Combined Therapy of Hypertensive Nephropathy with Breviscapine Injection and Antihypertensive Drugs: A Systematic Review and a Meta-Analysis

Objective To evaluate the beneficial and adverse effects of breviscapine injection in combination with antihypertensive drugs for treating hypertensive nephropathy in clinical practice. Methods We searched PubMed, the Cochrane Library, Embase, CNKI, Sino Med, VIP, and Wanfang Data for relevant literature. The timeframe of retrieval was set from the founding date of each database to September 28, 2018. Results Fourteen papers were included in this study. The quality of all the studies included was determined to be low. All studies were conducted with Chinese populations. Meta-analysis showed that, compared with single-use antihypertensive drugs, using breviscapine injection in combination with antihypertensive drugs to treat hypertensive nephropathy can reduce serum creatinine (Scr) [WMD = –35.16, 95% CI(–50.01, –20.31), P ≤ 0.001], blood urea nitrogen (BUN) [WMD = –2.00, 95% CI(–3.07, –0.94), P ≤ 0.001], 24-hour urinary total protein (24 h UTP) [WMD = –0.04, 95% CI(–0.05, –0.02), P ≤ 0.001], and the beta-2-microglobulin (B2M) [WMD = –0.09, 95% CI(–0.11, –0.07), P ≤ 0.001], improve creatinine clearance rate (Ccr) [WMD = 7.84, 95% CI(5.20, 10.49), P ≤ 0.001], and increase the clinical efficacy [RR = 1.27, 95% CI(1.05, 1.53), P = 0.014], but does not lower systolic blood pressure (SBP) [WMD = –1.02, 95% CI(–2.88, 0.84), P = 0.281]. There was no significant difference in adverse events between experimental groups and control groups. Conclusion Breviscapine injection in combination with antihypertensive drugs can improve clinical efficacy and Ccr and reduce Scr, BUN, 24 h UTP, and B2M in patients with hypertensive nephropathy. The present meta-analysis indicated that breviscapine injection can serve as a renal protective effect to patients with hypertensive nephropathy. However, the evidence of methodological quality and sample sizes is weak, and thus, further standardized research is required.


Introduction
Hypertension is a risk factor for stroke, coronary artery disease, heart failure, peripheral vascular disease, and chronic kidney disease (CKD) [1,2]. Elevated blood pressure (BP) was the leading global contributor to premature death in 2015, accounting for almost 10 million deaths and over 200 million disability-adjusted life years [2]. The main factors leading to the development of hypertensive nephropathy include (1) inappropriately elevated sympathetic nervous activity (SNA) [3]; (2) activation of the renin-angiotensin-aldosterone system (RAAS) [4]; (3) increased arterial stiffness [5]; (4) genetic susceptibility [6]; (5) impaired salt and water excretion by the kidney [7]. Hypertensive kidney disease is the second leading cause of end-stage renal disease (ESRD) after diabetes mellitus [8,9]. In Europe, according to the European Dialysis and Transplant Association registry, hypertensive nephropathy is accounted for 12% of new patients starting renal replacement therapy. However, the reported incidence varies among different countries, with France, Italy, and United Kingdom, reporting in 25%, 17%, and 6.1%, with both Japanese and Chinese reporting in 6% and 7%, respectively [10]. Chronic kidney failure is a global disease: at the end of 2016, approximately 3 million patients were 2 Evidence-Based Complementary and Alternative Medicine on dialysis. The incidence of chronic kidney failure varies between regions, North America 0.638 million, Europe, Middle East, and Africa 0.711 million, Asia-Pacific 1.343 million, and Latin America 0.288 million [11]. The diagnosis of hypertension-induced renal damage is based on the finding of reduced renal function and/or the detection of albuminuria (≥300mg/d, or ≥300mg/g albuminuria to-creatinine ratio in the first morning void). CKD is classified according to estimated glomerular filtration rate (eGFR), calculated by the 2009 CKD-Epidemiology Collaboration formula [12]. Current evidence suggests that, in patients with CKD, BP should be lowered to less than 140/90 mmHg and towards 130/80 mmHg. The combination of renin-angiotensin system blockers with calcium channel blockers (CCB) or diuretics should be used to achieve recommended blood pressure targets in CKD [9,13]. A recent meta-analysis has shown that BP lowering significantly reduced ESRD in patients with CKD, but only in patients with albuminuria, and had no beneficial effect on cardiovascular events [14]. In a large retrospective cohort containing 398419 treated hypertensive patients, the nadir systolic blood pressure (SBP) and diastolic blood pressure (DBP) for the lowest risk of ESRD and mortality were 137 and 71 mmHg, respectively, with a clear increase in mortality risk at SBP less than 120 mmHg [15]. The evidence with respect to BP targets in patients with CKD is complex. Lowering blood pressure may lead to a decrease in eGFR. The reduction of albuminuria is also considered to be a therapeutic target. However, there are also studies in which treatment that was less effective at reducing albuminuria was more effective at reducing cardiovascular events [16].
Breviscapine (Dengzhanhua) injection is extracted from Erigeron breviscapus (Vant.), Erigeron breviscapus also known as herba erigerontis or lamp chrysanthemum, is a traditional Chinese herb that has been in use for more than 600 years, found in Yunnan, Sichuan, Guizhou, and other southwest provinces of China. Breviscapine, as a purified flavonoid extract from this species, was first isolated by Zhang et al. [17]. Breviscapine mainly contains scutellarin (4 ,5,6,7tetrahydroxyflavone-7-O-glucuronide) and apigenin-7-Oglucuronide [18]. Studies have shown that breviscapine has significant effects on vasodilation; inhibition of platelet aggregation, scavenging free radicals, also has a protective effect on myocardial and endothelial structures because of its antiinflammatory effects and improves microcirculation; protection against ischemia/reperfusion (I/R); anticoagulation and antithrombosis; reduction of smooth muscle cell migration and proliferation; anticardiac remodeling; antiarrhythmia; and reduction of blood lipids [19][20][21][22]. Breviscapine has been demonstrated to possess a number of pharmacological functions in addition to its hemodynamic effects; it has been reported to serve as an antioxidative stress agent and a protein kinase C (PKC) inhibitor, can inhibit the glycogen synthase kinase 3 (GSK3 ) signaling pathway to promote neurobehavioral function following neurotrauma, and can improve renal function and reduce urinary microalbuminuria [23][24][25][26]. In the light of these pharmacological activities, an injection preparation of breviscapine (a traditional Chinese patent medicine) has been wildly used in clinical treatment for cerebral infarction, cardiovascular disease, diabetic nephropathy, renal impairment of essential hypertension, and stroke in China [18,25,27,28].
However, in the past decades, although numerous studies have compared breviscapine injection with antihypertensive drugs in the treatment of hypertensive nephropathy, the comparability of treatment protocols and evaluation methodologies among these studies remains to be proven, which greatly limits their clinical applicability. Furthermore, the current state of evidence of breviscapine injection for hypertensive nephropathy has so far been unknown. Therefore, we conducted this systematic review to evaluate beneficial and adverse effects of breviscapine injection in the treatment of hypertensive nephropathy.

Methods and Analysis
. . Search Strategy. We designed our systematic review and meta-analysis in accordance with the guidelines of the 2009 Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) statement. Electronic network databases were searched via computer. Foreign databases searched included PubMed, Embase, and the Cochrane Library. Chinese databases included the China National Knowledge Infrastructure (CNKI), China Biology Medicine Disc (Sino Med), the VIP information resource integration service platform (VIP), and the Wanfang Data knowledge service platform (Wanfang Data). The retrieval scheme was mainly based on a combination of subject words and free words. The searched words were "Dengzhanhua", "Dengzhanhua preparations", "Dengzhanhua Zhusheye", "Zhusheyong Dengzhanhua", "Gaoxueya Shenbing", "Gaoxueya Shenyan", "Gaoxueya Shenshunhai", while the searched English words were "Breviscapine", "Breviscapine Injection", "BVP", "Hypertensive Nephropathy", "Hypertension, Renal", "Hypertensive Kidney Lesion", "Hypertensive Renal Damage" and so on (Supplementary Tables S1 and S2 for the search strategy). The retrieval language was not limited, and the timeframe of the retrieval was from the founding date of each database to September 28, 2018. There was no language limitation. Manual searches of relevant literature supplemented the electronic search.

. . Inclusion and Exclusion Criteria
. . . Types of Studies. Randomized controlled trials (RCTs) that use breviscapine injection in combination with antihypertensive drugs to treat hypertensive nephropathy, regardless of blinding, were used in this study. Language was not restricted as to minimize publication bias.
. . . Types of Participants. There were no serious organic diseases or complications in the selected cases. The diagnosis of hypertension-induced renal damage is based on the finding of reduced renal function and/or the detection of albuminuria. CKD is classified according to eGFR, calculated by the 2009 CKD-Epidemiologyn Collaboration formula [12]. Hypertension was defined as SBP ≥ 140 mmHg or DBP ≥ 90 mmHg based on the Chinese Guidelines for the Prevention and Treatment of Hypertension (2010) and the 8th session of the Committee, Report of the Joint Commission/JNC8; secondary hypertension and primary heart, liver, and brain disorder were excluded. We did not limit inclusion based on age, sex, case source, disease course, hypertensive classification, or antihypertensives.
. . . Exclusion of Studies. Studies were excluded if they were (1) clinical trials from which no relevant data could be extracted; (2) studies that were published repeatedly; (3) populated with inconsistent baseline information (age, sex, case source, disease course, hypertensive classification, or antihypertensive drugs); (4) systematic review, important data report, and case reports; no reply from corresponding authors such that further data could not be obtained; and (5) therapeutic measures failing to meet the predetermined inclusion criteria.
. . . Intervention. Experimental group received breviscapine injection combined with antihypertensive drugs [captopril, amlodipine, lisinopril, benazepril, losartan potassium, felodipine, and nifedipine (medication dose, medication time and frequency, and treatment course)]. Control group received antihypertensive drugs including captopril, amlodipine, lisinopril, benazepril, losartan potassium, felodipine, nifedipine (medication dose, medication time and frequency, and treatment course). Age, sex, and other baseline conditions of research subjects were well matched.
. . . Data Extraction. Two evaluators independently performed a search according to the search strategy, and preliminary screening was based on independent topics and abstracts of the search results, excluding obviously unqualified documents. A full-text methodology screening was conducted on the literature that might meet the inclusion criteria, and the authors were contacted when there was incomplete information. Then, the studies were cross-checked by two evaluators. Any disagreement on the conclusion of two evaluators was resolved by discussion. If such disagreement could not be resolved through discussion, final judgment and arbitration were made by a third party. Extracted contents included authors' names, year of publication, number of samples, intervention, course of treatment, and observed indicators.
. . Quality Evaluation. The investigators simultaneously evaluated the bias risk of the included studies based on the "risk of bias" evaluation tool in the Cochrane Handbook for Systematic Reviews [42] of interventions and relevant assessment guideline regulations. This risk-evaluation tool contains seven items, (1) random sequence generation; (2) allocation concealment; (3) blinding of the participants and personnel; (4) blinding of the outcome data; (5) incomplete outcome data; (6) selective reporting; and (7) other bias, and were evaluated as having a "high risk of bias", "low risk of bias", or "unclear risk of bias" according to assessment criteria.
. . Data Analysis. (1) Stata 14.0 software was used to perform the statistical analysis for the meta-analysis [43]. (2) Select effect size: if an index of the included documents is a binary variable, the curative effect analysis statistics can be represented by relative risk (RR) and expressed by its confidence interval (CI); mean difference (MD) and 95% CI were used to represent continuous changes. (3) Homogeneity test: it tests the variation degree of original research results and clearly includes the degree of homogeneity of the experiment. (4) Meta-analysis: according to the result of the heterogeneity test, P ≥ 0.05 and 2 < 50 indicate that the results have good agreement and that the fixed effect model may be used.
< 0.05 and 2 ≥ 50 suggest that the heterogeneity of the results cannot be ignored. If the included studies still have clinical significance, the random effects model may be used. (5) Sensitivity analysis: in those meta-analyses of the comprehensive factors combined with multiple outcomes, possible anomalous studies were ruled out before reevaluation. The results were compared with those of meta-analysis before the exclusion to determine how the excluded studies would influence the combined effect size and the stability of metaanalysis. If there is little difference between the two results, then the sensitivity of the results is relatively low, and the results are stable, indicating high credibility. (6) Subgroup analysis: subgroup analysis was conducted on some indexes with high heterogeneity. For events in which quantitative synthesis was impossible and events with very low incidence, qualitative evaluation may be based on the description. In this study, Stata 14.0 software was used to conduct a sensitivity analysis and subgroup analysis and to create a sensitivity analysis chart.
. . Publication Bias. Publication bias occurs when positive data in similar research papers with statistical significance are more likely to be published in journals. This situation is hard to control. The funnel plot method is often used to detect publication bias. Egger's test was performed to detect publication bias in the outcome measures. If a large publication bias was found in a particular research index, the exact reason was identified.

Results
. . Search Results. A total of 411 documents [PubMed (n = 125), the Cochrane Library (n= 29), Embase (n = 124), Sino Med (n = 28), CNKI (n = 47), Wanfang Data (n =27), and VIP (n = 31)] met the data collection and search strategy conditions. NoteExpress, a professional document management software, was employed to check for duplication of the 411 obtained articles that met the relevance requirement. The majority of these trials were excluded because some papers were found in more than one database and some included irrelevant titles and abstracts. Only 159 studies were retrieved. Following a review of the titles and abstracts, several studies were excluded, and only 102 studies remained. Five trials were excluded because of duplicated publications. Twenty-seven trials were excluded for being animal studies, and twenty-five trials were excluded for being nonclinical trials, including case reports, pharmacokinetic studies, and conference abstracts. Eighty-eight out of the remaining 102 articles were excluded based on the inclusion criteria, leaving fourteen RCTs to be reviewed in Figure 1.

. . Summary of the Quality and Bias Risk of the Trials
Included. The quality of all studies included was low. All studies were carried out among the Chinese population. Fourteen studies mention the use of random allocation: All studies failed to mention the specific grouping method, and none of the studies discussed allocation concealment, blinding, or evaluator blinding. The quality assessment is shown in Figures 2 and 3   ≤ 0.001) (Figure 4), we employed a random effects model. A funnel plot analysis of the 12 trials suggested possible publication bias and inclusion of low quality studies as significant asymmetry is shown in Figure 5. We applied Egger's test to evaluate publication bias. A p (P = 0.634) value more than 0.05 was considered no publication bias (Supplementary Figure S1). The meta-analysis revealed that the experimental group performed better than the control group in reducing Scr [WMD = -35.16, 95% CI(-50.01, -20.31), ≤ 0.001] (Supplementary Figure S2).
. . . Sensitivity Analysis. We conducted a sensitivity analysis for Scr (Supplementary Figure S3). By seriatim excluding one trial each time and reperforming meta-analysis of the remaining trials, we could observe whether the outcomes have dramatically changed. Sensitivity analysis indicated that the outcomes of Scr were very similar, which had relatively good stability.
. . . Subgroup Analysis. Because of variability in evaluating point of the serum creatinine, we conducted subgroup analysis among studies using different doses of breviscapine injection (30ml, 20 ml, 12 ml, 10 ml, and 5 ml). Compared with the control groups, the results of subgroup analysis showed that there was no significant correlation between the decrease of serum creatinine and the dose of breviscapine injection (Figure 6).

. . . Comparison of Clinical Efficacy.
A total of 7 studies [31,33,35,36,[38][39][40] reported the results of the total effective rate, involving 657 patients (330 in the experimental group and 327 in the control group). After the test for heterogeneity ( 2 = 85.2%, ≤ 0.001) (Figure 12), we employed a random effects model. We conducted a sensitivity analysis and applied Egger's test ( = 0.181) to evaluate publication bias for effective rate (Supplementary Figures S13 and S14). Meta-analysis indicated that the total clinical effective rate of breviscapine injection plus antihypertensive drugs versus antihypertensive drugs alone in treating for hypertensive nephropathy was higher than that in the control group, which showed a statistically significant difference [RR = 1.27, 95% CI(1.05, 1.53), = 0.014] ( Figure S15). All included trials were published in Chinese academic journals. Since trials with negative or neutral results are less likely to be published,    the efficacy of published studies might be overestimated. Consequently, the possibility of publication bias could not be ruled out.
. . . Adverse Effects. Nine of the included trials [29][30][31][32][33][34][35][36]38] described adverse effects in detail, while the others did not mention adverse events. Only one showed mild facial flushing during intravenous dripping of breviscapine [30], and after the speed of transfusion was slowed down, the symptom got remitted. Two cases [38] showed redness of limb skin in the experimental group. There were two cases [31,32] of head inflation during the use of antihypertensive drugs. There were three cases [33,34] of head swelling and dizziness during infusion during the use of breviscapine injection. Twenty-one cases showed dry cough due to the use of ACEI in six trials [29, 31-33, 35, 36]. None of the adverse events were serious. These symptoms are self-limited and did not affect treatment.

Discussion
This meta-analysis included 14 studies with 1,170 total participants comparing breviscapine injection plus antihypertensive drugs versus antihypertensive drugs alone for hypertensive nephropathy. Meanwhile, the results demonstrated that the expression levels of SCr and BUN were significantly lower in patients treated with breviscapine injection in comparison with control subjects, and the improvement of creatinine clearance was higher in patients treated with breviscapine in comparison with control subjects, suggesting that the drug serves as a protective role in the renal system of patients with hypertensive nephropathy. Microalbuminuria is regarded as the earliest clinical sign of hypertensive nephropathy. It is defined as a urinary albumin excretion rate ranging from 30 to 300 mg/day, and the definitive measurement is based on a timed urine collection during a 24-h period. The present meta-analysis indicated that breviscapine injection can reduce 24-hour urinary protein and the urinary beta-2-microglobulin; a reduction in urinary protein may contribute towards the renal protective effect of breviscapine injection in patients with hypertensive nephropathy. Our analysis revealed that experimental groups showed better overall clinical efficacy than control groups.
Erigeron breviscapus is a kind of traditional Chinese medicine. It was first recorded in the book of "Yunnan Materia Medica". According to Chinese medicine theory, erigeron breviscapus is cold-natured, sweet, bitter, and pungent in taste, with the function of clearing heat, relieving toxicity, eliminating wind and dampness, activating blood and removing stasis, expediting channel and activating meridian, relieving inflammation, and alleviating pain. The monomer component of Chinese herbal medicine (CHM), also known as the natural pure compound drug, has recently attracted much attention. The natural extract artemisinin and its derivatives are good examples of monomer components of CHM that can treat diseases through various activities and can be a good starting point to uncover the mechanism of traditional Chinese medicine.
Hypertensive renal injury is major target-organ damage due to sustained high BP. Long-term hypertension can cause renal sclerosis and gradually progress to chronic renal failure. Positive control of hypertension is the key to preventing hypertensive renal damage. According to Chinese medicine theory, hypertensive renal injury is strongly related to fluid, phlegm, and dampness retention syndrome and liver-yang hyperactivity syndrome, which are caused by deficiency syndrome. Chinese medicines, which are used to treat fluid, phlegm, and dampness retention syndrome, deficiency syndrome, and liver-yang hyperactivity syndrome, respectively, have certain advantages with regard to treating hypertensive renal injury [45]. Clinical research indicates that Chinese herbal medicine might control increased SBP, inhibit the glomerular and tubular hyperplasia caused by high BP, and significantly reduce urinary albumin and beta-2-microglobulin by increasing the activity of renal rennin and the level of Ang II [46,47].

Limitations
Several limiting factors in this study should be considered. First, the quality of the included randomized controlled trials  was low. All included trials showed high or undefined risk of bias due to design, reporting, and methodology. There were no definitive, randomized, double-blind, placebo-controlled trials included in this meta-analysis. No trials reported detailed randomization methods or allocation concealment. No trial was double-blinded, and unfortunately, none of RCT (randomized controlled trial) has a placebo in the trial. Therefore, the reported strength of evidence should be reevaluated, and more rigorously designed, placebocontrolled trials are warranted to give high-level evidence in future studies [48]. Second, five trials did not report adverse reactions. Therefore, conclusions about safety cannot be made with confidence. Furthermore, certain active ingredients are chemically unstable, which limits large-scale synthesis. These pressing issues should be resolved in future research. The safety of breviscapine injection needs to be  strictly monitored and properly reported in future clinical trials. Third, all tests produced positive results, although most tests were conducted with small samples. We tried our best to avoid language bias and positional prejudice, but cannot rule out potential publication bias.
Overall, our results suggest that breviscapine injection is effective and safe for the treatment of hypertensive nephropathy, and this work has reference value for clinicians. More large-scale, multicenter, rigorously designed, randomized, controlled trials are needed to provide accurate data to further validate the effectiveness and safety of breviscapine injection.

Conclusions
Evidence from this systematic review shows that breviscapine injection in combination with antihypertensive drugs can improve clinical efficacy and creatinine clearance rate and reduce serum creatinine, blood urea nitrogen, 24hour urinary protein, and beta-2-microglobulin in hypertensive nephropathy patients. There is no evidence that breviscapine injection in combination with antihypertensive drugs can improve SBP in hypertensive nephropathy patients.

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

Acknowledgments
This work is supported by Social Development Key Programs of Science and Technology Commission Foundation of Jiangsu Province (grant BE2015730) and the first level of "333" Project of Jiangsu Province (grant BRA2016503).

Supplementary Materials
Supplementary Tables S1 and S2: search strategy. Supplementary Figure S1: Egger's test to evaluate publication bias