Development of a risk prediction model for permanent neurological dysfunction after total aortic arch replacement in patients with type A aortic dissection

Objective To explore the risk factors of permanent neurological dysfunction (PND) in patients with type A aortic dissection after total arch replacement, and to construct a nomogram model for predicting PND in these patients.

Methods A total of 337 patients with type A aortic dissection who underwent total arch replacement in the First Affiliated Hospital with Nanjing Medical University from January 2020 to October 2023 were selected as the study subjects. The patients were divided into PND group and non-PND group based on the occurrence of PND after total arch replacement. The risk factors of PND following total aortic arch replacement in patients with type A aortic dissection were screened using multivariate logistic regression analysis, and a nomogram model for predicting PND after total aortic arch replacement in these patients was constructed using R (version 4.2.3).

Results There were no statistically significant differences in gender, age, acute phase, hypoproteinemia, pericardial effusion, deep hypothermic circulatory arrest time, ventilator use time, overweight/obesity, anemia, smoking and other data between PND group and non-PND group (P＞0.05), while there were statistically significant differences in aortic cross-clamping time, cardiopulmonary bypass time, diabetes, hypertension and the severity of type A aortic dissection (P＜0.05). The multivariate logistic regression analysis indicated that aortic cross-clamp time (OR = 1.066, 95% CI: 1.040 to 1.092), cardiopulmonary bypass time (OR = 1.023, 95%CI:1.014 to 1.033), diabetes (OR = 4.003, 95%CI:1.612 to 9.941), hypertension (OR = 2.617,95%CI: 1.268 to 5.401), and ruptured dissection (OR = 2.855, 95%CI:1.35 to 6.032) were risk factors for PND following total aortic arch replacement in patients with type A aortic dissection (all P＜0.05). The nomogram model developed to predict PND in these patients demonstrated an area under the receiver operating characteristic (ROC) curve of 0.857 (95%CI: 0.797 to 0.918), with a calibration slope of 1.354 and a Brier score of 0.074. The model showed a sensitivity of 88.50% and a specificity of 71.20%. The Hosmer-Lemeshow test yielded χ² = 11.934 and P = 0.154, suggesting good agreement between predicted and observed probabilities in the calibration curve. Furthermore, both decision curve analysis and clinical impact curve evaluation demonstrated favorable clinical utility of the nomogram model.

Conclusions Aortic cross-clamp time, cardiopulmonary bypass time, diabetes, hypertension, and ruptured dissection are risk factors for PND following total aortic arch replacement in patients with type A aortic dissection. The nomogram model developed for predicting PND demonstrates acceptable accuracy and clinical utility.