Oral Administration of East Asian Herbal Medicine for Peripheral Neuropathy: A Systematic Review and Meta-Analysis with Association Rule Analysis to Identify Core Herb Combinations

This review aimed to comprehensively assess the efficacy and safety of oral East Asian herbal medicine (EAHM) for overall peripheral neuropathy (PN). In addition, an Apriori algorithm-based association rule analysis was performed to identify the core herb combination, thereby further generating useful hypotheses for subsequent drug discovery. A total of 10 databases were searched electronically from inception to July 2021. Randomized clinical trials (RCTs) comparing EAHM with conventional analgesic medication or usual care for managing PN were included. The RCT quality was appraised using RoB 2.0, and the random effects model was used to calculate the effect sizes of the included RCTs. The overall quality of evidence was evaluated according to the Grading of Recommendations Assessment, Development, and Evaluation. By analyzing the constituent herb data, the potential association rules of core herb combinations were explored. A total of 67 RCTs involving 5753 patients were included in this systematic review. In a meta-analysis, EAHM monotherapy and combined EAHM and western medicine therapy demonstrated substantially improved sensory nerve conduction velocity, motor nerve conduction velocity, and response rate. Moreover, EAHM significantly improved the incidence rate, pain intensity, Toronto clinical scoring system, and Michigan diabetic neuropathy score. The evidence grade was moderate to low due to the substantial heterogeneity among the studies. Nine association rules were identified by performing the association rule analysis on the extraction data of 156 EAHM herbs. Therefore, the constituents of the herb combinations with consistent association rules were Astragali Radix, Cinnamomi Ramulus, and Spatholobi Calulis. This meta-analysis supports the hypothesis that EAHM monotherapy and combined therapy may be beneficial for PN patients, and follow-up research should be conducted to confirm the precise action target of the core herb.


Description of the Condition
Peripheral neuropathy (PN) is one of the most common causes for a patient to visit the clinic [1]. The prevalence of diabetic peripheral neuropathy (DPN) or herpetic neuropathy, the commonly observed PNs, is at least 10-20% [2,3]. However, it is not easy to collect the available PN epidemiological data since the causes of pathology are very diverse. In addition, the symptoms can develop not only in a single affected area, but also in multiple nerves [4]. Symptoms that may occur due to this disease include chronic pain, decreased nerve conduction velocity (NCV), sensation loss, and abnormal sensations such as tingling, Incidence rate: Occurrence rate of PN due to multiple underlying causes. Pain intensity: Intensity of PN related to pain symptoms, as measured by instruments, such as the visual analog scale (VAS) or numerical rating scale (NRS). Toronto clinical scoring system (TCSS) [24]. Michigan diabetic neuropathy score (MDNS) [25]. Adverse events (AEs).

Risk of Bias in Individual Studies
Two review investigators (H.-G.J. and D.L.) independently evaluated the RoB of the included studies according to the revised tool for risk of bias in randomized trials, RoB 2.0 [26]. Disagreements between the two reviewers were resolved through discussion. R version 4.1.0 (R core Team (2021). R Foundations for Statistical computing, Vienna, Austria) was used with the 'robvis 'package to generate graphical presentations of biased risk assessments [27,28].

Statistical Analysis 2.5.1. Meta-Analysis
For continuous outcomes, the mean difference (MD) was calculated with a 95% confidence interval (CI). A standardized MD (SMD) of 95% CI was used to express the intervention effect when the same outcome was measured using different scales. Risk ratios or odds ratios with 95% CI were applied to represent results for dichotomous outcomes. Statistical heterogeneity across the included studies was tested using the χ 2 test and I 2 statistics. Heterogeneity was considered statistically significant when the p-value based on the χ 2 test was <0.10 or I 2 was ≥50%. If heterogeneity was identified, a subgroup analysis was performed to explore the possible causes. Statistical synthesis of individual research results was performed using R version 4.1.0, with the default settings of the 'meta' package and the 'metaprop' function [29]. Only the random effects model was adopted in this review to statistically examine the results conservatively. To distinguish publication bias, a contour-enhanced funnel plot was used for the outcome, which included most of the studies [30]. For the asymmetry on the visually confirmed funnel plot, Egger's test and Begg's test were additionally performed to specifically confirm the existence of publication bias.

Association Rule Analysis
By analyzing the constituent herb data of EAHM collected from the included studies, the potential association rules of core herb combinations were explored. Furthermore, prior to the association rule analysis, the frequency of individual herbs used in this analysis was checked. The R studio program (version 1.4.1106; Integrated Development for R. RStudio, PBC; Boston, MA, USA) was used for the Apriori association rule analysis and plot production. A data fit was performed using the R-package "arules" and the Rpackage "arulesViz"was applied to generate plots and charts according to the results [31,32]. The association rule analysis according to the Apriori algorithm is a data mining method for discovering meaningful correlations between two or more components included in one event [33]. This identifies the elements that compose the data and the relationship between the elements, and is used in various types of medical research aimed at predicting the variable characteristics [34][35][36]. This analysis does not identify a separate cause and aims to derive a rule from a combination of characteristics without a target variable. Support, confidence, and lift are the main metrics used to measure associations using the Apriori algorithm. The metric support evaluates the usefulness of the association rule and is the proportion of prescriptions containing a specific herb combination in the total EAHM prescription. This can be expressed as P(A∩B). The metric confidence indicates the likelihood that the consequent herb set will be included when an antecedent herb set is specified as an EAHM prescription. Therefore, support is the entire set of standard EAHM prescriptions, whereas confidence limits reference prescriptions to those that include a specific herb combination and is expressed as P(A∩B)/P(A) = P(B|A). The metric lift compensates for the fact that it is not known whether confidence is useful or a random result. The confidence of herbs A and B is divided by the confidence under the independent assumption that A does not affect B, and is expressed as P(A∩B)/P(A)·P(B) = P(B|A)/P(B). When the confidence is approximately 1, herbs A and B are considered irrelevant since they are close to independence in probability. Conversely, if the lift value is large, the correlation is interpreted as strong. In this review, the association rules were identified based on the minimum values for support and confidence of 20 and 80%, respectively. Among them, the core herb combination showing the most distinct association and its constituent herbs was searched.

Quality of Evidence According to Outcome Measurements
The overall quality of evidence for each outcome was evaluated using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) pro [37]. The GRADE assessment evaluates the overall quality of evidence in four levels: Very low, low, moderate, and high. The level of evidence is degraded according to factors, such as risk of bias, inconsistency, indirectness, imprecision, and publication bias.

Study Selection
A total of 903 studies were selected through an electronic database search, of which 37 duplicates were removed. After screening for titles and abstracts, 743 studies were excluded for at least one of the following reasons: (i) No clinical trial, (ii) studies unrelated to EAHM, (iii) case reports or reviews, and (iv) irrelevant to PN. A full text assessment was performed on the remaining 123 studies, and 56 studies were excluded for the following reasons: (i) No clinical trials or quasi-RCTs; (ii) no oral administration; (iii) undisclosed herb ingredients; (iv) combination of interventions other than oral administration of EAHM; (v) inappropriate control groups; (vi) not related to PN; (vii) duplicated. A total of 67 studies were identified. The screening process is summarized in Figure 1.

Study Characteristics
Four RCTs were published in English, and the rest were published in Chinese. Four studies were conducted in Japan, whereas the others were conducted in China. The etiology of PN included studies of 50 DPNs, 11 CIPNs, one HPN, one occipital neuralgia, one trigeminal neuralgia, and one supraorbital neuralgia. The sample size of the included studies ranged from 29 to 247, and a total of 5753 participants were separated into the experimental group (n = 2898) and the control group (n = 2855). The treatment duration ranged from 2 to 26 weeks. The characteristics of the included 67 studies are summarized in Table 1.

Risk of Bias
The methodological quality of the 67 included studies is summarized in Table 2 and Figure 2. The risk of bias was assessed using the RoB 2.0 tool [26]. Four studies were assessed to have a "low risk of bias" and the remaining 63 studies were assessed to have a 'high risk of bias'.

Risk of Bias
The methodological quality of the 67 included studies is summarized in Table 2 and Figure 2. The risk of bias was assessed using the RoB 2.0 tool [26]. Four studies were assessed to have a "low risk of bias" and the remaining 63 studies were assessed to have a 'high risk of bias.'   Figure 3). In the 21 studies comparing the effect of combined EAHM and WM therapy with the WM monotherapy control, the combined therapy significantly improved SNCV than the WM monotherapy control (n = 4454; MD 3.06, 95% CI 2.56-3.56, p < 0.0001; heterogeneity chi-square = 317.64, df = 43, p < 0.01; I 2 = 86%; Figure 4). In the 21 studies comparing the effect of combined EAHM and WM therapy with the WM monotherapy control, the combined therapy significantly improved SNCV than

AEs
Of the total 67 studies included in this review, 26 studies reported adverse event monitoring [ [38,45,49,65,69,72,77,84,101] reported multiple AEs possibly related to EAHM, and five studies [88,[90][91][92][93] reported AEs unrelated to EAHM. No AEs were observed in the 12 studies [40,46,47,[53][54][55]59,61,62,75,76,100]. The number of patients with AEs was 28/1322 (2.12%) in the experimental group and 31/1296 (2.4%) in the control group. Seven studies (seven in experimental groups and six in control groups) reported that the most frequent AEs were gastrointestinal symptoms, including abdominal pain, diarrhea, abdominal bloating, nausea, vomiting, anorexia, xerostomia, diarrhea, and constipation [38,45,69,72,77,84,101]. Skin rash was reported as an adverse event related to the integumentary system in two studies [49,69] (one in experimental group and two in control groups). Dizziness was reported as an adverse event related to the nervous system in three studies [65,84,101] (two in experimental groups and three in control groups). In all of the included studies, no severe AEs, which were life-threatening or required treatment for a long period of time, were reported. The details of the AEs reported in each study are presented in Table 1. Table 3 summarizes the results of subgroup analysis based on individual causative diseases of PN and NCV for each site measured in five or more studies. There were no substantial changes in the results of the subgroup analysis.     Table 4.

Apriori Algorithm-Based Association Rule Analysis
Nine association rules were identified in the analysis based on the composition of the 67 EAHM prescriptions included in this study (Table 5). Subsequently, the distribution of the lift value was recognized through a scatter plot consisting of the association rule, with the support value on the x-axis and the confidence value on the y-axis (Figure 14). Subsequently, the distribution of the lift value was recognized through a scatter plot consisting of the association rule, with the support value on the x-axis and the confidence value on the y-axis (Figure 14). The color depth of each association rule, determined by its lift value, confirmed that the distribution of the overall lift value ranged from 1.276 to 1.937. Meanwhile, a grouping matrix diagram was presented to examine the overall distribution of the identified association rule (Figure 15).  The color depth of each association rule, determined by its lift value, confirmed that the distribution of the overall lift value ranged from 1.276 to 1.937. Meanwhile, a grouping matrix diagram was presented to examine the overall distribution of the identified association rule (Figure 15). The horizontal ordinate shows eight association rules, and the vertical ordinate shows the items created by the eight rules. In this diagram, the depth of the color inside the circle represents the degree of lift, and the circle size represents the degree of support. From Figures 14 and 15 Caulis} => {Cinnamomi Ramulus}. Therefore, the constituents of the herb combinations with consistent association rules were Astragali Radix, Cinnamomi Ramulus, and Spatholobi Calulis. The relationship between these association rules is presented through a network graph (Figure 16). The horizontal ordinate shows eight association rules, and the vertical ordinate shows the items created by the eight rules. In this diagram, the depth of the color inside the circle represents the degree of lift, and the circle size represents the degree of support. From Figures 14 and 15 Calulis. The relationship between these association rules is presented through a network graph ( Figure 16). R PEER REVIEW 34 of 45 Figure 16. Network graph of the association rules in the meta-analysis of East Asian herbal medicine prescribed for peripheral neuropathy.

Publication Bias
The contour-enhanced funnel plot analysis was performed to explore the publication bias through the response rate, which is an outcome of most of the included studies (Figure 17). The pattern of the funnel plot with 47 studies shows a clear asymmetry, indicating that there might have been publication bias ( Figure 11). This was further confirmed by Egger's test (p < 0.0001) and Begg's test (p < 0.0001).

Publication Bias
The contour-enhanced funnel plot analysis was performed to explore the publication bias through the response rate, which is an outcome of most of the included studies ( Figure 17). The pattern of the funnel plot with 47 studies shows a clear asymmetry, indicating that there might have been publication bias ( Figure 11). This was further confirmed by Egger's test (p < 0.0001) and Begg's test (p < 0.0001). gure 17. Contour-enhanced funnel plot for the meta-analysis of East Asian herbal medicine prescribed for periph europathy.

Quality of Evidence According to the Outcome Measurements
In the comparison between the combination EAHM and WM therapy and W otherapy, the overall quality of evidence according to all of the outcome measu low to moderate. Meanwhile, the overall quality of evidence according to all of come measures was low to moderate in EAHM monotherapy compared with WM therapy. The results of the GRADE assessment are listed in Table 6. Table 6. Summary of findings for the studies in this meta-analysis.

Quality of Evidence According to the Outcome Measurements
In the comparison between the combination EAHM and WM therapy and WM monotherapy, the overall quality of evidence according to all of the outcome measures was low to moderate. Meanwhile, the overall quality of evidence according to all of the outcome measures was low to moderate in EAHM monotherapy compared with WM monotherapy. The results of the GRADE assessment are listed in Table 6.

Summary of the Main Finding
In this systematic review, 67 RCTs including 5753 PN patients were obtained and analyzed. The main finding of this study is that EAHM monotherapy or combined EAHM and WM therapy was superior to the control group without EAHM in improving nerve conduction velocity, response rate, incidence rate, pain intensity, and other overall symptoms. Additionally, EAHM is generally safe and tolerable for PN patients. Therefore, EAHM can be considered a recommended option for PN treatment in clinical practice based on the evidence presented in this study. On the contrary, in the association rule analysis of various EAHM prescription data included in this study, Astragali Radix, Cinnamomi Ramulus, and Spatholobi Calulis were identified as components constituting the core herb combination. It may be worthwhile to conduct further studies on whether EAHM containing the three individual herbs or their combination can exert a remarkable effect in the PN-treated group.

Limitations
This review has various limitations. Therefore, caution is required before using the results. First, most of the studies were conducted in China. As a result, additional well-designed multicenter clinical trials in East Asia are needed to generalize the positive results identified by the analysis. Second, the methodological quality of the clinical trials included in this study was generally poor. The overall risk of evaluated bias according to RoB 2.0 reported that only four included studies have a 'low risk of bias.' All of the other studies have a 'high risk of bias' due to methodological flaws in domains, such as the randomization process, deviations from the intended intervention, and missing outcome data. Therefore, it is difficult to draw firm conclusions, even though the review contains relatively large sample data and primary trials. Rigorous conclusions regarding EAHM can be drawn only in well-designed clinical trials to minimize the risk of future bias. Third, a high level of heterogeneity was observed in the meta-analysis of NCV, which is one of the primary outcomes of this study. This high heterogeneity is a problem that cannot be overlooked, as it reduces the significance of the synthesized evidence. In this study, the cause of heterogeneity could not be identified, even though a subgroup analysis was performed according to the underlying disease and treatment duration. It is estimated that the cause of the estimated troubleshooting is that the NCV at the basement and the amount of change in participants are different for each included study. This reflects the difficulty in diagnosing and measuring PN severity. Another possible cause of heterogeneity is the extreme diversity in the composition and dose of EAHMs used in individual clinical trials. This leads to inconsistency among interventions, except for the commonality of 'the combination of herbal medicines in East Asia.' In this review, the association rule analysis was performed on herb data to overcome this heterogeneity and derive useful information. The potential heterogeneity may be partially overcome in similar future systematic reviews by actively utilizing data mining methods. Fourth, the goal of this study was to identify valuable candidates for drug discovery or locate material information that may be employed in direct patient treatment in the clinic. Therefore, it was not possible to focus on quality control in the manufacturing process, such as pre-treatment, active ingredient extraction methods, and moisture content assessment, all of which significantly impact the efficacy of specific goods. Moreover, this was suspected to have influenced the heterogeneity of the results. In the future, an animal study meta-analysis on the same issue will be used to compensate for these flaws.

Implications of Clinical Practices
The evidence related to the use of EAHM therapy for PN supported by this study is consistent with the results of previous studies on similar subjects. A study analyzing the clinical data for DPN after using 216 EAHM prescriptions found that the combination of Astragali Radix-Cinnamomi Ramulus and Ligusticum Chuanxiong-Moutan Radicis Cortex highly correlated with MDNS improvement [105]. Moreover, considering that Astragali Radix and Cinnamoni Ramulus are among the top 10 herbs utilized and their combination was identified as a core herb pattern, the findings of the previous study are similar to this study. A systematic review in 2016 evaluating the effectiveness of EAHM formulation containing Astragali Radix as a central component for CIPN, also demonstrated a significant effect on the effective percentage and NCV [13]. Unlike this review, which dealt only with oral administration, this study elucidates the effects of topical preparation and injection. The differences between the two studies suggest that similar EAHMs can be effective when applied to PN even if they are administered through various routes. In 2020, a meta-analysis evaluating the effect of EAHM foot bath on DPN [106] was published, which also reported a significant improvement in SNCV, MNCV, and response rate in DPN patients after the EAHM treatment. However, the herbs frequently used in this study were Cinnamomi Ramulus, Carthami Flos, Herba Speranskiae Tuberculatae, and Cnidii Rhizoma, indicating that they are almost irrelevant to the frequently used herbs described in this review. According to the comprehensive evidence of this study and related topics to date, it is relatively clear that using various EAHM forms in clinical practice can be a meaningful treatment for PN patients. However, both the administration route and data from individual studies must be considered to identify whether a specific formulation or herb can be an effective choice. In addition, it cannot be concluded from only a few clinical studies. Therefore, further studies should focus on the possible mechanisms related to this topic.

Implications of the Research
The characteristic of EAHM to treat complex diseases by stimulating many networks of human interaction systems at the systematic level through a multi-targeted approach is being investigated [107]. Therefore, the multicomponent-derived EAHM exerts a synergistic effect between multiple compounds in the process of acting on multiple targets, resulting in efficacy with decreased toxicity and side effects [107,108]. Therefore, for efficient EAHM utilization, it is important to consider the synergistic combination of herbs rather than the primary mechanism of individual herbs. In this regard, the principle of prescription using botanical medicine called "Kun-shi-Choa-sa" has traditionally been used to combine two or more herbal medicines in East Asia, and recently, the simplest form of multi-herbal mixture, "herb-pair" is also studied [109]. As the associated case, the combination of Astragali Radix and Angelicae Gigantis Radix, which are mostly used, has been reported to improve axonal growth by primarily stimulating the neurotrophic signaling pathway against damage to the central nervous system [110]. The authors of this study argued that the combination of two drugs through a network pharmacology and a methodology could promote neurological recovery by inhibiting the expression of NogoA by triggering a multipath pathway. In another case, studies on the combination of Cinnamomi Ramulus and Glycyrrhyziae Radix et Rhizoma, the herbs frequently featured in this review, showed significant differences in pharmacokinetic parameters compared to the use of each single herb [111]. Based on this mechanism, high peak concentration, slow elimination, and great exposure were observed in Cinnamomi Ramulus and Glycyrrhyziae Radix et Rhizoma. According to several studies reviewed to date, appropriate herbal combinations are highly likely to produce excellent pharmacological and pharmacodynamic results. As a research hypothesis to develop efficient EAHM-based drugs for PN in the future, some core herb patterns identified by the association rule analysis in this review are meaningful.
Information on the pharmacological action of individual herbs is also important for achieving the above purpose. Research on various pharmacotherapeutic targets is required for an effective drug treatment for diseases in which the overall pathology, such as PN, is not fully understood. Combining this basic study with the action-related information of the individual active ingredient of EAHM will make it possible to clearly predict the direction of the synergistic effect expected from multiple herb combinations. Even for DPN, research on several molecular targets, including the polyol pathway, hexamine pathway, PKC signaling, oxidative stress, AGEs pathway, PARP pathway, MAPK pathway, NF-κB signaling, TNF-α signaling, and cyclooxygenase pathway is conducted [112]. In this review, the mechanisms by which major herbs induce PN pathology through various pathways are included. First, Astragali Radix downregulates the phosphorylation of heavy neurofilaments to prevent axonal damage and suppress pain hypersensitivity by reducing astrocytes and microglia scattered in the spinal cord and brain [113]. In another study, the mechanism by which APS protects against nerve damage is through miR-138 upregulation in rat neural stem cells [114]. Cinnamomi ramulus not only exerts neuroprotective effects by reducing oxidative damage and MDA and NO production, but also significantly suppresses pain hypersensitivity associated with inflammation [115,116]. Total glucosides in Paeoniae Radix protect against neurotoxicity, lower the level of neuronal nitric oxide synthase, and exhibit anti-nociceptive activity related to calcium channels [117,118]. Spatholobi Caulis demonstrated the therapeutic effects on neurological disorder-associated cell death by inhibiting JNK and p38 MAPK activation and reducing oxidative stress and apoptosis in a rat model of induced middle cerebral artery occlusion [119]. As mentioned above, studies related to the mechanism of action of herbs theoretically support the clinical effect of EAHM on PN, as confirmed in this review. However, in addition to these individual mechanisms, experimental studies are needed to identify targets that can reproducibly exert the synergistic effects of EAHM. Furthermore, future studies on whether the combined effects of EAHM actually produce clinical results distinguishable from the additive effects of individual agents, need to be performed.

Challenges and Perspectives
The following problems must be considered until the aforementioned discoveries are meaningfully exploited in clinical practice and medication discovery. Natural medicines, including EAHM and synthetic drugs, have significantly distinct modes of action, target pathways, and pharmacologically active components from a macroscopic point of view. The most well-known difference is that multiple compounds present in herbal medicine operate on many targets and single compound synthetic medications work on single targets [109]. As demonstrated in this review, most of the EAHM prescriptions comprise a blend of several components in specific amounts, frequently in a single formula. In this instance, each component alone frequently does not demonstrate several therapeutic actions, such as the entire combination. The pharmacological activity of EAHM is thought to be due to the synergetic action of several chemical components targeting multiple sites and the simultaneous action of multiple chemical components targeting a single site [120,121]. This is thought to be the most significant difference between synthetic medicines and EAHM. As a result of these EAHM characteristics, it has been difficult to discover possible indications and mechanisms in the past. In addition, there has been a belief that it is difficult to derive social and medical contributions as much as synthetic medications. Recent scientific research, on the contrary, has indicated that the combinatorial effect of mixed EAHM preparation can be particularly effective for complicated disorders, such as PN, autoimmune disease, degenerative disease, and cancer, which do not react well with single compound-based modern pharmaceutics [121][122][123][124]. As an example, recent research has demonstrated that EAHM can be used in large-scale public health emergencies, such as COVID-19 or preventive medicine using modern analytical tools, such as synthetic biology, data mining, and genomics [123,[125][126][127][128]. Future studies need to be conducted to identify the properties of EAHM to be utilized in actual drug discovery.
First, this review was undertaken with the goal of finding EAHM materials that may provide prospective advantages to PN patients, and it does not go into detail regarding the formulation process of the materials. However, as mentioned above, estimating a consistent impact even for the same herb material in a condition where there is a lack of adequate consensus and discussion on the processing technique, pre-treatment, and extraction method of individual EAHM materials may be challenging [129,130]. Since most of the EAHM dosage forms discussed in this study are decoctions, it is also important to include various methods for determining the water content of the product [131]. These issues need to be addressed in a review of animal studies focusing on this subject. Simultaneously, it is thought that providing standardized information on the above in future EAHM-related clinical trials would help in enhancing the quality control of herbal materials. Second, while comparing EAHM to natural materials with efficacy against PN is outside the scope of this review, it is deemed necessary. For example, several clinical trials have gathered early data for Cannabis sativa. Moreover, additional materials, such as mulberry, Citrullus colocynthis, Matricia chamomilla, and Myristica fragrans have promising benefits for PN [9,132]. It is envisaged that relevant drug discovery information will be generated through a comparison of the phytochemical and clinical effectiveness with those of conventional herbal medicine in follow-up investigations.

Conclusions
This meta-analysis supports the hypothesis that EAHM monotherapy may be beneficial for PN patients. Moreover, the combined EAHM and WM therapy may be recommended for these patients. EAHM monotherapy improves severe pain intensity and abnormal sensations, such as tingling, burning, and numbness, which impair the quality of life in PN patients. Additionally, unlike the PN treatment with WM alone, which has a poor prognosis, a combination of EAHM and WM treatment alleviated the symptoms of PN including tingling, burning, and numbness and prevented chronic PN. However, high quality RCTs evaluating the effects of EAHM are needed due to limitations, such as heterogeneity, to understand this result clearly. In addition, it is worth conducting a follow-up study to verify the specific action target of the core herb combination derived from the present review and the hypothesis of superiority in clinical practice.