标准版与左右侧版国际膝关节文献委员会主观膝关节表单评分的比较
这项队列研究(证据等级2)纳入100例患者,采用交叉设计比较标准版IKDC-SKF与左右侧版IKDC-SKF。标准版与患侧R/L评分平均差异1.4(P=.008;95%CI 0.4-2.4),但95%CI低于预设临床相关阈值5;两版本高度相关(rho=0.95),R/L版能持续区分损伤膝与未损伤膝(P<.001),55例偏好R/L版。
要点
- 标准版与患侧R/L评分平均差异1.4,95%CI 0.4-2.4,未超过临床相关阈值5。
- 标准版与患侧R/L评分高度相关,rho=0.95。
- R/L版能持续区分损伤膝与未损伤膝(P<.001),55例患者偏好R/L版。
核心结论
- 标准版IKDC-SKF与R/L版IKDC-SKF患侧评分之间未观察到临床相关差异,因此对有症状的单侧膝关节诊断,两种表单均可使用。
- R/L版IKDC-SKF在患侧与健侧之间显示出持续且临床相关的差异。
- 本研究中患者偏好R/L版IKDC-SKF。
研究设计与样本
- 队列研究(诊断性),证据等级2。
- 纳入100例患者,采用交叉研究设计,按简单随机化分为2组。
- 一组先完成标准版IKDC-SKF再完成R/L版IKDC-SKF,另一组顺序相反。
- 两次问卷之间设10分钟洗脱期,问卷后调查询问患者偏好表单。
- 预先设定5分作为临床相关阈值,该值低于已发表的标准版IKDC-SKF最小可检测变化和最小临床重要差异指标(范围为6.3至20.5)。
主要发现
- 标准版IKDC-SKF与R/L版IKDC-SKF患侧评分配对t检验显示平均差异1.4(P=.008;95%CI 0.4-2.4),差异有统计学意义,但95%CI低于临床相关阈值5。
- 标准版与R/L版患侧膝关节评分高度相关,rho=0.95。
- 患者通过R/L版IKDC-SKF能在一定评分范围内持续区分损伤膝与未损伤膝(P<.001)。
- 问卷后调查显示55例偏好R/L版IKDC-SKF,30例偏好标准版IKDC-SKF,15例无偏好。
- 事后分析显示对「R/L版 vs 标准版IKDC-SKF」的偏好有统计学意义(P=.014),对「R/L版IKDC-SKF vs 无偏好」的偏好有统计学意义(P<.003)。
关键数值与推荐
- 预设临床相关阈值为5分,低于标准版IKDC-SKF已发表的最小可检测变化和最小临床重要差异指标范围6.3至20.5。
- 标准版与患侧R/L版评分平均差异1.4(P=.008;95%CI 0.4-2.4)。
- 标准版与患侧R/L版评分相关系数rho=0.95。
- R/L版区分损伤膝与未损伤膝的P<.001。
- 偏好分布:R/L版55例,标准版30例,无偏好15例;事后分析P=.014和P<.003。
临床怎么用
- 作者认为,对有症状的单侧膝关节诊断,标准版IKDC-SKF与R/L版IKDC-SKF患侧评分均可使用。
- 本研究中患者偏好R/L版IKDC-SKF。
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: The International Knee Documentation Committee Subjective Knee Form (IKDC-SKF) is a validated patient-reported outcome used in clinical research. No studies exist directly comparing the standard unilateral adult version (IKDC-SKF) with the bilateral adult version (R/L IKDC-SKF). HYPOTHESES: The first hypothesis is that no clinically relevant difference would be observed between standard IKDC-SKF scores and involved R/L scores. The second hypothesis is that a relevant difference would be observed between involved and uninvolved scores on the R/L IKDC-SKF. STUDY DESIGN: Cohort study (Diagnosis); Level of evidence, 2. METHODS: One hundred patients were enrolled via a crossover study design. Participants were split into 2 groups per simple randomization. One group completed the standard IKDC-SKF first and the R/L IKDC-SKF second. The other group completed forms in the reverse order. A 10-minute washout period was administered between questionnaires. Participants reported their preferred form in a postquestionnaire survey. A 5-point threshold for clinical relevance was set a priori, which is less than multiple published minimal detectable change and minimal clinically important difference metrics of the standard IKDC-SKF, ranging from 6.3 to 20.5. Data were analyzed with Bland-Altman plots, paired t tests, correlations, and chi-square tests. RESULTS: Paired t tests between the standard IKDC-SKF and the involved R/L IKDC-SKF scores demonstrated a statistically significant mean difference of 1.4 ( P = .008; 95% CI, 0.4-2.4). However, the 95% CI falls under the clinically relevant threshold of 5. Standard and involved knee scores from the R/L IKDC-SKF were highly correlated, with a rho of 0.95. Patients consistently distinguished the injured knee from the uninjured knee across a range of scores via the R/L IKDC-SKF ( P < .001). The postquestionnaire survey showed that 55 patients preferred the R/L IKDC-SKF, 30 preferred the standard IKDC-SKF, and 15 had no preference. Post hoc analysis showed a significant preference for "R/L vs standard IKDC-SKF" ( P = .014) and "R/L IKDC-SKF vs no preference" ( P < .003). CONCLUSION: No clinically relevant difference was observed between the standard IKDC-SKF and the involved knee score of the R/L IKDC-SKF. Therefore, for symptomatic unilateral knee diagnosis, either form can be used. The R/L IKDC-SKF showed a consistent and clinically relevant difference between involved and uninvolved knees. Patients in this study preferred the R/L IKDC-SKF.
引用
- Magnuson JA, Strnad G, Smith C, et al. Comparison of Standard and Right/Left International Knee Documentation Committee Subjective Knee Form Scores. Am J Sports Med. 2019. PMID: 30896976. doi:10.1177/0363546519829770
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。