The Effect of Chinese Herbal Medicine on Albuminuria Levels in Patients with Diabetic Nephropathy: A Systematic Review and Meta-Analysis

To evaluate the effect of Chinese herbal medicine (CHM) on albuminuria levels in patients with diabetic nephropathy (DN), we performed comprehensive searches on Medline database, Cochrane Library, CNKI database, CBM database, Wanfang database, and VIP database up to December 2012. A total of 29 trials including 2440 participants with DN met the selection criteria. CHM was tested to be more effective in reducing urinary albumin excretion rate (UAER) (MD −82.95 μg/min, [−138.64, −27.26]) and proteinuria (MD −565.99 mg/24 h, [−892.41, −239.57]) compared with placebo. CHM had a greater beneficial effect on reduction of UAER (MD −13.41 μg/min, [−20.63, −6.19]) and proteinuria (MD −87.48 mg/24 h, [−142.90, −32.06]) compared with angiotensin converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARB). Combination therapy with CHM and ACEI/ARB showed significant improvement in UAER (MD −28.18 μg/min, [−44.4, −11.97]), urinary albumin-creatinine ratio (MD −347.00, [−410.61, −283.39]), protein-creatinine ratio (MD −2.49, [−4.02, −0.96]), and proteinuria (MD −26.60 mg/24 h, [−26.73, −26.47]) compared with ACEI/ARB alone. No serious adverse events were reported. CHM seems to be an effective and safe therapy option to treat proteinuric patients with DN, suggesting that further study of CHM in the treatment of DN is warranted in rigorously designed, multicentre, large-scale trials with higher quality worldwide.


Introduction
Diabetic nephropathy (DN), defined as the presence of micro-or macroalbuminuria in patients with diabetes, is the most common cause of end-stage renal disease (ESRD) across the world [1]. The prevalence of micro-and macroalbuminuria in patients with diabetes is as high as 37-40% in western countries and 57.4-59.8% in Asian countries [2][3][4]. Albuminuria is a well-established risk factor for cardiovascular disease and is also associated with ESRD [5,6]. Persistent albuminuria has toxic effect on tubular epithelial cells, causing tubulointerstitial inflammation and subsequent interstitial fibrosis. Angiotensin converting enzyme inhibitors (ACEI) and angiotensin receptor blockers (ARB) have been demonstrated to reduce albuminuria and delay the progression of DN by inhibition of renin-angiotensin system (RAS) and have become the standard of care for albuminuric patients [7,8]. Despite the renoprotective effects of ACEI and ARB, diabetic nephropathy progresses to ESRD in a large proportion of patients [9]. This indicates that in addition to the RAS, other pathways are involved in the pathogenesis of DN. Chinese herbal medicine (CHM), which can produce a potential effect of multitarget therapy and block these pathways, seems appropriate in the treatment of DN caused by multiple factors [10].
In traditional Chinese medicine, diabetic nephropathy is considered nearly equivalent to the term "Xiao Ke Bing, " which has been described in the "Yellow Emperor's Medicine Classic" (Chinese name in pinyin "Huang Di Nei Jing") more than 2000 years ago. Bawei Dihuang wan, originated from the "The Synopsis of Prescriptions of the Golden Chamber" in the Eastern Han Dynasty, is a famous Chinese herbal formula that has been used for a long time in the treatment of DN. In recent years, more and more herbal products are thought to be effective in reducing urinary protein in patients with DN. A number of randomised controlled trials (RCTs) have suggested that CHM alone or combined with ACEI/ARB has therapeutic potential in the treatment of DN in terms of reducing urinary albumin excretion, ameliorating proteinuria, and symptom improvement [11]. How about the effect of CHM on albuminuria alone or in combination with ACEI/ ARB as compared to ACEI/ARB? With a view to answering the question, the systematic review of randomized controlled trials evaluates the effects and safety of CHM on albuminuria in patients with DN.

Study Selection.
Studies were considered to be eligible for inclusion if they met all of the following criteria. (i) Patients included in the study were diagnosed with type 2 diabetes mellitus complicated with kidney disease, regardless of the stage of the DN (microalbuminuria defined as urine albumin excretion rate (UAER) of 20-200 g/min, or macroalbuminuria defined as UAER >200 g/min). (ii) The study was performed as a randomized controlled trial (RCT) describing a correct randomization procedure. Trials which used a clearly inappropriate method of randomization (e.g., open alternation) were excluded. (iii) The intervention of CHM included extract from herbs, single herbs, Chinese patent medicines, or a compound of herbs that was prescribed (individualized treatment) by Chinese practitioner. The control intervention included placebo or ACEI/ARB. Hypoglycemic therapy was used as a cointervention in both of the arms, including oral hypoglycemic drugs, insulin, and exercise. (iv) Outcomes included at least one of the following: urine albumin excretion rate, proteinuria, urinary albumincreatinine ratio, or urinary protein-creatinine ratio.

Data Extraction.
Two researchers independently extracted data, including study design, randomization, blinding and subject characteristics (e.g., age, sex, sample size, and albuminuria stage), and duration of treatment. Disagreements were resolved after discussion with other investigators.

Data Analysis.
Meta-analysis was carried out using Review Manager software (version 5.1), provided by the Cochrane Collaboration. The mean change in each study end point from baseline was treated as a continuous variable. Continuous data were presented as mean difference (MD), with 95% confidence interval (CI). The chi-squared test for heterogeneity was performed, and heterogeneity was presented as significant when 2 is over 50% or < 0.1. Random effect model was used for the meta-analysis if there was significant heterogeneity, and fixed effect model was used when the heterogeneity was not significant.

Search Results.
A total of 3937 publications were identified by both computer search and manual search of cited references. Of these, 1343 articles were determined to be duplicated. The remaining 2594 reports were retrieved in full text, of which 1991 were excluded on review of the titles and abstracts. After further reading, we excluded 530 for not describing randomization procedure, 25 non-ACEI/ARB or placebo comparators, 16 no outcome of interest, and 3 duplicated reports. Finally, a total of 29 studies were included in the meta-analysis. Figure 1 is a flow chart of study selection process.

Characteristics and Methodological Quality of Included
Trials. All 29 publications included were of a randomization procedure generated by a random number table or computer . Twenty-seven studies were published in Chinese and the other two in English. Numbers of participants of the individual studies varied from 40 to 409 with a total of 2440 participants included in this paper ( Table 1). The majority duration of treatment varied from one month to three months.
The Jadad scale is a 5-point scale for assessing the quality of RCTs in which three points or more indicate superior quality [41]. Of the 29 RCTs, 11 trials were of superior quality according to the Jadad score (≥3 points) [12,15,17,21,24,25,32,34,[38][39][40]. All studies described a correct randomization procedure, but only one of them mentioned allocation concealment [39]. Three out of 29 studies described blinding of participants [12,39,40]. Ten trials reported the dropouts information and mentioned follow-up, but this dropouts were not captured in the analysis [12,15,17,21,24,25,32,34,38,39]. Among all trials, the characteristics of participants in different treatment groups were similar at baseline (age, sex, race, and disease course).

Analysis of Chinese Herbal Medicine.
A total of 84 different kinds of herbs were included in 29 herbal preparations for treatment of DN. In Table 2, we listed the 14 herbs that were included most frequently in the 29 herbal preparations. For example, the herb used most often, Astragalus membranaceus (Huang Qi), was used 22 times in 29 different herbal preparations; the herb used second frequently, Salvia miltiorrhiza (Dan Shen), was used in 15 of 29 herbal preparations. Each compound prescription contained an average of 9 ingredients (range: [2][3][4][5][6][7][8][9][10][11][12][13][14]. The formulations of CHM were different and included tablet, capsule, oral liquid, and decoction.   Evidence-Based Complementary and Alternative Medicine       [12]. These symptoms could be tolerated by patients. One patient stopped the treatment of Tripterygium glycosides due to leucopenia [17]. Among 38 patients treated with Pishen Shuangbu tang, one patient developed mild diarrhoea, and one developed dizziness [19]. The symptoms were relieved after stopping the treatment. One patient developed mild diarrhea after taking Tangshen fang [24]. Adverse effects in ACEI/ARB treated patients included dry cough, hyperkalemia, and doubling of serum creatinine [15,17,19,28,35,39]. There was no significant difference between herbal treatment and ACEI/ARB regarding the incidence of adverse effects. No serious adverse events were reported.

Discussion
Based on the meta-analysis of 29 randomized controlled trials, CHM was tested to be more effective in reducing UAER and proteinuria compared with placebo or ACEI/ARB. Combination therapy with CHM and ACEI/ARB showed significant improvement in UAER, urinary albumin-creatinine ratio, protein-creatinine ratio, and proteinuria as compared to ACEI/ARB. It should be noted that there were no reported serious adverse events associated with CHM studied. To summarize, the results revealed that CHM is an effective and safe therapy option to treat albuminuric patients with DN. In TCM, diabetic nephropathy referred to as an intrinsically deficient but extrinsically excessive syndrome. Deficiency of qi and yin, and excess of stasis and dampness are believed to be the main mechanism responsible for development of DN [42]. Among the included 29 RCTs, 29 different herbal preparations were tested, including four extracts from a single herb, one Chinese patent medicine, and 24 Chinese herbal compound prescriptions. Of the 24 compound prescriptions, Bushen Huoxue decoction, Pishen Shuangbu tang, and modified Liuwei Dihuang tang were prescribed based on Liuwei Dihuang tang, which has the function of nourishing the kidney yin. A total of 84 different Zhang et al. 2011 [21] Zhong et al. 2012 [26] Total (95% CI)   kinds of herbs were included in 29 herbal preparations for treatment of DN. From the results of frequency distribution of categorized herbs according to their functions, herbs with qi-tonifying and yin-nourishing, blood-activating and stasisresolving, kidney-replenishing and water-draining appeared to be most frequently prescribed for the treatment of DN. The pathogenesis of diabetic nephropathy is complex and not yet fully clarified. In addition to the RAS, other pathways such as oxidative stress, inflammation, and excessive production of advanced glycation end products also contribute to the development of DN [43][44][45]. Therefore, although use of RAS antagonists appears to slow the progression of DN development to ESRD, it does not stop or reverse the pathology. Each herbal product within the TCM formulations could have several different active ingredients to attack a disease process in manifold ways. For example, astragalus polysaccharide has prophylactic and therapeutic effects on the progress of DN by decreasing the mRNA level of NF-B in renal cortex and increasing IkB mRNA expression in rats [46]. Additionally, the antioxidative effect of Astragalus membranaceus as a free radical scavenger implies its protective effect in the early stage of DN [47]. Salvia miltiorrhiza could be applicable for the treatment of DN by reducing the serum and kidney levels of transforming growth factor 1 (TGF-1) and the kidney levels of collagen IV, monocytes/macrophages (ED-1), and the receptor for advanced glycation end-products (RAGE) [48]. Corni Fructus has the potential to protect the animals from diabetic nephropathy by amelioration of oxidative stress and stimulation of PPAR expression [49]. These studies' results suggest that CHM can produce a potential effect of 8  Figure 4: CHM plus ACEI/ARB versus no treatment plus ACEI/ARB. multitarget therapy, which seems appropriate in the treatment of DN caused by multiple factors. It must be acknowledged, however, that the methodological quality of the trials evaluating the effect of CHM on DN was generally not high: 18/29 (62%) of the RCTs included in this review were scored as having mediocre methodological quality [Jadad scores = 2]. No trial was identified as a multicenter, large sample, prospective, double-blinded, controlled randomized trial. Furthermore, most of the studies did not report about allocation concealment process, which may have created potential selection bias. The possibility of publication bias in the reporting of RCTs is always of concern. Although we performed comprehensive searches and tried to avoid bias, since most of the studies were published in Chinese, there remained the possible existence of publication bias.
It is noteworthy that discrepancy in the herbal composition, drug formulation, and dose was observed between the studies, which may be the source of heterogeneity in the included RCTs. TCM formulas were composed of many herbs and the content and biological activities of these herbs can be influenced by many things, including where the herb was grown, and at what season it was harvested. Consequently, CHM for treating DN needs to equip standardized criteria for use to ensure the good reproducibility of the research result in real clinical practices.
The results of the present review provide strong evidence of the efficacy of CHM in reducing UAER, proteinuria, urinary albumin-creatinine ratio, and protein-creatinine ratio, suggesting that CHM can be used as an alternative therapy for the treatment of DN. However, majority of included studies were scored as having mediocre methodological quality. Future clinical trials of CHM on DN need to improve methodological quality and reported well according to the CON-SORT statement [50]. Hence, we conclude that further study of CHM in the treatment of DN is warranted in rigorously designed, multicentre, large-scale trials with higher quality worldwide.