An Investigation of the Prescription Patterns of Chinese Herbal Products for Chronic Glomerulonephritis Patients: A Hospital-Based Cross-Sectional Study

Chronic kidney disease (CKD) has a high incidence and prevalence worldwide, and chronic glomerulonephritis (CGN) is one of the main causes of CKD. Therefore, it is important to diagnose and treat CGN early. The purpose of this study is to analyze the prescription patterns and frequencies of Chinese herbal products (CHPs) for CGN by using a hospital-based database from the Chang Gung Memorial Hospital (CGMH), a large, tertiary hospital system in Taiwan, and to evaluate the safety and possible efficacy of CHPs by blood test. The International Classification of Disease Ninth Revision (ICD-9) code 582 was used to identify patients with CGN. From 2004 to 2015, a total of 54726 CHP prescriptions for CGN were provided. Association rule mining was used to analyze the prevalent of CHP combination patterns in treating CGN. Jia-Wei-Xiao-Yao-San (JWXYS) and Gorgon (Euryale feroxSalisb.) were the most frequently prescribed herbal formula (HF) and single herb (SH), respectively. The most frequently prescribed combination of CHPs was that of JWXYS with Bu-Yang-Huan-Wu-Tang (BYHWT) in CGMH. In statistical, the level of eGFR in Stage 3a and 3b group was increasing after treatment in 6 and 12 months and might not cause the renal function to worsen within 12-month treatments. To the best of our knowledge, this is the first pharmacoepidemiological study to review CHP treatments for CGN. However, additional studies and clinical trials are needed to provide data on the safety and efficacy of these CHPs.


Introduction
In the aging society, the proportion of patients with hypertension, diabetes, and other chronic diseases is increasing. The incidence and prevalence of chronic kidney disease (CKD) are very high worldwide [1]. There are no obvious symptoms of CKD, so patients are not aware that they have CKD.
According to a study from the Taiwan National Institutes of Health in 2008, 11.9% of people over the age of 20 in Taiwan had CKD, but only approximately 10% of them knew that they were suffering from kidney disease [2]. CKD was defined by the National Kidney Foundation Kidney Disease Outcomes Quality Initiative, and the definition was modified by the international guidelines of a group known as the 2 Evidence-Based Complementary and Alternative Medicine Kidney Disease Improving Global Outcomes. The diagnostic standards of CKD include kidney dysfunction or structural damage over a period of three months, and an impact on health. In the Kidney Disease Outcomes Quality Initiative research in 2012, the most common causes of CKD were chronic glomerulonephritis (CGN), diabetic nephropathy, and hypertensive renal sclerosis [3]. Caring for patients with CKD is a big burden for the national economy. Therefore, early diagnosis, treatment, and prevention as well as slowing the progression of CKD and CGN are very important.
CGN is a group of diseases that occur in the glomerulus. The mechanism of CGN is considered to be a series of inflammatory reactions caused by complex deposition that activate the complement system. There are many causes of CGN, such as bacterial and viral infections, immune responses by other systemic diseases, environmental toxins, drugs, cancers, thyroid disease, diabetes, and hepatitis B [4].
The current medications for CKD in patients with drug treatment are focused on drugs that slow down the residual renal function and those that avoid residual renal function deterioration and side effects. Treatment for CGN can include steroids and immunosuppressive agents, and partial patients can use angiotensin converting enzyme inhibitors (ACEi) and angiotensin II receptor blockers (ARB) to relieve microvascular pressure in the kidney to reduce proteinuria and improve CKD. However, the clinical applications of CGN drugs are limited by numerous adverse side effects such as poor control of diabetes mellitus, high levels of lipids in the blood, hypertension, insomnia, obesity, edema, osteoporosis, peptic ulcer, infection, hyperkalemia, and angioedema [4,5].
In recent years, complementary and alternative therapies have become increasingly popular for their potential efficacy and a small number of side effects for treating CKD [6][7][8]. The herbal formula (HF) known as Liu-Wei-Di-Huang-Wan (LWDHW) is reported to reduce the risk of dialysis and prolong the time to dialysis in CKD patients [6]. The CHPs of wind dampness-dispelling formulas and harmonizing formulas may have the effect of protecting kidney function before dialysis [7]. In CGMH, research has found that combinations of Jia-Wei-Xiao-Yao-San (JWXYS) and Bu-Yang-Huan-Wu-Tang (BYHWT) may improve the quality of life in CKD patients [8]. According to the traditional Chinese medicine (TCM) theory, TCM physicians evaluate the conditions of patients to prescribe one or more herbal formulas combined with several SHs for each prescription. However, few studies have reported on CGN in TCM. The aim of this study is to analyze the CHPs prescribed and the coprescription patterns for treating CGNs by using a hospital-based database and follow up the eGFR to evaluate the safety and possibly efficacy of CHPs coprescription patterns. Our hope is that these findings can provide reliable information on adjunctive therapy for CGN in the future.

Data Source.
We analyzed a sample of patients from CGMH Taipei, Linkou, Taoyuan Districts in Taiwan and determined the prevalence of prescribed CHPs in patients with CGN from 2004 to 2015. The electronic database of all claims obtained from the CGMH website included medical record files containing the patient's gender and date of birth, date of medical visits, medical care facilities and specialties, prescription drugs, management and treatment, unique identification numbers, and all diagnoses coded according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) format.

Study Subjects.
The study cohort comprised patients diagnosed with CGN (ICD-9-CM codes:582) from 2004 to 2015; information was obtained from TCM outpatient visit records. All CGN-related medical records were analyzed during the study period. In the CGN cohort, patients who had at least one CHP were defined as CHPs users. Those who underwent kidney dialysis treatment were excluded. Patients accepted blood test before treatment and after treatments for 6 and 12 months, totally three times. We observe eGFR in blood according to different stages: Stage 1-2 as eGFR ≥ 60 mL/min/1.73m 2 , Stage 3a as eGFR 45-59 mL/min/1.73m 2 , Stage 3b as eGFR44-30 mL/min/1.73m 2 , Stage 4 as eGFR 29-15 mL/min/1.73m 2 , and Stage 5 as eGFR< 15 mL/min/1.73m 2 .

Study Variables.
Information on patient characteristics included the patient's age, gender, and comorbidities. Patients were divided into three age groups as follows:<35years old, 35-55 years old, and >55 years old. Comorbidities commonly seen in kidney disease patients were chosen, including hypertension, diabetes, hyperlipidemia, gout, cardiovascular disease, ischemic heart disease, atherosclerosis, diseases of the digestive system, systemic lupus erythematosus, and rheumatoid arthritis. These comorbidities were considered as covariates in modeling.
Prevalence visits and average daily dose (g) of CHPs contained in prescriptions were the variables used in this study to identify single agents and combinations of CHPs commonly used for CGN. Quality control of CHPs is assured as SHs and HFs are all produced as concentrated powders by Good Manufacturing Practice pharmaceutical factories with advantages of high safety and stability, convenient use, and reduction in material change problems caused by poor storage.

Ethics Statement.
We obtained ethical approval for this study from the Institutional Review Board of the Chang Gung Memorial Foundation (CGMF) (IRB No: 201601457B0). Since deidentified data were used, a waiver of informed consent was granted by the Institutional Review Board of the CGMF.

Statistical
Analysis. Statistical analysis was performed using SAS Enterprise Guide 4.3 (SAS Institute Inc., Cary, NC, USA), which was used for data linkage and descriptive statistical analysis of drug utilization patterns. Figures were created using Sigma Plot 12.0 (Systat Software Inc., USA). Using paired-samples t test and P<0.05 indicated statistical significance between groups. TCM doctors commonly wrote a prescription of more than two CHPs to treat patients. Association rule mining (ARM) was used in this study to

Demographic Characteristics of CHPs Users.
We used the data from 2004 to 2015 of patients who had an ICD-9 code of 582 from the Taipei, Linkou, and Taoyuan Chang-gung districts. A total of 54726 CHP prescriptions used by 4438 CHPs users were included in this study. Patient demographic included age, gender, and potential comorbidities related to CGN, as shown in Table 1. The common comorbidities of patients with CGN receiving CHPs treatment were diseases of the digestive system, hypertension, diabetes mellitus, and hyperlipidemia.
3.3. Core Treatment for CGN. JWXYS was the core treatment for CGN because it accounted for the highest prevalence among all CHPs. Additionally, the central role of JWXYS can be found in the network of commonly used coprescriptions of CHPs ( Figure 2).  Evidence-Based Complementary and Alternative Medicine Figure 2: Social network analysis on associations between commonly used CHPs for CGN. JWXYS was the center of the CHPs network prescribed for CGN and became the core treatment for CGN. Larger circle represents higher frequency of prescription, and thicker connection line represents more common combination.  Table 4: The top ten most frequently used two Chinese herbal products (CHPs) in combination for chronic glomerulonephritis (n = 54726). Evidence-Based Complementary and Alternative Medicine 7 nP value for the use of CHPs before treatment and after treatment within 3-6-9 months. The P value < 0.05, indicating significant differences.  Figure 3 is the eGFR change according to the eGFR level before the initial treatment. All groups showed a significant upward trend within 6 months after CHPs use. Only the Stage 3a and 3b groups continued to increase significantly within 12 months.

Modern Blood Biochemical
Test. According to the above, the most commonly prescribed combination of two herbal formulas for CGN was JWXYS with BYHWT. In this study, 386 patients have taken JWXYS and BYHWT and accepted blood tests at least three times. According to different eGFR stages, we observe blood tests before treatment and after treatments for 6 and 12 months. Data analysis and chart are in Table 5 and Figure 3.

Discussion
This study is the first hospital-based cross-sectional study on CGN with CHPs treatment and did not include utilization of health services provided by other hospitals or clinics. By using a hospital-based database, we can not only collect all of the prescriptions but also ensure the quality of CHPs. In addition, the detection of comorbidities is more extensive. Comorbidities with CGN, such as diabetes, hypertension, and cardiovascular diseases, may affect each other and induce many side effects during treatment. Therefore, doctors also relieve patients' comorbidities and uncomfortable symptoms, such as loss of appetite, nausea, vomiting, fatigue, and edema and improve the quality of daily life.

Commonly Used Herbal Formulas and Combinations of CHP for CGN. JWXYS in modern pharmacological research
has been found to have anti-inflammatory, antitumor, antifibrosis, anticoagulant, and antiatherosclerotic effects. JWXYS has also been found to regulate immunity, reduce glycemic and lipid levels, induce analgesia, relieve emotional and neuropsychological disorders, and inhibit the central nervous system [29][30][31][32][33][34]. The antifibrotic effects and reduction in glycemic and lipid levels may improve kidney function by protecting blood vessels and microcirculation. CKD Stage 5 patients have high tendency of depression [35,36]. JWXYS has been shown to have antidepressive and antianxiety effects, which may help patients stabilize their emotions and improve their quality of life [37]. Some CKD and CGN patients have digestive syndromes such as a loss of appetite, nausea, and vomiting. The incidence of gastrointestinal (GI) symptoms in patients with renal dialysis has been found to be high, and patients can have nonulcerative and nonvaricose GI bleeding [38]. JWXYS may also regulate GI function to relieve symptoms. In the Chinese medicine theory, JWXYS is used in patients with the pattern of liver depression and spleen deficiency and transforms into heat. Indications include emotional disorders, GI problems, insomnia, palpitations, upset, dry mouth, and difficulty and pain in micturition. Clear studies on the relationship between JWXYS and CKD and CGN are still lacking. However, the eGFR has an increasing trend in Stage 3a and 3b group in this study. Therefore, JWXYS may be chosen as adjuvant therapy for Stage 3a and 3b group in CKD and CGN based on the above research and can be used in the future to do more complete and rigorous clinical trials to evaluate the effects on renal function and find the possible mechanisms.
Modern pharmacological research has found that BYHWT may protect nerves and vascular endothelial cells, regulate immunity, have anti-inflammatory, antiplatelet 8 Evidence-Based Complementary and Alternative Medicine aggregation, and antithrombotic effects, expand peripheral blood vessels, improve hemodynamics, and promote microcirculation [39][40][41][42][43]. BYHWT may regulate the performance of ICAM-1and VCAM-1 in diabetic mice to relieve inflammation and may reduce the levels of nitric oxide (NO). Therefore, BYHWT could improve renal microcirculation and alleviate pathological renal damage [19,20].The main ingredient of BYHWT, Astragalus membranaceus (Fisch.) Bge., may have effects on reducing blood lipid levels and platelet aggregation rates and can also inhibit thrombosis by regulating NO synthase expression and inhibiting endothelial NO synthesis in some studies. BYHWT may also improve renal hemodynamics, protect renal function, and enhance the immune response that could reduce urine protein [11,12,44]. This finding may also be the potential reason for high rates of BYHWT use in CKD and CGN. In the Chinese medicine theory, BYHWT is used in patients with the pattern of qi deficiency with blood stasis. Indications include fatigue, micturition, enuresis, hemiplegia, and aphasia after stroke. A few studies have indicated that BYHWT can improve renal circulation and reduce urine protein, but these findings need more evidence to confirm the efficacy and possible mechanisms in the future.
JWXYS used with BYHWT is an example of a pair of HFs that increase treatment efficacy, which may be due to the effect of the two drugs acting through different modern pharmacology mechanisms to improve renal microcirculation and protect the kidneys. This theory may explain why JWXYS is used in combination with BYHWT and why they are the two most frequently prescribed HFs in combination to treat CGN.

Commonly Used Single Herb for CGN. Salvia miltiorrhiza
Bunge, Astragalus membranaceus, and Rheum Palmatum L. are some of the top 10 most commonly used SHs. Angelica sinensis (Oliv.) Diels and Ligusticum striatum DC. are some ingredients of top ten HFs. These five SHs have effects such as antioxidation, antilipid peroxidation, antiplatelet aggregation, anti-inflammatory and antifibrotic effects, immune regulation, and regulation of renal vessels [45]. The pharmacological mechanisms of these findings may prevent the deterioration of CKD.

Modern Blood Biochemical
Analysis. In the study, the level of eGFR was significantly increasing within 6 and 12 months in statistics for Stage 3a and 3b group. However, only within 6 months it had significant increase in statistics for Stage 4 and 5 group. This finding for Stage 4 and 5 patients might have association with their renal function decline, disease progression, or using more drugs leading to drug interaction.
In this study, we observe that patients in Stage 3a and 3b with CHPs might not decrease eGFR level group. It is presumed that the use of CHPs treatment within 12 months may be safe and might not cause the renal function to worsen in Stage 3a and 3b group. However, more researches are needed in the future.

Effectiveness and Safety of CHPs.
Effectiveness and safety are the two important goals of drug use. The safety evaluation of CHPs often comes from personal observations, attempts, and clinical experience by ancient Chinese medical scientists and now could also be evaluated by modern pharmacological research and blood biochemical test. Table 6 shows the analysis of commonly used CGN treatments in CHPs pharmacological research, and there were no uncomfortable symptoms found. However, further research is needed in the future to assess the safety and effectiveness of CHPs.

Limitations.
There are some limitations to our study. First, this study restrictively used ICD-9 code 582 as the means of enrolling subjects only in the Chinese outpatient setting. We do not compare with the western medicine group. Meanwhile, we do not discuss whether the use of western medicine and the CHPs at the same time may affect the changes in the blood biochemical analysis, causing deviations in the evaluation of the results. Therefore, further well-designed, randomized, double-blinded, and placebocontrolled clinical trials are still needed to elucidate this problem.
Second, this study could not track the health outcomes of patients over time and could not rule out whether patients had other comorbidities along with CGN, including chronic diseases, pain, colds, or other complications. Whether these findings would influence the renal function and the data analysis on CHPs is unclear. Therefore, further studies that evaluate the efficacy and safety of CHPs in the CGN population are needed.

Conclusions
In conclusion, the most commonly prescribed combination for CGN in CGMH was that of JWXYS with BYHWT. After CHPs treatments, Stage 3a and 3b patients' eGFR level was increasing within 6 and 12 months and might not cause the renal function to worsen within 12 months treatments, but these still needed more evidences to confirm. The results of this study provide several directions for further cohort studies to elucidate the efficacy, safety, and drug-herb interactions of these commonly used CHPs among CGN patients.

Bu-Yang-Huan-Wu-Tang
(1) Suppresses the inflammatory cytokines and reactive oxygen species in the kidney of rats after induction of brain death [19]. (2) Lowers the index number of kidney hypertrophy (KW/BW) and the alleviate injury of the rat's kidney induced by proteinuria, lowers the quantity of nitric oxide, and reduces the damage of the kidneys, downregulates the expression of ICAM-1,VCAM-1 in renal tissue on diabetic kidney disease rats model [20]. (3) Inhibits angiogenesis and metastasis by affecting the expression levels of VEGF, RGS-5, and HIF-1 on nude mice model [21]. Du-Huo-Ji-Sheng-Tang (1) Anti-inflammatory effect and strikingly suppressed the expressions of Rho/NF-B signaling pathways as well as TGF-/smad-3 pathways on rat model [22,23]. (2) Promotes osteogenic differentiation and decreases the aging process of human mesenchymal stem cells in vitro model [24]. Zhi-Bo-Di-Huang-Wan (1) Enriches neuroendocrine immune pathways for immune-modulatory effects thus resulting in nephritis improvement [6,25]. (2) Antioxidant capacity by increasing SOD and GSH [26].
KDOQI: Kidney Disease Outcomes Quality Initiative ACEi: Angiotensin converting enzyme inhibitors ARB: Angiotensin II receptor blockers LWDHW: Liu-Wei-Di-Huang-Wan GMP: Good Manufacturing Practice ARM: Association rule mining GI: Gastrointestinal NO: Nitric oxide ESRD: End stage renal disease AKI: Acute kidney injury.

Data Availability
The data used to support the findings of this study are included within the article.

Disclosure
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of Interest
All authors declare that there are no conflicts of interest regarding the publication of this article.

Authors' Contributions
Wen Chen and Jiun-Liang Chen designed the experiments, performed data analysis and statistics, and drafted the manuscript. Hsing-Yu Chen and Yao-Hsu Yang helped with the data analysis and statistics. Jin-Wei Yang and You-Hung Wu participated in the study design and helped with the data analysis. Sien-Hung Yang and Jiun-Liang Chen provided professional comments for this work. All authors read and approved the final manuscript.