Rush University Medical Center, Chicago, IL.
J Arthroplasty. 2019 Jul;34(7):1369-1373. doi: 10.1016/j.arth.2019.02.048. Epub 2019 Mar 7.
Preoperative planning for total knee arthroplasty (TKA) is essential for streamlining operating room efficiency and reducing costs. Digital templating and patient-specific instrumentation have shown some value in TKA but require additional costs and resources. The purpose of this study was to validate a previously published algorithm that uses only demographic variables to accurately predict TKA tibial and femoral component sizes.
Four hundred seventy-four consecutive patients undergoing elective primary TKA were prospectively enrolled. Four surgeons were included, three of which were unaffiliated with the retrospective cohort study. Patient sex, height, and weight were entered into our published Arthroplasty Size Prediction mobile application. Accuracy of the algorithm was compared with the actual sizes of the implanted femoral and tibial components from 5 different implant systems. Multivariate regression analysis was used to identify independent risk factors for inaccurate outliers for our model.
When assessing accuracy to within ±1 size, the accuracies of tibial and femoral components were 87% (412/474) and 76% (360/474). When assessing accuracy to within ±2 sizes of predicted, the tibial accuracy was 97% (461/474), and the femoral accuracy was 95% (450/474). Risk factors for the actual components falling outside of 2 predicted sizes include weight less than 70 kg (odds ratio = 2.47, 95% confidence interval [1.21-5.06], P = .01) and use of an implant system with <2.5 mm incremental changes between femoral sizes (odds ratio = 5.50, 95% confidence interval [3.33-9.11], P < .001).
This prospective series of patients validates a simple algorithm to predict component sizing for TKA with high accuracy based on demographic variables alone. Surgeons can use this algorithm to simplify the preoperative planning process by reducing unnecessary trays, trials, and implant storage, particularly in the community or outpatient setting where resources are limited. Further assessment of components with less than 2.5-mm differences between femoral sizes is required in the future to make this algorithm more applicable worldwide.
全膝关节置换术（TKA）的术前规划对于简化手术室效率和降低成本至关重要。数字模板和患者特异性器械在 TKA 中显示出一定的价值，但需要额外的成本和资源。本研究的目的是验证先前发表的一种算法，该算法仅使用人口统计学变量准确预测 TKA 胫骨和股骨组件的大小。
前瞻性纳入 474 例接受择期初次 TKA 的连续患者。纳入 4 名外科医生，其中 3 名与回顾性队列研究无关。患者的性别、身高和体重被输入我们发表的关节置换尺寸预测移动应用程序。算法的准确性与来自 5 种不同植入物系统的植入股骨和胫骨组件的实际尺寸进行比较。使用多元回归分析确定我们模型中不准确离群值的独立危险因素。
当评估±1 尺寸内的准确性时，胫骨和股骨组件的准确性分别为 87%（412/474）和 76%（360/474）。当评估预测±2 尺寸内的准确性时，胫骨的准确性为 97%（461/474），股骨的准确性为 95%（450/474）。实际组件尺寸超出 2 个预测尺寸的危险因素包括体重小于 70 公斤（优势比=2.47，95%置信区间[1.21-5.06]，P=0.01）和使用股骨尺寸之间增量变化小于 2.5 毫米的植入物系统（优势比=5.50，95%置信区间[3.33-9.11]，P<0.001）。
本前瞻性患者系列验证了一种简单的算法，该算法基于人口统计学变量单独预测 TKA 组件的尺寸，具有很高的准确性。外科医生可以使用该算法通过减少不必要的托盘、试验和植入物存储来简化术前规划过程，特别是在资源有限的社区或门诊环境中。未来需要进一步评估股骨尺寸之间差异小于 2.5 毫米的组件，以使该算法在全球范围内更具适用性。
