A Retrospective Study Assessing the Risk Factors of Type 2 Diabetes Mellitus Linked to Diabetic Foot Ulcer Incidence in Jiaxing, China

Objective: Diabetic foot (DF) is one of the common serious complications of diabetes, which is an important cause of death and disability, and is associated with diabetic vascular disease and diabetic neuropathy. The purpose of this study was to assess the incidence and risk factors for diabetic foot among a diabetic population. Methods: The study was a retrospective cohort review. The population studied was 348 male examinees of type 2 diabetes mellitus (DM), in which age-matched equal examinees (174 patients each) were with diabetic foot (DF Group) and with no indication of diabetic foot (non-DF group), hospitalized in the same year at The First Hospital of Jiaxing, China. Medical records were reviewed to collect clinical prole, including duration of disease, smoking, previous diabetic foot incidence, and medication (such as metformin). Categorical data between groups were analyzed using chi-square test (χ2). Unconditional logistic regression analysis was used for multi-factor analysis to identify the risk factors of diabetic foot. Results: Comparing the baseline data of DF Group and non-DF Group, age, course of disease, LDL-C and use of metformin showed no signicant difference (P >0.05). The results of the logistic regression analysis showed that smoking history (OR=1.88, P=0.020), previous diabetic foot history (OR=2.290, P=0.016), Hcy (OR=1.194, p ≤ 0.001) were independent risk factors for diabetic foot disease, and HgB (OR =0.984, P=0.021) was the protective factor of diabetic foot disease in male subjects. Age (OR=0.985, P=0.304), course of disease (OR=1.048, P=0.233), history of metformin use (OR=0.851, P=0.509), HbA1c (OR=1.302, P=0.059), LDL-C (OR=0.936, P=0.698), Creatinine (OR =1.010, P=0.326), and ALB (OR =0.943, P=0.084) were not signicantly correlated with diabetic foot disease. Conclusions: Smoking, previous diabetic foot history, and homocysteine are independent risk factors for diabetic foot disease. HgB is the protective factor of diabetic foot disease in male patients. The following relevant clinical data collected, and analyzed: age, course of disease, smoking, past history of diabetic foot, metformin use, homocysteine (Hcy), hemoglobin (HgB), glycosylated hemoglobin (HbA1c), creatinine, low density lipoprotein cholesterol (LDL-C), serum albumin (ALB), etc. Fasting blood samples were collected on the rst sent tube transport The biochemical indices were tested on DDP biochemical Homocysteine assay kit (cyclic enzyme method) was Jiaxing Botai Ltd. The reference value range of homocysteine in our is (15~20μmol/L); Glycosylated hemoglobin was determined using Bio-Rad blood hemoglobin analyzer, CA, USA. The detection method is high pressure liquid chromatography.


Introduction
Diabetic foot (DF) is one of the common serious complications of diabetes, which is an important cause of death and disability, and is associated with diabetic neuropathy and diabetic vascular disease(2020).
Nontraumatic lower limb amputation is a serious complication of both diabetic neuropathy and peripheral vascular disease. On an average, 50% of non-traumatic amputees are from diabetic population. The annual mortality rate of diabetic foot ulcer patients is 14.4%. Even after amputation, the mortality rate of DF is up to 40% despite providing routine treatment(2018). Smoking, uctuation in blood glucose level, and chronic kidney disease have been con rmed as some of the diabetic foot risk factors (Qiang 2016a). Further search for other risk factors for DF, especially simple indicators, and the intervention of risk factors is particularly important.
Homocysteine (Hcy), a sulfur-containing amino acid, is an essential metabolic intermediate in methionine metabolism. The sulfhydryl group of homocysteine is a reactive group that can participate in redox reaction and causes oxidative stress. It is generally accepted that homocysteine promotes atherosclerosis (Borowska 2017) and it is a risk factor for cardiovascular disease and stroke(Vadini 2019). Earlier studies have shown that homocysteine is a risk factor for DF in type 2 diabetic patients.
(Bruce and Young 2008; Gazzaruso 2012; Qiang 2016b) Current evidence suggests that metformin reduces the absorption of vitamin B12, which, in turn, leads to increase in the risk of B12 de ciency and giant cell anemia. Vitamin B12 de ciency may also lead to elevated levels of homocysteine. (Orlenko 2018) The relationship between metformin, homocysteine, and DF is not well studied. Given the importance to diabetic complications and lack of de nitive risk factors for DF, our present study aimed to assess the risk factors for DF in men based on metformin treatment, smoking and other clinical factors.

Research design
This study was carried out on 348 diabetic patients admitted in Department of Endocrinology, The First Hospital of Jiaxing from January 2012 to June 2019 .The enrolled diabetic patients were divided into two groups. DF Group: 174 men of type 2 DM with DF. Non-DF Group (control): 174 patients without diabetic foot selected according to age (inter-group gap <4 years) and course (inter-group gap <2 years) with 1:1 matching design. Both groups had no signi cant difference in age and course of disease (P >0.05). To carry out this study, verbal consent was obtained from the patients. The ethical committee of The First Hospital of Jiaxing approved this study.
Inclusion criteria: Patients with type 2 DM, diabetic foot Wagner ulcer grade 1 to 5.

Data collection
The medical records were retrospective reviewed. The following relevant clinical data were collected, counted and analyzed: age, course of disease, smoking, past history of diabetic foot, metformin use, homocysteine (Hcy), hemoglobin (HgB), glycosylated hemoglobin (HbA1c), creatinine, low density lipoprotein cholesterol (LDL-C), serum albumin (ALB), etc. Fasting blood samples were collected on the rst day morning after admission and immediately sent to the laboratory in our hospital through the Taikang pneumatic tube transport system. The biochemical indices were tested on the DDP module of Roche Modular P800 biochemical analyzer, USA. Homocysteine assay kit (cyclic enzyme method) was obtained from Jiaxing Botai Biological Technology Development Co., Ltd. The reference value range of homocysteine in our hospital is (15~20μmol/L); Glycosylated hemoglobin was determined using Bio-Rad D10 blood hemoglobin analyzer, CA, USA. The detection method is high pressure liquid chromatography. Metformin usage refers to continuous use for at least three or more months in the rst half of the year of diabetes (Dose: ≥0.75 g/ day). The smoking history refers to present the smoker of ≥1 year with average daily numbers ≥ 5 or former smoker who quit smoking in less than 10 years. History of DF refers to the Wagner ulcer grade 1 before the patient was hospitalized.

Statistical analysis
Statistical analysis was performed with SPSS 16.0 (SPSS Inc., Chicago, USA). Data were expressed as mean±SD. t-test was used for the comparison of quantitative variables between two groups. Categorical data between groups were analysed using chi-square test (χ 2 ). Unconditional logistic regression analysis was used for multi-factor analysis. A P value <0.05 is considered signi cant.

Results
A total of 348 diabetic patients were involved in this study. There was no signi cant difference in age, course of disease, LDL-C, and metformin use between the two groups (P>0.05), DF group and non-DF group. There were signi cant differences observed in smoking history, previous history of diabetic foot, Hcy, HbA1c, HgB, Creatinine, and ALB (P<0.05). See

Discussion
Within the aged population, the incidence of diabetes has been gradually increased, and the chronic complications associated with diabetes also increased signi cantly. DF is one of the serious complications of diabetes. The First Hospital of Jiaxing has a multidisciplinary cooperative treatment team for treating DF. The patients with DF have been well treated and so the numbers of patients with diabetic foot visiting our hospital have been increased greatly. So, we have enough patients in our hospital to provide samples for this study and carry out a lot of research on DF.
The early prevention of DF risk factors is the key to reduce the death and disability of diabetic foot. Our previous studies suggest that homocysteine is a risk factor for DF (Qiang 2016b), but we did not include smoking, history of DF, etc., in previous studies. Recent studies reported that metformin can cause vitamin B12 absorption disorders, vitamin B12 de ciency, folic acid de ciency, and leads to elevated homocysteine level which may aggravate diabetic peripheral neuropathy. On the other hand, metformin administration is believed to improve diabetic peripheral neuropathy by reducing blood sugar level. Diabetic peripheral neuropathy is an important cause of DF (2020), so metformin is worth further exploration. Real and co-workers discussed the relationship between homocysteine and diabetic foot, but did not compare the course of disease, age and sex. Homocysteine is not only a risk factor for cardiovascular and cerebrovascular diseases(Naveed and Bokhari 2015; Vadini 2019). This study further con rmed that homocysteine is a risk factor for diabetic foot and is in consistent with previous studies (Qiang 2016b). It is believed that homocysteine can lead to endothelial dysfunction. (Austin et al. 2004) It can directly damage and destroy vascular endothelial cells by oxidative stress, promoting lipid peroxidation, inhibiting nitric oxide synthesis, inducing cell necrosis and apoptosis. In addition, it also promotes proliferation of vascular smooth muscle cells, damage platelets and coagulation-brinolysis balance, resulting in vascular endothelial damage, accelerate the progress of atherosclerosis and thrombosis. It is believed that homocysteine can lead to arteriosclerosis and occlusion through direct cytotoxicity, (Lahiri et al. 2013;Tripathi 2015) and then cause ischemic and hypoxic injury of nerve, and damage nerve cells directly through endoplasmic reticulum stress (Lehotský et al. 2016). Our study suggests that metformin is not a risk factor for DF and that metformin (OR = 0.851, POR = 0.509) has no direct correlation with diabetic foot. Metformin can cause vitamin B12 de ciency and may affect the metabolism of homocysteine, but the latest study also suggests that metformin treatment can reduce the level of homocysteine(Zhang 2019). Homocysteine is not only affected by drugs, but also associated with abdominal obesity and insulin resistance. Metformin treatment reduces homocysteine by reducing insulin resistance.
This study suggests that smoking history (OR = .1.88, POR = .0.020), previous history of DF disease (OR = .2.290, POR = .0.016) are risk factors for DF and are consistent with current view(2020). Of all the risk factors, the history of DF was the most correlated and the risk was the highest. After the rehabilitation of DF patients, how to prevent the recurrence of DF is particularly important.
This study suggests that hemoglobin is negatively correlated with the occurrence and development of diabetic foot (OR = 0.984, POR = 0.021). Hemoglobin is the protective factor of DF; and anemia is the risk factor of DF. The role of hemoglobin in the occurrence and development of DF cannot be ignored.
Hemoglobin is the protein responsible for oxygen transport in the body. Anemia often causes local ischemia and hypoxia, exacerbates the occurrence and development of diabetic foot. Hemoglobin is directly related to the prognosis of DF (Costa 2017).
This study suggested that age (OR = 0.985, P = 0.304), course of disease (OR = 1.048, P = 0.233), HbA1c (OR = 1.302, P = 0.059), LDL-C (OR = 0.936, P = 0.698), creatinine (OR = 1.010, P = 0.326), and ALB (OR = 0.943, P = 0.084) were not signi cantly correlated with DF. The occurrence of DF is closely related to the poor control(2020) of blood sugar, but our hospital did not suggest that glycosylated hemoglobin is related to DF, and the occurrence is affected by long-term chronic hyperglycemia. Glycosylated hemoglobin represents blood sugar for nearly 2 ~ 3 months.

Conclusions
Smoking history, previous diabetic foot history, and homocysteine are independent risk factors for DF disease. HgB is the protective factor of DF disease in male patients. This study suggests that metformin is not a risk factor for DF and has no direct correlation with DF. Thus, our retrospective study re ects that the usage of metformin to reduce blood sugar levels is safe in DM patients. We are planning for more prospective studies on metformin and DF in DM patients to validate our ndings further.

Availability of data and materials
The data used to support the ndings of this study are available from the corresponding author upon request.
Ethics approval and consent to participate