Construction and validation of a nomogram model for predicting the risk of recurrence in patients with lower extremity arteriosclerosis obliterans after surgical intervention.

Abstract

Objective: To explore and analyze the risk factors for recurrence in patients with lower extremity arteriosclerosis obliterans (ASO) after surgical intervention and to construct and validate a nomogram prediction model.

Methods: A total of 270 patients with ASO treated at our hospital were retrospectively selected as study subjects and divided into a training cohort (189 cases) and a validation cohort (81 cases) based on a 7:3 ratio. Patients in the training cohort were further divided into recurrence and non-recurrence groups based on whether they experienced recurrence within two years post-surgery. Univariate and multivariate logistic regression analyses were employed to identify independent risk factors for postoperative recurrence, which were then used to construct a predictive model and generate a nomogram.

Results: Of the 270 patients with ASO included in the study, the training cohort consisted of 189 patients, with 76 (40.21%) in the recurrence group and 113 (59.79%) in the non-recurrence group. The validation cohort consisted of 81 patients, with 32 (39.51%) in the recurrence group and 49 (60.49%) in the non-recurrence group. Univariate analysis in the training cohort revealed significant differences in age, body mass index (BMI), diabetes, hypertension, lesion location classification, use of antiplatelet drugs, triglycerides, fibrinogen (FIB), and di-dimer (D-D) (P < 0.05, respectively). Multivariate logistic regression analysis indicated that age ≥ 60 years, BMI ≥ 24 kg/m², diabetes, hypertension, discontinuation of antiplatelet therapy, FIB, and D-D were independent risk factors for recurrence after surgical intervention in patients with lower extremity ASO (OR = 2.471, 1.625, 4.568, 2.678, 5.974, 2.073 and 3.067; P < 0.05, respectively). When the training and validation cohorts were tested in the established nomogram model, the area under the curve (AUC) of the model was 0.832 (95% CI: 0.765-0.919) in the training cohort and 0.858 (95% CI: 0.745-0.964) in the validation cohort. Calibration curves indicated high consistency between the predicted and actual outcomes in both groups, suggesting good predictive accuracy of the model. Decision curve analysis showed that using this model significantly increased net clinical benefit for patients.

Conclusion: The nomogram model constructed for predicting the risk of recurrence in patients with lower extremity ASO after surgical intervention demonstrates good predictive and discriminative abilities, offering valuable guidance for clinical screening of high-risk populations.