Risk factors for diabetic foot ulcers: an Albanian retrospective study of inpatients with type 2 diabetes.

The aim of this study was to assess the impact of glucose control, diabetes-related complications and cardiometabolic risk factors on the risk of diabetic foot ulcers (DFUs) and DFU complications in Albanian adult inpatients with T2D.We conducted a retrospective case-control study on 482 Albanian adult inpatients with T2D. DFU was defined as a full-thickness skin lesion requiring ≥14 days for healing and was classified at the time of hospital admission. Demographic and biochemical parameters of the study participants, the presence of comorbidities and diabetes-related complications at the time of hospital admission were evaluated through a retrospective chart review.Mean age of study participants was 54.8±10.7 years. Participants (284 males and 198 females) were divided into two groups: DFU (cases; n=104) and non-DFU (controls; n=378). Multivariate analysis (performed by a logistic regression model) revealed that the most relevant independent variables associated with DFU were BMI [OR=0.62; p=0.007], HDL-cholesterol [OR=0.00; p<0.0001], triglycerides [OR=7.48; p=0.0004], cigarette smoking [OR=26.46; p=0.005], duration of diabetes [OR=1.53; p<0.0001], fasting plasma glucose (FPG) [OR=1.06; p<0.0001], systolic blood pressure (SBP) [OR=1.13; p=0.0004] and insulin therapy alone [OR=0.11; p=0.02]. ROC curve analysis showed that FPG (AUC=0.83), glycated hemoglobin (HbA1c) (AUC=0.75), triglycerides (AUC=0.78) and HDL-cholesterol (AUC=0.82) were the most reliable biomarkers able to detect DFU. In the DFU group, the most relevant independent variables associated with previous minor lower-extremity amputations (LEAs) were represented by HbA1c [OR=1.47; p=0.03], age <55 years [OR=0.12; p=0.05] and female sex [OR=4.18; p=0.03]; whereas the most relevant independent variables associated with diabetic peripheral neuropathy (DPN) were HbA1c [OR=1.70; p=0.006], SBP [OR=1.08; p=0.05], BMI [OR=1.20; p=0.03] and lack of cigarette smoking [OR=0.07; p=0.01]. Correlation analysis (performed through the nonparametric Spearman's rank correlation test or through the parametric Pearson test, as appropriate) revealed a significant positive relationship between HbA1c and FPG (r=0.58; p<0.0001), ulcer surface area (r=0.50; p<0.0001), ulcer grade (r=0.23; p=0.02), minor LEAs (r=0.20; p=0.04), DPN (r=0.41; p<0.0001), and metformin therapy alone (r=0.72; p<0.0001). There was a significant inverse correlation between HbA1c and insulin therapy alone (r=-0.31; p=0.01) and combined metformin and insulin therapy (r=-0.60; p<0.0001). Both DFU and non-DFU groups exhibited suboptimal mean LDL-cholesterol levels (>100 mg/dl) and mean HbA1c values >7.5%. Moreover, in DFU group HbA1c values were markedly elevated (≥10%) particularly in patients with a grade 3 ulcer and an ulcer surface area ≥4 cm2, as well as in patients with history of minor LEAs and in patients affected by DPN.The present study suggested that longer duration of diabetes, cigarette smoking, lower HDL-cholesterol levels, poor glucose control, and elevated triglyceride and SBP values may all represent major risk factors for the development of DFU in Albanian patients with T2D. Thus, community interventions and health policies aimed to improve the management of diabetes and related cardiometabolic risk factors should be urgently implemented in Albania, in order to prevent DFUs and other diabetes complications in patients with T2D.