To investigate the influencing factors and prediction modeling of contrast-induced nephropathy (CIN) after thrombectomy in diabetic patients with acute cerebral infarction.

A retrospective analysis was performed on 152 diabetic patients who underwent mechanical thrombectomy for acute cerebral infarction in the Affiliated Hospital of Xuzhou Medical University from September 2018 to June 2024. The patients were divided into CIN group (37 cases) and non-CIN group (115 cases) according to whether CIN was confirmed. Logistic regression method was used to analyze the influencing factors of CIN development. R language method was used to construct the prediction model nomogram, and clinical decision curve was constructed to evaluate its clinical application value. The receiver operating characteristic curve was drawn to evaluate the predictive role of the model in the development of CIN.

Univariate analysis showed that there were statistically significant differences in contrast dose, operation duration, white blood cell count, neutrophil count, fibrinogen, and triglyceride-glucose index between the two groups (P<0.05). Multivariate logistic regression analysis showed that the dose of contrast (OR=1.054, 95%CI: 1.014-1.096, P=0.008), triglyceride-glucose index (OR=3.554, 95%CI: 1.550-8.154, P=0.003), and fibrinogen (OR=3.358, 95%CI: 1.633-6.904, P=0.001) were independent risk factors for CIN. The receiver operating characteristic curve analysis showed that when triglyceride-glucose index, fibrinogen, and contrast dose were combined to predict CIN, the area under the curve was 0.857, with the sensitivity being 86.5%, and the specificity 77.4%.

The contrast dose, triglyceride-glucose index, and fibrinogen were the independent risk factors for CIN after thrombectomy in the diabetic patients with acute cerebral infarction. The nomogram model constructed had a certain value in predicting the occurrence of CIN after thrombectomy in the diabetic patients with acute cerebral infarction.