Risk factors for ulceration and amputation in diabetic foot: study in a cohort of 496 patients
- First Online:
- Cite this article as:
- Moura Neto, A., Zantut-Wittmann, D.E., Fernandes, T.D. et al. Endocrine (2013) 44: 119. doi:10.1007/s12020-012-9829-2
Treatment strategies for foot at risk and diabetic foot are mainly preventive. Studies describing demographic data, clinical and impacting factors continue to be, however, scarce. Our objective was to determine the epidemiological presentation of diabetic foot and understand whether there were easily assessable variables capable of predicting the development of diabetic foot. This was a retrospective study of 496 patients with established foot at risk or diabetic foot, who were evaluated based on age, gender, type and duration of diabetes, foot at risk classification, and the presence of deformities, ulceration, and amputation. The presence of deformities, ulceration, and amputation was recorded in 45.9, 25.3, and 12.9 % of patients, respectively. As for diabetic foot classification, the great majority of our cohort had diabetic neuropathy (92.9 %). Approximately 30 % had neuro-ischemic disease and only 7.1 % had ischemic disease alone. Sixty-two percent of patients presented neuropathy with no signs of arteriopathy. Foot classification was as a significant predictor for the presence of ulcer (p = 0.009; OR = 3.2; 95 % CI = 1.18–7.3). Only male gender was a significant predictor for ulceration (p < 0.001). Predictors of amputation were male gender (p < 0.001; OR = 3.44 95 % CI = 1.81–6.56) and neuro-ischemic diabetic foot (p < 0.049; OR = 4.6; 95 % CI = 1.01–20.9). The predictors for diabetic foot were male gender and the presence of neuropathy. The combination of neuropathy and peripheral vascular disease adds significantly to the risk for amputation among patients with the diabetic foot syndrome. Men, presenting combined risk factors, should be a group receiving special attention and in the foot clinic, due to their potentially worse evolution.
KeywordsDiabetic footIschemic diseaseNeuropathyUlcerationAmputation
Ischemic only group
Neuropathy only group
Diabetic foot is one of the most feared complications of diabetes mellitus. Consequences include significant pain, deformities, and ulcerations, as well as decreased life expectancy, especially after amputation ensues. Foot problems continue to be the most common cause for hospitalization in diabetic patients, and the small number of patients with diabetic foot among the general population accounts for nearly two-thirds of all nontraumatic lower extremity amputations . Treatment strategies are mainly preventive as interventions in individuals presenting sequelae often yield poor results [2, 3]. Multidisciplinary team approaches were capable of reducing the amputation rate among patients with ischemic disease [4, 5].
Studies describing demographic data, clinical and impacting factors continue to be, however, scarce. This is a worrisome fact given that the prevalence of diabetes is predicted to increase by 69 % in developing countries in the next 20 years .
The great majority of local studies is focused on patients presenting ulcerations and/or risk factors and is based on screening of a small sample of diabetic patients. There is an important lack of studies of patients with foot at risk and diabetic foot strictly speaking.
We assessed a large cohort of 496 patients with diabetes and established foot at risk or diabetic foot from an important diabetic foot center with the objective of determining the epidemiological presentation of diabetic foot and understand whether there were any easily assessable variables capable of predicting the development of diabetic foot.
Patients and methods
This was a retrospective study of 496 patients with diabetes and foot at risk or diabetic foot under follow up at the Endocrinology Division, University of Campinas from January 2007 to January 2010. Data were collected from patients’ charts. Age, sex, diabetes type, time since diabetes diagnosis, diabetic foot at risk classification (neuropathic disease only, vascular disease only, or neuro-ischemic disease), diabetic foot risk (1, 2 or 3) , and prevalence of deformities, ulceration, and amputation were assessed.
Influence of clinical characteristics on the presence of amputation and ulceration was investigated.
Clinical foot evaluation was made in all patients with diabetes presenting symptoms or signs of neuropathy and/or arterial disease. Neuropathy was diagnosed with the Michigan Neuropathy Screening Instrument  and physical examination using the 10 g (Semmes–Weinstein) monofilament showing the presence of two or more points of insensitivity (hallux, 3rd toe tip, 1st, 3rd, and 5th metatarsal heads). A 128 Hz tune fork insensitivity at hallux was also considered in the diagnosis of neuropathy. Arteriopathy was considered when patients presented diminished/absent pedal and/or posterior tibial pulses on physical evaluation. Diminished lower extremity blood flow was confirmed in all of these cases by assessment of ankle/brachial index (ABI) and was considered compatible with ischemia when lower than 0.9 .
Risk classification was as follows: (1) the presence of neuropathic symptoms alone; (2) neuropathy and the presence of diabetes related deformities and no current or previous history of lower extremity ulceration or amputation; and (3) any case with history of current or previous ulceration or amputation .
Ulceration and deformities were recorded as positive when present at evaluation on the most recent visit.
Deformities were considered present if related to diabetes, i.e., plantar callus, prominent metatarsal heads, hammertoes, claw toes, flattening of the plantar arch, and Charcot’s osteoarthropathy.
Patients were classified as neuropathic (NP group) when criteria for neuropathy were met but all pulses were normal to palpation and no ABI abnormality was found; ischemic only (IO group) when there were no criteria for neuropathy but pulses were diminished or absent on physical evaluation and arteriopathy was later confirmed by ABI; and neuro-ischemic (NI group) when both diseases were present simultaneously.
Demographic characteristics were analyzed by measurement of position and dispersion for continuous variables and frequency tables for categorical variables.
Comparison between two groups of continuous variables was done with the Mann–Whitney test. For association of two categorical variables the Chi Square test or Fischer’s exact test were used, as appropriate.
Identification of risk factors associated with diabetic foot characteristics was done by univariate and multivariate logistic regression analysis.
Data are reported as mean (SD) or as absolute numbers (percentage), unless otherwise stated. Significance level was fixed at 5 % (p < 0.05) for all tests. All calculations were performed using the SAS System for Windows (Statistical Analysis System—SAS Institute Inc, 2002–2003, Cary, NC, USA) version 9.1.3 Service Pack 3 and the SPSS for Windows (Statistical Package for the Social Sciences—SSPS Inc, 1989–2004, Chicago, IL, USA) version 13.0.
This study followed the Declaration of Helsinki. The University’s Ethics in Research Committee approved the study.
Cohort’s demographic characteristics
Clinical and demographical characteristics of all patients
n = 496
Age (SD) (years)
Diabetes duration (SD) (years)
n = 403
Mean age of patients was 60.2 (11.7) years and mean duration of diabetes was 16.8 (8) years. Mean HbA1c was 8.6 % (2.3 %). We had 241 males (48.6 %), and 30 patients (6 %) had type 1 diabetes. As for therapy, the majority of patients (80.8 %) were on insulin and 67.3 % on combination of oral drugs and insulin. Only 19.2 % were on oral therapy alone.
The presence of deformities, ulceration, and amputation was recorded at their last visit in 45.9, 25.3, and 12.9 % of patients, respectively. As for diabetic foot classification, the great majority of our cohort had diabetic neuropathy (92.9 %). Approximately 30 % had neuro-ischemic disease and only 7.1 % had ischemic disease alone. Sixty-two percent of patients presented neuropathy with no signs of arteriopathy (NP group). Also, 67.2 % of patients did not have any history of present or past ulceration or amputation.
We found a significant association of neuropathy and the presence of diabetes related deformities. Patients in the NI and NP groups had a significantly higher prevalence of deformities when compared to the IO group (46 and 48 vs 22 %, respectively; p = 0.02). There was no significant difference regarding age (p = 0.50), diabetes duration (p = 0.78), gender (p = 0.37), and type of diabetes (p = 0.09).
Uni and multivariate logistic regression analysis revealed only diabetic foot classification (NP and NI groups) as a significant predictor for the presence of deformities (p = 0.009; OR = 3.2; 95 % CI = 1.18–7.3).
There was a significant association between the presence of ulceration and male gender (33 vs. 18 % in males and females, respectively; p < 0.001) and type of diabetes (41 vs 24 % for type 1 and type 2 diabetes, respectively; p = 0.04). There was no significant association for other variables: age (p = 0.14), duration of diabetes (p = 0.94), and classification (p = 0.38).
Univariate logistic regression analysis showed that male gender (p < 0.001; OR = 2.15; 95 % CI = 1.36–3.40) and type 1 diabetes (p = 0.04; OR = 2.22; 95 % CI = 1.03–4.84) were predictors for the presence of ulceration. On multivariate analysis, however, only male gender remained as a significant predictor for this outcome (p < 0.001; OR = 2.15; 95 % CI = 1.36–3.40). The surrogate R2 (Nagelkerke) statistic for this multivariate model was 0.74.
We found significant associations between the presence of amputation and male gender (20 and 7 % in males and females, respectively; p < 0.001) and NI foot at risk classification (21 vs 9 % and 6 % for NI vs NP and IO, respectively; p = 0.006). There was no association between the presence of ulceration and age (p = 0.11), diabetes duration (p = 0.93), or type of diabetes (p = 1.0).
Risk factors for the presence of deformities, ulceration and amputation and number and percentage among all patients of all characteristics assessed for on univariate analysis
n = 403
M (% all cases)
F (% all cases)
Diabetes duration (years)
1 (% all cases)
2 (% all cases)
NP (% all cases)
NI (% all cases)
IO (% all cases)
Significant risk factors for the presence of deformities, ulceration, and amputation
95 % CI
95 % CI
95 % CI
Classification (NP/NI vs IO)
Gender (male vs female)
Diabetes type (1 vs 2)
Our study has shown that the prevalence of neuropathy without signs of ischemic disease was the most common form of presentation of diabetic foot in our cohort. Furthermore, neuropathy was present in more than 90 % of patients, whereas ischemic disease was identified in 37.7 % of cases, and only 7.1 % had ischemic disease without signs or symptoms of neuropathy.
Compared to other studies published, we found similar prevalences of type 1 and type 2 diabetes [10–16]. The mean age was just older than 60 years and diabetes duration was over 16 years, data more frequently found on countries of developed economies, where most commonly mean ages are well over 60 years [13, 15–21] and disease duration over 12–15 years [13, 15–18, 20–23]. Literature data from developing countries is much more scarce, but points to younger ages (commonly between 50 and 60 years) [11, 12, 14, 21, 24–28]. In these nations, time of diabetes vary between 6 and 15 years in patients presenting diabetic foot [11, 14, 21, 24, 28]. Some nations reported disease duration as short as 5 years . Countries with fast growing economies or those recently inserted in large economic communities such as Turkey, India, and Brazil show intermediate characteristics, i.e., patient age around 60 years and longer disease time, usually 10–15 years [10, 21, 25, 29–32].
Another finding was the higher prevalence of neuropathy when compared to arterial disease. This is also more commonly found in countries of developing economies, such as Arab and African nations, where ischemic disease/ulceration accounts for only 20–30 % of cases [11, 12, 21, 25, 31]. In contrast, nations from Western Europe and the USA, although reporting high prevalences of neuropathy, but not as high as those from the above-mentioned nations, show prevalence of ischemic disease usually between 40 and 50 % or more [13, 15, 18–21].
Factors influencing these differences are probably not due to diabetes type, as type 2 to type 1 diabetes ratio is similar in all studies, with type 1 representing 5–10 % of cases [10–16]. Rather, the high prevalence of obesity and consequently atherosclerotic disease as well as older mean age and possibly higher rates of smoking in the population from developed countries place diabetic individuals at higher risk for ischemic disease. In fast growing economies, such as Brazil, we are observing a steep rise in obesity and deaths from cardiovascular conditions in the last decades . As such, diabetic foot presentation could be changing as individuals are aging and gaining weight, and as better health care prolongs disease duration until complications present.
Our findings revealed that patients most likely to present amputation were those with neuro-ischemic disease, showing that combined risk factors put patients at a significantly higher risk. Also, our results showed that male patients are 3.4 times as likely to present amputation and 2.2 times as likely to present ulceration as females.
This difference in gender is attributable to better wound care in women, as men are more commonly involved in heavy physical work activities and deal with more social pressure to keep providing the family income. Also important, absenteeism is higher among men. These issues cause ulcerations to take longer to recover, being more likely to present at a consult at any given moment, and also more likely to terminate in amputation.
As for diabetes related deformities, as expected, the presence of neuropathy in NP and NI groups was associated with an increased prevalence, probably because neuropathy is required for alterations in weight distribution and muscle atrophy.
The main weaknesses of our study are its retrospective design and the lack of data on smoking habit and renal disease. Unfortunately, data on these parameters were lacking for too many patients, precluding any unbiased analysis. Additionally, as this was an audit of a clinical service, important data on lipid profile, insulin resistance as well as genetic background such as lipoprotein(a) and homocysteine are lacking. Perhaps these parameters, if added to those assessed in our study, could permit a more reliable stratification for the risk of diabetic foot .
In conclusion, the main predictors for diabetic foot complications in our cohort were male gender and the presence of neuropathy. The combination of neuropathy and peripheral vascular disease adds significantly to the risk of amputation among patients with the diabetic foot syndrome.
We thus believe that older men, presenting combined risk factors should be a group receiving more special attention and aggressive treatment in the foot clinic, due to their potentially worse evolution.
Conflict of interest
The authors declare that they have no conflict of interest.