医学
冠状面
矢状面
最小临床重要差异
正式舞会
外翻
口腔正畸科
外科
放射科
随机对照试验
产科
作者
Hassan Farooq,Evan R. Deckard,Justin Carlson,Nathan Ghattas,R. Michael Meneghini
标识
DOI:10.1016/j.arth.2023.04.040
摘要
Advanced technologies, like robotics, provide enhanced precision for implanting total knee arthroplasty components; however, optimal component position and limb alignment remain unknown. This study sought to identify sagittal and coronal alignment targets that correlate with minimal clinically important differences (MCIDs) in patient-reported outcome measures (PROMs).A total of 1,311 consecutive total knee arthroplasties were retrospectively reviewed. Posterior tibial slope (PTS), femoral flexion (FF), and tibio-femoral alignment (TFA) were measured radiographically. Patients were grouped based on whether they achieved multiple MCIDs for PROM scores. Classification and regression tree machine learning models were utilized to identify optimal alignment zones. The mean follow-up was 2.4 years (range, 1 to 11).The change in PTS and postoperative TFA were most predictive for achieving MCIDs in 90% of the models. Approximating native PTS within 4° correlated with MCID achievement and superior PROMs. Preoperative varus and neutral aligned knees were more likely to meet MCIDs and superior PROM scores when not overcorrected into valgus postoperatively (≥7°). Preoperative valgus-aligned knees correlated with MCID achievement when postoperative TFA was not overcorrected into substantial varus (<0°). Albeit less impactful, FF ≤ 7° correlated with MCID achievement and superior PROMs regardless of preoperative alignment. Sagittal and coronal alignment measurements had moderate to strong interactions in 13 of 20 models.Optimized PROM MCIDs correlated with approximating native PTS while maintaining similar preoperative TFA and incorporating moderate FF. Study findings demonstrate interactions between sagittal and coronal alignment which may optimize PROMs, highlighting the importance of three-dimensional implant alignment targets.III.
科研通智能强力驱动
Strongly Powered by AbleSci AI