文献速递关节外科

计算机辅助导航膝关节置换术研究质量评估

Quality Assessment of Computer-Assisted Navigation Knee Arthroplasty Studies.
PubMed 文献 · J Arthroplasty · 2025-06-06

本系统综述检索了PubMed、Ovid(Embase)和MEDLINE(2024年4月),纳入119项比较计算机导航与传统全膝关节置换术的随机对照试验,使用改良Detsky指数和改良Cochrane偏倚风险工具评估方法学质量。结果显示报告质量较高(平均改良Detsky评分92%,98%为高质量),但偏倚风险评分平均6.2,仅27%为高质量,分配隐藏、盲法和偏倚控制存在不足。

要点

  • 119项随机对照试验中,78%为单中心研究,42%来自亚洲,6.7%来自北美。
  • 平均改良Detsky评分为92%,98%的研究为高质量,评分随时间改善(P=0.003)。
  • 平均改良Cochrane偏倚风险评分为6.2,仅27%为高质量,分配隐藏和盲法存在缺陷。

核心结论

  • 随机对照试验报告标准有所改善,但方法学弱点持续存在,未来试验应强调稳健的随机化、盲法和足够的样本量。

研究设计与样本

  • 系统综述,检索PubMed、Ovid(Embase)和MEDLINE(2024年4月),纳入119项比较计算机导航与传统全膝关节置换术的随机对照试验。

主要发现

  • 119项随机对照试验中6.7%来自北美,42%来自亚洲,单中心试验占78%。
  • 平均改良Detsky评分为92%,98%的研究为高质量;评分随时间改善(P=0.003),多中心研究评分更高(P=0.034),不同试验地点(P=0.24)或主要结局(P=0.19)间无差异。
  • 平均改良Cochrane偏倚风险评分为6.2,27%为高质量;评分随时间改善(P=0.027)且因地点而异(P<0.001),澳大利亚试验评分最高,欧洲试验评分最低;单中心与多中心试验间(P=0.067)或不同主要结局间(P=0.36)无差异。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: Computer-navigated total knee arthroplasty (TKA) seeks to improve component alignment and reduce postoperative complications versus conventional TKA. However, the quality of studies evaluating outcomes varies. This review systematically assesses the methodological quality of randomized controlled trials (RCTs) comparing computer-navigated to conventional TKA using established quality assessment tools. METHODS: A systematic review was conducted by searching PubMed, Ovid (Embase), and MEDLINE in April 2024 using the terms "Computer Navigation and Knee Arthroplasty" or "Computer Assisted and Knee Arthroplasty." A total of 119 RCTs met the inclusion criteria. Trial quality was evaluated using the transformed Detsky index, which assesses reporting completeness, and the modified Cochrane risk-of-bias tool, which evaluates methodological rigor and biases. RESULTS: Of the 119 RCTs, 6.7% were from North America and 42% were from Asia. Single-center trials comprised 78%. The mean transformed Detsky score was 92%, with 98% of studies classified as high quality, reflecting strong adherence to reporting guidelines. Scores improved over time (P = 0.003) and were higher in multicenter studies (P = 0.034), with no differences by trial location (P = 0.24) or primary outcome (P = 0.19). However, the mean modified Cochrane risk-of-bias score was 6.2, with 27% rated as high quality, indicating deficiencies in allocation concealment, blinding, and bias mitigation. Modified Cochrane risk-of-bias scores improved over time (P = 0.027) and varied by location (P < 0.001), with the Australian trials scoring highest and European trials scoring lowest. There were no differences found between single- and multicenter trials (P = 0.067) or among primary outcomes (P = 0.36). CONCLUSIONS: While RCT-reported standards have improved, methodological weaknesses persist. Future trials should emphasize robust randomization, blinding, and adequate sample sizes to strengthen validity. Multicenter collaborations, standardized reporting, and increased transparency in industry-funded research are essential for improving quality. Addressing these limitations will ensure reliable evidence for TKA decision-making.

来源:J Arthroplasty,查看原文。版权归原作者所有。

引用

  1. Spooner A, Helton AG, Lee V, et al. Quality Assessment of Computer-Assisted Navigation Knee Arthroplasty Studies. J Arthroplasty. 2025. PMID: 40484052. doi:10.1016/j.arth.2025.05.108

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。