下腔静脉 (IVC) 滤器放置与重要的长期并发症相关。过滤器相关并发症的预测模型可能有助于指导临床决策，但仍然有限。我们开发了机器学习 (ML) 算法，可使用术前数据预测 1 年 IVC 过滤器并发症。血管质量倡议 (VQI) 数据库用于识别 2013 年至 2024 年间接受 IVC 过滤器放置的患者。当放置过滤器时，我们从住院指数中识别出 77 个术前人口统计/临床特征。主要结局是 1 年滤器相关并发症（滤器血栓形成、迁移、成角、骨折、栓塞或碎裂、静脉穿孔、新的腔静脉或髂静脉血栓、新的肺栓塞、通路部位血栓或回收失败的综合结果） 。数据分为训练集（70%）和测试集（30%）。使用具有 10 倍交叉验证的术前特征（极端梯度提升 [XGBoost]、随机森林、朴素贝叶斯分类器、支持向量机、人工神经网络和逻辑回归）训练了 6 个 ML 模型。主要模型评估指标是受试者工作特征曲线下面积 (AUROC)。使用校准图和 Brier 评分评估模型的稳健性。根据年龄、性别、种族、民族、农村地区、面积剥夺指数中位数、过滤器的计划持续时间、过滤器的着陆位置以及先前是否放置 IVC 过滤器，对各个亚组的表现进行评估。总体而言，14,476 名患者接受了 IVC 过滤器放置，584 名患者接受了 IVC 过滤器放置。 (4.0%) 经历了 1 年的过滤器相关并发症。具有主要结局的患者较年轻（59.3 [SD 16.7] vs. 63.8 [SD 16.0] 岁，p < 0.001），并且更有可能存在血栓形成危险因素，包括血栓形成倾向、既往静脉血栓栓塞 (VTE) 和 VTE 家族史。最佳预测模型是 XGBoost，其 AUROC (95% CI) 为 0.93 (0.92-0.94)。相比之下，逻辑回归的 AUROC (95% CI) 为 0.63 (0.61-0.65)。校准图显示预测/观察到的事件概率之间具有良好的一致性，Brier 得分为 0.07。 1 年过滤器相关并发症的前 10 个预测因素是 1) 血栓形成倾向，2) 既往 VTE，3) 抗磷脂抗体，4) 因子 V Leiden 突变，5) VTE 家族史，6) IVC 过滤器的计划持续时间（临时） )，7) 无法维持抗凝治疗，8) 恶性肿瘤，9) 近期/活动性出血，以及 10) 年龄。所有亚组的模型性能均保持稳健。我们开发的 ML 模型可以准确预测 1 年 IVC 滤器并发症，其性能优于逻辑回归。这些算法在指导患者选择滤器放置、咨询、围手术期管理和随访方面具有重要的实用价值，以减轻滤器相关并发症并改善结果。版权所有 © 2024。由 Elsevier Inc. 出版。

Inferior vena cava (IVC) filter placement is associated with important long-term complications. Predictive models for filter-related complications may help guide clinical decision-making but remain limited. We developed machine learning (ML) algorithms that predict 1-year IVC filter complications using pre-operative data.The Vascular Quality Initiative (VQI) database was used to identify patients who underwent IVC filter placement between 2013-2024. We identified 77 pre-operative demographic/clinical features from the index hospitalization when the filter was placed. The primary outcome was 1-year filter-related complications (composite of filter thrombosis, migration, angulation, fracture, and embolization or fragmentation, vein perforation, new caval or iliac vein thrombosis, new pulmonary embolism, access site thrombosis, or failed retrieval). The data were divided into training (70%) and test (30%) sets. Six ML models were trained using pre-operative features with 10-fold cross-validation (Extreme Gradient Boosting [XGBoost], random forest, Naïve Bayes classifier, support vector machine, artificial neural network, and logistic regression). The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). Model robustness was assessed using calibration plot and Brier score. Performance was evaluated across subgroups based on age, sex, race, ethnicity, rurality, median Area Deprivation Index, planned duration of filter, landing site of filter, and presence of prior IVC filter placement.Overall, 14,476 patients underwent IVC filter placement and 584 (4.0%) experienced 1-year filter-related complications. Patients with a primary outcome were younger (59.3 [SD 16.7] vs. 63.8 [SD 16.0] years, p < 0.001) and more likely to have thrombotic risk factors including thrombophilia, prior venous thromboembolism (VTE), and family history of VTE. The best prediction model was XGBoost, achieving an AUROC (95% CI) of 0.93 (0.92-0.94). In comparison, logistic regression had an AUROC (95% CI) of 0.63 (0.61-0.65). Calibration plot showed good agreement between predicted/observed event probabilities with a Brier score of 0.07. The top 10 predictors of 1-year filter-related complications were 1) thrombophilia, 2) prior VTE, 3) antiphospholipid antibodies, 4) Factor V Leiden mutation, 5) family history of VTE, 6) planned duration of IVC filter (temporary), 7) unable to maintain therapeutic anticoagulation, 8) malignancy, 9) recent/active bleeding, and 10) age. Model performance remained robust across all subgroups.We developed ML models that can accurately predict 1-year IVC filter complications, performing better than logistic regression. These algorithms have potential for important utility in guiding patient selection for filter placement, counselling, peri-operative management, and follow-up to mitigate filter-related complications and improve outcomes.Copyright © 2024. Published by Elsevier Inc.