临床试验：胶囊海绵作为嗜酸性食管炎非内镜监测工具——一项前瞻性真实世界多中心研究

文献信息

中文摘要

背景与目的： 嗜酸性食管炎（EoE）是一种慢性炎症性疾病，需要重复内镜检查和活检进行组织学监测，这具有侵入性、耗费资源且给患者带来负担。胶囊海绵（CS）提供了一种潜在的非侵入性替代方法。我们评估了CS与内镜活检相比在监测成人患者疾病活动度方面的诊断准确性、可行性和患者可接受性。
方法： 于2023年8月至2025年6月在英格兰三个NHS中心开展了一项真实世界前瞻性多中心队列研究。患有确诊EoE且无食管狭窄的成人接受了同日内镜检查和使用Endosign装置的CS。组织学活动度定义为内镜活检组织学每高倍视野≥15个嗜酸性粒细胞，或CS每0.3 mm²≥15个嗜酸性粒细胞。症状负担采用吞咽困难症状问卷评估，内镜严重程度采用EoE内镜参考评分评估，患者可接受性通过结构化问卷评估。
结果： 来自三家二级医疗医院的136例患者（164对配对检查）成功吞咽CS并获得足够采样；CS吞咽成功率为92.7%。中位年龄为45岁（IQR 34.8-56）；92/136（67.6%）为男性。CS的敏感性为76.3%，特异性为89.5%，PPV为80.4%，NPV为87.0%，总体准确率为84.8%（Cohen's κ = 0.67）。AUC为0.904（95% CI 0.856-0.953）。患者可接受性记录显示，96.7%认为CS可接受，78.8%更倾向于CS而非内镜，90.8%愿意重复该操作。未发生并发症。
结论： CS是一种安全、可行、准确且高度可接受的侵入性较小的工具，可用于监测EoE。其应用可改善诊疗路径，减少重复内镜活检的需求，同时保持诊断可靠性，并有可能减少服务负担、碳排放并改善患者体验。
试验注册： CoSiE：IRAS编号：314770；ClinicalTrials.gov标识符：NCT06501274；QuBiE：英国临床试验注册库标识符：ISRCTN14071734；试验编号：10083253。

英文摘要

BACKGROUND AND AIMS: Eosinophilic oesophagitis (EoE) is a chronic inflammatory condition requiring repeat endoscopy and biopsy for histological monitoring, which is invasive, resource-intensive and burdensome for patients. The capsule sponge (CS) offers a potential non-invasive alternative. We evaluated the diagnostic accuracy, feasibility and patient acceptability of CS versus endoscopic biopsies for monitoring disease activity in adult patients.
METHODS: A real-world prospective multicentre cohort study was conducted across three NHS sites in England from August 2023 to June 2025. Adults with established EoE without oesophageal strictures underwent same-day endoscopy and CS using the Endosign device. Histological activity was defined as ≥ 15 eosinophils per high-power field or 0.3 mm2 on endoscopic biopsy histology and CS respectively. Symptom burden was assessed using the Dysphagia Symptom Questionnaire, endoscopic severity using the EoE Endoscopic Reference Score and patient acceptability via structured questionnaire.
RESULTS: 136 patients (164 paired episodes) from three secondary care hospitals successfully swallowed CS with sufficient sampling; CS swallowing success rate was 92.7%. Median age was 45 years (IQR 34.8-56); 92/136 (67.6%) were male. CS demonstrated sensitivity 76.3%, specificity 89.5%, PPV 80.4%, NPV 87.0% and overall accuracy 84.8% (Cohen's κ = 0.67). AUC was 0.904 (95% CI 0.856-0.953). Patient acceptability recorded 96.7% found CS acceptable, 78.8% preferred it over endoscopy, and 90.8% would repeat the procedure. No complications occurred.
CONCLUSIONS: CS is a safe, feasible, accurate and highly acceptable less-invasive tool for monitoring EoE. Its implementation could improve care pathways, reducing the need for repeated endoscopic biopsies while maintaining diagnostic reliability, with potential to reduce service burden, carbon emissions and improve patient experience.
TRIAL REGISTRATION: CoSiE: IRAS number: 314770; ClinicalTrials.gov identifier: NCT06501274; QuBiE: UK Clinical Trial registry identifier: ISRCTN14071734; Trial number: 10083253.