Analysis of Risk Factors Associated with Organic Erectile Dysfunction in Patients with Type 2 Diabetes Mellitus and Erectile Dysfunction.

Background: Diabetes mellitus is a common metabolic disorder, and diabetic erectile dysfunction (DMED) is one of its common complications. The differentiation of the types of erectile dysfunction (ED) is fundamental to treatment, yet there is a lack of simple and efficacious tools for this purpose in clinical practice. In this study, we endeavor to predict ED types using commonly available clinical data from diabetic patients, aiming to develop and assess a risk prediction model for organic erectile dysfunction in individuals with type 2 diabetes.

Methods: The study was a retrospective analysis. Data were obtained from the hospital's internal medical record system. We selected and analyzed the clinical data of 250 patients with type 2 diabetes. Lasso regression was used for risk factor selection, and the selected variables were included in a multivariate logistic regression analysis to establish the risk prediction model. Internal validation was performed using the bootstrap method, and the discrimination, calibration, and clinical effectiveness of the model were evaluated using the C-index, calibration curve, decision curve analysis (DCA), and receiver operating characteristic (ROC) curve.

Results: Among the 250 patients, 168 (67.2%) were diagnosed with organic ED. The risk factors included in the logistic regression analysis were the duration of diabetes, low-density lipoprotein cholesterol (LDL-C), red blood cell distribution width (RDW), intima-media thickness of the carotid artery, diabetic retinopathy, diabetic nephropathy, and peripheral neuropathy. The C-index was 0.827 (95% confidence interval (CI) = 0.772-0.882). The distribution curve of the predicted values and the calibration curve of the model were well fitted. The decision curve analysis (DCA) suggested that using the model could be clinically beneficial when the threshold probability was between 28% and 100%.

Conclusion: By combining the duration of diabetes, carotid artery intima-media thickness, diabetic retinopathy, diabetic nephropathy, peripheral neuropathy, RDW, and LDL-C, this study preliminarily establishes a risk prediction model for organic ED in patients with type 2 diabetes mellitus. The model demonstrates good predictive performance.