全膝关节置换术中止血带使用的脆弱性:随机对照试验的系统综述
该系统综述检索了PubMed、EMBASE和MEDLINE,纳入23项比较全膝关节置换术中是否使用止血带的随机对照试验,共报告91个二分类结局。总体中位脆弱指数为4,中位脆弱商为0.0476;作者建议纳入脆弱性分析以提高研究结论解释的可靠性。
要点
- 23项随机对照试验共91个二分类结局,总体中位脆弱指数为4(IQR 3至6)。
- 总体中位脆弱商为0.0476(IQR 0.0291至0.0867)。
- 17.6%的结局中失访例数大于中位脆弱指数。
核心结论
- 改变少量结局事件通常足以逆转评估全膝关节置换术中止血带使用的随机对照试验的结论。
- 作者建议纳入脆弱性分析以提高研究结论解释的可靠性。
研究设计与样本
- 系统综述,检索PubMed、EMBASE和MEDLINE,纳入23项比较全膝关节置换术中止血带使用与否的随机对照试验,共报告91个二分类结局。
主要发现
- 总体中位脆弱指数为4(IQR 3至6),总体中位脆弱商为0.0476(IQR 0.0291至0.0867)。
- 11个统计学显著结局的中位脆弱指数为2(IQR 1.5至5),中位脆弱商为0.0200(IQR 0.0148至0.0484)。
- 80个无统计学显著结局的中位反向脆弱指数为4(IQR 3至6)。
- 17.6%的结局中失访例数大于中位脆弱指数。
关键数值与推荐
- 脆弱指数和反向脆弱指数分别定义为使显著性结果和不显著结果改变统计学显著性所需的结局事件逆转数。
- 脆弱商为脆弱指数或反向脆弱指数除以样本量。
临床怎么用
- 原文未据此给出临床推荐。
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: Physicians utilize P-values to interpret clinical trial data and guide patient-care decisions. Fragility analysis assesses the stability of statistical findings in relation to outcome event reversals. This study assessed the statistical fragility of recent randomized controlled trials (RCTs) investigating tourniquet use in total knee arthroplasty (TKA). METHODS: We queried PubMed, EMBASE, and MEDLINE for RCTs comparing outcomes in TKA based on tourniquet use. Fragility index (FI) and reverse fragility index (reverse FI) were calculated - for significant and nonsignificant outcomes, respectively - as the number of outcome reversals required to change statistical significance. The fragility quotient (FQ) was calculated by dividing the FI or reverse FI by the sample size. Median overall FI and FQ were calculated for all included outcomes, and sub-analyses were performed by reported significance. The literature search yielded 23 studies reporting 91 total dichotomous outcomes. RESULTS: Overall median FI was 4 with an interquartile range (IQR) of 3 to 6. Overall median FQ was 0.0476 (IQR 0.0291 to 0.0867). A total of 11 outcomes were statistically significant with a median FI and FQ of 2 (IQR 1.5 to 5) and 0.0200 (IQR 0.0148 to 0.0484), respectively. There were 80 outcomes that were nonsignificant with a median reverse FI of 4 (IQR 3 to 6). Loss to follow-up was greater than the median FI in 17.6% of outcomes. CONCLUSION: Altering a small number of outcomes is often sufficient to reverse findings in RCTs evaluating tourniquet use in TKA. We recommend including fragility analyses to increase reliability in the interpretation of study conclusions.
引用
- Cordero JK, Lawrence KW, Brown AN, et al. The Fragility of Tourniquet Use in Total Knee Arthroplasty: A Systematic Review of Randomized Controlled Trials. J Arthroplasty. 2022. PMID: 36566999. doi:10.1016/j.arth.2022.12.035
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。