A Nomogram for Predicting Symptomatic Intracranial Hemorrhage after Endovascular Thrombectomy.

BACKGROUND: Symptomatic intracranial hemorrhage (sICH) is a devastating complication of endovascular thrombectomy (EVT). We aim to develop and validate a nomogram for predicting sICH in patients with large vessel occlusion (LVO) in the anterior circulation. METHODS: We performed a single-center retrospective analysis on collected data from patients undergoing EVT for LVO in the anterior circulation between January 2018 and December 2021. Forward stepwise logistic regression was performed to identify independent predictors of sICH and establish a nomogram. The discrimination and calibration of the model was accessed using the area under the receiver operating characteristic curve (AUC-ROC) and calibration plot. The model was internally validated using bootstrap and 5-fold cross-validation. RESULTS: 243 patients were included, among whom 23 developed sICH (9.5%). After multivariate logistic regression, baseline glucose level (odds ratio [OR], 1.16; p = 0.022), Alberta Stroke Program Early CT Score (OR, 0.44; p < 0.001), regional Leptomeningeal Collateral score (OR, 0.74; p < 0.001) were identified as independent predictors of sICH, which were then incorporated into a predictive nomogram. The ROC curve of the model showed good discriminative ability with an AUC of 0.856 (95% CI: 0.785-0.928). The calibration plot of the model demonstrated good consistency between the actual observed and the predicted probability of sICH. The model was internally validated by using bootstrap (1000 resamples) with an AUC of 0.835 (95%CI: 0.782-0.887) and 5-fold cross-validation with an AUC of 0.831 (95%CI: 0.775-0.887). CONCLUSION: Our model is a reliable tool to predict sICH after EVT. Although the model was internally validated, further external validation is also warranted.