基于数据挖掘研究现代中医复方治疗股骨颈骨折用药规律

Study on the Medication Law of Modern Tra-ditional Chinese Medicine Compound Treatment for Femoral Neck Fracture

摘要: 基于“TCMISSV2.5”分析近30年中医复方治疗股骨颈骨折(Transcervical Fracture, TF)的用药处方规律，为临床用药提供参考。检索中国知网(CNKI)、中国生物医学文献数据库(SinoMED)和万方数据库(WF)中中医复方治疗TF的文献，时间范围为1993年01月01日~2023年01月01日。用Microsoft Excel 2021建立方剂数据库，并导入“TCMISSV2.5”。使用该系统中的频数分析、关联规则分析、熵聚类分析等方法进行综合分析。共搜索到文献1917篇，筛选后共纳入文献355篇，其中包含中药263味，使用频次 ≥ 20次的药物共计21味，中药使用频数分析显示，出现频次最高的前10位药物由高到低依次是当归、红花、骨碎补、续断、川芎、赤芍、乳香、没药、桃仁、熟地黄。中药性味、归经频数分析显示，药性以温、平、寒居多，五味多为苦、辛、甘，以归肝、心、脾经为主。高频药对频数分析显示，应用最多的前三组合为红花–当归、赤芍–当归、川芎–当归。通过高频药物关联规则分析，发现强关联性药对为桃仁–当归–红花、赤芍–川芎–当归。通过熵聚类分析，共演化出潜在新方组合9个。通过数据挖掘总结了现代中医复方治疗TF的用药规律，为临床可以提供一定的用药参考，并为新药的开发提供思路和依据。

Abstract: Based on the analysis of the prescription patterns of traditional Chinese medicine (TCM) compounds for the treatment of transcervical femoral neck fractures (TF) using “TCMISSV2.5”, this study pro-vides a reference for clinical medication. Literature on TCM compounds for the treatment of TF was retrieved from the China National Knowledge Infrastructure (CNKI), SinoMED, and Wanfang Data (WF) databases, with a timeframe from January 1, 1993, to January 1, 2023. A formula database was established using Microsoft Excel 2021 and imported into “TCMISSV2.5”. Frequency analysis, asso-ciation rule analysis, and entropy clustering analysis methods in the system were used for compre-hensive analysis. A total of 1917 articles were searched, and 355 articles were included after screening. These articles contained 263 traditional Chinese medicine ingredients, with 21 ingredi-ents used at least 20 times. The top 10 most frequently used drugs in descending order were An-gelica sinensis, Carthamus tinctorius, Dendrobium, Astragalus membranaceus, Ligusticum chuanxiong, Paeonia lactiflora, Myrrh, Frankincense, Persica, and Rehmannia glutinosa. The analysis of taste and meridian affiliation showed that the drug properties were mainly warm, mild, and cold; the five tastes were mostly bitter, pungent, and sweet, and the meridians primarily included the liver, heart, and spleen. The analysis of high-frequency drug pairs revealed the most commonly used three drug pairs and combinations to be Carthamus tinctorius-Angelica sinensis, Paeonia lacti-flora-Angelica sinensis, and Ligusticum chuanxiong-Angelica sinensis. Through high-frequency drug association rule analysis, it was found that strongly associated drug pairs were Persica-Angelica sinensis-Carthamus tinctorius and Paeonia lactiflora-Ligusticum chuanxiong- Angelica sinensis. Through entropy clustering analysis, nine potential new formula combinations were evolved. This study summarizes the medication rules of TCM compounds for the treatment of TF, providing a ref-erence for clinical use and offering insights and a basis for new drug development.