Search

Objective: Prediction of malignant middle cerebral artery infarction (MMI) could identify patients for early intervention. We trained and internally validated a ML model that predicts MMI following mechanical thrombectomy (MT) for ACLVO., Methods: All patients who underwent MT for ACLVO between 2015 - 2021 at a single institution were reviewed. Data was divided into 80% training and 20% test sets. 10 models were evaluated on the training set. The top 3 models underwent hyperparameter tuning using grid search with nested 5-fold CV to optimize the area under the receiver operating curve (AUROC). Tuned models were evaluated on the test set and compared to logistic regression., Results: A total of 381 patients met the inclusion criteria. There were 50 (13.1%) patients who developed MMI. Out of the 10 ML models screened on the training set, the top 3 performing were neural network (median AUROC 0.78, IQR 0.72 - 0.83), support vector machine ([SVM] median AUROC 0.77, IQR 0.72 - 0.83), and random forest (median AUROC 0.75, IQR 0.68 - 0.81). On the test set, random forest (median AUROC 0.78, IQR 0.73 - 0.83) and neural network (median AUROC 0.78, IQR 0.73 - 0.83) were the top performing models, followed by SVM (median AUROC 0.77, IQR 0.70 - 0.83). These scores were significantly better than those for logistic regression (AUROC 0.72, IQR 0.66 - 0.78), individual risk factors, and the Malignant Brain Edema score (p < 0.001 for all)., Conclusion: ML models predicted MMI with good discriminative ability. They outperformed standard statistical techniques and individual risk factors., Competing Interests: Declaration of Competing Interest None., (Copyright © 2023 Elsevier Inc. All rights reserved.)

Background: In the treatment of aneurysmal subarachnoid hemorrhage (aSAH), microsurgical clipping, and endovascular therapy (EVT) with coiling are modalities for securing the ruptured aneurysm. Little data is available regarding associated readmission rates. We sought to determine whether readmission rates differed according to treatment modality for ruptured intracranial aneurysms., Methods: The Nationwide Readmissions Database (NRD) was used to identify adults who experienced aSAH and underwent clipping or EVT. Primary outcomes of interest were the incidences of 30- and 90-day readmissions (30dRA, 90dRA). Propensity score matching was used to generate matched pairs based on age, comorbidities, hospital volume, and hemorrhage severity., Results: We identified 13,623 and 11,160 patients who were eligible for 30dRA and 90dRA analyses, respectively. Among the patients eligible for 30dRA and 90dRA, we created 4282 and 3518 propensity score-matched pairs, respectively. There was no difference in the incidence of 30dRA (12.4% for clipping versus 11.2% for EVT; P = .094). However, 90dRA occurred more frequently after clipping (22.5%) compared to EVT (19.7%; P = .003). Clipping was associated with poor outcome after 30dRA (odds ratio [OR] = 1.51, 95% confidence interval [CI] 1.21-1.88, P < .001) and after 90dRA (OR = 1.60, 95% CI 1.34-1.91, P = .001). Mean duration to readmission and cost of readmission did not vary, but clipping was associated with longer lengths of stay during readmission., Conclusions: Microsurgical clipping of ruptured aneurysms is associated with a greater incidence of 90dRA, but not 30dRA, compared to EVT. Poor outcomes after readmission are more common following clipping., (Copyright © 2020 Elsevier Inc. All rights reserved.)