Development and validation of a nomogram to predict the risk of peripheral artery disease in patients with type 2 diabetes mellitus

Abstract

ObjectiveTo develop and validate a nomogram for predicting the risk of peripheral artery disease (PAD) in patients with type 2 diabetes mellitus (T2DM) and assess its clinical application value.MethodsClinical data were retrospectively collected from 474 patients with T2DM at the Air Force Medical Center between January 2019 and April 2022. The patients were divided into training and validation sets using the random number table method in a ratio of 7:3. Multivariate logistic regression analysis was performed to identify the independent risk factors for PAD in patients with T2DM. A nomogram prediction model was developed based on the independent risk factors. The predictive efficacy of the prediction model was evaluated using the consistency index (C-index), area under the curve (AUC), receiver operating characteristic (ROC) curve, Hosmer-Lemeshow (HL) test, and calibration curve analysis. Additionally, decision curve analysis (DCA) was performed to evaluate the prediction model’s performance during clinical application.ResultsAge, disease duration, blood urea nitrogen (BUN), and hemoglobin (P<0.05) were observed as independent risk factors for PAD in patients with T2DM. The C-index and the AUC were 0.765 (95% CI: 0.711-0.819) and 0.716 (95% CI: 0.619-0.813) for the training and validation sets, respectively, indicating that the model had good discriminatory power. The calibration curves showed good agreement between the predicted and actual probabilities for both the training and validation sets. In addition, the P-values of the HL test for the training and validation sets were 0.205 and 0.414, respectively, indicating that the model was well-calibrated. Finally, the DCA curve indicated that the model had good clinical utility.ConclusionA simple nomogram based on three independent factors–duration of diabetes, BUN, and hemoglobin levels–may help clinicians predict the risk of developing PAD in patients with T2DM.

Predictive model and risk analysis for peripheral vascular disease in type 2 diabetes mellitus patients using machine learning and shapley additive explanation

“…Therefore, prompt diagnosis and treatment of PVD in DM patients are crucial. Currently, PVD diagnosis primarily depends on imaging techniques like B-ultrasound and MRI ( 52 ). However, these methods can be costly and time-consuming, and with the rising number of diabetes cases, imaging every diabetic patient presents a considerable challenge.…”

“…Liang et al. ( 52 ) developed a nomograph to predict the risk of PAD, however, nomograph method is inapplicable for large-scale datasets. Wilcox et al.…”

“…And the SHAP values plot indicate that the influence of Hb on PVD is complex and non-linear. Decreased Hb levels can lead to inadequate oxygen supply, particularly in already damaged blood vessels, which can exacerbate vascular damage and inflammatory responses, thus increasing the risk of cardiovascular events and lower limb arterial disease, Additionally, Hb is an important indicator for diagnosing anemia, and anemia can also impact diabetes control and the progression of complications ( 52 ). Individuals with both diabetes and anemia suffered more comorbidities, were more likely to be hospitalized and have a higher risk of death ( 55 ).…”

“…A retrospectively study indicated that BUN was an independent risk factor for PVD in patients with T2DM. BUN is an easily recognizable, widely accessible, and inexpensive marker that can identify patients at high risk for PVD ( 52 ). Our study obtained similar results concerning the aforementioned factors.…”

Association between blood urea nitrogen levels and diabetic retinopathy in diabetic adults in the United States (NHANES 2005-2018)

“…Multiple studies corroborate a correlation between younger diabetes onset and augmented diabetic complications (31,32), positing elevated glycated hemoglobin levels and hastened pancreatic b-cell failure (33) as putative mechanisms, engendering formidable glycemic control challenges in this demographic. Blood urea nitrogen is extensively utilized in the assessment and prognosis of renal insufficiency and heart failure, therefore many previous studies have concentrated on the connection between BUN and diabetic nephropathy (34) or diabetic adverse cardiovascular events (35). In recent years, studies on the correlation between BUN and DR have shown a positive correlation, implying that DR, diabetic nephropathy, and severe cardiovascular events may all have a shared etiology linked to elevated blood urea nitrogen levels.…”