceration to gangrene that is associated with significant long-term disability and premature mortality [1]. Complication rates for foot complications in Africa vary from 4 to 19%; peripheral neuropathy, 4–84.4%; peripheral vascular disease, 2.9–78.7%; gangrene, 0.6–69%; and amputation rates of 0.3–45% [1]. However, very few studies from Africa have specifically addressed Charcot foot arthropathy (CA).

Charcot foot arthropathy (CA), is a condition affecting the bones and joints of the foot in patients with peripheral neuropathy of various origins, characterized in the earliest phase by non-infective inflammation of bones and joints of the foot [8‐10] and in the later phases by progressive destruction of bones and joints in the weight-bearing neuropathic foot leading to acute fractures, dislocations and joint destruction [11, 12]. There are many reported aetiologies of CA, however, nowadays diabetes mellitus has become the leading cause [8, 12‐15].

Globally, several population-based studies have reported an estimated CA prevalence ranging from 0.08% of the general diabetes patient population to 13% of patients presenting at diabetic foot clinics [12, 14, 16, 17] though the exact prevalence of CA is unknown. This might partly be attributed to a high incidence of mistaken or delayed initial diagnosis [12, 14, 16]. Charcot foot arthropathy is associated with several complications [18, 19] yet there is a paucity of literature on CA from sub-Saharan Africa. We came across only one study from East Africa that was conducted in a Kenyan district hospital and reported foot complication rates as high as 29.2% in patients attending a diabetic clinic; with 95% of these having diabetic ulcers and 5% had CA or cellulitis [20]. To the best of the author’s knowledge, no study related to CA in Uganda has yet been published.

The acute phase of CA is often misdiagnosed and can lead to permanent foot deformity, ulceration thus increasing the risk of lower extremity amputation [8, 12, 21, 22]. Early diagnosis and management of acute CA are therefore imperative to avoid progression towards permanent foot deformity and its associated complications [12, 13]. In patients with diabetes and an established CA, patients can present with foot ulceration secondary to foot deformity and therapeutic efforts tend to be focused on ulcers management rather than arresting the disease process [23].

There are 3 types of Charcot foot classification: clinical, anatomical and radiological [11, 12, 17, 24‐26]. In clinical practice, Charcot foot can be classified into the acute and chronic stage [25]:

In the acute stage, the foot is remarkably red, warm and swollen; the midfoot is most affected and pain is not a prominent feature [25]. In the Chronic stage signs of local inflammation progressively recede [25, 27]; stable deformities may develop in this stage most commonly a rocker bottom deformity [25, 27]. The most frequently used anatomical classification was proposed by Sanders and Frykberg [11, 12, 17, 24‐26]. It describes five different patterns of destruction. The radiological classification is based on the natural history of the disease and is divided into four stages based on Eichenholtz’s work in 1966, and later modified in 1990 by Shibata et al. [17, 24‐28].

Potential risk factors associated with CA include age, duration of diabetes, body mass index, history of an instigating event such as foot trauma or foot surgery and peripheral neuropathy [11, 12, 14, 29, 30]. In patients with diabetes, CA typically presents during the fifth or sixth decade of life [31]. A longstanding history, at least a decade with diabetes is usual [32‐34]. Most commonly, at the time of onset patients with both Type 1 and Type 2 diabetes have been diagnosed for a period ≥10 years, however, for type 2 other studies found a diabetes duration of 5 to 9 years at the time of CA diagnosis [11, 12, 14, 29, 30]. The mean body mass index of Charcot patients, according to Pakarinen et al. [32] was 32.9 kg/m² and 34.5 kg/m² in men and women, respectively. A history of an instigating event (foot trauma, foot surgery) preceding the onset of Charcot's foot has been reported from 22 to 73% of the time [35]. Peripheral neuropathy is associated with all disorders that produce neuroarthropathy and is believed to be the prerequisite for the development of Charcot arthropathy, however not all patients with diabetes and peripheral neuropathy will develop Charcot foot arthropathy [8, 14, 17, 29, 30].

The prevalence of diabetes mellitus is currently increasing worldwide including Uganda, becoming, therefore, a major public health problem [3‐5]. If not managed in time, CA leads to permanent foot deformities, which will progress to foot ulcers, thus increasing the risk for lower extremity amputation. Charcot foot arthropathy is a contributing factor in the 15 to 40-fold increase in the risk of lower-extremity amputations in the population with diabetes [36]. Screening for diabetic complications in Uganda urban diabetic clinics is suboptimal [37]. Among patients with diabetes, peripheral neuropathy, the main risk factor of CA estimated to have a prevalence of 9 to 32% among patients with diabetes [14, 38, 39], was found to be 46.4% in Uganda at the time of diagnosis [5].

Despite the facts mentioned above, there was no baseline data in terms of extent and patterns of that foot disease in these clinics, there were no established protocol for its early diagnosis and management. Charcot foot arthropathy is overlooked and misdiagnosed; hence the increased morbidity increased risk of foot amputations and high economic burden among these patients.

Therefore, this study aimed at determining the extent of CA, the radiological patterns of CA and patient’s factors associated with CA among adult patients with longstanding diabetes attending MNRH.

One hundred adult patients with longstanding diabetes were recruited of which the majority were female (79/100, 79%) and a half was above 50 years of age. Sixty–four patients (64%) were either overweight or obese. Forty-three patients (43%) had suffered from diabetes mellitus for more than 10 years. The mean diabetes duration, mean body mass index and mean HbA1c were respectively 12.8 years (± 5.9 SD), 27.6 (± 5.2 SD) and 6.7% (± 6.7 SD). Patient demographic characteristics are summarized in Table 1.

This study aimed to determine the extent, the radiological patterns and associated factors of CA among adult patients with longstanding diabetes mellitus attending an outpatient’s clinic in an African setting.

The majority of participants with CA were in their sixth decade of life. This is in concordance with other studies where CA presents during the fifth and sixth decade [14, 29, 30]. Age is an important risk factor for diabetic neuropathy [46] which is very crucial in the pathogenesis of CA [8, 14, 17, 29, 30].

The proportion of CA among patients with longstanding diabetes in this study was 12%. Several authors have reported varying prevalence rates for CA in different populations ranging from 0.08% to 13 [12, 14, 17, 23, 26]. The prevalence is low when considering the general population with diabetes and increases when considering a specific group among patients with diabetes such as those consulting a foot clinic for a foot problem [14, 17]. Charcot foot arthropathy is not a frequent complication in the general population with diabetes; however, when present it affects the quality of life, threatens the lower limb for amputation and its management is challenging [23, 26]. Routine examination of the feet is very crucial for patients with diabetes because it can avert catastrophic complications. One-third of the participants in this study were in the acute phase of CA. Diagnosis of CA in the acute phase is difficult because clinical features are non-specific and foot radiographs are often normal. Making a diagnosis at this stage reduces morbidity and facilitates proper management [8, 12, 17, 23]. Clinicians should not only have a clear understanding of diabetic foot disease but should also have a high index of suspicion.

In this study, there was an equal distribution of CA in the fore- and mid-foot. This pattern is similar to what has been reported by other authors [14, 23]. In addition to diabetes, the pattern of disease is influenced by other biomechanical factors that act on the foot. For example, a high BMI predisposes the middle arch to collapse culminating in a rocker-bottom foot; tightness of the triceps surae complex leads to abnormal plantar pressure distribution in the forefoot during walking thus exposing the forefoot bones to more strain [8, 17, 23]. Hence, addressing the biomechanics of the foot is also important in the management of these patients [8, 17, 23, 47‐49].

Most of the lesions in this study were at the initial stages of the natural history of CA. The mainstay in the treatment of the initial stages (inflammatory and developmental) is non-operative management by offloading the limb in a total contact cast [17, 23, 50]. If these stages are treated judiciously, the achievement of a stable foot without surgery or skin breakdown is possible [17]. This shows the need for quick management intervention in the acute phase before latter stages where bone deformities are fixed and permanent, predisposing the foot to ulceration, infection and high risk of amputation [17]. Clinical symptoms in the inflammatory stage may precede radiographic changes by up to 1 year, hence treatment in this stage is critical in preventing the disease from progressing in further stages [14, 51].

The vast majority of patients in this study had abnormal weight-bearing foot radiographs with no obvious history of trauma. An abnormal weight-bearing foot radiograph in a patient with longstanding diabetes and peripheral neuropathy should raise suspicion of CA. When reading the film one must take time to confirm if the lesions seen are features of CA. Therefore clinicians should have a high index of suspicion and ought to routinely assess patient’s feet for CA.

Charcot foot arthropathy is a relatively common complication among patients with long-standing diabetes mellitus in our setting. Several cases present in the acute phase when there are no obvious pathological radiographic changes and diagnosis is relatively difficult. This notwithstanding, the majority of patients with a clinical diagnosis of CA may present with abnormal weight-bearing foot radiographs with the involvement of the mid-foot and forefoot. Therefore clinicians should have a high index of suspicion to enable early diagnosis and also prevent potentially disabling complications.