From the Department of Neurology (T.M., C.L., M.B., B.S., A.S., M.M., M.G., D.S., S.J., M.A.), University Hospital Bern, Inselspital, University of Bern; Department of Neurology and Stroke Center (U.F.), University Hospital Basel and University of Basel; University Institute of Diagnostic and Interventional Neuroradiology (A.M., C.K., N.F.B., T.D., J.G., R.W., R.M., J.K.), Support Center for Advanced Neuroimaging (R.W., R.M., J.K.), and Department of Diagnostic, Paediatric and Interventional Radiology (J.K.), University Hospital Bern, Inselspital, University of Bern, Switzerland.
Neurology. 2022 Sep 5;99(10):e1009-e1018. doi: 10.1212/WNL.0000000000200815.
Very poor outcome despite IV thrombolysis (IVT) and mechanical thrombectomy (MT) occurs in approximately 1 of 4 patients with ischemic stroke and is associated with a high logistic and economic burden. We aimed to develop and validate a multivariable prognostic model to identify futile recanalization therapies (FRTs) in patients undergoing those therapies.
Patients from a prospectively collected observational registry of a single academic stroke center treated with MT and/or IVT were included. The data set was split into a training (N = 1,808, 80%) and internal validation (N = 453, 20%) cohort. We used gradient boosted decision tree machine learning models after k-nearest neighbor imputation of 32 variables available at admission to predict FRT defined as modified Rankin scale 5-6 at 3 months. We report feature importance, ability for discrimination, calibration, and decision curve analysis.
A total of 2,261 patients with a median (interquartile range) age of 75 years (64-83 years), 46% female, median NIH Stroke Scale 9 (4-17), 34% IVT alone, 41% MT alone, and 25% bridging were included. Overall, 539 (24%) had FRT, more often in MT alone (34%) as compared with IVT alone (11%). Feature importance identified clinical variables (stroke severity, age, active cancer, prestroke disability), laboratory values (glucose, C-reactive protein, creatinine), imaging biomarkers (white matter hyperintensities), and onset-to-admission time as the most important predictors. The final model was discriminatory for predicting 3-month FRT (area under the curve 0.87, 95% CI 0.87-0.88) and had good calibration (Brier 0.12, 0.11-0.12). Overall performance was moderate (F1-score 0.63 ± 0.004), and decision curve analyses suggested higher mean net benefit at lower thresholds of treatment (up to 0.8).
This FRT prediction model can help inform shared decision making and identify the most relevant features in the emergency setting. Although it might be particularly useful in low resource healthcare settings, incorporation of further multifaceted variables is necessary to further increase the predictive performance.
尽管进行了静脉溶栓（IVT）和机械取栓（MT），仍有约 1/4 的缺血性脑卒中患者预后极差，且与高逻辑和经济负担相关。我们旨在开发和验证一个多变量预后模型，以识别接受这些治疗的患者中无效再通治疗（FRT）。
纳入单中心前瞻性观察性登记研究中接受 MT 和/或 IVT 治疗的患者。数据集分为训练集（N=1808，80%）和内部验证集（N=453，20%）。我们使用梯度提升决策树机器学习模型，对入院时可用的 32 个变量进行 k-最近邻插补，以预测 3 个月时改良 Rankin 量表 5-6 定义的 FRT。我们报告特征重要性、区分能力、校准和决策曲线分析。
共纳入 2261 例患者，中位（四分位间距）年龄 75 岁（64-83 岁），46%为女性，中位 NIH 卒中量表 9 分（4-17 分），34%仅接受 IVT，41%仅接受 MT，25%接受桥接治疗。总体而言，539 例（24%）患者接受了 FRT，其中 MT 单独治疗（34%）比 IVT 单独治疗（11%）更常见。特征重要性确定了临床变量（卒中严重程度、年龄、活动性癌症、卒中前残疾）、实验室值（血糖、C 反应蛋白、肌酐）、影像生物标志物（脑白质高信号）和发病至入院时间为最重要的预测因素。最终模型对预测 3 个月 FRT 具有良好的区分能力（曲线下面积 0.87，95%CI 0.87-0.88），且校准良好（Brier 0.12，0.11-0.12）。总体性能为中等（F1 评分 0.63±0.004），决策曲线分析表明，在较低的治疗阈值（最高 0.8）下，有更高的平均净收益。
该 FRT 预测模型有助于辅助决策，并在急救环境中识别最相关的特征。尽管它在资源匮乏的医疗环境中可能特别有用，但需要进一步纳入多方面的变量以进一步提高预测性能。
Zotero 插件
