Nomogram to predict unfavorable outcome of endovascular thrombectomy for large ischemic core

Objective: The prognosis for patients presenting with a large ischemic core (LIC) following endovascular thrombectomy is relatively poor. This study aimed to construct and validate a nomogram for predicting 3-month unfavorable outcome in patients with anterior circulation occlusion-related LIC who underwent endovascular thrombectomy. Methods: A retrospective training cohort and a prospective validation cohort of patients with a large ischemic core were studied. The diffusion weighted imaging related radiomic features and pre-thrombectomy clinical features were collected. After the selection of relevant features, a nomogram predicting modified Rankin Scale score of 3-6 as an unfavorable outcome was established. The discriminatory value of the nomogram was evaluated with a receiver operating characteristic curve. Results: A total of 140 patients (mean age 66.3 +/- 13.4 years, 35% female) were included in this study, consisting of a training cohort (n = 95) and a validation cohort (n = 45). The percentage of patients with an mRS scores of 0-2 was 30%, 0-3 was 40.7%, and 32.9% were dead. Age, National Institute of Health Stroke Scale (NIHSS) score, and two radiomic features, Maximum2DDiameter-Column and Maximum2DDiameterSlice, were identified as factors associated with unfavorable outcome in the nomogram. The nomogram demonstrated an area under the curve of 0.892 (95% confidence interval [CI], 0.812-0.947) in the training dataset and 0.872 (95% CI, 0.739-0.953) in the validation dataset. Interpretation: This nomogram, which includes age, NIHSS score, Maximum2DDiameterColumn, and Maximum2DDiameterSlice, may predict the risk of unfavorable outcome in patients with LIC caused by anterior circulation occlusion.

Automated summary

This study aimed to construct and validate a nomogram for predicting 3-month unfavorable outcome in patients with large ischemic core caused by anterior circulation occlusion who underwent endovascular thrombectomy. The study used retrospective and prospective cohorts to collect diffusion weighted imaging related radiomic features and pre-thrombectomy clinical features. After feature selection, a nomogram predicting unfavorable outcome was established and evaluated using receiver operating characteristic curve. The results showed that the nomogram had good discriminatory value in predicting the risk of unfavorable outcome in patients with LIC caused by anterior circulation occlusion.