Wake Forest University Baptist Medical Center, Winston-Salem, NC, USA.
Westchester Medical Center, Valhalla, NY, USA.
J Shoulder Elbow Surg. 2019 Dec;28(12):e410-e421. doi: 10.1016/j.jse.2019.05.017. Epub 2019 Aug 3.
We aimed to demonstrate that supervised machine learning (ML) models can better predict postoperative complications after total shoulder arthroplasty (TSA) than comorbidity indices.
The American College of Surgeons-National Surgical Quality Improvement Program database was queried from 2005-2017 for TSA cases. Training and validation sets were created by randomly assigning 80% and 20% of the data set. Included variables were age, body mass index (BMI), operative time, smoking status, comorbidities, diagnosis, and preoperative hematocrit and albumin. Complications included any adverse event, transfusion, extended length of stay (>3 days), surgical site infection, return to the operating room, deep vein thrombosis or pulmonary embolism, and readmission. Each SML algorithm was compared with one another and to a baseline model using American Society of Anesthesiologists (ASA) classification. Model strength was evaluated by calculating the area under the receiver operating characteristic curve (AUC) and the positive predictive value (PPV) of complications.
We identified a total of 17,119 TSA cases. Mean age, BMI, and length of stay were 69.5 ± 9.6 years, 31.1 ± 6.8, and 2.0 ± 2.2 days. Percentage hematocrit, BMI, and operative time were of highest importance in outcome prediction. SML algorithms outperformed ASA classification models for predicting any adverse event (71.0% vs. 63.0%), transfusion (77.0% vs. 64.0%), extended length of stay (68.0% vs. 60.0%), surgical site infection (65.0% vs. 58.0%), return to the operating room (59.0% vs. 54.0%), and readmission (64.0% vs. 58.0%). SML algorithms demonstrated the greatest PPV for any adverse event (62.5%), extended length of stay (61.4%), transfusion (52.2%), and readmission (10.1%). ASA classification had a 0.0% PPV for complications.
With continued validation, intelligent models could calculate patient-specific risk for complications to adjust perioperative care and site of surgery.
我们旨在证明，与合并症指数相比，监督机器学习（ML）模型可以更好地预测全肩关节置换术后（TSA）的术后并发症。
从 2005 年至 2017 年，美国外科医师学院-国家手术质量改进计划数据库（American College of Surgeons-National Surgical Quality Improvement Program database）被查询用于 TSA 病例。通过随机分配 80%和 20%的数据组创建了培训和验证集。包括的变量有年龄、体重指数（BMI）、手术时间、吸烟状况、合并症、诊断以及术前的血细胞比容和白蛋白。并发症包括任何不良事件、输血、住院时间延长（>3 天）、手术部位感染、返回手术室、深静脉血栓或肺栓塞以及再次入院。将每个 SML 算法与另一个算法以及美国麻醉师协会（ASA）分类进行比较。通过计算接收者操作特征曲线（receiver operating characteristic curve，ROC）下的面积（area under the receiver operating characteristic curve，AUC）和并发症的阳性预测值（positive predictive value，PPV）来评估模型的强度。
我们共确定了 17119 例 TSA 病例。平均年龄、BMI 和住院时间分别为 69.5±9.6 岁、31.1±6.8 和 2.0±2.2 天。百分比血细胞比容、BMI 和手术时间对结果预测最为重要。SML 算法在预测任何不良事件（71.0% vs. 63.0%）、输血（77.0% vs. 64.0%）、住院时间延长（68.0% vs. 60.0%）、手术部位感染（65.0% vs. 58.0%）、返回手术室（59.0% vs. 54.0%）和再次入院（64.0% vs. 58.0%）方面优于 ASA 分类模型。SML 算法对任何不良事件（62.5%）、住院时间延长（61.4%）、输血（52.2%）和再次入院（10.1%）的阳性预测值最大。ASA 分类对并发症的阳性预测值为 0.0%。
随着持续验证，智能模型可以计算患者特定的并发症风险，以调整围手术期护理和手术部位。
Zotero 插件