Author: Yili Hu
Abstract: To explore the method of using traditional Chinese medicine in the intervention of hyperuricemia through literature research, and to provide a basis for exploring the prescription of traditional Chinese medicine to reduce hyperuricemia.
Methods: Using "Chinese medicine" and "hyperuricemia" as the subject terms, we retrieved relevant journal documents from China CNKI from 2000 to 2020, and used data mining techniques (frequency analysis, association rule analysis, cluster analysis) to mine and study core drug combinations and new prescriptions.
Results: A total of 102 papers with 102 prescriptions and 115 traditional Chinese medicinal herb components were screened to meet the research conditions. The use frequency, drug pair combination, association rules, and cluster analysis showed that the core formula was Rhizoma Polygoni Cuspidati, Radix Clematidis, Herba Polygoni Avicularis, Lysimachia, Cortex Fraxini, a total of 5 herbs.
Conclusion: The best combination of traditional Chinese medicine intervention for hyperuricemia is able to remove dampness and turbidity, dispel phlegm and disperse blood stasis, and relieve arthritis.
Keywords: hyperuricemia, data mining, traditional Chinese medicine, core prescriptions
Hyperuricemia (HUA) is a metabolic disease in which the metabolism of purines in the body is disturbed, and the excretion of uric acid is reduced. With the improvement of people's living standards and changes in dietary structures, the incidence rate has increased year by year and is trending towards younger ages in recent years [1-2]. Hyperuricemia is a risk factor for various diseases such as gout, uric acid nephropathy, hypertension, diabetes, and cardiovascular and cerebrovascular diseases [3-4]. Western medicine has many side effects in the control and treatment of hyperuricemia, which is likely to cause secondary damage to the human body. Traditional Chinese Medicine has a unique effect in interfering with hyperuricemia [5-6]. Designating a high-efficiency combination formula for hyperuricemia is particularly urgent and important. This study summarizes the research and analysis of the law of Traditional Chinese Medicine interventions for hyperuricemia in the past 20 years. It explores the high-efficiency combination of traditional Chinese medicine for the intervention of hyperuricemia and provides a basis for medication and products to control or intervene with the disease.
1 Research materials
1.1 Literature sources
The subject headings of "Chinese medicine" and "hyperuricemia" are collected from the relevant journals of 2000-2020 from the CNKI and the Wanfang database.
1.2 Inclusion criteria 1) Must be clinical research literature regarding traditional Chinese medicine intervention for hyperuricemia or gout; 2) The number of cases observed and collected must be > 30 cases; 3) The intervention measures are oral administration of traditional Chinese medicinal herbs, and the prescription information is complete; 4) The effective rate of curative effect is greater than 70%
1.3 Exclusion criteria 1) The literature of research consists of hyperuricemia combined with other diseases; 2) Theoretical discussion articles; 3) Any literature of unreasonable design of experimental intervention measures.
2.1 Data processing 1) Refer to the 2015 edition of the "Chinese Pharmacopoeia" to uniformly standardize the names of medical herbs; 2) Only enter the main prescription herbs, and do not enter the medications with complications or concurrent certificates; 3) Use an Excel table to establish a database, double-check to ensure entry precisely.
The entire process was carried out with Python 3.8. Chinese medicine importance was obtained by calculating the frequency in a Python list. Connections between Chinese medicine were obtained by running social network analysis in Python. Group information was obtained by running an agglomerative clustering in Python.
3.1 Single herb frequency distribution
102 articles were included, involving 102 Chinese Medicine prescription formulas and 115 different herbs. As we all know, herbs that showed up more frequently in lists played a bigger role in Chinese medicine formulas than other combinations, i.e., if we observed more Cortex Fraxini than Rhizoma Polygoni Cuspidati, it is more likely that we will have Cortex Fraxini in a new Chinese medicine design. Levels of importance can be obtained by calculating the total number of connections for the given Chinese medicine combinations. Table 1 shows the component importance for every single herb. The top 15 herbs out of 115 are as follows: Cortex Fraxini, Rhizoma Polygoni Cuspidati, Radix Clematidis, Herba Polygoni Avicularis, Lysimachia, Rhizoma Dioscoreae Septemlobae, Rhizoma Smilacis Glabrae, Radix Cyathulae, Cortex Phellodendri, Rhizoma Alismatis, Semen Plantaginis, Rhizoma Atractylodis, Radix Salviae Miltiorrhizae, Radix Paeoniae Rubra, Semen Coicis, Pheretima, Poria. We also notice that Cortex Fraxini, Rhizoma Polygoni Cuspidati, Radix Clematidis, Herba Polygoni Avicularis, Lysimachia received more than 60 observations.
3.2 Multi-herb frequency distribution
By applying the same method, we can also understand the importance of 2, 3, 4, and 5 herb formulas. Results are summarized in Tables 2, 3, 4, and 5, respectively. Based on the results from tables 1 to 5, we concluded that the combination of Cortex Fraxini, Herba Polygoni Avicularis, Lysimachia, Radix Clematidis, and Rhizoma Polygoni Cuspidati appeared with the highest frequency.
3.3 Association rule learning
We further investigated the relationship between the two herbs. For each herb component, we are interested in two pieces of information, 1) the total number of connections with the given component; 2) the components connected to the given component and the total number of connections between them. For a given𝑥𝑖, 𝑁𝑥𝑖,𝑥𝑗 can be obtained by the algorithm proposed above for𝑖 ≠ 𝑗, and the total number of connections with 𝑥𝑖 is ∑ 𝑁𝑥𝑖 𝑗≠ ,𝑥𝑗𝑖. The results are shown in Table 6. In table 6, we filter out data with fewer than 5 connections for better visualization.
The same results are also presented by using social network analysis. The greater the number of connections between two nodes in the network, the larger the edge width. Based on the network, we concluded that Cortex Fraxini, Rhizoma Polygoni Cuspidati, Rhizoma Polygoni Cuspidati, Radix Clematidis, Herba Polygoni Avicularis, Lysimachia, Rhizoma Smilacis Glabrae had the greatest relationship.
3.4 Clustering Analysis
To explore the group information among all herb components, herb similarities are studied in addition to component importance. In this study, the Carbo index is used to compare the similarity of two components:
Group information is obtained by running an agglomerative clustering in Python with a similarity matrix as input which contains the Carbo index for every two-component pairings. The results are shown in figure 7. Based on results in figure 7, We found that the most comparable plants were Cortex Fraxini, Rhizoma Polygoni Cuspidati, Radix Clematidis, Herba Polygoni Avicularis, and Lysimachia. Furthermore, we determined that these five herbs are the best components for medication design.
Traditional Chinese medicine believes that hyperuricemia is caused by insufficient body spleen and kidney, coupled with improper diet, fatigue and internal injury, poor movement of the essence and qi in the body, stagnation of water, dampness, and internal growth, and phlegm and blood stasis over time. Dampness, phlegm, and blood stasis are blocked in joints, resulting in gout; stagnation in blood vessels, visible cardio-cerebrovascular damage; accumulation in liver and kidney, damage to liver and kidney function; piled in the body, lung, spleen, and kidney cannot be transported and transformed, and the essence is discharged Diabetes. Dampness, phlegm, and blood stasis are the main pathogenesis of this disease, and treatment should be to remove dampness and turbidity, dispel phlegm and blood stasis, and relieve arthralgia. In recent years, with the in-depth study of the effective material basis of herbs and the pathogenesis of hyperuricemia, active ingredients of traditional Chinese medicine able to regulate hyperuricemia have been continuously discovered . In this study, a large sample of prescriptions with a clinically effective rate of >70% in the intervention of hyperuricemia in the past 20 years was found, and a standard database of Chinese medicine prescriptions for intervention in hyperuricemia was constructed. On this basis, with an herb as the node, whether two or more herbs are connected in the same prescription, and the frequency of 2-5 herbs used in multiple prescriptions as weights were used to construct a specific formula for hyperuricemia. The Chinese medicine prescription correlation network model: select the node (herb) strength as the index to quantitatively research and extract the core prescriptions, and select the weight of the network connection as the index to study and analyze the correlation strength between the Chinese medicines. Finally, using Statistics-related knowledge, an in-depth exploration of the core prescriptions of traditional Chinese medicine was carried out. The core combination is usually composed of the most important drugs in the prescription. The analysis results show that Rhizoma Polygoni Cuspidati, Radix Clematidis, Herba Polygoni Avicularis, Lysimachia, Cortex Fraxini, are the 5 core parts of prescriptions used for hyperuricemia. In the prescription, Rhizoma Polygoni Cuspidati promotes blood circulation, dissipates blood stasis, relieves menstruation, and promotes dampness; Radix Clematidis dispels rheumatism and relieves meridians; Cortex Fraxini clears heat and dampness; Herba Polygoni Avicularis and Lysimachia diuresis and drains dampness; the common combination is to remove dampness and turbidity, dispel phlegm and dispel blood stasis. The effects shown are consistent with the pathogenesis and treatment of hyperuricemia in TCM. The research of the above-mentioned combination and compatibility provides extremely important data, more scientific conclusions, and a basis for the development of new effective intervention products for hyperuricemia.
 Li Jing. Epidemiological study of hyperuricemia [J]. Chinese Journal of Cardiovascular Diseases, 2016, 21 (2): 83-86.
 Yuchang Tang, Hong Liu, Bicheng Liu. Reflections on the basis of changes:the epidemiological data of hyperuricemia [J]. Drug evaluation，2015 (7) :8- 13.
 Zhu Y，Pandya BJ，Choi HK. Comorbidities of Gout and Hyperuricemia in the US General Population ：NHANES 2007-2008 [J]. American Journal of Medicine， 2012,125 (7): 679-687.  Kim SY. Guevara JP. Kim KM, et a1. Hyperuricemia and coronary heart disease: a systematic review and meta-analysis [J]. Arthritis Care Res, 2010, 62(2):170-180．
 Chunsheng Zhu, Bing Zhang, Zhijian Lin, et al. Research progress in the treatment of hyperuricemia with traditional Chinese medicine [J]. China Journal of Traditional Chinese and Pharmacy, 2015 (12):4374-4376.
 Mingyue Zhang, Yue Wang. General treatment of hyperuricemia by traditional Chinese medicine [J]. Guiding Journal of Traditional Chinese and Pharmacy, 2015 (10) :101-104.
 Jinchang Liu, Tao Wang. Present research situation and thinking of drug in treating hyperuricemia [J]. Tianjin Journal of Traditional Chinese medicine，2017, (5) :291-29