目的 全面收集中医药治疗肺结节（含早期肺癌）的随机对照试验证据，并分析和评价其现状，为临床研究设计与实践提供参考和建议。方法 全面检索智能化中成药临床证据数据库、中国知网、维普、万方、中国生物医学文献服务系统、PubMed、Web of Science、Embase、Cochrane Library，检索时间为建库至2025年9月。提取数据包括纳入研究的患者特征（如疾病亚型、中医证型）和发表年份、样本量、中医治则、治疗组与对照组干预措施、疗程、随访时长、详细方法学信息、结局指标等；对于涉及中成药的研究使用Aireview Agent进行标准化分类后人工核对；其他中医药研究的数据则人工提取与核对。结果 共纳入70项研究，2024年发文量最多。证型以气阴两虚证和痰瘀互结证为主，治法以化痰、益气、散结为主，干预措施以中药复方汤剂为主，结局指标中临床疗效出现频次最多。方法学评估中偏倚风险最高的是结局测量偏倚。结论 纳入研究的患者疾病类型不明确，中医辨证不规范、忽视终点指标、方法学质量存在偏倚，结局指标选择不统一。建议后续按疾病亚型分层开展去中心化临床研究或多中心、大样本的研究，规范试验设计，构建中医特色的评价体系，严格按照随访管理指南进行随访，提供更高质量的中医药治疗肺结节（含早期肺癌）的证据。

Objective To systematically evaluate the current evidence from randomized controlled trials (RCTs) on traditional Chinese medicine (TCM) for pulmonary nodules (including early-stage lung cancer), and to provide evidence-based recommendations for future clinical practice and research. Methods Comprehensive searches were conducted across the following databases up to September 2025: Intelligent Clinical Evidence Database of Chinese Patent Medicine, China national knowledge infrastructur (CNKI), VIP Journal Database, Wanfang Data Knowledge Service Platform, Chinese biomedical literature service system (SinoMed), PubMed, Web of Science, Embase, and the Cochrane Library. Extracted data included key characteristics of participants and studies, such as disease subtype, TCM syndrome, year of publication, sample size, TCM therapeutic principles, details of treatment and control interventions, duration of treatment and follow-up, methodological details, and outcome measures. For studies involving Chinese patent medicines, extracted data were categorized using the Aireview Agent and subsequently verified manually. Data from other TCM studies were extracted and verified manually. Results A total of 70 RCTs were included, with the greatest number published in 2024. The studies showed that deficiency of both qi and yin syndrome and phlegm-stasis mutual binding syndrome were the most common TCM syndromes. The main therapeutic principles were resolving phlegm, tonifying qi, and dispersing nodules. In the treatment groups, Chinese herbal decoctions were the predominant intervention. Clinical efficacy was the most commonly reported outcome measure. In the risk-of-bias assessment, the measurement of the outcome domain was most frequently rated at high risk of bias. Conclusion RCTs of TCM interventions for pulmonary nodules (including early-stage lung cancer) exhibited several critical limitations, such as inadequate specification of disease subtypes in participant eligibility criteria, inconsistent TCM syndrome differentiation, failure to report or analyze primary endpoints, and a high risk of bias—particularly in outcome measurement. We recommend developing a core outcome set for TCM interventions in pulmonary nodules. Future clinical trials—whether decentralized or large-scale, multicenter randomized controlled trials should be stratified by disease subtype, use standardized protocols, and strictly adhere to established follow-up guidelines to generate high-quality evidence, particularly for nodules at elevated risk of malignancy.

R285.64

2025现代中医药海河实验室第一批科技项目——基于深度推理模型的“中医药循证评价智能体”构建与应用（25HHZYSS00023）；天津中医药大学研究生科研经费项目（2024）