Traditional chinese medicine zheng in the era of evidence-based medicine: a literature analysis

Abstract

Zheng, which is also called a syndrome or pattern, is the basic unit and a key concept of traditional Chinese medicine (TCM) theory. Zheng can be considered a further stratification of patients when it is integrated with biomedical diagnoses in clinical practice to achieve higher efficacies. In an era of evidence-based medicine, confronted with the vast and increasing volume of TCM data, there is an urgent need to explore these resources effectively using techniques of knowledge discovery in databases. The application of effective data mining in the analysis of multiple extensively integrated databases can supply new information about TCM Zheng research. In this paper, we screened the published literature on TCM Zheng-related studies in the SinoMed and PubMed databases with a novel data mining approach to obtain an overview of the Zheng research landscape in the hope of contributing to a better understanding of TCM Zheng in the era of evidence-based medicine. In our results, contrast was found in Zheng in different studies, and several determinants of Zheng were identified. The data described in this paper can be used to assess Zheng research studies based on the title and certain characteristics of the abstract. These findings will benefit modern TCM Zheng-related studies and guide future Zheng study efforts.

Figure 1

Annual distribution of Chinese-language articles about TCM Zheng in 3 categories (animal experimental studies, clinical studies, and pure Zheng-related studies). The data are obtained from the SinoMed database (until December 11, 2011). In the calculation, some of the annual frequencies of animal studies are 0 and 1, which are too small to be clearly shown in the column diagram. Therefore, the values are converted by the natural logarithm function “Annual Value = ln(Annual Value_origin + 1).” Based on this function, the frequency of 0 is still 0, and the frequency of 1 (as well as other values) is ln(2).

Figure 2

The annual records of Chinese-language articles about TCM Zheng classification in 2 categories. The data were obtained from the SinoMed database (until December 11, 2011). The statistics are based on scanning studies as to whether they contain a disease name in the framework of Western medicine. The count of “independent Zheng” increases by one if a study does not contain a disease name, and the count of “Zheng in disease” increases by one if the study contains one or more disease names.

Figure 3

The annual records of English-language articles about TCM Zheng classification in 3 categories. The data were obtained from the PubMed database (until December 11, 2011). As in Figure 1, the annual values are also converted by the function “Annual Value = ln(Annual Value_origin + 1).” Therefore, a comparison between SinoMed and PubMed can be obtained based on the same standard.

Figure 4

The ten most common diseases in Chinese-language TCM Zheng-related clinical studies annually. The data were obtained from the SinoMed database (until December 11, 2011). Each line represents the annual studies of TCM Zheng for one particular disease. The annual values are converted by the function “Annual Value = ln(Annual Value_origin + 1)” to better display the tendency.

Figure 5

Overview of the Zheng-Zheng network. This network is generated from mining the SinoMed literature on TCM Zheng. The method of calculation is based on a data slicing algorithm that calculates the frequencies of the co-occurrence TCM Zhengs. Each node represents one type of Zheng. The size of the node indicates the frequency of Zheng publications; a larger node indicates more reports about the Zheng. The line width represents the frequency of co-occurrence of the connected Zhengs.

Figure 6

Overview of the disease-Zheng network. The disease-Zheng network is generated from a SinoMed literature analysis with the cooccurrence frequencies of disease. The method of calculating the frequency of co-occurrence is also based on a data slicing algorithm. In this figure, the square grey shape is a certain disease, and the round green shape is a TCM Zheng. If two diseases have a common Zheng, there is an edge connecting them. The upper left part identifies the relevant Zheng research on primary hypertension, with 2 TCM Zhengs for this disease. The upper right part represents the 6 most influential Zhengs in gastritis research. The section below each part represents the total number of shared Zheng among diabetes mellitus, hepatocirrhosis, and heart failure. The kidney yin deficiency Zheng can be found in both DM and Hepatocirrhosis, and Qi deficiency with blood stasis Zheng can be found in both DM and HF.