Evaluation of Risk Factors for Foot Ulceration in Individuals with Chronic Kidney Disease

Foot ulcerations are a worldwide public health problem and cause significant mortality and morbidity.^1,2 Ulcers are often infected, limiting mobility of the extremity, and may result in amputation.³ The incidence of foot ulceration increases in the presence of diseases such as DM and PAD.^2,4 CKD, ESRD, and HD have been shown to be among the independent risk factors for the development and course of foot ulcers, which are often caused by the consequences of DM, such as neuropathy.^5,6 Although the main determinants of injury and loss of lower extremity in the patient group with CKD, ESRD, and HD—whether or not accompanied by diabetes—are not fully understood, it has long been known that renal failure affects wound healing.^7-10 The literature indicates that foot ulceration, the prevalence of which varies in patients receiving HD, is associated with many conditions, such as uremia, uremic neuropathy, anemia, edema in the lower extremities, hypoalbuminemia, susceptibility to infection, and decreased pedal skin perfusion during HD.^11-14

Chronic foot ulcers lead to other serious lower extremity complications, such as deep infections that require hospitalization, extremity loss, and foot-related mortality; management of such complications is becoming increasingly challenging and is a significant burden on public and private payers, as are diabetes-related costs and resource use.¹⁵ Knowing the prevalence, causes, and risk factors of chronic foot ulceration is quite important in improving nursing care and outcomes in patients with CKD who are or are not receiving HD. Such complications are usually preventable when modifiable risk factors are identified and managed early.^16,17 

With the goal of raising awareness about the identification of foot ulceration in patients with CKD who are or are not receiving HD in Turkey, the authors of the current study compared the incidence and risk factors of foot ulceration in patients with stage 4 or 5 CKD who were or were not receiving HD and who did or did not have DM. 

Study design
The researchers obtained all necessary written permissions before starting the study. The Scientific Research Ethics Committee approved the study protocol on April 5, 2016 (number 24237859-266). After all patients included in the study were informed about the purpose of the study, their verbal and written consent was obtained indicating their acceptance. The study was conducted in accordance with the Declaration of Helsinki.

This cross-sectional and comparative study was conducted between January 2016 and June 2018 in Rize hospitals (Recep Tayyip Erdoğan University Training and Research Hospital nephrology outpatient clinic and HD unit and State Hospital HD unit) and a private dialysis center in Rize, Turkey. A total of 160 patients with or without DM with stage 4 or 5 CKD and who were or were not receiving HD were included in the study. The patients were divided into 4 groups according to whether or not they were receiving HD treatment and whether or not they had DM. The lowest possible sample size was calculated using the open-source statistical software OpenEpi (version 3). The frequency was estimated as 5% in the group receiving HD and 30% in the group not receiving HD, with 8% power, a 5% error prediction, and at least 36 patients in each group using the Fleiss method. A total of 40 patients per group was achieved by adding 10% to the number calculated using the Fleiss method.

Patients with CKD (stage 4 and 5) were divided into 4 groups, each consisting of 40 people: Group 1 consisted of patients who did not receive HD treatment and had diabetes (HD–DM+), and Group 2 consisted of patients who did not receive HD treatment (HD–DM–) and did not have diabetes. Group 3 consisted of patients who received HD treatment and had diabetes (HD+DM+), and Group 4 consisted of patients who received HD treatment and did not have diabetes (HD+DM–). 

Patients younger than 18 years, those with major lower extremity amputation, those who had problems communicating, those who received HD treatment for less than 1 year, those who were immobilized, and those who declined to participate were excluded from the study.

Data collection
The risk classification based on the diabetic foot risk assessment algorithm was determined based on all the data obtained by the researcher (V.A.Ö.) using the question-answer technique via face-to-face interventions, examination of the medical records, and lower extremity evaluation comprising a comprehensive foot examination.

Patient information form. The patient information form was prepared by the researchers (V.A.Ö., N.N.) in line with the relevant literature and included sociodemographic characteristics, disease and treatment information, lower extremity complications, and comorbidities.

Foot physical evaluation form. The form used in the physical assessment of patients’ lower extremities and feet was developed by the researchers (V.A.Ö., N.N.) after scanning the relevant literature.^18,19 The form consisted of 4 parts: foot deformities, and dermatological, neurological, and vascular evaluations. 

The foot examination was based on a report published by the American Diabetes Association Foot Care Group approved by the American Society of Clinical Endocrinologists³ and on the “Diabetic Foot Evaluation Form” developed by the Diabetes Nursing Association in Turkey and included in the 2015 National Consensus Report.²⁰ Evaluation was made by including the key components of comprehensive screening of the foot. 

1. Evaluation of deformities. The lower extremities were inspected for the presence of deformities such as hallux valgus and hammer or claw toe. 

2. Dermatological evaluation. Dermatological examination of the lower extremities was done to evaluate for the presence of skin pathologies (eg, hyperkeratosis, uremic pruritus, xerosis or dry skin, fissure) and nail pathologies (eg, half nail, absence of lunula, onychomycosis, ingrown nails).

3. Neurological evaluation. Three methods were used for neurological examination: the LANSS pain scale, the protective sensory evaluation test, and the vibration test. If at least 2 methods were positive for neuropathy, the patient was considered to have neuropathy. 

3.1. LANSS pain scale. The Turkish validity and reliability study of the LANSS pain scale, which was first developed by Bennett,²¹ was performed by Yucel et al.²² The sensitivity of the scale was 89.9%, and the specificity was 94.2%. The LANSS pain scale has a total score of 24 and consists of 5 questions and 2 sensory tests (allodynia test and pin-prick test) to help diagnose neuropathic pain. A score of 12 or more is indicative of pain associated with neuropathy.

3.2. Evaluation of protective sense. Semmes-Weinstein monofilament test was performed on patients using 5.07/10 g Semmes-Weinstein monofilament to evaluate protective sensation or sensory neuropathy in both feet.

3.3. Vibration test. In addition to the monofilament test, the vibration test was applied using a manual 128-Hz electronic tuning fork to detect foot neuropathy. This test had a sensitivity of 95% and a specificity of 76%. Vibration testing was done in the manner described by O’Brien and Karem,²³ on the dorsal aspect of the distal phalanx of the hallux. In the current study, if the examiner felt the vibration while holding the tuning fork on the dorsal aspect of the distal phalanx of the patient’s hallux but the patient did not feel the vibration, the test was positive for neuropathy. 

4. Vascular evaluation. Vascular evaluation involved assessment for the presence of intermittent claudication, evaluation of bilateral pedal pulses, and calculation of the ABI. If the findings of at least 2 of these assessments were positive for vascular insufficiency, the patient was considered to have peripheral vascular insufficiency. 

4.1. Evaluation of vascular symptoms. The Edinburgh Claudication Questionnaire was administered to detect intermittent claudication, which is a symptom of PAD.²⁴ The questionnaire has 52.5% sensitivity and 87.1% specificity for intermittent claudication. Patients with claudication were evaluated in 3 subgroups: patients with stage I and stage II typical claudication and patients with atypical claudication.

4.2. Evaluation of pedal pulses. To confirm the suspicion of PAD, pedal pulses (dorsalis pedis and posterior tibialis) were evaluated as “normal, weak, or absent” by palpation.²⁵  

4.3. Ankle-brachial index. Finally, the ABI was measured and calculated as described in the European Society of Cardiology Guidelines on the Diagnosis and Treatment of Peripheral Arterial Diseases.²⁵ The ABI provides information about the severity of PAD; this method is easy to use and is low cost. The results of studies included in a systematic review investigating the reliability of the ABI suggest that the inter- and intra-tester reliability of the ABI is acceptable.²⁶ The ABI is a simple, objective, and reliable tool for the diagnosis of PAD; however, its diagnostic sensitivity is low in patients with DM.²⁷ For this reason, in the current study 3 different methods were used in the vascular evaluation of patients’ feet.

The ABI values were obtained by proportioning the highest systolic blood pressure values obtained for both lower extremities to the highest systolic blood pressure obtained from upper extremity measurements separately for the right and left lower extremities. Measurements were obtained from a single extremity in patients with arteriovenous fistula.²⁵

Diabetic foot risk assessment form. In line with the answers received from the patients and the data obtained as a result of the physical examination of the feet, the appropriate risk category was determined for each patient according to the diabetic foot risk assessment algorithm. Risk assessment was carried out using the 2007 Reducing Foot Complications for People with Diabetes supplement to the Nursing Best Practice Guideline: Shaping the Future of Nursing guideline created in 2004 by the Registered Nurses Association of Ontario.²⁸ According to the algorithm, a patient was considered to be at low risk for foot ulcers in the presence of all of the following conditions: protective sensation, presence of foot pulses, absence of severe deformity, no history of foot ulcers, and absence of amputation. The patient was considered to be at high risk for foot ulceration in the presence of 1 or more of the following conditions: impaired protective sensation, inability to receive foot pulses, severe foot deformity, history of foot ulcer, and past amputation.²⁸

Statistical analysis

The data obtained in this study were analyzed using SPSS version 23 (IBM Corporation). Normality of quantitative variables was evaluated using the Kolmogorov-Smirnov test. One-way analysis of variance from multiple comparison tests and the Tukey Honestly Significant Difference test were used to compare normally distributed variables. The Kruskal-Wallis test was used for nonnormally distributed variables, and the Mann-Whitney U test with Bonferroni correction was used to determine the groups causing the difference. Categorical variables were compared using the chi-square test. Nonnormally distributed quantitative variables were presented as median (minimum; maximum), and normally distributed variables were presented as mean ± standard deviation. Alpha was set at .05 as the level of significance.

Sociodemographic characteristics and lower extremity findings of the patients are shown in Table 1. There were significantly more patients aged 56 to 75 years in group 3 (HD+DM+) than in other groups (P = .012). The ratio of patients with normal weight according to BMI value was significantly higher in group 4 (HD+DM–) than in other groups (P = .014). No significant difference was observed between the groups in terms of other sociodemographic parameters (for all, P > .05).

Ulceration was detected in 3 (7.5%) patients in group 1 (HD–DM+), 2 (5%) patients in group 2 (HD–DM–), 6 (15%) patients in group 3 (HD+DM+), and 3 (7.5%) patients in group 4 (HD+DM–). The number of patients with a history of ulceration was significantly higher in group 3 (HD+DM+) than in other groups (P < .001). Similarly, the number of patients with a history of amputation was significantly higher in group 3 (HD+DM+) than in other groups (P = .018). Although the duration of ulceration was less than 1 month in the majority of patients in all groups, this duration was between 1 and 6 months in 32.5% of patients in group 3 (HD+DM+) and in 10% of patients in group 4 (HD+DM–) and differed significantly by group (P < .001) (Table 1).

The distribution of CKD findings by group is shown in Table 2. As expected, the most common etiologic factor for CKD was diabetic nephropathy in groups 1 and 3 (60.0% and 65.0%, respectively). The leading etiological factor for 52.5% of Group 2 and 37.5% of Group 4, consisting of patients without diabetes, was other causes (neurogenic bladder, congenital bladder anomalies, cardiorenal syndrome, radiotherapy, familial Mediterranean fever, multiple drug use). 

Foot deformity and dermatological evaluation findings are presented in Table 3. Foot deformities such as hammer or claw toe, hallux valgus, bony prominence, and pes cavus were more common in patients who received HD (groups 3 and 4). Hammer or claw finger and bony prominence deformities were more common (30.0% and 32.5%, respectively) and significantly higher in group 3 patients, those with DM who received HD treatment (P < .05). 

Pathological skin findings xerosis/dry skin (60.0%) and fissure (30.0%) were significantly more common in patients in Group 3 than in the other groups (P < .05). Among the nail pathologies, absence of lunula (82.5%) and abnormal thickening of the nails (50.0%) were significantly more common in Group 3 patients than in the other groups. (P < .05) (Table 3).

Neurological and vascular evaluation findings are shown in Table 4. According to the LANSS pain scale, the presence of neuropathic pain was more common in the groups with DM (groups 1 and 3) than in the groups without DM (groups 2 and 4); this difference was significant (P < .001). Neuropathic pain was detected at a rate of 60.0% in Group 1 and 70.0% in Group 3, which included patients with diabetes. There was no protective sensation in 60.0% of patients in group 1 (HD–DM+) and in 65.0% in group 3 (HD+DM+); the difference between groups was found to be statistically significant (P < .001). 

Among the patient group receiving HD treatment, dorsalis pedis and posterior tibialis artery findings on palpation in both extremities of patients with DM were significantly weaker than patients without DM. (P < .05) (Table 4).

The risk of foot ulceration in patients with DM receiving HD treatment was significantly higher (72.5%) than in the other group (P < .001), as showin in Table 5. 

This study was conducted to determine both the incidence of foot ulceration in individuals with CKD and the risk factors affecting foot ulceration in these patients. More than half of the patients in all 4 study groups were males aged 56 to 75 years and were primary school graduates. In a study by Locking-Cusolito et al²⁹ investigating foot ulcer risk factors in 232 patients undergoing HD, half of the patients (50.4%) were between age 60 and 79 years, and 56% were male. Similarly, in a study investigating the risk factors associated with foot ulceration in patients undergoing HD, Kaminski et al³⁰ reported a mean patient age of 67.5 years and that more than half of the patients (64.7%) were male. In the current study, there was no significant difference between the groups in terms of sex. In the study by Kaminski et al, where they investigated the risk factors for foot ulceration in patients with end-stage renal failure, the male gender ratio of hemodialysis patients was between 60.6% and 77.6%.³¹ In this regard, the results of the current study are consistent with those of previous studies.

The majority of patients in groups 1, 2, and 3 were in the obese group according to BMI category (ie, underweight, normal, overweight, obese), while the majority of patients in group 4 were in the normal group. The difference between the groups in terms of BMI was statistically significant. Similar results have been reported in other studies.³²

The etiology of CKD varies globally. Although DM and hypertension are often considered comorbidities, kidney disease has many complex causes.³² As in all high- and middle-income countries and some low-income countries, DM and hypertension are the main causes of CKD in Turkey.³³ Although DM is among the important causes of CKD, in this study the patients with DM were consciously selected. In addition, hypertension was found in 71.9% of all patients in this study. In a study investigating the risk factors for foot ulcers in patients undergoing HD, Kaminski et al³⁰ found DM to be the etiological factor in 40% of patients, followed by glomerulonephritis (21.6%), other factors (15.6%), and hypertension (6.2%). Differences between studies can be attributed to patient selection.

As expected, in the current study the incidence of foot ulcers was higher in the groups with DM. However, receiving HD treatment increases the rate of foot ulcers, as stated in the literature. Studies have shown that there is a close relationship between initiation of HD and the incidence of foot ulcers in patients with DM.^34,35 In the current study, a history of ulceration (P < .001) and/or amputation (P = .018) was significantly higher in group 3, that is, patients with DM and undergoing HD. 

There are other specific risk factors that contribute to the incidence of foot ulcers in patients with both DM and CKD.¹⁹ Although DM is a well-known cause of foot ulcers, HD and CKD have proven to be independent risk factors for the development of lower extremity ulceration and amputation.^5,6 However, the presence of foot ulcers in patients with renal failure but without DM should not be underestimated. In fact, in the current study a history of foot ulcer was seen in patients with renal failure but without DM. These data are also important in predicting the risk of foot ulcers in patients with CKD but without DM.

In this study, only patients in the groups with DM had a history of amputation. History of amputation was significantly more common in group 3 (HD+DM+) (n = 5) than group 1 (HD–DM+) (n = 2), which consisted of patients with DM who did not receive HD treatment (P = .018). In the study by Locking-Cusolito et al,²⁹ 23.7% of patients undergoing HD had a history of ulceration and 13.4% had a history of amputation. 

In the current study, foot deformities such as hammer or claw toe, hallux valgus, bony prominence, and pes cavus were more common in patients undergoing HD and were especially so in group 3, which consisted of patients with DM undergoing HD. These findings show that patients with DM receiving HD are at high risk for developing foot ulcers.

One study reported that 95.74% of participants with ESRD who underwent HD presented with foot deformities; in addition, foot morbidity was found to be high.⁶ Kaminski et al³¹ reported a rate of foot deformity of 69% in patients with ESRD alone, 72.9% in patients with DM alone, and 71.4% in patients with ESRD and DM. Deformities are a risk factor for foot ulcer formation, and different findings have been reported in the literature.

Regardless of whether DM is present, the feet of patients at any stage of renal failure should be evaluated in terms of dermatological pathologies that may predispose to the development of ulceration and especially the cutaneous consequences of neuropathy. In the current study, both skin and nail pathologies were more common in patients who received HD. Hyperkeratosis, xerosis/dry skin, and fissures were more common in patients who received HD. However, uremic pruritus was more common in patients with CKD who did not receive HD treatment. Nail pathologies such as half nail, absence of lunula, abnormal thickening of the nail, and subungual hyperkeratosis were also more common in patients undergoing HD.

In the study by Locking-Cusolito et al,²⁹ skin and nail deformities were common. The most common finding was dry skin, followed by thickened toenails, fissures, claw toes, and ingrown nails. In another study, 76.6% of participants with ESRD who underwent HD had skin and nail problems.⁶

Peripheral neuropathy is a disorder that affects the cell body, axon, or myelin of motor or peripheral sensory neurons and occurs in 60% to 100% of patients who receive HD owing to CKD. Patients with reduced glomerular filtration rate are at risk for uremic neuropathy. This distal symmetric sensorimotor polyneuropathy is caused by the accumulation of uremic toxins (ie, organic waste) associated with oxidative stress-related free radical activity. It is believed that hyperkalemia plays an important role in the pathophysiology of uremic neuropathy.³⁶ In the current study, neuropathic pain was significantly higher in patients with DM (groups 1 [n = 24] and 3 [n = 28]) than in patients without DM (groups 2 [n = 5] and 4 [n = 13]) according to the LANSS pain scale (P < .001). Remarkably, in the groups without DM, neuropathic pain was more common in patients without DM who were receiving HD. In a study evaluating risk factors for foot problems in 232 patients undergoing HD in which neuropathy was diagnosed using the monofilament test, neuropathy was found in 74.6% of patients.²⁹ Neuropathies appear to be more common in patients receiving HD than in patients not receiving HD. However, neuropathy is an important risk factor for the development of foot ulcers in patients without DM who are receiving HD.

In the current study, the highest risk of foot ulcer was found in patients with DM, especially in those with DM who received HD. In the study by Kaminski et al,³¹ although the rate of patients with 1 or more risk factors for the development of foot ulcers was not statistically significant, this rate was 71.8% in patients with ESRD alone, 80.0% in patients with DM alone, and 89.8% in patients with ESRD and DM. In a study by Ndip et al,⁵ the percentage of patients at high risk of developing foot ulcers was higher in patients undergoing HD (37%) than in those who did not receive HD treatment (29%). Similarly, in the current study, patients with DM who were receiving HD constituted the highest risk group for foot ulcers. Excluding patients with DM, according to the foot assessment algorithm, the group that included HD-treated patients had a higher risk of foot ulcers than the non-HD-treated group.

This study is limited to the responses of patients with CKD and with or without DM who received or did not receive HD treatment in 1 province of Turkey. Therefore, it is not possible to generalize the results. In addition, other limitations of the study include the small number of patients in each group, the fact that each patient was evaluated only once and follow-up data could not be obtained, and the fact that the patients had been receiving HD treatment for at least 1 year. 

The results of this study show that patients with DM who are receiving HD are at high risk of developing foot ulcers. Risk factors associated with foot ulcers were HD treatment, age, BMI, history of lower extremity ulceration and/or amputation, foot deformities, skin and nail pathologies, neuropathy, and vascular insufficiency. The feet of patients with CKD who receive HD treatment should be evaluated regularly by nurses to improve patients’ quality of life and reduce morbidity, and patients should be educated about foot care. Comprehensive studies with long-term follow-up are needed to further evaluate the risk of foot ulceration in individuals with CKD.