文献速递疝与腹壁外科围术期管理与 ERAS

腹股沟疝修补术文献中统计学显著性与临床相关性的评估

Cutting through the p-value: evaluating clinical relevance in surgical literature analyzing the approaches for inguinal hernia repair.
PubMed 文献 · Surg Endosc · 2025-09-24

该研究检索了2018年以来六本外科期刊中比较开放、腹腔镜与机器人腹股沟疝修补术的文献,共纳入54篇(8篇RCT、46篇队列研究)。结果显示无一项研究定义临床相关性界值,仅6篇(11.1%)提及结果可能不具临床意义,但其中3篇仍仅凭P值<0.05提示某术式优越。

要点

  • 纳入的54篇疝修补比较研究中,无一定义临床相关性界值。
  • 仅6篇(11.1%)提及结果可能不具临床意义,其中半数仍仅凭P值推荐术式。
  • 16篇无统计学显著差异的研究中,2篇仍提示某术式优越。

核心结论

  • 疝外科文献中统计学显著性与临床相关性的区分仍处理不足,即使RCT亦如此。
  • 需定义复发、伤口相关并发症与术后疼痛等关键结局达到临床相关所需的关联指标差值。

研究设计与样本

  • 检索Hernia、Surgical Endoscopy、Annals of Surgery、Surgery、World Journal of Surgery与JAMA Surgery自2018年起的文献。
  • 纳入比较开放、腹腔镜与机器人腹股沟疝修补术的研究,排除成本效益等非临床结局分析。
  • 由两名作者独立筛选并评估临床相关性界值及是否仅凭P值提示术式优越。
  • 初检62篇,排除8篇,最终纳入54篇(RCT 8篇占14.8%,比较性队列研究46篇占85.2%)。

主要发现

  • 54篇纳入研究中无一项为所分析结局定义临床相关性界值。
  • 仅6篇(11.1%)研究指出其发现可能不具临床相关性。
  • 该6篇中3篇(50%)仍仅凭P值提示某术式优越,另3篇(50%)虽见统计学显著差异但未作此提示。
  • 16篇(29.6%)研究组间无统计学显著差异,但其中2篇(12.5%)仍提示某一术式优越。
  • RCT中仅1篇(12.5%)报告结果可能不具临床相关性,1篇(12.5%)在无统计学显著结果下仍提示获益。

临床怎么用

  • 阅读疝修补比较研究时,应区分P值<0.05与临床相关差异,避免仅凭统计学显著性改变术式选择。
  • 关注复发、伤口相关并发症与术后疼痛等关键结局是否有临床相关性界值支撑。

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 introduction of evidence-based medicine has challenged many concepts. In analyzing comparative study results, it is common to find narratives highlighting favorable outcomes based on a p-value of < 0.05, without understanding the clinical impact of the observed difference. Given the prevalence of this issue in hernia surgery research, we aimed to evaluate the prevalence of studies reporting a cutoff for clinical relevance in published comparisons of open, laparoscopic, and robotic inguinal hernia repair (IHR). METHODS: We searched Hernia, Surgical Endoscopy, Annals of Surgery, Surgery, World Journal of Surgery, and JAMA Surgery for articles comparing open, laparoscopic, and robotic IHR. Our search was performed according to the recent guidelines, comprising articles published since 2018. Articles analyzing non-clinical outcomes, such as cost-effectiveness, were excluded. Two authors independently screened the articles analyzing the presence of a clinical relevance cutoff definition of statistical significance, and if it suggested a superiority of a technique among others based solely on the p-value. RESULTS: The initial search resulted in 62 articles, of which 8 were excluded, resulting in 54 included manuscripts. Among the included studies, 8 (14.8%) were randomized controlled trials (RCTs), while 46 (85.2%) were comparative cohort studies. Surprisingly, none of the studies defined a clinical relevance cutoff for the outcomes analyzed. Furthermore, only 6 (11.1%) studies highlighted that their findings may not be of clinical relevance. However, even among those 6 studies, 3 (50%) suggested a superiority of the approach based solely on the p-value, while the other 3 (50%) studies, despite finding a statistically significant difference, did not make this suggestion. 16 (29.6%) studies showed no statistically significant differences between the groups, but 2 (12.5%) of those still suggested a superiority of one of the surgical approaches. Among the RCTs, only 1 (12.5%) reported that their findings may not be of clinical relevance, while 1 (12.5%) suggested a benefit despite not finding statistically significant results. CONCLUSION: Our study, encompassing the main journals in the surgical literature, demonstrated that the distinction between statistical and clinical relevance in hernia surgery, even in RCTs, remains inadequately addressed. There is a need for studies to define what difference in association measure is necessary to achieve clinical relevance for key outcomes in hernia surgery, such as recurrence, wound-related morbidity, and postoperative pain.

来源:Surg Endosc,查看原文。版权归原作者所有。

引用

  1. Balthazar da Silveira CA, Rasador ACD, Nogueira R, et al. Cutting through the p-value: evaluating clinical relevance in surgical literature analyzing the approaches for inguinal hernia repair. Surg Endosc. 2025. PMID: 40991045. doi:10.1007/s00464-025-12213-2

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。