Buddhiraju Anirudh, Shimizu Michelle Riyo, Chen Tony Lin-Wei, Seo Henry Hojoon, Bacevich Blake M, Xiao Pengwei, Kwon Young-Min
Bioengineering Laboratory, Department of Orthopedic Surgery, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, Boston, MA, 02114, USA.
Knee Surg Relat Res. 2025 Jan 13;37(1):3. doi: 10.1186/s43019-024-00256-z.
Unplanned readmission, a measure of surgical quality, occurs after 4.8% of primary total knee arthroplasties (TKA). Although the prediction of individualized readmission risk may inform appropriate preoperative interventions, current predictive models, such as the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) surgical risk calculator (SRC), have limited utility. This study aims to compare the predictive accuracy of the SRC with a novel artificial neural network (ANN) algorithm for 30-day readmission after primary TKA, using the same set of clinical variables from a large national database.
Patients undergoing primary TKA between 2013 and 2020 were identified from the ACS-NSQIP database and randomly stratified into training and validation cohorts. The ANN was developed using data from the training cohort with fivefold cross-validation performed five times. ANN and SRC performance were subsequently evaluated in the distinct validation cohort, and predictive performance was compared on the basis of discrimination, calibration, accuracy, and clinical utility.
The overall cohort consisted of 365,394 patients (training = 362,559; validation = 2835), with 11,392 (3.1%) readmitted within 30 days. While the ANN demonstrated good discrimination and calibration (area under the curve (AUC) = 0.72, slope = 1.32, intercept = -0.09) in the validation cohort, the SRC demonstrated poor discrimination (AUC = 0.55) and underestimated readmission risk (slope = -0.21, intercept = 0.04). Although both models possessed similar accuracy (Brier score: ANN = 0.03; SRC = 0.02), only the ANN demonstrated a higher net benefit than intervening in all or no patients on the decision curve analysis. The strongest predictors of readmission were body mass index (> 33.5 kg/m), age (> 69 years), and male sex.
This study demonstrates the superior predictive ability and potential clinical utility of the ANN over the conventional SRC when constrained to the same variables. By identifying the most important predictors of readmission following TKA, our findings may assist in the development of novel clinical decision support tools, potentially improving preoperative counseling and postoperative monitoring practices in at-risk patients.
计划外再入院是衡量手术质量的一项指标，在初次全膝关节置换术（TKA）后发生率为4.8%。尽管预测个体再入院风险可能有助于进行适当的术前干预，但目前的预测模型，如美国外科医师学会国家外科质量改进计划（ACS-NSQIP）手术风险计算器（SRC），效用有限。本研究旨在使用来自大型国家数据库的同一组临床变量，比较SRC与一种新型人工神经网络（ANN）算法对初次TKA后30天再入院的预测准确性。
从ACS-NSQIP数据库中识别出2013年至2020年间接受初次TKA的患者，并随机分层为训练队列和验证队列。使用训练队列的数据开发ANN，并进行了五次五折交叉验证。随后在不同的验证队列中评估ANN和SRC的性能，并基于区分度、校准度、准确性和临床效用比较预测性能。
整个队列包括365394名患者（训练组=362559名；验证组=2835名），其中11392名（3.1%）在30天内再次入院。在验证队列中，ANN表现出良好的区分度和校准度（曲线下面积（AUC）=0.72，斜率=1.32，截距=-0.09），而SRC表现出较差的区分度（AUC=0.55）且低估了再入院风险（斜率=-0.21，截距=0.04）。尽管两种模型的准确性相似（Brier评分：ANN=0.03；SRC=0.02），但在决策曲线分析中，只有ANN显示出比干预所有患者或不干预任何患者更高的净效益。再入院的最强预测因素是体重指数（>33.5kg/m²）、年龄（>69岁）和男性。
本研究表明，在使用相同变量时，ANN比传统的SRC具有更高的预测能力和潜在的临床效用。通过识别TKA后再入院的最重要预测因素，我们的研究结果可能有助于开发新型临床决策支持工具，潜在地改善高危患者的术前咨询和术后监测实践。
