Case Analysis and Diagnostic Model Construction in Patients with Transient Symptomatic Cerebral Infarction

Xuetong Sun ¹ Wei Quan ² Yan Li ³ Wanyue Sun ³ Fule Ren ¹ Yuqing Feng ¹ Jiao Kong ² Mingxi Li ² Zhiyong Sun ^4* Zhicheng Wang ^1*

• ¹ NHC Key Laboratory of Radiobiology, School of Public Health, Jilin University, Changchun, Hebei Province, China
• ² China-Japan Union Hospital, Jilin University, Changchun, Jilin Province, China
• ³ Department of Health Toxicology, School of Public Health, Jilin University, Changchun, Jilin Province, China
• ⁴ Department of Neonatology, Jilin Women And Children Health Hospital, Changchun, Jilin Province, China

Background: TSI is an independent disease distinct from TIA and ischemic stroke. TSI patients have a significantly higher risk of subsequent ischemic stroke within one year compared to TIA, imposing substantial burdens on society and families. Currently, the clinical characteristics and influencing factors of TSI remain unclear, contributing to potential confusion with TIA patients. Therefore, this article mainly analyzes the clinical characteristics and influencing factors of TSI patients, aiming to provide a basis for the early diagnosis and prevention of TSI.Methods: Between March 2019 and March 2024, data from 507 patients initially diagnosed with TIA who underwent DWI examination were retrospectively collected. Based on DWI sequence results, patients were categorized into the TIA group (DWI negative) and the TSI group (DWI positive). The demographic characteristics, clinical manifestations, past medical history, biochemical indicators upon admission, and imaging parameters were compared between the two groups. Risk factors associated with TSI using logistic regression models were investigated and a diagnostic model for TSI was developed. The model's discrimination and calibration were evaluated using the ROC curve and calibration plot.Results: A total of 507 patients were included in the study, with 293 in the TIA group and 214 in the TSI group. Large-artery atherosclerosis is the predominant cause of TSI, accounting for 53.3% of cases. Logistic regression identified several independent predictors for TSI: NT-proBNP (odds ratio 1.016, P<0.001); smoking history (odds ratio 2.040, P=0.003); ABCD3 score (odds ratio 1.495, P<0.001); Apo-A1 (odds ratio 0.123, P<0.001); and vascular stenosis degree. Based on these indicators, a diagnostic model for predicting TSI was established. The ROC curve and calibration plot demonstrated good discriminative ability (AUC=0.861, 95% CI: 0.829-0.892) and calibration (mean absolute error=0.023).Conclusion: High ABCD3 score, smoking history, elevated NT-proBNP levels, severe vascular stenosis, and lower levels of Apo-A1 are associated with increased risk of TSI. The diagnostic model established demonstrates discriminative and predictive value in diagnosing TSI.