Dias Correia Fernando, Nogueira André, Magalhães Ivo, Guimarães Joana, Moreira Maria, Barradas Isabel, Molinos Maria, Teixeira Laetitia, Pires Joaquim, Seabra Rosmaninho, Lains Jorge, Bento Virgílio
Neurology Department, Hospital de Santo António-Centro Hospitalar do Porto, Porto, Portugal.
SWORD Health, Porto, Portugal.
JMIR Rehabil Assist Technol. 2019 Jun 21;6(1):e14523. doi: 10.2196/14523.
The demand for total hip arthroplasty (THA) is rising. In the face of rapidly increasing health care costs, ensuring widespread, cost-effective rehabilitation is a priority. Technologies allowing independent home-based rehabilitation may be the key to facilitate access, improve effectiveness, and lower costs of care.
The aim of this study was to assess the feasibility of a novel artificial intelligence-powered digital biofeedback system following THA and compare the clinical outcomes against supervised conventional rehabilitation.
This was a single-center, parallel-group pilot study, with an 8-week intervention program. Patients were assessed at baseline, during the program (at 4 and 8 weeks), and 3 and 6 months after surgery. The primary outcome was the Timed Up and Go (TUG) score and secondary outcomes were the Hip dysfunction and Osteoarthritis Outcome Scale (HOOS; a patient-reported outcome) and hip range of motion (ROM).
A total of 66 patients were included: 35 digital physiotherapy (PT) versus 31 conventional. There were no differences at baseline between groups except for lower HOOS quality of life (QoL) subscale scores in the digital PT group. Clinically relevant improvements were noted in both groups at all time points. The digital PT group showed a retention rate of 86% (30/35). Per-protocol analysis revealed a superiority of the digital PT group for all outcome measures. Intention-to-treat analysis revealed the superiority of the digital PT group at all time points for TUG (change between baseline and 4 and 8 weeks: P<.001; change between baseline and 3 and 6 months: P=.001 and P=.005, respectively), with a difference between median changes of -4.79 seconds (95% CI -7.24 to -1.71) at 6 months post-THA. Between baseline and month 6, results were also superior in the digital PT group for the HOOS sports and QoL subscales and all ROM except for standing flexion.
This study demonstrates this novel solution holds promise in rehabilitation after THA, ensuring better clinical outcomes than conventional rehabilitation while reducing dependence on human resources.
ClinicalTrials.gov NCT03045549; https://clinicaltrials.gov/ct2/show/NCT03045549.
全髋关节置换术（THA）的需求正在上升。面对快速增长的医疗保健成本，确保广泛且具有成本效益的康复是当务之急。允许独立在家进行康复的技术可能是促进康复机会、提高康复效果并降低护理成本的关键。
本研究旨在评估一种新型人工智能驱动的数字生物反馈系统在全髋关节置换术后的可行性，并将临床结果与有监督的传统康复进行比较。
这是一项单中心、平行组的试点研究，有一个为期8周的干预计划。在基线、计划期间（第4周和第8周）以及术后3个月和6个月对患者进行评估。主要结果是计时起立行走（TUG）评分，次要结果是髋关节功能障碍和骨关节炎结果量表（HOOS；患者报告的结果）以及髋关节活动范围（ROM）。
共纳入66例患者：35例接受数字物理治疗（PT），31例接受传统治疗。除数字PT组的HOOS生活质量（QoL）子量表得分较低外，两组在基线时无差异。两组在所有时间点均观察到具有临床意义的改善。数字PT组的保留率为86%（30/35）。符合方案分析显示数字PT组在所有结局指标上均具有优势。意向性分析显示数字PT组在所有时间点的TUG评分上均具有优势（基线与第4周和第8周之间的变化：P<.001；基线与术后3个月和6个月之间的变化：分别为P=.001和P=.005），在全髋关节置换术后6个月时，中位数变化差异为-4.79秒（95%CI -7.24至-1.71）。在基线至第6个月期间，数字PT组在HOOS运动和QoL子量表以及除站立位屈曲外的所有ROM方面的结果也更优。
本研究表明，这种新方法在全髋关节置换术后的康复中具有前景，可确保比传统康复获得更好的临床结果，同时减少对人力资源的依赖。
ClinicalTrials.gov NCT03045549；https://clinicaltrials.gov/ct2/show/NCT03045549 。
