目的 基于数据挖掘技术探索中医临床治疗气虚血瘀型冠心病（coronary heart disease，CHD）的用药规律。方法 从中国知网（CNKI）、万方数据知识服务平台、中文科技期刊数据库（维普）、中国生物医学文献数据库及PubMed、Web of Science（WOS）数据库检索相关的临床研究，根据文献纳排标准筛选文献并建立中药数据库。采用Excel 2019对所有中药的使用频次、功效类别、性味归经、用药剂量进行分析，挖掘临床用药规律；采用IBM SPSS Statistics 21.0、IBM SPSS Modeler 18.0、Cytoscape 3.7.2软件对所有中药进行关联分析，得出核心药对及药物组合；采用Lantern 5.0软件对使用频次≥10的中药进行隐结构模型及综合聚类分析，得出核心复方，并在此基础上推断疾病兼证；采用Gephi 0.9.2软件对所有中药进行复杂网络分析，主要包括K-core分析和社群分析，得出核心中药及新复方。结果 最终筛选出558篇文献，其中，中医处方89首，中药197味，累计使用频次5 088次。中药使用频次及功效类别统计结果显示，使用频次排名前5的中药依次为黄芪、丹参、川芎、当归、赤芍。功效类别以补虚药、活血化瘀药、清热药、解表药、理气药为主。中药的性味归经分布结果显示，药性以温、寒、平为主，药味以甘、辛、苦为主，归经以肝、心、脾为主。对197味中药进行关联分析，结果表明黄芪、丹参、川芎、当归为核心中药。关联强度排前3位的药对依次为丹参-黄芪、川芎-黄芪、当归-黄芪。隐结构模型分析共获得15个隐变量、30个隐类、5个综合聚类模型，8个核心复方；以方测证推断出气虚血瘀型CHD的患者可能同时伴有气滞、痰湿、阴虚等兼杂证的临床表现。复杂网络K-core分析显示，黄芪、丹参、川芎、当归、甘草、赤芍、党参、红花等46味中药为核心中药。社群聚类得到3个核心中药群即3个新复方。结论 通过数据挖掘技术对近10年治疗气虚血瘀型CHD的文献进行探索，总结中医临床治疗气虚血瘀型CHD的用药规律，期望为中医临床治疗气虚血瘀型CHD提供标准化的用药参考和治疗新思路。

Objective To explore the medication rules of traditional Chinese medicines (TCMs) for the clinical treatment of coronary heart disease (CHD) with qi deficiency and blood stasis type based on data mining. Methods Relevant clinical researches were retrieved from China National Knowledge Infrastructure (CNKI), Wanfang Data Knowledge Service Platform, China Science and Technology Journal Database (VIP), China Biology Medicine disc (CBMdisc), PubMed, Web of Science (WOS) databases and the TCMS database was established by screening the literature in accordance with the inclusion and exclusion criteria of this study. Excel 2019 was used to analyze the usage frequency, efficacy categories, properties, flavours, meridians and dosages of all TCMs, and to explore the clinical medication rules. IBM SPSS Statistics 21.0, IBM SPSS Modeler 18.0 and Cytoscape 3.7.2 software were used to conduct association analysison of all TCMs to obtain the core drug pairs and drug combinations. Lantern 5.0 software was used to perform the latent structure model and comprehensive clustering analysis of TCMs with frequency of use ≥ 10 to obtain the core prescriptions, and to infer the concurrent syndromes of disease on this basis. Gephi 0.9.2 software was used to conduct complex network analysis of all TCMs, mainly including K-core hierarchical analysis and community analysis to obtain the core TCMS and new compound formulas. Results A total of 558 literature were finally screened out according to the inclusion and exclusion criteria of this study, which included 89 TCM prescriptions, 197 TCMs, and the cumulative frequency of use was 5 088 times. The statistical results of the frequency of use and efficacy categories of TCMs showed that the top five TCMs in terms of frequency were Huangqi (Astragali Radix), Danshen (Salvia Miltiorrhiza Radix), Chuanxiong (Chuanxiong Rhizoma), Danggui (Angelicae Sinensis Radix), and Chishao (Paeoniae Radix Rubra). The primary efficacy categories were tonic medicines, blood-circulating and stasis-resolving medicines, interior heat-clearing medicines, exterior-releasing medicines, qi-regulating medicines. The results of the distribution of the properties, flavors and meridians of TCMs showed that the medicinal properties were mainly warm, cold and neutral, the medicinal flavors were mainly sweet, pungent and bitter, and the meridians were mainly liver, heart and spleen. The association analysis of 197 TCMs showed that Astragali Radix, Salvia Miltiorrhiza Radix, Chuanxiong Rhizoma, and Angelicae Sinensis Radix were core TCMs. The top three medicine combinations in terms of association intensity were Salvia Miltiorrhiza Radix-Astragali Radix, Chuanxiong Rhizoma-Astragali Radix, and Angelicae Sinensis Radix-Astragali Radix. The analysis of the latent structure model obtained 15 hidden variables, 30 hidden categories, five comprehensive clustering models, and eight core prescriptions. According to the prescriptions the patients with CHD of qi deficiency and blood stasis type might both have the clinical manifestations of concurrent syndromes of qi stagnation, phlegm dampness, yin deficiency and other syndromes. The K-core hierarchical analysis of the complex network showed tha t46 TCMs such as Astragali Radix, Salvia Miltiorrhiza Radix, Chuanxiong Rhizoma, Angelicae Sinensis Radix, Gancao (Glycyrrhizae Radixet Rhizoma), Paeoniae Radix Rubra, Dangshen (Codonopsis Radix), Honghua (Carthami Flos) and other TCMs were the core TCMS. The three core TCMS were obtained by community clustering namely three new compound formulas. Conclusion This study explored the literature in the treatment of CHD of qi deficiency and blood stasis type in the past ten years through data mining, and summarized the medication rules of TCM for the clinical treatment of CHD of qi deficiency and blood stasis type, Which provided standardized medication references and new treatment ideas for the clinical treatment of CHD of qi deficiency and blood stasis type with TCMs.

R285

国家自然科学基金资助项目(82474278)