Micro‐ and macrovascular complications and risk factors for foot ulceration and amputation in individuals receiving dialysis with and without diabetes

Abstract Introduction This study examined the prevalence of microvascular and macrovascular complications in people receiving dialysis with and without diabetes and investigated independent risk factors for foot ulcers and lower‐extremity amputations. Methods We performed a cross‐sectional study of 119 individuals with diabetes and 219 individuals without diabetes receiving chronic dialysis during June 2019 at the Department of Nephrology, Rigshospitalet, University of Copenhagen, Denmark. Effects of diabetes and other risk factors were assessed by log‐binomial regression. Prevalence data were compared with a historical control group of 38 individuals with diabetes receiving dialysis examined in 2004 in the same department. Results We found that persons with diabetes had a twofold higher risk ratio of current (unadjusted risk ratio 2.2 [95% CI 1.1, 4.7]) and previous foot ulcer (2.5 [1.7, 3.7]) and a fourfold higher risk ratio of lower‐extremity amputation (4.2 [2.1, 8.6]) in comparison with persons without diabetes (all p < .05). Furthermore, persons with diabetes had a 70% increased risk ratio of myocardial infarction (1.7 [1.0–2.8], p = .041). In multivariable‐adjusted analysis, current foot ulcer was independently associated with previous foot ulcer (adjusted risk ratio 4.0 [95% CI 1.8, 8.9]), while lower‐extremity amputation was independently associated with diabetes (3.8 [1.8, 8.2]) and male sex (4.1 [1.5, 11.3]) (all p < .01). Conclusions Individuals with diabetes receiving dialysis had a higher prevalence of foot ulcer, lower‐extremity amputation and myocardial infarction compared to individuals without diabetes. Previous foot ulcer was the most important risk factor for current foot ulcer, while diabetes and male sex were important risk factors for lower‐extremity amputation.


| INTRODUC TI ON
Diabetic nephropathy remains the leading cause of end-stage renal disease (ESRD) in most countries. 1 The combination of diabetes and ESRD is associated with a high prevalence of macrovascular complications, including a particularly high risk of adverse cardiovascular outcomes. 2,3 Microvascular complications other than kidney failure, that is retinopathy and neuropathy, are also very frequent in this population. 3 In a study from 2007, we reported a high prevalence of current foot ulcers, lower-extremity amputations and advanced diabetic eye complications in individuals with diabetes receiving dialysis, as compared to both individuals receiving dialysis without diabetes and individuals with diabetes with normal kidney function. 4 Since this study was performed, there has been a general improvement in the prognosis of people with diabetes in terms of reduced complication and mortality rate. [5][6][7] These improvements also apply to individuals with diabetic nephropathy without ESRD, and an improved survival and renal prognosis for this population suggest that the individuals who reach ESRD have had diabetes for a longer duration than previously. 8,9 It is unclear what the consequences of these changes are in relation to the prevalence of micro-and macrovascular complications among people with ESRD receiving dialysis.
In the present study, we assessed the prevalence of micro-and macrovascular complications in a cross-sectional study of individuals receiving dialysis with and without diabetes, and examined whether the prevalence of complications has changed since the abovementioned study. Furthermore, we examined independent risk factors for current foot ulcers and lower-extremity amputations.

| Study population
We conducted a cross-sectional study of individuals receiving chronic dialysis therapy (haemodialysis and peritoneal dialysis) during June 2019 at the Department of Nephrology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark. Individuals were identified using the department's records of people receiving chronic dialysis therapy. Inclusion criteria were chronic dialysis therapy for more than 3 months and age ≥18 years. People living in Greenland were excluded as we did not have access to their medical records. Subsequently, individuals were classified according to diabetes status into the following two groups: (1) the group with diabetes that included individuals with a previous or current diagnosis of diabetes, and (2) the group without diabetes with no known diagnosis of diabetes.

| Data collection and definitions
Data were collected using the unique personal identification number assigned to all persons in Denmark. Data regarding the duration of ESRD were obtained from the Danish Society of Nephrology

| Data analysis
All analyses were performed in SPSS version 25 (SPSS Inc., Chicago, IL, USA). p Value <.05 was considered statistically significant. The mean value (standard deviation) was applied for normally distributed data and median (interquartile range) for non-normally distributed data.
BMI was calculated as weight in kilograms divided by height squared in metres. For individuals with major amputations, BMI was adjusted to account for the estimated weight of the missing limb. In this case, we used the estimated body weight calculated as weight divided by 1 minus P, where P designates the estimated percentage of total body mass set to p = .0326 (3.26%) for below-knee amputations and p = .0996 (9.96%) for above-knee amputations. 10 The two-sample t test was used to compare continuous normally The study was performed as an audit, which required no formal ethics permission, but was approved by the hospital management of Rigshospitalet (approval 2019-05-27).  with diabetes in 2019 with those in 2004, we found a significant reduction from 29% to 12% in the prevalence of current foot ulcer (p = .013), but no reduction in the prevalence of lower-extremity amputations (p = .825). The retinopathy severity level was improved with a significant reduction in the prevalence of proliferative retinopathy from 53% to 27% (p = .007), and significantly fewer individuals had a visual acuity below 0.3 with a prevalence reduction from 37% to 18% (p = .028). There was no significant difference in the prevalence of total retinopathy (background plus proliferative) (p = .387), background retinopathy alone (p = .059) or maculopathy (p = .476). Furthermore, there was no significant difference in the prevalence of angina pectoris (p = .537), myocardial infarction (p = .122) or stroke (p = .962). In addition, we found no significant difference when performing separate analysis for individuals with type 1 and type 2 diabetes, respectively, with regard to the preva- The unadjusted risk ratio for micro-and macrovascular complications comparing people receiving dialysis in 2019 with and without diabetes is reported in Table 3. People with diabetes had a twofold higher risk ratio of current and previous foot ulcer and a fourfold higher risk ratio of lower-extremity amputation in comparison with people without diabetes. Furthermore, the risk ratio of myocardial infarction was 70% increased among people with diabetes compared to those without.

| RE SULTS
Investigated risk factors for a current foot ulcer and lowerextremity amputation assessed in univariable and multivariableadjusted analyses are presented in Tables 4 and 5. Presence of diabetes, previous foot ulcer, increasing age and longer duration of diabetes were statistically significantly associated with an increased risk ratio of current foot ulcer, while the presence of hypertension was associated with a decreased risk ratio of current foot ulcer in the univariable analyses. In the multivariable-adjusted analysis, previous foot ulcer, age and hypertension remained significantly related to a current foot ulcer, with previous foot ulcer increasing the risk ratio fourfold. Diabetes lost its significance in the multivariable analysis.
For lower-extremity amputation, univariable analyses showed that the presence of diabetes and male sex were significantly associated with an increased risk ratio, and in multivariable-adjusted analysis, both were associated with a fourfold increased risk ratio of amputation. Presence of hypertension was, as for current foot ulcers, associated with a significantly decreased risk ratio of amputation, in both univariable and multivariable analyses. Figure 3 shows the temporal relationship between first foot ulcer or first amputation and the start of dialysis. For the group with diabetes, there was no significant increase or decrease in the incidence of first foot ulcer (p = .312) or first amputation (p = .376) after the start of dialysis. For the group without diabetes, there was a significantly higher incidence of first foot ulcer (p = .000) and first amputation (p = .000) after the start of dialysis compared to the incidence F I G U R E 1 Flow diagram of study groups prior to dialysis start. Regarding recurrence of foot ulcers, 13 of 14 (93%) individuals with a current foot ulcer in the group with diabetes had a previous foot ulcer, while only four of 12 (33%) without diabetes with current foot ulcer had a previous foot ulcer. Furthermore, individuals with diabetes had a significantly higher prevalence of major amputations compared to individuals without diabetes (9.2 vs 1.4%, p = .001).

| DISCUSS ION
The present study examined the prevalence of micro-and macrovascular complications in people receiving dialysis with and without diabetes. Overall, we found a high prevalence of both micro-and macrovascular complications in people receiving dialysis. Comparing complication prevalence between individuals with and without diabetes, individuals with diabetes had a significantly higher risk ratio of current and previous foot ulcer, lower-extremity amputation and myocardial infarction.
In the multivariable-adjusted analysis, previous foot ulcer was the single most important risk factor for having a current foot ulcer, increasing the risk ratio fourfold. This is consistent with previous studies reporting a strong correlation between previous foot ulcer and current foot ulcer. 11-15 Diabetes did not remain an independent risk factor in the multivariable-adjusted analysis,  Note: Data are presented as mean ± SD, median (interquartile range) or n (%). ESRD denotes end-stage renal disease, HD haemodialysis, and PD peritoneal dialysis. In contrast to current foot ulcers, both diabetes and male sex were considerable risk factors for lower-extremity amputation in the multivariable-adjusted analysis. This discrepancy could partly be explained by the fact that a part was amputated prior to dialy-  Interestingly, hypertension was a protective factor for both current foot ulcer and amputation. A possible explanation for this could be that our data are biased by the inclusion of individuals with hypotension in our non-hypertensive population, considering that chronic hypotension has been found to affect 5%-10% of people receiving haemodialysis, and is associated with a poor prognosis. 18 Chronic hypotension might be caused by diabetes-and uraemia-related cardiovascular autonomic neuropathy, a serious condition that is also associated with increased risk of other complications. [19][20][21] We found a significant reduction in the prevalence of current The substantially decreased prevalence of current foot ulcers was not accompanied by a decrease in lower-extremity amputations.
Other studies of people with diabetes with and without ESRD suggest a trend towards a decline in major amputations, while minor am-   28 This review reports that the incidence of both proliferative retinopathy and sight-threatening retinopathy was reduced by twofold to threefold over the last three decades.
In the present study, covering a 15-year period, we found that the prevalence of proliferative retinopathy and visual impairment decreased twofold, thus corroborating the review. The review suggests that these improvements are explained by better blood pressure and glycaemic control, as well as improvement in diabetic eye screening.
Furthermore, the introduction of anti-vascular endothelial growth factor treatment for diabetic eye disease might play a role. 29 We found no statistically significant reduction in the prevalence of macrovascular complications over the 15-year period. These findings are contrary to studies of people with diabetes in general, as well as people with diabetic nephropathy without ESRD, reporting improved prognosis including reduced risk of macrovascular complications as a result of multifactorial treatment. [6][7][8][9]30 However, in the present study, the mean age was increased by approximately 6 years, and the prevalence of myocardial infarction was reduced from 32% to 20%, indicating a trend towards improvement in terms of cardiovascular complications, although neither was statistically significant. The prevalence of stroke remained at 22%, possibly explained, at least partly, by the high prevalence of atrial fibrillation or flutter (at 31%). Note: All variables with p value <0.3 in the univariable analysis were included in the multivariable analysis.
The use of bold values indicates a p-value below .05.

TA B L E 5
Univariable and multivariableadjusted risk ratio (RR) for risk factors associated with lower-extremity amputation in individuals with and without diabetes receiving dialysis in 2019 To some extent, the lack of reduction in the prevalence of macro- and type 2 diabetes, respectively. Thus, it would have been relevant to perform analyses for type 1 and type 2 diabetes separately, but we only had sufficient data to perform separate analyses regarding angina pectoris and stroke. Finally, an important limitation of this study was the small sample size, and our findings should be interpreted with caution.
In conclusion, individuals with diabetes receiving dialysis had a higher prevalence of foot ulcer, lower-extremity amputation and myocardial infarction compared to individuals without diabetes.
Previous foot ulcer was the most important risk factor for current foot ulcer, while diabetes and male sex were important risk factors for lower-extremity amputation.

CO N FLI C T S O F I NTE R E S T
The authors have no conflicts of interest to declare for this study.

AUTH O R CO NTR I B UTI O N S
DHK, TPA, VRS, BFR and MH designed the study. DHK and TPA collected the data, performed data analysis and wrote the first draft.
All authors participated in the analyses and interpretation of data, and in the critical review of the manuscript. All authors approved the final version of the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data were available on request due to privacy/ethical restrictions.

O RCI D
Dea Haagensen Kofod https://orcid.org/0000-0003-0173-8456 F I G U R E 3 Temporal relationship between first foot ulcer or first lower-extremity amputation and start of dialysis in individuals with and without diabetes receiving dialysis in 2019. *p < .05