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

• Published in: Frontiers in endocrinology (Lausanne) Vol. 13; p. 1059753
• Main Authors: Liang, Jiemei, Song, Jiazhao, Sun, Tiehui, Zhang, Lanning, Xu, Shan
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 12.12.2022
• Subjects: Diabetes Mellitus, Type 2 - complications; Endocrinology; Humans; nomogram; Nomograms; Peripheral Arterial Disease - complications; Peripheral Arterial Disease - diagnosis; Peripheral Arterial Disease - epidemiology; peripheral artery disease (PAD); predictive model; Retrospective Studies; risk factor; Risk Factors; type 2 diabetes mellitus (T2DM); predictive model; nomogram; risk factor; type 2 diabetes mellitus (T2DM); peripheral artery disease (PAD)
• ISSN: 1664-2392; 1664-2392
• DOI: 10.3389/fendo.2022.1059753

To 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. Clinical 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. Age, disease duration, blood urea nitrogen (BUN), and hemoglobin ( <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 -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. A 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.