Department of Biomedical Sciences, Humanitas University, Pieve Emanuele, and Endoscopy Unit, Humanitas Clinical and Research Center IRCCS, Rozzano, Italy (C.H., M.S., A.R.).
Clinical Effectiveness Research Group, University of Oslo, Oslo, Norway, and Digestive Disease Center, Showa University Northern Yokohama Hospital, Yokohama, Japan (Y.M.).
Ann Intern Med. 2023 Sep;176(9):1209-1220. doi: 10.7326/M22-3678. Epub 2023 Aug 29.
Artificial intelligence computer-aided detection (CADe) of colorectal neoplasia during colonoscopy may increase adenoma detection rates (ADRs) and reduce adenoma miss rates, but it may increase overdiagnosis and overtreatment of nonneoplastic polyps.
To quantify the benefits and harms of CADe in randomized trials.
Systematic review and meta-analysis. (PROSPERO: CRD42022293181).
Medline, Embase, and Scopus databases through February 2023.
Randomized trials comparing CADe-assisted with standard colonoscopy for polyp and cancer detection.
Adenoma detection rate (proportion of patients with ≥1 adenoma), number of adenomas detected per colonoscopy, advanced adenoma (≥10 mm with high-grade dysplasia and villous histology), number of serrated lesions per colonoscopy, and adenoma miss rate were extracted as benefit outcomes. Number of polypectomies for nonneoplastic lesions and withdrawal time were extracted as harm outcomes. For each outcome, studies were pooled using a random-effects model. Certainty of evidence was assessed using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) framework.
Twenty-one randomized trials on 18 232 patients were included. The ADR was higher in the CADe group than in the standard colonoscopy group (44.0% vs. 35.9%; relative risk, 1.24 [95% CI, 1.16 to 1.33]; low-certainty evidence), corresponding to a 55% (risk ratio, 0.45 [CI, 0.35 to 0.58]) relative reduction in miss rate (moderate-certainty evidence). More nonneoplastic polyps were removed in the CADe than the standard group (0.52 vs. 0.34 per colonoscopy; mean difference [MD], 0.18 polypectomy [CI, 0.11 to 0.26 polypectomy]; low-certainty evidence). Mean inspection time increased only marginally with CADe (MD, 0.47 minute [CI, 0.23 to 0.72 minute]; moderate-certainty evidence).
This review focused on surrogates of patient-important outcomes. Most patients, however, may consider cancer incidence and cancer-related mortality important outcomes. The effect of CADe on such patient-important outcomes remains unclear.
The use of CADe for polyp detection during colonoscopy results in increased detection of adenomas but not advanced adenomas and in higher rates of unnecessary removal of nonneoplastic polyps.
European Commission Horizon 2020 Marie Skłodowska-Curie Individual Fellowship.
结肠镜检查中使用人工智能计算机辅助检测（CADe）可能会提高腺瘤检出率（ADR）并降低腺瘤漏诊率，但可能会增加对非肿瘤性息肉的过度诊断和过度治疗。
定量评估 CADe 在随机试验中的获益和危害。
系统评价和荟萃分析。（PROSPERO：CRD42022293181）。
通过 2023 年 2 月的 Medline、Embase 和 Scopus 数据库。
比较 CADe 辅助结肠镜检查与标准结肠镜检查在息肉和癌症检测方面的随机试验。
腺瘤检出率（每例患者检出≥1 个腺瘤的比例）、每例结肠镜检查检出的腺瘤数、高级别腺瘤（≥10mm 伴高级别异型增生和绒毛状组织学）、每例结肠镜检查检出的锯齿状病变数和腺瘤漏诊率作为获益结局。非肿瘤性病变的息肉切除术数量和退出时间作为危害结局进行提取。对于每个结局，使用随机效应模型对研究进行汇总。使用 GRADE（推荐评估、制定与评价）框架评估证据确定性。
纳入了 21 项涉及 18232 名患者的随机试验。CADe 组的 ADR 高于标准结肠镜组（44.0%比 35.9%；相对风险，1.24 [95%CI，1.16 至 1.33]；低确定性证据），相应的漏诊率降低了 55%（风险比，0.45 [CI，0.35 至 0.58]）（中等确定性证据）。CADe 组比标准组切除的非肿瘤性息肉更多（每例结肠镜检查 0.52 比 0.34 个；平均差值 [MD]，0.18 个息肉 [CI，0.11 至 0.26 个息肉]；低确定性证据）。CADe 仅使检查时间略有增加（MD，0.47 分钟 [CI，0.23 至 0.72 分钟]；中等确定性证据）。
本综述重点关注对患者重要结局的替代指标。然而，大多数患者可能认为癌症发病率和癌症相关死亡率是重要的结局。CADe 对这些患者重要结局的影响尚不清楚。
在结肠镜检查中使用 CADe 进行息肉检测可提高腺瘤的检出率，但不会提高高级别腺瘤的检出率，并增加非肿瘤性息肉的不必要切除率。
欧盟 Horizon 2020 玛丽·居里个体化奖学金计划。
