老年无并发症胆总管结石:观察与ERCP的目标试验模拟比较
该研究采用目标试验模拟,纳入TriNetX中2015至2023年诊断胆总管结石的75岁以上患者,倾向评分匹配后比较早期ERCP与观察。3年复合主要不良事件或12个月全因死亡在ERCP组为30.7%、观察组为22.4%,校正后无显著差异(aHR 1.13,95%CI 0.99-1.28);胜率比支持观察(0.71,95%CI 0.64-0.79)。结论支持对老年无并发症胆总管结石采取选择性「按需」ERCP策略。
要点
- 老年无并发症胆总管结石早期ERCP未改善复合结局(aHR 1.13,95%CI 0.99-1.28)。
- 胜率比支持观察优于早期ERCP(0.71,95%CI 0.64-0.79)。
- 敏感性分析中ERCP与更多主要不良事件相关,支持按需ERCP。
核心结论
- 在老年无并发症胆总管结石患者中,早期ERCP与观察相比未带来结局改善,支持选择性「按需」ERCP策略。
研究设计与样本
- 目标试验模拟,使用TriNetX健康记录数据,纳入2015至2023年诊断胆总管结石的75岁以上患者。
- ERCP组在Time Zero后60天内接受ERCP,观察组未接受,采用倾向评分匹配构建队列。
- 主要结局为3年内主要不良事件复合终点或12个月内全因死亡,采用校正Cox回归和胜率比分析。
- 敏感性分析采用30天ERCP窗口和仅门诊诊断患者。
主要发现
- 匹配后共5856例患者(每组2928例),复合结局在ERCP组为30.7%、观察组为22.4%,校正后无显著差异(aHR 1.13;95%CI 0.99-1.28)。
- 胜率比支持观察(0.71;95%CI 0.64-0.79)。
- 30天ERCP窗口敏感性分析中ERCP与更高主要不良事件相关(aHR 1.14;95%CI 1.00-1.31)。
- 仅门诊诊断敏感性分析中ERCP与更高主要不良事件相关(aHR 1.61;95%CI 1.13-2.29)。
- 两组死亡率无显著差异。
临床怎么用
- 对75岁以上无并发症胆总管结石患者,可考虑先观察、按需再行ERCP,而非一律早期ERCP。
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 & AIMS: The management of uncomplicated choledocholithiasis in elderly patients remains controversial due to uncertain clinical benefits of intervention and higher complication rates with endoscopic retrograde cholangiopancreatography (ERCP). We emulated a target trial comparing early ERCP vs observation in elderly patients with uncomplicated choledocholithiasis. METHODS: Target trial emulation using health record data from TriNetX. Patients >75 years with an incident choledocholithiasis diagnosis between 2015 and 2023 were included. The ERCP group underwent ERCP within 60 days of Time Zero; the observation group did not. Cohorts were constructed with propensity score matching. The primary outcome was a composite of major adverse events within 3 years or all-cause mortality within 12 months. Adjusted Cox regression was performed to compare the survival outcome between groups. Win ratio was calculated for reporting the hierarchical composite outcome. Sensitivity analyses were conducted with a 30-day ERCP window and patients diagnosed at ambulatory encounter only. RESULTS: Among 5856 matched patients (2928 per group), the composite outcome occurred in 30.7% of ERCP vs 22.4% of observation patients; however, there was no significant difference (adjusted hazard ratio [aHR], 1.13; 95% confidence interval [CI], 0.99-1.28). The win ratio favored observation (0.71; 95% CI, 0.64-0.79). In both sensitivity analyses, ERCP was associated with higher major adverse events (30-day ERCP window: aHR, 1.14; 95% CI, 1.00-1.31; ambulatory-only: aHR, 1.61; 95% CI, 1.13-2.29). Mortality did not differ significantly. CONCLUSIONS: In elderly patients with uncomplicated choledocholithiasis, early ERCP was not associated with improved outcomes. Findings support a selective, 'as-needed' approach to ERCP in this population.
引用
- Smith ZL, Amjad T, Hoard M, et al. Uncomplicated Choledocholithiasis in Elderly Patients: A Target Trial Emulation Comparing Observation and Endoscopic Retrograde Cholangiopancreatography. Clin Gastroenterol Hepatol. 2026. PMID: 41713828. doi:10.1016/j.cgh.2026.02.010
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。