Cole S, Fabiano G, Barea C, Cullati S, Agoritsas T, Gutacker N, Silman A J, Hannouche D, Lübbeke A, Pinedo-Villanueva Rafael
Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, UK.
Division of Orthopaedics & Trauma Surgery, Geneva University Hospitals and University of Geneva, Geneva, Switzerland.
BMC Musculoskelet Disord. 2025 Aug 20;26(1):804. doi: 10.1186/s12891-025-09024-w.
Patients receiving total hip arthroplasty (THA) have different expectations and concerns about their health outcomes after surgery. In this study we developed a tool based on registry data to inform patients and their clinicians about activity outcomes after THA.
We used data from the Geneva Arthroplasty Registry (GAR) on patients receiving a primary elective THA between 1996 and 2019. The information tool was developed around five activity outcomes: getting in/out of the car, getting dressed autonomously, independence in weekly tasks, interference in social activities, and activity levels. Based on baseline predictors, conditional inference trees (CITs) were used to create clusters of patients with homogeneous activity outcomes at one, five and 10 years after surgery, rather than to predict individual probabilities.
In total, 14 CITs were generated based on 6,836 operations included in the tool. Overall, activity outcomes substantially improved at all three times points after surgery, with 1-year values mostly being the highest. While before surgery only about 10% of patients had none/slight limitations in activities of daily living, about 70% did one year after surgery. The SF12 mental component score (MCS), SF12 self-rated health (SRH), BMI, ASA score, and comorbidity count were the most recurring predictors of activity outcomes. Predictors and their relative importance changed at different time points for the same outcome. For example, for ability to get in/out the car, whilst clusters at year 1 were generated based on WOMAC function, SRH, mental health, WOMAC difficulty walking, and SF12 physical interference, at year 5, ASA score, BMI, SF12 physical & mental health, activity level, and socio-economic status were significant. Outcome profiles varied by clusters.
Distinct activity outcomes clusters based on baseline patient characteristics were identified and knowing this can help inform patients' expectation and meaningful discussions with clinicians about treatment decisions.
接受全髋关节置换术（THA）的患者对术后健康状况有不同的期望和担忧。在本研究中，我们基于登记数据开发了一种工具，用于告知患者及其临床医生THA术后的活动结果。
我们使用了日内瓦关节置换登记处（GAR）1996年至2019年间接受初次择期THA患者的数据。该信息工具围绕五个活动结果展开：上下车、自主穿衣、每周任务的独立性、对社交活动的干扰以及活动水平。基于基线预测因素，使用条件推断树（CITs）在术后1年、5年和10年创建具有同质活动结果的患者聚类，而非预测个体概率。
该工具共纳入6836例手术，共生成14个CITs。总体而言，术后所有三个时间点的活动结果均有显著改善，1年时的值大多最高。术前只有约10%的患者日常生活活动无/轻度受限，术后1年约70%的患者如此。SF12精神成分评分（MCS）、SF12自评健康（SRH）、体重指数（BMI）、美国麻醉医师协会（ASA）评分和合并症数量是活动结果最常见的预测因素。对于相同的结果，预测因素及其相对重要性在不同时间点有所变化。例如，对于上下车能力，1年时的聚类基于西部安大略和麦克马斯特大学骨关节炎指数（WOMAC）功能、SRH、心理健康、WOMAC行走困难和SF12身体干扰生成，而在5年时，ASA评分、BMI、SF12身心健康、活动水平和社会经济地位具有显著性。不同聚类的结果概况各不相同。
基于患者基线特征识别出了不同的活动结果聚类，了解这一点有助于告知患者期望，并与临床医生就治疗决策进行有意义的讨论。
